- Email: [email protected]
Evidentiality and epistemic modality in witness testimony in the context of Italian criminal trials
+ Models
PRAGMA-4420; No. of Pages 10
Available online at www.sciencedirect.com
ScienceDirect Journal of Pragmatics xxx (2017) xxx--xxx www.elsevier.com/locate/pragma
Evidentiality and epistemic modality in witness testimony in the context of Italian criminal trials§ Paolo Greco Università di Napoli ‘‘Federico II’’, Via Porta di Massa 1, 80100 Napoli, Italy
Abstract Languages without grammatical evidentiality, such as Italian, often express the category of ‘information source’ using linguistic means that involve interactions between evidentiality and epistemic modality. This paper aims to investigate such interactions in the linguistic strategies adopted by various players in a criminal trial situation. In the context of a trial, both the expression of the source of information and the variation in the degree of the speakers’ commitment to what they are saying are key factors. This study is based on an analysis of one inferential and one reportive strategy in Italian, as they occur in the legal testimonies given in two different Italian trials. We will also discuss the rhetorical effects related to the underlying ambiguities between a purely evidential and an evidential/epistemic reading of the investigated strategies. © 2017 Elsevier B.V. All rights reserved.
Keywords: Forensic linguistics; Evidentiality; Epistemic modality; Commitment; Italian linguistics; Rhetorical effects
1. Introduction This paper aims to investigate the interactions between evidentiality and epistemic modality in the linguistic strategies adopted by the different parties involved in a criminal trial. The variation in the level of speakers’ commitment to the truth of the information they are communicating, as well as the explicit reference to the source of this information (first-hand, reported or inferred) are central aspects of many speech acts produced in criminal trials. Witnesses’ oral testimonies therefore represent a context in which both the source of the information provided, and the relations between speakers’ statements and the truth, play a crucial role.1 Yet, the relevance of the interactions between evidentiality and epistemic modality in the language used by lay witnesses, judges and attorneys in the context of a criminal trial has not yet been pointed out in the literature. Our analysis is based on a corpus of legal testimonies from two Italian criminal trials.2 The witnesses participating in these trials have very different socio-economic backgrounds. Through the analysis of their language, we will also emphasize that the speakers’ differing sociolinguistic level and rhetorical skill play a decisive role in their capacity to manipulate the subtle distinction between the expression of the source of information and the epistemic stance of the speaker. §
I wish to thank two anonymous reviewers for their extremely helpful comments. All errors are my own. E-mail address: [email protected]. To find out more about the characteristics of the pragmatic situation in which speakers, and witnesses in particular, find themselves during a trial, we refer the reader to, for instance, Atkinson and Drew (1979), O’Barr and Conley (1985), Conley and O’Barr (1990), and, for the Italian situation, to Bellucci (2002) and Galatolo (2007). 2 The Italian legal system, while not completely adversarial in nature, is more adversarial than inquisitorial. Witnesses are questioned by the public prosecutor and by the defendant's lawyers, and the system always provides for the possibility of direct examinations and crossexaminations (see also Galatolo, 2007:196--199). 1
https://doi.org/10.1016/j.pragma.2017.10.005 0378-2166/© 2017 Elsevier B.V. All rights reserved.
Please cite this article in press as: Greco, P., Evidentiality and epistemic modality in witness testimony in the context of Italian criminal trials. Journal of Pragmatics (2017), https://doi.org/10.1016/j.pragma.2017.10.005
+ Models
PRAGMA-4420; No. of Pages 10
2
P. Greco / Journal of Pragmatics xxx (2017) xxx--xxx
As a result, witnesses’ testimonies provide fertile ground for the analysis of phenomena related to the interactions between evidentiality and epistemic modality, and, conversely, linguistic investigations on such interactions are crucial to describing the rhetorical strategies employed by the parties involved in a trial. Finally, witnesses’ testimonies also provide fertile ground for the study of the impact of sociolinguistic variation on the linguistic realization of evidentiality. This paper is organized as follows: in section 2, we provide a brief overview of the main theoretical questions regarding the relations between evidentiality and epistemic modality (section 2.1), and then a sketch of the Italian epistemic/ evidential system (section 2.2); in section 3, after a description of the corpus investigated in this paper (section 3.1), we move on to the core of the analysis: the study of the use of two evidentiality strategies (the ‘epistemic future’ and the ‘reportive conditional’) in the trials that we have analyzed (sections 3.2 and 3.3). Finally, in section 4, we highlight the main results of this study and we propose some potential avenues for future research. 2. On evidentiality and epistemic modality 2.1. The relations between evidentiality and epistemic modality: a brief overview In this section, we shall briefly discuss two fundamental preliminary aspects of this paper: the view of evidentiality we shall adopt here, and the notion of speaker's commitment we shall take as our basis; and what, according to the literature on this subject, the relations are between these two categories. According to Aikhenvald (2004), evidentiality is a linguistic category ‘‘whose primary meaning is source of information’’ (Aikhenvald, 2004:3). But what is a ‘linguistic category’3? Does the term ‘evidentiality’ refer to a grammatical category or to a semantic-functional domain that can (but does not necessarily have to) take the form of a grammatical category4? This has been a hotly debated issue over the years, and we shall not enter into this discussion here.5 In this paper, we shall maintain that evidentiality is a functional domain whose linguistic realization can belong to four macro-types: a) b) c) d)
specific and purely grammatical, and sometimes obligatory encoding (that is, ‘grammatical evidentiality’); non-specific and non-obligatory, but grammatical (or semi-grammatical) encoding6; non-specific, non-obligatory, and non-grammatical encoding (that is, ‘lexical evidentiality’). non-specific, non-obligatory, non-grammatical, and non-lexical encoding (for instance, intonational encoding).7
Italian, like the other Romance languages, cannot mark evidentiality through specific and purely grammatical means. In this paper, we shall mainly address class b) phenomena (and their epistemic overtones): we aim to show that direct and cross-examinations of witnesses provide fertile ground for the analysis of the semi-grammatical encoding of evidentiality. This paper will use the definition of epistemic modality given by Nuyts (2006:6), namely ‘‘an indication of the estimation, typically, but not necessarily, by the speaker, of the chances that the state of affairs expressed in the clause applies in the world’’. The question of the relations existing between evidentiality and epistemic modality is one that has been a bone of contention since at least the 1980s.8 Behind the supposed connection between evidentiality and epistemic modality lies the idea that there is to some extent an obvious relationship between the source of our information and our evaluation of its likelihood. The more directly we have acquired a piece of information, the more certain we are meant to be of what we are saying. However, as pointed out by de Haan (1999:87), ‘‘evidentials are in fact a priori unmarked with respect to a commitment to the truth of the speech utterance on the part of the speaker. Evidentials merely assert that there is evidence to back up the speaker's utterance. Any connection between the two [. . .] is secondary in nature. They encode different things (source of information vs. attitude towards that information). Although they are closely enough related to cause overlap in some languages, this overlap is not universal’’.
The notion of ‘linguistic category’ itself is, as is well-known, problematic. We refer the reader to the discussion that took place in the Lingtyp list in January 2016 (available online in The Lingtyp Archives) and to the related papers published in Linguistic Typology (20.2, 2016, pp. 297--462). 4 Aikhenvald (2007) uses ‘information source’ as a technical term to refer to this semantic-functional domain, opposed to ‘evidentiality’, a term the author in question exclusively reserves for the grammatical category. 5 There is a considerable body of literature on this subject. A discussion of the various positions can be found in Greco (2012:55--75), and in Dendale and Van Bogaert (2012). 6 Aikhenvald (2004) refers to such means of encoding the source of information as ‘‘evidentiality strategies’’. 7 See Ifantidou (2001). 8 Detailed accounts of this debate can be found in Plungian (2010), Greco (2012:76--94) and, more recently, Brugmann and Macaulay (2015:205--208). 3
Please cite this article in press as: Greco, P., Evidentiality and epistemic modality in witness testimony in the context of Italian criminal trials. Journal of Pragmatics (2017), https://doi.org/10.1016/j.pragma.2017.10.005
+ Models
PRAGMA-4420; No. of Pages 10
P. Greco / Journal of Pragmatics xxx (2017) xxx--xxx
3
That is why nowadays a theoretical distinction is generally accepted between ‘information source’ and the commitment of the speaker to the truth of what he or she is saying. Yet the relations between these two domains are not always interpreted in the same way.9 The position that will be taken in this paper is that evidentiality and epistemic modality are two distinct but contiguous categories, which in many languages (including Italian) are characterized by overlapping zones with regard to preferential relations. These connections are based on the fact that some sources are generally considered more reliable than others (direct observation as opposed to indirect knowledge, for instance) and that a more reliable source often involves a stronger commitment on the part of the speaker to the truth of the information provided.10 2.2. Evidentiality and epistemic modality in Italian Italian is considered to have at least four semi-grammaticalized forms in which modal and evidential values are intertwined11: the expression si vede che (lit. ‘one sees that’); the verb dovere (‘must’) in the indicative mood; the verb dovere in the conditional mood; and the verb potere (‘can’) in the conditional mood. All these constructions carry inferential values, but they differ in their epistemic meanings: the order in which we presented them ranges from the highest degree of certainty (si vede che, dovere in the indicative mood) to the weakest degree of certainty ( potere in the conditional).12 Moreover, according to Pietrandrea (2005, 2007), Italian has two purely evidential forms: the semi-grammaticalized expression dice che (‘[someone] says that’),13 and the condizionale riportivo or the condizionale citativo (lit. the ‘reportive conditional’ or the ‘quotative conditional’, respectively), so the conditional used as a reportive marker.14 Finally, particularly in spoken language, an evidential extension of the values of the imperfect indicative has also been recognized, specifically in the use of the imperfect with a reportive function (see for instance Squartini, 2001; Calaresu, 2004:197--202; Giacalone Ramat and Topadze, 2007:24--25). Lastly, in some contexts the future tense conveys epistemic values. However, the status of the ‘epistemic’ future is somewhat controversial: Squartini (2008) considers it an evidential (inferential) marker, while Pietrandrea (2005, 2007) maintains that it has a purely epistemic function (pure conjecture). In this paper, we maintain that the epistemic future is an inferential marker. 3. Analysis 3.1. The corpus This study was conducted drawing on a corpus of witnesses’ testimonies in court. In particular, we have analyzed: a) the official transcripts of the direct examinations and cross-examinations of six lay witnesses conducted by the prosecutor and by the defense lawyers in one criminal trial (henceforth trial A) held between 2010 and 2011, for a total of around 42,000 words;
9 See the articulate paradigm and the accurate account proposed by Brugmann and Macaulay (2015). On the relationships between evidentiality and other categories (including epistemic modality), see also Aikhenvald (2015). 10 Konning (2003) provides a model that can account for the relationships between evidentiality and epistemic modality, which is particularly well suited to Romance languages. 11 The Italian modal/evidential system has been discussed in a number of contributions: see for instance Squartini (2001, 2004, 2005, 2008), Pietrandrea (2005, 2007) and Giacalone Ramat and Topadze (2007). See also Cruschina and Remberger (2008) for the domain of reported speech in some Italo-romance (and Hispano- and Daco-romance varieties). 12 This description is consistent with the one provided by Pietrandrea (2005, 2007), though not completely identical. 13 This expression can also occur in the form dice (‘says’) without the complementizer che (‘that’). For further details about this marker, see Lorenzetti (2002), and Calaresu (2004). For some comparable, but more grammaticalized, forms in other Romance varieties, see Travis (2006) and Cruschina and Remberger (2008). 14 Although Pietrandrea (2005, 2007) considers this use of the conditional to be a purely evidential marker, the condizionale riportivo is also known in Italian as the condizionale di riserva or the condizionale di dissociazione (lit. the ‘conditional of doubt’ or the ‘distancing conditional’, respectively) (see for instance Mortara Garavelli, 1995). The labels that have been given to this use of the conditional clearly emphasize the overlap between modality and evidentiality that is often to be found in this use of the conditional. This conditional is frequently (but not always) used by speakers when they want to stress that they are not committed to the truth of the information reported. The alternation between the use of, on the one hand, purely evidential reportive conditionals and, on the other, reportive conditionals with clearly epistemic overtones, is particularly significant in the context of a criminal trial (see section 3.3). For more detail on the reportive conditional in Italian, see, among others, Squartini (2001, 2005:247--250), Calaresu (2004), and Giacalone Ramat and Topadze (2007:21--24).
