Service telephone call openings: a comparative study on five European languages

Abstract

The paper presents the results of a comparative study on how speakers of different languages (English, French, German, Italian and Spanish) manage the opening of service phone calls. Previous research has focussed on cross-cultural variability in telephone conversations, but this is the first attempt to systematically compare several European languages at the same time. The communicative strategies speakers use in each language are analyzed qualitatively and quantitatively, allowing a systematic comparison across cultures and languages and the observation of intra-cultural variability. Comparative analysis is based on five fundamental moves that may be performed in a telephone call opening: summons-answer, identification, greetings, how-are-you’s, getting-down-to-business. Implications are drawn for cross-cultural research on interaction and for training staff working in multi-lingual and multi-cultural settings.

Keywords: telephone calls, conversation analysis, cross-cultural pragmatics, English, French, German, Italian, Spanish.

Telephone call openings have been the object of a considerable amount of cross-cultural and intercultural pragmatic research. The first systematic investigation in this area dates back to Schegloff’s (1968) analysis of telephone calls openings in the United States. This and much of subsequent research was carried out within the Conversation Analysis (CA) paradigm, which implies careful observation of the details of interactions in order to uncover how social order is created and reproduced in everyday life. The fundamental analytic units are moves and sequences, whereby conversation is seen as a social activity in which interactants produce actions and counter-actions in a highly coordinated manner.

Schegloff (1968, 1979, 1986) identified four core sequences in his corpus of North American telephone call openings:

These four core sequences have been used as a basic ‘template’ (Hopper 1992) for describing telephone call openings in a number of studies. Other researchers have proposed further developments of this initial model. For example, Bercelli & Pallotti (2002, see also Bowles & Pallotti 2004) add a fifth move which they call ‘getting down to business’, to indicate those actions - typically performed by the caller, but at times elicited and co-constructed by the receiver as well - in which the reason for call is stated and oriented to (see also Schegloff’s 1986:116 similar but not identical notion of ‘anchor position’). These works, as well as the present contribution, analyze landline telephone calls only. More recent research has focused on mobile phone conversations’ openings, which have some distinctive features of their own (Arminen 2005, Hutchby, & Barnett 2005, Arminen, & Leinonen 2006).

Schegloff’s seminal papers were based on a corpus of telephone calls collected in the US. Among his findings was that callers, when they are acquaintances, prefer not to provide explicit self-identification, but would rather play the game of mutual recognition based on the small voice samples produced in the first turns. Several subsequent studies, however, have shown that this generalisation does not apply to other contexts, cultures and languages. To begin with, the preference for other-recognition only holds for private homes in the US, whereas in service telephone encounters among strangers, for example, receivers customarily provide some form of personal and/or institutional identity (Schegloff 1986:122). Secondly, in some countries like Finland (Halmari 1993, Arminen 2006), Germany (Werlen 1984, Varcasia 2003), Sweden (Lindström 1994), and the Netherlands (Houtkoop-Steenstra 1991), the preference is for both callers and receivers to self-identify, even when they are at home and speaking with acquaintances. Finally, the opening may be more or less routinised. For instance, as Sifianou (1989) has shown, Greek openings tend to be more varied, and every person develops an individual style. Several other differences have been noted cross-culturally, regarding e.g. if, when and how people identify themselves, the length and shape of greeting sequences, the presence of other moves like apologies for disturbing, the more or less direct interactional style of participants, the number of turns composing the opening and so on (for reviews see e.g. Luke, & Pavlidou 2002, Leonardi 2003).

Telephone call openings thus represent an ideal object of study for cross-cultural pragmatics research. Since these social encounters are very specific and strongly constrained by technology, the range of actions that can be performed in them is limited so that one can thus observe how different cultures and languages vary in their realisation of the same interactional routine.

The present study is the first attempt to provide a systematic comparison among five European languages on the way service telephone calls are opened. Data come from a number of calls made to a variety of businesses and institutions, including bookstores, travel agencies, hairdressers, language schools, libraries, student dorms, university departments. Languages investigated are Italian (N=159 calls), Spanish (N=63), French (N=59), British English (N=56) and German (N=53)(1). The parties involved signed a written consent before the study began, so that receivers agreed on being recorded but would not know exactly which calls would have been made for the research project. Most of the calls were recorded between 2000 and 2003.(2>

Data were recorded on audio cassettes or digital media and transcribed in CA format (Ten Have 1999; see appendix for conventions used).

