Keywords

1 Introduction

A chef at a well-known sushi restaurant in Tokyo said, “I can see what customers are like by the very first order they make. It is a duel—either I overwhelm them or they overwhelm me.” For many Japanese, going to a sushi restaurant is a special experience. They feel anxious when opening the door and uneasy when ordering sushi. Typically, no written menu is provided and customers are required to know what fish is good for the season. No price tag is presented, and customers are anxious about how much the bill will amount to [1]. Customers eat at the counter in front of a chef, who is typically seen as an authoritative artisan. Having a sophisticated conversation with the chef is generally considered appropriate. Customers are expected to engage in many rituals, such as eating their sushi by hand and doing so as soon as it is served. Confident customers are adept at giving a good impression of themselves to the chef so that they are accepted as connoisseurs.

Such a duel is not necessarily peculiar to sushi restaurants. Describing any first-time service encounter, Solomon et al. [2] wrote, “… the first time customer will be especially vigilant as he/she assimilates such environmental clues as the appearance and demeanor of the service provider.” The provider does not know who the customer is, either. A restaurant’s service staff must “size up customers” to determine how to treat them [3]. Sushi restaurants are perhaps a setting in which these kinds of duel-like interactions can be observed more saliently than in other settings.

A service encounter is defined as “the direct interactions between a retail or service firm and its clients” [4 p. ix]. It is the moment of truth in which the real value is cocreated—and recent debates suggest that this is so for manufacturing as well as service industries [5]. The particular problem with service encounters stems from the fact that it is typically strangers who come into contact and make the transactions [4, 6]. Therefore, most service encounters are made predictable through routinization. Employing role theory, scholars have claimed that customers coming to acquire a service have learned the role so that they know how they must behave and how they will be treated [2]. The insight that initial routinized service encounters involve negotiation of selves appears at odds with the concept of routinized encounters being based on roles. Predictability has been emphasized in the service encounter literature. Scholars have certainly discussed personalized services as an opposite end of predictable services.

This study proposes overcoming such a sharp dichotomy of predictability and personalization. To do so, we draw on Ervin Goffman’s work on interaction rituals [7]. While service encounter scholars have used the notion of rituals, rites, and ceremonies in conjunction with roles and scripts [2, 8–10], they have not paid much attention to Goffman’s characterization of the social encounter. For Goffman, interactions are ritualized not only to maintain order but also because they involve “sacred objects,” that is, the self.

The goal of this study is to demonstrate this ritual nature of service encounters. We videotaped actual service encounters between sushi chefs and customers at two prominent sushi restaurants in Tokyo. Because customers’ and restaurants’ selves are a particular issue in the very first contact, we focus on the initial interaction that is invariably a sequence of ordering drinks. In this chapter, we do not attempt to report the full range of interactions observed, which would require much more space. Instead, we offer a detailed analysis of a few selected examples so that the duel-like interaction can be clearly revealed.

2 Rituals in Service Encounters

2.1 Predictability and Personalization

Services are particularly unpredictable because they are difficult to grasp firmly prior to experiencing them and because customers are involved in the core production of the service; this differs from the experience of buying products that have already been produced [6, 11]. Scholars have used role theory—“a dramaturgical metaphor” [2]—to explain how this predictability issue is managed. In service settings, “each participant has a role to play; the script from which he/she reads is often strictly defined” [2]. For example, “the customer role in an elegant restaurant involves very different actions than an appropriate role in a fast-food setting” [2]. Based upon this premise, it is claimed that “service encounters are role performances” (p. 101). Rituals or rites are used as another way to render services predictable [9]. “Rites are social dramas with well-defined roles for people to perform… Rites, in general, have been shown to aid in achieving consistency and predictability” [9].

Therefore, prior theory of service encounters puts much weight on predictability. The basic thesis is that if customers can predict what will happen based on clearly defined roles, better satisfaction can be attained: “Accurate mutual comprehension of role expectations is a prerequisite for a satisfying service experience” [2].

