Respectus Philologicus eISSN 2335-2388
2022, no. 41 (46), pp. 11–25 DOI: http://dx.doi.org/10.15388/RESPECTUS.2022.41.46.104

Insulting Rhetorical Questions – Mitigators or Amplifiers?

Džemal Špago
Dzemal Bijedic University of Mostar, Faculty of Humanities
Vrapčići 76, 88000 Mostar, Bosnia-Herzegovina
Email: dzemal.spago@unmo.ba
ORCID iD: https://orcid.org/0000-0001-9104-3994
Research interests: pragmatics, semantics

Abstract. The paper examines whether rhetorical questions (RQs) with insulting content or implications soften or intensify the insulting content that they express, as compared to corresponding direct statements with similar insulting content. The analysis is based on the results of two online surveys conducted among 276 Bosnian university students (182 and 94, respectively), who evaluated, in regard to their offensiveness, two sets of RQs and corresponding statements with insulting content or implications. Three types of insulting RQs were included in the surveys: insulting RQs without explicitly offensive terms, insulting RQs that incorporate derogatory words, and sarcastic RQs with insulting implications. The expected results were that: a) in line with Frank’s (1990) account of strengthening effects of RQs as their primary function, insulting RQs, with or without derogatory words, will function as amplifiers, and sound more offensive than corresponding declaratives; and b) sarcastic RQs, following Dews and Winner’s (1995) account of softening effects of sarcastic utterances, will function as mitigators, as compared to non-sarcastic declaratives with insulting content. The obtained results indicate that the first hypothesis cannot be verified (in spite of some indications that slight amplifying effects do exist), and the second hypothesis is completely rejected, with some likelihood that the opposite could be true.

Keywords: insulting rhetorical questions; insulting statements; derogatory words; sarcastic rhetorical questions; mitigators; amplifiers.

Submitted 08 July 2021 / Accepted 12 December 2021
Įteikta 2021 07 08 / Priimta 2021 12 12
Copyright © 2022 Džemal Špago. Published by Vilnius University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution License CC BY 4.0, which permits unrestricted use, distribution, and reproduction in any medium provided the original author and source are credited.

Introduction

Rhetorical questions (henceforth, RQs) represent a highly effective stylistic device characterized by function-form dichotomy and the ability to perform multiple communicative functions¹ (Schaffer, 2005; Ilie, 1994), both in friendly and aggressive communication (Špago, 2020). Whether viewed as indirect assertions (Han, 2002), questions with a constrained set of possible answers (Van Rooy, 2003), redundant interrogatives (Rohde, 2006), questions which point at an already known fact (Caponigro and Sprouse, 2007), or in some other way, one of the main features of RQs is that they are not posed in an attempt to elicit a verbal answer², but rather to achieve a number of other communicative goals, often those associated with persuasion (Frank, 1990). While in most cases, they share the same form with information-seeking questions, which makes context the crucial indicator of the rhetorical or non-rhetorical nature of questions, some RQs can contain elements (for instance, NPIs)³ that facilitate their rhetorical interpretation.

An interesting quality of RQs is that they can, depending on the context, be used both to strengthen and soften the content that they express (Frank, 1990; Ilie, 1994; Moshavi, 2009). Among a wide range of different uses of such questions, one that has received little, if any, attention so far is to express content that is (or may be interpreted as) insulting to the addressee or a third party.

While insults do not necessarily need to be used with hostile intentions⁴, they are most commonly defined as expressions that are used in an attempt to put down, humiliate or hurt the target (Mateo and Yus, 2013; Gabriel, 1998). Whether they appear in an innovative or conventional form, their interpretation and implications are highly context and culture-dependent (Mateo and Yus, 2013), and their perceived level of offensiveness often varies (Kremin, 2017).

The aim of this study is to explore, based on the results of two online surveys conducted among Bosnian university students, whether the offensive content is intensified or softened (and, if yes, to what extent) when expressed in the form of RQs, as well as whether different types of RQs (such as those that contain sarcasm or derogatory words) are perceived as less or more offensive than outright statements with similar insulting content.

1. Background

RQs have attracted much attention from researchers over the past few decades, as numerous studies focusing on different aspects of these unconventional questions have been done. In regard to the nature of RQs and/or the answers which they imply, several approaches have been proposed:

• RQs are (semantically) indirect statements in the form of questions, whose polarity is opposite to that of the questions⁵ (Han, 2002),

• RQs are essentially information-seeking questions, but all possible answers to them have more or less similar implications⁶ (Van Rooy, 2003),

• RQs are neither information-seeking nor information-providing, but redundant interrogatives whose purpose is to confirm the interlocutors’ “shared beliefs about the world” (Rohde, 2006, p. 135),

• RQs are semantically and syntactically the same as information-seeking questions but differ from them at a pragmatic level⁷ (Caponigro and Sprouse, 2007), etc.