The corpus size is relatively small but comparable to those used in previous research, which rarely went beyond a few scores of calls per language. This obviously raises issues of generalizability, and one should be very cautious about making general claims regarding millions of people based on a few scores of examples. Corpus size also prevents one from drawing finer distinctions among sub-categories. For example, Bowles & Pallotti (2004) have shown that there are a few differences between calls to bookstores and to other types of business in English and Italian, which may point to a more general claim that different activities exhibit systematic features in phone calls’ openings. However, in order to substantiate this claim, one would need robust corpora for each activity type and for each language, which would increase the size of the database considerably. Another area of possible differences may be related to geographical variations – it could be that speakers from different regions tend to open calls in specific ways. Again, this type of investigation, while certainly worth pursuing, would require region-specific corpora, which could be the subject of further studies on intra-linguistic variation. Some preliminary remarks on possible regional differences between Emilian and Sardinian calls in our larger Italian corpus were made by Bercelli (2003), who noted some subtle differences, for example, in the more common use of terms of address by Sardinians. In keeping with previous research, in this study we aggregated calls at the national level – and thereby at the linguistic level - and found some consistent trends for each language. More research on intra-linguistic variation is certainly needed, and we are currently working at making our corpus available on the Internet with automatized search options that will allow calls to be sorted by various parameters, such as geographical area, business size, caller’s gender and age (available soon on <linguistica.uniss.it>).

A first qualitative stage of analysis led to the identification of recurrent patterns in the data and the recognition of the most evident cross-cultural differences. In a subsequent phase, systematic comparison across languages was carried out by a quantitative analysis based on the core sequences framework presented above, with a few modifications leading to the following scheme, comprising five possible moves:(3)

In what follows, these categories will be used for a systematic cross-linguistic comparison, both at the level of individual moves and their realisation in different languages, and at the level of how such moves are combined in receivers’ and callers’ turns in the opening.

The use of quantification is uncommon in research inspired by the CA approach. However, as Schegloff (1993) notes, there is no intrinsic incompatibility between CA and quantification, which should be used cautiously, after a careful qualitative investigation has been carried out. One area Schegloff believes profitable for quantitative analysis is the comparison among different languages and communicative contexts (1993:116). Schegloff, like many other CA researchers, advocates the use of more informal, verbal types of quantification, i.e. words such as frequently, occasionally, massively. Much previous research comparing telephone calls in different languages has been stated in such terms, and the following citations exemplify how results were expressed in this type of studies.

There have been exceptions to this trend, and studies like Houtkoop-Steenstra (1991), Lindström (1994), Hopper & Chen (1996), Bangerter, Clark & Katz (2004), Arminen (2005, 2006), Bowles (2006) present results in both qualitative and quantitative terms. In this report, the exposition will proceed stating frequencies and percentages together with a conspicuous variety of examples, which contribute to illustrate the general points through the details of individual interactions. It should be made clear that, with this methodology, and type and amount of data, quantitative measures should be taken as indicative of trends and relative ratios, rather than exact values from a sample carefully designed to represent the whole population. This is why, in order to strike a balance between reporting accurate figures and avoiding the misplaced precision fallacy, we have chosen to round off figures to the first decimal in the table and to the integer in the text body. Furthermore, in order to increase reliability and replicability, the corpora which form the basis of our analysis are available on the Internet at the following address: www.gabrielepallotti.it.

The exposition is divided into two main parts, one devoted to the receiver’s first turn, the other to the caller’s - henceforth R and C in the transcriptions. For each part, individual moves are analyzed first, describing their frequency and how they are performed in different languages. An analysis follows of how these moves are combined, pointing to the most common turn formats for each language. In the conclusions, implications are drawn on how these findings may be used in the professions to improve intercultural communication practices.

After the caller’s summons, i.e. the phone ringing, the person at the other end of the line picks up the phone and starts speaking. These are the very first words in the call and a crucial interactional site, where several moves can be performed and social identities and roles begin to be defined. As Schegloff notes "The opening is a place where the type of conversation being opened can be proffered, displayed, accepted, rejected, modified – in short, incipiently constituted by the parts of it " (1979:25).

Minimally, one may simply display that the communication channel is open and that the conversation can proceed. However, it is often the case that other moves are produced in the receiver’s first turn, including greetings, self-identification, offers of availability and/or invitations to the caller to express the reason for the call. This lumping of moves in the first turn is functional to the pragmatic, business-oriented nature of a telephone service encounter. The receiver orients to the routinised character of the event, sorting out all the matters that need to be dealt with in the opening and rapidly proceeding to the core of the telephone call, the ‘business-at-hand’.

