A pragmatic analysis of l2 spanish reque

This research expands upon previous work byinvestigating the acquisition of second language requests during a short-termimmersion program 6 weeks in Madrid, Spain and in three situationa

Lori Czerwionka* and Alejandro Cuza

A pragmatic analysis of L2 Spanish

requests: Acquisition in three situational contexts during short-term study abroadDOI 10.1515/ip-2017-0016

Abstract: This study examines pragmatic acquisition of requests for speaking learners of Spanish This research expands upon previous work byinvestigating the acquisition of second language requests during a short-termimmersion program (6 weeks) in Madrid, Spain and in three situational contexts:food and drink, general merchandise, and familial Data were collected using anexperimental computerized oral discourse completion task Requests made bylearners (501 requests) and native speakers (224 requests) were compared con-sidering personal deictic orientation and directness of the requests For learners,shifts from speaker-oriented to hearer-oriented requests indicated greater prag-matic development in food and drink and familial contexts Results are dis-cussed considering pragmatic developmental stages and differential results inthe three contexts

English-Keywords: request, speech act, second language acquisition of pragmatics,study abroad, situational variation

1 Introduction

Pragmatic competence is essential to second language learners, especially incontexts where their second language is the primary language of use, as inthe case of study abroad Pragmatic competence encompasses the knowledge

of pragmalinguistic resources and sociopragmatic application of thoseresources to second language contexts (Barron 2003) “Pragmalinguisticsdeals with the linguistic resources of a language which can be employed toserve a specific communicative function, e.g the syntactic patterns whichcan be used to perform a particular speech act in a given language

*Corresponding author: Lori Czerwionka, Purdue University, West Lafayette, USA,

E-mail: czerwionka@purdue.edu

Alejandro Cuza, Purdue University, West Lafayette, USA, E-mail: acuza@purdue.edu

Sociopragmatics, on the other hand, examines the social circumstances underwhich a particular speech act can be performed Thus, pragmalinguistic stu-dies are language-specific, while sociopragmatic studies are culture-specific(cf Leech 1983: 10–13)“ (Schneider 2014: 114) Learners are faced with achallenge when their first and second languages use different linguisticresources to perform a given speech act in particular social circumstances, as

in the case of requests in English and Spanish English tends to rely more onspeaker-oriented, indirect requests like Could I have some water?; Spanishtends to use more hearer-oriented, direct requests like Pásame el agua ‘Pass

me the water’ (Ballesteros Martín 2001; Callahan 2011; Cenoz and Valencia1996; Félix-Brasdefer 2015; Fox and Heinemann 2016; Le Pair 1996; MárquezReiter 2000; Márquez Reiter 2002; Pinto 2005; Pinto and Raschio 2007;Placencia 2005; Shively 2011; Shively and Cohen 2008).1

Research on second language request acquisition during immersion grams abroad has provided varied findings There is evidence that English-speaking learners of Spanish shift to more Spanish-like, hearer-orientedrequests over a period abroad (Czerwionka and Cuza 2017; Shively 2011), yetcounter evidence often from specific request scenarios is also available(Bataller 2010; Hernández 2016; Shively and Cohen 2008) Mixed results onthe impact of study abroad exist when examining semester-long programsabroad (Bataller 2010; Shively 2011) and also short-term programs(Czerwionka and Cuza 2017; Hernández 2016) Furthermore, few studies havesystematically examined the impact of different situational contexts, with mostfocusing on service industry encounters (e.g., Félix-Brasdefer 2015)

pro-Therefore, the goal of the present study is to investigate the acquisition ofrequests among English-speaking learners of Spanish during a short-term,6-week program abroad in Madrid, Spain in three controlled situational con-texts A short-term program was selected considering that the majority of U.S.students study abroad in short-term programs (Institute of InternationalEducation n.d.) and that 57 % language learners who study abroad do so inshort-term programs (Allen and Dupuy 2013: 471) It is particularly relevant toexamine the educational outcomes of short-term programs in search for evi-dence to support the claim that “[e]ven shorter lengths of stay might helplearners become more targetlike, particularly with respect to highly salientconversational functions” (Bardovi-Harlig 1999: 685), as may be the case withrequests

