Abstract
This paper sets up to show how irony and reservations are explicitated in online media discourse, comparing their realizations in French and Hebrew online op-eds in leading journals. A corpus-based qualitative and quantitative analysis relies on two sets of big corpora for each language. The pragmatic analysis distinguishes between explicitating self- and other’s presumed ironic intents, the target of irony, its locus and overall speaker’s meanings. The findings indicate that the French data-set uses the verb ironiser, which has no comparable equivalent in Hebrew. More puzzling are the similarities between the two data-sets: both in French and in Hebrew journalists choose to explicitate irony and reservations, and they do so using similar discursive patterns. Conflicting forces are at play: interpretation paths are opened by irony, and are then narrowed down by the journalist’s interpretations. The results are interpreted in terms of informativeness, accountability and commitment to speaker’s meaning.
1 Introduction*
This paper sets up to show how irony and reservations are explicitated in online media discourse and to compare culture-specific preferences in French and Israeli Hebrew in this discourse type.1 The analysis focuses on op-eds in online leading journals. It starts with the premise that verbal irony has two defining features: it is inherently indirect, and it necessarily conveys implied criticism. Hence, the assignment of ironic intentions to the speaker involves a complex interpretation process, which requires that the interpreter rely on contextual cues to detect reasons to search for plausible speaker’s meanings that diverge from the utterance meaning, and on contextual clues which direct her in constructing alternative speaker’s meanings (Dascal and Weizman, 1987; Weizman and Dascal, 1991). Previous studies of indirectness in journalistic discourse – op-eds and commenting – indicated that ironic utterances may be embedded in textual environments which explicitate certain aspects of the irony or construct redundancy in textual markers which support the ironic interpretation (see section 2.2 below). In such cases, the range of plausible constructed speakers’ meanings assigned in the interpretation process is narrowed down. The discourse patterns used for explicitation and redundancy have hardly been dealt with systematically in the literature (but see Barbe, 1995; Weizman, 2011, 2020), and will be at the focus of the analysis in this paper, with special emphasis on comparing their realizations in French and Hebrew online op-eds.
Contrastive studies require that a comparison be made against the background of tertia comparationis, i.e. one or several shared components that encompass and transcend language-specific features. This common denominator, “a background of sameness” (Johansson, 2007: 39), viewed in semantic or functional terms (Chesterman, 1998; Egan, 2013), is determined by a certain degree of subjective perception and context-bound assessments of salience of the matching features (Chesterman, 1998). When the comparison is not based on complete matching, which is usually the case, it is determined based on background framing (Krzeszowski, 1990: 231). Hence, similarity, which is the basis for tertium comparationis, is “janus-faced. It simultaneously refers to a relation-in-the-world and a perception-in-the-mind” (Chesterman, 1989: 16).
The present contrastive analysis is based on functional comparability, i.e. the assumption that irony, as well as explicitation and redundancy in both languages, are comparable functions; it was hypothesized, however, that their discursive realizations may be different, and that they may fulfill different functions. The findings indicate that the discursive explicitations of irony and reservations in online op-eds in leading journals in French and Hebrew present some similarities and differences, but the similarities are more intriguing. I will argue that these similarities may be accounted for in terms of the growing tendency of journalists to avoid interpretive risks, and that this tendency involves an increase in the journalist’s accountability and commitment to speakers’ meanings (section 6).
Function-based contrastive analyses are not scarce. They include, to mention just a few, the large body of studies on speech acts, for example, the CCSARP (cross-cultural speech acts realization patterns) project studying requests and apologies (Blum-Kulka et al., 1989), Ogiermann’s focus on apologies in relation to positive and negative politeness (2009) and on perspective (2020), Kampf’s (2021) investigation of moral and ethical similarities and differences between Hebrew and U.S. American calls to condemn in news interviews and House and Kadar’s (2021) study of war apologies through the lens of speech act theory.
From a different angle, Weizman (2011, 2020) proposes a contrastive analysis of culture-specific preferences for redundancy and explicitation in conveying indirect meanings, including requestive hints, reservations and irony (Weizman, 2007, 2011, 2020), whereby the functional tertium comparationis is the perception of quantity of information (section 3). This study investigates the discursive explicitation and redundancy used to support the interpretation of irony in media discourse in French and Hebrew online op-eds.
The paper is structured as follows. The conceptual framework will first be presented (section 2). The present study will then be situated in the framework of an ongoing research (section 3), followed by methodological remarks (section 4). Patterns of discursive realizations of explicitation and redundancy marking irony and reservations in both data-sets will then be analyzed (section 5), highlighting similarities (5.1.) and differences (5.2). The findings will then be discussed (section 6).
2 Conceptual Framework
2.1 Irony as Non-attributive Meta-representation
Within the framework of relevance theory irony has been described as an echoic mention (Sperber and Wilson, 1981; Wilson and Sperber, 1992), or as a non-attributive meta-representation (Wilson, 2012). Utterances are considered as representations in as much as they have “a propositional form which is true of some actual or conceivable state of affairs” (Sperber and Wilson, 1986: 227). Thus, for example, the propositional form of an assertion represents some state of affairs in the real world, whereas the propositional form of a request represents a desirable state of affairs (ibid.). Meta-representation has been defined as “a representation of a representation: a higher-order representation with a lower order representation embedded within it” (Wilson, 2012: 230). Lower-order representations include public representations (e.g. utterances), mental representations (e.g. thoughts) and abstract representations (e.g. sentences, propositions) (Sperber, 2000; Wilson, 2012). In order to encompass all three types in a relevance-based theoretical framework, an analysis in terms of resemblance, rather than of identity between higher-order and lower-order representations, was postulated (Wilson, 2012; Weizman and Kohn, forthcoming).
Meta-representations may be attributive or non-attributive. Typical cases of attributive meta-representations include direct and indirect quotations, whereby attribution is textually marked by combinations of quotatives, some information about the quoted source and quotation marks. In non-attributive meta-representations, on the other hand, the connection between higher-order and lower-order representation is mostly unmarked. This is the case with mentions, when they represent utterances (e.g. Shut up is rude) or a proposition (e.g. Roses and daisies are flowers entails that roses are flowers) (Wilson, 2012: 232). Non-attributive meta-representations include echoic mentions, which, in addition to the attribution, convey the speakers’ attitude, be it agreement or disagreement, support or challenge (ibid.: 249; for cases of agreement and disagreement through resemblance, see Weizman and Kohn, forthcoming). A typical case of non-attributive meta-representation is that of ironic mentions which echo a prior saying, thought, belief or interpretation and necessarily convey criticism towards the first-order representation (Sperber and Wilson, 1981; Wilson and Sperber, 1992; Wilson, 2012).
In attributive meta-representations such as direct and indirect speech, attribution is textually marked by a quotative verb, some information about the quoted source, and possibly by the use of quotation marks. In non-attributive, ironic mentions, meta-representation is typically unmarked. Thus, in utterance (1), ‘The cabinet meeting was conducted quietly and calmly’, the proposition quietly and calmly may be interpreted as an ironic mention which echoes the belief of the chair of that meeting. The assignment of an ironic speaker’s meaning is entirely dependent upon contextual assumptions about the speaker, the meeting, conventions underlying cabinet meetings, and more. The ironic interpretation, however, may be supported by the use of partial quotation marks, i.e. quotation marks in which only part of the mentioned utterance or thought is embedded. This is the case, for example, in (2), ‘The cabinet meeting was conducted “quietly and calmly”’, where the quotation marks are not called for by the indirect quotations. The latter may be interpreted as conveying an ironic criticism towards the proposition in quotation marks, its content, its form or both (Weizman, 1984, 2008, 2011). As will be explained in the next section, I consider the use of indirect speech with partial quotation marks as redundant in the sense defined below.