Participants are required to have learned these roles so that they know what to expect. As such, these studies do not explain how participants behave when they have not learned the roles. Solomon et al. [2] in fact wrote:

Regardless, there are always behaviors which may come automatically to a veteran but which demand great cognitive activity by the novice. Anyone who recalls the adolescent apprehension over “doing the right things” on one’s first solo outing to an expensive restaurant can attest to this (how to talk to the waiter, what to order when the menu is written in another language, how much to tip).

Nonetheless, such cases are treated as exceptions and are not discussed much more than this. As the opening quote suggests, it is this interaction, in which roles are not necessarily completely transparent, which is particularly important for the service encounter.

Scholars at the same time have recognized that predictability is not enough to make service encounters possible. Customers expect personalized service encounters instead of being treated as anonymous role occupants. Prior studies treat predictability and personalization as conflicting goals. The choice between them is made based on contingency. Services that involve customers (such as high-end restaurants) are seen to require more personalization, while services with less involvement (such as fast-food outlets) require more predictability. In fact, more personalization in low-involvement cases leads to negative evaluation by customers because they value predictability, efficiency, and low cost [2, 9, 10].

There are two different meanings of personalization in the service encounter literature. The first definition emphasizes customization. Surprenant and Solomon [10] called this “customized personalization” based on “the desire to assist the customer in attaining the best possible form of the service offering for his or her needs.” Sushi restaurants are a prototypical example of this customized personalization. Customers choose each sushi item that they want to eat, one at a time—this style is called “Okonomi.” Although many sushi restaurants in Tokyo have adopted a course menu, or “Omakase,” the Okonomi customization is often emphasized as the most authentic way to eat sushi. The order is placed through interaction with the chef.

The second definition describes personalization as a way to treat a customer as an individual person with warmth, friendliness, and care. Mittal and Lasser [12] define personalization “as the social content of interaction between service employees and their customers. Thus, “personalization” concerns the manner in which service employees relate to customers as people—cold and impersonal at the one end to warm and personal at the other.” Surprenant and Solomon [10] characterized “programmed personalization” as the effort “to give the impression of personalized service by encouraging small talk, using customers’ names, and so on” (p. 89).

The first definition treats personalization as an adaptation to customers’ needs. The second definition also views the customers’ selves as fixed. Therefore, who the customer is matters for neither of these definitions of personalization. Nonetheless, several hints can be gleaned from the literature. Scholars have touched on the “self” only implicitly. Solomon et al. [2] wrote:

The self also can be thought of as a system of identities to which one is more or less committed (Jackson 1981), where commitment to a role implies a concern that one's role enactment be convincing (Sarbin and Allen 1968). Self-concept related to a role constitutes a role identity (McCall and Simmons 1966) (emphasis added).

Mittal and Lasser [12] briefly wrote, “customers may seek routine self-validation even from ephemeral encounters such as with a department store clerk (e.g., ‘doesn’t this suit look great on me?’)” (emphasis added). On this ground, this study seeks to bring the notion of self to the fore.

2.2 A Sociological Approach: Goffman and Ethnomethodology

These prior studies of service encounters used concepts of dramaturgy, role performance, and rites to emphasize predictability. Interestingly, Goffman used the same set of concepts to discuss seemingly opposite effects: presentation and negotiation of selves. That is, instead of reducing individual selves into roles, Goffman describes interactions in which individuals present their selves and negotiate them through interaction with others. Goffman’s sociological approach shifts attention to interactions themselves as opposed to individuals and their relations. In fact, service encounter researchers have emphasized interactions as a unit of analysis [2, 13]. Klaus [8] specifically called for analysis of interactions including both verbal and nonverbal actions based on audio and video recordings. However, the actual interactions—how people talk—have rarely been studied.

