3.1 Introduction
In Chapter 1 (Section 1.4) I presented my assumptions about an integrated pragmatic-discursive approach, and in Chapter 2 (Section 2.6) I described the approach to service encounters I will adopt in this book. Taking a pragmatic-discursive perspective, in this chapter I offer a contrastive analysis of the similarities and differences when negotiating service at comparable in-store supermarket delicatessens in the United States and Mexico. My data are representative of two regions: Indiana/the Midwest (United States) and Yucatán (Mexico) (see Introduction, Figures 2 and 3). As noted in the Introduction, cross-cultural pragmatics is contrastive, as it compares “data obtained independently from different cultural groups” (Spence-Oatey Reference Spencer-Oatey and Spencer-Oatey2000: 4).1 In this cross-cultural analysis, I adopt the “global” perspective of linguistic variation (see Fried Reference Fried, Fried, Östman and Verschueren2010) to examine pragmatic variation in two languages. I analyze cross-linguistic variation according to the following levels of pragmatic analysis described in Chapter 1 (Section 1.4.6): actional (request variants of the request for service), interactional (joint social actions), and organizational (turn-taking practices). The data are also analyzed with regard to gender differences between the customer and the server. Although the focus of this chapter is on the language used to negotiate the sales transaction (transactional talk), it will be shown that relational talk (e.g. small talk, jokes, laughter) is embedded in transactional talk in order to show changes of footing or alignment (Goffman Reference Goffman1981) (see Introduction, Figure 1).
This chapter is organized as follows: after a brief sociolinguistic description of the regions of the United States and Mexico under analysis (Section 3.2), the setting and the data collection procedures are described (3.3). The sequential structure of openings in service encounters is presented in Section 3.4.1, followed by an analysis of both the request variants and variation by gender in customer requests (3.4.2). An analysis of the request–response sequence is presented in Section 3.4.3, followed by an examination of terminal exchanges (3.4.4) and an analysis of the turn-taking structure of social action (organizational level) (3.4.5). In Section 3.5 I present the conclusions of the chapter.
3.2 Cross-cultural pragmatic variation in the United States and Mexico
The contrastive analysis is based on two regions, one in Indiana (United States) and one in Yucatán (Mexico). The Indiana English variety is part of the US Central or Midland region (Labov, Ash, and Boberg Reference Labov, Ash and Boberg2006), while the Yucatán variety of Spanish is considered a Southern dialect with individual linguistic characteristics that distinguish it from other Mexican dialects (Lipski Reference Lipski1994; Martín Butragueño Reference Martín Butragueño, Villanueva and Butragueño2014). Although previous research has concentrated on linguistic variation (e.g. phonology, morphology, syntax) in the Indiana variety (e.g. Ash Reference Ash, Murray and Lee Simon2006; Fogle Reference Fogle2008; Labov et al. Reference Labov, Ash and Boberg2006) and Yucatán variety (e.g. Lipski Reference Lipski1994; Michnowicz Reference Michnowicz, Holmquist, Lorenzino and Sayahi2007, Reference Michnowicz2011), this chapter centers on cross-cultural variation at the pragmatic and discursive levels in these two language varieties; namely, variation at the actional (speech acts), interactional (speech act sequences), and organizational (overall structure of the interaction and turn-taking format) levels (see Chapter 1, Section 1.4.6).
From a sociolinguistic perspective, Wolfram and Schilling-Estes (Reference Wolfram and Schilling-Estes2006) provide evidence for regional pragmatic variation in US regions with regard to differences in speech acts, address forms, politeness, openings, and overlapping talk. They make a strong case for sociolinguistic and pragmatic variation across cultural groups:
Speakers of all languages and dialects are quite capable of performing the same basic kinds of speech acts – directing, requesting, apologizing, and so forth – but how these speech acts are carried out and the conditions under which they are considered to be appropriate varies considerably across cultural groups.
Cross-cultural pragmatic variation is generally analyzed from a global perspective (Fried Reference Fried, Fried, Östman and Verschueren2010) by contrasting some aspect of the linguistic system – for example, the phonology, morphology, syntax, or pragmatics – of two or more languages. Although cross-cultural pragmatics often comprise situations of intercultural communication, interlanguage pragmatics, or lingua franca (Boxer Reference Boxer2002; Kecskes, Davidson, and Brecht Reference Kecskes, Davidson and Brecht2005; Schneider Reference Schneider and Fried2010), in this chapter I use the term cross-cultural pragmatics to refer to variation at some level of pragmatics (e.g. actional [speech acts] or interactional level [speech act sequences]). I contrast varieties of two languages that are part of monolingual communities of practice, that is, face-to-face interactions between native speakers of the same language, customers and servers at a designated location, such as Mexicans in Mexico and Americans in the United States (in the Introduction I explained my understanding of a community of practice in service encounters).
3.3 Setting and data collection procedures
The physical setting in which the transactions took place was comparable in the United States and Mexico. The supermarkets, which included on-site delicatessens, were national chains in both countries. The delicatessens in each country were similar in size and in products sold (e.g. cold meats, salads, cheeses, etc.), and each was located at a designated place inside the supermarket. In both locations, customers did not pay for the product at the conclusion of the transaction; instead, the purchase was completed at the checkout where all items were scanned and paid for. Figure 3.1 illustrates the description of the setting where the sales transactions took place, including the position of the customer and server, the display of the products for sale, and the placement of the audio tape recorder.
Figure 3.1 Supermarket delicatessen setting
The 1,400 recorded face-to-face interactions consisted of approximately 70 hours of audio-recorded data: 700 interactions in Indiana (40 hours) and 700 in Mexico (30 hours) (see Introduction, Table 1, for additional details regarding the description of the corpus). The population from each region can best be described as middle-class individuals from monolingual communities of practice, and the customers as regular clients who visit the neighborhood supermarket once or twice a week. The servers are also members of the same community of practice who are familiar with the customers as a result of their frequent transactions. In this book, the service encounters are considered to be representative of each sociocultural setting, specifically since, according to Mills, “decisions about what is appropriate or not are decided upon strategically within the parameters of the community of practice and within the course of the interaction rather than being decided upon by each individual once and for all” (Reference Mills2003: 235).
With regard to the distribution of gender, female customers outnumbered male customers in each region: Indiana (male [39% or 276/700 interactions]; female [61% or 424/700]) and Mexico (male [32% or 225/700]; female [68% or 475/700]). Although female customers predominated in the corpus, the distribution of male customers is fairly comparable in both sets of data. For the analysis of gender, a subset of the data from each setting was compared: 79% (552/700) of the American data, with 39% for each gender (males [276/700] and females [276/700]), and 64% (450/700) of Mexican data, with 32% for each gender (males [225/700]; females [225/700]).
3.4 Negotiating service in cross-cultural service encounters
3.4.1 Opening the interaction
This section examines the interactional resources used to open a sales transaction in the US and Mexican supermarket delicatessens. In both settings the request for service was initiated by prefatory interactional resources, such as a summons–answer sequence, presence or absence of greetings, and an offer–response sequence.