Other studies explored the use of RQs as a response to information-seeking questions (Schaffer, 2005), or other communicative functions of RQs (Ilie, 1994; Frank, 1990), as well as the use of sarcasm in RQs (Oraby et al., 2016; Oraby et al., 2017), and the prosody of RQs as opposed to that of information-seeking questions (Dehé and Braun, 2020; Braun et al., 2018; and others).

One aspect of RQs that is particularly of interest to us in this study is their communicative function related to strengthening or weakening the content they express. In their account of politeness strategies, Brown and Levinson (1987) list RQs as one of the off-record strategies which may be used to perform FTAs⁸ and soften excuses or criticism. On the other hand, Ilie (1994, p. 128) claims that RQs can function “as amplifiers or as mitigators, by emphasizing or toning down the addresser’s opinions, beliefs, assumptions”, which she later supports by the findings of her analysis of the use of RQs in courtroom discourse (p. 213). Frank (1990) adopts an approach which allows for a dual function of RQs (while they can be used to minimize face threats, and thereby have softening effects, they can also “enable speakers to make stronger statements, with greater implications, than would be possible if they had made straightforward assertions” (Frank, 1990, p. 726)), but concludes, based on the analysis of a set of spontaneous conversations, that the latter is the primary function of such questions.⁹ A number of studies of the use of RQs in the Bible also showed that their use is sometimes associated with strengthening statements, and, at other times, RQs serve as “a courteous means of issuing a corrective or criticism” (Moshavi, 2009, pp. 33–34).

While RQs with insulting content have been mentioned occasionally (for instance, Mateo and Yus (2013) list examples of such questions as forms of innovative insults¹⁰), there have been no studies that focused on this type of RQs yet.

Previous studies provided conflicting accounts of whether sarcastic utterances are more or less hurtful than corresponding non-sarcastic utterances. On the one hand, according to Dews and Winner’s (1995) tinge hypothesis, sarcastic utterances are less hurtful than literal (non-sarcastic) statements, as positive words, even when used ironically, “tinge” the negative implications of the utterance, and make it less hurtful (hence, the tinge hypothesis). Conversely, some other studies (for instance, Leggitt, Gibbs, 2000; Toplak, Katz, 2000) claim the opposite – sarcastic expressions sound more hurtful than literal (non-sarcastic) ones, as they intensify the hostile attitude¹¹.

2. Methodology

Two online surveys were conducted among 276 university students from different undergraduate study programs at Dzemal Bijedic University of Mostar, Bosnia-Herzegovina. The first survey, taken by 182 respondents¹², included 10 RQs with insulting content paired with corresponding statements with similar insulting content (the examples were not contextualized). The respondents were asked to assess the RQ and the outright statement in each pair in terms of their offensiveness, assuming that they were used in the same context and with the intention to insult and to mark either the RQ or the statement as less insulting, or, alternatively, both of them as equally insulting. The second survey aimed to strengthen and validate the findings of the first one and explore whether the assessment of contextualized examples would yield different results.

Three types of RQs with insulting content were included in the survey¹³:

• RQs with insulting content or implications, but without derogatory words (Ex.1:Why are you talking, who asked you anything? vs Stop talking, nobody asked you anything. Ex.7: And what are you like, why don’t you look at yourself first? vs You’re not any better, you should look at yourself first. Ex.8: What do you know about anything, man? vs You don’t know anything, man. Ex.11: Who cares what you think?! vs I don’t care what you think! Ex.13: Can you do anything right?! vs You can’t do anything right! Ex.17: Are you in your right mind?! vs You’re not in your right mind!)

• insulting RQs which include a derogatory word¹⁴ (Ex.2: What kind of fool would believe you? vs Only a fool would believe you. Ex.4: Am I going to listen to a jerk like you? vs I am not going to listen to a jerk like you. Ex.5: Are you really so stupid or what? vs You’re really stupid. Ex.10: Who would ever believe a liar like you? vs I never believe a liar like you. Ex.14: Who allows such incompetents to drive a car?! vs Such incompetents shouldn’t be allowed to drive a car! Ex.16: Am I supposed to be an ass-kisser like you?! vs I don’t want to be an ass-kisser like you! Ex.18: What kind of fool gave you a driver’s license?! vs The one who gave you a driver’s license is a fool! Ex.20: Are you really so insolent?! vs You’re so insolent!);

• sarcastic RQs with insulting implications¹⁵ (Ex.3: And who are you, some honest guys, I guess? vs You’re corrupted, too. Ex.6: And how do you look like, maybe like a model, like a beauty, right? vs You don’t look attractive, either. Ex.9: And what are you, maybe some kind of expert on that? vs You don’t know anything about that. Ex.12: When did you get so smart? vs Keep your advice to yourself! Ex.15: And what are you going to do in the meantime, watch a movie? vs You can clean it yourself, you’re not doing anything! Ex.19: So, what am I supposed to do, give you a medal for that? vs If you did it for me, don’t mention it all the time!).

