The language of integrative collective bargaining

Abstract

Although collective bargaining is essentially a communication process, the role of language (as distinguished from discourse) in bargaining exchanges has received little attention from industrial relations scholars. Building on the work of Karl Popper, this article proposes a decomposition of language into functions and values and analyses their relevance when parties to a collective bargaining encounter engage in an integrative process. The proposed framework provides labour negotiators seeking integrative outcomes with linguistic guidelines and scholars with a tool to analyse bargaining exchanges.

Introduction

Human affairs generally and employment relationships specifically occur in contexts of indeterminacy (Baldamus, 1961; Edwards, 1995; Smith, 2006). This is the case because no collective or individual agreement, no employment contract and no job description can determine precisely or conclusively how much effort is to be exerted in exchange for given wages or working conditions. Moreover, convincing evidence, compelling moral evaluations and logical demonstrations (i.e. propositions that have an overwhelming suasory power) are rarely available in workplace situations. When making claims, actors must therefore persuade their audience of the validity of their positions. In this sense, the employment relationship is essentially a linguistic arena (Hamilton, 2001).

Within industrial relations settings, persuasion is particularly salient during bargaining exchanges. As Martin (1992) noted, ‘collective bargaining is a process of persuasion, [in which each party attempts] to persuade the other side to improve its offer or to reduce its demands’ (p. 101). At some point in the process, one side delivers an assessment of the current state-of-affairs and advances propositions to amend it, supported by some form of reasoning calling on evidence, moral justification and application of the principles of logic. To this assessment and proposal, the other party responds and advances its own views. Arguments are thus weighed against each other and such exchanges normally resolve in agreements, formal or informal. These agreements are typically not definitive, though, for situations keep evolving and arrangements need to be revised at regular intervals by the parties to the employment relationship. In any case, the endeavour is less about closure (which remains temporary) than it is about opening, response and counterproposal. Besides, unsuccessful arguments rarely disappear; they often return, in one shape or another, in future disputes (Kirkbride and Durcan, 1987: 7).

Several scholars accept the central role of language in industrial relations (Blyton and Turnbull, 2004; Boden, 1994; Drew and Sorjonen, 1997). In existing research, however, when not referring to tongues or the ability to speak them (Almond and Connolly, 2020; Manzella, 2015; Ubalde and Alarcón, 2020), language is generally understood in an Orwellian sense. Indeed, as Hamilton (2001: 433–435) observed, for most industrial relations authors, language is synonymous with attempts to frame the context of discussions, manipulate opinion and structure perception of reality. For instance, Eaton and Kriesky (2001) analysed how managements used allegedly neutral language to control union organising and card-check agreements. Similarly, Kirkbride (1986) studied how the general manager of a medium-sized company employed seemingly common language to control discussions with worker representatives. In other words, industrial relations scholars have generally understood language as discourse, that is, as the way cognition (beliefs, ideas, etc.) is communicated, taking account of the location of communication, who speaks, why and when (van Dijk, 1997: 2). In so doing, these researchers have neglected the use of language (words and phraseology, what they achieve and what they reveal of speakers) as an object of analytic interest.

The identification of language with discourse and its neglect as a distinct focal construct are particularly visible in studies of collective bargaining exchanges.¹ Indeed, in these investigations, researchers have typically examined the way protagonists frame proceedings, before they take place, or during them. For instance, Wodak and van Leeuwen (2002) investigated the speeches of Austrian Chancellor Viktor Klima when he exposed his youth unemployment policies. Cooper and Ellem (2008) analysed the communications of the Australian government when it launched its employment relationship reforms (which, among other outcomes, reshaped the collective bargaining process). In the same vein, Cutcher-Gershenfeld and Isaac (2018) investigated the way legislative texts have shaped labour relations. Furthermore, while researchers have proposed comparative discursive analyses of integrative and distributive negotiations (Donohue and Roberto, 1996; Lillie, 2004; Olekalns et al., 1996; Rodríguez et al., 2016), linguistic analyses of such processes are rare. A notable exception to this observation is the study of Taylor and Thomas (2008), which found that successful negotiations (those resolving in settlements agreed by both parties) are associated with higher levels of linguistic style matching (word use coordination) by negotiators than unsuccessful negotiations.

