Misconduct, Malfunction and Dissatisfaction. A Social History of the Organisation of Complaint-Handling in Telephone Services

Disponible en ligne sur www.sciencedirect.com

ScienceDirect Sociologie du travail 58 (2016) e1–e20

Misconduct, Malfunction and Dissatisfaction. A Social History of the Organisation of Complaint-Handling in Telephone Services夽 Benoit Giry Centre Émile Durkheim, UMR 5116 CNRS et Université de Bordeaux, 3 ter, place de la Victoire, 33076 Bordeaux Cedex, France Available online 19 October 2016

Abstract The article’s subject is the genesis and development of complaint-handling practices in France’s telephone services. It shows that, far from being a new facet of contemporary production models, complaint-handling has its roots in a long historical tradition in the organisation of telephone services. Looking back to three important moments in the history of telecommunications (the telephone crisis at the beginning of the century, the decade of the “telephone modernisation plan” that began in 1974, and the introduction of the “new legal and commercial framework” in the 1990s), we suggest that complaints performed different organisational functions, which shaped complaint-handling in different ways and assigned varying meanings to the complaint. The text depicts the work of complaint-handling as both influencing and reflecting the changes in a public administration that is now a commercial firm. Beyond its descriptive interest, the article raises the question of the management of conflicts with customers and their influence on labour and its organisation. © 2016 Elsevier Masson SAS. All rights reserved.

Keywords: Complaint; Telephone Departments; Complaint-Handling; Organisation; Market; Client

Contrary to what a brief consideration of its modern meaning might suggest, the etymology of the word “complaint” refers not only to the voicing of a grievance but also to a social relationship. Its root is the Latin word plangere (“to lament”) and the prefix com (“with”), suggesting a

夽 First published in French: “La faute, la panne et l’insatisfaction. Une socio-histoire de l’organisation du travail de traitement des réclamations dans les services du téléphone”, Prix du jeune auteur 2014, Sociologie du travail 57 (3), 277-298 (http://dx.doi.org/10.1016/j.soctra.2015.05.001). Translation: John Crisp. E-mail address: [email protected]

http://dx.doi.org/10.1016/j.soctra.2016.09.022 0038-0296/© 2016 Elsevier Masson SAS. All rights reserved.

e2

B. Giry / Sociologie du travail 58 (2016) e1–e20

collective expression of grief or hurt, which came into the English language via the Old French complaindre, or complainte in its noun form. The word lost its usage as “lament” in the 17th century, eventually operating a semantic shift to its current English definition for our purposes, “a demand for something that you believe you have a right to, especially from a company, the government, etc.”,1 which implies a structured expression addressed to someone with the status to exercise coercive power over people and things. For the purposes of this article, we will consider the word in its most common modern usage where, since the rise of consumer rights, it has taken on a semi-legal meaning in parallel to its technical legal meaning, so that making a complaint (e.g. about poor service) in order to express displeasure (to “lament”) has evolved into making a complaint for the purpose of obtaining redress or compensation. Sociology has recognised this new usage by developing studies on the origin, formation and formalisation of complaints as part of the spectrum of action available to customers, consumers or users. The authors show how before verbalising their dissatisfaction by claiming, the customer-consumer-user must have named (perceived the problematic character of their experience) and blamed (identified a guilty party), to employ the categories established by Abel et al. (1981). The complainant may have been helped in this process by different actors who have contributed to the “social construction of the dispute” (Pinto, 1989). They will then have made their complaint, employing strategies that are to varying degrees political (Fijalkow, 2006) or consumerist (Barrey, 2002), while seeking to “grow” within coherent moral orders (Trépos, 1988, 1991), in accordance with the conception of critical activity proposed by Boltanski and Thévenot (1991). Although the shift of certain civil conflicts from the courts towards alternative dispute settlement methods, as well as the takeover of complaint-handling by organisations, give the subject a certain relevance today, corresponding scholarly efforts in France to study the way in which companies deal with complaints have been modest.2 French sociology, economics and history are generally interested in the productive and commercial functions of companies, rather than in their capacity to handle customer complaints, whereas a significant programme of research has developed since the 1970s in the English-speaking world on complaints (Day and Landon, 1977), on complaint-handling or complaint management (Fornell and Wernerfelt, 1988) and on service recovery (Krishna et al., 2011), in marketing and in management.3 It might be added that sociology, which came late to the study of the firm (Borzeix, 1986), has long thought of productive organisations as largely structured by conflicts specific to internal relations of production rather than those derived from commercial relations.4

1

According to the online Oxford English Dictionary: http://www.oxforddictionaries.com/. For the European Union, see in particular: “Commission Recommendation of 12 May 2010 on the use of a harmonised methodology for classifying and reporting consumer complaints and enquiries”, Official Journal of the European Union, L136/1. Consumer Empowerment, TNS Opinion & Social, 342. With regard to the organisational sphere, the AMARC (customer complaint management association), a grouping of large companies, was set up in 2004. 3 The handful of studies outside this framework were produced by legal scholars, sometimes using the tools of legal sociology (Littlefield and Ross, 1978). 4 Pierre Dubois published “Complaint handling in the textile industry” in 1968, but he was referring to “worker claims” (Dubois, 1968). The issue of Sociologie du travail dedicated to firms contained articles focusing on the question of industrial relations and on the changes brought about by the Auroux laws (Borzeix, 1986). Similarly, the issue of the “complaint” was recently explored by Olivia Foli (2008), but within the framework of employee relations. 2

B. Giry / Sociologie du travail 58 (2016) e1–e20

e3

Conversely, while some writings discuss the complaint-handling function (Joseph, 1988; Borzeix et al., 1995; Ughetto, 2011) almost exclusively in studies focusing on service relations, they approach it primarily as an “activity”, isolated from the organisation in which it takes place (Ughetto, 2013). These approaches therefore ignore the question of the connection between complaint-handling practices and the other practices coordinated by the organisers within the framework of a firm’s productive process, or quite simply exclude them on the grounds of their “non-productive” character (Ganem, 2011). By contrast, this article seeks to make an empirical contribution to the different fields of sociology that explore the position of the product’s recipient in the organisation of production. The various forms of customer inclusion are described by numerous authors as a salient feature of new production models (DiMaggio, 2001; Maugeri, 2006; Dujarier, 2008; Tiffon, 2013). Through a socio-historical detour (Inset 1), we will show that, in the telecommunications sector, the product recipient emerged very early in the figure of the “complainant”, and that this figure impelled the organisers to introduce the “complaint” as an object that gave rise to the work of “handling”. This article therefore seeks, through the prism of a study of the genealogy of “complaint-handling work”, to position itself within the lineage of developments in the sociology of market professionals (Cochoy and Dubuisson-Quellier, 2000; Kessous and Mallard, 2014) which, enriched by contributions from the sociologies of management, provide fresh insights into the connection between organisation and market. It illustrates something like the empirical construction of a “reaction function”, as understood by Albert O. Hirschman (1970): the establishment, within a productive organisation, of a system for adapting to the customer’s voice.