2.2 Redundancy and Explicitation
Following Weizman (2011, 296), redundancy is defined as follows:
If a discourse pattern P1 is used to achieve speaker’s meaning S, and discourse pattern P2 may equally achieve speaker’s meaning S (notwithstanding slight nuances), then the use of both P1 and P2 in a given textual environment will be conceived of as redundant. (ibid.; see also Weizman, 2017)
This reading of the term pertains to co-textual redundancy in discursive strategies.
If, for example, the verb claim conveys reservations (Dascal and Weizman, 1997; Weizman, 2011, see section 3), and partial quotations embedded in indirect speech convey critical reservations, then any combination of both patterns is considered redundant (Weizman, 2011, 2020). This is the case with (3): ‘Some people claim that the cabinet meeting was conducted “quietly and calmly”’. Moreover, indirect speech with embedded partial quotation-marks with any other quotative verb, as in (4): ‘Some people say that the cabinet meeting was conducted “quietly and calmly”’, is redundant, in as much as it combines the attributive meta-representation through indirect speech with the use of quotation marks in ironic utterances.
Unlike redundancy, explicitation refers to “any discursive procedure which is used to turn the implicit into explicit, off-record information and attitudes into on-record ones” (Weizman, 2020: 134). Thus, whereas redundancy increases the quantity of information, explicitation increases both its quantity and its visibility. Accordingly, explicitation may increase the level of redundancy, whereas redundancy does not necessarily increase explicitation (ibid.). An extreme case would be (5): ‘Some people say that the cabinet meeting was conducted “quietly and calmly”. It was not’. The present paper addresses the forms and functions of redundancy and explicitation, in French and Hebrew online op-eds, acknowledging that the results are limited to the discourse type under study.
3 Previous Findings
The study reported here is part of ongoing research which compares the forms and functions of irony in French and Hebrew. Specifically, it focuses on the forms and functions of redundancy and explicitation used to support the interpretation of irony. The comparison between these etymologically and typologically distinct languages was initially undertaken in the mid-1970s, when op-eds from the French daily Le Monde were translated regularly into Hebrew for publication in the Israeli daily Ha’aretz (Weizman, 1984, 1986). The comparison was motivated by the premise that the comparison between translations from language A to Language B in a given register presupposes a contrastive study of source language A and target language B in the same register (ibid.).
Initially, the contrastive analysis of French and Hebrew newspapers indicated that whereas in both of them ironic criticism was conveyed through the use of quotation marks, whether embedded in indirect speech or not. However, in Hebrew the irony conveyed by quotation marks was enhanced by stronger ties to the echoed utterance or by collocating with negatively-loaded emotional lexicon, whereas in the French data no such support was observed (Weizman, 1984). Later studies (Weizman, 2007, 2011, 2020) showed that the interpretation of ironic reservations was further secured in Hebrew by the use of the verb ‘claim’ (Hebrew ta’an), which, in journalistic discourse, conventionally implicates (Grice, 1975) the journalist’s reservations towards the echoed proposition, its content, its form or both (Dascal and Weizman, 1987; Weizman, 2020). Findings indicated that the equivalent of ‘claim’ in French (prétendre) was almost excluded from the French daily Le Monde (Weizman, 2011: 303). In 2020, a corpus-based analysis comparing two big corpora of journalistic discourse in French and Hebrew, each consisting of several millions of words, indicated that the use of the pattern ‘claim’ (prétendre) + partial quotations in indirect speech was not excluded from the French data, but it was still significantly lower than the use of its counterpart (ta’an) in the Hebrew data-set. This result partly confirms the findings of the previous studies mentioned above, which relied on much smaller data, and which indicated a higher preference in Hebrew for the disambiguation of ironic reservations. However, a pragmatic micro-analysis of 100 occurrences of ‘claim’ in the French and Hebrew corpora indicated that in half the cases the attitudinal function of the pattern under study was explicitated (see section 2.2.) in its co-textual environment. It was further found that in both languages, journalists had recourse to two explicitation strategies: stating rejection of the implied attitude and quoting conflicting views (Weizman, 2020). I interpreted these findings in terms of culture-specific preferences for redundancy and explicitation, as was explained in section 2.2.
This paper re-addresses the question of redundancy and explicitation in conveying irony and reservations, with one major difference: it focuses on the most informative end of the informativeness scale, i.e. on discursive patterns which include cues for indirectness side by side with lexical explictations through ‘ironically’, ‘ironic’,2 and then proceeds to explore other expliciations and redundancies in their textual environments. From a methodological viewpoint, this study benefits from the availability of big corpora.
The advantages of corpus-based discourse analysis and pragmatics have been discussed in several venues (see, among others, Baker, 2006; Partington et al., 2013; Aijmer and Rühlemann, 2015). The current research profits from the possibility to extend the analysis to larger, authentic, contextualized data and thus reveal more nuanced differences and similarities.
4 Data and Method
The study uses a mixed-methods approach that combines qualitative and quantitative analysis (also labeled horizontal and vertical methods, Rühlemann and Aijmer, 2015). This combination is typical of integrated, corpus-based pragmatic studies (Rühlemann and Aijmer, 2015). Two sets of corpora were used for each language. The first one is part of the Timestamped JSI Web Corpus Family, which is a set of more than 30 web corpora created by Jozef Stefan Institute, Slovenia (Trampus and Novak, 2012). Timestamped corpora consist of news articles crawled from sites across the world, updated with new texts daily. The second one belongs to the TenTen Corpus Family, currently denominated also Web corpora, which are available in more than 40 languages, Web crawled from the World Wide Web. The data-sets are accessible through Sketch Engine, a corpus-manager and analysis software (for a discussion of Sketch Engine by its founder, see Kilgarriff, 2014). Most corpora are tagged for parts of speech to facilitate complex searches which include lexical items (such as ironiquement [“ironically”], ironiser [“ironize”]), punctuations (such as quotation marks) and grammatical forms (such as indirect speech). However, whereas the French data-sets were fully tagged in both corpora, in Hebrew only a very small part was (Hebrew TenTen 2014, see list of corpora below). Consequently, we applied different methods to search for combinations of quotative verbs and quotation marks in French and in Hebrew. For the Hebrew we added the relatively small Hebrew General Corpus, since it consists of journalistic texts. The outcome of these searches is a list of utterances in which the form looked for is centralized. The possibility to extend the co-text of each utterance made it possible for us to read the utterances in their wider contexts. We applied the ‘text type’ meta-data filter tool available in the Sketch Engine corpora in order to retrieve results only from the online version of journals with nationwide circulation in both languages. Accordingly, the following outlets selected for French include Le Monde, Le Figaro, Libération, Le Parisien, Le Point, L’Express; and the outlets for Hebrew include haaretz.co.il; globes.co.il; maariv.co.il; nrg.co.il; inn.co.il; nana.co.il; calcalist.co.il; article.yedioth.co.il; 10tv.nana10.co.il; israelhayom.co.il; mako.co.il; nana10.co.il; news.walla.co.il; one.co.il; tve.co.il; ynet.co.il; mynet.co.il; rotter.co.il; news8.co.il; news1.co.il.
In order to ensure the selection of France-based and Israel-based journals only, we applied also the meta-data filter ‘source’, to search for all domains ending with ‘fr.’ and ‘il’.
To retrieve the patterns under study, we crawled the following corpora, last accessed on March 3, 2022:
French
Timestamped JSI web corpus 2014–2021, crawled from the Internet over the years 2014–2020. 8,166,710,150 words.
French Web 2017 (denominated also frTenTen17), crawled from the Internet in 2017. 5,752,261,039 words.
Hebrew
Timestamped JSI web corpus 2014–2021, crawled from the Internet over the years 2014–2020, 466,851,576 words.
Hebrew Web 2014 (denominated also heTenTen14), crawled from the Internet in 2014. tagged/tokenized, 895,876,116 words.
Hebrew Web 2021 (denominated also heTenTen21), crawled from the internet during Nov & Dec 2019, Nov & Dec 2020 and Jan 2021. 2,775,686,699 words.
Hebrew General Corpus, web crawled, mostly newspapers, crawling dates not documented on SketchEngine. 157,947,728 words.