Aristotle (1994) argued that suasory effectiveness has three intertwined elements: logic, pathos (what hearers expect) and ethos (speaker-based). He also insisted that persuasion demanded style, analogies and metaphors, which, on his view, are more persuasive than literal expressions – an insight which has found empirical support (Sopory and Dillard, 2002). Later authors have not deviated markedly from such recommendations, adjusting only their illustrious predecessor’s analysis to modern circumstances. For instance, Atkinson’s (1984: xvi) study of the ways in which politicians’ speeches arouse audiences is, on the author’s own admission, a simplification and adaptation of Greek (i.e. mostly Aristotelian) techniques. Similarly, Burke (1969) did not diverge from Aristotelian guidelines when he argued that hearers should identify with speakers if they want to persuade them. Despite their enduring merits and applicability, however, these suasory techniques do not include linguistic considerations. Specifically, they gloss over the fact that language can be decomposed into functions and what their use reveals of speakers’ and listeners’ intentions.

This article contributes to industrial relations scholarship by proposing a linguistic framework applicable to collective bargaining exchanges. Specifically, after a primer on integrative bargaining, the article presents a framework which decomposes language into functions and related values and analyses their contribution to the achievement of integrative outcomes. Such outcomes, it is argued, require that bargaining parties use effective signalling, true descriptions, valid arguments and authoritative advice about points at issue, rather than attempt to control their counterpart through authoritarian communications. In addition to its practical value for labour negotiators, this article’s linguistic framework opens avenues for further research.

Integrative collective bargaining

As opposed to distributive negotiations which are constant-sum games, integrative exchanges are discussions in which parties seek to cooperate by reconciling their interests to some degree, thereby increasing the total value of the sharing upon which they settle (Barrmeyer, 2017). Operating under adversarial preconceptions, people often do not look for integrative outcomes during negotiations because they believe such arrangements do not exist (Basadur et al., 2008; Thompson and Hastie, 1990). On their view, the pool of value is fixed and therefore one party’s gain is necessarily another party’s loss. In such distributive scenarios, while persuasion and agreement are desirable, they are not required when one party has means to impose its will unilaterally. However, negotiations have rarely, if ever, strictly binary outcomes; rather, settlements typically include degrees of integration and distribution (Barrmeyer, 2017; Putnam and Wilson, 1989). That is, in most cases, parties can achieve a better outcome than if they had simply exchanged compromises.

Within industrial relations settings, employers and employees seldom value the same point at issue in the same way. Such differences in valuation create mutually beneficial opportunities and thus integrative bargaining outcomes. For instance, discussions about the introduction of teamwork to improve productivity are opportunities to reduce workload or upskill employees. Even wage disputes can be turned into opportunities to reach agreement on flexitime or job security (Weiss and Schmidt, 2008). Alternatively, matters not initially included in the discussion can be brought into it to enlarge the field in which reciprocated concessions can be found. Discovering where such opportunities lie and how to obtain them is that with which integrative bargaining is concerned.

Integrative bargaining requires that parties communicate collaboratively. Indeed, the likelihood that parties discover, agree on and reciprocate beneficial opportunities increases when they reveal their interests, beliefs and values and seek options through information sharing (Barrmeyer, 2017). Furthermore, since integrative bargaining requires each party to balance its own interests with those of the other, integrative outcomes are typically discovered through proposals and counterproposals, argument and counterargument. Persistence is also required since exposing one’s interests, understanding those of others and creating mutually acceptable solutions takes more time and effort than simply trading concessions without revealing their costs.

Industrial relations scholars have tended to consider labour negotiations as a fundamentally adversarial exercise (Provis, 2000). Indeed, although they have generally recognised cooperation as a desirable objective, most researchers have assumed a fundamentally conflictual perspective on the employment relationship (Bray et al., 2020). On this view, integrative outcomes appear unrealisable. In fact, if they are to be achieved, some authors (e.g. Friedman and Shapiro, 1995; Fulmer et al., 2009) recommend resorting to bluffing, concealment, distortion and deception in labour negotiations.