Inset 1: Socio-historical investigation This article is based on archive and documentary research carried out as part of a sociology thesis on complaint-handling practices and their impact on the work and organisation of two big companies in France. The sources cited come from research conducted at the Library of Post and Telecommunications History and the National Library of France, the Library of the Conservatoire National des Arts et Métiers and the National Archives (documents from the National Archives are indicated by the abbreviation “AN” and are subsequently referenced in the text with their classification mark). Sociological research based on archives demands certain precautions, such as the need to consider the conditions under which the documents studied were produced. An archive sometimes appears to be speech without a speaker, a pure source, whereas “it reflects, above all, the reasonings and questions of those who produced it” (Israël, 2010, p. 175). In short, the archive has an angle: it is often difficult to investigate anything other than the history of the viewpoints of the people who left a record, who are often – in our case – the organisers. Our essential aim will therefore be to reconstruct these viewpoints which, in this sense, will tell us how practices were organised and the meanings that the organisers wanted to assign to them. It is helpful, in addition, to show the changes that those responsible for complaint handling made to those practices at the point when they were actually executed.

e4

B. Giry / Sociologie du travail 58 (2016) e1–e20

Our study shows that, as an object, the “complaint” seems to impose at least two constraints on the organiser in shaping the handling work. First, the complainant wants to communicate with someone in “authority” and tends to move around until they find such a person. In companies, where hierarchical structure coordinates the activity of individuals (Williamson, 1975), authority is distributed between several actors with varying degrees of clarity. Those responsible for handling complaints must be in a visible situation of authority, in other words appear to be sufficiently legitimate as interlocutors to assess the justice of a complaint, make a decision about it and have that decision implemented. At the same time – and this is the second constraint – the complaint has to find a place in a productive plan and an organisational configuration that may sometimes have been constructed without regard for it. The work of handling must therefore make the complaint useful and connect it with the other practices organised for the purpose of production and sales. It is this utility that, in the final analysis – in the eyes of the organisers – justifies the presence of such practices in the company. To study the construction of these practices is therefore to seek to understand the uses of complaints associated with them. Complaint-handling practices cannot therefore be understood outside the context of an organisation, of an economy of working practices. The organisation of complaint-handling is not the product of some fixed, once-and-for-all management rationality, frozen into a model or into a caricature with the traits of a Taylor or Ford. Its history is one of successive formulations, transformations in the object to be managed, reforms in management rationality. We should not be deceived by the apparent consistency of the figure of the organiser: true, with the exception of Georges Mandel’s central complaints department (see below), handling practices have always been organised by engineers. Depending on the period, however, the engineers have had to accommodate to other actors who limit their control over the organisation: at the beginning of the 20th century, they had still to collaborate with civil servants; after 1970, their hegemony over management positions was complete (Atten, 2013), but from the 1990s they had to adapt, in their practice, to commercial imperatives. An ahistorical or unsituated rationality needs to be replaced by historically situated reason in action, which tackles practical issues. This article is first of all the history of the emergence and evolution of a problem. In looking at the practices employed to organise complaint-handling within the framework of a pragmatic history (Cohen, 2001), it proposes to examine how the problem of complaints emerged in the organisation of the telephone service, together with the working practices developed in response by the organisers, at three key points in the history of the sector in France. It does not seek, therefore, to cover the whole 20th century. It looks first at the genesis and development of complainthandling, constructed in the 1920s at a time when the telephone was in “crisis”, depicted as a “national embarrassment” (Section 1). It is silent about the long period of “errance”, which lasted from the 1930s to the 1970s,5 but shows how the problem of complaints arose again in the heart of the decade of “telephone modernisation” (1974-1985), during an organisational transformation conducted with the aim of developing the network (Section 2). The article stops with the 1990s, when the organisation changed its legal status, redefined its business and saw its market deregulated (Section 3). In each of these periods, there was a change in the meaning of the complaint; subsequently, each of these changes gave rise to a different combination of a small number of elementary tasks in the managerial shaping of complaint-handling work. 5 Regarding her period of study (1910-1938), Alexandra Bidet writes: “This period initiated a long “errance” in the telephone sphere in France; from the 1930s to the dawn of the 1970s “modernisation”, the country’s telephone infrastructure and service remained essentially unchanged” (Bidet, 2005).

B. Giry / Sociologie du travail 58 (2016) e1–e20

e5

1. The Paris telephone network’s complaints boards and the central complaints department (1922-1935) In the early 1920s, “complaints boards” were installed in the Paris telephone network’s central offices.6 These pieces of equipment, which differed from the “multiples” where the telephone switchboard operators worked,7 were the stations where subscribers’ “claims” and “complaints” were handled. Gradually, a small group of operators separated from the body of ordinary operators, to come and work on the “boards”. An article of 1922 – when the Paris telephone network had around 120,000 lines (Bertho, 1984b) – suggested that a complaints board should have four operator positions per “multiple” of 10,000 subscribers.8 Based on the principle that this ratio was achieved in practice, we can estimate that some fifty positions were dedicated to the task of complaint-handling at that time. 1.1. Genesis of a “complaints service”: an economy of talk The establishment of this service was first of all linked to the network engineers’ desire to improve the “speed of the telephone”, in other words, in a system that relied almost entirely on manual switching, to increase the productivity of the switching operators. In the space of switching practices, complaints were “unnecessary words” (Rougier, 1927) exchanged during the connection process, which considerably slowed down production. The engineers considered that this reduced the productivity of the demoiselles du telephone (telephone girls),9 directly when one of them entered into conversation with a complainant, and indirectly when their supervisor, occupied in handling a complaint, could no longer monitor them. In 1920, therefore, the goal was to keep complaints away from the workers. In this respect, the boards service, at a time of “telephone anarchy” marked by permanent subscriber dissatisfaction, was a response to “real needs”:10 its introduction was a way to “relieve” the telephonists and supervisors, to whom complaints were particularly addressed, “from these tasks, which often prevented them from concentrating on their normal work”. From 1922, therefore, the instructions were clear: “under no circumstances should a switchboard telephonist enter into conversation with a complaining subscriber”; she was simply to “refer the subscriber” to the “complaints service”. In the same way, “if a complainant asks for the supervisor, the operator should automatically and without further discussion transfer them to the complaints board”.11 6 In provincial post offices, handling complaints was the responsibility of the Postmaster. For calls made from the Paris head post office, on rue du Louvre, a small room in the right wing of the building adjacent to the telephone kiosks, explicitly dedicated to “complaints and information”, had existed since the inauguration of the building in 1888 (source: “Le nouvel hôtel des Postes à Paris”, La Nature, revue des sciences et de leurs applications aux arts et à l’industrie, 16e année, 2e semestre, Masson, Paris, 1888). 7 In other words, the manual connection of subscribers using sockets and jacks. 8 “Les Tables de Réclamations du Réseau Téléphonique de Paris”, Annales des Postes, Télégraphes et Téléphones n◦ 11, 1922, p. 456. 9 Switchboard staff at the time, both operators and supervisors, were exclusively female, nicknamed “les demoiselles du téléphone” (telephone girls). 10 The post-war telephone service generated a degree of discontent that was much discussed in the specialist journals. The difficulties were depicted as a “national embarrassment” (“Une honte nationale. L’anarchie téléphonique”, article published by the Association of Telephone Subscribers, Texier, La Rochelle, 1909) or, more soberly, as a “telephone crisis” (“La crise du téléphone”, Revue des Téléphones, Télégraphes et TSF n◦ 18, 1925, pp. 87-91). 11 “Les Tables de Réclamations du Réseau Téléphonique de Paris”, op. cit., pp. 449-451, for all the passages cited in this paragraph.

