2.1Non-literal and attitudinal content of ironic utterances

Verbal irony comes in many different forms. A common definition holds that an ironic utterance articulates a meaning that is an alternative to the expression’s literal mea-ning; see, e.g., Grice (1989)Grice, H. Paul 1989 Studies in the Way of Words. Cambridge, MA: Harvard University Press.. In the present study, we focus on ironic utterances that contain non-literally used lexical material11.A common device to indicate non-literal, ironic uses of expressions are scare quotes; see, e.g., Predelli (2003)Predelli, Stefano 2003 “Scare Quotes and Their Relation to Other Semantic Issues.” Linguistics and Philosophy 26: 1–28. , Schlechtweg and Härtl (2023)Schlechtweg, Marcel, and Holden Härtl 2023 “Quotation Marks and the Processing of Irony in English: Evidence from a Reading Time Study.” Linguistics 61 (2): 355–390 . and exclude from the analysis other types such as ironic enthusiasm (e.g., Oh, yay, he crashed my brand new car!) or ironic rhetoric questions (How old did you say you were?), where irony operates on the illocutionary level.

An ironic interpretation of an utterance interacts with an incongruity between what is said and some element of the context or situation at hand. Consider the example in (2), which, when uttered ironically, asserts that the bar in question is not buzzing with people, that is, a meaning that deviates from the sentence’s literal meaning.

Drawing on Kaplan’s (1999)Kaplan, David 1999 The Meaning of ‘Ouch’ and ‘Oops’: Explorations in the Theory of Meaning as Use. Unpublished Ms. Los Angeles: University of California. distinction between descriptive and expressive content, we consider the negated22.For an approach that analyzes verbal irony to involve a form of (indirect) negation, see Giora (1995)Giora, Rachel 1995 “On Irony and Negation.” Discourse Processes: 19: 239–264. . version of the semantic content of (2) to represent the descriptive content of the sentence. We presume that, while the locution of (2) suggests that the bar in question is buzzing with people, the opposite meaning is intended as the assertion, producing an update of the common ground with the corresponding proposition, implying that the bar is not buzzing with people.

The non-literalness of the meaning of the expression is a central characteristic of ironic and sarcastic language; see, e.g., Sperber and Wilson (1981)Sperber, Dan, and Deirdre Wilson 1981 “Irony and the Use–Mention Distinction.” In Radical Pragmatics, ed. by Peter Cole, 295–318. New York: Academic Press.. A second element typically found in ironic utterances is the speaker’s intention to produce an evaluative comment reflecting beliefs concerning an entity’s value or significance, using criteria governed by a set of cultural standards, see, for example, Wilson and Sperber (1992)Wilson, Deirdre, and Dan Sperber 1992 “On Verbal Irony.” Lingua 87: 53–76. . We consider this content to represent expressive content, i.e., content that is non-truth-conditional and not at issue (Härtl and Bürger 2021Härtl, Holden, and Tatjana Bürger 2021 “ ‘Well, That’s Just Great!’ – An Empirically Based Analysis of Non-Literal and Attitudinal Content of Ironic Utterances.” Folia Linguistica 55: 361–387. ; Potts 2005Potts, Christopher 2005 The Logic of Conventional Implicatures. Oxford: Oxford University Press., 2007Potts, Christopher 2007 “The Expressive Dimension.” Theoretical Linguistics 33: 165–197. ). Consider the example in (3), uttered ironically.

With this utterance, the speaker communicates that the singer did not hit every note of the musical piece in question. At the same time, the speaker produces an evaluative comment that articulates a negative attitude towards the corresponding denotatum, that is, the lead singer’s performance. Ironically presented assertions conventionally express an evaluation, which is not the case with their negated, non-ironic equivalents. Hence, considering the following examples, we assume that only (4a), uttered ironically, conventionally implicates the speaker’s evaluation that the soloist sang poorly but not the non-ironic counterpart in (4b), which is used to describe a fact.

The speaker’s evaluative attitude expressed with an ironic utterance is typically negative, referred to as ironic criticism, but it can also be positive (ironic praise); see, e.g., Dews and Winner (1999)Dews, Shelly, and Ellen Winner 1999 “Obligatory Processing of Literal and Nonliteral Meanings in Verbal Irony.” Journal of Pragmatics 31: 1579–1599. , Kreuz and Link (2002)Kreuz, Roger, and Kristen Link 2002 “Asymmetries in the Use of Verbal Irony.” Journal of Language and Social Psychology 21: 127–143. . The latter type is illustrated in the following example.