Although the same set of moves can be recognized in all the languages examined, there are cross-cultural differences in the frequencies of individual moves and in how they are sequentially combined within the turn. Table 1 in the Appendix provides a summary of the most common turn formats found in the corpus.(4)

All the languages included in this study have simple, minimal tokens that can be uttered to perform the basic action of showing that the communication channel is open, i.e. the two speakers have established contact and they can proceed talking. There are however important differences in the way such tokens are used.

First of all, the pragmatic meaning of these forms is not exactly the same. French allô and Italian pronto are specialized discourse markers that can be used only in the context of a telephone call and only with the function of checking for an open channel, in the beginning or later on. In Italian, French and Spanish ‘yes’ can also be used as a minimal first turn response, which may be followed by another channel opener, as in sì pronto, allô oui or sí dígame. However, only in French does one find the opposite order, i.e. allô oui.

English hello performs all these functions too, but it can also be used in other ways, most notably as a greeting, both in face to face interactions and on the phone. The same holds for German hallo, which is much rarer in service encounters and appears only once in our corpus. Spanish dígame/díga, on the other hand, besides being conventionally used as a phatic signal at the beginning of a telephone call, also conveys the more specific and literal meaning of ‘(please) tell me’, that is, an invitation to speak. Hola and alò, which are common in Latin American varieties of Spanish, never occur as channel openers in our corpus recorded in Spain.

Hence, channel openers in different languages have different semantic and pragmatic meanings, producing certain restrictions on their position within the turn and implications for sequential development, as will be seen below. Secondly, there is substantial cross-linguistic variation in frequency of token usage, especially when they are the only expression uttered in the receiver’s first turn. While in French, Italian and Spanish it is quite common to answer a service telephone call with a bare channel opener (27%, 30% and 36% of the cases respectively), this pattern is found only once in the German corpus and is completely absent from the English calls. Answering the phone with a simple hello or hi is not at all infrequent in English, but is strictly confined to the domestic space of personal interactions (Schegloff 1986). This distinction seems to be more blurred in other languages, which may mean either that the two forms are viable alternatives for opening a service encounter on the phone, or that, at least for some offices and businesses, the distinction between behaviour in domestic and public domains is less clear-cut.

Receivers often identify themselves upon answering the phone at the workplace. This occurs in all the languages studied, with varying proportions - 95% of the times in English, 89% in German, 71% in French, 67% in Italian, 62% in Spanish. Not providing one’s identity at home can be a way of playing the game of mutual recognition from small voice samples, which exhibits, confirms and reinforces the intimate relationship among callers (Schegloff 1986). The opposite effect is produced when a stranger calls a public telephone number - here the caller may be legitimately uncertain of having contacted the right address, and the receiver’s self-identification in the first turn pre-empts further questions. As a matter of fact, if such self-identification is missing, it is often the case that callers immediately initiate a repair sequence in which they ask confirmation about the receiver’s identity, as in the following example.

Ex. 1

Providing one’s identity in the very first turn is thus a way of optimizing the opening, as the caller, being reassured that s/he has reached the right number, can proceed directly to the reason for call.

There are several ways in which self-identification can be performed. The simplest and most frequent is by stating the place name, as in the following examples.

Ex. 2

Ex. 3

Occasionally, and only in some languages (French, Italian, Spanish, German), receivers may self-identify by stating not the proper name of their business or institution, but rather the more general category they belong to, like ‘library’, ‘butcher’s’, ‘post office’.

Ex. 4

Another common self-identification format is providing one’s name, which is relatively common in English (23%) and German (20%), less so in other languages.

Ex. 5

Ex. 6

As can be seen in these examples and in Table 1, with the exception of four German calls the receiver’s proper name is always uttered in association with a more institutional identification, be it the place name or its category.

Another frequently occurring move in the receiver’s first turn are greetings. Although they appear in all the languages investigated, their lexicalization and meaning differ in various ways. For example, while in English the greeting must be appropriate to a specific time of the day (good morning, good afternoon, good evening) languages like French and Spanish have a single expression (bonjour, buenos días) for the whole period from early in the morning to around 5 p.m.. While good morning in English is valid until noon, and is then followed by good afternoon, German guten morgen can only be said until around 10 a.m., after which guten tag becomes the appropriate formula until sunset. Furthermore, while some languages, like German and English, include a generic and relatively informal greeting such as hallo or hello, which can also be heard as a phatic channel opener, Italian and Spanish equivalents such as salve and hola never occur in the receiver’s first turn in our corpus.

