Speech Acts Bibliography:
Advice / Refusal of Advice
Bardovi-Harlig, K. & Hartford, B. S. (1990). Congruence in native and nonnative conversations: Status balance in the academic advising session. Language Learning, 40 (4), 467-501.
Looks at 32 advising sessions between faculty advisors and NS and NNS grad students. Found that nonnative subjects (NNSs) may lack status-preserving strategies that minimize the force of non-congruent speech acts -- i.e., those strategies that allow students to take out-of-status turns without jeopardizing their relationship with their advisors. The linguistic competence of the NNSs was thought to be good, but the lack was in context-specific pragmatic competence.
Bardovi-Harlig, K. & Hartford, B. S. (1991). Saying "no" in English: Native and nonnative rejections. In L. F. Bouton (Ed.), Pragmatics and Language Learning. Volume 2 (pp. 41-57). Urbana, IL: University of Illinois at Urbana-Champaign.
This article examines rejections used by 7 NSs and 39 proficient NNSs of English from 39 audio-taped academic advising sessions. Refusals were analyzed in terms of linguistic marking, semantic formulae, and context specific content of the refusals. While NSs were able to reject an advisor's suggestion and still maintain the status balance, NNSs were less predictably able to do so. NNSs tended to employ a wider variety of refusals than the NSs, but were more successful when following the NS tendencies. Explanation was the most common semantic formulae for both groups. Giving alternatives was the second most common strategy for NSs, and avoidance was the second most common strategy for NNSs.
Bardovi-Harlig, K. & Hartford, B. S. (1993). Refining the DCT: Comparing open questionnaires and dialogue completion tasks. Pragmatics and Language Learning. Volume 4 (pp. 143-165). Urbana, IL: University of Illinois at Urbana-Champaign.
Compares the influence of two forms of discourse completion tasks on the elicitation of rejections of advice. An open questionnaire, which provided scenarios alone, was compared with a classic dialogue completion task in which a conversational turn was provided. The study was based on their Hartford & Bardovi-Harlig 1992 study but the DCT has authentic language in it. Thirty-two grad students participated (19 NSs, 13 NNSs), all having been in an actual counseling situation with an adviser. 17 did the open questionnaire first and then the discourse completion task, 15 did it vice versa. Both NSs and NNSs showed task influence, although it was greater for the NNSs. In many cases NNS's responses were very similar to those of NSs on the dialogue completion task. Their conclusion was that for reactive speech acts (i.e., those that never stand alone) such as rejections, the inclusion of conversational turns is the preferred format. The increased specificity of the dialogue completion task over the open questionnaire is particularly important to the NNSs. NSs were more adept at imagining a plausible conversational turn given a scenario than the NNSs, so for them the presence of talk made less difference. Refining of DCTs should be to include more authentic speech in the conversational turn(s) provided.
Bardovi-Harlig, K. & Hartford, B. S. (1993). Learning the rules of academic talk: A longitudinal study of pragmatic change. Studies in Second Language Acquisition, 15 (3), 279-304.
Reports on a longitudinal study of the acquisition of pragmatic competence. Ten advanced adult nonnative speakers of English were taped in two advising sessions over the course of a semester -- an early and a later session. Two speech acts, suggestions and rejections, were analyzed according to their frequency, form, and successfulness and compared with similar data gathered for six native speakers. The nonnative speakers showed change toward the native speaker norms in their ability to employ appropriate speech acts, moving toward using more suggestions and fewer rejections, and became more successful negotiators. However, they changed less in their ability to employ appropriate forms of the speech acts, continuing to use fewer mitigators than the native speakers. Furthermore, unlike native speakers, they also used aggravators. The investigators claimed that these results may be explained by the availability of input: learners received positive and negative feedback from the advisor regarding the desirability and outcome of particular speech acts, but they did not receive such feedback regarding the appropriateness of the forms of such speech acts.
Bardovi-Harlig, K. & Hartford, B. S. (1996). Input in an Institutional Setting. Studies in Second Language Acquisition, 18 (2), 171-88.
Investigates the nature of input available to learners in the institutional setting of the academic advising session. The advisory session is viewed as an unequal status encounter that by nature is a private speech event and cannot be observed by other learners. They report on the longitudinal study of sixteen graduate students (Bardovi-Harlig & Hartford 1993). Whereas the advisor may teach the advisees about the form of the interview and content that is appropriate, the advisees do not receive negative feedback regarding form, so they cannot learn from that.
Hartford, B. S. & Bardovi-Harlig, K. (1992). Experimental and observational data in the study of interlanguage pragmatics. Pragmatics and Language Learning. Volume 3 (pp. 33-52). Urbana, IL: University of Illinois at Urbana-Champaign.
Compares data on rejections of advice by NSs and NNSs from natural conversations in academic advising sessions with data collected from a discourse completion test (DCT) -- 13 NSs and 11 NNSs on the DCT vs. 39 advising sessions (18 NSs and 21 NNSs). Their studies found that NNSs used more semantic formulas (hence, made more turns) for a rejection and made more rejections while NSs made suggestions more than twice as often as they rejected advice. They found three common semantic formulas (explanations, alternatives, rejections) and more than ten less common ones. Although the use of DCTs has benefits such as availability of large samples and experimental controls, there are biases: respondents used a narrower range of semantic formulas on the DCT (as Beebe & Cummings, 1985, found), used fewer status preserving strategies, and lacked extended negotiations found in the natural data. This is because the DCT does not promote the turn-taking and negotiation strategies found in natural conversation. Also, the DCT allows students to be less polite, and more bald-on-record statements are used than in the natural situation — even in the status-unequal situation that these investigators used, because of the anonymity involved in the measure. Also, respondents can opt out with the DCT, which is unlikely in a natural conversation. However, the DCT facilitates testing hypotheses derived from instances in natural conversation where there is insufficient data. It provides data to help explain and interpret the natural data. Conclusion — need for more observational data but DCT has an important role as well.