e6

B. Giry / Sociologie du travail 58 (2016) e1–e20

The complaint-handling space created by the boards was initially designed on the same principle as the switchboard section (Bidet, 2005). In fact, it was not separate, but acted as an overflow space. In order to “economise” the board operator as well as the switchboard operator she replaced,12 the instructions prohibited: “[...] any exchange of words that is not strictly necessary to the performance of the service [, which] delays not only the current communication, but also those that follow, [and which] subsequently exposes the operator to further complaints and finally increases her fatigue and by extension the fatigue of all her colleagues” (“Le service des réclamations et le rôle des surveillantes”, Revue des Téléphones, Télégraphes et TSF n◦ 70, 1929, p. 360). This imperative of silence was not solely driven by economics. Its function was also to leave some uncertainty about the status of the person handling the complaint, so that she could be construed as an “authority”. That is why, when the complainant “asks for the supervisor”, the “operator automatically and without further discussion” was to transfer them to the boards service. This is also the reason why the “complaints board must not give its title in answering the subscriber”, but reply “only with the words ‘Hello. Yes?’ [literally ‘I’m listening’]”.13 In short, therefore, the procedure was a recording task, in which any exchanges had to be productive: “[the board operator] asks the subscriber for their telephone number and invites him to describe his complaint. She notes in her ledger (with the time of the call, the number of the initial group and the subscriber’s telephone number) all the facts given to her.”14 This economy of discursive exchanges was also reinforced by a veto on revealing the service’s operational secrets. Madeleine Campana, an operator from 1921 to 1934, illustrates this point in her memoirs with the story of a day when the Gutenberg exchange had to be partially evacuated for rat control, and she had found herself “unable to tell [her] dear subscribers that here, in the 20th century, the temple of progress [had] experienced a mediaeval invasion of rodents” (Campana and Jaubert, 1976, p. 134). The main effect of these instructions was to deprive the operators of supporting arguments in their exchanges with complainants, who would persist, to varying degrees, in holding them responsible for telephone malfunctions.15 The containment of certain exchanges with customers contributed to the outlines of a new object in telephone management. At this stage, however, while the board operator was useful, complaints were not. Managing complaints was seen as a negative externality, an undesirable work residue that constantly resurfaced in the midst of normal activity. The only desirable outcome was its evacuation. This “negativity” carried over even into the accounts, where the boards service, intrinsically a debit item in the balance sheet,16 first appeared to be a necessary evil: the aim was to contain an undesirable proportion of interchanges with early century subscribers brought up “with a horror of the telephone” (Londres, 1930, p. 53), in the hope that the costs generated by the operation of the “boards” would be offset by an increase in the productivity of the operators of 12

The expression “operator economy” comes from A. Bidet (2005). “Les Tables de Réclamations du Réseau Téléphonique de Paris”, op. cit., pp. 451-452. 14 Ibid, p. 452. 15 This representation of the responsibility of telephone operators in the production of the service would be written into literature by Marcel Proust, who depicted them as “Omnipotent”, as “ironic Furies”, as “ever irritated servants of the Mystery”, as “shadowy priestesses of the Invisible”, as “capricious Guardians”, all epithets that highlight the immense importance of their “character” in the sweep of production (Proust, 1920, my emphasis). 16 “In other words, that it was recorded as an item of expense, with no counterpart on the revenue side. However, the organisation of such a service must fulfil two essential conditions, which I will study further on, which are: 1. They must be strictly economical, i.e. cost as little as possible; 2. But “fully achieve the intended goal” (Pilloned, 1925, p. 813). 13

B. Giry / Sociologie du travail 58 (2016) e1–e20

e7

the “multiples”. Nevertheless, the first managerial use of the complaint would gradually emerge in the course of this period. For technical and moral reasons, linked with the poor quality of contact with subscribers, this utility would initially be played out less on an external stage, that of relations with the complainant, than on an internal stage, that of the organisation, through the practices of “enquiry”. 1.2. The Department tribunal: the complaint as misconduct As noted by an engineer early in the century, the complaints boards were a way to “seek out immediately, where they exist, the responsibilities involved, the misconducts and errors committed” to which complaints might point.17 This was not a matter of taking complaints at face value but – in the course of a procedure in which elements raised in the complaint were compared with elements produced by the organisation (incident book, supervisor’s opinion, etc.) – of testing the hypothesis of an “operator error”. The description of the activity thus borrowed its terminology from the law: the complaints service “institutes an enquiry”, “investigates complaints”, draws up “statements” on the basis of “evidence”.18 The object of this enquiry was both commercial and professional; it was authoritative both for subscribers and for the organisation: the aim was to establish “a record of incidents of service error”,19 to “rule in full possession of the facts on a request for a reimbursement or rebate”, to “give the public the explanations needed” but also “to assign responsibilities, to assess the professional value of the agents at fault, to apply disciplinary penalties fairly proportioned to the seriousness of the errors made and where applicable to prescribe for the department the measures that the enquiry had shown to be required”.20 Repair, compensate, explain, punish, prescribe: the different elementary tasks of complaint-handling were thus laid down in the 1920s. They designated a range of social relations, between professionals or between professionals and customers, likely to be produced within the framework of complaint-handling. Nevertheless, at that time, questions of repair, explanation, prescription and compensation, though present, largely came second to the issue of punishment. For the organisers of the 1920s, the complaint was less to do with technical faults or prejudice to customers than with assigning blame. Indeed, they developed an aetiology of complaint that focused on misconduct. A 1928 booklet addressed to postmasters describes the different reasons for complaints as follows: “The subject of these complaints is usually a delay in answering calls, a mishandling or mistake (untimely call terminations, wrong numbers, etc.), impatience or rudeness on the part of the operator” (“Réponses aux réclamations. Notice à l’usage des receveurs”, Secretary-General of Posts, Telegraphs and Telephones, Ministry of Trade and Industry, 1928. Document with no visible editor, calligraphic manuscript, unnumbered pages). Fault might be attributable to the subscriber or the operator, but never to the appliances, despite the technical unreliability of the interwar telephone. This conception largely determined differences in the implementation of the tasks in complainthandling practices. For example, compensation – extremely rare because of the barriers erected by 17

“Les Tables de Réclamations du Réseau Téléphonique de Paris”, op. cit., p. 449. “Le service des réclamations et le rôle des surveillantes”, Revue des Téléphones, Télégraphes et TSF n◦ 70, 1929, p. 360. 19 “Les Tables de Réclamations du Réseau Téléphonique de Paris”, op. cit., p. 449. 20 “Le service des réclamations et le rôle des surveillantes”, op. cit., p. 361, my emphasis. 18

e8

B. Giry / Sociologie du travail 58 (2016) e1–e20

a series of legal provisions – was conditional on proof of an “operator error”. The forms in which customers were given explanations and the instructions addressed to employees also depended on whether or not an error had been made that legitimised the complaint.21 This conception would seem to imply that the organisers essentially exploited complaints for disciplinary purposes. In the event of a presumption of fault, the board operators would refer it to the central office controller who, on the basis of the file, was authorised to punish one of his subordinates. Indeed, this method of dealing with complaints was broadly consistent with the employment principles promoted in the 1920s documentation, implicit in the statement of the qualities expected of an operator, from “fear of the boss” to “pride in the work”.22 Punishment, which arouses “fear” and stings “pride”, was one more method of tightening control over the switchboard operator’s activity, the objective of all organisational practices at the time. The propagation of complaints through the practice of the enquiry would grow in importance throughout the 1920s, but it was in the mid-1930s that these management principles were most spectacularly applied. On 12 December 1934, Georges Mandel, then recently appointed Minister of Posts, Telegraphs and Telephones (PTT), set up the Central Complaints Department (SCR) on rue de Grenelle, in order to remedy the failings of the decentralised complaints services, which he considered ineffective because they were both “judges and judged”.23 The SCR was independent of the production units, but attached to the ministry. Its purpose was to centralise the handling of written complaints,24 but also to implement enquiries and sanctions, and to publicise their results. In the words of the director of the Minister’s office, Georges Wormser, the SCR was “something entirely new”. It was headed by a general inspector of posts, a certain Mr. Girodet, whom G. Wormser described as “an extremely energetic man, with deep familiarity with the activity”, who was charged with the “mission of settling everything within forty-eight hours, or three days at a maximum”. G. Wormser described how the department worked: “[Girodet] received some 120 or 125 letters a day, sorted then, and extracted perhaps 15 or 20 that he would refer to me for personal investigation. Of those 15 or 20, I would bring perhaps 2 or 3 to the attention of Mandel. All this worked very effectively” (Wormser, 1975, p. 90). Indeed. In its first week of operations, between 14 and 21 December 1934, the SCR received 701 complaints, of which 235 related to the telephone service. The Revue des Téléphones, Télégraphes et TSF commented: “The volume of complaints addressed to this department was so large that the number of agents initially assigned to it had immediately to be increased” (“L’activité du service des réclamations aux PTT”, Revue des Téléphones, Télégraphes et TSF n◦ 132, 1935, p. 76). Between 1 and 15 January 1935, according to a report in L’Express du Midi, “1240 complaints were received”, 346 of them concerning the telephone service.25 Every week, the SCR 21