When uttering (5) ironically, the speaker literally says something negative in order to express something positive. Wilson (2013)Wilson, Deirdre 2013 “Irony Comprehension: A Developmental Perspective.” Journal of Pragmatics 59: 40–56. argues that ironic praise is subject to stronger use restrictions than ironic criticism, which is often considered the default in verbal irony; see also Wilson and Sperber (1992)Wilson, Deirdre, and Dan Sperber 1992 “On Verbal Irony.” Lingua 87: 53–76. . The standard of verbal irony to involve a negative attitudinal polarity has been explained by means of a normative bias (Kreuz and Glucksberg 1989Kreuz, Roger, and Sam Glucksberg 1989 “How to Be Sarcastic: The Echoic Reminder Theory of Verbal Irony.” Journal of Experimental Psychology: General 118: 364–386.). Due to this normative bias, which describes people’s general aspiration to fulfill social norms rather than defy them, an ironically praising utterance must refer to a preceding assertion, event, situation, etc., while ironic criticism can be used without such a specific reference point. Thus, ironic praise standardly requires an echo of or allusion to some explicit previous statement, while ironic criticism does not require such an anchor point; see also Dews and Winner (1999)Dews, Shelly, and Ellen Winner 1999 “Obligatory Processing of Literal and Nonliteral Meanings in Verbal Irony.” Journal of Pragmatics 31: 1579–1599. .

2.2Echoic mention and pretense in ironic utterances

Ironic utterances have been analyzed as instances of echoic language (see Wilson 2006Wilson, Deirdre 2006 “The Pragmatics of Verbal Irony: Echo or Pretence?” Lingua 116: 1722–1743. ; Wilson and Sperber 1992Wilson, Deirdre, and Dan Sperber 1992 “On Verbal Irony.” Lingua 87: 53–76. ), that is, as utterances that reproduce certain assumptions or statements. By using echoic language ironically, the speaker makes reference to states of affairs that are, based on mutually shared knowledge, expected or desired but that are not the case (Gibbs and Colston 2007Gibbs, Raymond W., and Herbert L. Colston 2007 Irony in Language and Thought. A Cognitive Science Reader. New York: Lawrence Erlbaum Associates. ). The echoic component of verbal irony is closely linked to the notion of mentioning an expression, i.e., using an expression metalinguistically. In their echoic mention theory, Sperber and Wilson (1981)Sperber, Dan, and Deirdre Wilson 1981 “Irony and the Use–Mention Distinction.” In Radical Pragmatics, ed. by Peter Cole, 295–318. New York: Academic Press. argue that ironic utterances are used to express an attitude towards the utterance itself, thus giving rise to an implicature aiming at substituting the expression’s literal meaning with the implicated meaning.

An ironic utterance can echo a specific previous remark or echo an assumption about a general norm. Consider the examples in (6), in which scare quotes are used around the respective ironically mentioned materials.

With the ironic utterance in (6a), presumably quoting the travel agent, the speaker echoes a previous reference to a certain denotatum as grand hotel for something that would better be described as Bed & Breakfast. Under this reading, (6a) echoes a specific remark. In contrast, an ironic utterance like (6b) quotes a general assumption about a norm, which, in this case, is the norm that the weather should be lovely when going on a picnic. In both cases, a negative attitude is expressed towards the corresponding denotata, that is, the accommodation facilities and the weather, respectively. We follow Wilson and Sperber (1992)Wilson, Deirdre, and Dan Sperber 1992 “On Verbal Irony.” Lingua 87: 53–76. here, who argue that the attitudinal meaning is produced as the speaker dissociates themselves from the thought they are echoing by delivering the echo in the form of mockery. Similarly, Wilson (2013Wilson, Deirdre 2013 “Irony Comprehension: A Developmental Perspective.” Journal of Pragmatics 59: 40–56. , 41) maintains that a speaker using verbal irony is “echoing a thought [that] she attributes to an individual, a group, or to people in general, and expressing a mocking, scornful or contemptuous attitude to this thought”.