Various resources were employed to initiate a sales transaction prior to the request for service. Three main sequences were identified to open the transaction, namely, a summons–response sequence, greeting exchanges, and offer of product-response (Mexican data only). (Instances of pre-requests [pre-sequences] will be analyzed in the interactional level [Section 3.4.3]). These sequences occurred in strict adjacency pairs and in different combinations. The summons–response sequence signals the beginning of the interaction and secures the attention of both parties prior to the opening of the sales transaction. This is an instance of a preliminary (Schegloff Reference Schegloff1968, Reference Schegloff1986) or realignment of the interaction that leads to the sales transaction. Verbal and non-verbal resources were used in the summons–response sequence, as in the following examples (customer [C] and server [S]). (See Transcription conventions at the beginning of this book.) In the English translation, the formal address form (you-formal) is marked as V (V), while the informal address form is marked as T (T). The arrow (→) signals the summons–response sequence:
[1] US summons–answer sequences
a. Female server; male customer
01 → ((Customers standing)) 02 S: → Can I help who's next? 03 C: can I have the honey turkey ↑, sliced thin? b. Male server; male customer
01 → ((Customer standing)) 02 S: → Have you been helped, sir? 03 C: → Nope 04 S: can I help you? 05 C: a pound of the mesquite smoked turkey breast Private Selection ↑
[2] Mexican summons–answer sequences
a. Female interlocutors
01 → ((Customers standing)) 02 S: → ¿Quién sigue? Who's next? 03 buenas tardes good afternoon 04 C: jamón de pierna Kir. Kir ham. b. Female server; male customer
01 → ((Customer standing)) 02 S: → Setenta y dos ↑ seventy-two ↑ 03 ((bell rings to signal ticket number)) 04 ((customer approaches server's area)) 05 S: ¿Qué va a llevar señor? What are youV gonna have sir? 06 buenas tardes, jamón de pavo ↑ good afternoon, turkey ham ↑ 07 C: doscientos gramos de jamón Virginia Fud ↓ two hundred grams of Fud Virginia ham ↓ c. Female interlocutors
01 ((customer standing)) 02 S: Setenta y cinco, setenta y seis Seventy-five, seventy-six 03 ((bell rings to signal ticket number)) 04 C: yo – setenta y cinco ↑ me – seventy-five ↑ 05 S: ¿qué va a llevar? La atiendo – jamón de pavo ↑ what are youV gonna have? I can help youV – turkey ham ↑ 06 C: pimiento del Fud. Fud pimiento loaf.
Examples (1)–(2) show various formats of the summons–answer sequence. A summons–answer sequence is also referred to as an “attention-getting device,” and it often occurs at the beginning of a two-part sequence (Schegloff Reference Schegloff1968: 1080). In the examples in (1a–b) and (2a) the customer's presence (while waiting his/her turn to be served) (line 01) summons the server's attention to reorient the interaction (line 02). Unlike in the US setting, in the Mexican delicatessen the response to the customer's presence was realized by a bell ringing and a server calling out the customer's number. The server's call ended with a high rising terminal that functioned to secure the interlocutor's response (e.g. seventy-two ↑, example 2b, line 02). In example (2c), the Mexican customer identifies herself in response to the server's call (lines 01–04), following which the server opened the transaction (line 05). In these examples, the summons–answer sequence serves to realign the participants’ roles prior to the server's opening of the transaction (example 1a, line 02; example 1b, line 02; example 2a, line 02; example 2b, line 05; example 2c, line 05) and the customer's request for service (example 1a, line 03; example 1b, line 05; example 2a, line 04; example 2b, line 07; and example 2c, line 06). More specifically, the summons–answer sequence serves to establish common ground (Clark Reference Clark1996) between the customer and the server based on common expectations and assumptions with regard to the sales transaction, such as the interactional roles of both customer and server, and their assumptions about the service setting.
After establishing common ground and alignment of the participants’ interactional roles through the summons–answer sequence, the sales transaction often opened with the server's allocation device (e.g. ‘Can I help whoever is next?’). No gender variation was noted in the selection of this strategy in either country. The following examples represent variation in the expressions commonly used by the server to open the transaction (example 3, US data; example 4, Mexican data) (address forms are marked in bold):
[3] Pragmalinguistic variation in the server's allocation devices (US data)
a. Male servers
Have you been helped, sir/ma'am? I can help whoever's next Can I help who's next/somebody? OK, may I help who's next? Alright can I help whoever's next? Does anyone need help? b. Female servers
May I help who's next? Can I help you, sir? Who's next? Can I help the next person? OK, who's next here? Have you been helped? Have you been waited on?
[4] Pragmalinguistic variation in the server's allocation devices (Mexican data)
a. Male servers
Buenas tardes amiga, ¿ya le atendieron? Good afternoon, girlfriend, have youV been helped? Buenas tardes señora, ¿qué va a llevar? Good afternoon ma'am, what are youV gonna have? ¿Ya le atendieron? Have youV been helped? ¿Ya atienden a la señora? Is anyone helping the lady? Buenas noches, ¿qué va a llevar? Good evening, what are youV gonna have? ¿Usted qué va a llevar? What are youV gonna have? Buenas noches nena, ¿qué le servimos? Good evening babe, what can we get youV? b. Female servers
Buenas tardes reina, ¿qué va a llevar? Good afternoon queen, what are youV gonna have? Seño, ¿qué le sirvo? buenas tardes Ma'am, what can I get youV? good afternoon ¿Qué buscaba, señorita? What were youV looking for, Miss? ¿Qué va a llevar doña*? What are youV gonna have, miss/madam ¿Qué vas a llevar, amigo? What are youT gonna have, friend? Buenas tardes corazón – ¿ya le atienden? Good afternoon sweetheart – are they already helping youV? ¿Qué va a llevar? buenas tardes – ¿qué le sirvo señora? What are youV gonna have? Good afternoon – what can I get youV ma'am? ¿Algo que le pueda servir, amiga? Can I get youV anything, girlfriend? ¿Qué desea, caballero? What would youV like, sir? Señora, ¿la atiendo? Ma'am, can I help youV? Buenas tardes, amiga – ¿le puedo attender? Good afternoon, girlfriend – can I help youV? * Doña is a deferential address form that is used before a woman's first name to express respect. It can be loosely translated as ‘miss’ or ‘madam.’
One difference between the two settings is reflected in the presence and variation of address forms (I take up the issue of the pragmatic function and variation in address forms in Chapter 8). US servers mainly utilized the deferential forms, ‘sir/ma'am’ (example 3), while Mexican servers used a wide inventory, ranging from formal expressions (e.g. caballero ‘gentleman’; señora ‘madam’; señor ‘sir’; reina ‘queen’; doña ‘Ms/madam’ [with respect]) to forms of address that express involvement or solidarity toward the customer (e.g. amigo/a ‘male or female friend’; corazón ‘sweetheart’; nena ‘babe’). The Mexican servers often opened the transaction with a greeting (e.g. buenas tardes ‘good afternoon’) and the deferential address style to express distance, as shown in the verbs or in the pronominal form V (e.g. ¿Usted qué va a llevar? ‘what are youV going to have?’). Based on the selection of the address form, the Mexican servers adopt an interactional style that expresses involvement (person-orientedness), while the US servers adopted a business-like style in the transaction (task-orientedness) (Fant Reference 260Fant, Ehlich and Wagner1995; Placencia Reference Placencia2005). The frequent deferential forms used in Spanish express respect and involvement with the interlocutor.
Immediately following the summons–answer sequence, and prior to the request for service, the opening of the transaction included a variety of greeting formulas and an offer–response sequence. Table 3.1 shows the distribution of the opening devices identified in the US and Mexican data prior to the issuance of the request (1,400 transactions [700 in each country]).
Table 3.1 Distribution of opening devices in US and Mexican supermarket delicatessens
| Opening device | US delicatessens | Mexican delicatessens |
|---|---|---|
| Customer initiates transaction with greeting | 7.3% (51/700) | 4.9% (34/700) |
| Server initiates transaction with greeting | 4.6% (32/700) | 14.4 % (101/700) |
| Reciprocal greeting | 10.7% (75/700) | 11.4% (80/700) |
| Customer initiates transaction with no greeting | 40% (280/700) | 41.9% (293/700) |
| Server initiates transaction with no greeting | 37.4 (262/700) | 27.4% (192/700) |
| Total | 100% (700) | 100% (700) |
Greetings were often used to open the transaction by both the server and the customer. The examples below show the placement of the greeting in the opening sequence:
[5] Greetings in the opening sequence
a. Reciprocal greeting (English data: male server; female customer)
01 ((Customer standing)) 02 S: → Hi 03 C: how are you doin’? 04 S: good, yourself ↑ 05 C: Alright 06 I would like a half a pound of the lemon pepper chicken breast. b. Reciprocal greeting sequence (Mexican data: females)
01 ((Customer standing)) 02 S: → Buenas tardes ↑ Good afternoon ↑ 03 C: Buenas good afternoon 04 S: ¿qué le servimos? what can we get youV? 05 C: queso de ese – mi marca Manchego. that kind of cheese – Manchego brand. c. Non-reciprocal greeting (Mexican data: females)
01 ((Customer standing)) 02 S: → Buenas tardes, reina, ¿qué va a llevar? Good afternoon, queen, what are youV gonna have? 03 C: me da 100 gramos de:: jamón. give meV100 grams of:: ham. d. Non-reciprocal greeting (English data: male server; female customer)
((Customer standing)) C: Hi, can I have a pound of your, ah Cajun turkey, please? S: how would you like that sliced?
