INTRODUCTION
The vital role played by encounters between different kinds of service personnel and their clients has attracted the attention of various groups of researchers. Among those are language analysts who are drawn by the regularities in the form of the interaction to see service encounters as constituting a genre, and to seek to describe its schematic structure. To begin, the researchers create a dataset for analysis by collecting instances of service encounters that they have intuited as being of “the same kind” (Ventola 1987:3). The analysis that follows consists of identifying the common linguistic features shared by all instances of service encounters. Hasan 1989, for example, focuses on the level of conversational moves and identifies the Generic Structure Potential (GSP) for service encounters, which comprises a set of elements that defines the “total range of textual structures available within a genre” (Hasan 1996:53). Ventola 1987 approaches the topic by considering similarities in lexico-grammatical choices made on the planes of conversational structure, lexical cohesion, reference chain, conjunction, and boundary marking in instances of service encounters. The linguistic description produced of the genre serves two further purposes: classificatory and predictive. To classify an event as belonging to the genre of service encounters, Hasan 1989 looks within the speech event for the obligatory elements in the service encounter GSP: Sale request, Sale compliance, Sale, Purchase, Purchase closure. The presence or absence of optional elements such as Greeting or Finis would classify the events as belonging to different subgenres. Similarly, Ventola 1987 argues that the similar lexico-grammatical realizations of the different planes of discourse (e.g., conversational structure, lexical cohesion, reference chain) would classify the events in her data as belonging to the genre of service encounters; the differences in the realizations would classify the events as having different registers.1
The classificatory purpose is seen in several passages in Ventola 1987. Commenting on Hasan's work, she writes:
The GSP allows a systematic and consistent description of texts in terms of the elements which have been included in the text … This means that one is able to specify the typeness of instances of texts by the permissible sequences and the inclusion/exclusion of optional elements. Secondly, the GSP allows for a classification of text instances into types on the basis of the nature of their obligatory elements (the texts where elements X, Y, and Z, appear are from a different genre class from the texts with elements A, B, and C). (1987: 45)
With regard to her own work, Ventola (1987:6) promises that her analysis “will present a comprehensive view of how these discourse systems collectively function in the texts and enable one to draw conclusions about the genre classification of the analyzed texts.” In summarizing her own analysis, Ventola (1987:227) again writes: “The discourse system realizations should project the fact that all three texts analyzed belong to one and the same genre, that of service encounters, but that at the same time each text represents texts belonging to different registers.”
In the foreword to Ventola's book, Robin Fawcett (Ventola 1987:xviii) welcomes the approach to genre analysis developed by researchers such as Hasan and Ventola as being “predictive” or “generative.”
Genre studies that focus narrowly on describing textual forms and typologies have given rise to some misapprehensions. In a review of Ventola 1987, Ellis questions the theoretical status of her notion of genre: “It remains unclear, however, whether genres are generative or actualized systems, since V addresses this point only parenthetically” (1989:859). He goes on to criticize her theoretical position in the book as being “circular”:
It is difficult to shake the feeling that V uses her knowledge of the service-encounter genre to predict certain patterns of cohesion and reference. Then, when she finds these patterns, she uses their existence as proof of the linguistic realization of generic structure. (Ellis 1989:861)
What Ellis finds problematic in Ventola's work, I suggest, results from a lack of social orientation in the narrow conception of genre in some studies of “typeness” in instances of service encounters. A purely linguistic or textual approach to genre is prone to the danger of confusing the description of the “what” of genre with explanation of the “why.”
A broader conception of genre is found in the work of researchers such as Kress 1985, 1987, Swales 1990, Martin 1992, Bhatia 1993, Lemke 1994, and Hanks 1996. Here, the communicative purpose of the social event is taken to be the defining property of a genre and a privileged criterion for member texts (Swales 1990), which typically exhibit patterns of similarity in terms of content, speaker orientation, and textual organization. This is so because social events, as Kress 1985 observes, are often more or less thoroughly structured in terms of the participants' goals, the conditions that give rise to them, the expectations and constraints on behaviors, and so on. From this perspective, a genre is a “conventionalized communicative event” (Bhatia 1993), or as Johnstone (2002:158) puts it, “a recurrent verbal form (or ‘text type’) associated with a recurrent purpose or activity.” A more elaborate definition of genre that highlights its dual nature as both a linguistic entity and a cultural artifact is found in Swales 1990. The definition is later adapted by Bhatia, who claims that a genre is
a recognizable communicative event characterized by a set of communicative purpose(s) identified and mutually understood by the members of the professional or academic community in which it regularly occurs. Most often it is highly structured and conventionalized with constraints on allowable contributions in terms of their intent, positioning, form and functional value. These constraints, however, are often exploited by the expert members of the discourse community to achieve private intentions within the framework of socially recognized purpose(s). (Bhatia 1993:13)
Following this definition, a proper analysis of genre should involve asking the following three questions:
- What is a distinct grouping of texts recognized by a community as constituting instances of the same communicative event?
- What are the linguistic forms and structures characteristic of the group of texts?
- What do these textual characteristics tell us about the socioculturally recognized purpose(s) that are being fulfilled by the genre and the conditions of the situation within which the texts occur?
The present study aims to provide a socially oriented characterization of the spoken genre found in retail encounters in traditional Chinese local markets. The culturally familiar genre is sometimes referred to metalinguistically in the local dialect3
The local dialect spoken in Guangdong Province, where the data were collected, is Cantonese. Data used in this article were transcribed using the Cantonese Romanization Scheme established by the Linguistic Society of Hong Kong. Tones are indicated with arabic numerals.
TRADITIONAL LOCAL MARKETS AND SMALL SHOPPING MALLS
To outsiders, the local market can be a daunting place. It appears chaotic, with some vendors who rent designated spots from local market authorities, and others who have more informal arrangements, simply displaying their wares on the ground. A broad center aisle divides the market into two halves: On one side is the wet market, with stalls that sell meat, vegetables, fruit, and so on; on the other side is the dry market, with stalls that sell inexpensive clothes, shoes, household products, and various gadgets. The stalls are rarely partitioned and are identified by a casually displayed license with the vendor's number on it. The interactions are brief and functional. Shoppers rarely spend more than a couple of minutes at a stall and move swiftly from one to another. Except between vendors and longtime customers, interaction is usually minimal. During the bustling periods from 7:30 to 10:30 a.m. and 4:00 to 6:30 p.m., the market is a noisy, dirty, and rather chaotic place, but nevertheless robust, dynamic, and efficient. The market brings together a mix of people with a wide range of regional backgrounds. Some merchants and shoppers are from the local area, but there are also many immigrants from other provinces who can be heard speaking Cantonese, the local dialect, with heavy accents from their home towns. Recently, a lot of the small businesses in the dry market have moved into small, downscale shopping malls for the comfort of air conditioning and the convenience of adjacent facilities such as restrooms and eateries.
