‘Which side are you on – Mr. Westerberg?’ Reason, affect, and division in public debate

ABSTRACT

This article studies affect and reason in practices of persuasion through a rhetorical-performative framework. Focusing on the Swedish economic crisis 1988–1993, this piece argues that the successful rise and installation of neoliberal discourse at the elite policy-making level – particularly regarding privatization – can be explained by the prominent role passion played in the presentation of its case outside mainstream media channels. In describing and explaining this shift in discourse away from a previously hegemonic social-democratic regime, the author draws on post-foundational discourse theory and rhetorical political analysis, as well as corpora from a range of fora during this period. Affect and reason are operationalized as the rhetorical concepts pathos and logos. The analysis reveals how they operate in entangled ways as combinations and constellations of etho-pathetic and logo-pathetic argumentation. A significant conclusion is that the eschewing of passion by conventional, mainstream media and hegemonic actors, made it possible for advocates of neoliberal policies to establish inroads into the Swedish political and economic establishment. The article contributes by anchoring the rhetorical-performative theory in empirical research to produce new insights into the role of affect and reason in discursive change and continuity.

KEYWORDS:

• Discourse
• rhetorical-performative analysis
• hegemony
• Sweden
• crisis
• affect
• reason
• privatization

Introduction

The quick shift from a social-democratic welfare state ‘par excellence’ (Blyth 2002, 12) to a welfare-market society in Sweden still puzzles researchers (Svallfors and Tyllström 2018; Hort 2014b; Andersson 2015; Larsson, Letell, and Thörn 2012). This is because previous analyses overlook the significance of discourse, rhetoric, and affective investments. To find out how the social-democratic project loses its grip and how the neoliberal project grips Sweden, this article analyses a historical process in which the social-democratic hegemony is overturned in favour of neoliberal hegemony: the debate on privatization in the years leading up to and through the 1990s fiscal crisis.

The neoliberal project that eventually hegemonizes the debate on privatization understands, and mobilizes through, the power of rhetoric. In The Politically Impossible (1988) right-wing researchers and think-tank intellectuals discuss political strategies to persuade an internal audience. Westholm’s analysis of the long-lived social-democratic consensus summarizes the importance of affect, reason, and rhetoric:

The labour movement stood for faith in the future, was carried by demands for justice, and perceived by its followers as natural power holders in society. ‘The Internationale carries fortune for all' was not a battle song for friends of free trade.¹ (Westholm 1988, 23)

He traces its success to affective inscriptions of ‘hope’ and ‘fortune’ – even in song; the construction of a seemingly natural authority; and symbolic collective unification in battle. ‘The bourgeoise’ has accepted ‘the opponents’ description’ of itself as ‘unfair and undemocratic’ and the market as ‘individualistic, egoistic, and competitive’ – ‘which nobody wants to be’ (Westholm 1988, 35; 33). Because feelings mobilize voters, these descriptions must be disrupted, he explains. ‘Naming’ and ‘impressions’ become crucial tools in the neoliberal struggle for ‘pluralism in opinions’ (Westholm 1988, 40). The re-description strategy strives to ‘permanently organize and fortify the understanding of and engagement for’ political liberalization (Westholm 1988, 42). The neoliberal struggle takes aim at ‘ideas’, the ‘content of concepts’, and the ‘imaginary’: ‘Once private property is anchored in the public majority, it will be difficult to abolish, [it will be] “politically impossible”’ (Westholm 1988, 42). Privatizations, he thinks, will break up ‘collectivist idea-traditions and political forms of organization’, and ‘influence people’s perception of the world’ (Westholm 1988, 41).

This article explores the roles of affect and reason in the public debate on privatization 1988–1993. This period captures fundamental shifts in the medialized public political discourse in a time of political transformation, division, and fluctuating public opinion. I analyse public utterances and rhetorical strategies employed to persuade or move the audience to act. Rather than approaching affect and reason as opposites, I explore these as co-constitutive forms of political persuasion and argumentation. As the empirical analysis shows, affect and reason are entangled phenomena in political discourse. In favouring rational forms of political expressions over passionate engagements, the very terms of public political debate are changed for decades to come.