The use of deceptive tactics when seeking achieves integrative outcomes is vulnerable to two rejoinders. First, any settlement reached by way of manipulative techniques cannot be called integrative. Except in those rare cases where they benefit people against their will, agreements reached through deception are exploitive in nature, even if the deceived party does not immediately realise that this is so. As such, these outcomes forego the additional value that could have been generated by way of a genuinely integrative process (Townsend and Loudoun, 2016). Second, even in adversarial situations, ad hoc mutually beneficial agreements exist that improve on simple ‘I win, you lose’ outcomes. Indeed, irrespective of whether the work environment is ‘high trust’ or ‘low trust’ (Fox, 1974), parties to industrial disputes, as a matter of practical necessity, need to find at least temporary agreement on day-to-day basic issues. This conclusion holds whichever perspective on the employment relationship one adopts, be it radicalism, adversarial pluralism, collaborative pluralism, consultative unitarism, autocratic unitarism and egoism (Bray et al., 2020: 128).

More generally, when employers and employees enter collective bargaining negotiations, they discuss rules and policies through which organisational goals are to be met. Each side appoints representatives to participate in this process and make decisions on its behalf. Such a delegation of decision-making power reduces the individual freedom of the delegates, because employers and employees alike will remonstrate and ultimately remove their mandate if they come to believe that their representatives no longer act in what they perceive to be their best interests. Delegates on either side thus retain their mandate for as long as they can secure agreements deemed to be advantageous by those they represent. Integrative settlements belong to this category and strengthen the mandate of those who obtain them. The stability of the work organisation, as well as the general impression among its members that it is in capable hands, is also enhanced thereby (Townsend and Loudoun, 2016). Such consequences explain, in part, why integrative bargaining outcomes are superior, in terms of staff morale and workplace performance, to distributive ones (Deery and Iverson, 2005). Why and how these improvements are achieved become clearer when analysing the functions of language and their related values.

Language functions and values

The main function of language – communication – can be decomposed into (sub)functions. Building on the work of German linguist Bühler (2011 [1934]), Popper (1989 [1963]: 134ff) analysed language into the following four functions: expressive, signalling, descriptive and argumentative. The expressive function serves to convey the inner states of the speaker, such as emotions and feelings. The signal function aims to elicit reactions from others (for instance, linguistic responses). The descriptive function involves statements about how a state-of-affairs is perceived by the speaker. Finally, the argumentative function concerns the presentation of reasons in connection with certain questions or problems. According to Popper (1972: 120–121), the powers of critical argument are the powers of reasoning. They are, therefore, the basis of collective development, if only in the form of a bargaining outcome.

Popper, like Bühler before him, held that language functions form a hierarchy in which each presupposes or implies its predecessor but not its successor. For instance, description requires signalling (whether implicitly or explicitly, descriptions rely on sense data) but not argument. However, argument implies expression since it conveys a belief; it is a signal, since it calls for a response (agreement or rebuttal); finally, argument is also description since it is advanced with regard to a specific situation.

Popper (1989: 135) further argued that speakers activate language functions differently, according to antithetical values. Specifically, self-expression is either revealing or concealing, depending on whether it reveals the genuine emotions or feelings of the speaker. Signalling is either effective or ineffective when speakers confuse their hearers. While descriptions are either true or false, arguments are either valid or invalid.

Spillane (1987) completed Popper’s hierarchy with the advisory function of language. Since advising presupposes the giving of reasons to justify, confirm, refute or seek new recommendations, it follows that the advisory function entails the argumentative function: in giving advice, speakers argue. When there are good reasons for accepting and following advice, such communication is authoritative; conversely, when a ‘recommendation’ is imposed upon a listener, this communication (which is in fact an order) is better qualified as authoritarian. Table 1 summarises the proposed hierarchy of language functions and details their content and related values.

Table 1. A hierarchy of language and values; adapted from Spillane (1987).

In a collective bargaining exchange, protagonists instantiate language functions in one of their two possible antithetical values according to their preferences. These preferences are not necessarily consistent. For example, a newly appointed human resource manager might declare ‘There is no legal obligation to provide more than three shift breaks’ (true descriptive statement); to this declaration, a union representative may respond disingenuously, trying to take advantage of her interlocutor’s recent arrival in the organisation ‘A fourth break has been company policy for the last ten years’ (false descriptive statement).