Over a nine-month period in 1998, I conducted ethnographic observations of a local market in southern China and a small shopping center in the neighborhood. I was able to observe the goings-on in the market and the shopping mall rather inconspicuously because these are naturally busy places during certain hours of the day. The data for this article include 38 audio recordings of naturally occurring interactions between salespersons and customers recorded over several weeks. Friends and family members who are local residents (except for two females who were visitors) agreed to help with collecting the data for the project by carrying small tape recorders on their daily trips to buy food and other necessities, as well as weekend visits to shops that sell leisure items. Occasionally I accompanied them (with my companion being the primary shopper). Other times I observed them at a distance. The customers in the data include both males and females, and teenagers as well as middle-aged persons, all of whom are native speakers of Cantonese. The recordings were played back to the shoppers, who confirmed that they represented typical instances of the everyday event of buying and selling in local markets and shopping malls.
SUBSTANTIVE ACTS: CONTENTS OF MARKET RETAIL ENCOUNTERS
Previous studies of service encounters in different cultural contexts (e.g., Mitchell 1957, Hasan 1989, French 2001) have shown that the array of activities that salespersons and customers may engage in fall into three generally groups. The GSP for service encounters proposed in Hasan 1989, for example, contains the following obligatory and optional elements: Greeting, Sale Initiation, Sale Enquiry, Sale Request, Sale Compliance, Sale, Purchase, Purchase Closure, Finis. Using the different kinds of activities that previous studies have identified as the basis, I posit three potentially operative frames in service encounters. The first is a valet frame, in which an attender waits on the attended, ascertaining his or her wants and needs and striving to satisfy them. The second is a consultation frame, in which the professional provides expert opinions and advice to the client, with or without prompting by the latter. The last is a transaction frame, in which the seller and buyer seek to satisfy their competing economic interests and maximize their personal gain. The activities involved in the valet and consultation frames make up what is generally thought of as “service,” which is ancillary to and supportive of the activity of negotiating an exchange in the transaction frame. The three frames, thus, are often intertwined in service encounters.
In this section I focus on substantive acts, those actions whose performance pertains to the task(s) to be accomplished in the encounter. I will show that a wide range of verbal and nonverbal forms that commonly occur within market retail encounters are consistently and predictably produced and interpreted within a transaction frame as having one of a small set of illocutionary forces. In the first subsection, I examine talk that revolves around the topic of price, which is by far the most common topic in market interactions. In the second subsection, I turn to talk related to the product in two different environments: pre- and post-initial offer. In the third subsection, I summarize the findings on substantive acts.
Price talk
Interaction in market negotiations often revolves around the topic of price: asking about the price of merchandise, naming the price, complaining about and justifying the price, and so on. This section will show how different kinds of price talk (including verbal and nonverbal actions) are produced and interpreted as acts of soliciting, making, or rejecting an offer, which make up the minimal set of bargaining moves belonging to a transaction frame in which buyer and seller negotiate price and quantity.
Answer to a question
Example (1) is a typical opening of interactions taped at a local market. A customer walks up to a hawker selling fruit. She points at some oranges:
(1) (E1/1–4)4
Lines in the excerpt that are being considered in the analysis are marked with arrows.
Lines 1 and 2, one may argue, form a typical adjacency pair. The customer's utterance in line 1 satisfies all the conditions for a question to elicit information from the salesperson. It has all the linguistic markings of a question: final rising intonation, the sentence-final question particle aa1, and the wh- word gei2 cin4 ‘how much’. Questions function to indicate that the speaker lacks knowledge of a particular state of affairs and to elicit that information from the hearer. The salesperson's response in the next turn (line 2), which contains the sought-for information – loeng5 man1 go3 ‘two dollars each’ – is interpretable as an answer, which would support the interpretation of the customer's utterance in line 1 as a question. Such an interpretation of lines 1 and 2 would make the pair of utterances acts that belong to a valet frame: The customer seeks service, in this case price information, and the salesperson provides it.
Yet this is not the end of the sequence. The interaction continues.
(2) (E1/1–4)
Upon receiving the price quote in line 2, the customer initiates a purchase. Gam2 ‘then’ in line 3 marks the move of initiating a purchase as resulting from the salesperson's price quote in line 2. The salesperson's utterance ‘two dollars each’ is hence not only an answer to the customer's question about price, but is also an offer of the merchandise at the named price. Offers, in contrast to statements (or answers), are classified as commissives (Searle 1975), which oblige the speaker to follow a future course of action and effect a world-to-words fit. The fact that the salesperson's utterance in line 2 is followed by an initiation of an exchange suggests that what may initially appear as an answer to an inquiry about price is not a statement but essentially an offer, which opens up the options for the customer either to accept it and hence make a purchase, or to reject it in the next turn. An utterance that contains the form x dollars, then, is treated in the event as an elliptical version (Austin 1962) of the expanded utterance I offer X dollars for the merchandise. The interpretation of price quotes as offers is further attested by the fact that no salesperson or customer in my data refused to make a sale or purchase at the price they named in an answer to a price question. While it might not be surprising that naming a price counts as making an offer in the setting of a local market, it does speak to the strength of a transaction frame: Price quotes are not given as information, they are made as offers to be accepted or rejected.
Question on price
If an answer to an inquiry about price is produced and interpreted as an offer, the price question itself is a solicitation of an offer. That a price question does not function as an information-seeking device that belongs in a valet frame is supported by the observation (my own and that of some salespersons I interviewed) that customers would frequently inquire about price even when such information can be readily gathered from price tags or other signs.5
The use of price tags, however, is a rather rare practice in local markets, and is only found in a small number of shops.
