“I’M ALWAYS SINCERE”: WHAT IS SALIENT IN SARCASM?
1
“I’M ALWAYS SINCERE”: WHAT IS SALIENT IN SARCASM? GINA M. CAUCCI, ROGER J. KREUZ, & EUGENE H. BUDER Introduction How does a speaker let a listener know that he/she is speaking sarcastically? Recently, researchers have suggested that specific words, collocations and even phrases (e.g., Kreuz & Roberts, 1994; Utsumi, 2000), can act as cues to the listener that the speaker’s utterance is meant sarcastically. For example... Adjective-adverb collocations: (e.g., That’s a lovely, pink, uh, satin-ish dress.) Interjections: (e.g., Oh, Um, Uh,Well) Frozen phrases (Haiman, 1998: e.g., “Don’t you just love it when...”) There have also been claims that paralinguistic cues (i.e., gesturing, eye-rolling, smiling) are cues to sarcastic intent (e.g., Kreuz, 1996). Furthermore, it has been suggested that the relationship of the interlocutors (e.g., Kreuz & Caucci, 2008) can influence how and how often speakers use sarcasm. The purpose of this project was to identify the specific facial and lexical cues used, as well as assess the role of common ground, in the production of natural sarcastic utterances. Procedure Participants were recorded in a conversational setting in the Social Interaction Lab while engaging in tasks designed to elicit natural sarcasm (see Materials). Common ground was manipulated to determine whether it affects the cues used to signal sarcastic intent. Naturally produced sarcastic utterances were compared to sincere/literal utterances from the same speaker in the same task. This was done in order to have a baseline for comparisons. Results (2): Lexical Sarcasm Sarcastic and literal utterances were coded for the presence of adjectives, adverbs, adjective-adverb collocations, and pauses. Results showed that there were significant differences between the sarcastic and literal utterances for presence of adjectives, adjective-adverb collocations, and pauses (see Table 1). Materials Participants were given three tasks designed to elicit sarcasm in a natural way. Tasks were randomized between sessions and included commenting on badly-dressed celebrities (Hancock, 2004), creating a meal for a person that they hate (Coates, 1991) and discussing bad restaurant experiences (see pictures to the right). Participants 29 pairs (17 friend and 12 stranger) were recruited from the Audiology and Psychology departments at the University of Memphis. Participants were told the purpose of the study was to investigate social interaction. 5063 Social Interaction Lab at the School of Audiology and Speech Language Pathology Discussion Importantly, all sarcastic utterances in our data set were elicited naturally. The results from the current project suggest that speakers indicate their sarcastic intent with facial as well as lexical cues in spontaneous sarcasm. Furthermore, we were able to show that common ground affects the facial cues participants produce to indicate sarcastic intent. Interestingly, we found a complicated relationship among facial, lexical and social cues to sarcasm. Future research should attempt to tease apart the individual contributions of these factors to the comprehension of a sarcastic statement. References Caucci, G. M., Kreuz, R. J., & Buder, E. H. (2007, November). Acoustic analyses of the sarcastic tone of voice. Paper presented at the 48th annual meeting of the Psychonomic Society, Long Beach, CA.. Coates, L.J, (1991). A collaborative theory of inversion: Irony in dialogue. Unpublished master’s thesis, University of Victoria,Victoria, British Columbia, Canada. Haiman, J. (1998). Talk is cheap: Sarcasm, alienation, and the evolution of language. New York: Oxford University Press. Hancock, J.T. (2004).Verbal irony use in face-to-face and computer-mediated conversations. Journal of Language and Social Psychology, 23, 447-463. Kreuz, R. J. (1996). The use of verbal irony: Cues and constraints. In J. S. Mio & A. N. Katz (Eds.), Metaphor: Implications and Applications, (pp. 23-38). Mahwah, NJ: Lawrence Erlbaum Associates. Kreuz, R. J., & Caucci, G. M. (2008, July). Do lexical factors affect the perception of sarcasm? Paper presented at the 18th annual meeting of the Society for Text and Discourse, Memphis, TN. Kreuz, R. J., & Roberts, R. M. (1995). Two cues for verbal irony: Hyperbole and the ironic tone of voice. Metaphor and Symbolic Activity, 10, 21-31. Louwerse, M.M., Benesh, N., Hoque, M.E., Jeuniaux, P., Lewis, G. , Wu, J., & Zirnstein, M. (2007). Multimodal communication in face-to-face conversations. Proceedings of the 28th Annual Conference of the Cognitive Science Society (pp. 1235-1240). Mahwah, NJ: Erlbaum. Utsumi, A. (2000).Verbal irony as implicit display of ironic environment: Distinguishing ironic utterances from nonirony. Journal of Pragmatics, 32, 1777-1806. “Prepare a meal for a person you