Ibid., pp. 359-362. “Les bureaux-écoles téléphoniques en Allemagne”, Revue des Téléphones, Télégraphes and TSF n◦ 19, 1925, p. 170. 23 Appointed on 8 November 1934, G. Mandel was described as an authoritarian minister. He ordered multiple dismissals and transfers at different levels of the hierarchy. His general objectives were to submit the administration to the “vigilant censorship of the public”, to “tighten control” within it, as he put it to the senior committee of the PTT (Favreau, 1996). 24 The creation of the SCR did not change the methods of handling “ordinary” complaints— i.e. oral complaints (“Central complaints department”, Bulletin d’informations, de documentation et de statistique n◦ 2, Ministère des PTT, Paris, Imprimerie Nationale, pp. 57-58). 25 “Le service central de réclamations des P.T.T.”, L’Express du Midi, Toulouse edition, Sunday 20 January 1935. 22

B. Giry / Sociologie du travail 58 (2016) e1–e20

e9

sent its activity report to the press. The report covered the number of complaints received, the services concerned, and also the sanctions arising from the enquiries. This public airing of professional punishment, a component already present in the letters from postmasters responding to complaints,26 thus received national coverage: “Since a large number of these complaints were justified, Mr. Mandel had to apply 105 sanctions, going as far as temporary exclusion” (L’Express du Midi, Toulouse edition, Sunday 20 January 1935). “Is there any need to add that this not only upheld the complaints that appeared justified, which unfortunately remain far too high in number, but that sanctions were applied each time it was appropriate” (“L’activité du service des réclamations aux PTT”, op. cit., p. 76.). The better-known the SCR became, the greater the influx of complaints. In the first fortnight of April 1935, for example, 1699 complaints were handled, giving rise to 202 sanctions, ranging from recommendation to dismissal. The department was nevertheless wound up after the Minister’s departure. Mandel’s SCR was a borderline experiment designed to serve political interests. However, it represented the culmination of a form of managerial logic that was already implicit in the “boards” section of the Paris network, holding up a magnifying mirror to a system already in place. In this system, the term complaint was understood in its legal sense, as something needing to be “upheld” and likely to result in punishment. The complaint concerned potential professional misconduct whose grounds were to be judged from information collected in an enquiry, and the form in which the complaint was embodied was the “statement”. Having formulated the complaint, the complainant was largely excluded from the procedure, which remained entirely in the hands of professionals, who took on the role of “judges” whose action was directed towards their subordinates, who for their part became “culprits” within the framework of the disciplinary procedure. 2. Complaints in the “telephone modernisation plan” (1974-1985) The documents covering the 1940s, 1950s and 1960s reveal a certain languor in the management of the task of complaint-handling. Overall, the 30 years following the Second World War were in no way “glorious” for the telephone services. They were a period of stagnation (Bertho, 1984a).27 The automation of the switching process, underway since the 1920s, was slow and behind schedule. The number of subscribers remained stable at a low level. The instructions relating to complaints in the PTT remained unchanged in their different versions, in terms of the allocation of tasks and the roles of provincial postmasters, telephonists, supervisors, etc.28 These 30 years were also marked by technical stagnation. A chief engineer’s report tells us that while “equipment without jacks or sockets was designed in 1967, its use presents a certain complexity, since the boards are fitted

26

“Réponses aux réclamations. Notice à l’usage des receveurs”, op. cit. At the same time, the PTT’s administration was, for its part, undergoing significant upheavals, which led to a gradual increase in autonomy in the managements of postal services and telecommunications services. These changes would initially benefit the postal service and are illustrated by the alterations in the role of the postmaster (Join-Lambert, 2001). 28 The decree of 10 May 1946, which created separate managements for Posts and Telecommunications, did not initially influence the assignment of roles in respect of telephone complaints, especially outside Paris. 27

e10

B. Giry / Sociologie du travail 58 (2016) e1–e20

with some sixty keys”. In consequence, until at least the mid-1970s, “responses to complaints” were still “made, in the old centres, by ‘manual’ switchboards with sockets and jacks”.29 Over the period, the complaint-handling process nevertheless became dispersed as the organisation of the telephone service itself dispersed, with sharper distinctions between its “urban”, “interurban” and “international” activities. From the 1970s onwards, progress in the development of automatic switching made it possible to look afresh at the organisation of production, abandoning the categories inherited from the old manual switching method. It was only at this time, when the postal and telecommunication services established separate managements, that a significant effort was made to modernise and extend the network – the “telephone modernisation plan” – bringing profound changes in work and organisation. In order to fulfil the productivist imperatives of the “delta LP”,30 the service was reorganised to reflect a distinction between “technical” roles, relating to the development and maintenance of the network, grouped into “main operating centres”, and “commercial” roles, based in “commercial agencies”, relating to the management of contractual relations with the ever-growing numbers of subscribers. This division left complaints as something of a Cinderella and, in 1976, at a time when their numbers were increasing with the growth in the number of lines, they once again came under the spotlight. 2.1. The General Telecommunications Directorate (DGT) working group and the “Diderot experiment” In fact, the archives mention the establishment of a working group on complaints at the end of 1976. Appointed by CEO Gérard Théry, made up of senior civil servants from the corps of telecommunications engineers placed at the head of the administration to organise the “modernisation”, it was set up to resolve the “problems of the 13”, the dedicated number for “service malfunction” issues, which had teams in the main operating centres. The letter of engagement of 3 November 1976 in fact raises only one problem: complaints.31 At the time, the “13” was suffering from the absence of a single, explicitly identified national contact point for complaints. Complaints were supposed to be distributed between different departments, accessible by different phone numbers, on the basis of the technical competences specific to each department.32 The task of finding the right interlocutor was left to the subscriber. However, the working group’s letter of engagement noted that, “except in particular cases, it is difficult to ask [the subscriber] [...] to make subtle distinctions according to the nature of the complaint”. As a result, complaints tended to come to the “13”, since subscribers did not always clearly perceive the “subtle differences between complaints and service failure issues”.33

29

“Rapport de l’ingénieur en chef chargé de la D.O.T. Nord Est, 15 septembre 1976”, AN, fonds Thery, 19870355/25. “Delta LP” refers to the target figure for the increase (delta) in main lines (LP), set over the period of the “telephone modernisation plan” that ran from 1974 to 1985, covering the end of the 6th and the whole of the 7th ten-year plan. Norbert Alter and Christian Dubonnet describe the 1970s as being dominated by this productivist priority: for managers of the time, “only the ‘delta LP’ matters” (Alter and Dubonnet, 1994, p. 38). 31 “Lettre de mission de la DPR : problèmes du 13”, AN, fonds Thery, 19870355/25. 32 Automatic switching meant that it was no longer possible to formulate complaints in the course of the connection process. The subscriber was therefore required to follow a specific complaint procedure. The investigation conducted following this letter of engagement would reveal that, depending on the nature of their problem, complainants might be asked to dial 10, 12, 13, 15, 15.13.13 or 19.33.33. In addition, the instructions varied from one region to another. 33 An estimate by the DGT concluded that almost a third of activity on the “13” related not to service problems but to “miscellaneous complaints”. 30