1 Speaker-oriented requests contain first person-reference and hearer-oriented contain person reference.

second-2 Speech acts

Speech acts have been studied in a variety of different sociocultural contexts,including linguistically and geographically distinct contexts within the Spanish-speaking world (Félix-Brasdefer 2005; Félix-Brasdefer 2009; Félix-Brasdefer2010; Márquez Reiter 2002; Márquez Reiter 2005; Méndez Vallejo 2013;Placencia 1998; Placencia 2005; Ruzickova 2007; Stapleton 2004) Comparativestudies have addressed questions related to cross-linguistic communication(Barron 2005; Blum-Kulka et al 1989; Márquez Reiter et al 2005; Marti 2006),including comparisons of English and Spanish (Ballesteros Martín 2001;Ballesteros Martín 2002; Bou Franch and Lorenzo-Dus 2008; Callahan 2011;Koike 1994; Márquez Reiter 1997; Pinto 2010; Pinto and Raschio 2007) Priorresearch has also addressed speech acts as they vary according to particularsituations (Bataller 2010; Cenoz and Valencia 1996; Félix-Brasdefer 2004) andlocal situational factors such as social distance and power (Blum-Kulka et al.1989; Brown and Levinson 1987; Félix-Brasdefer 2004)

Speech acts provide insight into the interactional norms and expectations

of a given community, examined often through the lens of politeness ories Brown and Levinson (1987) initially proposed a universal understand-ing of politeness, but further investigation in diverse socioculturalcommunities led to criticisms of ethnocentricism (Bravo 2002; Hernández-Flores 2004; Matsumoto 1989; Pizziconi 2003) For the current purposes,politeness is understood as“a form of social interaction that is conditioned

the-by the sociocultural norms of a particular society” (Félix-Brasdefer 2006:2159) “[L]inguistic structures do not in themselves denote politeness”(Watts 2003: 168), rather “types of social interaction are sanctioned notonly by ourselves as individuals but by society as a whole” (143) In light ofthis understanding of politeness, second language requests over a short-termprogram abroad in Spain provide insight into American learners’ adoption ofsocial interactions (i.e., requests) sanctioned by the local, Spanish society

2.1 Requests in Spanish and English

Requests express a desire that attempts to make the hearer take action; theymay range in illocutionary force, depending on the linguistic features used tomake the request (e.g., direct imperative forms to indirect hints) (Bach 2003;Blum-Kulka et al 1989; Searle 1976) In Spanish and English, requests differmost saliently with respect to their personal deictic orientation and directness(Table 1)

With respect to deictic orientation, requests in Peninsular Spanish are monly hearer-oriented (i.e., second-person reference, ¿Me traes un café? ‘Willyou bring me a coffee?) and ambiguously-oriented—when no verb is used tomake the request (e.g., verbal ellipsis, un café por favor ‘a coffee please’)(Marquez Reiter 2000; 2002; Pinto 2005; Placencia 2005; Shively 2011; Shivelyand Cohen 2008) American English speakers typically rely on speaker-orientedrequests (i.e., first-person reference – I’d like a coffee) (Fox and Heinneman2016; Pinto and Raschio 2007) With respect to directness, indicated with clausetype in this study, Peninsular Spanish requests are overwhelmingly direct, atleast in food and drink contexts (Placencia 2005) Placencia (2005) indicatedthat native speakers of Spanish rely on imperative clauses and verbal ellipsismost often Spanish speakers have broadly been found to use more directrequests than English speakers (Ballesteros Martín 2001; Callahan 2011; Félix-Brasdefer 2015; Márquez-Reiter 1997) Spaniards have also been shown to rely

com-on oriented, ccom-onventicom-onally indirect requests, in the form of oriented interrogatives (Bataller 2010; Cenoz and Valencia 1996; Le Pair 1996).Spanish requests are quite formulaic across the peninsula (Stapleton 2004), anduse of hearer-oriented interrogatives, imperatives, and verbal ellipsis are com-mon request strategies in Peninsular Spanish In comparison to Spanishrequests, English-speaking Americans use a broader range of requests fromdirect imperatives to indirect preparatory directives (e.g., Do you have coffee?)

hearer-to permission directives (e.g., May I … ) (Ervin-Tripp 1976) Two of the mostcommon request forms in English are speaker-oriented“want/need-declarativesand can-interrogatives” (Fox and Heinemann 2016: 499) Pinto and Raschio(2007) found can-interrogatives, also called permission directives, to be com-mon in English and uncommon in Spanish

Table 1: Examples of requests by orientation and directness in Spanish and English.

‘A beer.’ Conventionally

indirect

¿ Me pones una ensalada?

‘Can you get me a salad?’

Interrogative oriented)

(Speaker-Can I have a salad?