The article proceeds in four stages. First, I review the affective turn in the social sciences and humanities. Linking this field of research to that of Rhetorical Political Analysis and post-foundationalist discourse-theory, I present a rhetorical-performative approach to analysing affect and reason. I subsequently present the key concepts used in the study. Second, I account for the research strategy and methodology employed to collect and analyse utterances in the debate on privatization. Third, I contextualize the empirical case by focusing on (a) the emergent crisis; (b) the hegemonic shift; (c) the issue of privatization; and (d) how different positions within the privatization debate develop. Fourth, these theoretical and methodological concepts are confronted with the data to arrive at an empirically informed and problem-driven analysis engaged in further theorization. The final section discusses the role of affect and reason in political discourse and change.

The affective turn: a brief overview

The affective turn, which initially focused on bodily sensations and intensity beyond rationality (Massumi 1995), now designates a heterogeneous body of literature in sociology (e.g. Goodwin, Jasper, and Polletta 2001; Turner and Stets 2005; Gould 2009), political science (e.g. Clough and Halley 2007; Hoggett 2012; Cossarini and Vallespín 2019), philosophy (e.g. Sousa 1990; Massumi 1995), and beyond. Several of these theories conceptualize affect and reason in binary, oppositional terms. The affective turn arguably reproduces distinctions between affect/emotion and rationality/reason (Eklundh 2019). While reason is attached to the mind and thought, affect and emotion are attached to the bodily and the biological. Critical readers, such as Hall (2005) and Grosz (1994), problematize the subordination of the latter in Western philosophical thought. Emotions and affect, they argue, are often treated as detached from so-called real politics, which instead is associated with reason and rationality.

Proving this point, Ahmed (2004) explores the links between rhetoric, politics, and affect. Her analysis emphasizes the role of emotions in the construction of social and political identities. She stresses that ‘emotion and sensation cannot be easily separated’ (Ahmed 2004, 6), and she acknowledges the disputed meaning of each term, but makes no clear-cut distinction between emotion and affect. She argues that emotions move between signs, figures, and objects through processes of association and displacement – and thus produce affect. In contrast, Gottweis (2007) contends that emotions/affect should not be seen as things residing in an object or being carried by the vehicle of discourse and rhetoric, ‘but as a form of rhetorical praxis that creates effects in the world' (Gottweis 2007, 240). He also criticizes the prevalent philosophical suspicion that passion is a threat to the moral and social order – and that reason is ‘the path to freedom and truth’ (Gottweis 2007, 239). To dispel the contradiction between reason and emotion Gottweis (2007) draws on Aristotle and proposes to revalorize passion.

Queer theorist Rosenberg (2013) differentiates between conscious emotion and affect, treating the latter as more immediate bodily expressions and reactions. When analysing protest songs which provoke strong corporeal responses in the audience, she recalls Aristotle’s theory of catharsis (the purification or purgation of the emotions). Affects, she argues, are inscribed in our memory of music. Inscribed affects can reappear when we least expect, overruling our awareness. Hearing the protest song Hasta siempre, she says ‘makes me feel the revolution’ (Rosenberg 2013, 186). Hence, affective responses are determined not only by the immediate and individual context, but by culture, common symbols, and history.

Cossarini and Vallespín’s (2019) compilation of contemporary studies of passion in social and political sciences shows that emotion, affect and passion are still treated as opposite axes of consciousness, experience, and identification. Whereas emotion is regarded as conscious, affect is seen as unconscious. Similarly, emotion is treated as individual experiences and identifications, whereas affects and passions are regarded as collective (Cossarini and Vallespín 2019). In summary, passion, affect, and emotion are somewhat conflated terms. The following theoretical discussion centres on the applicability of these concepts in analyses of political argumentation, persuasion, and discourse.