Hebrew General Corpus, web crawled, mostly newspapers, crawling dates not documented on SketchEngine. 157,947,728 words.
Based on the sources detailed above, discursive realizations of explicitations and redundancy in both data-sets will be analzyed.
5 Discursive Realizations of Explicitation and Redundancy in a Cross-Cultural Perspective
The interpretation of ironic speaker’s meaning is a complex procedure. I have already mentioned that it requires the identification of cues for indirectness, i.e. the identification of a reason to believe that the speaker has intended to convey a meaning which diverges from the utterance meaning,3 and that this meaning conveys dissociation, derogatory judgment or criticism (Aristotle, 1960; Grice, 1975, 1978; Sperber and Wilson, 1981 among other. For a discussion, see Weizman, 2008). But full reconstruction of attributed ironic meanings, whether in line with the speaker’s intentions or not, involves making assumptions about three additional components: the victim (Barbe, 1995; Dews and Winner, 1995; Jorgenesen, 1996) or target (Sperber & Wilson, 1981; Holdcroft, 1983) of the ironic criticism, i.e. who is being criticized, the locus of irony, i.e. what is being criticized, and the ironist, i.e. the author of the ironic criticism (Weizman, 2008). The analysis which follows shows that each one of these components may be explicitated. Moreover, it will be shown that even the ironic meaning itself is sometimes explicitly spelled out. Still, the cases discussed here are interesting because the explicitations do not replace cues for indirectness. Rather, they are added to them. In other words, we focus only on explicitations which occur in contextual environments that include also patterns of indirectness.
Under these conditions, I consider explicitation of indirect meanings as redundancy, whereas redundancy is not necessarily an explicitation (section 2.2). In accordance with the aims of the paper, four patterns of explications are presented, considering the explicitated components and the cross-cultural comparison. All the utterances retrieved from the corpora include the French and Hebrew counterparts of the tokens ironic, irony, and ironically. The utterances retieved from the French corpora include also the verb ironiser (see section 4). A basic distinction is drawn between culture-specific similarities (5.1) and differences (5.2). Sub-categories within each of them are categorized by the combinations of discourse patterns. The analysis highlights the components of the explicitation – the ironic intent, the ironist, the target of irony and the locus of irony.
5.1 Comparing French and Hebrew: Similarities
5.1.1 Explicitating a Third Party’s Ironic Meaning: the Pattern ‘3rd Person/O + Quotative [Excluding ‘Claim’]+ Ironically/Ironic+ Meta-represented Speech’
In utterances which fall under this category the journalist diagnoses, as it were, the ironic intent of a third party. In some cases, additional components of the ironic interpretation are foregrounded, including the target of irony (i.e. who is being ironically criticized), the locus of irony (what is being criticized), the cue for irony (e.g. untruthfulness or echoic mention), and even the reconstructed speaker’s meaning. It will be argued that by so doing, the journalist shifts the responsibility for the interpretation of the speaker’s meaning of a third party from the reader to himself, and thus increases his accountability for speaker’s meaning (section 6). In some cases, the ironic interpretation is further supported by additional discourse patterns, in which case the explicitation is considered redundant. Whether redundant or not, however, through the explicitation the journalist ensures the ironic interpretation and at the same time limits the scope of the reader’s search for indirect meanings.
This category is represented in both the French and Hebrew data-sets.
Example (6) is an extract from an article whereby the journalist criticizes German Chancellor Merkel’s open-door policy and quotes public figures who oppose it.
(6)
Police union chief Reiner Wendt said rape and murder could have been prevented if Germany “had been better prepared for the dangers that always stem from massive immigration […]”. On social media, they wrote an ironic “thank you” to Chancellor Merkel for the open door policy that brought so many immigrants and refugees to Germany last year. (Hebrew, Ynet.co.il, Timestamped JSI Web Coprus 2014–2021)
The article discusses Merkel’s immigration policy. Following several remarks against it, the paragraph analyzed here opens with the unequivocal affirmations of the German Police union chief, who enumerates the dangers of immigration, associating them with no less than rape and murder (‘rape and murder could have been prevented if Germany “had been better prepared for the dangers that always stem from massive immigration […]”’). In this context, thanking the Chancellor for her open door policy would be most plausibly interpreted as an insincere expression of gratitude, and hence – as ironic (Haverkate 1990, see footnote 2). The journalist ensures the assignment of an ironic speaker’s meaning to the commenters, explicitly qualifying the expression of gratitude as ‘an ironic “thank you”’. The evaluation ‘so many immigrants and refugees’, which, in this context, conventionally implicates “too many”, supports the ironic criticism of the commenters if it is assigned to them; but it can also be assigned to the journalist, in which case he acts as an ironist too.
On this reading, then, there seem to be two levels of irony, with two plausible ironists – the commenters and the journalist, one locus – the Chancellor’s open door policy, and one target – the Chancellor. Textual support is provided for each interpretation. An ironic interpretation could have been reached without the qualifier ‘ironic’, relying on the use of partial quotations and the recognition of insincerity, but the interpretation is nevertheless ensured through the explicitation of the commenters’ ironic intent and an additional conventional implicature.
In example (7) the use of quotation marks is more ambiguous than in the preceding one, and the explicitation guides the reader towards a preferred meaning:
(7)
I easily found on the internet Edouard Frédéric-Dupont, died in 1995 (!), who had filed between 1946 and 1951 no less than 99 bills, of which many for families, the right of women and the status of guardians ... nicknamed ironically by his opponents “the deputy of the concierges” [“doormen”], of which he was very proud! (Lemonde fr., French Web 2017)
The person mentioned here, Frédéric-Dupont, is presented as highly willing to help ordinary citizens through legislation. Therefore, calling him ‘the deputy of the concierges’, whereby ‘concierges’ [“doormen”] is a metonymy for ordinary people, is not necessarily a criticism, nor is it necessarily an insult. Rather, it might be understood as a humoristic echoic mention of Frédéric-Dupont’s own preferences, as confirmed by ‘for which he was very proud’. This meaning, however, is not favored, given the contextual specific information that the nickname in question was assigned to him by his opponents. Unlike in the previous example, then, by specifying the identity of the ironists (‘his opponents’) and their ironic intent (‘ironically’) the journalist leads the readers towards his preferred interpretation.
In example (8), from the French data-set, the ironic interpretation would have been entirely dependent on contextual clues, had it not been for the explicitation of intent. In other words, it is not secured by redundant textual markers:
(8)
This Wednesday, August 7, the team goes to a small square in the 15th arrondissement, ironically nicknamed “Fouquet’s”. Seven men, of various ages and origins, are seated on the stone benches, a beer in their hands. (lefigaro.com., 12.9.2017, French Web 2017)
Fouquet’s Paris is an upscale, exclusive brasserie in Paris located at the prestigious Avenue des Champs-Élysées, which has hosted for more than 100 years famous public figures from all around the world, for example the actresses Simone Signoret and Catherine Deneuve, U.S. President Theodore Roosevelt, and many more. For the readers who possess this contextual knowledge, nick-naming a small café in a disadvantaged neighborhood after the luxurious Fouquet echoes the immense gaps in society, and would most plausibly be relied on as a cue for the ironic meaning of the nickname in question. Those who are not familiar with this cultural background could miss the speaker’s meaning and the criticism it implies. By explicating the ironic intent (ironically nicknamed), the journalist ensures the preferred ironic interpretation.