Languages also vary in how frequent greetings are in the receivers’ first turns, from 79% of the cases in German, 73% in French, to 55% in English, , 40% in Italian and a mere 19% in Spanish.

Some qualifications to these data are in order.

Firstly, the figure reported for English includes only greetings such as good morning and good afternoon. The addition of turns containing hello raises the proportion of turns with greetings to 76%(5). Secondly, the low figure for Spanish can be accounted for by a particular conversational pattern found in this language, whereby the caller’s first turns often consists of simply uttering a greeting, which is then reciprocated by the receiver, as in the following example. In other words, Spanish receivers can greet either in their first (19%) or in their second turn (48%), and sometimes even in both (6%), which brings the overall proportion of calls with greetings by the receiver to 73% (Colamussi & Pallotti 2003).

Ex. 7

Another move that can be found in receivers’ first turns are offers of availability such as (how) can I help you? English is by far the language in which this move is most often produced, being present in 41% of receivers’ first turns. Examples like the following were uttered in a variety of offices and businesses, of various sizes and in different areas of the United Kingdom.

Ex. 8

Ex. 9

Expressions like (how) can I help you? seem to be not just stereotyped routines, frozen politeness formulas, but rather a way of actually expressing the receiver’s orientation towards the caller’s needs. This can be seen quite clearly in the following example, where the bookshop assistant uses the rather uncommon and creative can I do something before falling back on the more usual how can I help.

Ex. 10

This type of interactional move is also found in other languages, but with much smaller frequencies and confined to larger firms. This might be the result of training programs based on materials originally conceived for the English-speaking world, yielding such awkward and unnatural prefabricated sequences as in the following Spanish example:

Ex. 11

Having examined the moves that can be produced in the receivers’ first turn, we will now focus on how these moves are combined and linearly ordered. For each language, special attention will be given to the most common patterns and to those that seem to be peculiar to it.

As is clear from Table 1, French receivers use very few formats for producing their first turns. The most common by far is place identity + greetings, which alone accounts for nearly two thirds of the openings. Another 27% of the calls is answered with simple channel openers, for the most part single tokens of allô or oui, and occasionally a combination of both as in allô oui and oui allô. Three calls were initiated with the place identity, followed by the receiver’s identity and greeting, a pattern that wasn’t found in other languages except German. These three formats account for 95% of the calls, with the following being a characteristic French opening:

Ex. 12

The two most common formats in Italian are the same as in French, namely place identity + greetings (28% of the cases) or a simple channel opener (30%). However, Italian receivers tend to employ a wider array of answers in their first turn. Among these, the simple identification of the place being reached is relatively common (15%), which may also be accompanied by a channel opener such as pronto.

Ex. 13

English is the only other language in our corpus in which this phenomenon is represented, in openings like hello meridian school. However, hello here can be seen both as a channel opener and a greeting, so that the sequence channel opener (proper) + place identity seems to be peculiar to Italian.

Spanish is the language in which service telephone calls are more frequently opened with a simple phatic signal, like sí, díga or dígame, a choice that appears in over 36% of the calls in our corpus. Callers seem to be used to it and they don’t treat these cases as particularly problematic. In fact, while French and Italian callers confronted with a bare allô or pronto tend to ask for confirmation of the receiver’s identity one time out of two, Spanish callers carry on producing their turns more than 90% of the times without any special repair work.

Another common opening move is simply stating the place name, as in banco de españa or farmacia. This seems to be coherent with a general attitude to conciseness in Spanish openings, a language in which one finds exchanges like the following, containing only the minimally necessary moves, without greetings, identification or pre-requests.

Ex. 14

"R: hello?
C: hello (0.2) uh to reserve a table?"

(ACS 25)

Another turn format which is relatively common in Spanish (11%) and peculiar to this language is place identity + dígame, as in the following example.

Ex. 15

Here one can see how dígame is a different channel opener from French allô or Italian pronto, or even Spanish sí. In fact, the literal meaning of dígame is ‘tell me’, a meaning that is still transparent to speakers despite the ‘ritualization’ (Haiman 1994) of the form as a conventional way to answer the phone. This is why dígame always occurs at the end of the turn when this contains other moves, such as channel openers proper (like sí) or the receiver’s identification. In other words, one finds openings like peluquerìa dígame? or sí dígame?, but the reverse would just be impossible, as dígame peluquerìa and dígame sí literally mean ‘tell me hairdresser’ and ‘tell me yes’. The following example shows how dígame, besides being a channel opener, is also used and heard as an invitation to speak. The receiver in her first turn answers with a greeting and her personal identity; after reciprocation of the greeting by the caller, the receiver encourages her to speak with a dígame.