A rhetorical-performative approach

I now turn to the theoretical approach and the key elements of the theoretical contribution of this article. I explain the article’s central concepts and their interrelations, namely rhetoric and performativity; passion; pathos and affect; reason, rationality, and irrationality; and logos. The nodal point in this framework is the concept of discourse, minimally defined as relational arrangements of meaning-making elements that facilitate socially meaningful conceptions of reality (Laclau and Mouffe 1985). Discourse can be analysed through the practices which shape and reshape meaning, such as conceptual articulations, rhetorical strategies, arguments, images, practices and patterns, enunciative modalities, narratives, and so on (Marttila 2015; Glynos and Howarth 2007; Torfing 2005; Finlayson 2007; Wodak and Meyer 2016). Post-foundational discourse theory, also known as the Essex School, after Laclau and Mouffe (1985) and subsequent related works, has been known for its lack of a systematic methodology. This deficit is tackled in Glynos and Howarth’s (2007) ‘logics approach’, Marttila’s (2015) post-foundational research programme, and in a growing turn to rhetoric (De Cleen et al. 2021; Palonen 2018; Laclau 2014; Martin 2013; Finlayson 2007; see also Sunnercrantz 2017).

Whereas discourse theory and the logics approach allow for analyses of change and stabilization of discursive processes and subjectivities, a rhetorical analysis provides conceptual and methodological tools to analyse the practices of persuasion, styles of argumentation and appeal, relational constructions of speakers and addressees, which constitute a discourse. To study affect and reason in linguistic practices I use a rhetorical-performative approach (see also Palonen 2018; Sunnercrantz 2017), which builds on the works of discourse theorists (e.g. Laclau, Mouffe, Norval, Stavrakakis, Glynos and Howarth), rhetorical analysts (Gottweis and Finlayson), and a psychopolitical approach to public speech (Martin). This allows for an exploration into the rational as well as affective elements of public political rhetoric. I now proceed to explain the different elements of this framework in further detail.

Rhetoric and performativity

Rhetoric generally refers to the art of persuasion through speech and arguments (Finlayson 2007; Martin 2013). In contingent discourses, rhetorical practices of naming, redescription, argumentation, and persuasion are part of the contestation around meaning and conceptual change (Skinner 2002). Focusing on tropes rather than persuasion, Laclau (2014) claims that rhetoric has an ontological, constitutive character, as a practice which establishes, negotiates, and institutionalizes the meaning of social relations, identities, and objects.

Through language philosophers like Austin (1975), Pocock (1973) and Derrida (1988; 1986), we can further understand the constitutive aspects of rhetorical practices. In speaking we describe, define, and redefine concepts; produce points of identification and particular ways of perceiving the world and its objects; argue for a cause; negotiate demands, et cetera. All utterances in the debate have a performative aspect in that they ‘do’ something: they link together elements in a discourse. Laclau (2014) makes no distinction between signification and action; or speaking and doing. Rhetoric can hence be treated as a performance including persuasive speech. Even constative, assertive, or descriptive utterances are part of a rhetorical-performative practice. In persuading the public in favour of, or against, e.g. privatization politics, a language for speaking about privatization is invented, negotiated, and provided in medialized public debate (Sunnercrantz 2017).

In addressing others, we position ourselves in relation to those elements that are discursively available to us. Norval (2007) explains how an ‘other’ or a ‘they’ is constituted in the act of addressing. As there is no pregiven subject of address, it is through the articulation of a claim that the ‘other’ is constituted as the subject of the address. Simultaneously, the self is posited in relation to that other. Such is the constitutive character of rhetoric. It manifests when a political statement identifies an addressee and attempts to create a group to, for or on behalf of whom they speak – a group which is different from another group (Norval 2007; cf. Sunnercrantz 2017).