B. Giry / Sociologie du travail 58 (2016) e1–e20

e11

Léon Enkaoua, a DGT chief engineer on the working group, was therefore asked to formulate “concrete proposals” in answer to the following two questions: starting from the principle that a “single complaints service” was necessary, “where should it be located?” and “what facilities should be available in the exchanges that receive these calls?” In addition, the service reform should facilitate the subscriber’s task by creating a single and visible complaints service, compatible with the delta LP production targets and the new organisational setup, in particular the distinction between technical functions and commercial functions. L. Enkaoua outlined the different possible options on the basis of experiments conducted in the regions. Poitiers proposed, for example, using the manual units on the “10” (manual interurban calls), where activity was declining. The idea was to “revive the role” of operators whose skills, with the arrival of automatic switching, were becoming obsolete. In this way, the technical skills of the manual operators, their capacity to replace the automatic switching system, would be exploited. In this model, however, which was based on an obsolete technology and organisation, complaints generally concerned a failed or lost connection, which could immediately be resolved by manual connection. At the time, however, automatic exchanges were responsible for only 20 % of reported service breakdowns.34 The proposal therefore reflected a dated conception of complaints and complaint-handling, and seemed tainted with an attitude of mistrust towards automatic switching. The working group ruled it out. The model finally chosen was the “Diderot experiment”, conducted in the exchange of that name in Paris. As in Poitiers, the experiment was consonant with the imperative of centralisation. However, the approach was slightly different: it consisted in bowing to the habits of subscribers by making the “13” a one-stop service for “miscellaneous complaints”, as described by the working group. This single entry point did not erase the division between technical and commercial: the job of the “13” would be to distinguish technical complaints (“loss of service”, “poor sound quality”, “lost connection”, etc.), which it would retain, from commercial complaints (“perceived overcharging”), which it would transfer to the agencies. 2.2. The complaint as a “working tool” of technical alert: technical faults This complaints service replicated the same hierarchical structure as the boards service, with female operators and supervisors, but it employed new equipment, “produced by a team that worked in concert with the staff of the Diderot manual exchange”, coming into operation in June 1976. It consisted of “small mobile operator desks, a supervisor’s desk, a service quality monitoring panel and two control boxes”, all “designed for a Main Operating Centre with capacity of 100,000 lines, with the highest rate of complaints observed in Paris at the present time”.35 Since the equipment could accommodate ten operators and one supervisor, if we extrapolate the service to the national level on the basis of the target number of main lines in the delta LP (i.e. 20 million lines), we obtain a projection of 2000 operators handling complaints, and 200 supervisors. The meaning of complaint had changed. With automatic switching and the development of the network, the managers were no longer looking for human misconduct but for technical malfunctions. The administrative disciplinary enquiry was replaced by the technical investigation. A 1981 article in a dedicated publication for DGT employees describes the role of the operator and the use of complaints as follows:

34 35

“‘13’ : aux petits soins du téléphone”, Messages n◦ 304, 1981, p. 7. “Ici, le service des réclamations”, Bulletin d’information des télécommunications d’Île-de-France n◦ 3, 1976, pp. 15-19.

e12

B. Giry / Sociologie du travail 58 (2016) e1–e20

“The primary role of the operators is, of course, to register the complaint. But also, to obtain as much information as possible about the service malfunction. Hence a whole battery of questions along the lines of: ‘Is the fault continuous or intermittent? How long has it been going on? Does it occur at busy times?’, etc. This information, which would subsequently help in identifying the fault more quickly, would be recorded by the operator on a special form. A form that is almost incomprehensible for the neophyte, so stuffed is it with mysterious acronyms that nevertheless mean very specific things for initiates. [...] From the moment the operator registers it, the complaint shifts from being something imprecise to a signal, a working tool” (“‘13’: aux petits soins du téléphone”, op. cit., p. 6, my emphasis). The technical investigation was distributed between three main players: the operators of the “13”, responsible for recording the complaint on a form called “Sext000”; the operators at “Tests and Measurements”, responsible for carrying out tests on the line and on the subscriber’s equipment, and recording the result on the Sext000 form; and finally the referral agent who, on the basis of information recorded on the Sext000, identified “the location of the fault [...] on the basis of the nature of the complaint, the result of the tests and measurements, the “past” history of the line, his knowledge of the network, as well as other indications”.36 Here, the only value of the handling of the complaint is as the first stage in a technical fix. It becomes a “working tool” that needs to be distributed. Its transcription into a “Sext000 form” incorporates it into a rigid graphic framework and thereby converts the complaint into a piece of technical information, which is then connected with prior information and integrated, in standardised form, into the “technical subscriber file” in which, to employ an expression from the journal Messages, “Sext000 forms end their careers”.37 Unlike the “statements” of the 1920s, which focused on mistakes by the human manual switchboard operators, the Sext000 considered only the failings of the three instruments involved in the production of automatic switching: the “automatic exchange”, “the line” and “the installation”. “Example: a subscriber is unable to get a ringtone. That may be caused by the telephone set, in particular a loose connection. It may be caused by a ‘break’ in the line. It may be caused by the exchange: perhaps some equipment has remained connected to the line” (“‘13’: aux petits soins du téléphone”, op. cit., p. 7). These instruments are never defective in themselves: overhead lines, for example, may be damaged by “bad weather”, by “a hunter’s shotgun pellets”, by “lorries and agricultural machinery”; underground lines may be cut by “mechanical diggers”. As operator of the network, the company externalises responsibility for malfunctions and can therefore assume the role of legitimate authority in dealing with complaints. The silence of the boards service has given way to the automatic announcement: “This is the Complaints Service, an operator will be with you shortly”.38 The subordination of complaint-handling practices to the necessities of managing the telephone network is present even in the methods of handling “commercial” complaints, primarily “billing disputes”, which can be legitimate only in so far as they indicate a technical problem.39 In the

36

“‘13’ : aux petits soins du téléphone”, op. cit., pp. 7-8. Ibid, p. 8. 38 “Ici, le service des réclamations”, op. cit., p. 17. 39 The DGT estimated that it received around 400,000 per year in the 1980s (“La contestation des taxes téléphoniques”, PTT info n◦ 588, 1987, p. 4). 37

B. Giry / Sociologie du travail 58 (2016) e1–e20

e13

years 1970-1980, the reimbursement of overcharges became a possibility, provided that the reality of this “overcharge” was confirmed by an “in-depth technical investigation”: “Tests on the line, checks for damage to the overhead lines or cables linking to the subscriber’s line, work on the main distribution frames, cables cut or under water, opening of a distribution box, checks on the call timing system, [...] line monitoring [...] For two weeks” (“La contestation des taxes téléphoniques”, op. cit., p. 4). The commercial agencies themselves were thus also subject to the hegemony of the “technical subscriber file” (Hochereau, 2014), the sole reference for the business. However, this is not the only factor behind the success of this conception of investigation. At the time, disciplinary practices were precluded by a series of social measures, in particular those that prevented subscribers from knowing the identity of operators. On 8 June 1982, a technical adviser to the Posts and Telecommunications Ministry wrote to the management of the Île-de-France network: “A user has drawn the attention of the Office to the behaviour of the operators in the telephone Directory Enquiries Service. Apparently, these employees are impolite to the point of rudeness, talking amongst themselves when they have a caller on the line. In addition, the supervisor of the service has apparently stated that she has no means to reprimand the operators concerned, and that recently – under pressure from the unions – it has become impossible to identify these employees, since they are no longer obliged to give their number when answering. Is there nothing that can be done? If true, the supervisor’s response seems to me even more serious than the behaviour of the operators” (“Lettre du 8 juin 1982”, AN, fonds F. Aron, cote: 19860381/9). The director of the Île-de-France network could only confirm his subordinate’s account: “the unions” had forced the removal of the “discreet listening posts located outside the operations rooms”. A division controller could still listen to an operator under certain conditions (“novice operator or operator whose work was unsatisfactory”) but, when he did so, “the operator was alerted by a warning light on her desk”. It was equally true that, in the DGT’s different services, the operators “no longer provide their position number” and answered calls with the words “hello, you are through to Télécom”. The incidents described by the Minister could therefore “be attributed to the fact that, to a certain extent, the operators feel protected by a degree of anonymity”. The director had himself conducted an enquiry which showed that “incidents of the kind mentioned in [the technical adviser’s letter] do occur”. He knew this because complaints of this type “are noted in an incident book at each Centre”. At this time, however, no further action could be taken.40 3. Complaints in the private company and the deregulated market (from 1990) At the beginning of the 1990s, DGT became a company. As a symbol, whereas the administration had only ever been headed by X-Télécoms (telecommunications engineering graduates), a graduate of the Higher School of Economics and Business was appointed to run the company in 1995. Along with the change in status, its business was redefined: it was “revealed” to itself as a “service enterprise” in an internal document haunted by the “spectre” of the forthcoming