[6] Opening transaction with no greeting (English data: female server; male customer)
01 S: → Can I help you, sir? 02 C: Uh– I'd like, uh, a:bout a pound of tavern ha:m ↑
Opening the transaction with a reciprocal greeting (one or two adjacency pairs) was realized by means of common routine formulas (5a–b), and it was preferred almost equally in English (10.7% or 75/700) and Spanish (11.4% or 80/700). As shown in Table 3.1, cross-cultural differences were noted in the person who initiated the interaction: Mexican servers initiated the transaction with a non-reciprocal greeting more frequently (14.4% or 101/700) than the customer (4.9% or 34/700) (5c), whereas in the American delicatessen it was the customer who opened the transaction with a greeting (7.3% or 51/700) more frequently than the server (4.6% or 32/700) (5d). The most common greeting forms in English included ‘hi,’ ‘hello,’ ‘hey,’ and ‘how're you doing?’, while the Mexican greetings comprised buenos días ‘good morning,’ buenas tardes ‘good afternoon,’ and buenas noches ‘good evening.’ The hola + greeting opening (hola, buenas tardes ‘hello, good afternoon’) was infrequent in the Mexican data. However, the majority of encounters in both the Mexican (69.3% or 485/700) and American delicatessens (77.4% or 542/700) did not contain any greeting expression (these numbers combine both categories shown in Table 3.1: customer- and server-initiated transaction with no greeting). A Chi-square test for independence showed that the differences are significant, X2 (1, N = 1,400): 11.46, p < .001. That is, at a delicatessen the near absence of a greeting expression to preface a request for service is not culture-specific; instead, opening a transaction with no greeting represents a cultural expectation in the regions examined in Mexico and the United States. This absence of greetings does not represent an instance of impolite behavior. In both contexts, the transactions were frequently initiated by the server, as in example (6).
Cross-cultural variation in opening sequences should be examined through the lens of previous research. While greetings were almost always present (90%) when opening service encounters in small shops in France (Kerbrat-Orecchioni Reference Kerbrat-Orecchioni2006) and in small shops in Ecuador (Placencia Reference Placencia2005, Reference Placencia, Schneider and Barron2008), in the context of supermarket delicatessens in the United States and in Mexico, greetings are hardly ever used. This difference, however, can be due to the type of setting: greeting sequences appear to be more frequent in less formal service encounter settings, as in small shops, while the near absence of greeting sequences is not expected in delicatessens due to the formality of the setting (a deli section inside a supermarket).
Finally, in addition to the summons–answer sequence and the greeting exchanges, the opening sequence included a suggestion on the part of the Mexican server to buy a particular product or brand. The placement of this action occurred in different positions: prior to the customer's initial request for service, or prior to or after the greeting exchange. Server-initiated offers were present in 23% of the Mexican data (or in 160/700 interactions), and absent from the American data. Examples of server-initiated offers in the Mexican data are shown in (7):
[7] Initiated offers prior to the request for service
a. Mexican data (females)
((Customer standing waiting to be called)) S: → Va a llevar jamón, doña ↑ – ¿qué jamón le sirvo? Are youV having ham, madam ↑ – what kind of ham can I get youV? ¿cuánto va a llevar? how much are youV having? C: Deme – 300 gramos. give meV – 300 grams. b. Mexican data (females)
01 ((Customer standing waiting to be called)) 02 S: → Va a llevar jamón ↑, buenos días = Are youV gonna have ham↑, good morning = 03 C: = buenos días = good morning 04 S: → de pavo, de pierna, pechuga ↑ ¿de cuál buscaba? turkey ham, leg, breast ↑ which one were youV looking for? 05 C: jamón ahumado ↓ smoked ham ↓ c. Mexican data (females)
01 ((Customer standing waiting to be called)) 02 S: ¿Qué buscaba señorita? What were youV looking for miss 03 C: quiero – medio de jamón americano Yasa ↓ I want – half (a kilo) of Yasa American ham ↓ 04 S: → tengo el americano Ibermet, también en oferta I also have the American Ibermet on sale 05 lo tengo a treinta it costs 30 ((pesos)) 06 le gustaría probarlo ↑ would youV like to try it ↑ 07 C: sí, por favor ↓ yes, please ↓ ((after trying product, customer orders product))
Server-initiated offers were often employed to open the transaction in the first turn, followed by the customer's acceptance of the offer (7a). Other offers were realized by means of two turn-constructional units (TCUs) (Sacks et al. Reference Sacks, Schegloff and Jefferson1974) (7b, lines 02 and 04), followed by the customer's compliance (line 05). Example (7c) shows the offer placed in the third turn (line 04), after the customer's request (line 03). In this case, the offer was used by the server to persuade the customer to change her mind and buy the suggested brand. This represents a realignment of the participants’ roles: after the issuance of the offer for service-response (7c, lines 04–05), the server redirects the interaction with a suggestion to try the product, followed by the customer's acceptance (lines 06–07). All server-initiated offers (7c, line 06) ended with a high terminal (↑) to secure a compliance response, which often ended in a low terminal (↓) (7c, line 07).
In contrast, in the US delicatessen the server complied with the customer's request and did not try to dissuade the customer with unsolicited offers for a different brand or product. In this asymmetric interaction, the interactional roles are negotiated differently in each cultural setting: while the server in the US delicatessen respects the customer's initial request for service (a preference for independence [Scollon and Scollon Reference Scollon and Scollon2001]), an offer from the Mexican server represents an expected sociocultural move that is successfully received by the customer as an instance of affiliation (Scollon and Scollon Reference Scollon and Scollon2001).
In the Mexican server–customer relationship, the role of the server is that of advice-provider, while in the US context it is that of information-giver. In her discussion of self-service bookshop encounters in Italy and England, Gavioli (Reference Gavioli and Caliumi1993) initially noted this difference: the Italians favored the roles of advice-seeker (client) and advice-giver (assistant), while in the British data the client and the assistant played the roles of information-seeker and information-giver, respectively. In the Mexican delicatessens analyzed, Mexican servers focused on the needs of the customers and expressed alignment as advice-givers, while the customers aligned as advice-seekers, ultimately complying with the server's advice to buy a different brand or product. After successful alignment of the opening sequence, the customer issued the request sequence, as shown in the next section.
3.4.2 Pragmalinguistic variation in the request for service
This section focuses on pragmalinguistic variation in the use of the linguistic resources utilized to perform a request for service, as well as the elements used to internally modify the request. Figure 3.2 shows the distribution of the customer-initiated requests found in the US and Mexican data (1,400 requests [700 per group]). The percentages are presented above each bar, and the frequencies for each variant can be seen at the bottom of the figure. This figure shows the seven request variants available for customers to issue a request for service in the US and Mexican delicatessens, along with different degrees of preference (see Table 1 in Chapter 1 for the classification of requests for service).