Rhetorical strategies are key factors to be considered in the processes that shape public, political discourse as attempts to persuade people and arrive at a mutual understanding, shared perception and interpretation (Martin 2013; Gottweis 2007). Methodological turns to political rhetoric (Martin 2013; Gottweis 2007; Finlayson 2007) enable an inquiry into the roles of affect and reason in political argumentation. In this perspective, affect and reason are appealed to and performed through persuasive utterances and arguments in a debate. In the Aristotelian tradition, appeals to and expressions of emotions, feeling, and passions are tied to the function of moving the audience. This is conceptualized as pathos. Applications of reason and demonstration of logic as a mode of persuasion belongs to logos. Appeals to the character or authority of the speaker are conceptualized as ethos (Aristotle 1991). Yet, Gottweis (2007) and Martin (2013) emphasize that argumentation rarely functions via either logos, pathos, or ethos – but through combinations and various emphases on these modes of persuasion.

Ethos typically accompanies an appeal to either pathos or logos. Ethos refers to the speaker’s character or authority and involves giving the listener a sense of the speaker’s entitlement to speak (Martin 2013). The establishment of credentials is a procedural, performative phenomenon created in action (Gottweis 2007). An utterance asserting the expertize, formal qualifications or experience of a speaker may imply authority over a specific field, for instance. Authority can also be exercized through the lack of style, tropes, and emotion, which allows formality to obscure facts and claims – presenting demands as descriptions (Finlayson 2007). Attempts to posit the speaker in relation to other actors, also belongs to ethos. Because affect and reason are the foci of this article, I will describe these forms of appeal in detail, starting with theoretical advancements on passion and affect, which is then operationalized as pathos.

A turn to passion

Post-foundationalist discourse theorists – e.g. Mouffe, Laclau, Stavrakakis, Norval, Glynos and Howarth – strive to undermine dichotomies of reason/affect, representation/real, and meaning/being. Their take on affect therefore differs from the affective turn introduced twenty years ago. Instead of tying affect to the material world, and reason to the symbolic, Mouffe (2005, 2013, 2018) and Laclau (2014) treat affect as both corporeal and discursive. Linguistic and affective components are, moreover, inseparable in signifying practices. Owing to the strong influence of psychoanalysis on the development of discourse theory (see Tønder and Thomassen 2005), they underscore the role of affect in the construction of political and collective identities.

Mouffe (2005, 2013, 2018) conceptualizes common affects as passion. Borrowing from Freud, she demonstrates how libidinal investments foster the affective attachments and communal bonds necessary in constructing political identities and collective wills. Passions and mobilization of affective energy aid in the construction of an abstract ‘we’ distinguished from a ‘they’ and ‘create a strong identification among the members of a community’ (Mouffe 2013, 46). This involves inscriptions in ‘discursive/affective signifying practices’ (Mouffe 2018, 39), such as words and actions. Divisions of ‘us’ and ‘them’ are then affectively supported through sentiments of anger, frustration, and indignation, for instance.

Stavrakakis (2007) treats the affective dimension as passionate attachments. These might take on the character of love, solidarity, dignity, or pride; or aversion, distrust, or hate. He sees affect and discourse as co-constitutive – meaning that the two influence and construct one another (Stavrakakis 2007). He verifies that political identification requires both symbolic articulations as well as affective, libidinal investments. Any successful identification accordingly needs to operate in an affective dimension to acquire long-term stability. A certain ideology or political grammar affectively invests, for instance, the nation with fantasies of enjoyment – blaming the lack of their ultimate realization on the activity of certain enemies (Stavrakakis 2007).

Norval (2005) explains that both affective force and rhetorical form constitutes the character of investments in a hegemonic project. Affect, understood as emotional investment or enjoyment, is constituted through the differential investments of emotional energy (cathexes), in a signifying chain. Subsequently, practices of naming are investments in a partial object, a cathected rallying point of passionate attachments. Processes of investing hate or love in an object necessarily belongs to the order of affect. Affective components are in that sense integral to the signifying process (Norval 2005).