Goffman [7, 14] revealed more nuanced ways in which roles are performed. Goffman analyzed how one’s self is presented and negotiated through social encounters. He wrote, “the individual will have to act so that he intentionally or unintentionally expresses himself, and the others will in turn have to be impressed in some way by him” [14]. One seeks to “control the conduct of the others, especially their responsive treatment of him,” by “influencing the definition of the situation which the others come to formulate” [14]. Any social encounter involves such an interactional construction of self. When one presents oneself and proposes a certain definition of the situation, others also project their own definition of the situation, which in most cases is in line with that of the initial person. Typically, individuals are oriented to avoid undermining others’ “face” [7].

The meaning of ritual here is different from the rituals discussed in service encounter literature [2, 8, 9]. Goffman [7] wrote:

I use the term ritual because I am dealing with acts through whose symbolic component the actor shows how worthy he is of respect or how worthy he feels others are of it… One’s face, then, is a sacred thing, and the expressive order required to sustain it is therefore a ritual one.

Prior studies used the concept to highlight how individuals come to appreciate service without surprises—that is, predictability. Goffman, conversely, emphasizes the order in which individuals deal with their and others’ selves as sacred objects. This image of rituals implies not so much predictable and unsurprising encounters as duel-like encounters. Goffman [7] wrote:

As sacred objects, men are subject to slights and profanation; hence as players of the ritual game they have had to lead themselves into duels, and wait for a round of shots to go wide of the mark before embracing their opponents.

Although Goffman was the first to pay attention to how interactions actually occur, he did not have a fully developed methodology for empirically analyzing interactions. Ethnomethodologists developed the empirical program to study social interactions based on recordings of naturally occurring interactions [15–17]. These ethnomethodologists established the field of conversation analysis.

Ethnomethodology seeks to describe the methods by which individuals produce orderly practices [18]. Like Goffman, it rejects external criteria to understand a particular practice; instead, it analyzes how participants themselves display their actions as understandable in certain ways. Here, an ethnomethodological notion of accountability is important. Actions are accountable because they are produced to be observable and understandable to others in the situation. Therefore, a social order is not found outside the practices, but within them. Individuals display an action methodically so that others can understand it. Others then present their own understanding of the action by means of their action. Through this sequence, routines become routine.

Ethnomethodology offers a way to examine how individuals render their actions routinely and orderly. It therefore enables a systematic analysis of what Goffman described in only impressionistic ways. Using this ethnomethodological approach, this study examines the initial service encounters in which customers’ and providers’ selves are presented and negotiated.

3 Method

Conversation analysis was chosen as the methodology to analyze social interactions. Interactions between chefs and customers were videotaped with multiple camcorders. Because it was not feasible to videotape customers without permission, those study participants who agreed to video recording were recruited. Observations were made in the evening at two sushi restaurants in Tokyo—Restaurants A and B. Both were prominent sushi restaurants repeatedly featured in books and magazines. The typical bill amount at both restaurants ranged from 20,000 to 30,000 Japanese yen. The sessions lasted about 2 h.

Eleven customers were observed in Restaurant A. There were three two-person groups, one three-person group, and two individuals. Therefore, there were six groups of customers. One was a regular customer. There were three chefs and a few assistant staff members. In Restaurant B, six sets of customers were observed. There were two two-person groups. Two customers were highly experienced in sushi, and there were no regular customers. In total, 17 customers were observed.

Conversation analysis begins with transcribing interactions word for word and proceeds to repeated review to understand the interactions in detail. The standard transcription system is explained in Table 1. It is important to analyze each instance of interaction on its own and ground the analysis in what is observable in the sequential data prior to comparing it with other data. After each interaction was understood properly, structures of the interaction were specified. Structures in conversation analysis consist of sequences of pairs called adjacency pairs—for example, question/answer and request/decline. When structures of all instances were described, instances with the same structures were grouped together.