Another element that has been identified to play a role in the comprehension of verbal irony is pretense. Clark and Gerrig (1984)Clark, Herbert H., and Richard J. Gerrig 1984 “On the Pretense Theory of Irony.” Journal of Experimental Psychology: General 113: 121–126. argue that with an ironic utterance the speaker pretends to be an uninformed person, that is, they act as if they were unaware of actual circumstances, with the intention for the addressee to see through this pretense. By using an utterance ironically, a speaker simulates a speech act and, thus, produces an attitudinal comment towards the corresponding assertion (Currie 2006Currie, Gregory 2006 “Why Irony is Pretence.” In The Architecture of the Imagination: New Essays on Pretence, Possibility, and Fiction, ed. by Shaun Nichols, 111–133. Oxford: Oxford University Press. ). In the examples in (6) above, such an analysis implies that the speaker acts as if they actually believed that the accommodation is a grand hotel and that the weather is lovely, respectively, and intend the addressee to identify the corresponding speech acts as pretense. An important element in Clark and Gerrig’s analysis is the assumption that speaker and addressee must share a certain common ground for an ironic utterance to be comprehended as such (Clark and Gerrig 1984Clark, Herbert H., and Richard J. Gerrig 1984 “On the Pretense Theory of Irony.” Journal of Experimental Psychology: General 113: 121–126. , 124).

Echoic mention and pretense analyses of verbal irony share certain characteristics (cf. Attardo 2000Attardo, Salvatore 2000 “Irony as Relevant Inappropriateness.” Journal of Pragmatics 32: 793–826. ; Wilson 2013Wilson, Deirdre 2013 “Irony Comprehension: A Developmental Perspective.” Journal of Pragmatics 59: 40–56. ), the component of mockery being one of them. Both echo and pretense can be seen as the basis of the mockery effect ironic utterances create. An utterance communicating a certain statement’s echo produced by a speaker evidently pretending the assertion to be true will give rise to the implicature that the utterance is intended to mock. Crucially, however, the two notions have been shown not to be identical; see, for example, Wilson (2006)Wilson, Deirdre 2006 “The Pragmatics of Verbal Irony: Echo or Pretence?” Lingua 116: 1722–1743. and Declercq (2017)Declercq, Dieter 2017 “A Philosophy of Satire: Critique, Entertainment, Therapy.” PhD Thesis. University of Kent. for analyses of cases of (ironic-)echoic language that do not involve pretense. We follow this view and assume that while ironic utterances are typically echoic and used as reminders of a thought, only a subset of ironic utterances involves pretenses of speech acts. This perspective is in line with integrated accounts that combine both echoic mention as well as pretense as complementary notions in their analysis of verbal irony (Kumon-Nakamura et al. 1995Kumon-Nakamura, Sachi, Sam Glucksberg, and Mary Brown 1995 “How about Another Piece of Pie: The Allusional Pretense Theory of Discourse Irony.” Journal of Experimental Psychology 124: 3–21. ; Popa-Wyatt 2014Popa-Wyatt, Mihaela 2014 “Pretence and Echo: Towards an Integrated Account of Verbal Irony.” International Review of Pragmatics 6: 127–168. ).

One such type of verbal irony that involves the pretense of a speech act, we assume, is ironically delivered surprise as instantiated, for example, in ironically uttered exclamations like What a pleasant day!. Surprise is a reaction to something that exceeds expectation. Pretended surprise can be seen as a means of exaggeration, which has been argued to have an amplifying effect that increases the gap between what is said and what is meant and, therefore, makes an utterance easier to be understood in a non-literal way (Colston and Keller 1998Colston, Herbert L., and Shaune B. Keller 1998 “You’ll Never Believe This: Irony and Hyperbole in Expressing Surprise.” Journal of Psycholinguistic Research 27: 499–513. ; Kreuz 2020Kreuz, Roger J. 2020 Irony and Sarcasm. Cambridge, MA: The MIT Press. ).

To sum up, we view ironic utterances as instances of echoic language used to produce a mockery effect. Certain ironic utterances, like ironic surprise, involve the pretense of a speech act. For an utterance to be identified as ironic, speaker and addressee must share a certain common ground.