Expressing – the language of feelings

The expressive function of language is at the bottom of Table 1 because it is the most primitive and least regulated form of language (it is also a form that does not require the presence of other people). As mentioned, expressive statements are either revealing (sincere) or concealing (insincere), although listeners cannot ascertain this status. The reputation of theatre and movie actors depends to a large degree on their ability to make confected expressive statements (such as ‘I am upset’) sound sincere. However, should a general uncertainty develop about expressive language beyond the borders of the acting world, questions of descriptive truth (and falsity) become intractable and arguments about material conditions impossible.

The last five decades have been marked by a growing emphasis on emotive language and a corresponding decline of argument (Bloom, 1987; Joullié and Spillane, 2020: 285ff; Palmer and Hardy, 2000). By starting their propositions with ‘I feel’, ‘it seems to me’ and other expressions of comparable meaning, postmodern speakers make their utterances invulnerable to criticism since it is impossible to argue about another person’s feelings. In the context of collective bargaining, when speakers elevate feelings to the status of facts (i.e. treat them as descriptions), they not only place their claims on fragile grounds, but also expose themselves to the charges of political incorrectness and hypocrisy. Political incorrectness, because statements about personal feelings are, in varying degrees, likely to upset other people; hypocrisy, because expressions of feelings cannot be judged sincere or insincere. When one interlocutor advances an expressive statement, the other is thereby placed in the unenviable position of a referee having to decide about the status of a move without access to a rulebook. A polite refusal to consider the merits of the statement is a popular response for those with a penchant for rational argumentation (Popper, 1989).

The subjective nature of expressive statements and the impossibility to decide whether they are sincere or insincere acquire special relevance when such communications are employed in bargaining discussions. For instance, because there is no observable entity or force called ‘psychological stress’ or ‘job dissatisfaction’, managers and union representatives have no choice but to rely on communications (spontaneous or collected by way of interviews and questionnaires) from employees to evaluate stressful or unpleasant work conditions. Moreover, while stressors typically originate from the external environment, they cannot be divorced from the concerned individuals’ perception of them (Karasek, 1979; Lazarus, 2006). Indeed, while noise is stressful for most people (and is harmful at extreme levels), for some, it is a source of stimulation (modern ‘music’). Employer representatives can therefore reject expressive statements like stress reports as resulting from employees’ psychological maladaptation or incompetence rather than pertaining to work (or the workplace) itself. Similar comments apply to Repetitive Strain Injury (RSI) or mental health claims: in the absence of identifiable physical lesions or trauma - lacking in RSI cases (Spillane, 2017; Spillane and Deves, 1987), ruled out by definition for mental health problems (Szasz, 2004) - such claims are vulnerable to out of hand dismissal by the party against which they are directed (DeFrank and Ivancevich, 1998: 61).

Signalling – the language of effectiveness

According to Popper (1989: 134), signalling takes place whenever an expressive statement of one individual operates upon another as a stimulus and triggers a response from the other individual. Signalling can therefore be voluntary or involuntary. For example, a cry of despair may induce despair in other people. Not all expressions are so ‘contagious’, however, since expressions of fear can encourage aggressors, while signs of courage typically discourage them. Humans share expressive and signalling language with non-human animals.

Since the two lower functions are always present when the higher ones are present, it is always possible to analyse linguistic phenomena in terms of the lower functions, that is, as expressions or signals. Popper (1972: 120ff) believes this approach is disastrous. In his view, when communication fixates at the level of feelings and signals (or stimuli), the higher functions of language are not activated. Consequently, critical evaluation and reasoning cannot take place and psychological (individual and collective) development becomes impossible. That is, when speakers do not go beyond expression and signalling, they are bound to view the problems they are trying to address as resulting from ‘poor communication’, a universal (but superficial) pseudo-explanation which does not require speakers to engage in the higher and distinctively human aspects of linguistic interaction. Moreover, by failing to develop arguments, speakers are likely to address so-called ‘communication problems’ by way of orders: ‘My way, or else’. This primitive form of language, which only seeks obedience, does not achieve cooperation and therefore does not belong to an integrative exchange.