Taking into consideration the fact that insulting RQs represent the harshest form of aggressive RQs, and that, accordingly, the speaker’s intention is not to pay heed to the addressee’s “face“ needs when using such questions, I predict, in line with Frank’s (1990) view that the primary function of RQs is to strengthen statements, that insulting RQs, with or without derogatory words, should act as amplifiers. Therefore, the following hypothesis will be tested in regard to non-sarcastic insulting RQs:

1) RQs with insulting content will function as amplifiers, i.e. they will sound more offensive than corresponding insulting statements.

Regarding sarcastic RQs with insulting implications vs non-sarcastic statements with insulting content, the following hypothesis, in line with Dews and Winner (1995), will be tested in this study:

2) Sarcastic RQs will function as mitigators, i.e. they will sound less offensive than corresponding non-sarcastic statements with insulting content.

3. Results and discussion

This section is divided into three subsections: the findings related to the first hypothesis are presented and analysed in subsections 4.1. (insulting RQs which do not include derogatory terms) and 4.2. (insulting RQs which incorporate derogatory terms), while the data related to the second hypothesis, regarding sarcastic RQs, is explored in subsection 4.3.

3.1. Insulting RQs without derogatory words

The respondents from the first survey evaluated three pairs of RQs and corresponding direct statements with insulting content which do not incorporate derogatory words (the offensiveness of the RQs/statements is based on the denigrating implications of the content – the addressee should keep quiet; the addressee is not a good person; the addressee doesn’t know anything). The pattern which has been noticed in responses to different examples of such RQ-direct statement pairs is characterized by three things:

a) there is a divided opinion regarding the harshness of such RQs and corresponding statements, as all three potential answers (the RQ is less insulting / the statement is less insulting / they are equally insulting) received some attention;

b) the respondents who understand the direct statement in each pair as less insulting outnumber those who find the RQ as less insulting, which can indicate that such RQs tend to strengthen the insulting content;

c) in each pair, the number of those who interpret the RQ as either less or equally offensive as the corresponding statement is higher than the number of those who find statements as less insulting, which reveals that amplifying effects of insulting RQs are not obvious.

In two out of the three pairs, the most common answer was that the RQ and the statement are equally insulting, and in one pair, the most frequently chosen option was that the statement is less insulting.¹⁶ Additionally, in one example, there was a predominant response (the RQ and statement are equally insulting) provided by more than two-thirds of the survey participants, as shown in Table 1:

Table 1. RQs with insulting implications compared to corresponding insulting statements (no derogatory words being used) – first survey (182 respondents)

RQ-statement pair	The RQ is less insulting	The statement is less insulting	They are equally insulting
Ex.1: Why are you talking, who asked you anything? vs Stop talking, nobody asked you anything.	14 (7.6%)	46 (25.2%)	122 (67%)
Ex.7: And what are you like, why don’t you look at yourself first? vs You‘re not any better, you should look at yourself first.	49 (26.9%)	74 (40.6%)	59 (32.4%)
Ex.8: What do you know about anything, man? vs You don‘t know anything, man.	31 (17%)	70 (38.4%)	81 (44.5%)

Although no striking differences between the responses of the male and female respondents have been noticed, an interesting finding is that the female respondents were more likely to disregard potential differences between such RQs and corresponding statements in terms of their harshness, and the male respondents were more likely to recognize amplifying effects of the RQs, as shown in Table 2:

Table 2. RQs with insulting implications compared to corresponding insulting statements (no derogatory words being used) – comparing male and female responses from the first survey (182 respondents)

RQ-statement pair	The RQ is less insulting M vs F		The statement is less insulting M vs F		They are equally insulting M vs F
Why are you talking, who asked you anything? vs Stop talking, nobody asked you anything.	10 (11%)	4 (5%)	30 (32%)	16 (18%)	54 (57%)	68 (77%)
And what are you like, why don’t you look at yourself first? vs You‘re not any better, you should look at yourself first.	28 (30%)	21 (24%)	41 (44%)	33 (38%)	25 (27%)	34 (39%)
What do you know about anything, man? vs You don‘t know anything, man.	21 (22%)	10 (11%)	40 (43%)	30 (34%)	33 (35%)	48 (55%)

While the results from the first survey hint at the possibility that RQs with insulting implications are more likely to be interpreted as harsher, rather than milder, when compared to outright statements with similar content, on average close to two-thirds of the respondents (over 65%) did not recognize the provided examples of statements as less insulting if used in the same context as the corresponding RQs. There was a number of points that needed to be further clarified with the second survey: whether some RQ-statement pairs from the second survey will be predominantly marked as equally offensive (as was the case with Ex.1); whether the evaluation of contextualized examples would significantly affect the results; and, finally, if the noted differences in responses of male and female respondents from the first survey were just a coincidence.