Please cite this article in press as: Greco, P., Evidentiality and epistemic modality in witness testimony in the context of Italian criminal trials. Journal of Pragmatics (2017), https://doi.org/10.1016/j.pragma.2017.10.005
+ Models
PRAGMA-4420; No. of Pages 10
4
P. Greco / Journal of Pragmatics xxx (2017) xxx--xxx
b) the audio recordings of the direct and cross-examination of one witness in a trial against former Italian Prime Minister Silvio Berlusconi (henceforth trial B).15 As far as trial A is concerned, we have systematically changed all the names and toponyms for privacy reasons. In this trial, attorneys and witnesses clearly come from different sociolinguistic backgrounds. All the witnesses come from deprived socio-economic backgrounds, and most of them have been involved in different kinds of criminal activities (or at least they are closely acquainted with criminals). By contrast, we did not change the names in the trial against Berlusconi. In this case, the identity of the witness is of some significance: he is Antonio Di Pietro, a former public prosecutor and leading Italian political adversary of Silvio Berlusconi who previously served as a government minister.16 As far as the structures that we have analyzed are concerned, while we acknowledge that in Italian both evidentiality and epistemic modality are mainly linguistically encoded by lexical means, we decided to limit the scope of our study exclusively to the grammatical encoding of these values. This is mainly because linguistic exchanges in the context of a trial are regulated by a series of constraints that prevent the use of certain expressions.17 In particular, we will be concerned with some uses of two evidential strategies in which evidential and epistemic values are tightly intertwined: the epistemic future, and the reportive conditional. We will show how more educated speakers (such as prosecutors or attorneys, but also one of the witnesses) exploit the ambiguous boundaries between the evidential and the epistemic reading of these strategies to try to influence the judge's decision. This study is therefore intended to be a qualitative analysis of a selection of excerpts that demonstrate the relevance of the interactions between evidentiality and epistemic modality during a criminal trial, and the ways in which more educated, skilled and trained speakers are able to exploit the ambiguity of two Italian linguistic strategies, the interpretation of which can be purely evidential or evidential and epistemic. However, a brief overview of the quantitative data that we have gathered will allow us to pinpoint some fundamental issues regarding the use of reportive conditionals and epistemic futures in the trials that we have investigated. Epistemic futures occur infrequently in the corpus: we found only 17 instances of this use of the future.18 More generally, it seems that the scope for inferences is very limited in our trials19: we found only 139 occurrences of inferences. Seventy-six of these instances are formulated through lexical means (54.7%), the other 63 are realized through semi-grammaticalized strategies: modal verbs (such as potere or dovere) on 46 occasions (33.1%); epistemic futures, as we have already seen, in the remaining 17 cases (12.2%). Epistemic futures are therefore very rare. Yet, as we shall see in the next section they represent a strategy for attorneys to implicitly communicate what they cannot state explicitly, playing on the boundaries between evidentiality and epistemic modality. Reportive strategies occur much more frequently in our corpus: we found 517 contexts in which they are used. In most cases, the verb encoding the reported information is expressed in the indicative (293 instances, 56.7% of the total). Another common strategy is the direct quotation (152 cases, 29.4% of the total). Reportive conditionals occur less frequently: we only found 46 instances of this strategy (8.9% of the total).20 Yet their use is of particular interest to us. They represent a marked choice, and therefore an ideal strategy to suggest (without any explicit statement) that we are communicating something more than pure reportive evidentiality.21 3.2. Evidentiality and speaker's commitment in court: the epistemic future The reliability of witnesses’ statements has a fundamental role to play in a trial situation. Of course, the lawyers of the defendants and the prosecutor frequently try to demonstrate the contradictions arising in witnesses’ accounts, and/or the weaknesses of the evidence they provide. Yet, they are not allowed to directly question the accounts provided by the witnesses (which by law must be considered truthful unless a clear contradiction arises). Given these constraints,
15 The audio recordings are freely available on the Radio Radicale internet site (www.radioradicale.it), and are published under the Creative Common license. The examination and the cross-examination of the witness took place on July 9th, 2014 and lasted for four hours and fifteen minutes. 16 Di Pietro became very well known in Italy in the 1990s, when he was one of the foremost prosecutors in the Mani pulite (‘Clean Hands’) judicial investigations into political corruption. 17 See for instance Galatolo (2007:196--199). 18 Epistemic futures in our corpus occur mainly in the future perfect form. We found only 4 occurrences of simple futures. 19 This statement holds probably true for the overall context of examination and cross-examination phase of trials. See O’Barr and Conley (1985:665--667) and Galatolo (2007:196--199). 20 Less frequent is the use of the infinitive form of the verb to encode the reported information (26 occurrences, 5% of the total). 21 See also the example provided by Calaresu (2004:193), in which a speaker reports someone else's words, alternating the indicative and the conditional on the basis of his estimation of the likelihood of the truth of the information provided by the original speaker.
Please cite this article in press as: Greco, P., Evidentiality and epistemic modality in witness testimony in the context of Italian criminal trials. Journal of Pragmatics (2017), https://doi.org/10.1016/j.pragma.2017.10.005
+ Models
PRAGMA-4420; No. of Pages 10
P. Greco / Journal of Pragmatics xxx (2017) xxx--xxx
5
the more grammaticalized expressions of evidentiality (and their epistemic overtones) are particularly important, since they allow prosecutors, attorneys and witnesses to capitalize on the interplay between evidentiality and epistemic modality in order to reach their communicative goals. In this framework, the epistemic future represents a useful strategy for speakers to encode the variation in their commitment to the propositional content of what they are saying. (1) Di Pietro: Prob. . . presumibilmente, probabilmente ne avrò parlato con i capigruppo, con lo stesso Formisano, ne avrò parlato con altri parlamentari dell’Italia dei Valori, certamente ne ho parlato, appunto, con i diretti interessati con i pubblici ufficiali che stavano al Ministero, a cui l’ho consegnata. Di Pietro: Prob. . . presumably, probably I spoke22 about it [an audio recorded tape] with the parliamentary group leaders, with Formisano -- I spoke23 about it with other Members of Parliament belonging to Italia dei Valori -- I definitely spoke about it with the people directly concerned, with the public officials at the Ministry, whom I gave it to. The excerpt in (1) shows very clearly how the witness switches from an epistemic future (used with the adverb probabilmente (‘probably’)) to a present indicative (used with the adverb certamente (‘definitely’)), when he moves on from general recollections to the main point of his testimony.24 The passage in (2) shows how lawyers can use the epistemic future to surreptitiously challenge a witness's testimony without explicitly calling his or her claims into question. (2) Avvocato: Allora io voglio dire, lei sta lottando disperatamente, ce l’ha detto prima, per tenere i voti tutti i giorni. L’avrà chiesto a Caforio ‘‘ma ti ha detto chi sono le persone, soprattutto della mia parte politica, pronti a fare questo governo di larga coalizione?’’?. Lawyer: So, I mean, you are desperately struggling, as you told us before, every day to keep your votes. You must have asked25 Caforio, ‘‘But did he tell you who the people are, especially from my party, who are ready to form a broad coalition government?’’. The use of the epistemic future (in the perfect) with a peculiar prosodic pattern (an interrogative intonation and a strong stress on the auxiliary's tonic vowel), allows the lawyer to specify that the alleged behavior of the witness was somewhat strange, without explicitly questioning the truth of his claims. This type of context allows the speaker to set up a framework in which the actions described by the predicate in the future perfect are considered logical and obvious, to such an extent indeed that any other behavior sounds strange or even suspect. As a result -- and this is clearly the implicit aim of the strategy being pursued by the lawyer -- the account given by the witness is considered unreliable. 3.3. Evidentiality and speaker's commitment in court: the reportive conditional In section 3.1 we have shown that in our corpus reportive evidentiality is generally encoded by the simple indicative, and that the conditional occurs in less than 10% of the reportive contexts. However, the reportive conditional has a very different frequency of occurrence in the two trials that we have analyzed. In trial A, out of 291 reportive contexts, 162 (55.7%) show an indicative and only 13 (4.5%) are characterized by a conditional. Therefore, despite the significant number of reportive contexts, in trial A the reportive conditional seldom occurs (and never in witness testimony). In trial B, meanwhile, we found 33 occurrences of reportive conditionals and 131 indicatives in a total of 226 reportive contexts. While the percentage of use of the indicative is similar in the two trials (55.7% in trial A, 58% in trial B), in trial B there is a significant rise in the frequency of the reportive conditional (14.6% of cases, compared to 4.5% in trial A). The main difference between the two trials lies in the fact that in trial A, reportive conditionals only occur in the language of lawyers, prosecutors and judges, while in trial B the witness employs this linguistic strategy too. The reportive conditional is therefore a marked stylistic choice in both trials, but in trial B, where both the attorneys and the witness are trained speakers coming from an educated background, its use is more frequent. The functional values of the reportive conditionals can be very different. On certain occasions, the evidential value of the reportive conditional is clearly prominent, and the possible epistemic overtones, if present, are secondary in nature. The clearest cases are perhaps those in which the conditional is used by judges as a sign of impartiality (although they also use the indicative in similar contexts):
22
The verb parlare (‘to speak’) is in the future perfect. See footnote 22. 24 This witness's testimony is in fact based on his claim that he had received an audio tape allegedly featuring data that could be relevant to the trial. According to the witness, he never listened to the tape, which he immediately passed on to some public officials so that they could check whether the material on the tape had criminal repercussions or not. This tape has now gone missing. 25 The verb chiedere (‘to ask’) is in the future perfect. The intonation of the utterance is interrogative. 23
Please cite this article in press as: Greco, P., Evidentiality and epistemic modality in witness testimony in the context of Italian criminal trials. Journal of Pragmatics (2017), https://doi.org/10.1016/j.pragma.2017.10.005
+ Models
PRAGMA-4420; No. of Pages 10
6
P. Greco / Journal of Pragmatics xxx (2017) xxx--xxx
(3) Presidente: Allora la difesa dice: ‘‘Come mai lei non ha mai fatto cenno nel corso degli interrogatori resi precedentemente a questa circostanza e cioè che Maria avrebbe anche visto, avrebbe anche incrociato, c’è questo riferimento anche alla presenza di suo figlio’’. Judge: Then the defense said: ‘‘Why in the previous hearings didn’t you mention this fact, namely that Maria also saw,26 that she came across27 [these people], and there is also this reference to the presence of her son’’. Conditionals are also used when an attorney rephrases the words of a witness who is contradicting himself/herself. (4) Presidente: Qui dice ‘‘la mia convivente mi riferì che a sparare all’impazzata contro la folla nella zona di Posillipo erano stati Francesco e Mario quest’ultima mi aggiunse che aveva appreso tale notizia in famiglia e direttamente dagli interessati’’. Avvocato A: Quindi non l’aveva visti! [. . .] Avvocato B: Sul verbale di oggi [. . .] ha detto che la convivente gli avrebbe riferito che li avrebbe visti quando stava a Posillipo. Questo è il verbale di oggi.28 Judge: Here you say, ‘‘My partner told me that Francesco and Mario wildly shot into the crowd in the Posillipo area. She added that she found out about this from her family and directly from the people involved’’. Lawyer A: So she didn’t see them! [. . .] Lawyer B: In today's official transcript [. . .], he said that his partner told him29 that she saw30 them when she was in the Posillipo area. This is today's official transcript. In the example above, the conditional is used not only to indicate the second-hand nature of the information provided, but also to suggest that the witness is unreliable. The contradictions between the differing versions of the same story given by the witness on different occasions are somewhat emphasized by the use of the conditional in the lawyer's account. In other contexts, the epistemic overtones are even more prominent. As we have seen in the previous section, in certain contexts the use of the epistemic future allows attorneys to challenge the words of the witnesses without resorting to more explicit means, for instance lexical items, that would probably lead to a formal request by the court to rephrase the passage in question.31 This aspect is particularly clear in the use of the reportive conditional in trial B. In this trial, the conditional is widely used by lawyers when they want to surreptitiously discredit the information provided by the witness, or at least when they wish to emphasize that this information is based on someone's account of events and that there is no hard evidence for what has been said.32 (5) Avvocato: Abbiamo appreso che Caforio sarebbe stato avvicinato da Tomassini, e di questo abbiamo già parlato io non intendo più riproporre il problema, e da De Gregorio. Lei stamattina ha detto al pubblico ministero che Caforio le avrebbe detto che De Gregorio avrebbe rappresentato che dietro di lui vi era Berlusconi, che i soldi sarebbero stati dati, anzi ha parlato di Forza Italia per la verità, e di Berlusconi. Questo ha detto al Pubblico Ministero stamattina. Lawyer: We found out that Caforio was contacted33 by Tomassini, we have already talked about this and I will not raise the issue again, and by De Gregorio. This morning, you told the public prosecutor that Caforio told34 you that De Gregorio said35 that Berlusconi was behind him and that the money had been paid,36 and indeed you talked about Forza Italia,37 and about Berlusconi. This is what you said to the public prosecutor this morning.