Unlike in the previous examples, in (9), the journalist goes as far as spelling out the ironic speaker’s meaning he attributes to a third party, in this case – the football-club Barcelona forward Leo Messy. The exchange reported here takes place a day after Barcelona has reached the quarter-finals of the Champions League for the eighth consecutive year, in the Barcelona vs. Manchester City game held on March 18, 2015. Several days before the Sunday’s game against Real Madrid, Messy expresses his confidence in his team but acknowledges the strengths of their opponent. In the following extract, he speaks about his special status as an undisputed star:
(9)
After playing one of his best games this season, Messi is the first to warn Real that they are more dangerous when under pressure and criticism: “Real are a great team with amazing players, we must respect them more than ever”. When asked about his good performance, the Argentine replied with irony: “I quickly went from being a disaster to being in my best moment, right?” Messi is not an actor who used to joke in front of the media. He does nothing in front of it except give boring answers. But there was a clear message in his words: on the one hand, he is highly aware of what is being written about him, and on the other hand, it is obligatory to take things in proportion, even when it comes to him. (Hebrew, Israelhayom.co.il, Timestamped JSI Web Corpus 2014–2021)
When answering journalists about his excellent performance, Messy alludes to the public’s intolerance towards his failures, echoing the exaggerated real or presumed evaluations of his critics, mostly the negative ones (‘I quickly went from being a disaster to being in my best moment’). As in the previous examples, the journalist foregrounds Messy’s ironic intent (‘replied with irony’). Furthermore, he goes as far as clarifying the speaker’s meaning of Messy’s ironic utterance, i.e. the latter’s expectation that the public will not judge him so harshly, which he qualifies as a ‘clear message’, i.e. ‘ it is obligatory to take things in proportion, even when it comes to him’, and ‘if one day he is not at his level, […] it means nothing’.
Interestingly, the French data-set presents an exceptional case of the explication of all the components of ironic interpretation:
(10)
In the first stacks of envelopes, the members of the Superior Council of the Judiciary first recognized those who are called, in the trade, the “habitual complainants” [“plaignants d’habitude”] – an expression which ironically echoes the “habitual delinquents” [“délinquants d’habitude”]. They are those who have already sent their criticisms – rarely justified – to the president of the court and to the prosecutor, even to the first president and to the general prosecutor of the court of appeal, sometimes even to the parliamentarians, and to the guard of the Seals[...] (lefigaro.com., French Web 2017)
Here, the journalist talks about citizens who habitually send their complaints to the judicial authorities, and are criticized for doing so. In this exceptional case, the author spells out the ironic speaker’s meaning conveyed by the nickname assigned to them, ‘habitual complainants’, foregrounds the ironic intent (ironically), spells out the main cue for ironic interpretation (‘“habitual complainants” – an expression which ironically echoes the “habitual delinquents”’), and points out to the locus of irony (‘rarely justified’) – the futility of repeatedly sending out these complaints to the superior court. While this textual maneuvering indicates the journalist’s high pragmatic awareness, it also shows the commitment he undertakes for the entire process of interpretation of a third party’s speaker’s meaning (see section 6).
So far we saw how journalists explicate various components of ironic messages that are authored by a third party (6–10), and sometimes join the ironist in his or her implied criticism (6,8 9). Their stance, however, remained implicit. In the next section journalists show commitment to their speaker’s ironic meanings by explicitating their own intents.
5.1.2 Explicitating the Journalist’s Ironic Speaker’s Meaning: The Pattern ‘1st Person + Quotative [Excluding ‘Claim’] + Ironically + Meta-represented Speech’
The pattern ‘I+ quotative (excluding ‘claim’)+ ironically+ meta-represented speech’ is particularly interesting. This is because, as I mentioned (section 1), irony is inherently indirect, which means that a speaker’s meaning is considered ironic only if the hearer reconstructs it from utterance meaning as a result of a search for cues and clues. In utterances pertaining to the category discussed here, the explicitation of the ironist and his or her intent may go as far as to cancel its ironic force. The potential cancellability of the ironic force pertains, to varying degrees, to all the explicitation patterns, including those presented in the previous section, but here it is most pronounced. The journalist positions herself as the animator, the author and the principal (Goffman, 1981) of the ironic criticism: he or she produces the ironic utterance, structures its form and content, and is therefore committed to the ironic meaning. the distanciation, which is part and parcel of ironic meaning may therefore be at risk. At the same time, the search for indirect meaning is not canceled altogether, since, as will be shown, the search for cues still applies at some point.
This complexity might account for the fact that the pattern ‘I + quotative + ironically + meta-represented speech’ is scarcely used in both French and Hebrew data -sets. As we shall see, just as with the previous pattern, different interpretation components may be explicated in addition to the ironist’s identity and the ironic intent.
In example (11), only the ironic intent is spelled out:
(11)
Reading Marx, even if his analysis of capitalism (and its shortcomings) is excellent, he commits a series of errors (the generalized Stockholm syndrome of the working masses is passed over in silence – I speak of the syndrome ironically), in particular on his historical analysis of the capitalism […] (lefigaro.com., French Web 2017)
In the opening paragraphs of this article, the journalist criticizes explicitly several aspects of Marx’s doctrine. He has recourse to indirectness only when arguing that Marx did not consider the admiration of the working class towards the bourgeoisie, which he calls ‘Stockholm syndrome’. This term, habitually signifying the psychological bond hostages develop towards their captors, is exaggerated in the context described here, and may plausibly be interpreted as a flouting of the maxim of quantity (Grice 1975), in which case it may lead to the re-construction of an ironic criticism addressed at Marx’s shortsightedness. At this point, the reader is still engaged in the search for a plausible speaker’s meaning. However, to ensure the desired reading, the journalist spells out his ironic intent as explicitly as possible in the first person (‘I speak of the syndrome ironically’).
The next example illustrates the explication of the very speaker’s meaning, in addition to the ironist’s identity and his intent. The article discusses what the author considers as the socialists’ past misconception of money and its role in establishing the social order.
(12)
We are like that because for years we used to say “you shouldn’t spend your life earning it”, and the socialists managed to have this “achievement” (I say “achievement” ironically because in any case with the franchouillarde mentality of the French people it was not difficult to make the French believe that we could live as before or even better by working less. (lefigaro.com., French Web 2017)
The journalist, who self-positions as one of the naïve socialists, uses the typical textual irony marker – quotation marks embedding a single word (‘the socialists managed this “achievement”’) (see also example 2, section 2.1). The irony may be detected if the word in quotation marks is read as an echoic mention which meta-represents the socialists’ self-esteem and the pride they take in their deeds, as represented by the conventional implicature of “achievement”, i.e. an accomplishment of a complex undertaking. In this interpretation, the target of irony is the echoed source, i.e. the socialists. The journalist, on his part, seems to have little trust in the readers, spells out his ironic intent ((‘I say “achievement” ironically’), and goes even further to explain his criticism of what he conceives as the socialists’ uncalled for self-esteem (‘because in any case with the franchouillarde mentality of the French it was not difficult [..]’).
5.1.3 Rejecting the Pattern ‘3rd Person + Quotative ‘Claim’+ Ironically+ Meta-represented Speech’
I started (section 3) with the observation that the use of the verb ‘claim’, which conventionally implicates the journalist’s reservations towards the echoed proposition, its content, its form, or both, was found to differentiate between French and Hebrew op-eds: in early studies, its Hebrew counterpart ta’an was amply used to ensure the interpretation of ironic reservations, whereas its equivalent prétendre was almost excluded (Weizman, 2007, 2011). In a later study, which compared the redundant patterns ‘ta’an/ prétendre + partial quotations in indirect speech’, both ta’an and prétendre were found in the data-sets, but prétendre was significantly less frequent than its Hebrew counterpart (Weizman, 2020). I interpreted these findings, among others, as representing a preference in Hebrew for higher redundancy and disambiguation of ironic reservations (section 3).
Against this background, in the current study I searched for the patterns ‘1st/3rd person + claim + ironically + meta-represented speech’, i.e. which contain the verb ‘claim’ as quotative. Unlike its counterpart which contains other quotative verbs (pattern 5.1.1) and concerns the ironic stance of the meta-represented source, the pattern analyzed here is used to convey both the meta-represented speaker’s ironic attitude (through ‘ironically’), as well as the journalist’s reservations (through the conventional implicature of ‘claim’ and the meta-represented speech). In such cases, the locus of the journalist’s irony and its victim may diverge with those of the quoted source (as in 13 below), converge with them, or remain ambiguous (as in 14 below).