Ex. 16

A variety of opening formats is found in the German corpus, none of which appears to be particularly prevalent over the others. A common feature however is the receiver’s self-identification, which occurs in 91% of the cases. This preference for self-identification has been described for other North-European languages like Dutch (Houtkoop-Steenstra 1991) and Swedish (Lindström 1994) and can also be observed in the callers’ first turns, as will be shown below.

Among the most common formats one finds simple place identity (25%), possibly followed by greetings (25%). Receivers are also quite ready to provide their personal identity (20% of the calls), more frequently associated to that of their workplace, but in a few cases as the only form of self-identification. Providing one’s name is also a standard opening in private telephone calls (Werlen 1984), and when it is done in a service encounter it is delivered in a fast, automatic way, often without any accompanying lexical material like am Apparat or am Telefon (the equivalent of ‘[xyz] speaking’), as in the following examples, where the receivers’ names (Dittman and Katy Grimm) are given immediately after the place identification.

Ex. 17

Ex. 18

In English too there seems to be no frequently recurrent pattern in the way telephone calls are opened. A large variety of formats is found, none of which accounts for more than 16% of the cases. This figure refers to an opening that occurs in all languages, the association of place identity and greetings. However, while in the other languages greetings tend to follow identification, the reverse order is normally found in English.

Ex. 19

As we said, another common move in the English corpus is the offer of availability, which is found even in relatively small businesses, such as a fishing shop (boat rod and parrot can I help you?) or a florist (hello melany can I help you?). In one case, a car accessories shop, the offer of availability was the only move in the first turn:

Ex. 20

A move that occurs only in the English corpus is thanking the caller, as in the following example.

Ex. 21

After having examined how receivers answer the phone, in the next sections we will focus on callers’ first turns, beginning with an analysis of individual moves followed by the description of the most common turn formats in different languages, summarized in Table 2 in the Appendix.

Speakers of languages having specialized channel openers for the phone, such as Italian pronto or French allô, also use such forms when they act as callers. For example, 51% of the French callers’ first turns begin with allô or oui and 15% of such turns are initiated with pronto in Italian. With respect to Italian, French callers may choose which of the two forms to use to respond to the summons, and they seem to prefer oui as second part, used in two thirds of these calls.

Ex. 22

English and Spanish have words that can be used both as channel openers and greetings, namely hello/hi and hola. Their basic function is that of an informal greeting, which was subsequently extended to the role of phatic signal in phone conversations. These forms may be used on their own or in combination with another greeting formula, which is usually a more formal and explicit greeting, as in hello good morning or hola buenos días.

When the second greeting form is also produced, it reinforces the greeting made in the first form, so that its main role is that of channel opener. In English hello is used together with another greeting form only in 3% of the calls, whereas it is found on its own in the majority of the caller’s first turns (86%). The following examples, 23 and 24, show the prototypical use of this move in the two languages.

Ex. 23

Ex. 24

In Spanish the use of hola together with another form of greeting is much wider than in English, accounting for over 75% of the calls. Whereas in English the channel opener is the same phatic signal for receiver and caller, ie. hello, in Spanish the caller’s response to dìgame, acting both as channel opener and preliminary greeting form, is hola.

In German, instead, the callers open their turns either with greetings or identification, as their interlocutors do, because there is nothing that works specifically as channel opener in the institutional context. Hallo, which is the equivalent signal for opening the channel, is restricted to the informal context.

Greetings are a very frequent move in callers’ first turns. French and German callers greet 98% of the times, Spanish 95%, Italians 94.5% and the English 89%. The less frequent use of greetings by English callers might depend on the fact that, as pointed out in the previous sections, English receivers display their availability to satisfy the interlocutor’s request in a number of cases, and this may lead to speed up the opening phase towards the reason for calling. In fact, in all cases where callers directly get into the business of the conversation, skipping the reciprocation of the greetings (9% of the English corpus), the receiver’s turn contained an explicit offer of availability, as in the following example.

Ex. 25

In all instances of this type the caller replies to the offer of availability with an acknowledgement token, yeah, directly followed by the request.

When it comes to identification, considerable differences emerge among languages, as can be seen in Table 3 in the Appendix. For example, while English callers never self-identify, in Germany self-identification is almost categorical and its lack may not only be of relevant absence, but also sanctionable (Schegloff 1968) if receiver has identified her/himself. The principle that comes into play is that even if they do not know each other, speakers need to introduce themselves before starting to talk about anything.