Describing – the language of truth

Descriptive language conveys statements about the world. Such statements are required for collective bargaining exchanges to commence because parties first need to agree about the nature of the issues they seek to address. More generally, the quality of a bargaining relationship has been defined as the extent to which parties are able to describe to each other the constraints under which they operate (Brown, 2009: 435). Inevitably, however, disagreement about descriptions arise when parties misrepresent facts (willingly or not). While a divergence of descriptions strains bargaining discussions, it also represents an opportunity to consolidate them provided that the parties agree on a method to resolve their differences. When carried out successfully, such remedial measures dissolve misunderstandings, clarify the issues at hand and build trust. Since integrative exchanges require cooperation about points at issue and since cooperation (as opposed to manipulative control) cannot be achieved through lies, the bargaining parties’ preference for truth or falsity determines whether the proceedings are integrative or distributive.

In bargaining discussions, the issue of truth (or falsity) of descriptions is often difficult to settle. This difficulty arises in part from the metaphorical nature of human language generally and the language of bargaining exchanges specifically. As noted, metaphorical statements possess more suasory power than literal statements; since their intention is to persuade, negotiating parties tend to use them liberally. It is, therefore, unsurprising that the language of negotiations is replete with metaphors, typically derived from the military, gaming and sports arenas (Cohen, 2003; Watson, 2004). Hence, Gordian knots need to be cut and kites flown, while cards that are not kept close to one’s chest are at risk of being shot down by negotiators playing hardball or taking the role of the Devil’s advocate.

Notwithstanding their suasory effectiveness, metaphorical statements are figures of speech that are not literally true. They are therefore false. Consequently, when metaphors multiply, discussions become confused because speakers no longer mean what they say or say what they mean (Palmer and Dunford, 1996). In terms of the hierarchy presented in Table 1, when confronted with metaphors or ambiguous descriptions, bargaining parties either progress to argument or revert to signalling and expressive language. In the current (postmodern) times of ‘post-truth’, they are likely to regress. Ambiguous language and metaphorical statements have at least one merit; however, they point to the value and necessity of arguing about descriptions. Cross-critical evaluation of descriptions (the Socratic method) transforms acrimonious exchanges into more mature conversations in which speakers engage in constructive argument. At this stage of a discussion, the higher levels of the linguistic hierarchy come into play.

Arguing – the language of cooperation

Contrary to postmodern belief, arguing is the language of cooperation because arguments enable people to expose justifications and to explain why their opinion and recommendations are valid (Cattani, 2008). Critical arguments clarify problems, expose misconceptions and reveal lies. They also unmask dogmatists and extremists and those who try to present contentious situations as one-sided affairs. Furthermore, argument supports integrative outcomes (Putnam and Wilson, 1989; Roloff et al., 1989). Indeed, bargaining is a process of proposal exchange, development and debate. Integrative bargaining is tentative and exploratory since it aims at enlarging the pool of value to be shared between parties rather than imposing settlement. Since the critical assessment of recommendations and assumptions reveals what alternatives exist, parties seeking integrative outcomes will find in critically reviewing propositions a common process from which new ideas are generated and mutually beneficial solutions appraised.

Employer and employee representatives benefit from developing their argumentative skills. On one hand, valid argumentation enables speakers to convince others of the strength of their case without having recourse to lying, misrepresentation or coercion. On the other hand, those skilled in critical argumentation see through attempts on the part of others to act in these ways. Unscrupulous labour negotiators can try to set the terms of the debate, frame reality or confuse their counterparts by way of misleading statements, tangential or unrelated considerations and fallacious arguments. Studies such as Kirkbride (1986, 1988) offer examples of exchanges that executives controlled through use of such tactics, which union representatives failed to challenge when they could. Conversely, Rackham and Carlisle (1978a, 1978b) identified labour negotiations in which one party successfully resisted attempts by the other to side-track the discussion by way of compelling and repeated argument.

More generally, the bargaining table is, for some, a forum to demonstrate their verbal skills and impose their views. Argumentative language is therefore a protection against efforts on the part of a speaker to direct discussions towards distributive bargaining when an integrative settlement is in fact possible. Furthermore, knowing that arguing and counter-arguing will take place during discussions minimises the likelihood that bargaining parties will perceive proceedings in terms of domination and submission. In this sense, a shared critical inquiry is a language of mutual recognition (Emmel, 2008) that builds trust between management and union representatives. The sharing of critical arguments and justifications also exemplifies reciprocity in communication, essential when negotiators want to avoid conflict escalation (Putnam and Jones, 1982: 191).