Figure 3.2 Cross-cultural variation in requests for service in US and Mexican supermarket delicatessens
There is cross-cultural variation in two variants used to make a request for service: US customers showed a strong preference for conventional indirectness (46.3% or 324/700 requests) (e.g. ‘Can/could I get a pound of roast beef?’), while the Mexican customers preferred elliptical (verbless) requests (42.9% or 300/700). In addition, the US customers used want statements as the second most frequent variant (19.1% or 134/700), followed by three additional requests (each with 9–12% of use): elliptical (12% or 84/700), need statement (9.9% or 69/700), and assertion (9.3% or 65/700). The imperative and Yes/No interrogative forms were the least frequent in English (3.3% and below). On the other hand, the Mexican customers employed two variants that were each used 24% of the time (assertion [24.1% or 169/700] and imperative [23.9% or 167/700]). Four variants were utilized infrequently in the Mexican data (5% and below): want statement, need statement, Yes/No interrogative, and conventional indirectness. A Chi-square test for independence indicates that the difference in the use and non-use of each variant is significant in both cultures: conventional indirectness, X2 (1, N = 1,400) = 379.35, p < .0001; elliptical, X2 (1, N = 1,400) = 165.88, p < .0001; imperative, X2 (1, N = 1,400) = 124.52, p < .0001; assertion, X2 (1, N = 1,400) = 54.44, p < .0001; and, want statement, X2 (1, N = 1,400) = 62.98, p < .0001. This finding indicates a greater probability that the use and non-use of each of these variants is culture-specific.
The examples in (8) show the most frequently used customer-initiated requests for service in the American data (8 a–e [conventional indirectness, need statement, want statement, elliptical, and assertion]) and Mexican data (8 f–h [elliptical, imperative, and assertion]) (the pragmatic effect of rising terminals is examined in the analysis of the request–response sequence [interactional level, Section 3.4.3]).
[8] Cross-cultural variation in requests for service
a. Could I – can I get a pound of the Maple Glazed, um, turkey?
(United States, conventional indirect, female)
b. Um – I need a pound of Boar's Head sweet sliced shaved ham, please
(United States, need statement, female)
c. Um, I'd like a pound of the uh, Maple Glazed Honey Coat ↑, shredded please?
(United States, want statement, male)
d. honey-smoked turkey breast (United States, elliptical, female)
e. I'll take three-quarters of a pound of the Oven Gold turkey breast ↑
f. Pechuga de pavo San Rafael 300 gramos
bien delgadito por favor (Mexico, elliptical, female)
San Rafael turkey breast 300 grams
very thinlyD sliced please
g. Deme medio de salchicha Iberomex (Mexico, imperative, female)
Give meV half (a kilo) of Iberomex sausage
h. Me da 300 gramos de salchicha de pavo Swan ↑ por favor
(Mexico, assertion, female)
Give meV 300 grams of Swan turkey sausage ↑ please
Cross-cultural differences were also found with regard to the placement of the request in the interaction. Although the initial customer request was frequently placed in the second turn in both cultural settings, the US customers used this second position more frequently (71% or 494/700) than the Mexicans (41% or 289/700). Among the US customers the initial request was also placed, although less frequently, in the first (7.4%), third (2.6%), fourth (11.9%), or fifth and later turns (7.6%). Unlike the US requests, the Mexican customer-initiated requests were often placed in the first (25%), third (8.7%), fourth (6.9%), or fifth or later turns (18%). As I will explain in the analysis of the request–response sequence (Section 3.4.3), pre-sequences (i.e. requests for availability) were more frequent in the Mexican (23% or 160/700) than in the US transactions (11% 75/700). Thus, more variation was noted in the Mexican data with regard to the placement of the initial request, where the request for service was predominantly placed in the second, first, and fifth or later turns.
Although infrequently used, a direct question (Yes/No interrogative) followed by the server's response was more frequent in the Mexican data (1.9%; 13/700), while it was used only once in the US corpus. Yes/No interrogatives (e.g. ‘do you have … ?’) require a pragmatic interpretation on the part of the server: the request is understood as a request for service, not as a request for information. In addition, elliptical (or verbless) requests that ended with a final rising intonation (↑) were also interpreted as a request for service. Examples (9a–b) show a Yes/No interrogative (marked by →) in Spanish and English. This interrogative is interpreted as a request for service by the server (line 03 [9a] and line 02 [9b]). And example (9c) is an elliptical request with a high terminal (↑, line 02) that is also interpreted as a request for service by the server (line 03) (formal address forms are marked V and informal as T. Nominal address forms [vocatives] are in bold).
[9] Questions following questions
a. Yes/No interrogative (Mexico: female customer; male server)
01 S: Buenas tardes señora, ¿qué va a llevar? Good afternoon ma'am, what are youV gonna have? 02 C: → tiene del jamón virginia de Fud ↑ do youV have Fud Virginia ham ↑ 03 S: ¿cuánto? how much? 04 C: 300 pero de este 300 (grams) but this kind 05 S: OK. OK. b. Yes/No interrogative (United States: female customer; male server)
01 C: → Do you have the honey ham? 02 S: do you want Boar's Head or Private Selection? 03 C: um: Boar's Head sounds good. c. Elliptical request (Mexico: female interlocutors)
01 S: Buenas tardes Good afternoon 02 C: → jamón selecto de Fud ↑ Fud select ham↑ 03 S: ¿cuánto, corazón? how much, sweetheart? 04 C: un cuarto a quarter (kilo) 05 S: un cuarto ↓ a quarter ↓
Yes/No interrogatives or elliptical questions with a final rising intonation may be interpreted as a request for information or as a request for service. The high terminal (↑) serves as an interactional (prosodic) resource in discourse to secure hearer's uptake. In the examples in (9), the server's response shows that the customer's Yes/No interrogative is understood as a request for service, and this is achieved through pragmatic interpretation (lines 02–03 [a], lines 01–02 [b], lines 02–03 [c]). The phenomenon of questions following questions was initially identified by Merritt (Reference Merritt1976b) with data of service encounters at a US self-service store. A Yes/No interrogative in the question–question adjacency pair may pose an apparent violation of conditional relevance in the organization of turn-taking. According to Schegloff, “[t]he relevance of some turn which can be a second pair part is conditional on the occurrence of a first pair part from the same pair type” (Reference Schegloff2007: 20). In the adjacent sequences in the examples in (9) marked with →, an apparent absent move on the part of the server (e.g. C: Do you have the honey ham?, S: ‘yes, we do’) does not constitute a violation of conditional relevance. Instead, the second-pair part (i.e. the server's response in the form of a question) is conditional to the occurrence of the first-pair part (i.e. the Y/N interrogative). The adjacency pair is organized in terms of the absence of an elliptical direct move that can be recovered during the interlocutor's final interpretation (second-pair part). As stated by Schegloff, relevance rules represent a central sequential organization device “that binds actions together into coherent sequences” (p. 21). The second-pair part of the question-following-question sequence requires a level of pragmatic and sociolinguistic competence necessary for the interpretation of the request.
With regard to internal modification of the request, requests for service in English and Spanish were infrequently modified (see Chapter 1, Section 1.4.6 [actional level] for the classification of request variants and examples of lexical and syntactic internal modifiers). Lack of internal modification of the request was more frequent in the Mexican data (83.9% or 587/700 interactions) than in the English data (66.4% or 465/700). Requests for service were more likely to be modified in English (33.6% or 235/700) than in Spanish (16.1% or 113/700). The politeness marker, ‘please’ (por favor), was the most frequent internal modifier in both the US (14.6% or 102/700) and the Mexican initial requests (11.7% or 82/700), used once in each customer-initiated request (see examples 8b, c, h). The conditional form was frequently used in English for conventional indirect requests and in want statements (e.g. ‘I'd like’) (14.9% or 104/700) (example 8c), whereas the Mexicans sometimes used a diminutive form to soften the request (2.4%; 17/700) (example 8f). Two or more modifiers were also infrequently employed in one request in English (2% or 14/700) and in Spanish (1.4% or 10/700), as shown in (8f), with a diminutive form (delgadito ‘thinlyD sliced’) and por favor (‘please’).