40

“Réponse de la direction du réseau IDF au cabinet du ministre, le 8 septembre 1982”, AN, fonds F. Aron, 19860381/9.

e14

B. Giry / Sociologie du travail 58 (2016) e1–e20

market.41 It was thus part of a movement that other public companies and administrations were experiencing at the same time, prompting them to consider the question of their relations with their users and giving rise to a large body of research on the topic of the service relationship (Weller, 1998). 3.1. The “new legal and commercial framework” In the case of telecommunications, these relations were weakened by market liberalisation, which gave customers the option of choosing a different operator (deregulation was completed in 1996), but also by the act of 2 July 1990 which placed them under the common law regime, brought disputes under legal (rather than civil service) jurisdiction, and reversed the burden of proof to the detriment of the company, in particular in billing disputes. As a result, customer relations – under the combined effect of market liberalisation processes and the change in the legal status of the DGT – came under particular scrutiny: the essential was to avoid customer defections and lawsuits. In this respect, the redefinition of the company’s business appears to have been an attempt to redefine production targets, which were no longer to be assessed in terms of material production (such as “delta LP”), but in terms of a relational service (Gadrey, 1994) largely governed by the commercial imperatives of attracting and retaining a customer base that was no longer captive. In the documentation, the organisers refer to these upheavals with a litotes: the “new legal and commercial framework”.42 In order to pass the test of this new framework, each product was calibrated against the standard of “customer satisfaction”. This concept, imported from marketing studies and transposed into indicators in the organisation from the mid-1980s onwards, was associated with the quest for customer loyalty and harmonious client relations.43 Under this new regime, dissatisfaction became particularly problematic, especially when it was unexpressed and resulted in customers moving elsewhere or instituting legal procedures. The consequence of these changes would be to shift complaint-handling practices to the heart of the production sphere, or even into certain management activities, whereas they had previously been kept on the periphery, both in the technical domain – as ancillary activities – and in the accounting domain – as “loss-making services”. At this time, complaint-handling was moved out of the “13” services and assigned to “consumer departments” organised in “call centres”. The latter were formed in the midst of a period that would see the consecration of this organisational form as the essential place where “customer relations” were produced (Cousin, 2002; Russell, 2008). They were housed in multipurpose operational units, with both technical and commercial functions, set up during an organisational reform called “EO2”.44 A new profession emerged: “customer advisers” and “account managers” 41 Judging by the terminology of the “Cap 98” plan (drawn up in 1995), the use of the word revelation is not excessive: “Our business has always been, by nature, a service business: communication is not a physical product, and customers themselves play a big role in the satisfaction of their needs. But for a long time, this fact was masked by our dominant preoccupation with developing our networks. Our organisation shows the deep imprint of our highly technical culture” (p. 8). The word “spectre” highlights the fact that the market question is not tackled head-on in the documents. The reference to the market is implicit: it is both ubiquitous and tacit. 42 “Instruction du 21 janvier 1993”, AN, fonds ENSPTT, 20020187/34. 43 The system was called SATCLI. Other measures along these lines were outlined in “Plan du discours de M. Gérard Longuet, ministre des PTT, lors de la conférence de presse “consommateur”, le 2 mars 1988” (AN, fonds Longuet, 19900155/3). 44 EO2 (Evolution of the Organisation no. 2) followed EO1 (1993), which referenced the different jobs and skills with a view to a reorganisation implemented by EO2 (1995).

B. Giry / Sociologie du travail 58 (2016) e1–e20

e15

replaced “operators”. These new workers were responsible for building, managing and maintaining commercial relations through different activities subsumed under the category of “customer relations” (sales, account management, disputes, complaints, etc.). From the mid-1990s, all the documents defining the company’s development priorities included provisions for complaint-handling practices and for the different ways of using complaints. While the initial aim was simply to prepare for the “new legal framework” by producing complaint files that could “be used directly in the event of legal proceedings”,45 by the following year the prospect of the competitive market prompted the company to envisage “exploiting complaints” in order to “direct [the] efforts” needed with respect to “commercial quality”.46 Similarly, a few years later the “Cap 98” corporate plan announced the systematisation of “complaints analysis” in order to develop a form of “direct customer participation in the definition or testing [of] services”. 3.2. The devaluation of investigatory practices and the “assumption of trust”: the complaint as dissatisfaction In the complaints of the 1990s, the focus of managers was less on human error or technical failure than on dissatisfaction. The complaint was no longer legitimised by an administrative or technical substrate but simply acknowledged as expressing a dissatisfaction, and therefore a risk of defection or court action. In this sense, it was automatically legitimate. It even became desirable by comparison with other ways of settling conflicts. This new definition led to a rearrangement in the practical components of complaint-handling. They were no longer implemented within the framework of administrative or technical investigations, but were divided into practices of “defusing” and “analysis”, the former focusing on the tasks of explanation and compensation, the latter on prescription. The conception of the complaint as dissatisfaction raised entirely new practical problems. Take the example of compensation: in the 1920s, as in the 1970s, billing was an administrative act that was largely protected by the law. Reductions, reimbursements and, a fortiori, compensation payments were made difficult by the one-sided provisions of the subscription contracts. In cases where they occurred, they were conditional on extensive technical investigations. In the 1990s, however, the organisers realised that this kind of intransigence was expensive. According to one observer from the national directorate of consumer services at the time, the amounts at stake rarely exceeded 300 francs, whereas the fixed cost of a technical investigation was estimated at around 1500 francs: “it therefore seems reasonable to balance this cost against the sum in dispute”.47 The intransigence of investigation also proved commercially damaging. A survey conducted by SOFRES on behalf of the Telecommunications Regulation Agency found that customers calling to make a complaint wanted “a more conciliatory and less suspicious attitude” on the part of the historical operator or that they wanted the company to give “the customer the benefit of the doubt”.48 Reflecting these remarks, the instruction of 21 January 1993 on the “Prevention and handling of billing complaints” introduced an innovation: “It is important to assume that customers are in good faith, to listen to them and to give credence to their grievances or requests for explanation”. The priority was not, or not so much, to compare the customer’s version with an internal source 45 46 47 48

Executive Management, “Lettre d’orientation générale 1989-1990”. Executive Management, “Lettre d’orientation générale 1989-1990”. “Comment faire progresser FT avec les réclamations de ses clients”, 1995, AN, fonds ENSPTT, 20020187/34, p. 33. “The relations of the telecommunications operators with their ordinary customers”, SOFRES, 1999, pp. 20-24.

e16

B. Giry / Sociologie du travail 58 (2016) e1–e20

of information or technical factors. Instead, the instruction emphasised the importance of what it called “defusing”, in other words the response to the customer: “In these contacts, it is important to find a solution that fits the customer’s situation and to provide a reasoned response, while inviting him to think about it. If the customer is not convinced, he will confirm his complaint. For this reason, all ambiguities should be removed in this phase” (“Instruction du 21 janvier 1993”, AN, fonds ENSPTT, 20020187/34). For this purpose, management gave account managers “leeway” to compensate customers more easily by offering reimbursements and then, later on, by means of the first “commercial gestures”. Moreover, the asymmetry between the technical and commercial functions was reversed: customer advisers were to “refer cases for investigation” by an agent of the technical services when the complaint concerned a “technical problem”; but the handling of the complaint remained in the commercial sphere, entailing a verbal and possibly a monetary exchange. As a result, there was a shift in the value placed on different skills: the emphasis was less technical, more relational, with a particular focus on communication – communication intended first to convince, and therefore to “defuse”, and then to convey the sense of “authority”. A new professional toolkit developed, made up in particular of booklets of stock paragraphs to be used in constructing arguments in letters of response. These booklets also gave advice on how to formulate arguments, suggested different methods of self-presentation, depending on the authority that the customer manager might be required to portray: “If I (customer) address the Chairman or a director by name to express my dissatisfaction, I will understand if he personally replies that he is “sorry” or “deeply regrets” what has happened to me through the fault of his company, a company to whose constant improvement – as he misses no media opportunity to assert – he is fully committed. But if I address myself to an entity such as the [consumer department], which I do not even know to be a department [of the company], I will be surprised and probably irritated if an interlocutor whose status is unknown to me expresses his or her apology or understanding: it is not the role of an arbitrator to deplore the attitude or failings of one of the parties, but to arbitrate” (Extract from an archive of the consumer department, private collection). Another sign of the importance of communication was the annual competitions held in the mid1990s to reward “the best written reply to a customer complaint”, or the existence of a lexicon of “no-no words” to be avoided in written or spoken exchanges with customers, which notably included the word “complaint”. So the skill of the customer adviser lay in his or her eloquence and “writing style”. It was a function that the organisers would continuously update. The instructions regarding apologies, for example, would evolve: initially, apology was forbidden, because of its legal implications – an acknowledgement of fault that might be cited in evidence against the company – and then, after 2010, encouraged on the grounds of its positive commercial impact. 3.3. Analysis practices and their incorporation into a management tool: prescription This authority was thus no longer embodied in punishment.49 By the end of the 1980s, there is less and less reference in the documents to the theme of “responsibility” in relation to complaints and, where it is present, it is the collective nature of the responsibility that is emphasised. 49 Which does not, on the other hand, imply the disappearance of the topic of “professional misconduct” (Chateauraynaud, 1991).