Advising – the language of authority

Advising is the language of authority because asking for (and providing) advice assumes that the advice-giver possesses authoritative knowledge that the advice-seeker does not possess. Authority is especially visible in cases where advice is accepted without decisive evidence or conclusive reason. So-called ‘arguments from authority’ achieve only so much, however; recommendations normally require justifications, implying that, in the linguistic hierarchy, the advisory function of language is above the argumentative function. Justifications for advice can be implicit or explicit, easily identified or requiring sustained research. For instance, why workers should wear their hardhats on construction sites does not require much, if any, explanation. Why employers should provide workers who wear prescription glasses with free corrective safety spectacles does call for some elaboration, which may (or may not) be effective. Whatever the case, when advice is accepted, authority has been established. As Friedrich (1963) argued, authority arises from reasoned elaboration, that is, from exchanges between individuals cooperatively seeking, evaluating and disseminating knowledge by way of argument, counterargument and critical evaluation.

As Spillane (1987: 221) noted, the language of advice, which relies on such expressions as ‘should’, ‘ought to’, ‘in the interest of’, ‘I recommend’ and others of comparable intent is close to the language of power, with which it is often confused. In the workplace generally and during collective bargaining exchanges specifically, the difference between authoritative and authoritarian language matters. Indeed, the difference between the language of advice and the language of power reflects the difference between situations where employees accept and behave according to recommendations (because they believe that such behaviour is appropriate) and those situations where employees comply and behave according to orders. The source of authority thus differs from that of power: those who exercise or aspire to authority seek consent. Those who are happy to coerce others content themselves with commands accompanied by threats.

While employers have power over those they employ so long as they can dismiss them, employees can form or join a trade union to generate power counter to that of their employer. However, even within an adversarial context, might is not right: power not constrained by authority leads to brutality and disregard of human rights. That a collective bargaining process takes place attests to the existence of a desire, shared by employers and employees alike, to bring power under the mitigating control of authority. Indeed, for an organisation to operate, not only does a degree of consensus have to exist within its members on what actions are acceptable and desirable (Barnard, 1968: 168–169), but employers and employees alike must control their behaviour according to general and ad hoc rules, by-laws and implicit norms. Authority is therefore a concession requiring moral agreement: to grant authority is to authorise the exercise of power and to hold individuals responsible for it.

If the minimum condition for the emergence and acceptance of authority is a shared elaboration of reasons, authority is not exclusively vested in the employer, but is found throughout the work organisation. While employers authorise executives to act on their behalf and employees to work, employees authorise employers to direct them. While such authorisations are normally formalised in writing, documents such as employment contracts do not achieve anything in and of themselves. They affect reality only insofar as people adhere to them. Authority within the workplace thus requires some form of continuous, open discussion in which employers and employees (or their appointed representatives) participate. In this respect, a collective bargaining negotiation is an ideal forum in which to establish authority. During bargaining exchanges, employers and union representatives seeking integrative outcomes establish their authority, avoid coercion, share knowledge and appreciate the ethical aspects of the organisation to which they belong. Indeed, when the proceedings resolve in an outcome that is agreed by the parties, that outcome becomes authoritative (Stinchcombe, 1986: 152). In other words, authoritative employer and employee representatives value integrative outcomes over distributive ones.

To executives and employee representatives who believe that negotiations are best run according to a distributive approach because employment is a power struggle, one will retort that the language of authority does not dissolve the conflict in the employment relationship. Rather, the view advanced here is that such a conflict is temporarily and partially subsumable under agreement about a specific policy or management decision. Such situations arise because, even in an adversarial outline, there are cases where policies and decisions are found reasonable (in the sense of rational, ethical, desirable and acceptable) by employers and employees alike.

While it is true that a slack job market, tight profit margins and unfavourable economic conditions favour employer representatives during bargaining proceedings, these factors do not, in and of themselves, establish executives’ authority. What establishes authority in bargaining discussions is a capacity for reasoned elaboration (argument and critical evaluation), steeped in factual knowledge and logical thinking therefrom. Even if executive and union representatives do not start from equal premises, establishing authority on work-related problems during bargaining is an objective achievable by all.