The results obtained from the second survey were in line with the pattern observed in the first one (divided opinions; more responses in favour of amplifying rather than mitigating effects of RQs; yet, the majority of the respondents did not recognize the provided RQs as amplifiers). In all three assessed pairs, the most common answer (in two of them chosen by more than half of the respondents) was that the RQs and statements are equally offensive, as shown in Table 3.

It can be concluded that the assessment of contextualized RQ-statement pairs from the second survey did not yield significantly different results, but there was less fluctuation in responses to different examples. The findings of the second survey further strengthened the view that amplifying effects of this kind of insulting RQs are not apparent to about two-thirds of the respondents. In regard to the above-mentioned gender differences, the second survey also indicated the possibility that women are more likely to disregard differences between RQs and statements in terms of their strength. However, a limitation of the second study is the fact that only 24 male respondents participated in it, so further research on this is needed.

Table 3. RQs with insulting implications compared to corresponding insulting statements (no derogatory words being used) – second survey (94 respondents)

RQ-statement pair	The RQ is less insulting	The statement is less insulting	They are equally insulting
Ex.11: Who cares what you think?! vs I don’t care what you think!*	3 (3.2%)	36 (38.2%)	55 (58.5%)
Ex.13: Can you do anything right?! vs You can’t do anything right!**	19 (20.2%)	26 (27.6%)	49 (52.1%)
Ex.17: Are you in your right mind?! vs You’re not in your right mind!***	24 (25.5%)	30 (31.9%)	40 (42.5%)

* Context: You are talking to your friend about her new boyfriend (you are genuinely worried): „I think you should be careful with this guy.” She gets offended and responds angrily.
** Context: An angry mom is shouting at her child who just spilled milk on the floor.
*** Context: A man is starting to cross the street without noticing an approaching car. The driver slams on the brakes and yells at the man angrily.

Table 4. RQs with insulting implications compared to corresponding insulting statements (no derogatory words being used) – comparing male and female responses from the second survey (94 respondents)

RQ-statement pair	The RQ is less insulting M vs F		The statement is less insulting M vs F		They are equally insulting M vs F
Who cares what you think?! vs I don‘t care what you think!	1 (4%)	2 (3%)	10 (42%)	26 (37%)	13 (54%)	42 (60%)
Can you do anything right?! vs You can‘t do anything right!	8 (33%)	11 (16%)	7 (29%)	19 (27%)	9 (38%)	40 (57%)
Are you in your right mind?! vs You‘re not in your right mind!	5 (21%)	19 (27%)	9 (38%)	21 (30%)	10 (42%)	30 (43%)

Based on the responses of the two groups of respondents (182 and 94, respectively) to six RQ-statement pairs with insulting implications, both contextualized and non-contextualized, it can be concluded that, on average, only 33.6% of the respondents recognize amplifying effects of this type of insulting RQs. In five cases, the most common answer was that the RQs and statements sound equally offensive (on average, 49.5% of the respondents did not notice any difference between RQs and statements from this section when it comes to their strength). All of this leads to the conclusion that, despite Frank’s (1990) account of strengthening effects of RQs as their primary function, amplifying effects of insulting RQs cannot be verified in such instances. A reason for that could be that, in such examples, recognizing the speaker’s intention to insult is crucial in determining the addressee’s understanding of potentially insulting utterances, and it makes little difference whether they are worded as RQs or outright statements.

On the other hand, in each of the six RQ-statement pairs from this section, the number of those who recognize amplifying effects of RQs exceeds the number of those who see them as mitigators (by the average margin of 16.9%), which indicates that slight amplifying effects could exist.

3.2. Insulting RQs which include a derogatory word

The respondents from the first survey assessed four RQ-statement pairs with insulting content, which include derogatory words. In two cases, the derogatory words (jerk and liar) were used as a presupposition in reference to the addressee, and the majority of the respondents (over 64%) evaluated those RQs and the corresponding statements as equally insulting. This finding counters the hypothesis that insulting RQs should act as amplifiers. A reason for this could be that derogatory words directed at addressees determine the harshness of such utterances, and the form in which the insult is realized (an RQ or a statement) becomes more or less irrelevant in regard to its harshness.