26
The verb vedere (‘to see’) is in the perfect conditional. The verb incrociare (‘to come across’) is in the perfect conditional. 28 The conditional was not used, of course, by the witness in the original account of the facts. 29 The verb riferire (‘to tell’) is in the perfect conditional. 30 See footnote 26. 31 On the strategies adopted by attorneys in courtroom cross-examinations, see also Drew (1992). 32 Indirect evidence (specifically, provided by a de relato, hearsay, witness) can be accepted in a criminal trial under certain circumstances, in accordance with Article 195 of the Italian Codice di Procedura Penale (Criminal Procedure Code). 33 The predicate essere avvicinato (in this context, ‘to be contacted’) is in the perfect conditional. 34 The verb dire (in this context, ‘to tell’) is in the perfect conditional. 35 The verb rappresentare (in this context, ‘to say’) is in the perfect conditional. 36 The verb dare (‘to give’) is in the passive perfect conditional. 37 An Italian political party, led by Silvio Berlusconi. 27
Please cite this article in press as: Greco, P., Evidentiality and epistemic modality in witness testimony in the context of Italian criminal trials. Journal of Pragmatics (2017), https://doi.org/10.1016/j.pragma.2017.10.005
+ Models
PRAGMA-4420; No. of Pages 10
P. Greco / Journal of Pragmatics xxx (2017) xxx--xxx
7
(6) Avvocato: Secondo le. . . secondo la sua narrazione di oggi, le avrebbero narrato solo una parte e la parte vera sarebbe contenuta nella registrazione Lawyer: According to. . . according to the account you have given today, they told38 you only part of the story, and the true part is given39 in the recording. It is clear from the examples above that the reportive conditional can be used to question the account of a given fact provided by a witness without explicitly challenging his or her words. The distancing effect of the conditional allows lawyers to treat the witness's testimony as an uncertain account of events, or, as the attorney says in (6), as a narrazione (a word which can mean ‘account’ but which is often used in Italian as a synonym for ‘tale’ or ‘story’). In the excerpts presented above, lawyers take advantage of the fact that the use of the conditional, as we have seen in example (3), may also convey a purely evidential meaning,40 while the modal meaning (non-commitment, uncertainty), although often associated with this, may be seen as an extension of it41: the modal value of uncertainty is only implied, and cannot be explicitly attributed to the lawyers. However, it is clear to any Italian speaker that the use of the conditional in the contexts presented in (5) and (6), and particularly in the latter case, strongly fosters an interpretation of the information provided as being something not far removed from a lie -- or at least an imaginative story. We have already noted that the reportive conditional never occurs in the witnesses’ testimonies, with one exception: that of Antonio Di Pietro. (7) Di Pietro: Lui dice, ha riferito, si riferiva al fatto che Lavitola avrebbe ricevuto, avrebbe detto che anche lui, Formisano avrebbe ricevuto 50000 euro da Berlusconi. Di Pietro: He says, he told me, that he was referring to the fact that Lavitola received,42 [and that he also] said43 that also Formisano received44 s50,000 from Berlusconi. (8) Di Pietro: A me ha sconvolto non la vicenda di Caforio [. . .] mi ha sconvolto, e ribadisco che mi sconvolge ancora adesso molto, che chi ha accompagnato il Senatore Caforio, e cioè Formisano, né in quel momento, né davanti a me, né davanti alla pubblica accusa ha ritenuto di dire un fatto importantissimo e cioè, ribadisco, che Lavitola è andato a trovarlo nell’Ufficio del Senatore, di Formisano, poi sarà vero non sarà vero, lo diranno loro, e davanti a Formisano avrebbe telefonato a Marinella Brambilla dicendo di farsi ritelefonare da Berlusconi, e subito dopo Berlusconi l’avrebbe ritelefonato, a dire del Formisano, io non lo so, non c’ero lì, sono de relato, sentirete Formisano, [Berlusconi] avrebbe telefonato e gli avrebbe detto, a Lavitola, di dire a Formisano che c’era davanti a lui [Lavitola] «We, promettigli qualsiasi cosa che siamo disponibili a tutto».45 Di Pietro: I wasn’t shocked by what Caforio did [. . .] I was shocked -- and I repeat that I’m still very shocked about this -- by the fact that Formisano, who accompanied Senator Caforio, decided to hide on that occasion, as well as when he was heard by the public prosecutor, a very important fact, namely, I repeat, that Lavitola met him in the office of the Senator, of Formisano. Now, they will say if this is true or not,46 and in the presence of Formisano he [Caforio] phoned47 Marinella Brambilla, asking to be called back by Berlusconi, and immediately afterwards Berlusconi rang back48; this is Formisano's version -- I don’t know, I wasn’t there, I am just saying what I heard; you will hear Formisano's account, he [Berlusconi] phoned49 and told50 him, Lavitola, to tell Formisano, who was standing right in front of him [Lavitola]: «Hey, promise him anything -- we’re willing to do anything». As clearly demonstrated by examples (7) and (8), in Di Pietro's account of events, reportive conditionals are relatively frequent (15 occurrences in 118 possible contexts, meaning 12.7% of the instances). He is a de relato witness, he knows
38
The verb narrare (‘to tell’) is in the perfect conditional. The verb contenere (‘to contain’) is in the perfect conditional. 40 See also the final statement of the attorney in (5): ‘‘Questo ha detto al Pubblico Ministero stamattina’’ (‘‘This is what you said to the public prosecutor this morning’’). With this statement, the attorney explicitly qualifies his own previous affirmations as reported. 41 This position is held for instance by Giacalone Ramat and Topadze (2007:22). 42 The verb ricevere (‘to receive’) is in the perfect conditional. 43 The verb dire (‘to say’) is in the perfect conditional. 44 See footnote 42. 45 On the use of direct reported speech in legal testimonies see Galatolo (2007). We cannot discuss the issue here, but Di Pietro's use of direct reported speech is also particularly interesting. 46 The expression used by Di Pietro could be literally translated as ‘it will be true, it will not be true’ with an epistemic future. 47 The verb telefonare (‘to phone’) is in the perfect conditional. 48 The verb ritelefonare (‘to ring back’) is in the perfect conditional. 49 See footnote 47. 50 See footnote 34. 39
Please cite this article in press as: Greco, P., Evidentiality and epistemic modality in witness testimony in the context of Italian criminal trials. Journal of Pragmatics (2017), https://doi.org/10.1016/j.pragma.2017.10.005
+ Models
PRAGMA-4420; No. of Pages 10
8
P. Greco / Journal of Pragmatics xxx (2017) xxx--xxx
what this means, and he knows that it is very likely that his words will carry no weight unless the individuals he is talking about confirm them. Therefore, he uses the conditional as a sign of impartiality. On many occasions, Di Pietro takes great care to use the present indicative for all the events he was party to, and the conditional for all the events that were reported to him. In this way, he clearly establishes a difference between what he wants to present as fact (namely that he received a recorded tape which, he was told, contained recordings of an attempt at bribery) and what he presents as second-hand information (namely accounts of this attempted bribery that were provided to him by some of the parties involved). A corollary of this strategy is, of course, that the only part of his account that could be called into question is the hearsay evidence, and not the facts he had a direct knowledge of, as the defendant's attorneys were trying to do. In the next excerpt, it is interesting to note that Di Pietro uses a conditional to express a piece of information he provided himself. In this case, the defendant's attorney is surreptitiously questioning Di Pietro's account of an event by using the conditional. In the passage presented in (9), Di Pietro adopts the shrewd strategy of using and repeating the conditional used by the lawyer (with an echoing effect),51 accepting the evidential value of the form (the attorney was referring to Di Pietro's claims, and he was therefore quoting second-hand information) but clearly rejecting the modal value of the verb form. He adds a precise temporal setting which strongly militates against an interpretation of non-commitment/uncertainty for the use of the conditional here. This strategy allows Di Pietro to reveal the lawyer's ruse: (9) Avvocato: [in a text written by Di Pietro and submitted to the court] Lei riferisce con un corsivo quello che le avrebbe narrato il Caforio. Di Pietro -- Che mi avrebbe narrato il 4 marzo 2013. Lawyer: [in a text written by Di Pietro and submitted to the court] You put what Caforio told52 you in italics. Di Pietro: What he told53 me on March 4, 2013. In these last examples, we have seen how evidentiality and epistemic modality interact in the actual uses of the reportive conditional in the examination and in the cross-examination of a witness in a criminal trial in which both the attorneys and the witness are trained speakers. In the light of the discussion above, we maintain that the core value of the reportive conditional is purely reportive (see example (3)). However, since the epistemic overtone of non-commitment is often associated with the reportive conditional, trained speakers can exploit this preferential relationship between the reportive conditional and a low degree of speaker's commitment, to surreptitiously question the truth and the validity of certain statements without overtly questioning them (above all in contexts in which such questioning would be inappropriate, or, as in the case of an attorney talking about the testimony given by a witness, simply not allowed). 4. Discussion and conclusions This study aimed to investigate the rhetorical functions of the variations in the interactions between evidentiality and epistemic modality in the linguistic strategies adopted by attorneys, prosecutors and witnesses in two different Italian criminal trials. We also aimed to highlight the impact of different sociolinguistic backgrounds on the speakers’ abilities to exploit the possibilities offered by these interactions in two evidential strategies, namely the epistemic future and the reportive conditional. The particular characteristics of the codified interactions between attorneys and witnesses provide fertile ground for the study of both the relationships between evidentiality and epistemic modality in Italian, and the rhetorical strategies adopted by the different parties involved in a criminal trial in some specific contexts. In criminal trials, the law forbids certain kinds of statement. For instance, attorneys and prosecutors are not allowed to directly question the words of a witness in the absence of clear contradictions. However, as we saw in sections 3.2 and 3.3, they often challenge the information provided by witnesses by using linguistic strategies such as the reportive conditional or the epistemic future. This paper represents one of the first explorations of the use of evidential markers, and of the rhetorical strategies exploiting the ambiguous boundaries between evidentiality and epistemic modality, as they occur in criminal trials. The analytic results are language-specific, but we hope to have shown the potentialities of the study of evidential markers and of the relationships between evidentiality and epistemic modality in the context of criminal trials. We can in fact assume that aspects related to the source of information, the speaker's commitment, and the semantic and pragmatic continuities
51 52 53
On echo utterances, and on their pragmatic functions, see Yamaguchi (1994). The verb narrare (‘to tell’, ‘to narrate’) is in the perfect conditional. See footnote 52.