3.1Mirativity in ironic utterances

Ironic utterances of the type in (7) and (8) are inclined to involve modal particles like ja (lit. ‘yes’) or aber (lit. ‘but’) in German. These noticeably assist in interpreting an utterance as ironic, which – ignoring intonational or non-verbal features here – is more difficult to construe in out-of-the-blue contexts if such an expression is not present. Consider the following examples.

Notice that our claim is not that sentences of this type cannot trigger ironic interpretations without the corresponding particles. Without them, however, the ironic interpretation is palpable to a lesser degree, particularly in the written mode. Utterances like (7) or (8) convey an expressive meaning and adopt an exclamative function. We claim that the expressivity is rooted in a mirative reading of these sentences, i.e., a reading that articulates the speaker’s surprise. Crucially, with ironically interpreted sentences of this sort, the speaker’s surprise is pretended. Thus, a speaker ironically uttering a sentence like (7) or (8) intends to simulate (positive) surprise and pretends that the assertion’s content exceeds their expectation.

We follow accounts that argue that particles expressing mirativity33.We stay agnostic regarding the question whether mirativity should be considered an independent linguistic category or not. See, e.g., Hengeveld and Olbertz (2012)Hengeveld, Kees, and Hella Olbertz 2012 “Mirativity Does Exist!” Linguistic Typology 16: 487–503. for discussion. do not trigger the surprise reading by virtue of their lexical meaning. Instead, we assume that these particles assist in the construal of surprise pragmatically; see Zimmermann (2011Zimmermann, Malte 2011 “Discourse Particles.” In Semantics: An International Handbook of Natural Language Meaning – Vol. 2, ed. by Klaus von Heusinger, Claudia Maienborn, and Paul Portner, 2011–2038. Berlin: Mouton de Gruyter., 2027). In its lexical, non-mirative meaning, ja, for example, marks a proposition p uttered at time t_u as true and uncontroversial. It links the utterance’s content to some pretext as a reference point and instructs the addressee to retrieve p, not currently being considered, from the shared common ground (Karagjosova 2003Karagjosova, Elena 2003 “Modal Particles and the Common Ground: Meaning and Functions of German ‘ja’, ‘doch’, ‘eben’/‘halt’ and ‘auch’.” In Perspectives on Dialogue in the New Millenium, ed. by Peter Kühnlein, Hannes Rieser, and Henk Zeevat, 335–349. Amsterdam: Benjamins. ; Repp 2013Repp, Sophie 2013 “Common Ground Management: Modal Particles, Illocutionary Negation and VERUM.” In Beyond Expressives: Explorations in Use-Conditional Meaning, ed. by Daniel Gutzmann, and Hans-Martin Gärtner, 231–274. Leiden: Brill.). Ja may also be used to introduce information that is new to the addressee. In this case, the state of affairs corresponding to p needs to be accessible to the addressee and be able to be confirmed by them (see Kraus 2018Kraus, Kelsey N. 2018 “Great Intonations.” PhD Thesis. UC Santa Cruz. as cited in Kroll and Roberts 2019Kroll, Margaret, and Tom Roberts 2019 “Stating the Obvious: ‘Of Course’ as a Focus-Sensitive Marker of Uncontroversiality.” Proceedings of Sinn und Bedeutung 23. 37–53., 45). Importantly, under certain speaker-related conditions, the common-ground update to be induced by the retrieval of p is expressed as exceeding expectation (Rett 2011Rett, Jessica 2011 “Exclamatives, Degrees and Speech Acts.” Linguistics and Philosophy 34: 411–442. ). In such cases, the particle assists in a surprise reading of the utterance.44.The surprise component is typically realized with exclamative intonation in oral language. We presume that the surprise readings of the utterances in question also arise in the written mode.

3.2Mirative particles and the background proposition

An ironic reading of a sentence gives rise to an alternative interpretation of what the proposition p expresses literally. Thus, ja, as used in (7) above, does not mark p as true at t _u . Instead, we assume, it is used to simulate that the proposition is true at t _u .