In another RQ-statement pair, a derogatory word (fool) was used in reference to a third party (only a fool would believe the addressee), and, according to the obtained results, it still makes little difference whether the utterance is worded as an RQ or a statement. A potential explanation for this could be that, in such instances, another derogatory term in reference to the addressee is implied (if only a fool would believe the addressee, then the addressee must be a liar), which erases differences between an insulting RQ and a corresponding statement in terms of the degree of their offensiveness.

As for the fourth pair from this section, in which a derogatory word (stupid) was used in reference to the addressee, but was not formulated as a presupposition, provided answers indicate that the RQ has slightly mitigating effects noted by 37% of the respondents. The explanation could be that the RQ is not as definitive as the statement assigning the negative quality to the addressee. Table 5 presents the collected responses which relate to these insulting RQ-statement pairs:

Table 5. Insulting RQs with derogatory words compared to corresponding insulting statements with the same derogatory words – the first survey (182 respondents)

RQ-statement pair	The RQ is less insulting	The statement is less insulting	They are equally insulting
Ex.4: Am I going to listen to a jerk like you? vs I am not going to listen to a jerk like you.	34 (18.6%)	31 (17%)	117 (64.2%)
Ex.10: Who would ever believe a liar like you? vs I never believe a liar like you.	22 (12%)	37 (20.3%)	123 (67.5%)
Ex.2: What kind of fool would believe you? vs Only a fool would believe you.	45 (24.7%)	43 (23.6%)	94 (51.6%)
Ex.5: Are you really so stupid or what? vs You‘re really stupid.	67 (36.8%)	39 (21.4%)	76 (41.7%)

No significant differences have been spotted between the responses of the male and female respondents in this section.

The results from the first survey did not support the first hypothesis in regard to insulting RQs which include a derogatory word used as a presupposition in reference to the addressee, as almost no differences have been noted between such RQs and corresponding statements with insulting content. Points that needed to be further clarified or validated by the second survey included the following: whether the assessment of contextualized examples would yield similar results; whether additional examples of RQs which include a derogatory word used in reference to a third party (such as Ex.2) and those referring to the addressee without being used a presupposition (such as Ex.5) would be assessed in a similar way, in regard to their offensiveness.

The respondents from the second survey evaluated four contextualized examples from this category. In two of them, derogatory words (incompetent and ass-kisser) were used as a presupposition in reference to the addressee. In one case (Ex. 16), the assessment was completely in line with the results from the first survey, with close to two-thirds of responses marking the RQ and statement as equally insulting. In another one (Ex. 14), the margin was not as high, most likely because it was the only example in which the addressee was not explicitly linked to the derogatory word (such incompetents vs a jerk like you – Ex. 4; a liar like you – Ex.10; an ass-kisser like you – Ex.16). However, more than half of the respondents did not notice any difference between the RQ and the corresponding statement.

The assessment of an additional example of an RQ with a derogatory word (fool) referring to a third party (Ex.18) shows that it behaves similar to RQs with insulting implications: while slight amplifying effects of the RQ seem to be present, the majority of the respondents (over 60%) do not recognize them.

Table 6. Insulting RQs with derogatory words compared to corresponding insulting statements with the same derogatory words – the second survey (94 respondents)

RQ-statement pair	The RQ is less insulting	The statement is less insulting	They are equally insulting
Ex.14: Who allows such incompetents to drive a car?! vs Such incompetents shouldn’t be allowed to drive a car!*	15 (15.9%)	31 (32.9%)	48 (51%)
Ex.16: Am I supposed to be an ass-kisser like you?! vs I don’t want to be an ass-kisser like you!**	9 (9.5%)	24 (25.5%)	61 (64.8%)
Ex.18: What kind of fool gave you a driver’s license?! vs The one who gave you a driver’s license is a fool!***	14 (14.8%)	35 (37.2%)	45 (47.8%)
Ex.20: Are you really so insolent?! vs You’re so insolent!****	36 (38.2%)	22 (23.4%)	36 (38.2%)

* Context: Following a minor car accident, one angry driver is talking to the other, who caused it.
** Context: After an employee had an argument with his boss, his colleague is trying to tell him that he shouldn’t have argued with him. He responds angrily.
*** Context: the same as in Ex.17 – this time the pedestrian is shouting back at the driver.
**** Context: Parents are upset by their teen’s disrespectful behavior.

Conversely, slight mitigating effects, in line with the results of the first survey, have been noted in the example in which a derogatory word (insolent) used in reference to the addressee was not formulated as a presupposition. The complete results from the second survey are shown in Table 6.

The evaluation of the two sets (contextualized and non-contextualized) of insulting RQ-statement pairs which include derogatory words, shows that there is little difference between an insulting RQ and a corresponding statement in terms of their strength if they both include the same derogatory words used as presuppositions in reference to the addressee (in three out of four such examples, more than 64% of the respondents marked them as equally insulting). Nevertheless, just as it was the case with the RQs from the preceding section, such RQs seem to be slightly more inclined towards the amplifying rather than mitigating side, as in three out of four assessed examples, there were more responses marking statements as less insulting.