# i¿ Pue do tener una ensalada?

i

This request form is felicitous in English and infelicitous in Spanish as indicated by #.

2.2 Requests of English-speaking second language learners

of Spanish in immersion settings

The differences between Spanish and English request forms have the potential tobeget transfer from first language norms to second language requests, leading torequests in the second language that are infelicitous in the local community Thestudy of second language learners’ requests in Spanish highlight learners’ usagetendencies and acquisition of pragmatic norms (Alcón-Soler 2015; Bataller 2010;Félix-Brasdefer 2004; Félix-Brasdefer 2007; González-Cruz 2014; Hernández2016; Le Pair 1996; Pinto 2005; Shively 2011) In the study abroad context ofSpain, some research indicates that learners adjust their request strategiesduring immersion programs to approximate native speaker populations(Czerwionka and Cuza 2017; Shively 2011) Yet Shively and Cohen (2008) foundmixed results depending on individual situational scenarios, and Hernández(2016) found that learners maintained a preference for speaker-oriented requestsover a short-term program abroad Even comparing the two prior studies ofEnglish-speaking learners of Spanish over short-term programs abroad, therewere divergent findings (Czerwionka and Cuza 2017; Hernández 2016)

Analyses of requests during study abroad in the Peninsular Spanish contexthave tended to focus on service-industry contexts in which food or drinks areexchanged (Bataller 2010; Czerwionka and Cuza 2017; Shively 2011) In food anddrink contexts specifically, there is evidence that English-speaking learners ofSpanish in Spain shift away from speaker-oriented requests over the periodabroad Shively (2011) demonstrated this trend for learners over a semesterabroad, finding a decrease in the dependence on speaker-oriented requests and

an increase in ambiguously-oriented requests Czerwionka and Cuza (2017) found

a decrease in speaker-oriented requests in favor of hearer-oriented requests infood and drink scenarios over a 6-week, short-term program abroad Thus, similarrequest development was found after a semester-long and 6-week program, andpragmatic intervention does not seem to be required to provoke this pragmaticdevelopment as evidenced in the 6-week program data Conversely, Bataller (2010)found distinct trends for a coffee request scenario (Bataller 2010) In a reanalysis

of the Bataller (2010) data to examine the personal deictic orientation of requests,the learners used speaker-oriented requests for coffee most often at the beginningand end of the semester using declarative want-statements (I want… ‘Quiero … ’),need statements (I need … ‘Necesito … ’), and query permission interrogativerequests (Can I have… ‘¿Puedo tener … ?2)

2 This structure is pragmatically inappropriate in Spanish.

While focusing on food and drink contexts, other contexts have also beenaddressed Bataller (2010) analyzed requests in a shoe exchange scenario, inwhich learners relied on speaker-oriented requests 84.6 % of the time at thebeginning of the semester and 81.3 % at the end Hearer-oriented request usageincreased from 0 % to 6.3 % over the semester Thus, in her study, learnersexperienced small shifts towards the use of hearer-oriented requests, but overallmaintained a strong reliance on speaker-oriented requests independent of thecontext Hernández (2016) analyzed the deictic orientation of requests in fivesituations (i.e., Request of slower speech, airplane seat, paper extension, lessfood, leaving school—originally used in Shively and Cohen 2008) He found alearner preference for speaker-oriented requests at the beginning and end of ashort-term program.

Considering the prior research collectively, the mixed findings prompt sideration of the impact of the social contexts of requests and methodologicalapproaches Regarding social contexts, it is questioned whether learners aremore apt to adopt hearer-oriented requests in food and drink situations com-pared to others There is little evidence to support this hypothesis, partially due

con-to the fact that in investigations that have addressed a variety of contexts, oftenthe variables considered have been unbalanced, making it difficult to makecomparisons (cf Ballesteros Martín 2001) For example, in the scenarios pro-vided by Shively and Cohen (2008) and used by Hernández (2016), relative socialstatus, social distance, and imposition were considered, but their representationwas not balanced For relative status of the hearer, three scenarios included ahigher status listener, one was of equal or higher status, and one was of lowerstatus This type of imbalance makes it difficult to understand how these vari-ables impact the request norms This same issue has been observed regardingthe analysis of social contexts (e.g., university setting, familial) Thus, to inves-tigate the impact of the social context (e.g., food and drink, general merchan-dise, familial) on learners’ requests, sufficient data and a balanced design areimperative