In summary, integrative bargaining outcomes require that each side of the bargaining table practises a language of effective signalling, true descriptions and valid reasoning, culminating in authoritative advice. This conclusion assumes that signalling effectiveness is preferred to ineffectiveness, truth to falsity, rationality to irrationality, authoritative advice to authoritarian commands and cooperation to coercion. Although widespread, such preferences are not universal. In any workplace, there will be employees, be they executives, employee representatives, managers or operatives, ready to lie, misrepresent, confuse, manipulate and coerce others in the name of perceived personal, collective or organisational interests. However, since integrative outcomes are recognised as mutually beneficial by the bargaining parties, they are not achieved through manipulative language.

Research implications

Over the last few decades, industrial relations and their governance mechanisms have fundamentally changed. During the 1980s and 1990s, the post-war systems of collective bargaining that Dunlop (1958) and Flanders (1970) theorised as having for objectives the elaboration of rules governing workplace relations have almost disappeared. Human Resource Management (HRM) practices applied across industries and growing body of legislation now dominate industrial relations regimes (Dobbin and Sutton, 1998; Weil, 2014). These changes have been accompanied by a marked decline in union membership across members of the Organisation for Economic Co-operation and Development (OECD, 2020.) and a corresponding weakening of the legitimacy of traditional forms of representation (Upchurch et al., 2016). In the European Union, collective bargaining outcomes on such matters as training, wage moderation and flexibilisation of wage setting and working time have become progressively distributive (Glassner and Keune, 2012).

Since the late 1990s, employers have increasingly adopted flexible ‘organic’ organisational structures in which decision-making is decentralised, hierarchies flattened and unity of command abandoned for cross-functional teams and matrix structures (Raelin, 2011). The HRM regime still prevails, but the workplace it governs has morphed. Whereas, modernity had replaced farm labour with factory labour and separated the economic sphere (the workplace) from the domestic one (the household), postmodernity (the advent of which coincides with the start of the HRM era) has replaced the bureaucracies the rise of which Weber predicted with ever-changing adhocracies (Joullié and Spillane, 2020: 299–302). Simultaneously, the virtual office and pervasive mobile telecommunications have all but abolished the ideas of fixed working hours and workplace. At home, on the road and ever more rarely at work, employees are now contactable around the clock, 7 days a week, 12 months a year. The work-life balance debates of the Weberian workplace have become discussions about work-life integration (Williams et al., 2016).

If a degree of agreement exists on the nature of the changes that industrial relations settings have endured, analyses diverge about their causes. While most authors locate them in the growing popularity of ‘organic’ structures (Cosh et al., 2012), others have analysed them as stemming from the rise of identity politics (Bullard and Johnson, 2000; Piore and Safford, 2006). According to this alternative account, the axis of social and political enrolment has shifted. Rather than being rooted in the workplace allegiances (class, status, profession and industry) that the traditional forms of representation embodied, employee mobilisation now originates from considerations external to the workplace, notably sex, gender, religion, ethnicity, age, disability and other interests promoted by informal ‘grassroots’ communities. It is noteworthy that, owing to their nature, such considerations rest at least partly on feelings that individuals harbour about characteristics which they believe define them at work and elsewhere.

Deciding which of the aforementioned explanations of changes in industrial relations settings is correct is not a required undertaking of the current essay. More relevant is to observe that the linguistic framework proposed in these pages strengthens the voice of employees seeking solutions to work-related problems and thus contributes to the legitimacy of those who represent them, regardless of their overall setting. Indeed, while integrative outcomes cannot be guaranteed when the other party has decided against them as a matter of principle, the more the voice of employees is conveyed in a language of effective signalling, true descriptions, valid reasoning and authoritative advice, the more it is difficult to ignore. This is especially the case when the issues at hand stem not from industry-wide problems to be dealt with by way of abstract, general legislation or blanket HRM-inspired policy, but rather find their source in organisational specifics the solutions to which are to be found out of practical necessity. Employee representative bodies, regardless of their form, which achieve integrative outcomes legitimate their existence, in the eyes of those they defend as well as in the eyes of those against whom they bargain. To the extent that this article defines the language of integrative discussions, it provides a linguistic basis for such (re)legitimation effort.