In cases where a derogatory word is used in reference to a third party, although the results from the two surveys were not identical (slight amplifying effects of the RQ seem to be present only in the second survey), we can still conclude that no apparent differences between such RQs and statements can be verified. Finally, in cases where a derogatory word used in reference to the addressee is not expressed as a presupposition, both surveys indicate that such RQs are slightly more inclined towards the mitigating interpretation.

3.3. Sarcastic RQs with insulting implications

The respondents from the first survey evaluated the offensiveness of three sarcastic RQs with insulting implications paired with corresponding non-sarcastic statements with insulting content. In two cases, the most common answer was that the statement is less insulting (in one of them, it was chosen by almost 60% of the respondents), which hints at strengthening effects of sarcastic RQs when used to insult someone:

Table 7. Sarcastic RQs with insulting implications compared with corresponding non-sarcastic statements with insulting content – the first survey (182 respondents)

RQ-statement pair	The RQ is less insulting	The statement is less insulting	They are equally insulting
Ex.3: And who are you, some honest guys, I guess? vs You‘re corrupted, too.	44 (24.1%)	109 (59.8%)	29 (15.9%)
Ex.6: And how do you look like, maybe like a model, like a beauty, right? vs You don‘t look attractive, either.	42 (23%)	64 (35.1%)	76 (41.7%)
Ex.9: And what are you, maybe some kind of expert on that? vs You don‘t know anything about that.	56 (30.7%)	82 (45%)	44 (24.1%)

No significant differences between the responses of the male and female respondents have been noticed.

Although there was much fluctuation in responses to different questions from this subsection, it is evident that the results from the first survey reject the view of sarcastic RQs as mitigators. In spite of the finding that more than half of the respondents have not recognized the amplifying effects of sarcastic RQs, the fact that in two out of three examples, the most common answer was that insulting statements are less offensive indicates that it is very likely that sarcastic RQs intensify negative implications of insulting content.

The results from the second survey were even more varied than those from the first one. In one case (Ex.12), the results indicate slight amplifying effects of the sarcastic RQ, and in another one (Ex.19), more than two-thirds of the respondents recognized the RQ as an amplifier. The most surprising of the obtained results is related to Ex.15, which is the only example in which a sarcastic RQ displayed some mitigating effects:

Table 8. Sarcastic RQs with insulting implications compared with corresponding non-sarcastic statements with insulting content – the second survey (94 respondents)

RQ-statement pair	The RQ is less insulting	The statement is less insulting	They are equally insulting
Ex.12: When did you get so smart? vs Keep your advice to yourself!*	30 (31.9%)	37 (39.3%)	27 (28.7%)
Ex.15: And what are you going to do in the meantime, watch a movie? vs You can clean it yourself, you’re not doing anything!**	45 (47.8%)	32 (34%)	17 (18%)
Ex.19: So, what am I supposed to do, give you a medal for that? vs If you did it for me, don’t mention it all the time!***	6 (6.3%)	63 (67%)	25 (26.5%)

* Context: You just tried to give your friend some advice – that he should study more and go out less, but he gets offended, and responds angrily.
** Context: You just told your sister/brother to clean the apartment, and (s)he gives you the angry response.
*** Context: Parents are angry at their disrespectful teenage son. They are telling him that they do not deserve this, as they did so much for him. The teen responds with the RQ/statement.

Although it is difficult to give a clear account of why there is so much difference in responses to different RQ-statement pairs from this subsection (especially Ex.15 vs Ex.19), a potential explanation could be linked to the pretended answer provided by the speaker. Namely, this answer sounds much more sarcastic in Ex.19, as it relates to something impossible (a teen giving his parents a medal for doing so much for him), whereas in Ex.15 it sounds much more realistic (your brother/sister watching a movie instead of cleaning the apartment). Another possible reason could be related to different roles of interlocutors, which might affect the interpretation of the strength of sarcastic RQs with insulting implications (a disrespectful teen being sarcastic with his parents vs one angry brother/sister being sarcastic with another).

While the evaluation of the six sarcastic RQ / insulting declarative pairs, contextualized and non-contextualized, did not yield uniform results, it can still be concluded that the second hypothesis does not hold, as only in one case some mitigating effects have been noted. On the other hand, in four examples, the most common response (in two of them selected by the majority of the respondents) was that sarcastic RQs act as amplifiers, which points to potential amplifying effects of sarcastic RQs with insulting implications.

However, further research on this is needed, especially considering the fact that responses from this subsection were highly varied and that the outright statements, which have been paired with sarcastic RQs, did not contain harsh or derogatory words, which could have affected their interpretation, and, possibly, could have elicited different answers from the respondents.