B. Giry / Sociologie du travail 58 (2016) e1–e20

e17

Social control has become more “incremental” less “heroic”.50 This factor is sometimes expressed positively by a call for professional “solidarity”: “Customers see us all as responsible for quality. Let us work together to satisfy them” (“Projet d’entreprise des télécoms”, DGT, 1986, p. 29). Nevertheless, by the 1990s, this exhortation is already overtaken by new methods of management control and employment founded on the competences model (Zarifian, 2001) and based on the use of quantitative indicators for monitoring production. The organisers would initially attempt to graft complaints handling onto this new technology of organisation, by making it a “cogent indicator” (Boussard, 2001). On 15 March 1993, the executive management launched the construction of ARTIC, a computer application designed to enable “the implementation of accounting and operations for complaints in all regions”. The document produced for the occasion by the “customer data exploitation” section, housed within the equally evocatively named “analysis, feedback and follow-up” division, emphasises the dual nature of the complaint: “Every letter of complaint received by [the company] contains firstly a complaint that has to be handled (investigation, reply), and secondly raw information that [the company] needs to exploit in order to improve its service” (“ARTIC”, AN, fonds ENSPTT, 20020187/34). The extraction of “raw information” entails a process of translation, since the complaint “cannot be exploited as it stands”: “In order [for it to be] exploitable, it needs to be reformulated to remove stylistic effects, aggression, to provide a clear, easily reused and relatively standardised synthesis, [and to] be codified for recording in the database of reasons for complaint” (“ARTIC”, AN, fonds ENSPTT, 20020187/34). Complaint handlers are therefore invited to “track” complaints in the information system. This is because, once processed, the complaint provides “short-term inputs for the management of the hubs, the Regional Divisions and the Sales Division [...]; and medium and long-term elements for the Sales Division and Central Management, in drawing up the group’s future national strategy”. The first indication related to numbers of complaints. In the early 1990s, the company “estimates that it receives [...] between 100,000 and 500,000 complaints [per year], but does not know the exact number”. However, the organisers were above all interested in the “reasons” for customer dissatisfaction: at a time when the commercial viability of its products was no longer guaranteed by monopoly status, “it is becoming more and more necessary to use them accurately”. Complaints are utilised to assess the quality of the different products and services, as a basis for prescription.51 Prescription was to take two forms in the company. The first was what the organisers called the “customer’s cry” (Zerbib, 2001). The practice was to present a complaint to the company’s executive committee that was perceived as particularly significant. However, it had no implications as to what was to be done about this “cry”. While this practice had the virtue of edification, it was not prescriptive in the strict sense. 50

On the subject of incremental and heroic industrial control, see Coleman, 1990. Complaints form part of quality processes and international standards encourage their prescriptive use. ISO 9000 includes the handling and analysis of complaints. In 2004, another standard was introduced (ISO 10002:2004), and updated in 2014, dedicated to complaint-handling. These two standards emphasise the prescriptive dimension of complaints. 51

e18

B. Giry / Sociologie du travail 58 (2016) e1–e20

During the 2000s, complaints would give rise to another practice, founded on an accounting exercise. Complaints tracked by the customer advisers were extracted from the information system on a monthly basis and subjected to a so-called “customer focus” analysis, designed to link each complaint to a “root cause”. Each root cause in reality referred to a “process”, in other words a fragment of the organisation endowed with relative autonomy and responsible for carrying out part of the business – for example the “order-delivery” process. From the end of the 2000s, those responsible for “customer focus” would be authorised to allocate the costs associated with handling complaints – the costs of the labour and the capital as well as those linked with compensation – to the balance sheet of the processes deemed responsible for those complaints. Complaint analysis thus became part of “management control”, i.e. of a “process by which managers influence other members of the organisation to implement the organisation’s strategies” (Anthony, 1988, p. 10). Through this shift, the costs of complaint-handling become the costs of complaints, borne by the processes deemed to be “at fault”. Theoretically, therefore, the complaints process was no longer a debit item in the balance sheet. By a strange reversal, the cost of complaints, previously a sign of their non-productive character, has today become a driver of their influence in the organisation. 4. Conclusion It was in the 1920s that professional practices first emerged around “complaint” in the telephone industries. Initially, the aim was to keep it away from the workers on the grounds that it was “unnecessary speech”. Then, relocated to a dedicated space for “handling”, the complaint came to refer to a distinct aspect of the interchanges between the user and the professional, which was classified neither with interchanges immediately useful to the production of the telephone service (announcement of the number to be called, information on the state of the service: “number engaged”, etc.), nor with the unnecessary chatter condemned by early twentieth-century engineers. The complaint gradually acquired a certain legitimacy: bit by bit, it was granted the right to handling and efforts were made to integrate the activities associated with it into the space of professional practices in uncluttered telephone exchanges dedicated to the rationalisation of tools, minds and bodies. In the 1920s, the different components of complaint-handling practices – punishment, repair, explanation, compensation, prescription – were laid down. They would be arranged in different ways according to what the organisers judged it appropriate to see in a complaint: human misconduct, technical failure, dissatisfaction. In a first configuration (1920), in which production of the service was dependent almost exclusively on the work of the switchboard operators, the handling of complaints consisted mainly in working on “misconduct”: investigating, identifying a culprit, applying a sanction. In a second configuration (1970), implemented to support the deployment of the line network and the development of automatic switching in the decade of “telephone modernisation”, it constituted the first moment in a process of technical support. Finally, in the third configuration (1990), in the midst of a period marked by the opening up of the market and the gradual privatisation of the company, it consisted in avoiding court proceedings and customer defections. Depending on the organisational configuration they encounter, complaints can – from a different perspective – be the source of varying sequences of relations. Those relations, disciplinary, technical or otherwise, constitute the “outcomes” of complaints. They contribute to the formation of the sociological reality of the productive and commercial organisation, in its attempt to bind the relations between workers and customers into a stable network.