Please cite this article in press as: Greco, P., Evidentiality and epistemic modality in witness testimony in the context of Italian criminal trials. Journal of Pragmatics (2017), https://doi.org/10.1016/j.pragma.2017.10.005
+ Models
PRAGMA-4420; No. of Pages 10
P. Greco / Journal of Pragmatics xxx (2017) xxx--xxx
9
between evidentiality and epistemic modality are crucial in criminal trials in many countries (particularly those with an adversarial system). As far as the analytical results are concerned, in this paper we have described how trained public speakers (such as attorneys and prosecutors) capitalize on the mutual relationship between the evidential and epistemic values of verbal forms such as the future and the conditional. For instance, in their use of reportive conditionals, in many contexts they seem to adeptly make use of the subtle boundaries between what is simply reported as fact and what is reported but implicitly considered uncertain. This double interpretative possibility of the Italian reportive conditional, which is strictly related to a more general typological characteristic of reportive evidentials in many languages of the world,54 is at the basis of the different interpretations of the values of the Italian reportive conditional that are to be found in the literature.55 In the light of the discussion presented in this paper, we maintain that the core function of the Italian reportive conditional is that of producing a distancing effect. This effect can be of a purely evidential nature, or, more often, of an evidential nature with epistemic overtones. Therefore, the reportive conditional is an evidential marker which is often associated with an epistemic overtone of non-commitment/ uncertainty.56 The structures conveying inferential values occur less frequently in our corpus. However, we have observed a particular use of the Italian epistemic future perfect (example (2)), that we would like to further discuss here. The Italian future perfect, in its epistemic interpretation, generally conveys conjectures or inferences (see Bertinetto, 1995:123). Yet, in some contexts, the use of the future perfect with a peculiar intonational pattern (interrogative, with a strong accent on the auxiliary) allows the speaker to set up a frame in which every deviation from the state of affairs described by the speaker should be regarded as very strange (or even suspicious, as in the excerpt discussed in example (2)). The presupposition that lies behind these statements is that on the basis of our general knowledge of the world we can assume that it is almost obvious that the state of affairs described in the proposition has to be true. As a result, these future perfect constructions carry strong inferences, and convey the speaker's total conviction that the information they contain is true. We suggest calling this use of the epistemic future (which, to the best of our knowledge, has not yet been described in the literature) the ‘polemical future’,57 because its rhetorical effect is to argue that any state of affairs other than the one described in the sentence would be very strange. Through this study, we also intended to address a more general question, namely that of the fundamental role played by discourse traditions in the analysis of evidential and modal markers58: different contexts can lead to linguistic strategies being used more (or less) consciously or carefully. Therefore, we believe that the use and functions of evidential and modal markers in a given language should be studied in their context.59 In this paper, we have analyzed their occurrence in the context of a trial, but studies conducted in other situations from everyday life will undoubtedly point to different uses of these markers. The choice of which linguistic means to use to encode evidential and/or epistemic values is also certainly sensitive to sociolinguistic variation: people coming from different sociolinguistic backgrounds show different abilities in the use of inferential and reportive markers. Speakers coming from more educated backgrounds are able to exploit the ambiguities related to the interactions between evidentiality and epistemic modality, in order to achieve different communicative goals. Thus far, these factors have generally been neglected in studies on the evidentiality/epistemic modality interface. We are convinced that this research path is promising, and that further studies analyzing evidential and modal markers in the light of sociolinguistic and discourse tradition factors will shed light on aspects of the use of these markers that have not yet been recognized. They will also help us to reach a better understanding of how we exploit the interactions between evidentiality and epistemic modality, by playing on the ambiguities of the markers encoding these values. References Aikhenvald, Alexandra Y., 2004. Evidentiality. Oxford University Press, Oxford. Aikhenvald, Alexandra Y., 2007. Information source and evidentiality: what can we conclude? Ital. J. Linguist. 19, 209--227. Aikhenvald, Alexandra, 2015. Evidentials: their links with other grammatical categories. Linguist. Typol. 19 (2), 239--277.
Aikhenvald (2004:180) highlights that in certain evidential systems the speaker ‘‘may choose to employ the reported evidential for two reasons. Firstly, to show his or her objectivity; that the speaker was not the eyewitness to an event and knows about it from someone else. Secondly, as means of ‘shifting’ responsibility for the information and relating facts considered unreliable’’. 55 See footnote 14. 56 For a discussion of the evidential and epistemic functions of conditionals in a typological perspective, see Aikhenvald (2004:105--109). 57 The ‘polemical future’ shares some characteristics with the so-called ‘‘valeur de renforcement’’ (Rossari et al., 2017) of the Italian epistemic future. However, these values only partially overlap, in that only some of the cases of epistemic futures indicated by Rossari et al. (2017) can be regarded as (at least partially) fitting the description we have given of the ‘polemical future’. 58 On the notion of discourse tradition see Koch (1997). 59 Grossmann and Tutin (2010) and Cornillie (2010) have previously emphasized the importance of such an approach. 54
Please cite this article in press as: Greco, P., Evidentiality and epistemic modality in witness testimony in the context of Italian criminal trials. Journal of Pragmatics (2017), https://doi.org/10.1016/j.pragma.2017.10.005
+ Models
PRAGMA-4420; No. of Pages 10
10
P. Greco / Journal of Pragmatics xxx (2017) xxx--xxx
Atkinson, J. Maxwell, Drew, Paul, 1979. Order in Court. The Macmillian Press, London. Bellucci, Patrizia, 2002. A onor del vero. Fondamenti di linguistica giudiziaria. UTET, Torino. Bertinetto, Pier Marco, 1995. Il verbo. In: Renzi, L., Salvi, G. (Eds.), Grande Grammatica Italiana di Consultazione, vol. II. Il Mulino, Bologna. pp. 13--161 (quoted from the second edition, 2001). Brugmann, Claudia M., Macaulay, Monica, 2015. Characterizing evidentiality. Linguist. Typol. 19 (2), 201--237. Calaresu, Emilia, 2004. Testuali parole: la dimensione pragmatica e testuale del discorso riportato. Franco Angeli, Milano. Conley, John M., O’Barr, William M., 1990. Rules versus Relationships. The Ethnography of Legal Discourse. The University of Chicago Press, Chicago. Cornillie, Bert, 2010. An interactional approach to evidential and epistemic adverbs in Spanish conversation. In: Diewald, G., Smirnova, E. (Eds.), Linguistic Realization of Evidentiality in European Languages. Mouton de Gruyter, Berlin/New York, pp. 309--330. Cruschina, Silvio, Remberger, Eva-Maria, 2008. Hearsay and reported speech: evidentiality in Romance. In: Benincà, Paola, Damonte, Federico, Penello, Nicoletta (Eds.), Selected Proceedings of the 34th Incontro di Grammatica Generativa. Special Issue of the Rivista di Grammatica Generativa, vol. 33. pp. 99--120, http://arcaold.unive.it/handle/10278/2504. de Haan, Ferdinand, 1999. Evidentiality and epistemic modality: setting boundaries. Southwest J. Linguist. 18, 83--101. Dendale, Patrick, Van Bogaert, Julie, 2012. Réflexions sur les critères de définition et les problèmes d’identification des marqueurs évidentiels en français. Lang. Fr. 173, 13--29. Drew, Paul, 1992. Contested evidence in courtroom cross-examination: the case of a trial for rape. In: Drew, P., Heritage, J. (Eds.), Talk at Work. Cambridge University Press, Cambridge. Galatolo, Renata, 2007. Active voicing in court. In: Holt, E., Clift, R. (Eds.), Reporting Talk. Reporting Speech in Interaction. Cambridge University Press, Cambridge, pp. 195--220. Giacalone Ramat, Anna, Topadze, Manana, 2007. The coding of evidentiality: a comparative look at Georgian and Italian. Ital. J. Linguist. 19, 7-38. Greco, Paolo, 2012. Evidenzialità. Storia, teoria e tipologia. Aracne, Roma. Grossmann, Francis, Tutin, Agnès, 2010. Evidential markers in French scientific writing: the case of the French verb ‘‘voir’’. In: Smirnova, E., Diewald, G. (Eds.), Linguistic Realization of Evidentiality in European Languages. Mouton de Gruyter, Berlin/New York, pp. 279--307. Ifantidou, Elly, 2001. Evidentials and Relevance. Benjamins, Amsterdam. Koch, Peter, 1997. Diskurstraditionen: zu ihrem sprachtheoretischen Status und ihrer Dynamik. In: Frank, B., Haye, Th., Tophinke, D. (Eds.), Gattungen mittelalterlicher Schriftlichkeit. Narr, Tübingen, pp. 43--79. Konning, Hans, 2003. Modalité et évidentialité. In: Birkelund, M., Boysem, G., Kjaersgaard, P.S. (Eds.), Aspects de la Modalité. Niemeyer, Tübingen, pp. 131--151. Lorenzetti, Luca, 2002. Sulla grammaticalizzazione di ‘dice’ nell’italiano parlato. In: Heinemann, S., Bernhard, G., Kattenbusch, D. (Eds.), Roma et Romania. Festschrift für Gerhard Ernst zum 65. Geburststag. Niemeyer, Tübingen, pp. 211--221. Mortara Garavelli, Bice, 1995. Il discorso riportato. In: Renzi, L., Salvi, G., Cardinaletti, A. (Eds.), Grande Grammatica Italiana di Consultazione, vol. III. Il Mulino, Bologna. pp. 429--470 (quoted from the second edition, 2001). Nuyts, Jan, 2006. Modality: overview and linguistic issues. In: Frawley, W. (Ed.), The Expression of Modality. Mouton de Gruyter, Berlin/New York, pp. 1--26. O’Barr, William B., Conley, John M., 1985. Litigant satisfaction versus legal adequacy in small claims court narratives. Law Soc. Rev. 19, 661--701. Pietrandrea, Paola, 2005. Epistemic Modality. Functional Properties and the Italian System. Benjamins, Amsterdam. Pietrandrea, Paola, 2007. The grammatical nature of some epistemic-evidential adverbs in spoken Italian. Ital. J. Linguist. 19, 39--63. Plungian, Vladimir A., 2010. Types of verbal evidentiality marking: an overview. In: Diewald, G., Smirnova, E. (Eds.), Linguistic Realization of Evidentiality in European Languages. Mouton de Gruyter, Berlin/New York, pp. 15--58. Rossari, C., Ricci, C., Siminiciuc, E., 2017. Les valeurs rhétoriques du futur en français, italien et roumain. In: Baranzini, L. (Ed.), Le futur dans les langues romanes. Peter Lang, Berne, pp. 49--77. Squartini, Mario, 2001. The internal structure of evidentiality in Romance. Stud. Lang. 25, 297--334. Squartini, Mario, 2004. Disentangling evidentiality and epistemic modality in Romance. Lingua 114, 873--895. Squartini, Mario, 2005. L’evidenzialità in rumeno e nelle altre lingue romanze. Z. Romanische Philol. 121, 246--268. Squartini, Mario, 2008. Lexical vs. grammatical evidentiality in French and Italian. Linguistics 46, 917--947. Travis, Catherine, 2006. Dizque: a Colombian evidentiality strategy. Linguistics 44, 1269--1297. Yamaguchi, Haruhiko, 1994. Echo utterances. In: Asher, R.E. (Ed.), Encyclopedia of Language and Linguistics, vol. 3. Pergamon Press, Oxford, pp. 1084--1085.