Conclusion

The present study examined potential softening or intensifying effects of insulting RQs compared to corresponding insulting statements. The research was motivated by previous accounts (Frank, 1990; Ilie, 1994; etc.), which found that RQs can sometimes serve as mitigators and, at other times, as amplifiers, depending on the intention of the speaker. Based on the results from two surveys taken by 276 Bosnian university students, two hypotheses regarding the harshness of RQs with insulting content or implications have been tested: compared to corresponding insulting declaratives, insulting RQs, whether with or without explicitly offensive terms, will act as amplifiers, and sarcastic RQs with insulting connotations will act as mitigators.

Based on the obtained results, we can conclude that the first hypothesis has not been verified. Namely, despite the finding that more respondents recognized strengthening, rather than softening, effects of RQs in cases where the insulting content is implied or stated without the use of derogatory words or expressions, the fact that about two-thirds of the respondents failed to recognize such RQs as amplifiers shows that the hypothesis cannot be confirmed when it comes to such RQs with insulting content. In the case of insulting RQs that incorporate derogatory terms used as presuppositions in reference to the addressee, there is almost no difference between RQs and statements in terms of their offensiveness, which indicates that, in such instances, derogatory words mostly determine the harshness of insulting content. The form of such utterances becomes irrelevant. However, suppose derogatory terms are not used as presuppositions in reference to the addressee or used in reference to a third party. In that case, then the results hint at slight mitigating or amplifying effects, respectively, of insulting RQs when compared to corresponding declaratives.

Regarding the second hypothesis (that sarcastic RQs with insulting implications will have softening effects when compared to outright non-sarcastic statements with insulting content), although the results from both surveys were surprisingly varied, it is still evident that the hypothesis does not hold, as some mitigating effects have been noted only in one out of the six evaluated examples. What is more, the finding that in four examples, some, or even strong, amplifying effects of sarcastic RQs have been identified indicates that such RQs could act as amplifiers.

The overall results indicate that, when it comes to expressing insulting content, although slightly more inclined towards the amplifying rather than mitigating interpretation, non-sarcastic RQs have only a minor influence on the harshness of such content. When insulting content is expressed in the form of sarcastic RQs, the results indicate that such RQs could act as amplifiers rather than mitigators. However, further research is needed on the potential amplifying effects of sarcastic RQs, compared to insulting declaratives, especially in light of varied responses to different examples of sarcastic RQs evaluated in this study.

References

Bianchi, C., 2014. Slurs and appropriation: An echoic account. Journal of Pragmatics, 66, pp. 35–44. https://doi.org/10.1016/j.pragma.2014.02.009.

Braun, B., Dehé, N., Neitsch, J., Wochner, D., and Zahner, K., 2018. The prosody of rhetorical and information seeking questions in German. Language and Speech, 62 (4), pp. 779–807. https://doi.or/0.117/023830918816351.

Brown, P., Levinson, S., 1987. Politeness. Some Universals in Language Usage. Cambridge: Cambridge University Press.

Caponigro, I. and Sprouse, J., 2007. Rhetorical questions as questions. In: Puig-Waldmüller, E. ed. Proceedings of Sinn und Bedeutung, 11, pp. 121–133. Available at: <https://ojs.ub.uni-konstanz.de/sub/index.php/sub/article/view/635/570> [Accessed 1 July 2021]. https://doi.org/10.18148/sub/2007.v11i0.635.

Dehé, N., Braun, B., 2020. The prosody of rhetorical questions in English. English Language and Linguistics, 24 (4), pp. 607–635. https://doi.org/10.1017/S1360674319000157.

Dews, S., and Winner, E., 1995. Muting the meaning: A social function of irony. Metaphor and Symbolic Activity, 11 (1), pp. 3–19. https://doi.org/10.1207/s15327868ms1001_2.

Filik, R., Turcan, A., Thompson, D., Harvey, N., Davies, H., Turner, A., 2016. Sarcasm and emoticons: Comprehension and emotional impact. The Quarterly Journal of Experimental Psychology, 69 (11), pp. 1–42. Available at: <https://www.researchgate.net/publication/283323798_Sarcasm_and_Emoticons_Comprehension_and_Emotional_Impact> [Accessed 1 July 2021]. https://doi.org/10.1080/17470218.2015.1106566.

Frank, J., 1990. You call that a rhetorical question? Forms and functions of rhetorical questions in conversation. Journal of Pragmatics, 14, pp. 723–738. https://doi.org/10.1016/0378-2166(90)90003-V.