B. Giry / Sociologie du travail 58 (2016) e1–e20

e19

In this article, we have focused on the complaint as an infra-legal method of characterising and managing commercial disputes. However, such a study could form part of a more general programme of research that would explore the influence of conflicts with customers, in their different forms, on changes in work and its organisation. This would bring an additional perspective to research into workplace conflicts. Acknowledgements This article originated in a doctoral thesis funded by CIFRE EDF R&D. The author would like to thank the members of the jury of the 2014 Young Author Prize and of the editorial committee of Sociologie du travail, together with Jeremy Bouillet, Olivier Cousin, Jean-Luc Metzger, Pierre Naves, Amélie Petit and Alina Surubaru, for their comments on earlier versions of this text. References Abel, R.L., Felstiner, W.L.F., Sarat, A., 1981. The Emergence and Transformation of Disputes: Naming, Blaming, Claiming. Law & Society Review 15 (3/4), 631–654. Alter, N., Dubonnet, C., 1994. Le manager et le sociologue. L’Harmattan, Paris. Anthony, R.N., 1988. The Management Control Function. Harvard Business School Press, Boston. Atten, M., 2013. Les ingénieurs des télécommunications (1844-1999) : un grand corps ? In: Griset, P. (Ed.), Les ingénieurs des Télécommunications dans la France contemporaine. Réseaux, innovation et territoires (xixe -xxe siècles). Institut de la Gestion Publique et du Développement Économique, Comité pour l’histoire économique et financière de la France, Paris, pp. 17–32. Barrey, S., 2002. Les grimaces du client. Des figures du consumérisme aux figures du consommateur “écrivain”. Sciences de la société 56, 167–184. Bertho, C. (Ed.), 1984a. Histoire des télécommunications en France. Érès, Paris. Bertho, C., 1984b. Les réseaux téléphoniques de Paris–1879-1927. Réseaux 4, 25–53. Bidet, A., 2005. La mesure du travail téléphonique. Le cas des opératrices (1910-1938). Histoire et mesure 20 (3–4), 15–48. Boltanski, L., Thévenot, L., 1991. De la justification. Les économies de la grandeur. Gallimard, Paris. Borzeix, A. (Ed.), 1986. Retour sur l’entreprise. Sociologie du travail 28 (3), 231–386 [special issue]. Borzeix, A., Fischer, S., de Fornel, M., Lacoste, M., 1995. Les lettres de réclamations. In: Quin, C. (Ed.), L’administration de l’équipement et ses usagers. La documentation franc¸aise, pp. 71–107. Boussard, V., 2001. Quand les règles s’incarnent. L’exemple des indicateurs prégnants. Sociologie du travail 43 (4), 533–551. Campana, M., Jaubert, J., 1976. La demoiselle du téléphone. Delarge, Paris. Chateauraynaud, F., 1991. La faute professionnelle. Une sociologie des conflits de responsabilité. Métailié, Paris. Cochoy, F., Dubuisson-Quellier, S. (Eds.), 2000. Les professionnels du marché. Sociologie du travail 42 (3), 359–504 [special issue]. Cohen, Y., 2001. Organiser à l’aube du taylorisme. La pratique d’Ernest Mattern chez Peugeot, 1906-1919. Presses universitaires Franc-comtoises, Besanc¸on. Coleman, J., 1990. Foundations of Social Theory. Harvard University Press, Cambridge. Cousin, O., 2002. Les ambivalences du travail. Les salariés peu qualifiés dans les centres d’appels. Sociologie du travail 44 (4), 499–520. Day, R.L., Landon Jr., E.L., 1977. Towards a Theory of Consumer Complaining Behavior. In: Woodside, A.G., Sheth, J.N., Bennett, P.D. (Eds.), Consumer and Industrial Buying Behavior. North-Holland Publishing Company, New York, pp. 425–437. DiMaggio, P. (Ed.), 2001. The Twenty-First Century Firm: Changing Economic Organisation in International Perspective. Princeton University Press, Princeton. Dubois, P., 1968. Le traitement de la réclamation dans l’industrie textile. Sociologie du travail 10 (4), 393–406. Dujarier, M.-A., 2008. Le travail du consommateur. De McDo à Ebay : comment nous coproduisons ce que nous achetons. La Découverte, Paris. Favreau, B., 1996. Georges Mandel ou la passion de la république, 1885-1944. Fayard, Paris.

e20

B. Giry / Sociologie du travail 58 (2016) e1–e20

Fijalkow, Y., 2006. Usagers ou consommateurs ? France Télécom ou la dérégulation du service public. Presses universitaires du Mirail, Toulouse. Foli, O., 2008. Plaintes, normes et intégration. Le cas d’une organisation bureaucratique (Thèse de l’université Paris Dauphine) Available online: http://basepub.dauphine.fr/bitstream/handle/123456789/192/THESE Foli.pdf. Fornell, C., Wernerfelt, B., 1988. A Model for Customer Complaint Management. Marketing Science 7 (3), 287–298. Gadrey, J., 1994. Les relations de service et l’analyse du travail des agents. Sociologie du travail 36 (3), 380–389. Ganem, V., 2011. Seul le travail rentable est évalué. Travailler 25, 129–143. Hirschman, A.O., 1970. Exit, Voice, Loyalty: Response to Decline in Firms, Organisations and States. Harvard University Press, Cambridge. Hochereau, F., 2014. La création d’une entreprise de service. In: Kessous, E., Mallard, A. (Eds.), La fabrique de la vente. Presses des Mines, Paris, pp. 191–219. Israël, L., 2010. L’usage des archives en sociologie. In: Paugam, S. (Ed.), L’enquête sociologique. Presses universitaires de France, Paris, pp. 167–185. Join-Lambert, O., 2001. Le receveur des Postes, entre l’État et l’usager (1944-1973). Belin, Paris. Joseph, I., 1988. La relation de service. Les interactions entre agents et voyageurs. Les annales de la recherche urbaine 39, 43–55. Kessous, E., Mallard, A. (Eds.), 2014. La fabrique de la vente. Presses des Mines, Paris. Krishna, A., Dangayash, G.S., Jain, R., 2011. Service Recovery: Literature Review and Research Issues. Journal of Service Science Research 3 (1), 71–121. Littlefield, N.O., Ross, H.L., 1978. Complaint as a Problem Solving Mechanism. Law and Society Review 12 (2), 199–216. Londres, A., 1930. Le juif errant est arrivé. Arléa, Paris. Maugeri, S. (Ed.), 2006. Au nom du client. Management néo-libéral et dispositif de gestion. L’Harmattan, Paris. Pilloned, A., 1925. Le service des dérangements dans les téléphones suisses (Extrait du Bulletin Technique des Télégraphes et Téléphones Suisses). Revue des Téléphones, Télégraphes et TSF 28, 813–834. Pinto, L., 1989. Du “pépin” au litige de consommation. Une étude du sens juridique ordinaire. Actes de la recherche en sciences sociales 76-77, 65–81. Proust, M., 1920. Le côté de Guermantes. Gallimard, Paris. Rougier, E., 1927. Le vade-mecum de la téléphoniste. Publications de l’indicateur universel des PTT, Paris. Russell, B., 2008. Call Centers: A Decade of Research. International Journal of Management Review 10 (3), 195–219. Tiffon, G., 2013. La mise au travail des clients. Economica, Paris. Trépos, J.-Y., 1988. La construction sociale des conflits de consommation. Thèse d’État soutenue à l’Université de Lille 1. Trépos, J.-Y., 1991. Contraintes et coûts d’investissement dans les réclamations écrites. Revue franc¸aise de sociologie 32 (1), 103–127. Ughetto, P., 2011. Gardien d’immeuble, sentir et ressentir. Communications 89, 89–101. Ughetto, P., 2013. L’enjeu paradigmatique de la relation de service au sein de la sociologie du travail. La nouvelle revue du travail 2 [online: http://nrt.revues.org/726]. Weller, J.-M., 1998. La modernisation des services publics par l’usager: une revue de littérature (1986-1996). Sociologie du travail 40 (3), 365–392. Williamson, O., 1975. Markets and Hierarchies. Analysis and Antitrust Implications. Free Press, New York. Wormser, G., 1975. Témoignage. In: Antoine, M., Barral, P., Favier, J., Guiral, P., Legendre, P., Michaud, H., Thuillier, G., Tudesq, A.-J., Tulard, J., Vidalenc, J. (Eds.), Origines et histoire des Cabinets des ministres en France. Actes du colloque consacré à l’histoire des cabinets ministériels tenu aux Sciences politiques. Droz, Genève, pp. 83–92. Zarifian, P., 2001. Le modèle de la compétence. Trajectoire historique, enjeux actuels et propositions. Liaison, Paris. Zerbib, D., 2001. Service National Consommateur. L’ambassadeur de la relation client. Fréquences Télécoms 156, 10–11.