Statement and counterstatement
Sometimes, in response to a price quote, a party simply names a modified price, creating what may appear to be a statement-counterstatement sequence. The following is an example taken from an interaction between two customers and a salesperson at a kitchenware stall:
(3) (E31/1–5)
I have argued that naming a price counts as making an offer in market negotiations. Following this argument, the price quote in line 3 is a counteroffer made by the customer in response to an earlier offer by the salesperson in line 2. Implied in line 3 is hence a rejection of the previous offer, as the issuance of an immediate counteroffer can be taken as tacit evidence of the answer that would make the counteroffer relevant. Lines 2 to 4 can be expanded to reveal the following underlying structure:
(3a)
The customer's utterance ‘fifteen dollars’ in line 3 does not form a pair with the salesperson's offer of ‘twenty dollars’ in line 2. Rather, it is a counteroffer predicated on her elided rejection of the salesperson's previous offer, ‘twenty dollars’. The three acts packed into the two utterances in lines 2 and 3 – offer, rejection, and counteroffer – are typical of a transaction frame in which participants negotiate to get the best deal.
Evaluation of price
We have just seen how a rejection may be implied or elided by virtue of the rejecting party's making a counteroffer. Inference of rejection, however, can also be made in relation to other kinds of price talk. In my data, this is most commonly done by means of negative evaluation of the price.6
While a negative evaluation of the price of a commodity may be conventionally taken as a rejection of the offer, I still consider the interpretation to be a matter of inference. It is perceivable that a party may still make a purchase after complaining about the price being too high.
(4) (E15/21–24)
In response to the customer's query on price, the salesperson in line 22 answers that the jeans cost ‘eighty-nine dollars’. In the next turn (line 23), the customer makes a negative evaluation of the price with an exaggerated exclamation waa1! and a sarcastic remark – that he ought to get into the business himself since it is so lucrative. Instead of making a counteroffer, as we have seen in example (3), the rejecting party (the customer) in this example extended the interaction by asking the question peng4 di1 dak1 m4 dak1 aa1? ‘How about cheaper?’ to explicitly solicit a modified price – a next offer – from the salesperson.
Sometimes a party simply issues a rejection without making a next offer or explicitly soliciting one. Example (5) was taped at a small shop that sells crystal ornaments.
(5) (E5/1–6)
The price – ‘eighty dollars’ – that the salesperson names in line 2 is evaluated as unreasonable by the customer, using exaggerated stress and rising intonation to signal disbelief (line 3), and explicit comment on the merchandise as being ‘too expensive’ (lines 4–5). The customer's rejection, expressed by means of a negative evaluation of the price, stimulates the salesperson to make a next offer.
Sometimes a negative evaluation is made of the person who makes the offer, not the offer itself. Example (6) continues from (4):
(6) (E15/24–26)
The customer's utterance in line (25) – jau5 mou5 gaau2 co3 aa1 ‘did you make a mistake?’ is an idiomatic expression regularly used to express a hearer's disbelief and outrage, as it directly challenges the addressee's credibility. The customer's negative evaluation of the salesperson is followed by volunteered information about the price of the pair of jeans he is wearing. The fact that he does not leave the shop right after rejecting the offer, however face-threatening the rejection may be, is enough to stimulate the salesperson to make a modified offer of ‘forty-five dollars’ in line 26. Highly face-threatening negative evaluations of the other party or his or her offer, therefore, do not necessarily foreclose further interaction.
Sometimes negative evaluation of the price is done in very indirect ways. Example (7), taped at an accessories shop, shows how the salesperson's initial offer triggers a side sequence (Jefferson 1972) between the customers about a previous purchase.
(7) (E8/1–10)
In response to the salesperson's offer of ‘two for ten dollars’ in line 4, the customers launch into what is superficially a question-answer side sequence (Jefferson 1972) in lines 6 and 7, about a previous purchase that excludes the salesperson's participation. Unlike a typical side sequence that holds off a yet-to-come response, however, the question-answer pair is nevertheless considered as a relevant response to the ongoing (or matrix) sequence. The salesperson's justification of the price in line 8 is meaningful only when it is seen as countering a rejection by the customers.
So far we have considered only negative evaluations. Positive evaluations are frequently used by a party to reinstate an earlier offer that has been rejected. Example (8) shows a customer and a salesperson negotiating the price for some household items.
(8) (E27/1–6)
The salesperson makes an initial offer of ‘five dollars’ in line 2. The customer implicitly rejects this and makes a modified offer of ‘four dollars’ in line 3. The salesperson, then, has the options in line 4 of accepting or rejecting the customer's offer (implicitly or explicitly) with or without making a new offer. Her utterance in line 4, ‘It's already very cheap’, can be seen as doing two things: By not accepting the customer's offer she implicitly rejects it, and by positively evaluating an earlier offer she made (and depicting herself as being not greedy for profits) she reinstates it as a new offer. The customer accepts the reinstated offer of ‘five dollars’ in line 5 and moves to complete the purchase.
In examples (4) to (8), we see how negative and positive evaluation of price (and the party who makes the price quote) are regularly produced and interpreted as a small set of acts belonging in a transaction frame – rejections and reinstated offers. What I did not find in my data is a party agreeing with another's price evaluation,7
I have observed that salespersons in department stores in the United States sometimes concur with a customer's opinion that a commodity is too expensive.
Silence and departure
We have looked at how different kinds of utterances are produced and interpreted as acts of soliciting, making, and rejecting an offer. In this subsection, we will see that nonverbal actions such as silence and attempts to depart also can be taken as moves belonging in a transaction frame.
Prior to the beginning of excerpt (9), the salesperson and two customers have made a series of offers and counteroffers.
(9) (E9/35–40)
The salesperson rejects the customer's offer of ‘three dollars' in line (35) by making a next offer of ‘three fifty each’, which is the same as an earlier offer she made. The customers issue no verbal response in the next turn (line 37), but maintain their physical position. From the salesperson's modified offer in the next turn (lines 38–39), we can see that the customers’ 10-second silence is nevertheless interpreted as a relevant and meaningful response, namely a rejection of the offer, which stimulates a new offer from the salesperson.