Table 1 Standard transcription system
Full size table

The analysis is based on how participants, themselves, understand actions. As ethnomethodology assumes that actions are accountable, participants’ understanding of each action is observable in the action. In particular, whether a particular analysis is correct is confirmed by the subsequent action that responds to the action. The subsequent action manifests a certain understanding of the previous action. When someone says, “Why don’t you come and see me sometime?” and the other person says, “Yes, I will,” this latter action shows his understanding of the original action as an invitation. If the response is something like “Oh, I am sorry. I have been busy,” then the second person is shown to have understood the initial action as a complaint. If the first person continues, “No, I didn’t mean that,” then it is shown that she took the second person’s response as an apology but corrected the meaning of her initial action as an invitation.

Because of its sociological interests, this analysis does not seek to specify individuals’ psychological constructs such as intention, emotion, and satisfaction. Participants themselves cannot be sure what others have in mind. They take action in such ambiguity. If an analyst pins down an intention when participants themselves are not sure, then the analysis is external to the phenomenon. This does not mean that conversation analysis cannot be used to investigate issues concerning concepts of the “individual,” such as the self or personal feeling. If such “individual” components are analyzable as relevant features of interactions between participants and, therefore, become observable and analyzable objects for participants, conversation analysis can reveal how they are constructed as recognizable matter, through a detailed examination of the actual interaction. This study focuses on the concept of self and demonstrates how it becomes relevant in service encounters.

When identifying sequential structures of actions, analysts will undoubtedly encounter exceptions that do not fall under the identified structures. These are not exceptions only to the analyst. The analyst should first examine how they are exceptions to the participants themselves. If participants recognize a particular event as an exception to the norm and take action accordingly, then the analysis is further enriched. Therefore, the identified structures of action should not be seen as factual, but as normative.

4 Initial Service Encounters in Sushi Restaurants

4.1 Nonexpert Customers

A typical example is shown here. We contrast the case of less experienced nonexperts with that of experienced sushi connoisseurs. The first line of each turn denotes original Japanese in Roman letters and the second line is a literal translation. “Ch” refers to the chef and “C1” is a customer. There was another customer in the party, namely, C2.

Fragment 1

01Ch:e::: .hhh          sassoku          desu          ga (0.5) [ono] mimono ha dou shima shou ka.=

U:::m .hhh to begin (0.5) [how] would you like to do about drinks.

02 C1 : [hai]

[Yes]

03C1:=a::::(.)<mu>site ru n de: nnama biiru de:

Mm::::(.)because          it          is          <humid> ((today)) : ((I’ll have)) ddraft beer:

04 Ch : >nama          biiru          iki mash[ou<

>Let’s go with draft b[eer<

05 C1 : [nama de ii desu [ka¿

[Is dra[ft ((beer)) ok[ay¿

06 Ch : [>sou          desu          ne<

[>Yes          it          is<

The first question in line 01 is literally translated. It is a “how” (dou) question and includes “doing” (shimashou). In almost all the cases observed, this format was used. The “how” question is more open ended than a “what” question. It does not necessarily ask for a specific drink item. The answer could also include not having drinks. Although this question appears to be benign, it is actually not an easy question to answer. The customer had just been seated. This was the first contact after the greeting. Further, no menu was provided. There was no hint of the available drinks, from the surrounding environment, except for what other customers were drinking.

Although it may appear easy to order something common, that is, beer, which is available in any restaurant in Japan, it is not an obvious choice. First-time customers would not even know the system of this restaurant, for example, how the meal is organized, whether they need to order appetizers or whether appetizers are served without having to be ordered, what would be the best drink to begin with, and so on. The choice of beer, itself, is not trivial. There are different kinds, including draft and bottled beer, as well as different brands and sizes. On top of all this, some people say that beer does not go well with raw fish, and sushi is typically associated with Japanese rice wine (sake). The customer was forced to respond to the question without knowing all these details.

Therefore, with this first standard question, the chef projected a definition of the situation, as Goffman put it. This definition also includes the self-image of this restaurant. The question is assumed to be an unproblematic and a natural one for the restaurant’s customers. That is, customers who come to this restaurant should be able to order, in response to that question. This analysis does not suggest that the chef was deliberate in posing this difficult question. It is enough to describe how the question works in the factual interaction. The analysis is validated by the response of the customer, as seen below.