The presence of high terminals (↑) was realized by means of a prosodic downgrader (Félix-Brasdefer Reference Félix-Brasdefer, García and Placencia2011; Placencia Reference Placencia, Schneider and Barron2008), and it occurred in cases where the request for service ended with a final rising intonation, as in the examples in 8 (e and h) and 9 (a [line 02], and c [line 02]). While the low final rising intonation is characteristic of assertions, elliptical requests, and imperatives (declarative statements) (unmarked intonation), a high terminal represents a marked intonation and conveys a pragmatic (interpersonal) meaning that is recovered via an inferential process, such as a politeness effect to the vendor, or compliance with the request. Although it is difficult to show an acoustic analysis of the request due to background noise in the supermarket from music or simultaneous transactions, both the author and the Mexican assistant who transcribed and coded the data agreed that there was a clear final rising intonation. Thus, based on our judgment, the melodic curve of an assertion (me da ‘you give meV’) is represented by 10a (line 01), and the melodic curve of an imperative is shown in 10b (line 01). Both requests end with a high terminal:
[10] Requests for service with a high terminal
a. Assertion (female server; male customer)
b. Imperative (females)
In these requests, the upward arrow represents an interactional (prosodic) resource to express that the request will be warranted by the server (line 02 in each example). Following previous research on prosody and pragmatics (Culpeper Reference Coupland2011; Wichmann and Blakemore Reference Wichmann and Blakemore2006), the high terminal conveys a polite function and secures the uptake by the interlocutor. In these examples the prosodic downgrader is reinforced with an additional internal modifier, namely, ‘please’ (por favor) (10a, line 01) and ‘a little bit’ (poquito) (10b, line 01) in each example. The internal modifiers express a polite interpretation by softening the effects of a direct request.
In the context of supermarket delicatessens, unmitigated requests for service predominate in both the Mexican and US settings. The lack of internal modification in the US and Mexican requests for service represents a sociocultural expectation and does not show an instance of inappropriate (or impolite) behavior. Although other studies of service encounters have shown that requests for service are generally unmitigated, as in Ecuador and Madrid (Placencia Reference Placencia2005, Reference Placencia, Schneider and Barron2008), in sociocultural settings such as small shops in France, requests are largely modified by the conditional form (Kerbrat-Oreccioni Reference Kerbrat-Orecchioni2006). However, one must also acknowledge the pragmatic and discursive function of high terminals in assertions and imperatives to signal compliance and secure the uptake on the part of the server. Interpretation of the high terminal as an interactional resource is a necessary condition to reinforce the interpersonal relations between the customer and server.
Gender analysis of the request for service This section presents the analysis of pragmatic variation by gender with regard to the preferred request for service (request variant) and internal modification. For comparability purposes, the same number of male customer requests was compared with female requests in each group. That is, 552 transactions for the English data (276 for each gender) and 450 for the Mexican corpus (225 for each gender). It should be noted that there were more male servers in the US delicatessens than in the Mexican ones, where female servers predominated. In both countries, interactions with a female server were more frequent than interactions with a male server.
Table 3.2 shows pragmatic variation with respect to both server and customer gender in the use of the three most frequent types of requests for service in US and Mexican delicatessens (see Figure 3.2 for distribution of all seven request variants).
Table 3.2 Distribution of requests for service by gender in US and Mexican supermarket delicatessens
| US group (552 male/female customer interactions with male/female servers [276 for each gender]) | |||
|---|---|---|---|
| Gender (customer → server) | Conventional indirectness | Want statement | Elliptical |
| M → M (138 interactions) | 49.3% (68/138) | 17.4% (24/138) | 12.3% (17/138) |
| M → F (138 interactions) | 43.5% (60/138) | 18.8% (26/138) | 13.8% (19/138) |
| F → F (77 interactions) | 44.2% (34/77) | 24.7% (19/77) | 16.9% (13/77) |
| F → M (199 interactions) | 51.8% (103/199) | 16.6% (33/199) | 10.1% (20/199) |
| Mexican group (450 male/female customer interactions with male/female servers [225 for each gender]) | |||
|---|---|---|---|
| Gender (customer → server) | Elliptical | Imperative | Assertion |
| M → M (28 interactions) | 53.6% (15/28) | 32.1% (9/28) | 7.1% (2/28) |
| M → F (197 interactions) | 50.8% (100/197) | 27.9% (55/197) | 15.2% (30/197) |
| F → F (208 interactions) | 33.7% (70/208) | 21.2% (44/208) | 33.2% (69/208) |
| F → M (17 interactions) | 35.3% (6/17) | 29.4% (5/17) | 29.4% (5/17) |
With respect to the US group, conventional indirectness is the strategy most frequently used; the gender of the interlocutors has little effect on the type of strategy employed. The gender of the interlocutor had little effect on the type of strategy employed. Similarly, the presence of the interlocutor had little effect on the selection of the two less frequently employed strategies (want statement and elliptical). In contrast, the Mexican data displayed more variation in preference for the three most frequent strategies. For example, elliptical requests predominated among male customers interacting with male (53.6%) or female (50.8%) servers. Elliptical requests and assertions were mainly used in female–female interactions, and although imperatives were the second most frequent strategy used by Mexican male customers addressing a male (32.1%) or a female (27.9%) server, female customers also selected this form when addressing a female (21.2%) or a male (29.4%) server.
Overall, for the US group a Chi-square test for independence revealed that differences in the preference for request variants (see Table 3.2: conventional indirectness, want, elliptical) are significant with a male customer interacting with a male or female server: M [Customer] → M [Server], X2 (1, N = 138) = 57.14, p < .0001, and in M → F interactions, X2 (1, N = 138) = 36.83, p < .0001. Significant differences were also obtained when a female customer interacted with a female or male server: F [Customer] → F [Server], X2 (1, N = 77): 14.89, p < 0.001, and in F → M interactions, X2 (1, N = 199) = 103.7, p < .0001. In contrast, for the Mexican group a Chi-square test for independence showed that differences in the preference for different request variants (see Table 3.2: elliptical, imperative, assertion) are significant with a male customer interacting with a male or female server: M [Customer] → M [Server], X2 (1, N = 28) = 14.12, p < .001) and in M→ F interactions, X2 (1, N = 197) = 59.39, p < .0001). Significant differences were also obtained when a Mexican female customer interacted with a female or male server: F [Customer] → F [Server], X2 (1, N = 208): 10.06, p < .001. These findings indicate a greater probability that the use and non-use of each of the request variants is not gender specific, since conventional indirectness predominated in the US group and the elliptical form in the Mexican group.
Given the relatively high frequency of assertions and imperatives used to issue a request by female customers with female servers, the stereotype that women are more polite or indirect (Holmes Reference Holmes1995; noted in Mills Reference Mills2003) does not hold in the Mexican context. In this context imperative or assertive requests by female or male customers represent a sociocultural expectation for the members of this community of practice. However, because of the low number of interactions with a male server (28 male–male interactions and 17 female–male interactions), the results of the current analysis (with regard to the preference of strategy used when issuing a request for service with a male server) are tentative (in Chapter 5 I discuss in more detail the issue of gender in service encounters at an open-air market). Finally, as far as internal modification is concerned, no major gender differences were found. Although infrequently used overall, the politeness marker, ‘please’ (por favor), was almost equally used by both Mexican and US male and female customers. A slightly higher frequency of this form was employed by US women (17.4%; 48/276) than US men (12.3%; 34/276). The other internal modifiers (e.g. conditional, diminutive) were almost absent from results in both sociocultural contexts.
3.4.3 Request–response sequence
This section examines the structure of the request–response sequence during the opening of sales transactions in US and Mexican delicatessens (interactional level [see Chapter 1, Section 1.4.6]). This sequence comprises various joint actions that are used to negotiate the interaction, such as optional requests for availability (pre-sequences), the request for service-response, clarification requests, and the delivery-acceptance of the product (sales purchase). As mentioned above, the placement of the request for service varied in each culture: in the US encounters it was placed in the second turn (71% or 494/700]), while in the Mexican interactions it occurred less frequently in second position (41% or 289/700). The request for service in the Mexican interactions was also delayed until the fifth or later turns (18% or 126/700), while delayed responses were less frequent in the US data (8% or 53/700). Mexican customers were more likely to delay the request by means of pre-sequences (23% or 160/700), while pre-sequences in the US data were infrequent (11% or 75/700). The request–response sequence can be preceded by preliminary information about the product (request for availability), in which case the request is delayed for one or more turns.