I have so far described the occurrence of a non-solicited next offer as being “stimulated” by the rejection that occasions it. (See example 4 for an instance of a solicited next offer, and examples 5 and 6 for instances of a non-solicited next offer.) The description glosses over something that the analysis has taken for granted so far: that the two negotiating parties remain in each other's presence after an offer has been rejected. Of course, they do not have to. One thing that contributes to the possibility of extended price negotiation is the parties' tacit agreement to stay within the engagement, availing themselves of more talk. In the next example, however, I show how even an apparent breach of the consensus to stay engaged by attempting to depart can be produced and interpreted as a bargaining move that, paradoxically, furthers the negotiation. Example (10) shows an interaction between two customers and a wicker mat hawker:
(10) (E34/15–27)
The customer's utterance in line 21 – Gam2, syun3 laa3 ‘Well then, let it be’ – after their offer has been rejected in the previous turn acknowledges the rejection and completes the transaction, and their departure officially closes the encounter. The re-summons and answer that follow (lines 22–25), I would argue, are not typical. A re-summons after the closing would normally open up a new state of talk (Goffman 1981) with a new focus (Goffman 1961). If the previous focus is resumed because a speaker has more to say on the topic, some kind of repair would be in order (Schegloff & Sacks 1973). The answerer of a summons, as Schegloff (1972:379) observes, typically employs an assent term such as uh huh to signal availability, leaving the summoner to produce an utterance that contains real propositional content in the next turn. In example (10), the interaction after the summons is answered – the customers' act of initiating the first question in line 25 without skipping a beat – suggests that the departure in line 21 has an ambiguous meaning. It may be a resolute move to terminate the negotiation (if the salesperson had not re-summoned, or the customers not answered, the encounter would indeed terminate), or it may be a bargaining move that pressures an opposing party into proposing more acceptable terms for the exchange. Example (10a) shows the sequential organization of the interaction:
(10a) (E34/15–27)
If the attempt to depart (line 21) is the move of soliciting a new offer, the re-summons of a departing customer (lines 22 and 23)8
I consider the helper's utterance in line 23 as ancillary to the nonverbal act of tapping on the customers' shoulders in line 22. The two lines are considered together as one move of summoning.
In this section, we see how common nonverbal actions serve functions that are atypical: Absence of a response to an offer does not trigger a “re-run” (Goffman 1981), and the attempt to depart is not always a resolute act to terminate further engagement. The fact that silence and departure are regularly produced and interpreted as bargaining moves – rejecting and soliciting an offer – is further evidence for a transaction frame that dominates the encounter.
The argument I wish to make is that in Chinese local markets, people engage in price talk to reach a deal. This might be a rather obvious statement at first glance, but surely there are other retail contexts where salespersons and customers talk about price for the sake of exchanging information, or even just to have a conversation. The consistent use of price talk for the exclusive purpose of negotiating a deal demonstrates the dominance of a transaction frame for the event.
Product talk
While most talk in interactions between buyers and sellers in local markets revolves around the issue of price, sometimes talk shifts to the merchandise, often as an evaluation of it. In the following subsection, I show how product talk in the pre-initial offer environment is regularly produced and interpreted as a bargaining move. In the next subsections, I show how product talk that occurs in the post-initial environment is infrequent and rarely elaborated. The different treatment of product talk in the two environments lends further support to the argument that a transaction frame prevails in the encounter.
Evaluation of the merchandise
Example (11) shows an interaction between a hawker who sells sundries and a customer; it contains a customer's evaluation of the merchandise.
(11) (E28/1–10)
Responding to the customer's solicitation in line 4, the salesperson makes a next offer of ‘ten dollars each’ in line 5. The customer responds by commenting on the less than satisfactory quality of the bowls in line 6. Negative evaluation of the merchandise may signal a lack of interest in most retail contexts; in Chinese local markets, however, it is interpreted as soliciting a modified offer. Instead of giving up on the interaction, the salesperson pursues the negotiation by making a modified offer of ‘six dollars’ in the next turn (line 7). The marker of compliance – hou2 laa1 ‘all right’ – that prefaces her turn signals that the salesperson interprets the customer's negative evaluation of the merchandise as a solicitation, to which her modified offer is a response.
Example (12) shows a salesperson's positive evaluation of the merchandise as a rejection of a customer's offer.
(12) (E3/1–7)
By referring to a better offer from a different hawker in line 5, the customer implicitly rejects the salesperson's offer and, one can argue, implicitly makes a counteroffer of ‘ten dollars for two’. The salesperson responds to the implicit offer by pointing out that his pumelos are of superior quality and are juicier. The positive evaluation of his merchandise in line 5 implicitly rejects the customer's offer and reinstates his last offer (line 3) of ‘six dollars each’, which the customer accepts in the next turn (line 6).
Example (13) is taken from an interaction between two customers and a shop assistant at an accessories shop.
(13) (E12/10–19)
In lines 14 and 15, the salesperson positively evaluates the earrings – they ‘look nice’ (hou2 leng3) – that the customer has tried on and recommends another pair that she claims is very popular. The customers issue a negative evaluation of both pairs of earrings in the next turns (lines 16–17) in an unmitigated fashion, using a simple negation of the salesperson's utterance – ‘don't look nice’ (m4 hou2 tai2). Despite the face-threatening disagreement, the salesperson continues to pursue the negotiation, hinting at a modified offer in line 18: If the customers take both pairs of earrings she will give them a ‘lower price’ (peng4 di1). The customer's interest in pursuing the exchange is demonstrated in her follow-up question in line 19 – ‘how much cheaper?’ (peng4 gei2 dou1 aa1) – which serves as a request for clarification of the rather vague offer.
Examples (11), (12), and (13) show that although the merchandise may become a topic of talk, it is not recognized as a relevant topic in and of itself. If product talk functioned as a statement of opinion, we could expect to see more elaborate positive evaluation by the salesperson, a bigger sales pitch. We could also expect a customer's negative evaluation of the merchandise to signal a lack of interest in it, and the salesperson to ascertain the customer's desire and to make appropriate suggestions. We might even expect disagreeing parties to engage in protracted talk about the quality of the merchandise until some common ground or resolution was reached. Instead, the data show evaluation of merchandise as an act of rejecting and/or reinstating an offer. Positive and negative evaluations of the merchandise, then, constitute bargaining moves belonging in a transaction frame.
Non-response to pre-initial offer evaluation of merchandise
We have seen how negative evaluation of the merchandise by the customer is consistently produced and interpreted as the bargaining move of rejecting a current offer (with or without implicitly soliciting a next offer). It is notable, however, that not all negative evaluation is interpreted as a bargaining move. In my data, I found similar evaluations of the merchandise that fetch no response at all. In example (14), two customers who have just stopped at an accessories stall make a negative comment about the merchandise.