Customers typically did not answer this question in a straightforward manner, except for regular customers who knew the restaurant and experienced sushi connoisseurs, as shown below. In the fragment, the customer ordered draft beer in line 03. Some observations can be made. First, he delayed the answer with the filler sound “Mm….” Second, he stretched the word beer (in Japanese, “de:” indicates an incomplete sentence). The stretched sound indicates that he did not complete his utterance. He needed something to complete it: a reaction from the chef. These two factors were observed consistently in many other cases.

Other typical responses included displaying one’s inability to order so that the chef could offer some hints. When asked the initial question, a customer looked right and then left. This bodily behavior indicated that he needed some information in order to be able to make his order. Another customer responded with laughter, to which the chef responded with a few hints.

Coming back to the fragment, we see the customer further provided a reason for his order: “because it is humid.” This action is awkward if we consider this sequence as a simple exchange of information. Why does the customer provide a reason for his order? Plainly, by this action, he presented that his order required an explanation. In other words, he showed that his order was not immediately obvious and might not be completely appropriate.

By this action then, the customer projected himself as someone who could order but gave, or gave off in Goffman’s terminology, the impression that he was not ready to be a fully integrated customer. Some kind of negotiation of self was evident.

In line 04, the chef accepted the order, but his response was emphatic. The “let’s” format has “we” as the subject. Therefore, the chef involved himself in the decision that the customer made. In other words, this can be heard as the chef’s emphatic agreement with the customers’ choice. What was this action doing? The chef was confirming the definition of the situation that was presented before him. He did more than confirm the order; he confirmed the customer’s performance.

This kind of emphatic acceptance of an order was not common in our data. However, providers typically accepted the order in most cases, just in a less emphatic way. Accepting an order appears to be commonsensical. However, the providers needed to ask a subsequent question to specify the order further. “Beer” was not enough because there were different sizes and kinds of beer. Therefore, if they accepted the order, it meant that they could not start the specification sequence immediately. In most cases, providers accepted the order first before jumping on to the subsequent question.

In short, this example shows that although the very first interaction was highly standardized and ritualized, the way this interaction was done involved more than simply making the ordering process more predictable and easier. What the customer and the restaurant were like were at issue in this interaction.

4.2 Regulars and Sushi Connoisseurs

The above analysis can be made clearer if we look at cases of experienced customers. Because of the lack of space, we report one example of a non-regular sushi connoisseur. The abbreviation AS refers to an assistant, and not the chef, in this case. C refers to a customer. Kirin is one of the major beer brands.

Fragment 2

01 AS : onomimono          ikaga          itashi mash[ou ka

How          would you like to do about d          r[ink

02 C : [biiru wo

[Beer please

03 (.)

04 AS : biiru          ga (.)          oobin          kobin to [gozai masu ga

For            beer (.) ((We have)) large and small [bottles

05 C : [n↑:

[We↑:ll

06 C : jyaa kobin de.

All right then a small bottle.

07 (0.2) ((AS nods))

08 AS : kobin wa (.) ( ) ga

For a small bottle (.) we have ( )

09 C : e:            Kirin            de o[negai] shima#su#

Umm: Kirin beer [plea]#se#

10 AS : [hai]

[Yes]

The first question was in the same format as in Fragment 1. The customer’s response, however, was different from the previous example. In this case, neither the filler nor the stretch was observed, nor was a reason for the order. The customer simply uttered a complete minimal sentence: “Beer please.” In fact, the original Japanese was even simpler because the “please” part (“wo”) was not complete.

The assistant’s acceptance was also marked. Unlike in the previous case, the assistant did not accept the order and instead went straight into specifying the order, that is, the size of the bottle. The same assistant exhibited a different pattern with less experienced customers who exhibited similar filler and stretch.