Example (11) shows US transactions with a pre-sequence (→) delaying the request:
[11] United States: Pre-sequences prefacing the request for service
a. Pre-sequence (male interlocutors)
01 S: Can I help you? 02 C: → yeah, you have any Tavern Ham ↑ 03 S Tavern Ham ↑ Yes, we do. 04 C: I'll take half a pound ↓ b. Pre-sequence (male interlocutors)
01 S: Can I help you? 02 C: → um, can I taste the low sodium ham? 03 S: yeah 04 ((clerk hands patron the slice)) 05 C: thank you 06 S: no problem 07 C: alright. Let me have a pound.
Pre-sequences in the US data were infrequent and were realized by means of preliminary information about the product (11a, line 02 [brand]). Pre-requests were also realized through requests to taste the product that were initiated by the customer (11b, line 02). The request for service was delivered after one or two pre-sequences, followed by the request (11a, line 04; 11b, line 07). In contrast, in the Mexican data the request for service often included various preliminaries requesting information about the product (brand, price, or offers by the server). Example (12), taken from the Mexican data, features a long series of pre-sequences that were negotiated prior to the request for service (line 22):
[12] Mexican pre-sequence: negotiation of the request (female interlocutors)
01 S: Buenas [tardes Good [afternoon 02 C: [¿tienes jamón de pavo de pierna? [do youT have turkey leg ham? 03 S: ¿de cuál? what kind? 04 C: de pavo turkey 05 S: de pierna de pavo turkey leg 06 C: de pierna – de pavo–de pechuga ((laughs)) turkey leg – turkey – breast ((laughs)) 07 S: ah : pechuga quie:re ((laughs)) ah: youV want turkey breast ((laughs)) 08 C: perdón ↑ pardon me ↑ 09 S: tengo la deli, si gusta I have the deli brand, if you'dV like 10 tengo la [San Rafael I have the [San Rafael 11 C: [¿cuál es la de deli? ¿puedo verla? [which is the deli one? Can I see it? 12 S: la redondita the roundD one 13 C: ¿a cómo está? – ¿esa la deli? how much is it? that one, the deli brand? 14 S: le sale a $75 ((pesos)) el kilo it costs (youV) 75 ((pesos)) a kilo 15 C: y ¿qué otro me decías? and what other one were youT telling me about? 16 S: San Rafael $133 – la clásica, 103 la:: turkey line, San Rafael 133 ((pesos)) – the classic one, 103 ((pesos)) the:: Turkey brand, 17 la:: pechuga Balance – [$99= the Balance (turkey) breast – [$99= 18 C: [¿puedo probar esta? [can I try this one? 19 S: y la de Swan está saliendo a $114 ((pesos)) and the Swan, Swan (turkey) breast costs $114 ((pesos)) 20 C: ¿la de qué? which one? 21 S: Swan pechuga de swan $114 Swan (turkey) breast $114 ((pesos)) 22 C: → dame 300 gramos ↓ (you) T give me 300 grams ↓
The pre-sequence in (12) is realized by a series of preliminaries before the main request is issued. After the server's non-reciprocated greeting, the customer inquires about a particular type of ham (lines 02–12). In the next pre-sequence, the customer asks about the price (lines 13–21), followed by the request for service, which is delayed until line 22. Pre-sequences are used to accomplish coordinated joint actions prior to the request, such as requests for information (or queries about the product), clarification requests, initiations of repair (line 08), and requests for action, such as asking the server to offer a sample (line 18). After the preliminaries have been negotiated by both parties, the request is finally presented in line 22.
Negotiation of the request is crucial during the pre-sequence phase, as both parties arrive at a mutual resolution before the request is delivered. Delaying the request by means of preliminaries allows the interlocutors to establish alignment and gain common ground (Clark Reference Clark1996) prior to the realization of the request. Note that the customer and the server co-construct the pre-sequence until an agreement is accomplished. In the Mexican pre-sequences, customers often asked the server to provide a sample of the product (more frequently than in the US data) and made several inquiries about the price and current specials (e.g. ¿qué oferta tienes? ‘what's on special?’ ¿es la más económica? ‘is it the cheapest?’).
The format and discursive structure of the request for service varied with regard to whether the customer's initiated request was explicit or general. Example (13) shows the format and structure of frequent US request–response sequences: example (13a) shows an explicit request followed by a short response sequence, while (13b) illustrates a generic request with subsequent co-construction to accomplish a successful request (the request is marked with →):
[13] US response sequences
a. Explicit request (female server; male customer)
01 S: Can I help you? 02 C: → yeah, can I have, um, a pound of the Boar's Head Deluxe 03 ham sliced just above shaved ↑ 04 S: OK, how much of it? 05 C: a pound ↑ 06 S: a pound ↓ 07 C: Deluxe Ham ↑ 08 S: ((nodding, yes)) b. Generic request (female server; male customer)
01 C: Can I help you? 02 S: → yeah can I get some of uh-tavern ham-a pound ↑ 03 C: how much? 04 S: can I get a pound please? 05 C: a pound ↑ how do you want it sliced? 06 C: uh-thin 07C : thin ↑ 08 S: is that OK or do you want it [thinner or…? 09 C: [yeah that'll work 10 S: you said a pound? 11 C: yes ma'am 12 S: anything else for you? 13 C: that was six ninety-nine, right? 14 S: yeah 15 C: there you go– anything else? 16 ((delivers product)) 17 C: that'll be it 18 S: alright 19 C: thank you.
The explicit request in (13a) includes the four required components of a complete request: the amount, the product type, the brand, and the thickness of the slice (lines 02–03) (e.g. ‘can I have, um, a pound of the Boar's Head Deluxe ham sliced just above shaved ↑’). Notice that the request ends with a final rising intonation (↑) to secure the interlocutor's uptake, as evidenced by the server's clarification request (line 04), which completes the sequence and confirms understanding of the customer's request (lines 05–08). If the request is complete, the remainder of the transaction is brief and is accomplished through the coordinated actions of both participants, initiated by the server (e.g. clarification requests [line 04]). On the other hand, (13b) shows a generic request (line 02), followed by the server's queries for clarification requests (lines 03–04, 10–11), or requests for additional information that are accomplished through various turns (lines 05–09). Notice the overlapping talk in lines 08–09, which features collaborative talk during the negotiation of the transaction. The pre-closing phase begins in line 12 to 17, followed by the terminal exchange (lines 18–19).
Although the Mexican encounters followed the US sequential structure, the negotiation of the request included additional social actions on the part of both the customer and the server. The server often issued an offer to try a different brand, as shown in example (14, line 05 [→]) (the requests for service are marked in bold):
[14] Mexican negotiation of the request–response sequence (females)
01 S: Hola, [buenas tardes Hello, [good afternoon 02 C: [buenas [good afternoon 03 eh me da – medio kilo de jamón eh (you)V give me – half a kilo of ham 04 premium turkey lite ↑ – turkey Swan premium turkey lite ↑ – Swan turkey 05 S: → si gusta, tengo el turkey lite de Fud, es Virginia if youV like, I have lite turkey – the Fud one, it's Virginia 06 C: ¿a cómo está? how much is it? 07 S: a 70 pesos, amiga, 71.90 70 pesos, girlfriend, 71.90 08 C: hay uno más barato ↑ is there a cheaper one ↑ 09 S: ¿por qué? ((laughs)) de pavo, el Iberomex de 56 pesos why? ((laughs)) turkey, the Iberomex costs 56 pesos 10 C: ay dame ese, es que lo comen como botana ah (you) T give me that one, the thing is they eat it as a snack 11 S: ((laughs)) ¿cuánto? ((laughs)) how much? 12 C: medio kilo – ay sí half a kilo – ah yes 13 S: ¿aparte le sirvo algo más, seño? besides that can I get youVanything else, ma'am? 14 C: no, está bien, gracias no, it's OK, thanks 15 S: aquí tiene there youV go 16 ((delivers product)) 17 C: gracias thanks 18 S: de nada ↓ you're welcome ↓
In the exchange in (14), the customer's initiated request (lines 03–04) is answered by the server's offer (line 05), followed by the customer's response (lines 06–07). In the next sequence, the customer indirectly rejects the offer and asks for a lower price, followed by the server's response (line 08–09). The beginning of the server's response represents a relational sequence including a joke and laughter (¿por qué? ‘why?’ [laughs], line 09). This sequence functions as the pre-sequence of the reformulated request, from an assertion (me da ‘youV give me’, line 03) to an imperative (dame ‘giveT me’, line 10), followed by the response (lines 11–12). It is also worth noting the switch in address form from V (me da ‘youV give me’, line 03) to T (dame ‘youT give me’, line 10) which is probably due to the successful transaction, and to emphasize the solidarity aspect of the interaction (in Chapter 8, Section 8.5, I address the issue of pragmatic variation in pronominal use). The pre-closing sequence occurs in lines 13–16, followed by the terminal exchange (lines 17–18). Overall, this interaction is co-constructed through sequential exchanges that are coordinated by joint actions. These actions include the opening (greeting exchange), a request–response sequence, an initiated offer–response sequence, a reformulated request–response, and the terminal exchange. The negotiation of social actions is the result of the customer's and the server's individual actions (e.g. speech acts) that are co-constructed to achieve common ground (Clark Reference Clark1996).