(14) (E10/1–6)
The salesperson opens the interaction with a question in line 2 – Soeng2 wan2 di1 mat1 je4 aa1? ‘What are you looking for?’ – to identify the desired product, to which the two customers offer no verbal or nonverbal response. In lines 4 and 5, the customers issue their first utterances, one echoing the other's negative evaluation of the merchandise they are looking at as ‘cheap-looking’ (hou2 ‘cheap’). (The customers are looking for something specific and so are coordinated in their attention.) Unlike in previous examples, the negative evaluation does not stimulate a modified next offer, since no initial offer has been made and the merchandise does not have price tags. Although the salesperson is clearly within earshot, the negative evaluation simply fetches no response from her. The customers move on to the next stall without so much as a pause, suggesting that it is unlikely that they themselves expect much by way of response.
The difference in the salesperson's response to negative evaluation of the merchandise seems to lie in the environment in which the negative evaluation is made. In examples (11) and (13), the customers' negative evaluations occur only after price talk has begun; they are interpreted as rejections of the current offer and solicitations for the next. The negative evaluation of the merchandise in example (14), in contrast, occurs before price talk has begun; it is produced and interpreted as no more than the customers' opinion about the product. There is no example in my data (and personal experience) of customers negatively evaluating a piece of merchandise whose price they then go on to inquire. There is also no example in my data, though this is often observed in other retail contexts, of salespersons asking follow-up questions to find out what the customer might be looking for in order to make a recommendation for purchase. The fact that salespersons rarely behave this way in Chinese local markets suggests that a valet frame in which salespersons attempt to ascertain and anticipate customers' wants and needs is suppressed. As statements of opinion, customers' negative comments on the merchandise that occur before the initial offer is made are not acted on. In fact, they happen very infrequently.9
The customers in examples (14), (15) and (16) are not local residents. They are, however, native speakers of Cantonese from Hong Kong who have had some experience shopping in the kind of local market where the data were collected.
Minimal response to pre-initial offer inquiry about merchandise
The significance of the pre-/post-initial offer divide is also demonstrated by another kind of product talk: inquiry about the merchandise. Questions about the product from the customer before the initial offer is made often receive minimal response. Example (15) was taped at a stall that sells crystals and semiprecious stones in a small shopping mall.
(15) (E7/1–9)
In response to the customer's question in line 2 about availability of other colors, the salesperson responds (line 3) by using deictics such as ha6 min6 ‘below’ and gaak3 le4 ‘next to’ and casually gesturing at the display case without establishing exactly where she is referring to. In response to the question on availability of other styles (line 4), the salesperson gives a brief and unelaborated answer in line 5 – hai6 aa3, saam1 fun2 ‘yeah, three styles’. Although the salesperson's responses seem to satisfy the conditional relevance of the customer's yes/no questions, the customer likely expects answers that go beyond what is already obvious; if she does not, the questions would be unnecessary. It is likely that the customer is looking for the salesperson to offer some sales services, perhaps to take the pendants out of the display case for her to see up close. The cues, however, pass unnoticed.
Example (16), taped at another stall on the same floor in the shopping mall, shows more clearly how questions that occur before the initial offer often elicit minimal information.
(16) (E6/1–13)
Although there is no way to determine from the elliptical question in line 2 – ni1 go3 hai6? ‘This is?’ – what kind of information the customer is seeking, it seems unlikely that she is bidding for information readily available on sight, such as the shape of the pendant. The salesperson's answer in line 3 nevertheless supplies just such obvious information. That the customer's call for an appreciable undertaking is not heeded is evident in line 6, where the question ‘what stone is this?’ is prefaced by the phrase ngo5 ji3 si1 waa6 ‘I mean’, a marker with a remedial function of clarifying the speaker's misinterpreted intention (Schiffrin 1987:297). The salesperson's reply in line 7 that ‘it's crystal’ (hai6 seoi2 zing1), again, hardly seems informative, given that the stall is located on a floor in a mall designated for stalls selling crystal and semiprecious stones. The customer expresses dissatisfaction with the answer and reissues the same question in line 8, placing stress on the category of required information, mat1 je4 ‘kind’. The salesperson's reply in line 9 – Hai6 zi2 seoi2 zing1 ‘It's purple crystal’ – is syntactically ambiguous in Chinese. It can be interpreted as [qualifier+noun], i.e. ‘purple color crystal’, in which case it is again rather uninformative, as the color of the crystal is readily discernible by the customers. The utterance may also be interpreted as a proper noun like ‘amethyst’. In this interpretation, the salesperson's answer would seem to have complied with the customers' request for information about the stone.
The subsequent interaction suggests, however, that the customers are looking for more elaborated talk. The side sequence in lines 11 and 12 between the customers – ngo5 dim2 zi1 hai6 mi6 zan1 gaa3? ‘How do I know if this is real crystal?’ – expresses doubt about the genuineness of the crystal and is undoubtedly meant to be heard by the salesperson. All the customers' questions up to this point can be seen as attempts at subtly stimulating the salesperson into providing expert knowledge on how to distinguish between genuine crystals and fakes. A direct, closed-end question such as Is this crystal real? would be self-defeating. After all the trouble of hinting indirectly at the sought-for information, however, the salesperson in line 13 provides barely any expert knowledge to support her claim that the stone is ‘really crystal’ (zan1 seoi2 zing1).
Examples (15) and (16) show instances of the customer trying unsuccessfully to elicit expert knowledge from the salesperson. In other retail contexts, such as a jewelry shop, such expert knowledge is often dispensed freely by salespersons or via printed materials such as pamphlets. The data show, however, that salespersons in local markets do not expect to demonstrate expertise in the merchandise they deal in. One may suspect that salespersons just do not have any more knowledge than customers, as the goods involved are everyday necessities and other low-end leisure items. Many have observed that Chinese consumers rarely rely on salespersons for information on goods, but tend to rely on recommendations from friends or on published information. Most customers simply do not expect salespersons to provide expert knowledge. Since little is demanded, little is offered.
I have argued here that product talk, like price talk, is consistently produced and interpreted as acts of rejecting and/or reinstating an offer – that is, as bargaining moves belonging in a transaction frame. Product talk that does not fit in a transaction frame is rarely attended to or elaborated. In fact, such talk happens very infrequently. Unlike interactions in other retail contexts, customers do not usually explain overtly what they might be looking for or ask for recommendations; salespersons, on the other hand, rarely try to elicit information about customers' wants and needs or offer professional opinions and knowledge. A valet and a consultation frame are therefore muted.