Both ja and aber require a background proposition to be accessible in the common ground (Fischer 2007Fischer, Kerstin 2007 “Grounding and Common Ground: Modal Particles and Their Translation Equivalents.” In Lexical Markers of Common Grounds, ed. by Anita Fetzer, and Kerstin Fischer, 47–66. Amsterdam: Elsevier.).55.This background proposition is not necessarily mentioned explicitly in the pretext; cf. Section 2.2. In contrast to ja, however, abe r has been argued to point to presupposed content contrary to what is expressed in the utterance at hand. According to Diewald and Fischer (1998)Diewald, Gabriele, and Kerstin Fischer 1998 “Zur diskursiven und modalen Funktion der Partikeln ‘aber’, ‘auch’, ‘doch’ und ‘ja’ in Instruktionsdialogen.” Linguistica 38: 75–99. , aber adds a meaning component that points to a disparity between the speaker’s expectation and the actual reality perceived by the speaker. In other words, the use of aber indicates that the speaker assumes a certain proposition to be given but that the current situation departs from that (Diewald and Fischer 1998Diewald, Gabriele, and Kerstin Fischer 1998 “Zur diskursiven und modalen Funktion der Partikeln ‘aber’, ‘auch’, ‘doch’ und ‘ja’ in Instruktionsdialogen.” Linguistica 38: 75–99. , 82; see also Abraham 2017Abraham, Werner 2017 “Modalpartikel und Mirativeffekte.” In Grammatische Funktionen aus Sicht der japanischen und deutschen Germanistik [Linguistische Berichte Sonderheft 24], ed. by Shin Tanaka, Elisabeth Leiss, Werner Abraham, and Yasuhiro Fujinawa, 76–108. Hamburg: Buske.).

Considering the non-ironic interpretation of the example in (8) above, this assumption implies that the speaker did not expect the apartment to be spacious, which, however, the speaker observes not to be the case, given the current situation. The interpretation supports a surprise reading of the sentence, typically realized as an exclamative.66.As it is not in the scope of the current study, we leave open the question of whether the surprise reading of the particle is restricted to certain sentence types. For an analysis of sentence type restrictions for modal particles as the result of an interaction between the particle and specific grammatical constructions, we refer the reader to Alm et al. (2018)Alm, Maria, Janina Behr, and Kerstin Fischer 2018 “Modal Particles and Sentence Type Restrictions: A Construction Grammar Perspective.” Glossa 3: 1–32. . With the ironic reading of the sentence, an alternative interpretation emerges, where the speaker, as we claim, communicates a staged feeling of surprise, thus implicating that they are observing the apartment to be the opposite of spacious. At the same time, the surprise is manifested by the speaker with the intention for the addressee to see through the pretense and to construe the utterance as a form of mockery.

3.3Mockery and evaluative attitude

The question is still open as to what content the mockery in ironic utterances containing mirative particles is based on. As stated above, both ja and aber in their relevant interpretations require a background proposition to be accessible in the common ground. The background proposition refers to states of affairs that are expected based on knowledge shared in the common ground by the interlocutors. Our claim is that it is this background proposition that is echoed in ironic utterances of the type in (7) or (8). Crucially, as the echo is delivered in the form of mockery, the speaker signals a dissociation from the content of the proposition, giving rise, thus, to the implicature of an evaluative attitude the speaker has towards the corresponding denotation.

(11) says that, with an ironic utterance u containing ja or aber, the corresponding utterance u reproduces proposition p as an echo, which gives rise to the implicature that speaker s expresses an evaluation of p.

3.4Hypothesized effect of mirative particles in ironic utterances

We view ironic utterances articulating the speaker’s surprise as instances of pretense. Modal particles (ja, aber) with a mirative interpretation assist in the reading of a staged (positive or negative) feeling of surprise and the utterance’s interpretation as ironic. In ironic utterances of this type (cf. the examples in (7) and (8) above), the speaker echoes a background proposition shared in the common ground. The echo is delivered in the form of mockery, which implicates the speaker’s evaluative attitude towards the expression’s denotation. Based on these assumptions, we can formulate hypotheses as to what effects modal particles have in ironic utterances. The empirical perspective pursued here implies that the attitudinality of verbal irony is a gradual feature and, thus, present to certain degrees in an utterance (cf. Härtl and Seeliger 2019Härtl, Holden, and Heiko Seeliger 2019 “Is a So-Called ‘Beach’ a Beach? An Empirically Based Analysis of Secondary Content Induced by Ironic Name Use.” In Secondary Content: The Linguistics of Side Issues, ed. by Daniel Gutzmann, and Katharina Turgay, 200–221. Leiden: Brill. ). The notion is motivated by experimental results as reported in Smith and Hall (2011)Smith, E. Allyn, and Kathleen Currie Hall 2011 “Projection Diversity: Experimental Evidence.” In Proceedings of the ESSLLI 2011 Workshop on Projective Content. 156–170., who found substantial heterogeneity among projective meanings (i.e., meanings like presuppositions that “project” through, for instance, negation or interrogation) with respect to their projective strengths.