Please cite this article in press as: Greco, P., Evidentiality and epistemic modality in witness testimony in the context of Italian criminal trials. Journal of Pragmatics (2017), https://doi.org/10.1016/j.pragma.2017.10.005
PRAGMA-4420; No. of Pages 10
Available online at www.sciencedirect.com
ScienceDirect Journal of Pragmatics xxx (2017) xxx--xxx www.elsevier.com/locate/pragma
Evidentiality and epistemic modality in witness testimony in the context of Italian criminal trials§ Paolo Greco Università di Napoli ‘‘Federico II’’, Via Porta di Massa 1, 80100 Napoli, Italy
Abstract Languages without grammatical evidentiality, such as Italian, often express the category of ‘information source’ using linguistic means that involve interactions between evidentiality and epistemic modality. This paper aims to investigate such interactions in the linguistic strategies adopted by various players in a criminal trial situation. In the context of a trial, both the expression of the source of information and the variation in the degree of the speakers’ commitment to what they are saying are key factors. This study is based on an analysis of one inferential and one reportive strategy in Italian, as they occur in the legal testimonies given in two different Italian trials. We will also discuss the rhetorical effects related to the underlying ambiguities between a purely evidential and an evidential/epistemic reading of the investigated strategies. © 2017 Elsevier B.V. All rights reserved.
Keywords: Forensic linguistics; Evidentiality; Epistemic modality; Commitment; Italian linguistics; Rhetorical effects
1. Introduction This paper aims to investigate the interactions between evidentiality and epistemic modality in the linguistic strategies adopted by the different parties involved in a criminal trial. The variation in the level of speakers’ commitment to the truth of the information they are communicating, as well as the explicit reference to the source of this information (first-hand, reported or inferred) are central aspects of many speech acts produced in criminal trials. Witnesses’ oral testimonies therefore represent a context in which both the source of the information provided, and the relations between speakers’ statements and the truth, play a crucial role.1 Yet, the relevance of the interactions between evidentiality and epistemic modality in the language used by lay witnesses, judges and attorneys in the context of a criminal trial has not yet been pointed out in the literature. Our analysis is based on a corpus of legal testimonies from two Italian criminal trials.2 The witnesses participating in these trials have very different socio-economic backgrounds. Through the analysis of their language, we will also emphasize that the speakers’ differing sociolinguistic level and rhetorical skill play a decisive role in their capacity to manipulate the subtle distinction between the expression of the source of information and the epistemic stance of the speaker. §
I wish to thank two anonymous reviewers for their extremely helpful comments. All errors are my own. E-mail address: [email protected]. To find out more about the characteristics of the pragmatic situation in which speakers, and witnesses in particular, find themselves during a trial, we refer the reader to, for instance, Atkinson and Drew (1979), O’Barr and Conley (1985), Conley and O’Barr (1990), and, for the Italian situation, to Bellucci (2002) and Galatolo (2007). 2 The Italian legal system, while not completely adversarial in nature, is more adversarial than inquisitorial. Witnesses are questioned by the public prosecutor and by the defendant's lawyers, and the system always provides for the possibility of direct examinations and crossexaminations (see also Galatolo, 2007:196--199). 1
https://doi.org/10.1016/j.pragma.2017.10.005 0378-2166/© 2017 Elsevier B.V. All rights reserved.
Please cite this article in press as: Greco, P., Evidentiality and epistemic modality in witness testimony in the context of Italian criminal trials. Journal of Pragmatics (2017), https://doi.org/10.1016/j.pragma.2017.10.005
+ Models
PRAGMA-4420; No. of Pages 10
2
P. Greco / Journal of Pragmatics xxx (2017) xxx--xxx
As a result, witnesses’ testimonies provide fertile ground for the analysis of phenomena related to the interactions between evidentiality and epistemic modality, and, conversely, linguistic investigations on such interactions are crucial to describing the rhetorical strategies employed by the parties involved in a trial. Finally, witnesses’ testimonies also provide fertile ground for the study of the impact of sociolinguistic variation on the linguistic realization of evidentiality. This paper is organized as follows: in section 2, we provide a brief overview of the main theoretical questions regarding the relations between evidentiality and epistemic modality (section 2.1), and then a sketch of the Italian epistemic/ evidential system (section 2.2); in section 3, after a description of the corpus investigated in this paper (section 3.1), we move on to the core of the analysis: the study of the use of two evidentiality strategies (the ‘epistemic future’ and the ‘reportive conditional’) in the trials that we have analyzed (sections 3.2 and 3.3). Finally, in section 4, we highlight the main results of this study and we propose some potential avenues for future research. 2. On evidentiality and epistemic modality 2.1. The relations between evidentiality and epistemic modality: a brief overview In this section, we shall briefly discuss two fundamental preliminary aspects of this paper: the view of evidentiality we shall adopt here, and the notion of speaker's commitment we shall take as our basis; and what, according to the literature on this subject, the relations are between these two categories. According to Aikhenvald (2004), evidentiality is a linguistic category ‘‘whose primary meaning is source of information’’ (Aikhenvald, 2004:3). But what is a ‘linguistic category’3? Does the term ‘evidentiality’ refer to a grammatical category or to a semantic-functional domain that can (but does not necessarily have to) take the form of a grammatical category4? This has been a hotly debated issue over the years, and we shall not enter into this discussion here.5 In this paper, we shall maintain that evidentiality is a functional domain whose linguistic realization can belong to four macro-types: a) b) c) d)
specific and purely grammatical, and sometimes obligatory encoding (that is, ‘grammatical evidentiality’); non-specific and non-obligatory, but grammatical (or semi-grammatical) encoding6; non-specific, non-obligatory, and non-grammatical encoding (that is, ‘lexical evidentiality’). non-specific, non-obligatory, non-grammatical, and non-lexical encoding (for instance, intonational encoding).7
Italian, like the other Romance languages, cannot mark evidentiality through specific and purely grammatical means. In this paper, we shall mainly address class b) phenomena (and their epistemic overtones): we aim to show that direct and cross-examinations of witnesses provide fertile ground for the analysis of the semi-grammatical encoding of evidentiality. This paper will use the definition of epistemic modality given by Nuyts (2006:6), namely ‘‘an indication of the estimation, typically, but not necessarily, by the speaker, of the chances that the state of affairs expressed in the clause applies in the world’’. The question of the relations existing between evidentiality and epistemic modality is one that has been a bone of contention since at least the 1980s.8 Behind the supposed connection between evidentiality and epistemic modality lies the idea that there is to some extent an obvious relationship between the source of our information and our evaluation of its likelihood. The more directly we have acquired a piece of information, the more certain we are meant to be of what we are saying. However, as pointed out by de Haan (1999:87), ‘‘evidentials are in fact a priori unmarked with respect to a commitment to the truth of the speech utterance on the part of the speaker. Evidentials merely assert that there is evidence to back up the speaker's utterance. Any connection between the two [. . .] is secondary in nature. They encode different things (source of information vs. attitude towards that information). Although they are closely enough related to cause overlap in some languages, this overlap is not universal’’.
The notion of ‘linguistic category’ itself is, as is well-known, problematic. We refer the reader to the discussion that took place in the Lingtyp list in January 2016 (available online in The Lingtyp Archives) and to the related papers published in Linguistic Typology (20.2, 2016, pp. 297--462). 4 Aikhenvald (2007) uses ‘information source’ as a technical term to refer to this semantic-functional domain, opposed to ‘evidentiality’, a term the author in question exclusively reserves for the grammatical category. 5 There is a considerable body of literature on this subject. A discussion of the various positions can be found in Greco (2012:55--75), and in Dendale and Van Bogaert (2012). 6 Aikhenvald (2004) refers to such means of encoding the source of information as ‘‘evidentiality strategies’’. 7 See Ifantidou (2001). 8 Detailed accounts of this debate can be found in Plungian (2010), Greco (2012:76--94) and, more recently, Brugmann and Macaulay (2015:205--208). 3
Please cite this article in press as: Greco, P., Evidentiality and epistemic modality in witness testimony in the context of Italian criminal trials. Journal of Pragmatics (2017), https://doi.org/10.1016/j.pragma.2017.10.005
+ Models
PRAGMA-4420; No. of Pages 10
P. Greco / Journal of Pragmatics xxx (2017) xxx--xxx
3
That is why nowadays a theoretical distinction is generally accepted between ‘information source’ and the commitment of the speaker to the truth of what he or she is saying. Yet the relations between these two domains are not always interpreted in the same way.9 The position that will be taken in this paper is that evidentiality and epistemic modality are two distinct but contiguous categories, which in many languages (including Italian) are characterized by overlapping zones with regard to preferential relations. These connections are based on the fact that some sources are generally considered more reliable than others (direct observation as opposed to indirect knowledge, for instance) and that a more reliable source often involves a stronger commitment on the part of the speaker to the truth of the information provided.10 2.2. Evidentiality and epistemic modality in Italian Italian is considered to have at least four semi-grammaticalized forms in which modal and evidential values are intertwined11: the expression si vede che (lit. ‘one sees that’); the verb dovere (‘must’) in the indicative mood; the verb dovere in the conditional mood; and the verb potere (‘can’) in the conditional mood. All these constructions carry inferential values, but they differ in their epistemic meanings: the order in which we presented them ranges from the highest degree of certainty (si vede che, dovere in the indicative mood) to the weakest degree of certainty ( potere in the conditional).12 Moreover, according to Pietrandrea (2005, 2007), Italian has two purely evidential forms: the semi-grammaticalized expression dice che (‘[someone] says that’),13 and the condizionale riportivo or the condizionale citativo (lit. the ‘reportive conditional’ or the ‘quotative conditional’, respectively), so the conditional used as a reportive marker.14 Finally, particularly in spoken language, an evidential extension of the values of the imperfect indicative has also been recognized, specifically in the use of the imperfect with a reportive function (see for instance Squartini, 2001; Calaresu, 2004:197--202; Giacalone Ramat and Topadze, 2007:24--25). Lastly, in some contexts the future tense conveys epistemic values. However, the status of the ‘epistemic’ future is somewhat controversial: Squartini (2008) considers it an evidential (inferential) marker, while Pietrandrea (2005, 2007) maintains that it has a purely epistemic function (pure conjecture). In this paper, we maintain that the epistemic future is an inferential marker. 3. Analysis 3.1. The corpus This study was conducted drawing on a corpus of witnesses’ testimonies in court. In particular, we have analyzed: a) the official transcripts of the direct examinations and cross-examinations of six lay witnesses conducted by the prosecutor and by the defense lawyers in one criminal trial (henceforth trial A) held between 2010 and 2011, for a total of around 42,000 words;
9 See the articulate paradigm and the accurate account proposed by Brugmann and Macaulay (2015). On the relationships between evidentiality and other categories (including epistemic modality), see also Aikhenvald (2015). 10 Konning (2003) provides a model that can account for the relationships between evidentiality and epistemic modality, which is particularly well suited to Romance languages. 11 The Italian modal/evidential system has been discussed in a number of contributions: see for instance Squartini (2001, 2004, 2005, 2008), Pietrandrea (2005, 2007) and Giacalone Ramat and Topadze (2007). See also Cruschina and Remberger (2008) for the domain of reported speech in some Italo-romance (and Hispano- and Daco-romance varieties). 12 This description is consistent with the one provided by Pietrandrea (2005, 2007), though not completely identical. 13 This expression can also occur in the form dice (‘says’) without the complementizer che (‘that’). For further details about this marker, see Lorenzetti (2002), and Calaresu (2004). For some comparable, but more grammaticalized, forms in other Romance varieties, see Travis (2006) and Cruschina and Remberger (2008). 14 Although Pietrandrea (2005, 2007) considers this use of the conditional to be a purely evidential marker, the condizionale riportivo is also known in Italian as the condizionale di riserva or the condizionale di dissociazione (lit. the ‘conditional of doubt’ or the ‘distancing conditional’, respectively) (see for instance Mortara Garavelli, 1995). The labels that have been given to this use of the conditional clearly emphasize the overlap between modality and evidentiality that is often to be found in this use of the conditional. This conditional is frequently (but not always) used by speakers when they want to stress that they are not committed to the truth of the information reported. The alternation between the use of, on the one hand, purely evidential reportive conditionals and, on the other, reportive conditionals with clearly epistemic overtones, is particularly significant in the context of a criminal trial (see section 3.3). For more detail on the reportive conditional in Italian, see, among others, Squartini (2001, 2005:247--250), Calaresu (2004), and Giacalone Ramat and Topadze (2007:21--24).