Task-focused goal and participant roles
The analyses of price talk and product talk above can be summarized as follows:
- Many kinds of verbal and nonverbal actions (i.e., questions and answer about price, evaluation of price and merchandise, non-utterances such as silence and attempts to disengage from interaction) are regularly produced and interpreted as acts of soliciting, making, rejecting an offer – the minimal set of moves that constitute basic bargaining.
- The initial offer is a significant divide. While post-initial offer talk is constituted of bargaining moves, pre-initial offer talk that cannot be interpreted as bargaining moves receives little elaboration and is rarely acted on. It in fact occurs infrequently.
The analysis of substantive acts shows that a transaction frame dominates interactions between salespersons and customers in Chinese local markets to the extent that valet and consultation frames are suppressed. The retail encounters are almost completely filled out by bargaining moves; the initial offer for a piece of merchandise serves as the catalyst for extended and focused interaction and sets in motion a chain of bargaining moves between participants who are focused on maximizing their individual gain in the transaction. Without the initial offer, bargaining cannot begin, and there is little for the participants to go on. The prevailing transaction frame, once activated, persists until the cessation of the speech activity (by virtue of an agreement being reached or termination of the encounter) and serves as the interpretive context for verbal and nonverbal actions. Any talk that falls outside the restrictive frame is relegated to the periphery.
Compared with their counterparts in other retail outlets (e.g., in big department stores or car dealerships) or other cultural contexts (e.g., in American flea markets or Middle Eastern open markets), salespersons in Chinese local markets seem to play a very restricted role. As Goffman (1961:86) explains, within a role, the activities involved fall into “different, somewhat independent parcels or bundles,” which are often differently apportioned to occupants of the same role in different contexts. While a salesperson in, say, a typical American department store can be expected to do a number of things (including providing information on the merchandise, offering expertise on specialized products, or advising customers on their choices), it seems clear that salespersons in typical Chinese markets define their role restrictively as that of a seller and their job as little more than responding to the customer's bid for a piece of merchandise. In sum, retail encounters in Chinese local markets are highly task-focused, with a tightly circumscribed goal of determining the terms of a transaction, and narrowly defined participant roles of buyer and seller in the strictest sense of the terms.
RITUAL ACTS: BOUNDARIES OF MARKET RETAIL ENCOUNTERS
In the previous section, I attempted to characterize retail encounters in Chinese local markets in terms of the substantive acts that occur within the communicative event. In this section, I turn to the boundaries of the event. One can argue that before shoppers and vendors are engaged with each other as customers and salespersons to negotiate a sale, they are already engaged with each other in a more general way; shoppers who come to the market walk along thoroughfares lined by vendors, eyeing the merchandise displayed by the latter, who eagerly anticipate business. Sometimes customers give no more than a passing glance at the stalls and little attention to the vendors' promotional cries. Sometimes they pause briefly to inspect the merchandise in full view of the vendor. Sometimes they may initiate negotiation with the vendor and even make a purchase. The broader speech situation (Hymes 1972) in which the speech event of a retail encounter (wherever its boundaries might be) is nested might be described as “co-mingling in a public place of commerce,” to adapt an expression from Goffman 1971. Within the speech situation, many activities that are conducive to a commercial exchange can be found. Which activities are socially recognized as constituting the speech event of the retail encounter can be determined by examining how the members of the community themselves draw the boundaries of the retail encounter.
In this section, I focus on ritual acts, those actions that, through their performance, provide bracketing for a period of mutual access and coordination – that is, a focused gathering (Goffman 1971). I will show that shoppers and vendors in Chinese local markets typically regard each other as outgroup persons whose only reason for face engagement is to transact business rather than to cultivate personal relations. I first analyze the general practice of avoidance rituals through aversion of gaze, and the timing of the initial eye-to-eye look. Next, I examine the general absence of engagement rituals and the rare occurrence of closing interchanges. I then consider the notion of a retail encounter as subjectively defined by the participants themselves. Finally, I summarize the findings on ritual acts.
Practice of avoidance rituals
Despite mingling in full view of each other, shoppers and vendors do not typically make eye contact with each other in the market. The use of avoidance rituals (Goffman 1971:62–63) is perhaps not unusual. As members of a collectivistic culture, Chinese shoppers and vendors regard each other as outgroup persons (Hofstede 1980) with whom the normative relationship is dissociative or even competitive (Triandis & Vassiliou 1972). What is more notable, however, is the common practice of shoppers studiously avoiding eye contact with the vendor upon entering a shop or stopping in front of a stall and even while inspecting the merchandise. Example (17) is an instance of such behavior.
(17) (E2/1–3)
The customers offer no overt response to the salesperson and avert their eyes as they examine the fruit. The salesperson, in contrast, remains watchful.
Of the many crucial functions that eye gaze serves in social interaction, one is described by Kendon (1967:24) as its use as “an expressive sign and regulatory signal by which [one] may influence the behavior of the other.” Kendon writes:
… when one perceives that another is looking at one, one perceives that the other intends something by one, or expects something of one. In a word, one perceives that one is being taken account of by another … To receive his gaze is to receive an indication that one is being taken account of. (1967: 59)
Goffman (1963:92) makes a similar observation, explaining that meeting of the eyes “ritually establish[es] an avowed openness to verbal statements and a rightfully heightened mutual relevance of acts.” The expressive and regulatory function of eye gaze is particularly important at the beginning of a focused gathering between shoppers and vendors. Here, I draw on Goffman's distinction of two kinds of recognition: cognitive and social. The former is “a private act that a concealed spy can engage in”; the latter is “a ceremonial gesture of contact with someone” (Goffman 1963:112). The two kinds of recognition are often taken to occur simultaneously in most situations we call “encounters.” Imagine that one person knocks on the door and steps into another's office. Such a move is duly marked as the beginning of a social encounter by the instantaneous meeting of the eyes (and often the exchange of verbal rituals) that opens up a two-way channel and establishes a basic orientation between two persons as co-interactants. As we see in example (17), however, social recognition of contact underway does not automatically follow when a shopper stops at a stall and even inspects the merchandise with the vendor looking on. By not allowing her gaze to be caught, the shopper subjectively and symbolically denies any encounter-in-progress.