Finally, in addition to server-initiated offers during the opening of the request, customers asked for additional information about the product, followed by the server's response. Example (15) shows how both parties co-construct the request–response sequence across multiple turns (the customer's request for information is marked with an arrow [→] and the requests for service are bolded). The first part of the request (lines 01–08) is accomplished successfully. The second part of the request shows a change of alignment with regard to the reformulation of the initial request.
[15] Co-construction of the request for service (Mexican female interlocutors)
The second part of the request for service is realized in lines 09–23. This sequence begins with the customer's request for additional information about the product (line 09), followed by a repair sequence (10–15). The trouble source (line 10) is probably the result of the server not hearing or comprehending the word rebajado ‘on sale’ (line 09), which leads to further clarification of the request in lines 14–15 and 17. In line 18 the customer issues another confirmation request, followed by the server offering another option (line 19). These joint actions (lines 09–19) function as pre-sequences that lead to the reformulated request for service, which is delivered in line 20 (me das este ‘(you)T give me this one’) and completed in lines 21–23. The co-construction of this sequence is the result of coordinated joint actions that are negotiated to achieve a successful transaction.
After the request–response sequence was completed, the interaction ended by means of various interactional resources as terminal exchanges, as shown in the next section.
3.4.4 Closing the interaction
Closing the sales transaction showed more similarities than differences between the US and Mexican delicatessens. The closing sequence comprised various joint actions, such as pre-closings, thank-you expressions, offers to reopen the transaction, a closing sequence without the expression of gratitude, or relational talk during the terminal exchange. Terminal exchanges displayed various formats, ranging from the conventional pre-closing + closing sequence to terminals with insertions of relational talk to reopen the closing. The end of the transaction was marked by the server's question regarding additional service (e.g. ‘Can I get you anything else?’) and the customer's response in the form of a pre-closing device that ended the transaction, such as ‘no, that's it’, or the Mexican equivalent es todo ‘that's all’ (and other variants). If the customer's response is affirmative, then the sales transaction continues until the customer does not have anything else to order. Table 3.3 shows the distribution of closing devices used in the 1,400 transactions (US: 700; Mexican: 700).
Table 3.3 Distribution of closing devices in US and Mexican supermarket delicatessens
| Closing device | US delicatessens | Mexican delicatessens |
|---|---|---|
| No closing found | 21.9% (153/700) | 41.7% (292/700) |
| Customer closes with thanks | 13.7% (96/700) | 13.3 % (93/700) |
| Server closes with thanks | 3.7% (26/700) | 2.0 % (14/700) |
| Reciprocal thank-you sequence | 58.0% (406/700) | 35.1% (246/700) |
| Closing only – no ‘thank-you’ forms | 2.7% (19/700) | 7.9% (55/700) |
| Total | 100% (700) | 100% (700) |
As seen in Table 3.3, more than 50% of the interactions in each setting closed with either a reciprocal thank you, or a thank you on the part of the customer or server only (combining the middle three categories). Closing a transaction with a reciprocal thank-you sequence was the norm in both settings (Americans: 58% or 406/700; Mexicans: 35.1% or 246/700), and closing the interaction with the customer's thank-you expression also predominated in both data sets. Very few interactions closed without any form of ‘thank-you’ expression in the US (2.7% or 19/700) and Mexican data (7.9% or 55/700). It should be noted that the end of the transaction could not be identified in some portion of the Mexican (41.7% or 292/700) and US encounters (21.9 or 153/700) due to background noise or presence of simultaneous transactions. In both contexts it was more frequently the customer who ended the transaction.
In both sociocultural settings a transaction was often closed with some pre-closing and closing sequences, and the expressions used varied in both form and sequential placement. Examples of terminal exchanges are shown in (16 a–d) and (17) for English and Spanish, respectively.
[16] Closing sequences in US delicatessens
a. Male interlocutors
S: Here you go, sir ((delivers product)) C: thank you ((receives product)) S: is that it ↑ C: → yep. b. Male interlocutors
S: Anything else I can get for you? C: → no, thank you. c. Female server; male customer
01 S: Anything else ↑ 02 C: → that's it 03 thank you ↓ 04 S: thank YOU ↑ 05 you have a nice day 06 C: you have a good Easter ↓ d. Male interlocutors
01 S: OK, anything else ↑ 02 C: → that's all ↓ 03 thank you very much 04 S: you're welcome 05 sorry about that wait 06 C: that's OK ↓ 07 S: there you go 08 ((delivers product)) 09 you have a great day 10 C: you too, thanks ↓ 11 ((accepts product))
Closing devices varied with respect to form and sequential placement in the final phase of the interaction. The closing sequence (with no expression of gratitude) was realized by means of one or two adjacency pairs. This sequence may end abruptly without any form of gratitude in one adjacency pair, as in (16a), or with a thank-you form only, without the server's acknowledgment (16b). However, most closings in the English data ended with two or three adjacency pairs across various turns, including reciprocal ‘thank-you’ and ‘well-wishing’ exchanges. In (16c), after the pre-closing device, ‘that's it’ (line 02), the terminal exchange includes a ‘thank-you’ exchange (lines 03–04), and a ‘well-wishing’ exchange (lines 05–06). Notice that both thank-you forms differ prosodically: while the customer's ‘thank you’ (line 03) is realized with a downward final intonation (↓), the server's response is stressed with a high terminal (↑), as well as a nuclear stress on YOU. And in (16d), in addition to the pre-closing device (line 02) and the ‘thank-you’ exchange (lines 03–04), there is a side sequence that repairs the long wait (line 05–06). This sequence (non-transactional talk) is embedded within transactional talk and strengthens the links of solidarity between the server and the customer (see Chapter 7 for the analysis of non-transactional talk). Notice that the side sequence represents a realignment of the participants’ roles; that is, a change of footing from business to relational talk. The interaction ends with the server handing the product over to the customer (lines 07–08), a good-wish exchange, and the customer's acceptance of the product (lines 09–11).