The findings indicate that this pattern is extremely dispreferred in both data-sets. In both of them, no occurrences were found in the leading journals under study. In view of the interest presented by the changes over time described above, we extended the corpus to include the full corpora, including local news venues and public documentation, and came up with extremely small numbers of occurrences. These include 4 results in Hebrew ( i.e. 3.699e–7 per million in Hebrew Web 2014, 1,081,393,920 tokens) and 5 results in French (6.122e–8% per million in 8,166,710,150 million tokens), none of which featured in any of the national, leading journals which constitute our corpus. In our conceptual framework, this means that in both languages enhancing the explicitations and redundancy described in 5.1, 5.2 by adding the verb ‘claim’ is unacceptable in mainstream journals. Two short examples will briefly illustrate this pattern:
(13)
When Charles Gave ironically claims [prétend ironiquement] that the French are “divided on the subject of picking green beans or collecting snails”, we tell him that he should ask himself the question of which subject he is on. (French, Timestamped JSI Web Corpus 2014–2021, flux.md)
In this extract, the journalist discusses the political involvement of Charles Gave, a liberal economist and supporter of Eric Zemour, leader of the extreme right-wing party, in the 2022 general elections in France. Two stances are conveyed here. Gave ironically criticizes the French society by echoing presumed ordinary persons’ insignificant and trivial small talk, challenging the absurdity of assigning importance to these issues. His ironic intent is explicated by the journalist (‘ironically’). At the same time, the journalist conveys his criticism towards Gave through the use of the quotative ‘claim’, combined with meta-represented speech. He further enhances his dissociation by self-positioning as one of Gave’s numerous opponents (‘we’), and by so doing – positions Gave as ‘the other’.
This exceptional pattern is used also in (14) below. The article discusses allegations against the Pritzker Prize-winning architect Frank Gehry as concerns the Ray and Maria Stata Center, an academic complex he designed for the Massachusetts Institute of Technology (MIT). In 2007, a lawsuit was filed by MIT against Gehry and two building companies for serious flaws in the design. In the extract analyzed here, the journalist mentions former Boston University president John Silber:
(14)
According to [leta’anat] Silber, the building is absurd in several ways. First of all, for its price […] He ironically claimed that with this building Gehry had reached a climax, an expensive building that looked shabby, while other architects had managed to design cheap buildings that looked magnificent. Silber had a simple explanation for why this architecture is absurd […] In his opinion, no client would allow an architect to build him such an absurd building, neither for residence, nor for a commercial company.
Hebrew Web 2014, wordpress.com
An ironic criticism is addressed at Gehry’s work both by Silber and by the journalist. Silber’s irony is conveyed through the mismatch between the conventional implicature of ‘climax’, which creates expectations for positive evaluations, and the negative assessments which follow. The irony is enhanced by the non-logical, context-dependent paradox (Shukrun-Nagar, 2013) constructed between ‘expensive’ and ‘shabby’. The quoted source’s ironic intent is further explicitated through ‘ironcially’ and the reptitive negative evaluations (‘the building is absurd […]’, ‘this architecture is absurd’, ‘such an absurd building’). The journalist’s use of ‘claim’ (‘he ironically claimed’) seems to convey the distanciation expected from reports on judicial affairs. As such, his stance is rather ambiguous. However, the redundant rendering of the absurdity and paradoxicality of the situation, which may be interpreted as situational irony, might also imply the journalitst’s affiliation with Silber’s ironic stance and its target, Frank Ghery.
The rejection of this pattern in mainstream journals deserves some attention, and will be discussed later (section 6).
So far we examined the similarities between the two data-sets. The comparison between them yielded one marked difference: the use of the verb ironiser in French, to be discussed in 5.2.
5.2 Comparing French and Hebrew: Differences Explicating a Third Party’s Ironic Meaning: the Verb ‘Ironiser’
In her analysis of what she calls “explicit irony” in English, Barbe (1995) notes: “Explicit irony is becoming more and more forceful a part in everyday conversation. Perhaps soon, the nonce-verb ‘ironize’ can be accepted in colloquial speech or academic language and then find its way into dictionnaries” (Barbe 1995, 143). What seemed to Barbe as a nonce-verb in English nearly 30 years ago has no parallel form in Hebrew, as there is no verb coined after the Hebrew noun ironya (“irony”). In French, on the other hand, ironiser is a grammatically regular verb, which features in dictionaries as well as in our corpora.4 Noun-based derivation of verbs is less productive in Hebrew than it is in French, since it is subject to more severe morphological constraints. Here lies the most remarkable difference between our two data-sets: not only does the French use an explicitation marker which does not exist in the Hebrew language, it does so more often than it uses any of the other available explicitation strategies, with 699 occurrences (0.1 per million tokens) in French Web 2017 (6,845,630,573 tokens), and 8681 0ccuurences (1.06 per million tokens) in Timestamped JSI web corpus 2014–2021 French (8,166,710,150 tokens).
In French, ironiser is an intransitive verb that signifies “to treat something or someone with irony”, and it either features with no compliment or collocates with contre (“against”]) and sur (“about”) (Trésor de la Langue Française Informatisée). Thus, unlike categories 5.1.1, 5.1.2, and 5.1.3 it can not be used with represented speech as its compliment. In our French data-set, it is used mostly to explicate the function of quotations in direct speech.
I argued before (section 2.2) that the explicitation discussed here is essentially redundant since it provides information about components of the ironic meaning for which there are already cues in the text. We have also seen that it sometimes contributes to the disambiguation of potential ambiguities (example 7). This is the case in (15) below.
In an article published in the left-wing daily Libération, the journalist discusses the changes initiated by Jean-Marie Le Pen, president of the far-right party Le Front National between 1972–2011, in his party, mostly the handing over of his executive powers to the party’s vice president, his daughter, Marine le Pen. The article opens with a metaphor of a house in the process of slow destruction and proceeds to a quote of a party member, who elaborates on the semantic field of a house which is being torn apart and its furniture sold in pieces. The journalist alludes to the Paquebot, i.e. the general headquarters of the Front National, which was sold in 2011 in order to cover the party’s debts, thus preparing for the 2012 general elections.
(15)
Not an explosion, nor a split like the one in 1998, just a slow dislocation. The walls of the Front National house are cracking on all sides. “In the absence of being able to sell the Paquebot in one piece, it is as if we were proceeding to a sale by cutting. Each one carrying a small piece of the common house and some furniture as a souvenir”, ironized a member of the far-right party. The announcement by Le Pen of his decision to hand over control in 2010 to the next congress and to entrust the helm of the Paquebot, the headquarters of the FN in Saint-Cloud (Hauts-de-Seine), to the youngest of his daughters and vice-president of the FN, Marine Le Pen, has accelerated the internal crisis. (French, Libération, French Web 2017)
The metaphor of a house torn apart, authored by a far-right activist, is brought up by the journalist in direct speech (“[…] it’s as if we were proceeding to a sale by cutting. Each one carrying a small piece of the common house and some furniture as a souvenir”). The metaphor is further enhanced by the allusion to the iconic Paquebot, which symbolized its power, until it had to be sold. The quotation could be interpreted either as a sentimental expression of disappointment with the leader’s growing weakness or as a bitter, ironic criticism expressed by a political activist towards his leader and his actions. The journalist opts for the second meaning and makes it clear through the use of the quotative (‘ironized a member of the far-right party’). In a way, he imposes his own interpretation of a third party’s intention on the reader, or at least directs her to his preferred reading.