Reflecting the perception of pretended surprise, we hypothesize ironic utterances containing a modal particle (ja or aber) to be perceived as more pretended than ironic utterances not containing a particle. No such difference is expected for non-ironic utterances.

Second, based on the assumption that aber but not ja points to presupposed content contrary to what is expressed in the utterance, a stronger pretense effect can be conjectured for ironic utterances containing aber as compared to ironic utterances containing ja.

We are also interested in the interplay between mirativity and the polarity of the ironic utterance, manifested as either ironic criticism or ironic praise.77.With the type of ironic utterance under discussion here, ironic criticism materializes as pretended positive surprise (cf. This is a spacious apartment!) and ironic praise as pretended negative surprise (cf. This is a tiny apartment!). Our hypothesis here is the null hypothesis (H_C). Last, we hypothesize a main effect for ironic utterances to be perceived as more pretended than non-ironic utterances (H_D). To test our hypotheses, we devised a rating study, which we will now present.

4.1Method

4.2Results

The statistical analysis was conducted using the Minitab^® software package (Minitab, LLC. 2020Minitab, L. L. C. 2020 Minitab 19 [computer software]. Retrieved from https://​www​.minitab​.com. Minitab 19). A repeated-measures variance analysis (General Linear Model) by subject was conducted for the dependent variable (Rating). The independent variables were included as fixed (within-subject) factors. The factor Subject was treated as random. Table 1 summarizes the mean ratings for each condition.

The analysis revealed a highly significant difference, F(1, 49) = 2838.95, p < .0001, for the mean ratings of ironic (IRO) and non-ironic (NON-IRO) utterances, i.e., ironic utterances were rated to be more pretended than non-ironic ones. The rating differences between all other relevant conditions are comparatively small, but not negligible. The analysis showed that ironic utterances containing a particle (PRT) were rated as more pretended (F(1, 49) = 7.06, p = .008) than the ones which did not include a particle (NO-PRT). In a planned post-hoc comparison (Fisher LSD), the latter effect was found to be significant only for ironic utterances (p < .05) but not for non-ironic utterances. However, the interaction between utterance type (IRO / NON-IRO) × particle (PRT / NO-PRT) failed to reach statistical significance. Importantly, only ironic utterances (IRO) containing the particle aber (ABER) were rated as more pretended (F(1, 49) = 10.04, p = .002) in comparison to ironic utterances (IRO) without a particle (NO-PRT). No such difference was found for ironic utterances containing ja (JA) in comparison to ironic utterances without a particle. Last, the difference between ironic criticism (NEG) and ironic praise (POS) was not significant.

4.3Discussion

The results clearly reveal that ironic utterances were perceived as more pretended than non-ironic utterances. This leads us to accept hypothesis H_D. Furthermore, the results lead us to accept hypothesis H_A. As ironic utterances containing a modal particle were indeed perceived as more pretended than the ones without, it can be assumed that modal particles do play a role in enhancing the discrepancy between two propositions. Especially the modal particle aber, in confirmation of our hypothesis H_B, seems to enhance this effect, due to its primary function of signaling a discrepancy between two propositions. The modal particle ja, however, primarily reminds the hearer of an accessible proposition instead of contrasting it, which we assume to be the reason why there was no effect for ja. We interpret these results as support for the assumption that ironic utterances of the type under discussion require a background proposition contrary to what the utterance asserts and that aber is better suited to simulate surprise about something that is staged as exceeding expectation. Additionally, the results showed no significant difference between ironic criticism and ironic praise. Even though ironic criticism might be more common than ironic praise (cf. Section 2.1 above), the perception of pretense associated with either seems to be identical.