hate” (Coates, 1991). “Discuss with your partner the fashions you see” (Hancock, 2004). Table 1. Number of Lexical Items Found in Utterances __________________________________________________________ Lexical Characteristic Sarcastic Baseline wetoioiuoiwu Interjections 18 18 Adjectives (only)* 28 8 Adverbs (only) 8 4 Adjective/Adverb collocations* 7 0 Pauses* 7 0 __________________________________________________________ Results (1): Facial Cues Facial gestures were coded using a taxonamy adapted from Louwerse et al. (2007). Results showed a significant difference in number of smiles, laughs, nods, and looks to partner for sarcastic statements compared to the literal utterances (see below). HEAD forward down left tilt right tilt fast nod slow nod* left turn right turn EYES brows up brows down asymmetrical rapid blink squinting widening looks away looks to partner* MOUTH smile** lip tighten* laugh* 0 0.1 0.2 0.3 0.4 0.5 DIFFERENCES IN FACIAL CUES FOR SARCASTIC & LITERAL STATEMENTS ✓ ✓ ✓ ✓ ✓ ✓ Sarcastic Literal * p < .05, ** p < .001 Results (3): Common Ground In previous work, we showed the importance of common ground in determining how, as well as how often, people speak sarcastically (e.g., Caucci, Kreuz, & Buder 2007). There were 12 sessions of friend pairs, 7 of which contained at least one sarcastic utterance. Three of the 12 stranger-pair sessions included at least one sarcastic utterance. Common ground affected the frequency of facial cues used to signal sarcastic intent (see Figure 2). * p < .05 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 looks to partner smiles laughs COMMON GROUND DIFFERENCES FOR FACIAL CUES Friends Strangers ✓ ✓ ✓ ✓ p < .05 Figure 1. Figure 2.
Transcript of “I’M ALWAYS SINCERE”: WHAT IS SALIENT IN SARCASM?
FRIENDS STRANGERS
LOOKS TO PARTNER
SMILES
LAUGHS
0.53 0.14
0.63 0.21
0.37 0
“I’M ALWAYS SINCERE”: WHAT IS SALIENT IN SARCASM?GINA M. CAUCCI, ROGER J. KREUZ, & EUGENE H. BUDER
Introduction
How does a speaker let a listener know that he/she is speaking sarcastically?
Recently, researchers have suggested that specific words, collocations and even phrases (e.g., Kreuz & Roberts, 1994; Utsumi, 2000), can act as cues to the listener that the speaker’s utterance is meant sarcastically.
For example...
Adjective-adverb collocations: (e.g., That’s a lovely, pink, uh, satin-ish dress.) Interjections: (e.g., Oh, Um, Uh, Well)Frozen phrases (Haiman, 1998: e.g., “Don’t you just love it when...”)
There have also been claims that paralinguistic cues (i.e., gesturing, eye-rolling, smiling) are cues to sarcastic intent (e.g., Kreuz, 1996).
Furthermore, it has been suggested that the relationship of the interlocutors (e.g., Kreuz & Caucci, 2008) can influence how and how often speakers use sarcasm.
The purpose of this project was to identify the specific facial and lexical cues used, as well as assess the role of common ground, in the production of natural sarcastic utterances.
Procedure Participants were recorded in a conversational setting in the Social Interaction Lab while
engaging in tasks designed to elicit natural sarcasm (see Materials).
Common ground was manipulated to determine whether it affects the cues used to signal sarcastic intent.
Naturally produced sarcastic utterances were compared to sincere/literal utterances from the same speaker in the same task. This was done in order to have a baseline for comparisons.
Results (2): Lexical Sarcasm
Sarcastic and literal utterances were coded for the presence of adjectives, adverbs, adjective-adverb collocations, and pauses.
Results showed that there were significant differences between the sarcastic and literal utterances for presence of adjectives, adjective-adverb collocations, and pauses (see Table 1).
Materials Participants were given three tasks designed to elicit sarcasm in a natural way.
Tasks were randomized between sessions and included commenting on badly-dressed celebrities (Hancock, 2004), creating a meal for a person that they hate (Coates, 1991) and discussing bad restaurant experiences (see pictures to the right).
Participants 29 pairs (17 friend and 12 stranger) were recruited from the Audiology and Psychology
departments at the University of Memphis.
Participants were told the purpose of the study was to investigate social interaction.
5063
Social Interaction Lab at the School of Audiology and Speech Language Pathology
Discussion Importantly, all sarcastic utterances in our data set were elicited naturally.
The results from the current project suggest that speakers indicate their sarcastic intent with facial as well as lexical cues in spontaneous sarcasm.
Furthermore, we were able to show that common ground affects the facial cues participants produce to indicate sarcastic intent.
Interestingly, we found a complicated relationship among facial, lexical and social cues to sarcasm.