The use of explicitations for disambiguation, as in 15, is rather rare in the two data-sets (but see also example 7). More typically, they collocate with cues that might less equivocally lead toward the identification of irony. This is the case in (16) below. The journalist judgementally explicates the ironic intent of a third party in an utterance which is intertwined with several challenging comments addressed at the victim of irony, and thus accentuates the derogatory attitude inherent in irony:
(16)
The first words of the prosecutor, Bernard Beffy, freeze the audience. “The day of glory has arrived, Mr. Mayor! You have your reward: your reputation and your status as an expiatory victim!” He reproaches the defendant for having refused mediation and a guilty plea that had been offered to him, subject to the payment of a fine of 600 euros. “You wanted the court, because you wanted a platform?” he says. He attacks with the same aggressiveness the elected officials – for the most part the mayors without labels of small municipalities and other, sympathizers or militants of the PS – who have come to support their colleague and their “somewhat primary corporatism”. He ironizes heavily: “We would like to know, Coulsore, is this Chicago of the 1930s? Fortunately, you are not the mayor of Montfermeil, Monsieur Boisars?”. On his bench, Maurice Boisard seems stunned. (French, lemonde.fr., French Web 2017)
The Mayor of Coulsore, a small community of 2500 inhabitants, has been accused of slapping a young man who destructed public property, acted violently in public and defied the Mayor personally. Consequently, the Mayor is being charged and brought to trial. The journalist, appalled at the undue severity of the judicial authorities’ reaction, joins the inhabitants and several mayors of other mall cities in expressing his indignation and addresses the humiliation to which the Mayor of Coulsore was subject in the trial. Specifically, he criticizes the prosecutor’s aggressive attack against the defendant and his supporters, which he quotes in some detail. The prosecutor accuses the Mayor directly of taking advantage of the situation and exploiting it for political benefits (“You wanted the court, because you wanted a platform!”) and for his incompetence (“Have you set up a local security and crime prevention council?”). Then, the prosecutor challenges him ironically, through exaggeration, i.e. the flouting of Grice’s (1975) maxim of quantity (“We would like to know, Coulsore, is this Chicago of the 1930s?”), and through the echoic mention of the Mayor’s presumed thoughts, unduly assigning him complaints about the disorder in his own community (“Fortunately, you are not the Mayor of Montfermeil, Monsieur Boisard!”). The journalist, engaged in transmitting not only the content of the prosecutor’s allegations but also his condescending tone (Bakloudi and Bres, 2016), explicates the ironic intent and even criticizes it (‘ironizes heavily’ [ironise avec lourdeur]).
The textual analysis will be concluded with a typical case of explicitation intertwined with highly redundant cues:
(17)
On Tuesday morning, the centrist deputy, previously close to François Bayrou, disclosed that he expected that with this debate “the campaign will start”. He said that he was able to persuade “simply by speaking during the 19 minutes to which I am entitled”. He protested a few hours before the debate against the title “small candidate” [“petit candidat”] attributed to those who do not reach a double-digit score in the polls. “‘The small ones’ [“les petits”], the ‘tiny ones’ [“les minuscules”], who will disturb the game of the ‘huge ones’ [“les immenses”]. I get on my knees every morning looking at the ‘huge ones’, they are so beautiful”, he ironized. (French. lefigaro.fr, frtenten17)
The article discusses the status of political candidates for the presidential elections who represent small parties. To illustrate the problem, the journalist tells the story of Jean Lassalle, founder of the Resistons! (Res) party, who presented his candidature in the 2017 presidential elections and received 435,301 votes (1.21%). Lasalle is quoted as firmly contesting the despising titles assigned to leaders of small parties (‘The small ones’ [“les petits”], the ‘tiny ones’ [“les minuscules”], who will disturb the game of the ‘huge ones’.) He dramatizes the situation, alluding to worship practices (‘I get on my knees every morning’), and echoically mentions the nickname ‘small candidates’. By so doing he elaborates on its semantic field (“tiny ones” [“minuscules”], “huge ones” [“immenses”]), marked by quotation marks embedding expressions which represent his interpretation of the accepted naming, “small ones”. This redundant use of cues is further enhanced by the explicitation of the ironic intent through the verb ‘ironiser’ [he ironized].
The verb ironiser has been shown to provide a form of explicitation which is inexistent in Hebrew. As we have seen, it further contributes to the construction of redundancy, by collocating with explicitation patterns that do exist in both the Hebrew and French data-sets. In the concluding remarks I will discuss some of the implications of the similarities and differences highlighted above.
6 Discussion: Irony Explicitation, Informativeness and Accountability
Comparing the two data-sets under study, the main difference is the use of the verb ironiser to explicitate the ironic intent of a third party, a category which is inexistent not only in the Hebrew corpora, but in the Hebrew language, since there is no comparable verb which fulfills a similar function (section 5.2). Other than that, the two data-sets are similar in various ways. In both of them, journalists explicitate their own ironic intents (5.1.2) as well as those of other parties (5.1.1) through the use of similar patterns (‘1st/ 3rd person + quotative [excluding ‘claim’] + ironically + meta-represented speech’ in its various forms); in both of them, each and every component which plays a part in the process of interpreting irony was found to be explicitated, including the identity of the ironist, the target and the locus of irony. Journalists go as far as explicitating all the ironic components, and even spelling out their interpretation of the speaker’s meaning. In most cases, the interpretation of ironic meanings is not only explicitated, but is further supported by redundant cues. Furthermore, the two data-sets resemble in the inacceptability of the pattern ‘3rd person + quotative ‘claim’+ ironically+ represented speech’, in which explicitations and redundancies are further enhanced by the quotative ‘claim’, which conventionally implicates reservations and criticism.
These results call for some pondering. I will discuss them in terms of informativeness, accountability and commitment to speaker’s meaning, and will thus consider also their development over time.
As noted above (section 3), this study is part of ongoing research that compares the discursive realizations of irony and reservations in op-eds in French and Hebrew mainstream online journalism. Previous studies indicated that whereas in both French and Hebrew ironic criticism was conveyed mostly through the use of echoic quotation marks, whether embedded in partial indirect speech or not, in Hebrew the irony conveyed by quotation marks was enhanced by stronger ties to the echoed utterance or by collocating with negatively-loaded emotional lexicon, and in the French data no such support was observed (Weizman, 1984). Later studies (Weizman, 2007, 2011, 2020) showed that the interpretation of ironic reservations was further secured in Hebrew by the use of the verb ‘claim’ (Hebrew ta’an), which, in journalistic discourse, conventionally implicates the journalist’s reservations and criticism (Dascal and Weizman, 1987; Weizman, 2020). No such support was provided in French, since the comparable verb (prétendre) was almost excluded from the French daily Le Monde (Weizman, 2011). These findings, as well as others, which concern the realization of requestive hints (Weizman, 1989), led me to advocate a cross-cultural comparison in terms of quantity scales. I suggested to locate each language on a scale of its relative informativeness within a set of the languages under study, thus indicating its preferences for degree of informativeness as regards certain forms and functions in a given discourse type (Weizman, 2007). Accordingly, Hebrew was found to be located near the high-informative end of the scale as compared to French in terms of the realization of irony and reservations in journalistic op-eds5 (Weizman, 2007, 2011). In a later study, a corpus-based analysis comparing two big corpora of journalistic discourse in French and in Hebrew, each consisting of several million words, showed a growing use of the verb prétendre in similar environments in French, but it was still significantly lower than the use of its counterpart (ta’an) in the Hebrew data-set. The differences between French and Hebrew op-eds in degree of informativeness grew smaller, in that in both of them the reservations conveyed indirectly were explicated by similar strategies – rejection of the implied attitude and quoting conflicting views (Weizman, 2020). Still, Hebrew showed marked preference for discursive informativeness as compared with French (ibid.). The conceptualization of explicitation and redundancy in terms of degree of informativeness informed the present contrastive analysis, considering the discursive patterns analyzed in section 5 and identified through a bottom-up corpus-based analysis as their discursive realizations.
The current study differs from the previous ones in that it focuses on the most informative end of the informativeness scale, i.e. on discursive patterns which include cues for indirectness side by side with lexical explicitations through ‘ironically’, ‘ironic’, and then proceeds to explore other expliciations and redundancies in their textual environments. The findings surprisingly indicate that the op-eds in the French sub-data are located at the high-informativeness end with the use of ironiser (section 5.2), which has no comparable form or equivalent function in Hebrew. As mentioned earlier, the pragmatic functions of this verb have been under-investigated and we can get a wider picture of their implications if their uses are further explored.