German callers’ identification is such a routinised move that it may be produced without any ‘frame’ (Schegloff 1968:353), but just by uttering the name of the person speaking, as in the following call.

Ex. 26

German callers identify in 87% of the cases, a pattern that has been described for other northern European countries, such as Sweden (Lindström 1994) and the Netherlands (Houtkoop-Steenstra 1991), both in private and service calls.

Callers identify in their first turns 19% of the times in Italian, 12% in Spanish and 3% in French. This seems to happen only when it is relevant for the request that is going to be made, so that self-identification should be seen as part of the preliminary work to the request, like in ex. 27.

Ex. 27

In this specific instance the caller gives her identity because she is calling to communicate her miles card number. Although her identification will be explicitly requested by the receiver later in the call, the caller also provides it spontaneously in her first turn. It should also be reported that the caller had already been to the travel agency and had promised to call back for providing the code number. Hence, caller’s self-identification may also be seen as prompting recognition from the interlocutor. This interpretation is strengthened by the micro-pause following identification, which seems to project a possible recognition turn, which however doesn’t occur.

Apart from these more specific cases, it seems that English, French, Italian and Spanish callers don’t find lack of self-identification particularly ‘accountable’ (Garfinkel 1967), as opposed to what happens in Germany and other Northern European countries. In these contexts an asymmetry is produced whereby the receiver’s identification is preferred and required, while that of the caller can normally be omitted. In other words, callers are automatically and implicitly recognised by their interlocutors as (possible) customers and this seems to be taken as sufficient for the conversation to proceed.

As in the previous section, we will now look at how single moves are combined in callers’ first turns, pointing to commonalities and differences among languages. Table 2 summarizes the most common patterns.

75% of French calls are initiated by the caller’s greeting and getting down to business, which may be preceded by a channel opener like oui or allô. Ex. 22 in the previous section is an example of this latter category, while ex. 28 below shows the most frequent pattern of greetings followed by the request.

Ex. 28

In both instances we have quite long request turns. Speakers deal with the accomplishment of the opening moves at the very beginning of their turns and pass on to the reason for calling quickly.

Use of the channel opener at turn beginning is quite frequent in French. One also finds, for example, turns composed of just the channel opener and greetings (8%), or channel opener and greetings followed by a request of confirmation of the receiver’s identity (8%). Altogether, in 51% of the French calls the caller begins his or her turn with oui and/or allô.

Although Italian callers exhibit a wider variety in their first turns, the most common pattern (46%) is the same as in French, i.e. greetings directly followed by the reason for the call, as in this example.

Ex. 29

"R: alma good evening
C: e: good evening mh i’d like apiece of information madame
listen i recently had a perm (.) but now
i would like to colour ((my hair)) >is it possible<?
R: yes"

(ATI 02)

Of the five groups studied, Italians are, after Germans, those who most often provide their identity, which they do 18% of the times. As noticed for the Receiver’s identification, callers that self-identify in their first turn do it either by providing their name, or by stating their belonging to a social category, such as clients, mothers, students, etc. Italian callers in our corpus identify themselves with a category almost half of the times they provide an identification. Below follows an example of turns with greetings followed by identification, the second most used format (10%).

Ex. 30

In this example the caller self-identifies by providing the category she belongs to, student of the faculty of Modern Languages of the University of Bologna, and this kind of identification is prior and embedded to the discount request.

The most frequent turn format by far in Spanish, and quite peculiar to this language, is the caller’s simply greeting, with no additional moves. Most of the times a double form of greeting is produced, as hola buenos días, a sequence found in 51% of the calls. Another 6% consists of a form of greetings not prefaced by hola. After this turn consisting of greetings only, the receiver greets back, and the caller then proceeds with other moves - reason for call, but also negotiation of the receiver’s identity - in their second turn. This opening ritual is very frequent in Spanish and seems to be peculiar to this language (Colamussi & Pallotti 2003).

Ex. 31

In this example the caller greets twice in her first turn, then she asks the receiver to confirm his identity in the second, and formulates her request in the third turn only.

Finally, another 25% of the calls is represented by turns in which the moves are compressed in the first one with the use of hola plus another greeting form plus the request, such as in ex. 24 previously analysed.

As we have already said, German callers very frequently identify themselves. This usually occurs after the greetings and before the reason for call (64%), but one also finds identification preceding the greetings and getting down to business (11%) and even a few examples of turns composed of just greetings and identification (6%). Example 32 can be taken as a quite typical way of call opening in German.