Gabriel, Y., 1998. An introduction to the social psychology of insults in organizations. Human Relations, 51 (11), pp. 1329–1354. Available at: <https://www.researchgate.net/publication/240276297_An_Introduction_to_the_Social_Psychology_of_Insults_in_Organizations> [Accessed 1 July 2021]. https://doi.org/10.1023/A:1016946332565.

Han, C., 2002. Interpreting interrogatives as rhetorical questions. Lingua, 112, pp. 201–229. Available at: <https://www.sciencedirect.com/science/article/pii/S0024384101000444?via%3Dihub> [Accessed 1 July 2021]. https://doi.org/10.1016/S0024-3841(01)00044-4.

Ilie, C., 1994. What Else Can I Tell You? A Pragmatic Study of English Rhetorical Questions as Discursive and Argumentative Acts. Stockholm: Almqvist & Wiksell International.

Kremin, L. V., 2017. Sexist swearing and slurs: Responses to gender-directed insults. LingUU, 1 (1), pp. 18–25. Available at: <https://linguujournal.nl/wp-content/uploads/2017/10/Kremin-2017-Sexist-swearing-and-slurs.pdf> [Accessed 1 July 2021].

Leggitt, J. S., Gibbs, R. W., 2000. Emotional reactions to verbal irony. Discourse Processes, 29 (1), pp. 1–24. https://doi.org/10.1207/S15326950dp2901_1.

Mateo, J., Yus, F., 2013. Towards a cross-cultural pragmatic taxonomy of insults. Journal of Language Aggression and Conflict, 1 (1), pp. 87–114. Available at: <https://www.researchgate.net/publication/245539008_Towards_a_cross-cultural_pragmatic_taxonomy_of_insults> [Accessed 1 July 2021]. https://doi.org/10.1075/jlac.1.1.05mat.

Moshavi, A., 2009. Two types of argumentation involving rhetorical questions in Biblical Hebrew dialogue. Biblica, 90 (1), pp. 32–46. Available at: <https://www.researchgate.net/publication/290029169_Two_Types_of_Argumentation_Involving_Rhetorical_Questions_in_Biblical_Hebrew_Dialogue> [Accessed 1 July 2021]. https://www.jstor.org/stable/42614874.

Oraby, S., Harrison, V., Hernandez, E., Reed, L., Riloff, E. and Walker, M., 2016. Creating and characterizing a diverse corpus of sarcasm in dialogue.  In: 17th Annual Meeting of the Special Interest Group on Discourse and Dialogue, pp. 31–41. Available at: <https://www.researchgate.net/publication/312250769_Creating_and_Characterizing_a_Diverse_Corpus_of_Sarcasm_in_Dialogue> [Accessed 1 July 2021]. DOI: 10.18653/v1/W16-3604.

Oraby, S., Harrison, V., Misra, A., Riloff, E. and Walker, M., 2017. Are you serious? Rhetorical questions and sarcasm in social media dialog. In: 18th Annual SIGdial Meeting on Discourse and Dialogue, pp. 310–319. Available at: <https://www.researchgate.net/publication/319162869_Are_you_serious_Rhetorical_Questions_and_Sarcasm_in_Social_Media_Dialog> [Accessed 1 July 2021]. DOI: 10.18653/v1/W17-5537.

Rohde, H., 2006. Rhetorical questions as redundant interrogatives. San Diego Linguistics Papers, 2, pp. 134–168. Available at: <https://escholarship.org/uc/item/4xd7t5ww> [Accessed 1 July 2021].

Schaffer, D., 2005. Can rhetorical questions function as retorts? Is the Pope Catholic? Journal of Pragmatics, 37, pp. 433–460. https://doi.org/10.1016/j.pragma.2003.12.007.

Špago, Dž., 2020. Rhetorical questions as aggressive, friendly, or sarcastic/ironical questions with imposed answers. ExELL, 8 (1), pp. 68–82. Available at: <https://www.sciendo.com/article/10.2478/exell-2020-0014> [Accessed 1 July 2021]. https://doi.org/10.2478/exell-2020-0014.

Toplak, M., Katz, A. N., 2000. On the uses of sarcastic irony. Journal of Pragmatics, 32 (10), pp. 1467–1488. https://doi.org/10.1016/S0378-2166(99)00101-0.

Van Rooy, R., 2003. Negative polarity items in questions: Strength as relevance. Journal of Semantics, 20, pp. 239–273. Available at: <https://www.researchgate.net/publication/2534444_Negative_Polarity_Items_in_Questions_Strength_as_Relevance> [Accessed 1 July 2021]. https://doi.org/10.1093/jos/20.3.239.

Zwarts, F., 1996. A hierarchy of negative expressions. In: Wansing, H. ed. Negation: A notion in focus. Berlin: Walter de Gruyter, pp. 169–194. https://doi.org/10.1515/9783110876802.169.