5 Discussion

The first question is the same for both less experienced and more experienced customers—it is typically different with regulars because the chefs know who the customers are. The subsequent sequence also appears to be the same at the surface level. Yet, the subtle differences reveal the customer type. This understanding of customers’ selves is also evident in how the providers respond to their orders.

The initial service encounters are highly routinized. Roles are clearly defined and participants can read from scripts [2]. Nonetheless, predictability is not the same as a lack of drama. Goffman used the ritual metaphor to show that individuals present and negotiate their sacred selves. Rituals are both a source of predictability and a social drama in which individuals’ selves are constructed.

Without this subtle understanding of the dual nature of interaction rituals, prior theory on service encounters has underscored predictability and relegated all other unpredictable components to the notion of personalization [2, 10]. Predictability and personalization were seen as a trade-off. The analysis above has revealed that personalization is central in the highly standardized interactions. These two are therefore not in a trade-off relationship.

To see this more clearly, we can provoke a new conceptualization of personalization. The personalization discussed in the literature takes one of two forms. First, encounters are personalized with friendliness, warmth, and sympathy [10, 12]. Second, encounters are personalized to the extent that each individual’s needs are understood and addressed with customized services [10]. Although the second form of personalization is relevant to sushi restaurants, the initial encounters were highly standardized.

The first definition is one-sided because in sushi restaurants, service encounters are more like a duel than a situation in which warmth, friendliness, and sympathy are appropriate. Of course, we do not argue that chefs and restaurant staff members were rude. They were polite and courteous. The analysis focuses on the subtle and nuanced duel behind the courteous interactions. Interactions in which one’s self is at issue are not necessarily uncomfortable. Customers would not pay the restaurants’ high rates to receive unsatisfactory service.

Why, then, are sushi restaurants organized this way? By creating a community of sushi connoisseurs, restaurants can create cultural value in addition to the value of the services offered at each restaurant. The cultural value lies in the fact that customers feel they are experiencing a special service that only sushi connoisseurs are fully qualified to enjoy. They might be oriented toward learning to become connoisseurs themselves or at least to become able to behave confidently in such restaurants. Customers then enjoy the experience of developing their selves further through the service encounter. They have gained one round of experience and hope to do better the next time.

If this is the case, then we should understand services in a nuanced manner. Being friendly with customers or tailoring services to meet customers’ needs is only one way to create value. In these approaches, customers’ selves are outside the scope of research. This study seeks to highlight an alternative approach using Goffman’s notion of an interaction ritual through which selves are presented and negotiated.

6 Conclusion

We analyzed the very first interactions of service encounters in sushi restaurants in Tokyo. The detailed analysis revealed that the encounter began with an open-ended “how” question regarding drinks without any information as to what drinks were available and how the meal would be organized. With the first question, chefs projected a definition of the situation: Their customers should be capable of answering the question. With the answer, then, the customers presented their own definition of the situation. The subtle cues showed that they presented their order in such a way that they probed how the chef would react to it. The chefs then accepted the order first, sometimes emphatically, before moving on to the further refinement of the order.

Based on this analysis, we argue that in service encounters at sushi restaurants, customers’ and chefs’ or restaurants’ selves are at issue. We argue that such presentation and negotiation of selves is a critical aspect of service personalization. The existing concepts of personalization were shown to be limited, and an alternative conceptualization was proposed: Services are personalized to the extent that participants’ selves are negotiated.

Although the analysis was based on data particular to sushi restaurants, the theoretical argument is not restricted to that context. In other kinds of restaurants, scholars have discussed the importance of providers’ “sizing up” customers [3]. Scholars have also discussed the experience of going to elegant restaurants: Doing the right things is often problematic [2]. Therefore, similar interactions should be observable in other services. The findings should also be relevant to non-restaurant services. For example, taking a course with a well-known professor at a university may be similar, as students are not only gaining knowledge but also developing their selves.