Ex. 32

English callers seem to be quite uniform in their behaviour, contrary to what happens when they act as receivers. In fact, the most common turn format by far is a greeting directly followed the reason for call (82%). Otherwise, in another 9% of cases, greetings may be omitted and the caller begins by simply stating their request. This however occurs only after a receiver’s offer of availability, which in a certain way urges the caller to rush to the request, as in the following example.

Ex. 33

In other words, upon hearing how can I help you?, callers have two competing motivations in producing their first turn. They might reciprocate the greeting, which was however produced quite a while before in the receiver’s first turn, but they also have to provide a second pair part to the request to formulate how they wish to be helped, which is the most recent move in the interaction. It is thus not surprising that this latter format may somehow push them to skip greetings in order to answer as soon as possible the receiver’s question.

Several practical implications can be drawn from these results. The first is that, although telephone call openings appear at first sight to be a relatively simple and invariable routine, there is actually great variation in the way they are realized across languages, institutions and individuals. Paraphrasing Schegloff (1986), the ‘routine is an achievement’, and every time we are confronted with the apparently simple and straightforward task of opening a phone call, we must find an appropriate way, which cannot be determined a priori, as the outcome is always co-constructed with the other interlocutor and is sensitive to various contextual variables. These constraints are weaker on the receiver’s side, as his or her first turn is not dependent on previous talk, which explains the lesser variety found in these turns. However, even such cases may exhibit a degree or variation, depending e.g. on how close one is to the phone, on the activities one may be engaged with when the phone is ringing, on the calls that were made in the preceding seconds or minutes, and so on (Schegloff 1986:118ff). The caller’s task is more complex, as his or her turn is contingent upon previous talk by the receiver and must be tailored to the variety of requests to be made.

Even among speakers of the same language there is thus a high variability in the way telephone calls begin. Given this inherent variability, does it make sense to compare different languages and cultures, as if Germans, French, Italians all behaved in the same way? The present study has shown that there are no single ways in which speakers of a language open a telephone call. However, quantitative analysis shows that certain patterns are more frequent in one language than in others and such differences can be considerable. Some moves and turn formats which are quite ordinary in one language are virtually absent from another, and even when the same moves are produced, they may appear in different orders with different sequential implications.

Some of these differences meet the eye of the researcher and the layperson alike quite easily. Such is the case, for example, of the strong tendency for German callers to self-identify, vis á vis the lack of such a practice in English and French, or the frequency with which Italians, French and Spaniards answer the phone with a simple ‘hallo’, a format that is virtually absent from German and English service calls. Other differences are less evident, and only a careful analysis can make them apparent. Among them, the tendency for Spanish speakers, and to a certain extent for the French, to produce greetings in separate turns - i.e. the caller’s first and the receiver’s second - without inserting them in turns containing other moves; or the fact that in some language like French there seem to be only two or three main ways to answer the phone on the workplace, whereas much greater variation is found in German and English.

Other areas of non obvious differences concern the linear order in which various moves occur within turns. For example, channel openers like French allô and Italian pronto always occur at the beginning of the turn, while Spanish dígame only at the end. In English there is a strong tendency for greetings to occur at the beginning of the receiver’s first turn, a format that is much rarer in German and virtually absent from Spanish, Italian and French, where greetings are produced at the end of the turn or in any case after uttering the place and/or receiver’s identity.

Such findings may influence intercultural communication training programs in several ways. First, one should acknowledge the fact that there is no pan-European standard way of answering the phone. Even a relatively widespread pattern - identifying the place and greeting - is realised in the reverse order in at least one language, English. From this follows the dubious validity of training programs based on literal translations of materials originally written in other languages. An opening like good afternoon travelrunners independent can I help you?, which is quite natural in the UK and uttered even in small businesses, becomes almost ludicrous when it takes the Spanish form super líneas santander buenos días la atiende ricardo dígame en que puedo ayudarle, which strongly contrasts with the usual patterns for that language, characterized by very short turns in the opening phase.

Staff working in multilingual settings should thus be made aware of the peculiarities of the pragmatic routines in each of the languages spoken, including their own native language. This would make them realize what is unmarked, appropriate behaviour and what might be seen as more or less sanctionable deviations from such standards, both on the receiver’s and the caller’s end. Opening a telephone call is just one small communicative ritual in professional daily life, albeit a frequent and important one. There are hundreds, thousands, of similar rituals which display such cross-cultural variation, and it would be virtually impossible to provide in-depth training for each one of them. However, taking one exemplar case like telephone calls and unravelling all of its complexity might be a way to sensitize workers to the fine details of cross-cultural variation in communicative rituals.(7)

Table 2. Caller’s first turn formats in different languages (percentages).