More puzzling are the similarities between the two data-sets, and the somewhat surprising understanding that both in French and in Hebrew journalists choose to explicitate irony and reservations, though not frequently, and that they do so using similar discursive patterns. I argue that when journalists opt for high-level explicitation of irony and reservations, culture-specific discursive preferences are put on hold, and general, possibly universal genre-dependent tendencies are brought in. On this account, journalists ensure what seems to them the preferred interpretation, and by so doing they assume discursive responsibility for it. Let me clarify this point.
Irony and explicitation may seem to be a contradiction in terms. Irony is inherently indirect and as such – its identification and understanding presuppose the decoding of off-the-record cues. Moreover, the attribution of presumed ironic speaker’s meanings, whether intended by the speaker or not, requires making assumptions not only about an ironic intent, but also about possible ironist, target and locus of irony. The search for an ironic speaker’s meaning is mostly “an interpretive experience” (Tobin and Israel, 2012: 44); it allows for searches for cues and clues while going back and forth, making assumptions in context, rejecting some of them and embracing others (Dascal and Weizman, 1987; Weizman and Dascal, 1991). This search may even result in an ambiguous reconstruction, or in unstable irony, whereby “no stable reconstruction can be made out of the ruins revealed through the irony” (Booth, 1974: 240). Discursive redundancy in the text may provide the reader with shortcuts in this challenging journey; explicitation may put an end to this entire quest for meanings. This possibility does not apply to the cases discussed here, since I have chosen to focus only on those utterances where explicitations and redundancies collocate with cues for indirectness. Still, through explicitation readers’ options are being narrowed down and the scope of the reader’s search for indirect meanings is limited.
Why would the journalist want to play this give-and-take game? Due to its inherently indirect nature, irony is always risky; and it is specifically risky in the genre of online journalism, which involves multiplicity and heterogeneity of audiences. Through explicitation journalists cater to the needs of the readers, which include an increase in disambiguation and the minimization of discursive risks, even at the expense of interpretative challenges and alignment with culture-dependent preferences. Interestingly, the finding that in both languages enhancing explicitation and redundancy by the redundant ‘claim’ is unacceptable in mainstream journals (section 5.1.3) seems to indicate that disambiguation of irony has its limits, and that these limits are similar in both languages.
The disambiguation of irony puts additional weight on the journalists’ shoulders. By spelling out what they consider as the ironic intent of others and the various components of its meaning, journalists self-position as manifestly accountable and committed to the presumed ironic meanings they assign to the discourse of others; by spelling out their own ironic intent and its components, they give up on the possibility to deny the implicit criticism conveyed in their own indirect discourse (Weizman, 1989). Furthermore, the analysis shows that in cases of ambiguity journalists may even assume responsibility for foregrounding one speaker’s meaning over the other (e.g. extracts 7, 15). In such cases, they seem to assume that their readers share the same presuppositions and knowledge, and, consequently, their interpretations of the situations (Barbe, 1995).
It has long been accepted that speakers’ accountability and commitment to communicative meanings have social, moral and ethical implications, which concern the participants’ rights and obligations (Katriel and Dascal, 1989; Dynel, 2016; Garfinkel, 1967; Haugh, 2013; Searle 2010, Weizman and Fetzer, 2021). At the same time, as noted by Haugh (2013: 43), “the relative degree of these respective rights, responsibilities and so on can be modulated through the different ways in which talk can be formulated by participants”. By explicitating irony journalists engage in a give-and-take game with their readers. Conflicting forces are at play: interpretation paths are opened by irony, and are then narrowed down by the journalist’s interpretations. The findings analyzed in this paper foreground this tendency and shed light on the potential contribution of cross-cultural contrastive analysis to the understanding of pragmatic developments.
Acknowledgements
This paper is dedicated to the late Prof. Shoshana Blum-Kulka, my teacher, advisor, colleague and friend, with deep appreciation for her guidance and inspiration.
I am most grateful to Prof. Il-Il Yatziv-Malibert for delving with me into the context of the French data, and to the anonymous reviewers for their valuable comments on this paper.
References
Aristotle. 1960. The rhetoric. (tr. Lane Cooper). New York: Appleton-Century-Crofts.
Baker, Paul. 2006. Using Corpora in Discourse Analysis. London/New York: Continuum.
Baklouti, Elodie and Jacques Bres. 2016. Ce qu’ironiser veut dire… De l’usage métadiscursif des termes ironie, ironiser, ironique(ment) dans le texte théâtral et dans le texte journalistique. In: Geneviève Salvan and Amir Biglari (eds.), Figures en Discours, Paris: L’Harmattan. 978-972-8061-0267-6. . Accessed 20 August 2022.
Barbe, Katharina. 1995. Irony in Context. Amsterdam/Philadelphia: John Benjamins.
Blum-Kulka, Shoshana, Juliane House, and Gabriele Kasper (eds.). 1989. Cross-Cultural Pragmatics: Requests and Apologies. Norwood, New Jersey: Ablex.
Booth, Wayne C. 1974. A Rhetoric of Irony. Chicago: University of Chicago Press.
Chesterman, Andrew. 1998. Contrastive Functional Analysis. Amsterdam: John Benjamins.
Dascal, Marcelo and Elda Weizman. 1987. Contextual exploitation of interpretation clues in text understanding: An integrated approach. In: Jef Verschueren and Marcella Bertucceli-Papi (eds.), The Pragmatic Perspective. Amsterdam/ Philadelphia: John Benjamins, 31–46.
Dews, Shelly and Ellen Winner. 1995. Muting the meaning: A social function of irony. Metaphor and Symbolic Activity 10 (1): 3–19.
Dynel, Martha. 2016. With or without intentions: Accountability and (un)intentional humor in film talk. Journal of Pragmatics 95: 67–78.
Egan, Thomas. 2013. Tertia comparationis in multilingual corpora. In: Karin Aijmer and Bengt Altenberg (eds.), Advances in Corpus-based Contrastive Linguistics: Studies in Honour of Stig Johansson. Amsterdam/Philadelphia: John Benjamins, 7–24.
Garfinkel, Harold. 1967. Studies in ethnomethodology. Cambridge: Polity Press.
Goffman, Erving. 1981. Forms of Talk. Oxford: Blackwell.
Grice, Herbert Paul. 1975. Logic and conversation. In: Peter Cole and Jerry L. Morgan (eds.), Syntax and Semantics: Speech Acts, vol. III. New York, San Francisco and London: Academic Press, 41–58.
Grice, Herbert Paul. 1978. Further notes on logic and conversation. In: Peter Cole (ed.), Syntax and Semantics: Pragmatics, vol. IX. New York: Academic Press. 113–127.
Haugh, Michael. 2013. Speaker meaning and accountability in interaction. Journal of Pragmatics 48: 41–56.
Haverkate, Henk. 1990. A speech act analysis of irony. Journal of Pragmatics 14: 77–109.
Holdcroft, David. 1983. Irony as trope, and irony as discourse. Poetics Today 4 (3): 493–511.
House, Juliane and Daniel Kadar. 2021. Cross-Cultural Pragmatics. Cambridge: Cambridge University Press.
House, Juliane and Daniel Kadar. 2022. Political language in contrast: An introduction. Journal of Pragmatics 188: 132–137.
Johansson, Stig. 2007. Seeing through Multilingual Corpora: On the Use of Corpora in Contrastive Studies. Amsterdam: John Benjamins.
Jorgensen, Julia. 1996. The functions of sarcastic irony in speech. Journal of Pragmatics 26: 613–634.
Kampf, Zohar, 2021. Political speech acts in contrast: the case of calls to condemn in news interviews. Journal of Pragmatics 180: 203–211. https://doi.org/10.1016/j.pragma.2021.04.031.