Beyond a strengthening of employee voice, the linguistic framework exposed here outlines ways of analysing the language of the employment relationship generally. Three distinct yet interrelated avenues of research present themselves. First, the model acknowledges the communication of feelings, such as those arising from work situations found stressful, yet does not include such a language as an acceptable basis for the formulation of grievances and the justification of policies. Instead of expressions, integrative outcomes rest on statements (reports on organisational performance, number of hours lost to accidents, etc.) and arguments the respective truth and validity of which are critically evaluable and thus acceptable by anyone. While executives and union representatives can use the linguistic analysis offered in the preceding pages as guidelines in their bargaining discussions, researchers can use it to conduct post hoc reviews of these discussions and explain why they succeeded (or failed) to achieve integrative outcomes. It is this usage of the framework that has been emphasised in this article to this point.

Second, the linguistic guidelines outlined in this article allow considerations external to the employment relationship to be brought into bargaining discussions, but only insofar as they can be critically evaluated by the other party and found to be relevant to the workplace. Indeed, if the lowest level of language allowed in bargaining discussions are descriptions that are in principle refutable by those against whom they are directed, parties cannot advance claims stemming solely from their being members of a specific group. Speakers who want to keep identity politics out of bargaining exchanges can press those advancing claims stemming from what they feel is their ‘essence’ to formulate instead factual descriptions. For their part, researchers wanting to appreciate how identity politics have made their way into bargaining exchanges can use this article’s linguistic framework to analyse the content of these discussions to show the extent to which feelings were allowed in the proceedings. Conducted retrospectively, research conducted on these lines will help answer the question about the causes of the changes in industrial relations settings broached earlier. That is, a historical analysis across industries of the language of bargaining exchanges will reveal the degree to which expressions have come into bargaining proceedings, their nature (i.e. whether they are about sex, ethnicity, religion, etc.), when they started to dominate discussions – assuming they have – and whether this dominance preceded or succeeded the rise of ‘organic’ structures.

Third, this article’s linguistic framework emphasises as bases for integrative bargaining discussions propositions that can be critically evaluated by employer and employee representatives. As such, it provides a firm grounding for assessing the challenges that organic structures generate (rarefaction of career prospects, confusion caused by multiple lines of command, requirement for lifelong learning, blurring of the work-home frontier, etc.) for employees. In this sense, while advocating rationality of argument and soundness of advice, the framework proposed here can be viewed as protection from bureaucratic alienation insofar as it safeguards the identity, integrity and legitimated existence of workers by way of integrative bargaining outcomes. Expressed differently, to the extent that it recommends a rational approach to workplace issues over an irrational one, the analysis offered here outlines is compatible with Weberian industrial relations settings while offering protection against ‘iron cage’ alienation. Industrial relations scholars attracted by such a prospect can use this article’s linguistic framework in their analysis of workplace problems and in the formulation of their solutions.

Conclusion

To the extent that industrial relations settings generally and collective bargaining processes specifically are essentially linguistic arenas, parties to the employment relationship have little choice but to be sensitive to language and its effects on those to whom it is directed. Specifically, parties to collective bargaining will note that integrative outcomes require use of a distinctive language, namely a language of effective signalling, true descriptions, valid reasoning and authoritative advice. While using such language does not prevent distributive outcomes altogether, it conveys the strongest resistance to speakers committed to imposing their views. Moreover, even if ignored during a round of bargaining exchanges, a language of effective signalling, true descriptions, rational argument and authoritative advice strengthens the voice of speakers in future discussions. Authoritative language thus does not ignore or belittle the conflict that lies in the employment relationship. Rather, authoritative language improves the legitimacy of those who speak it and the likelihood that disputes are settled according to their preferences.

Finally, if industrial relations settings require novel forms and processes of employee representation because the postmodern workplace is inimical to traditional trade unions, then research into ways to consolidate such new representative arrangements is especially indicated. Achieving bargaining outcomes that are in the interest of employees legitimates whichever form of representation they use. The linguistic analysis advanced here has therefore practical and research value.