Future research should attempt to tease apart the individual contributions of these factors to the comprehension of a sarcastic statement.
References
Caucci, G. M., Kreuz, R. J., & Buder, E. H. (2007, November). Acoustic analyses of the sarcastic tone of voice. Paper presented at the 48th annual meeting of the Psychonomic Society, Long Beach, CA.. Coates, L.J, (1991). A collaborative theory of inversion: Irony in dialogue. Unpublished master’s thesis, University of Victoria, Victoria, British Columbia, Canada.
Haiman, J. (1998). Talk is cheap: Sarcasm, alienation, and the evolution of language. New York: Oxford University Press.
Hancock, J.T. (2004). Verbal irony use in face-to-face and computer-mediated conversations. Journal of Language and Social Psychology, 23, 447-463.
Kreuz, R. J. (1996). The use of verbal irony: Cues and constraints. In J. S. Mio & A. N. Katz (Eds.), Metaphor: Implications and Applications, (pp. 23-38). Mahwah, NJ: Lawrence Erlbaum Associates.
Kreuz, R. J., & Caucci, G. M. (2008, July). Do lexical factors affect the perception of sarcasm? Paper presented at the 18th annual meeting of the Society for Text and Discourse, Memphis, TN.
Kreuz, R. J., & Roberts, R. M. (1995). Two cues for verbal irony: Hyperbole and the ironic tone of voice. Metaphor and Symbolic Activity, 10, 21-31.
Louwerse, M.M., Benesh, N., Hoque, M.E., Jeuniaux, P., Lewis, G. , Wu, J., & Zirnstein, M. (2007). Multimodal communication in face-to-face conversations. Proceedings of the 28th Annual Conference of the Cognitive Science Society (pp. 1235-1240). Mahwah, NJ: Erlbaum.
Utsumi, A. (2000). Verbal irony as implicit display of ironic environment: Distinguishing ironic utterances from nonirony. Journal of Pragmatics, 32, 1777-1806.
“Prepare a meal for a person you
hate” (Coates, 1991).
“Discuss with your partner the fashions you see” (Hancock,
2004).
Table 1. Number of Lexical Items Found in Utterances__________________________________________________________
Lexical Characteristic Sarcastic Baseline wetoioiuoiwuInterjections 18 18
Adjectives (only)* 28 8
Adverbs (only) 8 4
Adjective/Adverb collocations* 7 0
Pauses* 7 0
__________________________________________________________
Results (1): Facial Cues
Facial gestures were coded using a taxonamy adapted from Louwerse et al. (2007).
Results showed a significant difference in number of smiles, laughs, nods, and looks to partner for sarcastic statements compared to the literal utterances (see below).
Sarcasm BaselineHEADforwarddownleft tiltright tiltfast nodslow nodleft turnright turnEYESbrows upbrows downasymmetricalrapid blinksquintingwideninglooks awaylooks to partner
MOUTHsmilelip tightenlaugh
3.03% 3.03%33.33% 30.30%
3.03% 0%6.06% 0%6.06% 3.03%
15.15% 3.03%18.18% 12.12%18.18% 15.15%
15.15% 6.06%3.03%
3.03% 0%3.03% 6.06%6.06% 3.03%6.06% 0%
18.18% 24.24%36.36% 15.15%
45.45% 6.06%24.24% 3.03%21.21% 6.06%
*
HEADforward
downleft tilt
right tiltfast nod
slow nod*left turn
right turnEYES
brows upbrows downasymmetrical
rapid blinksquintingwidening
looks awaylooks to partner*
MOUTHsmile**
lip tighten*laugh*
0 0.1 0.2 0.3 0.4 0.5
DIFFERENCES IN FACIAL CUES FOR SARCASTIC & LITERAL STATEMENTS
✓
✓✓✓
✓✓
SarcasticLiteral
* p < .05, ** p < .001
Results (3): Common GroundIn previous work, we showed the importance of common ground in determining how, as
well as how often, people speak sarcastically (e.g., Caucci, Kreuz, & Buder 2007).
There were 12 sessions of friend pairs, 7 of which contained at least one sarcastic utterance. Three of the 12 stranger-pair sessions included at least one sarcastic utterance.
Common ground affected the frequency of facial cues used to signal sarcastic intent (see Figure 2).
* p < .05
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
looks to partner smiles laughs
COMMON GROUND DIFFERENCES FOR FACIAL CUES
FriendsStrangers
✓✓
✓
✓p < .05
Figure 1.
Figure 2.
![Sarcasm & Thwarting in Sentiment Analysis [IIT-Bombay]](https://static.fdocuments.in/doc/165x107/5559a7f6d8b42a5b2a8b4d58/sarcasm-thwarting-in-sentiment-analysis-iit-bombay.jpg)