Closing sequences in the Mexican data were generally similar to those from the US setting: they were realized through fixed adjacency pair sequences, as shown in (17 a–d):
[17] Closing sequences in Mexican delicatessens
a. Female interlocutors
S: Algo más aparte ↑ Anything else ↑ ((delivers product)) C: → no, es todo ↓ no, that's all ↓ b. Female interlocutors
01 S: Algo más, nena ↑ Anything else, babe ↑ 02 ((delivers product)) 03 C: → no, es todo ↓ no, that's all ↓ 04 ((receives product)) 05 gracias thanks 06 S: gracias. thanks. c. Male server; female customer
01 S: Algo más doña ↑ Anything else ma'am ↑ 02 ((delivers product)) 03 C: es todo ↓ that's all ↓ 04 gracias ((receives product)) thanks 05 S: [buenas tardes [good afternoon 06 C: [gracias [thank you 07 S: a usted. thank YOU. d. Female interlocutors
01 S: Algo más ↑ Anything else ↑ 02 C: es todo ↓ that's all ↓ 03 S: → quesito manchego no se le ofrece ↑ wouldn't youV like some manchego cheeseD ↑ 04 La Villita – ¿no le gustaría probar la Villita? La Villita – wouldn't youV like to taste la Villita? 05 cuarto, cuesta veinticinco pesos a quarter (kilo) costs twenty-five pesos 06 C: deme cuarto (youV) give me a quarter (kilo) 07 S: → ¿se le ofrece algo más? would youV like anything else? 08 ((delivers product)) 09 C: eso es todo ↓ that's all ↓
The expression es todo ‘that's all’ was used with two functions in Mexican closings: as a pre-closing device followed by the terminal exchange, and as a closing device to end the transaction. In (17a) the terminal sequence ends with es todo (‘that's all’), while in (17b) this closing expression functions as a pre-closing device (line 03), followed by the terminal exchange (lines 05–06). In (17c) we can see that the pre-closing device is followed by a four-turn thank-you exchange (lines 04–07). In this example, a usted ‘thank YOU’ (line 07) represents the appropriate response to the customer's ‘thank you’ (line 06), similar to its English equivalent, ‘thank you,’ with a high terminal (↑). Finally, the example in (17d) features an instance of opening-up closings (Hartford and Bardovi-Harlig Reference 263Hartford, Bardovi-Harlig, Bouton and Kachru1992; Schegloff and Sacks Reference Schegloff and Sacks1973). In their analysis of closing sequences in advising-student interactions, Hartford and Bardovi-Harlig (Reference 263Hartford, Bardovi-Harlig, Bouton and Kachru1992) showed the culmination of the advising session by the initiation of new conversation (or post-session talk), namely, “new conversations, with new topics not directly related to the advising session” (Hartford and Bardovi-Harlig Reference 263Hartford, Bardovi-Harlig, Bouton and Kachru1992: 115). In (17d), after the customer's pre-closing device (line 02), the server issues an offer (lines 03–05), which the customer accepts (line 06). The closing move in this transaction is finally accomplished in lines 07–09 by means of es todo ‘that's all.’
Relational talk represents a discourse strategy to open up the closing, such as episodes of small talk or brief relational sequences, as shown in example (18):
[18] Opening up the closing (US data: female server; male customer)
In this exchange, the closing is accomplished in lines 01–05 with the discourse marker, ‘OK,’ which signals completion. Then the server realigns the closing with a relational sequence (line 06) to express good wishes (lines 07–09). Instead of bringing this sequence to an end, the server continues with small talk (lines 10–12), followed by the closing (lines 13–14). In this example, relational sequences (lines 06–09) are used as a bridge in the discourse to trigger small talk (lines 10–12). Small talk (talking about the holidays, weather, school, children) was used to open up the closing in order to reinforce the links of affiliation and secure successful transactions in the future. (In Chapter 7 I explain in more detail the pragmatic and discursive functions of relational talk.)
Finally, male and female customers did not show any differences in the preference for the devices they used to open or close the transaction (see Tables 3.1 and 3.3). In both the US and the Mexican data, the great majority of encounters opened with no greeting (by either customer or server), whereas most of the transactions closed with a thank-you exchange. While the presence of greetings or thank-you exchanges can be perceived as an instance of polite behavior, the negotiation of the opening and closing sequence also represents an instance of politic (or appropriate) behavior (Locher and Watts Reference 267Locher and Watts2005) in the context of service encounters. Following previous work on face and facework (Bargiela-Chiappini and Haugh Reference Bargiela-Chiappini and Haugh2009), the negotiation of face in service encounters is discursively co-constructed by the server and the customer, as evidenced by the various turns used to accomplish both the opening and the closing sequences.
3.4.5 Organizational level
In this section I provide an analysis of one complete interaction in the U.S. setting at the organizational level (see Chapter 1, Section 1.4.6). In the delicatessens the turn-taking pattern was predominantly A-B-A-B, and largely by means of strict adjacency pairs. Instances of overlap and backchanneling were more frequent in relational talk and in phatic exchanges (i.e. greetings and terminal exchanges). The interaction in (19) illustrates the turn-taking patterns during the negotiation of service (male customer [C]; female server [S]):
[19] Turn-taking structure in a US delicatessen (Indiana) (male customer [C]; female server [S])
The turn-taking pattern is A-B-A-B with brief delays, latching with previous turn (=), short turns, and overlaps. Turns are largely designed in one or several TCUs (Sacks et al. Reference Sacks, Schegloff and Jefferson1974; Schegloff Reference Schegloff2007) within the same turn, as in line 16 with two TCUs (‘oh’ and ‘it's the Private Selection it's just a honey ham’), and in lines 26–27 with two TCUs, namely, a repetition and a question. Another characteristic was the presence of brief acknowledgment responses, as in lines 17 (‘oh’), lines 21 and 30 (‘yeah’ and ‘yup’), and line 34 (‘OK’).
The organization of turn-taking in the Mexican and US data was accomplished through adjacency pairs featuring coordinated sequences of actions. The interaction in (19) begins with a summons–answer sequence to establish identification of the participants’ roles and to create common ground (lines 01–03) (Clark Reference Clark1996). The transaction begins with the server's solicitation for service (line 04) followed by the customer's request, which is interpreted (by pragmatic inference) on the part of the server as a request for service (05–07). The customer's response to the server's question is delayed for two seconds (lines 07–08). The co-construction of the request is realized by means of unsolicited advice from the server and the customer's response (lines 09–23). This sequence features the A-B-A-B pattern, immediate turns (marked by =), and a relational sequence (lines 20–23). In the next sequence the server asks questions about the amount and thickness of the product (lines 24–30), followed by a clarification–response sequence (lines 32–36). Responses to the server's information/clarification questions were frequently realized with a two- or three-turn structure. For example, the response to the server's information question (line 24) is completed in two turns (lines 25–26), and the next question (line 27) is completed in three turns (lines 28–30). Another discourse strategy was high and low terminals, which signal a Transition Relevance Place (TRP) (Sacks et al. Reference Sacks, Schegloff and Jefferson1974) and secure uptake (lines 25–26 and 29–30). Specifically, high terminals indicate a projection of more information to come, and this was achieved by means of pragmatic inference (Whichmann Reference Wichmann2004). The terminal sequence includes the pre-closing (lines 37–39) and the closing sequence in two overlapping turns (lines 40–41).
Similar turn-taking patterns were noted in the Mexican service encounters. What differed between the Mexican and US interactions was the type of sequences embedded in different places of the interaction. For example, in the Mexican data, advice–response sequences (with a fixed A-B-A-B turn-taking pattern) were embedded in the opening and closing sequences. Another difference was the offer–response sequence (offer to taste product), which was initiated by the American server (e.g. do you want to taste it?), whereas this move was frequently initiated by the customer in the Mexican context (e.g. can I taste it?). In both cultures, turns were short (comprising one or two TCUs), some customer responses were delayed, and there was evidence of immediate latching and overlapping turns, mainly in the opening and closing sequences.
3.5 Conclusion
This chapter looked at cross-cultural variation in US and Mexican service encounters (supermarket delicatessens) during the negotiation of service at various levels of pragmatic analysis (actional, interactional, and organizational). It examined variation in the discursive patterns that US and Mexican customers use to open and close a sales transaction, the different choices (request variants) that customers use to issue a request for service, and the co-construction of the request–response sequence. The data were analyzed with regard to the gender of both customer and server to show variation at the actional level (requests for service and internal modification). This chapter ended with an analysis of the turn-taking structure (organizational level) in one complete service encounter.
Since the present chapter focused on the “global” view of linguistic variation (across languages) (Fried Reference Fried, Fried, Östman and Verschueren2010), the next chapter examines intra-lingual pragmatic variation (the “local” view) in two US and two Mexican regions. It analyzes pragmatic variation from a sociolinguistic regional perspective in Mexican small shops and in US delicatessens.