Hernández-Flores, N. (1999). Politeness ideology in Spanish colloquial conversation: The case of advice. Pragmatics, 9(1), 37-49.
This article addresses the lack of universal application of Brown and Levinson's (1987) notion of politeness as related to positive and negative face by analyzing colloquial conversations in Peninsular Spanish, specifically the case of advice. The author proposes that politeness is not always used to mitigate face threatening acts. Adapting Bravo's (1996) classification of face, the data was analyzed utilizing the notions of autonomy (perception by people as being one's own within a group, self-affirmation) and affiliation (to be perceived as an integrated part of the group) as well as the cultural notion of confianza. This study asserts that, in the case of advice in Peninsular Spanish, politeness is often used to "enhance the relationship between interactants” and not necessarily to mitigate face threatening acts, as proposed in Brown and Levinson's model. This conclusion is supported by examples from the corpus indicating the common and friendly expression of unsolicited advice, clear assertions, and rejections by the hearer.
Hinkel, E. (1997). Appropriateness of advice: DCT and multiple-choice data. Applied Linguistics, 18 (1), 1-26.
Looks at what can be learned about L2 speech acts from data obtained by means of English language multiple-choice and DCT instruments in two experiments. The focus was on the L1 responses of NSs and the L2 responses given by speakers of Chinese to DCTs multiple-choice questionnaires (constructed using the DCT responses as alternatives) dealing with the appropriateness of advice in common and observed situations. The finding was that significantly more of the 40 NSs than of the 40 Taiwanese Chinese preferred direct and hedged advice in response to the DCTs. When responding to the multiple- choice questionnaire, another group of 40 NSs selected substantially fewer options with either direct or hedged advice than another group of 40 Taiwanese Chinese subjects did, which is congruent with the body of research on NSs and Chinese L1 sociolinguistic behaviors. The investigator determined that since the multiple-choice responses were consistent with the literature, the DCTs may not be the best elicitation instrument for L1 and L2 data pertaining to ambiguous and situationally constrained sociolinguistic acts.
Jeon, M. (2003). Closing the advising session. Working Papers in Educational Linguistics, 18(2), 89-106.
The study looked at the four subsections of closings (shutting down, preclosing, thanking -- expressions of gratitude, terminal exchange) in advising sessions and the effect of ESL proficiency on the ability to do these without having the speech behavior marked. It built on the work by Bardovi-Harlig and Hartford. There is a delicate balance between overly extending closings and terminating the conversation too abruptly with "gotta go." The study involved four native English-speaking advisors and 32 ESL learners from eight language backgrounds and at three proficiency levels. Abbreviated closings were seen as missing any of the four subsections, and extended closings added another step after preclosing, namely reopening or making arrangements. Results: off the 32 interactions, 15 had extended closings, 10 "completed" (or appropriate), and 6 were abbreviated. The results showed that cross-sectionally, abbreviated closings decreased as proficiency increased. Also, the least proficient were most likely to engage in extended closings. Since the most advanced students were least likely to have marked closings, it was seen not only as an indication of advanced proficiency, but also more experience with such advising situations. The author points out that the participants did not provide retrospective information which might have enhanced her understanding of their perceptions of them. She suggests the need to provide instruction on proper closings.
Martinell-Gifre, E. (1992). Preguntas que no preguntan. E.L.U.A., 8, 25-35.
The author provides an exploratory look at utterances that appear to be interrogatives (e.g., rising intonation, punctuation, etc.) that do not actually function as questions. That is, they are not used to obtain information. Instead, many of these utterances are used to suggest, threaten, request, etc. In an analysis of the circumstances and purposes for which these types of utterances are used, the author classifies five distinct contexts of use to: (1) anticipate attitude or response, (2) elicit response or reaction, (3) react to what has been said, (4) react to attitudes of actions of others, and (5) demonstrate a physical presence that is not dispensable. Examples and discussion of each type of use are given.
Stewart, M. (2004). Written pedagogic feedback and linguistic politeness. In Márquez Reiter, R. & Placencia, M. E. (Eds.), Current Trends in the Pragmatics of Spanish (pp. 99-120). Amsterdam, Netherlands: John Benjamins Publishing.
This qualitative investigation analyzes written feedback given by native Spanish-speaking (Peninsular Spanish) tutors to Spanish language learners on written assignments (NS of British English). Fifty-eight tokens of written feedback were analyzed for presentational (level of the text) and selectional (level of the utterance) politeness. Results of the analysis show that extra-linguistic variables (e.g., power, roles) help determine weight accorded to the face threatening act, types of strategies used, and level of attention given to the speaker and hearer. As opposed to linguistic feedback, where the NS felt a very high level of authority, in this study, the tutors opted for more strategy use and face care when giving conceptual feedback, a subject on which they likely felt less authority. Linguistic feedback was given with little mitigation and included imperatives and present tense indicative statements (familiar form). However, conceptual feedback was given at the presentational (justification, exemplification, and appeal to a higher authority) and selectional levels (hedging, presupposition, and defocusing of the agent) in order to focus on protecting both the face of the speaker and the hearer.