To explicate the role of discursive/affective inscriptions Mouffe turns to Spinoza’s use of ‘conatus’ and Freud’s ‘libido’. The conatus, or ‘the general striving to persevere in our being’, ‘will experience affects that will move it to desire something and to act accordingly’ (Mouffe 2018, 39). Glynos and Howarth (2007) also argue that emotions and passions can constitute a motivating force causing agents to engage politically. Passions and emotions need to be articulated through the available vocabularies in their immediate context to be interpreted and made meaningful. They then become discursive – that is, rendered meaningful as an object of discourse. Since discourses are constituted through meaningful articulations (linguistic and non-linguistic) when subjects attempt to interpret and make sense of the world around them, discourses function to make sense of people’s affective states. A political project can thus provide a grammar for expressing emotion and channel affect into action. Moreover, it may authorize selected feelings and actions while downplaying and invalidating others, as Gould (2009) suggests. An actor may translate her feelings into a desire for political change, retaliation or revenge and anger can be launched as a channel of political action (Glynos and Howarth 2007). Affect thus carries a potential of representation and politization.

Moreover, it is desire that moves human beings to act, and affects make us act in one direction rather than another (Mouffe 2018). Martin’s psychoanalytic analysis affirms that ‘desire exerts a more powerful force on us than that of knowing a truth’ (Martin 2019, 18). That reason alone cannot motivate human beings to act, and that it cannot oppose passions (as the only true motive of the will), is an old argument (e.g. Hume 1896), but nonetheless important. Martin sees affect, like persuasive speech, as co-constitutive of meaning-making and reason – and not merely as non-conscious and corporeal (Martin 2019; cf. Massumi 1995).

Like Mouffe, Norval (2007) emphasizes the role of passionate engagement and persuasion as practices that shape politics – an emphasis that has largely been excluded from liberal conceptions of politics (see also Hall 2002). ‘What is needed, however, is a connection between the domain of language and the passions’, argues Norval (2007, 210). Martin (2019) does precisely this and points to the role of rhetoric as that of ‘not merely “telling the truth” but putting truth into a form that makes it desirable’ (Martin 2019, 18). Language, speech, and discourse provides the means and sites in which emotions and affectivity take on significance, are conveyed, and mobilize the public, as Gottweis (2007) and Martin (2019) explain. Emotions and affect ‘belong to the repertoire of rhetoric’ (Gottweis 2007, 240) and it is through speech and argument that ‘citizens are affectively engaged with formal and informal politics’ (Martin 2019, 20). Martin argues that public speech serves to evoke and find ways to orient ourselves towards the desires that propel us. It aims to grip people enough that they rethink what their interests are and how they may best be served. Rhetoric has an affective potential as it ‘provokes and mobilizes subjective commitments and aligns them to public positions’ (Martin 2019). Rhetoric is central to understand how affects and passions are incited and mobilized. The significance of rhetoricity and affective force in the constitution of political identities – and hence in hegemonic projects – comes to the fore in such analyses (e.g. Norval 2005; Laclau 2014; Stavrakakis 2007).

Pathos, affect, and passion

I treat affect and passion as parts of persuasive performances and modes of argumentation – rather than actual emotions carried by the vehicle of rhetoric (Gottweis 2007; cf. Ahmed 2004). In rhetoric, an appeal with or to emotion/affect, is conceptualized as pathos. It provides the link between language and the passions, sought by Norval (2007). Pathos is ‘probably central to any appeal that seeks to motivate others to act’ (Finlayson 2007, 558). Used to shape political goals and strategies, pathos involves identifying the shared feelings of the audience. Gottweis (2007) sees the analysis of pathos in policy analysis as a focus on the emotions/affects implemented in discourse. A speaker can employ a rhetoric that leads the audience to react affectively and feel emotions. These are intelligent responses open to reasoned persuasion, he argues. Affective responses to unjust treatment, such as anger, are based on a belief which may be criticized and altered by argumentation. By analysing pathos, we can pinpoint the use of affect and passions in the mobilization of political opinion, desire, and action.