Please cite this article in press as: Greco, P., Evidentiality and epistemic modality in witness testimony in the context of Italian criminal trials. Journal of Pragmatics (2017), https://doi.org/10.1016/j.pragma.2017.10.005
+ Models
PRAGMA-4420; No. of Pages 10
4
P. Greco / Journal of Pragmatics xxx (2017) xxx--xxx
b) the audio recordings of the direct and cross-examination of one witness in a trial against former Italian Prime Minister Silvio Berlusconi (henceforth trial B).15 As far as trial A is concerned, we have systematically changed all the names and toponyms for privacy reasons. In this trial, attorneys and witnesses clearly come from different sociolinguistic backgrounds. All the witnesses come from deprived socio-economic backgrounds, and most of them have been involved in different kinds of criminal activities (or at least they are closely acquainted with criminals). By contrast, we did not change the names in the trial against Berlusconi. In this case, the identity of the witness is of some significance: he is Antonio Di Pietro, a former public prosecutor and leading Italian political adversary of Silvio Berlusconi who previously served as a government minister.16 As far as the structures that we have analyzed are concerned, while we acknowledge that in Italian both evidentiality and epistemic modality are mainly linguistically encoded by lexical means, we decided to limit the scope of our study exclusively to the grammatical encoding of these values. This is mainly because linguistic exchanges in the context of a trial are regulated by a series of constraints that prevent the use of certain expressions.17 In particular, we will be concerned with some uses of two evidential strategies in which evidential and epistemic values are tightly intertwined: the epistemic future, and the reportive conditional. We will show how more educated speakers (such as prosecutors or attorneys, but also one of the witnesses) exploit the ambiguous boundaries between the evidential and the epistemic reading of these strategies to try to influence the judge's decision. This study is therefore intended to be a qualitative analysis of a selection of excerpts that demonstrate the relevance of the interactions between evidentiality and epistemic modality during a criminal trial, and the ways in which more educated, skilled and trained speakers are able to exploit the ambiguity of two Italian linguistic strategies, the interpretation of which can be purely evidential or evidential and epistemic. However, a brief overview of the quantitative data that we have gathered will allow us to pinpoint some fundamental issues regarding the use of reportive conditionals and epistemic futures in the trials that we have investigated. Epistemic futures occur infrequently in the corpus: we found only 17 instances of this use of the future.18 More generally, it seems that the scope for inferences is very limited in our trials19: we found only 139 occurrences of inferences. Seventy-six of these instances are formulated through lexical means (54.7%), the other 63 are realized through semi-grammaticalized strategies: modal verbs (such as potere or dovere) on 46 occasions (33.1%); epistemic futures, as we have already seen, in the remaining 17 cases (12.2%). Epistemic futures are therefore very rare. Yet, as we shall see in the next section they represent a strategy for attorneys to implicitly communicate what they cannot state explicitly, playing on the boundaries between evidentiality and epistemic modality. Reportive strategies occur much more frequently in our corpus: we found 517 contexts in which they are used. In most cases, the verb encoding the reported information is expressed in the indicative (293 instances, 56.7% of the total). Another common strategy is the direct quotation (152 cases, 29.4% of the total). Reportive conditionals occur less frequently: we only found 46 instances of this strategy (8.9% of the total).20 Yet their use is of particular interest to us. They represent a marked choice, and therefore an ideal strategy to suggest (without any explicit statement) that we are communicating something more than pure reportive evidentiality.21 3.2. Evidentiality and speaker's commitment in court: the epistemic future The reliability of witnesses’ statements has a fundamental role to play in a trial situation. Of course, the lawyers of the defendants and the prosecutor frequently try to demonstrate the contradictions arising in witnesses’ accounts, and/or the weaknesses of the evidence they provide. Yet, they are not allowed to directly question the accounts provided by the witnesses (which by law must be considered truthful unless a clear contradiction arises). Given these constraints,
15 The audio recordings are freely available on the Radio Radicale internet site (www.radioradicale.it), and are published under the Creative Common license. The examination and the cross-examination of the witness took place on July 9th, 2014 and lasted for four hours and fifteen minutes. 16 Di Pietro became very well known in Italy in the 1990s, when he was one of the foremost prosecutors in the Mani pulite (‘Clean Hands’) judicial investigations into political corruption. 17 See for instance Galatolo (2007:196--199). 18 Epistemic futures in our corpus occur mainly in the future perfect form. We found only 4 occurrences of simple futures. 19 This statement holds probably true for the overall context of examination and cross-examination phase of trials. See O’Barr and Conley (1985:665--667) and Galatolo (2007:196--199). 20 Less frequent is the use of the infinitive form of the verb to encode the reported information (26 occurrences, 5% of the total). 21 See also the example provided by Calaresu (2004:193), in which a speaker reports someone else's words, alternating the indicative and the conditional on the basis of his estimation of the likelihood of the truth of the information provided by the original speaker.
Please cite this article in press as: Greco, P., Evidentiality and epistemic modality in witness testimony in the context of Italian criminal trials. Journal of Pragmatics (2017), https://doi.org/10.1016/j.pragma.2017.10.005
+ Models
PRAGMA-4420; No. of Pages 10
P. Greco / Journal of Pragmatics xxx (2017) xxx--xxx
5
the more grammaticalized expressions of evidentiality (and their epistemic overtones) are particularly important, since they allow prosecutors, attorneys and witnesses to capitalize on the interplay between evidentiality and epistemic modality in order to reach their communicative goals. In this framework, the epistemic future represents a useful strategy for speakers to encode the variation in their commitment to the propositional content of what they are saying. (1) Di Pietro: Prob. . . presumibilmente, probabilmente ne avrò parlato con i capigruppo, con lo stesso Formisano, ne avrò parlato con altri parlamentari dell’Italia dei Valori, certamente ne ho parlato, appunto, con i diretti interessati con i pubblici ufficiali che stavano al Ministero, a cui l’ho consegnata. Di Pietro: Prob. . . presumably, probably I spoke22 about it [an audio recorded tape] with the parliamentary group leaders, with Formisano -- I spoke23 about it with other Members of Parliament belonging to Italia dei Valori -- I definitely spoke about it with the people directly concerned, with the public officials at the Ministry, whom I gave it to. The excerpt in (1) shows very clearly how the witness switches from an epistemic future (used with the adverb probabilmente (‘probably’)) to a present indicative (used with the adverb certamente (‘definitely’)), when he moves on from general recollections to the main point of his testimony.24 The passage in (2) shows how lawyers can use the epistemic future to surreptitiously challenge a witness's testimony without explicitly calling his or her claims into question. (2) Avvocato: Allora io voglio dire, lei sta lottando disperatamente, ce l’ha detto prima, per tenere i voti tutti i giorni. L’avrà chiesto a Caforio ‘‘ma ti ha detto chi sono le persone, soprattutto della mia parte politica, pronti a fare questo governo di larga coalizione?’’?. Lawyer: So, I mean, you are desperately struggling, as you told us before, every day to keep your votes. You must have asked25 Caforio, ‘‘But did he tell you who the people are, especially from my party, who are ready to form a broad coalition government?’’. The use of the epistemic future (in the perfect) with a peculiar prosodic pattern (an interrogative intonation and a strong stress on the auxiliary's tonic vowel), allows the lawyer to specify that the alleged behavior of the witness was somewhat strange, without explicitly questioning the truth of his claims. This type of context allows the speaker to set up a framework in which the actions described by the predicate in the future perfect are considered logical and obvious, to such an extent indeed that any other behavior sounds strange or even suspect. As a result -- and this is clearly the implicit aim of the strategy being pursued by the lawyer -- the account given by the witness is considered unreliable. 3.3. Evidentiality and speaker's commitment in court: the reportive conditional In section 3.1 we have shown that in our corpus reportive evidentiality is generally encoded by the simple indicative, and that the conditional occurs in less than 10% of the reportive contexts. However, the reportive conditional has a very different frequency of occurrence in the two trials that we have analyzed. In trial A, out of 291 reportive contexts, 162 (55.7%) show an indicative and only 13 (4.5%) are characterized by a conditional. Therefore, despite the significant number of reportive contexts, in trial A the reportive conditional seldom occurs (and never in witness testimony). In trial B, meanwhile, we found 33 occurrences of reportive conditionals and 131 indicatives in a total of 226 reportive contexts. While the percentage of use of the indicative is similar in the two trials (55.7% in trial A, 58% in trial B), in trial B there is a significant rise in the frequency of the reportive conditional (14.6% of cases, compared to 4.5% in trial A). The main difference between the two trials lies in the fact that in trial A, reportive conditionals only occur in the language of lawyers, prosecutors and judges, while in trial B the witness employs this linguistic strategy too. The reportive conditional is therefore a marked stylistic choice in both trials, but in trial B, where both the attorneys and the witness are trained speakers coming from an educated background, its use is more frequent. The functional values of the reportive conditionals can be very different. On certain occasions, the evidential value of the reportive conditional is clearly prominent, and the possible epistemic overtones, if present, are secondary in nature. The clearest cases are perhaps those in which the conditional is used by judges as a sign of impartiality (although they also use the indicative in similar contexts):
22
The verb parlare (‘to speak’) is in the future perfect. See footnote 22. 24 This witness's testimony is in fact based on his claim that he had received an audio tape allegedly featuring data that could be relevant to the trial. According to the witness, he never listened to the tape, which he immediately passed on to some public officials so that they could check whether the material on the tape had criminal repercussions or not. This tape has now gone missing. 25 The verb chiedere (‘to ask’) is in the future perfect. The intonation of the utterance is interrogative. 23
Please cite this article in press as: Greco, P., Evidentiality and epistemic modality in witness testimony in the context of Italian criminal trials. Journal of Pragmatics (2017), https://doi.org/10.1016/j.pragma.2017.10.005
+ Models
PRAGMA-4420; No. of Pages 10
6
P. Greco / Journal of Pragmatics xxx (2017) xxx--xxx
(3) Presidente: Allora la difesa dice: ‘‘Come mai lei non ha mai fatto cenno nel corso degli interrogatori resi precedentemente a questa circostanza e cioè che Maria avrebbe anche visto, avrebbe anche incrociato, c’è questo riferimento anche alla presenza di suo figlio’’. Judge: Then the defense said: ‘‘Why in the previous hearings didn’t you mention this fact, namely that Maria also saw,26 that she came across27 [these people], and there is also this reference to the presence of her son’’. Conditionals are also used when an attorney rephrases the words of a witness who is contradicting himself/herself. (4) Presidente: Qui dice ‘‘la mia convivente mi riferì che a sparare all’impazzata contro la folla nella zona di Posillipo erano stati Francesco e Mario quest’ultima mi aggiunse che aveva appreso tale notizia in famiglia e direttamente dagli interessati’’. Avvocato A: Quindi non l’aveva visti! [. . .] Avvocato B: Sul verbale di oggi [. . .] ha detto che la convivente gli avrebbe riferito che li avrebbe visti quando stava a Posillipo. Questo è il verbale di oggi.28 Judge: Here you say, ‘‘My partner told me that Francesco and Mario wildly shot into the crowd in the Posillipo area. She added that she found out about this from her family and directly from the people involved’’. Lawyer A: So she didn’t see them! [. . .] Lawyer B: In today's official transcript [. . .], he said that his partner told him29 that she saw30 them when she was in the Posillipo area. This is today's official transcript. In the example above, the conditional is used not only to indicate the second-hand nature of the information provided, but also to suggest that the witness is unreliable. The contradictions between the differing versions of the same story given by the witness on different occasions are somewhat emphasized by the use of the conditional in the lawyer's account. In other contexts, the epistemic overtones are even more prominent. As we have seen in the previous section, in certain contexts the use of the epistemic future allows attorneys to challenge the words of the witnesses without resorting to more explicit means, for instance lexical items, that would probably lead to a formal request by the court to rephrase the passage in question.31 This aspect is particularly clear in the use of the reportive conditional in trial B. In this trial, the conditional is widely used by lawyers when they want to surreptitiously discredit the information provided by the witness, or at least when they wish to emphasize that this information is based on someone's account of events and that there is no hard evidence for what has been said.32 (5) Avvocato: Abbiamo appreso che Caforio sarebbe stato avvicinato da Tomassini, e di questo abbiamo già parlato io non intendo più riproporre il problema, e da De Gregorio. Lei stamattina ha detto al pubblico ministero che Caforio le avrebbe detto che De Gregorio avrebbe rappresentato che dietro di lui vi era Berlusconi, che i soldi sarebbero stati dati, anzi ha parlato di Forza Italia per la verità, e di Berlusconi. Questo ha detto al Pubblico Ministero stamattina. Lawyer: We found out that Caforio was contacted33 by Tomassini, we have already talked about this and I will not raise the issue again, and by De Gregorio. This morning, you told the public prosecutor that Caforio told34 you that De Gregorio said35 that Berlusconi was behind him and that the money had been paid,36 and indeed you talked about Forza Italia,37 and about Berlusconi. This is what you said to the public prosecutor this morning.