Table 3. Caller’s self-identification in the first turn (percentages).

¹ All callers were native speakers of the languages investigated. To the best of our knowledge, all receivers were native speakers too; at least none of them had a clearly non-native pronunciation.

² We wish to thank Lorena Bissiri, Anna Colamussi, Paola Contu, Gloria Deriu, Silvia Nieddu, Anna Tilocca, Manuela Riu for sharing their corpora in the present research project.

³ The scheme appears to be reliable: besides having been replicated in a number of previous studies, it was easily understood by us and our co-workers. Very few doubts arose as to how a particular move should have been classified, and consensus was reached for all of them. A few categories are inherently ambiguous, like the double function of English hallo as channel opener and greeting, or that of Spanish dígame as channel opener or prompt for formulating a request. However, this is due to structural reasons rather than to the unreliability of raters’ judgments. Such special cases are discussed in the text.

⁴ For technical accidents, in nine German phone calls the receiver’s first turn was not recorded, partially or entirely. Quantitative analysis of the receivers’ first turns will be carried out only on the remaining 44 complete calls. To increase sample size, the German corpus consists of 35 calls recorded for this study plus 9 calls taken from articles by Jäger (1979) and Werlen (1984).

⁵ Schegloff (2004) provides a thorough discussion on whether hello at the beginning of a phone call is to be considered a greeting or an answer to a summons. His conclusion is that it can be heard as either or both, the interpretation largely depending on what participants do in subsequent talk.

⁶ This is one of the very few cases in our corpus in which caller and receiver seem to know each other. However, as becomes clear in subsequent turns, the caller only knew he had to contact Mrs. Rossi, although he had never talked to her before. More generally, callers and receivers in our corpus did not know each other and only in exceptional cases had they been in professional contact before.

⁷ An anonymous reviewer asked whether our results might be ‘related to something deeper and more general in the structural aspects of those cultures’. As conversation analysts, we tend to be very cautious about making broad generalizations at the level of ‘culture’ or ‘ society’. However, in different research paradigms the observation of small details in everyday practices can indeed be used as grounds for more general observations on a culture’s ethos. Some ethnographers, for example, have used CA methods to analyze the micro-level of interactional organization, which was then related to wider-ranging sets of practices, attitudes, and beliefs relevant for ‘macro’ constructs such as person, responsibility, and social distance (Moerman 1988, Duranti 1997: 264-79). .Everyday routines such as opening a phone call may contain ‘rich points’ (Agar 1996) which open broader perspectives on a culture’s basic structuring. For example, the finding that there is a strong tendency for Germans and speakers of other Nordic countries to self-identify when they make or answer a call may point to broader conceptualizations of what is appropriate behaviour in the public domain and the types of information that can be easily made accessible to others. Similarly, the British a ssistants’ frequent statements about their readiness to help and their consistent self-identification seem to display a strong concern for customer satisfaction, which stands in contrast to the more i nformal Spanish and Italian openings, where callers are often addressed with a simple hallo. This attitude is also consistent with other types of behaviour: for instance, British shop assistants are frequently ready to provide suggestions and alternative solutions when they cannot satisfy a customer’s request, whereas Italians tend to just say no, letting the customer initiate the repair work. This was found by Varcasia (2007) in our telephone corpus, but also by Zorzi (1990) in face-to-face bookshop encounters, showing that there are indeed recurrent cultural patterns that cut across contexts, media and time. This type of investigation is certainly worth pursuing, but goes beyond the scope of this paper. We can only hope that our results may be useful for further investigations relating the micro and the macro levels of social organization.

Gabriele Pallotti is associate professor of Applied Linguistics at the University of Modena and Reggio Emilia. His research interests include second language acquisition, cross-cultural and intercultural pragmatics, conversation analysis, with special attention to methodological and epistemological issues.

Cecilia Varcasia is Researcher at the Centre for Language Studies of the Free University of Bolzano (Italy). Her research interests are conversation analysis, inter- and cross-cultural pragmatics and second language acquisition.

Mail address for correspondence:

Gabriele Pallotti, Università di Modena e Reggio Emilia, Facoltà di Scienze della Formazione, Via Allegri 9, 42100 Reggio Emilia (Italy).

name.surname@unimore.it
name.surname@unibz.it

Journal of Intercultural Communication, ISSN 1404-1634, issue 17, June 2008.
Editor: Prof. Jens Allwood
URL: http://www.immi.se/intercultural/.