The mixed findings in prior investigations of learners’ request developmentmay also potentially be explained by the impact of the methodologicalapproach Maintenance of speaker-oriented, English-like, requests was foundmost often using a written discourse completion task (DCT) (Bataller 2010;Hernández 2016) Shifts from speaker- to hearer-oriented requests—becomingmore Spanish-like—were found in studies that relied on naturally-occurring data(Shively 2011) and a computerized oral DCT (Czerwionka and Cuza 2017).While data elicitation methods could have had an effect (Golato 2003),Bataller and Shively (2011) indicated more similarities between elicited andnaturally-occurring data than differences (see also Félix-Brasdefer 2003)

Supporting the limited impact of elicitation method is the fact that both Shively(2011) and Czerwionka and Cuza (2017) found similar shifts from speaker- tohearer-oriented requests using different methods Thus, it is unclear that meth-odological effects have had a primary role in producing distinct results related tolearner request development, and there is evidence for similar pragmatic devel-opment analyzed via computerized oral DCTs and spontaneous data.

In summary, students can make progress towards hearer-oriented requestsduring a period abroad in Spain, at least in food and drink scenarios(Czerwionka and Cuza 2017; Shively 2011) It is less clear whether learnersexperience a shift from speaker- to hearer-oriented requests in other situations(e.g., Bataller 2010; Hernández 2016) Therefore, the current study examinedlearner requests in food and drink, general merchandise, and familial situationsduring a short-term immersion program in Spain, using the same methodology

as Czerwionka and Cuza (2017) to provide comparative data The rationale forthese particular contexts is addressed in Section 3 This study provides insightinto second language acquisition of requests during short-term programs abroadand examines the impact of three situational contexts

3 The current study

The goal of this investigation is to examine the second language acquisition ofrequests by English-speaking learners of Spanish during a short-term immer-sion program in Madrid, Spain The current investigation expands upon priorresearch to consider request acquisition in three situational contexts: (1) ser-vice encounter contexts of food and drink retailers, (2) service encountercontexts in which general merchandise or services are exchanged, and (3)familial contexts in which food and drink objects are exchanged betweenfamily members These are referred to as food and drink, general merchandise,and familial contexts These three contexts have been approached in priorliterature on requests in study abroad settings (Shively 2011; Bataller 2010;Shively and Cohen 2008), indicating that they are common interactions forlearners abroad Learners identify service encounters and family interactions

as common contexts in which learning occurs during study abroad programs(Misfeldt and Plews 2017), again indicating that experiences in these contextsare frequent for learners We include food and drink and general merchandise

as distinct types of service encounters, since it is assumed that food and drinkinteractions are more common for learners as food and drink are part of dailyevents that occur multiple times per day Requesting other items in serviceencounters (e.g., shoes, movie tickets, stamps) likely occurs less frequently,

but students in the program studied mentioned having these types of tions as well Family contexts were included because the students in thisprogram indicated on questionnaires that they spent large amounts of timewith host families Per week, students reported spending an average of 5.8hours going out for food or drink (SD = 5.6), 16.0 hours doing tourist activitiesand shopping which may include general merchandise encounters (SD = 9.9),and 16.8 hours doing activities with their host families (SD = 12.6).

interac-The following research questions guided the analyses, with research tion 1 addressing learners’ development over time and research question 2addressing the comparison of learners and native speakers of PeninsularSpanish The research questions are applied first to the analysis of requestorientation (i.e., hearer-, speaker-, and ambiguously-oriented requests) andthen to request directness (i.e., direct and conventionally indirect requests).Research question 1: Do the request strategies of English-speaking learners ofSpanish change over the immersion program in food and drink, general mer-chandise, and familial contexts?

ques-Research question 2: Do learners’ request strategies become more felicitousover the period abroad in food and drink, general merchandise, and familialcontexts considering Peninsular Spanish norms?

Regarding hypotheses for research question 1 for request orientation, weexpect learners’ requests to change over the period abroad considering theincreased exposure to Peninsular request norms in food and drink, generalmerchandise, and familial contexts Based on prior research on food and drinkcontexts, a shift away from speaker-oriented requests is expected We hypothe-size that learners will exhibit a similar shift in all three contexts The hypothesesrelated to research question 2 for request orientation are that learners willdevelop more felicitous requests over the period abroad, thus exhibiting signifi-cant differences from native speakers at the beginning of the program and nosignificant differences at the end of the program While prior research hasindicated this trend in food and drink contexts, the development of felicitousrequests is expected in all contexts