26
The verb vedere (‘to see’) is in the perfect conditional. The verb incrociare (‘to come across’) is in the perfect conditional. 28 The conditional was not used, of course, by the witness in the original account of the facts. 29 The verb riferire (‘to tell’) is in the perfect conditional. 30 See footnote 26. 31 On the strategies adopted by attorneys in courtroom cross-examinations, see also Drew (1992). 32 Indirect evidence (specifically, provided by a de relato, hearsay, witness) can be accepted in a criminal trial under certain circumstances, in accordance with Article 195 of the Italian Codice di Procedura Penale (Criminal Procedure Code). 33 The predicate essere avvicinato (in this context, ‘to be contacted’) is in the perfect conditional. 34 The verb dire (in this context, ‘to tell’) is in the perfect conditional. 35 The verb rappresentare (in this context, ‘to say’) is in the perfect conditional. 36 The verb dare (‘to give’) is in the passive perfect conditional. 37 An Italian political party, led by Silvio Berlusconi. 27
Please cite this article in press as: Greco, P., Evidentiality and epistemic modality in witness testimony in the context of Italian criminal trials. Journal of Pragmatics (2017), https://doi.org/10.1016/j.pragma.2017.10.005
+ Models
PRAGMA-4420; No. of Pages 10
P. Greco / Journal of Pragmatics xxx (2017) xxx--xxx
7
(6) Avvocato: Secondo le. . . secondo la sua narrazione di oggi, le avrebbero narrato solo una parte e la parte vera sarebbe contenuta nella registrazione Lawyer: According to. . . according to the account you have given today, they told38 you only part of the story, and the true part is given39 in the recording. It is clear from the examples above that the reportive conditional can be used to question the account of a given fact provided by a witness without explicitly challenging his or her words. The distancing effect of the conditional allows lawyers to treat the witness's testimony as an uncertain account of events, or, as the attorney says in (6), as a narrazione (a word which can mean ‘account’ but which is often used in Italian as a synonym for ‘tale’ or ‘story’). In the excerpts presented above, lawyers take advantage of the fact that the use of the conditional, as we have seen in example (3), may also convey a purely evidential meaning,40 while the modal meaning (non-commitment, uncertainty), although often associated with this, may be seen as an extension of it41: the modal value of uncertainty is only implied, and cannot be explicitly attributed to the lawyers. However, it is clear to any Italian speaker that the use of the conditional in the contexts presented in (5) and (6), and particularly in the latter case, strongly fosters an interpretation of the information provided as being something not far removed from a lie -- or at least an imaginative story. We have already noted that the reportive conditional never occurs in the witnesses’ testimonies, with one exception: that of Antonio Di Pietro. (7) Di Pietro: Lui dice, ha riferito, si riferiva al fatto che Lavitola avrebbe ricevuto, avrebbe detto che anche lui, Formisano avrebbe ricevuto 50000 euro da Berlusconi. Di Pietro: He says, he told me, that he was referring to the fact that Lavitola received,42 [and that he also] said43 that also Formisano received44 s50,000 from Berlusconi. (8) Di Pietro: A me ha sconvolto non la vicenda di Caforio [. . .] mi ha sconvolto, e ribadisco che mi sconvolge ancora adesso molto, che chi ha accompagnato il Senatore Caforio, e cioè Formisano, né in quel momento, né davanti a me, né davanti alla pubblica accusa ha ritenuto di dire un fatto importantissimo e cioè, ribadisco, che Lavitola è andato a trovarlo nell’Ufficio del Senatore, di Formisano, poi sarà vero non sarà vero, lo diranno loro, e davanti a Formisano avrebbe telefonato a Marinella Brambilla dicendo di farsi ritelefonare da Berlusconi, e subito dopo Berlusconi l’avrebbe ritelefonato, a dire del Formisano, io non lo so, non c’ero lì, sono de relato, sentirete Formisano, [Berlusconi] avrebbe telefonato e gli avrebbe detto, a Lavitola, di dire a Formisano che c’era davanti a lui [Lavitola] «We, promettigli qualsiasi cosa che siamo disponibili a tutto».45 Di Pietro: I wasn’t shocked by what Caforio did [. . .] I was shocked -- and I repeat that I’m still very shocked about this -- by the fact that Formisano, who accompanied Senator Caforio, decided to hide on that occasion, as well as when he was heard by the public prosecutor, a very important fact, namely, I repeat, that Lavitola met him in the office of the Senator, of Formisano. Now, they will say if this is true or not,46 and in the presence of Formisano he [Caforio] phoned47 Marinella Brambilla, asking to be called back by Berlusconi, and immediately afterwards Berlusconi rang back48; this is Formisano's version -- I don’t know, I wasn’t there, I am just saying what I heard; you will hear Formisano's account, he [Berlusconi] phoned49 and told50 him, Lavitola, to tell Formisano, who was standing right in front of him [Lavitola]: «Hey, promise him anything -- we’re willing to do anything». As clearly demonstrated by examples (7) and (8), in Di Pietro's account of events, reportive conditionals are relatively frequent (15 occurrences in 118 possible contexts, meaning 12.7% of the instances). He is a de relato witness, he knows
38
The verb narrare (‘to tell’) is in the perfect conditional. The verb contenere (‘to contain’) is in the perfect conditional. 40 See also the final statement of the attorney in (5): ‘‘Questo ha detto al Pubblico Ministero stamattina’’ (‘‘This is what you said to the public prosecutor this morning’’). With this statement, the attorney explicitly qualifies his own previous affirmations as reported. 41 This position is held for instance by Giacalone Ramat and Topadze (2007:22). 42 The verb ricevere (‘to receive’) is in the perfect conditional. 43 The verb dire (‘to say’) is in the perfect conditional. 44 See footnote 42. 45 On the use of direct reported speech in legal testimonies see Galatolo (2007). We cannot discuss the issue here, but Di Pietro's use of direct reported speech is also particularly interesting. 46 The expression used by Di Pietro could be literally translated as ‘it will be true, it will not be true’ with an epistemic future. 47 The verb telefonare (‘to phone’) is in the perfect conditional. 48 The verb ritelefonare (‘to ring back’) is in the perfect conditional. 49 See footnote 47. 50 See footnote 34. 39
Please cite this article in press as: Greco, P., Evidentiality and epistemic modality in witness testimony in the context of Italian criminal trials. Journal of Pragmatics (2017), https://doi.org/10.1016/j.pragma.2017.10.005
+ Models
PRAGMA-4420; No. of Pages 10
8
P. Greco / Journal of Pragmatics xxx (2017) xxx--xxx
what this means, and he knows that it is very likely that his words will carry no weight unless the individuals he is talking about confirm them. Therefore, he uses the conditional as a sign of impartiality. On many occasions, Di Pietro takes great care to use the present indicative for all the events he was party to, and the conditional for all the events that were reported to him. In this way, he clearly establishes a difference between what he wants to present as fact (namely that he received a recorded tape which, he was told, contained recordings of an attempt at bribery) and what he presents as second-hand information (namely accounts of this attempted bribery that were provided to him by some of the parties involved). A corollary of this strategy is, of course, that the only part of his account that could be called into question is the hearsay evidence, and not the facts he had a direct knowledge of, as the defendant's attorneys were trying to do. In the next excerpt, it is interesting to note that Di Pietro uses a conditional to express a piece of information he provided himself. In this case, the defendant's attorney is surreptitiously questioning Di Pietro's account of an event by using the conditional. In the passage presented in (9), Di Pietro adopts the shrewd strategy of using and repeating the conditional used by the lawyer (with an echoing effect),51 accepting the evidential value of the form (the attorney was referring to Di Pietro's claims, and he was therefore quoting second-hand information) but clearly rejecting the modal value of the verb form. He adds a precise temporal setting which strongly militates against an interpretation of non-commitment/uncertainty for the use of the conditional here. This strategy allows Di Pietro to reveal the lawyer's ruse: (9) Avvocato: [in a text written by Di Pietro and submitted to the court] Lei riferisce con un corsivo quello che le avrebbe narrato il Caforio. Di Pietro -- Che mi avrebbe narrato il 4 marzo 2013. Lawyer: [in a text written by Di Pietro and submitted to the court] You put what Caforio told52 you in italics. Di Pietro: What he told53 me on March 4, 2013. In these last examples, we have seen how evidentiality and epistemic modality interact in the actual uses of the reportive conditional in the examination and in the cross-examination of a witness in a criminal trial in which both the attorneys and the witness are trained speakers. In the light of the discussion above, we maintain that the core value of the reportive conditional is purely reportive (see example (3)). However, since the epistemic overtone of non-commitment is often associated with the reportive conditional, trained speakers can exploit this preferential relationship between the reportive conditional and a low degree of speaker's commitment, to surreptitiously question the truth and the validity of certain statements without overtly questioning them (above all in contexts in which such questioning would be inappropriate, or, as in the case of an attorney talking about the testimony given by a witness, simply not allowed). 4. Discussion and conclusions This study aimed to investigate the rhetorical functions of the variations in the interactions between evidentiality and epistemic modality in the linguistic strategies adopted by attorneys, prosecutors and witnesses in two different Italian criminal trials. We also aimed to highlight the impact of different sociolinguistic backgrounds on the speakers’ abilities to exploit the possibilities offered by these interactions in two evidential strategies, namely the epistemic future and the reportive conditional. The particular characteristics of the codified interactions between attorneys and witnesses provide fertile ground for the study of both the relationships between evidentiality and epistemic modality in Italian, and the rhetorical strategies adopted by the different parties involved in a criminal trial in some specific contexts. In criminal trials, the law forbids certain kinds of statement. For instance, attorneys and prosecutors are not allowed to directly question the words of a witness in the absence of clear contradictions. However, as we saw in sections 3.2 and 3.3, they often challenge the information provided by witnesses by using linguistic strategies such as the reportive conditional or the epistemic future. This paper represents one of the first explorations of the use of evidential markers, and of the rhetorical strategies exploiting the ambiguous boundaries between evidentiality and epistemic modality, as they occur in criminal trials. The analytic results are language-specific, but we hope to have shown the potentialities of the study of evidential markers and of the relationships between evidentiality and epistemic modality in the context of criminal trials. We can in fact assume that aspects related to the source of information, the speaker's commitment, and the semantic and pragmatic continuities
51 52 53
On echo utterances, and on their pragmatic functions, see Yamaguchi (1994). The verb narrare (‘to tell’, ‘to narrate’) is in the perfect conditional. See footnote 52.