The practice of gaze avoidance in local markets is observed to breach at specific moments. Example (18) continues from example (17) and shows how two customers make the first eye contact with the salesperson.
(18) (E2/1–4)
The breach of avoidance as the customer raises her gaze is timed to occur at the moment she issues the first utterance, in this case price talk that solicits an initial offer from the salesperson. The concurrence of the initial eye-to-eye look with the onset of bargaining talk, but not before, provides more evidence for the argument that retail encounters in Chinese local markets are dominated by a transaction frame: A customer and a salesperson are ratified as an interactional pair only when their mutual interest in pursuing a transaction and readiness for the undertaking of negotiating its terms are established.10
The salesperson and the customer are on an unequal footing at the outset of a negotiation episode, which explains their contrasting behaviors. By virtue of being open for business, the salesperson makes known his or her intention to sell the merchandise on display and readiness for negotiation. In contrast, the customer's complementary intentions to buy or to negotiate a purchase cannot be ascertained superficially.
Absence of engagement rituals
Not only are avoidance rituals practiced up to the point when both parties are ready to begin negotiating; engagement rituals that have regularly been observed to preface service encounters in other contexts (e.g., Tsuda 1984, Gavioli 1997) are also noticeably absent. By “engagement rituals” I refer to both what Goffman 1971 terms “supportive interchanges,” such as greetings and farewells used between acquainted persons, and “remedial interchanges,” used between unacquainted persons, which include requests for attention or apologies for an intrusion and thanks-for-the-audience. We have seen in previous examples (e.g., examples 1, 3, 5, 8, 11, 12, and 15) how interactions between customers and salespersons typically open without any engagement ritual. The same observation is made by Pan 2000 in regard to encounters between clerks at a Chinese state-owned stamp store and its customers. The absence of engagement rituals, as Pan argues, reflects the typically instrumental nature of encounters between outgroup persons in collectivistic cultures, where verbal rituals that are not germane to accomplishing the task-at-hand are perceived as unnecessary. Pan's explanation supports the argument that retail encounters in Chinese local markets are highly focused on the primary goal of negotiating a transaction, in relation to which the secondary task of exchanging service is marginalized.
There are, however, two rare examples of the use of engagement rituals, both closing rituals, in my data. Example (19) was taped at a shop that sells household sundries.
(19) (E35/21–26)
Having agreed over the price of a small cart and completed the transaction, the salesperson expresses gratitude – Do1 ze6 ‘Thanks’ – and a wish for future contact (line 25). The customer responds in line 26 with well-wishing – Hou2 saang1 ji3 ‘Good business’. Together, the closing rituals in lines 25 and 26 sum up the positive consequence of the encounter for the relationship: that the two parties hold each other in good regard.
A similar instance of the use of closing rituals is seen in example (20). After a lot of bargaining, the customers and the salesperson finally agree on a price of fifteen dollars for five bangles.
(20) (E9/43–47)
After the transaction is completed, the salesperson evaluates the deal as favorable for the customer and expresses her desire for future contact (line 46). The customer replies with an agreement token okay in line 47 before departing.
In both examples (19) and (20), the engagement rituals occur at the closing of an encounter in which a transaction has been successfully negotiated and completed. It should be noted that both encounters have begun, like many others, with an absence of engagement rituals at the opening (i.e., no greetings or apology for intrusion), which marks the encounters as being between outgroup persons. The exchange of interpersonal rituals typically used between acquainted persons (e.g., well-wishing and projection of future contact) at the closing, then, ceremoniously marks a perceived change in the social relationship between the participants. It seems that having successfully closed a deal, outgroup, unacquainted persons may now claim some form of acquaintance.
The general absence of engagement rituals, except in the restricted post-exchange environment where they may sometimes occur, is further evidence for the dominance of a transaction frame and subjugation of a valet frame in retail encounters in local markets. Customers and salespersons typically regard each other as outgroup persons, engagement with whom is motivated by the instrumental purpose of negotiating a transaction. An ingroup/acquaintance-like relationship may be claimed if and only if a transaction has been successfully conducted.
Clearly, there are advantages for outgroup persons to claim acquaintanceship, especially in business dealings, which are one of the most common reasons for outgroup engagement. Acquainted persons, Goffman (1963:119) points out, enjoy certain privileges from each other. As someone who is acquainted with the customer, the salesperson can, for example, demand that the customer give the salesperson his or her business. In local markets, this is seen in shoppers who would carefully avoid certain acquainted vendors seeing them buying from a different vendor. On the other hand, a customer can also demand from an acquainted salesperson discounted prices, better selection, and other kinds of benefits due to a regular customer. In fact, observations have been made of changing relationships and rituals used between customers and salespersons in privately owned stores and businesses that have flourished as a result of the economic reform that began in the late 1970s (see Pan 2000).
Recognition, rituals, and non-encounters
So far, I have discussed how shoppers avert their gaze before they are ready to negotiate, and how negotiation typically begins without any engagement rituals. There are, of course, many times in local markets when one can observe a shopper enter a shop or stop at a stall, look at the display for a while, and then leave, all without so much as a word or glance in the salesperson's direction. Example (21) is an instance of such a mundane occurrence.
(21) (E10/1–5)
The shoppers offer no response to the vendor's question and avert their gaze. After determining that there is nothing of interest to them at the stall, they simply depart. No eye contact is made with the vendor, and no talk is exchanged at all.
Instances when no talk happens often do not get included into a dataset at all, much less analyzed alongside other instances where talk does occur. Yet it is clear that the participants in example (21) did partake in some joint activity; a summons to negotiate was issued and answered,11
One can argue that by virtue of being open for business, a salesperson is issuing a summons to passing shoppers. A customer who stops at a stall or enters a shop can be said to be answering the summons.
Transaction-based relationship between participants
The analysis of ritual acts in retail encounters in Chinese local markets can be summarized as follows:
- Cognitive and social recognition of an encounter-in-progress are not coterminous.
- Social recognition of a retail encounter-in-progress is only given, and social ratification of the participants as an interactional pair of buyer and seller only made, when both parties are ready to negotiate the terms of a transaction.
- A relationship beyond that between outgroup persons is only possible when the individual goals of the participants as buyer and seller have been satisfied in a transaction.