For research question 1 and request directness, it is hypothesized thatlearners in this program will use more direct requests over time in all contextsconsidering that Czerwionka and Cuza (2017) found a learner shift towardsdirectness, at least in Spanish food and drink contexts Contrary to Czerwionkaand Cuza’s (2017) finding, Félix-Brasdefer’s (2007) examination of universitylevel Spanish learners at four different levels in an American university settingindicated that learners used more indirect requests over the four years,

identifying a trend that followed the developmental stages proposed by Kasperand Rose (2002) For the current hypothesis we rely on the Czerwionka and Cuza(2017) data that better represent the student population, educational program,and Spanish variety under consideration in this paper For research question 2related to directness, we predict that learners’ requests will diverge from nativespeakers’ request norms at the beginning of the program and that learners willdevelop more felicitous requests over the period abroad, thus exhibitingrequests that align with native speakers’ requests at the end of the program.This hypothesis stands for all three contexts.

3.1 Participants and immersion context

Seventeen (n = 17) English-speaking learners of Spanish and fifteen (n = 15)native speakers of Peninsular Spanish participated in the study The secondlanguage (L2) learners (mean age = 21.2, SD = 2.0) were undergraduate students

at a large American university in the Midwest All learners were enrolled in astudy abroad program in Madrid, Spain, the criteria that determined thesample size of this population They were majors or minors in Spanish,and they had all taken a fifth semester Spanish course or beyond when startingthe program Learners responded to a survey about their Spanish language useduring the program (Cuza 2013)

The learners lived with host families in Madrid, and they reported usingSpanish during an average of 52.5 hours per week (SD = 35.6) Of this time, 37 %represented time using Spanish with their host families Per the programrequirements, students had at least two meals per day with their host families

In response to a separate question and as stated in Section 3 above, learnersestimated spending 16.8 hours per week doing activities with their host families(SD = 12.6), 5.8 hours per week going out for food or drink (SD = 5.6), and 16.0hours per week doing tourist activities and shopping (SD = 9.9) They took twouniversity courses, selecting from a sixth semester Spanish language course, anadvanced culture course on current events and policies in Spain, and anadvanced art class that included weekly visits to the Prado museum None ofthe courses included pragmatics or requests as specific course topics The sixthsemester Spanish course addressed the past subjunctive and conditional forms,yet no specific attention was paid to the possibility of using these structures forrequests Learners also participated in excursions organized by the program andexplored Madrid during their free time The Spanish native speaking (NS)participants were from Madrid, Spain (mean age = 23.93, SD = 6.7) They all haduniversity education or were university students at time of testing

3.2 Procedures and instruments

Learners completed a computerized oral discourse completion task at thebeginning and end of the program, within three days of the beginning andend of the 6-week program.3 The same task was used at both times with theexception of a different randomized order of test items The native speakers ofSpanish completed the same production task once during the program Therewere fifteen test items and seven distractors Five test items represented thefood and drink context, five represented the general merchandise context, andfive represented the familial context (Table 2) By eliciting five repetitionswithin each context from each participant, this design provided 510 learnerand 225 native speaker requests for analysis Nine learner responses and onenative speaker response were not included in the analysis due to lack ofresponse, absence of a request, or missing data in one case The resultinglearner (501) and native speaker (224) requests offered a satisfactory number oftokens per participant per context for analysis The distractor items related tospeech acts other than requests

During the task, participants heard a preamble and saw a photo of the scenario,collectively functioning to frame the request in the food and drink, generalmerchandise, or familial context Following a prompt, which indicated theaddressee, participants made a request (Example 1) All requests were audiorecorded

Example 1 Oral discourse completion task text

Preamble: Rosa está en una cafetería y quiere pedir una ensalada mixta

‘Rosa is in a cafeteria and wants to order a salad.’

Table 2: Requested items and addressee by context.

establishments General

Merchandise

movie ticket, phone card, shoe store, tobacco

store, location in city

customer service agents

3 For other elicited production tasks, see Cuza (2013), Czerwionka (2010), Félix-Brasdefer (2007), or Schauer (2004).

Prompt: Ahora viene el camarero y Rosa le dice…

‘The waiter is coming and and Rosa says to him … ’Sample Request: Para mí, una ensalada mixta

‘For me, a salad’