Please cite this article in press as: Greco, P., Evidentiality and epistemic modality in witness testimony in the context of Italian criminal trials. Journal of Pragmatics (2017), https://doi.org/10.1016/j.pragma.2017.10.005
+ Models
PRAGMA-4420; No. of Pages 10
P. Greco / Journal of Pragmatics xxx (2017) xxx--xxx
9
between evidentiality and epistemic modality are crucial in criminal trials in many countries (particularly those with an adversarial system). As far as the analytical results are concerned, in this paper we have described how trained public speakers (such as attorneys and prosecutors) capitalize on the mutual relationship between the evidential and epistemic values of verbal forms such as the future and the conditional. For instance, in their use of reportive conditionals, in many contexts they seem to adeptly make use of the subtle boundaries between what is simply reported as fact and what is reported but implicitly considered uncertain. This double interpretative possibility of the Italian reportive conditional, which is strictly related to a more general typological characteristic of reportive evidentials in many languages of the world,54 is at the basis of the different interpretations of the values of the Italian reportive conditional that are to be found in the literature.55 In the light of the discussion presented in this paper, we maintain that the core function of the Italian reportive conditional is that of producing a distancing effect. This effect can be of a purely evidential nature, or, more often, of an evidential nature with epistemic overtones. Therefore, the reportive conditional is an evidential marker which is often associated with an epistemic overtone of non-commitment/ uncertainty.56 The structures conveying inferential values occur less frequently in our corpus. However, we have observed a particular use of the Italian epistemic future perfect (example (2)), that we would like to further discuss here. The Italian future perfect, in its epistemic interpretation, generally conveys conjectures or inferences (see Bertinetto, 1995:123). Yet, in some contexts, the use of the future perfect with a peculiar intonational pattern (interrogative, with a strong accent on the auxiliary) allows the speaker to set up a frame in which every deviation from the state of affairs described by the speaker should be regarded as very strange (or even suspicious, as in the excerpt discussed in example (2)). The presupposition that lies behind these statements is that on the basis of our general knowledge of the world we can assume that it is almost obvious that the state of affairs described in the proposition has to be true. As a result, these future perfect constructions carry strong inferences, and convey the speaker's total conviction that the information they contain is true. We suggest calling this use of the epistemic future (which, to the best of our knowledge, has not yet been described in the literature) the ‘polemical future’,57 because its rhetorical effect is to argue that any state of affairs other than the one described in the sentence would be very strange. Through this study, we also intended to address a more general question, namely that of the fundamental role played by discourse traditions in the analysis of evidential and modal markers58: different contexts can lead to linguistic strategies being used more (or less) consciously or carefully. Therefore, we believe that the use and functions of evidential and modal markers in a given language should be studied in their context.59 In this paper, we have analyzed their occurrence in the context of a trial, but studies conducted in other situations from everyday life will undoubtedly point to different uses of these markers. The choice of which linguistic means to use to encode evidential and/or epistemic values is also certainly sensitive to sociolinguistic variation: people coming from different sociolinguistic backgrounds show different abilities in the use of inferential and reportive markers. Speakers coming from more educated backgrounds are able to exploit the ambiguities related to the interactions between evidentiality and epistemic modality, in order to achieve different communicative goals. Thus far, these factors have generally been neglected in studies on the evidentiality/epistemic modality interface. We are convinced that this research path is promising, and that further studies analyzing evidential and modal markers in the light of sociolinguistic and discourse tradition factors will shed light on aspects of the use of these markers that have not yet been recognized. They will also help us to reach a better understanding of how we exploit the interactions between evidentiality and epistemic modality, by playing on the ambiguities of the markers encoding these values. References Aikhenvald, Alexandra Y., 2004. Evidentiality. Oxford University Press, Oxford. Aikhenvald, Alexandra Y., 2007. Information source and evidentiality: what can we conclude? Ital. J. Linguist. 19, 209--227. Aikhenvald, Alexandra, 2015. Evidentials: their links with other grammatical categories. Linguist. Typol. 19 (2), 239--277.
Aikhenvald (2004:180) highlights that in certain evidential systems the speaker ‘‘may choose to employ the reported evidential for two reasons. Firstly, to show his or her objectivity; that the speaker was not the eyewitness to an event and knows about it from someone else. Secondly, as means of ‘shifting’ responsibility for the information and relating facts considered unreliable’’. 55 See footnote 14. 56 For a discussion of the evidential and epistemic functions of conditionals in a typological perspective, see Aikhenvald (2004:105--109). 57 The ‘polemical future’ shares some characteristics with the so-called ‘‘valeur de renforcement’’ (Rossari et al., 2017) of the Italian epistemic future. However, these values only partially overlap, in that only some of the cases of epistemic futures indicated by Rossari et al. (2017) can be regarded as (at least partially) fitting the description we have given of the ‘polemical future’. 58 On the notion of discourse tradition see Koch (1997). 59 Grossmann and Tutin (2010) and Cornillie (2010) have previously emphasized the importance of such an approach. 54
Please cite this article in press as: Greco, P., Evidentiality and epistemic modality in witness testimony in the context of Italian criminal trials. Journal of Pragmatics (2017), https://doi.org/10.1016/j.pragma.2017.10.005
+ Models
PRAGMA-4420; No. of Pages 10
10
P. Greco / Journal of Pragmatics xxx (2017) xxx--xxx
Atkinson, J. Maxwell, Drew, Paul, 1979. Order in Court. The Macmillian Press, London. Bellucci, Patrizia, 2002. A onor del vero. Fondamenti di linguistica giudiziaria. UTET, Torino. Bertinetto, Pier Marco, 1995. Il verbo. In: Renzi, L., Salvi, G. (Eds.), Grande Grammatica Italiana di Consultazione, vol. II. Il Mulino, Bologna. pp. 13--161 (quoted from the second edition, 2001). Brugmann, Claudia M., Macaulay, Monica, 2015. Characterizing evidentiality. Linguist. Typol. 19 (2), 201--237. Calaresu, Emilia, 2004. Testuali parole: la dimensione pragmatica e testuale del discorso riportato. Franco Angeli, Milano. Conley, John M., O’Barr, William M., 1990. Rules versus Relationships. The Ethnography of Legal Discourse. The University of Chicago Press, Chicago. Cornillie, Bert, 2010. An interactional approach to evidential and epistemic adverbs in Spanish conversation. In: Diewald, G., Smirnova, E. (Eds.), Linguistic Realization of Evidentiality in European Languages. Mouton de Gruyter, Berlin/New York, pp. 309--330. Cruschina, Silvio, Remberger, Eva-Maria, 2008. Hearsay and reported speech: evidentiality in Romance. In: Benincà, Paola, Damonte, Federico, Penello, Nicoletta (Eds.), Selected Proceedings of the 34th Incontro di Grammatica Generativa. Special Issue of the Rivista di Grammatica Generativa, vol. 33. pp. 99--120, http://arcaold.unive.it/handle/10278/2504. de Haan, Ferdinand, 1999. Evidentiality and epistemic modality: setting boundaries. Southwest J. Linguist. 18, 83--101. Dendale, Patrick, Van Bogaert, Julie, 2012. Réflexions sur les critères de définition et les problèmes d’identification des marqueurs évidentiels en français. Lang. Fr. 173, 13--29. Drew, Paul, 1992. Contested evidence in courtroom cross-examination: the case of a trial for rape. In: Drew, P., Heritage, J. (Eds.), Talk at Work. Cambridge University Press, Cambridge. Galatolo, Renata, 2007. Active voicing in court. In: Holt, E., Clift, R. (Eds.), Reporting Talk. Reporting Speech in Interaction. Cambridge University Press, Cambridge, pp. 195--220. Giacalone Ramat, Anna, Topadze, Manana, 2007. The coding of evidentiality: a comparative look at Georgian and Italian. Ital. J. Linguist. 19, 7-38. Greco, Paolo, 2012. Evidenzialità. Storia, teoria e tipologia. Aracne, Roma. Grossmann, Francis, Tutin, Agnès, 2010. Evidential markers in French scientific writing: the case of the French verb ‘‘voir’’. In: Smirnova, E., Diewald, G. (Eds.), Linguistic Realization of Evidentiality in European Languages. Mouton de Gruyter, Berlin/New York, pp. 279--307. Ifantidou, Elly, 2001. Evidentials and Relevance. Benjamins, Amsterdam. Koch, Peter, 1997. Diskurstraditionen: zu ihrem sprachtheoretischen Status und ihrer Dynamik. In: Frank, B., Haye, Th., Tophinke, D. (Eds.), Gattungen mittelalterlicher Schriftlichkeit. Narr, Tübingen, pp. 43--79. Konning, Hans, 2003. Modalité et évidentialité. In: Birkelund, M., Boysem, G., Kjaersgaard, P.S. (Eds.), Aspects de la Modalité. Niemeyer, Tübingen, pp. 131--151. Lorenzetti, Luca, 2002. Sulla grammaticalizzazione di ‘dice’ nell’italiano parlato. In: Heinemann, S., Bernhard, G., Kattenbusch, D. (Eds.), Roma et Romania. Festschrift für Gerhard Ernst zum 65. Geburststag. Niemeyer, Tübingen, pp. 211--221. Mortara Garavelli, Bice, 1995. Il discorso riportato. In: Renzi, L., Salvi, G., Cardinaletti, A. (Eds.), Grande Grammatica Italiana di Consultazione, vol. III. Il Mulino, Bologna. pp. 429--470 (quoted from the second edition, 2001). Nuyts, Jan, 2006. Modality: overview and linguistic issues. In: Frawley, W. (Ed.), The Expression of Modality. Mouton de Gruyter, Berlin/New York, pp. 1--26. O’Barr, William B., Conley, John M., 1985. Litigant satisfaction versus legal adequacy in small claims court narratives. Law Soc. Rev. 19, 661--701. Pietrandrea, Paola, 2005. Epistemic Modality. Functional Properties and the Italian System. Benjamins, Amsterdam. Pietrandrea, Paola, 2007. The grammatical nature of some epistemic-evidential adverbs in spoken Italian. Ital. J. Linguist. 19, 39--63. Plungian, Vladimir A., 2010. Types of verbal evidentiality marking: an overview. In: Diewald, G., Smirnova, E. (Eds.), Linguistic Realization of Evidentiality in European Languages. Mouton de Gruyter, Berlin/New York, pp. 15--58. Rossari, C., Ricci, C., Siminiciuc, E., 2017. Les valeurs rhétoriques du futur en français, italien et roumain. In: Baranzini, L. (Ed.), Le futur dans les langues romanes. Peter Lang, Berne, pp. 49--77. Squartini, Mario, 2001. The internal structure of evidentiality in Romance. Stud. Lang. 25, 297--334. Squartini, Mario, 2004. Disentangling evidentiality and epistemic modality in Romance. Lingua 114, 873--895. Squartini, Mario, 2005. L’evidenzialità in rumeno e nelle altre lingue romanze. Z. Romanische Philol. 121, 246--268. Squartini, Mario, 2008. Lexical vs. grammatical evidentiality in French and Italian. Linguistics 46, 917--947. Travis, Catherine, 2006. Dizque: a Colombian evidentiality strategy. Linguistics 44, 1269--1297. Yamaguchi, Haruhiko, 1994. Echo utterances. In: Asher, R.E. (Ed.), Encyclopedia of Language and Linguistics, vol. 3. Pergamon Press, Oxford, pp. 1084--1085.
Please cite this article in press as: Greco, P., Evidentiality and epistemic modality in witness testimony in the context of Italian criminal trials. Journal of Pragmatics (2017), https://doi.org/10.1016/j.pragma.2017.10.005