The analysis of ritual acts shows that customers and salespersons in Chinese local markets relate to each other as outgroup persons whose primary reason for engagement is to satisfy their goals by negotiating a deal. As outgroup persons, the tacit norm of behavior between customers and salespersons in the market is that of avoidance. This explains to some extent the impressions that outside observers may have of Chinese salespersons and customers as being curt, rude, and even contentious. In sum, negotiation encounters in Chinese local markets are understood by members of the community as perfunctory, task-focused engagements that are not designed to encourage the cultivation of long-term, personal relationships between their participants.
Normative rules, Goffman (1971:61, n. 54) explains, inherently carry a kind of duality that creates “a meaningful set of non-adherences.” The breach of the avoidance rule between customers and salespersons – including behaviors such as making eye contact, initiating and responding to talk, sampling the merchandise, and many others – signals a change in the outgroup norm of dissociation. The more intrusive the behavior is, the more significant is the non-adherence, and the stronger is the projection of a transaction. This explains why novice shoppers are often advised not to initiate talk with salespersons and to lay off the merchandise if they have no real intention of making a purchase.12
I observed an instance in the local market of a fruit seller chastising a scrutinizing shopper for handling her fruit without eventually making a purchase.
These aggressive behaviors of salespersons are more often observed in low-end shopping malls and computer arcades. Salespersons would wait outside of their shops, and sometimes at the top or bottom of an escalator, so as to position themselves to seize shoppers passing by.
CONCLUSION
This study aims to characterize the spoken genre found in retail encounters in traditional Chinese local markets in terms of the social actions performed through highly routinized verbal and nonverbal behaviors between shoppers and vendors, and the communicative purpose(s) achieved by such social actions. To that end, I have examined both substantive acts that are found within, and in fact constitute, the speech event, and ritual acts that mark its boundaries. Interaction between customers and salespersons is characterized by domination of a transaction frame and subjugation of valet and consultation frames. Shoppers and vendors typically regard each other as outgroup persons between whom face engagement is motivated by the instrumental purpose of negotiating a transaction. The analysis shows that routinized retail encounters in Chinese local markets constitute a primary genre (Bakhtin 1986) that consists of a single kind of practice for achieving the tightly circumscribed communicative purpose of negotiating commercial exchange, and not interpersonal relationship.
At the beginning of the article, I reviewed some problems with taking a purely linguistic perspective on typed interactions in service encounters. The main criticism of circularity of argument is evident in how the data were selected and used. Instances of service encounters are included in the dataset that are deemed by the analyst to be illustrative of the genre and are then analyzed for the schematic structure that defines the genre. To reinject a sociocultural orientation, as I have attempted to do in this essay, is to begin by asking a different question: Why is this specific genre created and used by members of the community in the way it is? A different research question requires a different methodology. Instead of relying on the analyst's intuition, the spoken genre that is the object of this study is one that is identified by the community members themselves,14
Hanks 1996 notes, however, that socioculturally meaningful categories do not always have metalinguistic labels. Even when labels are available, Gumperz 1982 points out that the mapping between them and actual events are at best imprecise. Swales (1990:57) points out that there are even genre names used by a community with no genres attached them. Hence, the process of identifying a folk category is often not straightforward. For a discussion on how ethnographers can elicit folk categories in a community, see Saville-Troike (1989:129–90, 157–59).
As the focus of analysis of genre shifts from describing the form to explaining the form in terms of its rationale, so does the purpose of the analysis. Instead of classifying individual instances of text or predicting what speakers are likely to do in certain contextual configurations, the purpose of socially oriented form-of-genre analysis is to elucidate individual instances of use of generic resources and to investigate the meaning-making practices of the community as a whole.
As “conventionalized communicative events” (Bhatia 1993), genres serve as mental models for members of a community that both enable and constrain their discourse production and interpretation. Specific instances of service encounters, from the prototypical to the less so, are examined in their own right to reveal the strategic process that speakers apply to shape the socio-rhetorical context, which may include the communicative purpose itself (or more precisely, the degree of complexity of the purpose). Speakers often follow standard practices within the boundaries of the genre to reproduce a conventional communicative event, but they may also be innovative by deviating from expectations or violating constraints of the genre. Bhatia (1993:15) explains, for example, that generic knowledge is often exploited by members of specialist communities such as newspaper reporters and attorneys to achieve special effects that satisfy “private intentions within the framework of socially recognized purpose(s).”
As “social semiotic formations” (Lemke 1994) that represent culturally specific ways in which Field, Tenor, and Mode values are combined by a community to form recognizable communicative events (Eggins & Martin 1997), distinctive genres reveal the community's unique “intertextual-relations-construing practices that are a significant part of its culture” (Lemke 1994). In this article, I have examined only one formation, the bargaining genre, as it appears in retail encounters in Chinese local markets. I have not attempted to delimit the scope of its occurrence (i.e., in what other settings for commercial exchange besides the local market does the genre appear?), nor have I attempted to identify its contrastive genres. However, if we set up an analytical category using the communicative goal of commercial exchange as a defining attribute, a host of other events besides market retail encounters can be included: a sales encounter in a department store between sales clerk and customer, an exchange in a restaurant between waiter and customer, a visit at a traditional medicine shop between herbal doctor and patient, a sales meeting between representatives from two companies, and so on. The fact that some of these events likely exhibit a different textual form from the bargaining genre, and that they are likely perceived by the community as being distinctive and contrastive types of communicative events, reveals the criteria to which the community is oriented in meaning-making.15
Lemke 1994 points out that a community often construes intertextual relations along other dimensions of meaning, for example the interpersonal.
Last, as social constructions that are created, maintained, and changed by the discursive practices of individual members of a community, distinctive genres and the order of related genres are powerful indicators of changes happening in the larger social context. The economic scene in the People's Republic of China has been undergoing tremendous change since the economic revolution beginning in the late 1970s, a process fueled by the all-consuming trend of globalization. Retail marketing is on the front line of the change. New kinds of outlets, such as giant modern supermarkets and malls, are competing with traditional open-air markets and small local shops, and winning. High-end goods have grown in market share. Customers are willing to pay more for better service,16
“Retail Banking in China.” McKinsey Quarterly, 11.04.04, 10:25 AM ET 〈http://www.forbes.com/business/2004/11/04/cx_1104mckinseychina4.html〉 Accessed August 20, 2005.