Katriel, Tamar and Marcelo Dascal. 1989. Speaker’s commitment and involvement in discourse. In: Yishai Tobin (ed.), From Sign to text: A Semiotic View of Communication, Amsterdam/Philadelphia: John Benjamins, 275–296. https://doi.org/10.1075/fos.20.21kat.
Kilgarriff, Adam, Vit Baisa, Jan Bušta, Miloš Jakubíček, Vojtěch Kovář, Jan Michelfeit, Pavel Rychlý and Vit Suchomel. 2014. The Sketch Engine: Ten years on. Lexicography: Journal of ASIALEX 1 (1). 7–36. https://doi.org/10.1007/s40607-014-0009-9.
Krzeszowki, Tomasz P. 1990. Contrasting Languages: The Scope of Contrastive Linguistics. Berlin: De Gruyter Mouton.
Ogiermann, Eva. 2009. On Apologising in Negative and Positive Politeness Cultures. Amsterdam/Philadelphia: John Benjamins.
Ogiermann, Eva and Bella Spyridoula. 2020. An Interlanguage Study of Request Perspective: Evidence from German, Greek, Polish and Russian Learners of English. Contrastive Pragmatics 1(2): 180–209. doi: https://doi.org/10.1163/26660393-BJA10003.
Partington, Alan, Duguid, Alison and Charlotte Taylor. 2013. Patterns and Meanings in Discourse: Theory and Practice in Corpus-Assisted Discourse Studies (CADS). Amsterdam/Philadelphia: John Benjamins. https://doi.org/10.1075/scl.55.
Rühlemann, Christoph and Karin Aijmer. 2015. Corpus pragmatics: Laying the foundations. In: Karin Aijmer and Christoph Rühlemann (eds.), Corpus Pragmatics: A Handbook. Cambridge: Cambridge University Press, 1–26.
Searle, John. 2010. Making the Social World. The Structure of Human Civilization. Oxford: Oxford University Press.
Shukrun-Nagar, Pnina. 2013. The construction of paradoxes in news discourse: The coverage of the Israeli Haredi community as a case in point. Discourse Studies 15(4): 463–480.
Sperber, Dan and Deirdre Wilson. 1981. Irony and the use-mention distinction. In: Peter Cole (ed.), Radical Pragmatics. New York: Academic Press, 295–318.
Sperber, Dan and Deirdre Wilson. 1986. Relevance: Communication and Cognition. Cambridge, Mass.: Harvard University Press.
Sperber, Dan. 2000. Metarepresentations in an evolutionary perspective. In: Dan Sperber (ed.), Metarepresentations: A Multidisciplinary Perspective. Oxford: Oxford University Press, 117–137.
Tobin, Vera and Michael Israel. 2012. Irony as a viewpoint phenomenon. In: Barbara Dancygier and Eve Sweetser (eds.), Viewpoint in Language: A Multimodal Perspective. Cambridge: Cambridge University Press, 25–46.
Trampus, Mitja and Blaz Novak. 2012. Internals of an aggregated web news feed. In: Proceedings of 15th Multiconference on Information Society. http://ailab.ijs.si/dunja/SiKDD2012/Papers/Trampus. (Accessed 20 June 2022).
Trésor de la Langue Française Informatisée, http://www.atilf.fr/tlfi.
Weizman, Elda. 1984. Some register characteristics of discourse structure in journalistic language. Applied Linguistics V/1: 39–50.
Weizman, Elda. 1986. An interlingual study of discourse structures: Implications for the theory of translation. In: Juliane House and Shoshana Blum-Kulka (eds.), Interlingual and Intercultural Communication: Discourse and Cognition in Translation and Second Language Acquisition Studies. Tübingen: Narr, 115–126.
Weizman, Elda. 1989. Requestive hints. In: Shoshana Blum-Kulka, Juliane House and Gabriele Kasper (eds.), Cross-Cultural Pragmatics: Requests and Apologies. New Jersey: Ablex, 71–95.
Weizman, Elda. 2001. Addresser, addressee and target: Negotiating roles through ironic criticism. In: Edda Weigand and Marcelo Dascal (eds.), Negotiation and Power in Dialogic Interaction. Amsterdam: John Benjamins, 125–137.
Weizman, Elda. 2007. Quantity scales: Towards culture-specific profiles of discourse norms. In: Marion Grein and Edda Weigand (eds.), Dialogue and Culture. Amsterdam/Philadelphia: Benjamins, 141–152.
Weizman, Elda. 2008. Positioning in Media Dialogue: Negotiating Roles in the News Interview. Amsterdam/Philadelphia: John Benjamins.
Weizman, Elda. 2011. Conveying indirect reservations through discursive redundancy. Language Sciences 33: 295–304.
Weizman, Elda. 2020. The discursive pattern ‘claim+ indirect quotation in quotation marks’: Strategic uses in French and Hebrew online journalism. Journal of Pragmatics 157: 131–141.
Weizman, Elda and Marcelo Dascal. 1991. On clues and cues: Strategies of text understanding. Journal of Literary Semantics XX/1,18–30.
Weizman, Elda and and Ayelet Kohn. Forthcoming. Resemblance in comments/posts interaction: its dialogcitiy and functions. Pragmatics and Society.
Weizman, Elda and Anita Fetzer. 2021. The discursive construction of accountability for communicative action to citizens: A contrastive analysis across Israeli and British media discourse. Intercultural Pragmatics 18(5): 605–632. DOI:10.1515/ip-2021-5002.
Weizman, Elda, Hasan Atifi and Michel Marcoccia, Work in progress. Corpus-based contrastive analysis of meta-comments on irony in French, Moroccan Arabic and Hebrew.
Wilson, Deirdre, 2012. Metarepresentation in linguistic communication. In: Deirdre Wilson and Dan Sperber (eds.), Meaning and Relevance. Cambridge: Cambridge University Press, 230–258.
Wilson, Deirdre and Dan Sperber. 1992. On verbal irony. Lingua 87: 77–90.
Biographical Note
Elda Weizman is professor emerita at Bar Ilan University, Israel, where she teaches cross-cultural pragmatics. Her research interests, anchored in socio-pragmatics, focus on the negotiations of positionings in mediated political discourse, with special interest in political interviews, online commenting, the construction of ordinariness and the negotiation of ironic meanings. She conducts a comparative corpus-based research on irony and reservations in French and Hebrew online media discourse.
She is associate editor of Pragmatics and Topics on Humor Research, and member of several editorial boards, including Internet Pragmatics, Contrastive Pragmatics and Hebrew Linguistics. She has published and edited in English, Hebrew and French.
This paper is written in memory of Shoshana Blum-Kulka.
The notion of ‘culture-specific’ preferences is used here in the sense House and Kadar’s (2021, 2022) ‘linguaculture’, i.e. “conceptualisations of culture through conventional patterns of language use” (House and Kadar, 2022: 133).
“Irony” and “cynicism” have been found to be used interchangeably in online discourse, with different culture-dependent preferences (see, e.g. the preference for “cynicism” as compared to “irony” in online commenting in Hebrew, Weizman 2015, versus the preference for “ironie” as compared to “cynicism” in French, Weizman, Marcoccia and Atifi, work in progress). The comparison of “irony” and its derivations in terms of explictation has yielded the most interesting results as regards the French/Hebrew contrastive analysis.
In addition to the identification of an echoic mention (section 2.1), major cues include a blatant flouting of Gricean maxims (Grice, 1975, 1978 ) and the violation of the sincerity condition underlying the felicitous performance of a speech act (Haverkate, 1990).
To my knowledge, the pragmatic uses of the verb ironiser have not received special attention in the literature. For example, Baklouti et al (2016), in their discussion of meta-talk about irony in drama and journalistic discourse, make no distinction between ironiquement and ironiser.
Additional evidence for the preferences of relatively high informativeness in Hebrew was provided by the comparison of discursive realizations of requestive hints in Hebrew, Canadian French and Australian English, Weizman, 1989, 2007).