3.3 Data coding and analyses

Each request was coded according to the personal deictic orientationand directness of the speech act Within orientation, requests were hearer-oriented (i.e., second-person reference), speaker-oriented (i.e., first-personreference), or ambiguously-oriented (i.e., verbal ellipsis) For directness,hearer-oriented requests were produced with either imperative or hearer-oriented interrogative (Interrogative-H) clauses, where imperatives aredirect and interrogatives are conventionally indirect (Blum-Kulka et al.1989; Blum-Kulka 1989) Speaker-oriented requests were speaker-orientedinterrogatives (Interrogative-S) or declaratives; Interrogative-S clauses areconventionally indirect, and declaratives are direct requests, followingBlum-Kulka et al (1989) Blum-Kulka (1989) provided various examples

of conventionally indirect requests including those with modal verbs(¿Me puedes pasar la sal? ‘Can you pass me the salt?’) and without modalverbs (¿Me pasas la sal? #‘You pass me the salt?’); Both types of interroga-tives were coded as hearer-oriented interrogatives and conventionally indir-ect These, like the speaker-oriented interrogatives (Can I have the salt?), can

be interpreted literally as interrogatives or as requests, demonstrating theconventionality of these utterances when interpreted as requests (Blum-Kulka

et al 1989: 41–43) Examples of each combination of orientation and ness are listed in Table 1 (Section 2)

direct-The data were collapsed by participant and context, resulting in tional data related to orientation and directness for each participant, in each ofthe three contexts The data analyses addressed orientation (Section 4.1) anddirectness (Section 4.2) separately

propor-For orientation and directness, three steps to analysis were conducted First,

a descriptive analysis of the requests produced by learners at the beginning ofthe program (L2-pre), learners at the end of the program (L2-post), and nativespeakers (NS) was conducted Second, learners’ change in requests was ana-lyzed by comparing L2-pre and L2-post requests, using individual linear mixedeffects models for each of the dependent variables (Orientation: hearer-oriented,speaker oriented, ambiguously oriented; Clause directness: imperative, interro-gative-H, interrogative-S, declarative) with context (food and drink, general

merchandise, familial) and group (L2-pre vs L2-post) as fixed effects and subject

as a random effect Subsequent planned multiple comparisons using leastsquare means analyses with Tukey-Kramer adjustments were also conducted

to compare group performance in the three contexts Third, as a measure of thepragmatic appropriateness of learners’ requests at the beginning and end of theprogram, learner data were compared to the native speaker data (L2-pre vs NSand L2-post vs NS) A similar set of linear mixed effects models was used, andindividual models for each dependent variable listed above were conducted withcontext (food and drink, general merchandise, familial) and group (L2-pre vs NSand L2-post vs NS) as fixed effects and subject as a random effect Subsequentplanned multiple comparisons using least square means analyses with Tukey-Kramer adjustments were also conducted to compare group performance in each

of the three contexts All statistics were conducted using SAS 9.3.4

4 Results

4.1 Request orientation

Based on the descriptive analysis for request orientation (Figure 1), learnersincreased the use of hearer-oriented requests over the program, approaching NSuses in the food and drink, general merchandise, and familiar contexts Whileincreasing the use of hearer-oriented requests, learners decreased the use ofspeaker-oriented requests in all three contexts, moving towards NS norms

Figure 1: Descriptive data for request orientations for L2-pre, L2-post, and NS.

4 The limited ambiguously-oriented data did not allow for statistical tests, and thus only descriptive data are presented for ambiguously-oriented requests.

Learners’ use of ambiguously-oriented requests seemed to differ in the threecontexts, but remained low for all groups in all contexts Taken jointly, thesegeneral trends indicate pragmatic improvement for learners in their use of hearer-and speaker-oriented requests since in all contexts learners’ usage approachedthat of native speakers over the program For samples of learners’ hearer- andspeaker-oriented requests to aid understanding of the descriptive and statisticalresults, see examples 2–8 and 9–15 in Section 5.

To investigate change in learners’ request orientations, linear mixedmodels and multiple comparisons were conducted Table 3 includes groupcomparisons A significant increase in hearer-oriented requests was foundoverall, and more specifically within the food and drink and familial con-texts A significant decrease was found with the use of speaker-orientedrequests overall, which can be attributed to a significant decrease withinthe familial context

Comparing learners to native speakers (Table 4), the learners’ use of oriented requests overall did not align with that of native speakers at thebeginning of the program At the end of the program, there was still a sig-nificant difference between learners and native speakers overall, but learnersachieved more appropriate uses of hearer-oriented requests in the food anddrink and general merchandise contexts over the program For speaker-oriented requests, learners and native speakers differed significantly in their

hearer-Table 3: Orientation: Pragmalinguistic group analyses (L2-pre vs L2-post linear models and Tukey-Krammer comparisons).