Specific words can be used to invite certain responses and, as we shall see, signifiers can be invested with affect. Some words have violent connotations (‘fight’, ‘war’); others invite us to feel concerned (‘crisis’, ‘disaster’), joy (‘hope’, ‘love’), et cetera (Martin 2013). In the empirical analysis, affect is operationalized as (a) a type of rhetorical appeal and passionate engagement, and (b) affective/emotional attachments and investments in/to certain signifiers.

Reason, rationality, and irrationality

Reason and rationality are less contested concepts. Associated with the Enlightenment, they feature heavily in Rawlsian political philosophy and Habermasian deliberative-democracy approaches. According to Gottweis, such strands in political science and political philosophy are ‘obsessed with the idea to eliminate passion and anything remotely irrationally sounding in politics’ (Gottweis 2007, 239). Mouffe (2005, 2013, 2020) has developed a pointed critique of their avoidance of ‘irrational’ affects and passions in favour of communicative rationality. Liberal thoughts’ individualistic interpretation of politics overlooks collective identification and the antagonistic nature of the political, she argues. Because collective identifications are central to the construction of political relations, a failure to understand collective identities leads to incomprehension of political movements and expressions of irrational forces (Mouffe 2020; 2013). Norval (2007) also criticizes Habermasian accounts for favouring rationality over other elements of democratic argumentation, such as affective forms of persuasion. In envisaging political change as a result of rational argumentation, provisions of reasons, and consequent processes of deliberation, they fail to understand situations in which a common understanding is absent and consensus is impossible (Mouffe 2005; Norval 2007). Norval raises an important question which the deliberative approach fails to answer, namely, what is it that pulls human beings into the formation of a ‘common space of reasons or political grammar’? (Norval 2007, 122)

Mouffe (2005) and Norval (2007) both draw on Wittgenstein and the notion that ‘at the end of reason comes persuasion’ (Wittgenstein 1972, para. 6012). While Wittgenstein separates reason and persuasion (which accordingly relies on our feelings, emotions, and values) he also points to the limits of our reason-giving practices. From a rhetorical perspective, persuasion is better understood as the beginning, rather than the end, of reason – but to Norval, processes of identification start where reason-giving ends. This moment of change might explain how a certain political project loses its grip or holds us captive and guides our judgement:

This moment of subjective assent (‘I am a democrat!’) involves a process of identification – a picture gripping us, being occupied by something – that escapes the linguistic reductionism and excessively rational, disembodied account of much deliberative democratic theory. This ‘identification-as’ is the embodied act of a subject passionately involved. (Norval 2007, 124)

Allegiance to a political grammar, identification, and political participation entails engagement in a range of practices, include dimensions of argumentation, such as persuasion and rhetoric, as well as affective inscriptions, following Norval (2007; 2005). On the other hand, a provision of ‘reasons’ of all sorts is involved in the formation of a political consensus which involves, ‘not the “discovery” of a shared interest or opinion but its creation’ (Finlayson 2007, 551). Reason is in that sense constructed and can be affective. Arguments provide instrumental but also rational-legal, affective, and valuational reasons.

Reason as logos

Reason is here operationalized through the rhetorical concept logos. Logos signifies a type of persuasion through logical justification tactics and argument in itself by a process that seemingly proves the true state of things. Logocentric arguments appeal to reason through rational argumentation, deduction, or induction (see Martin 2013; Gottweis 2007; Finlayson 2007; Aristotle 1991). Quasi-logical arguments utilize commonplace pre-conceptions and provide parts of a syllogism while relying on the audience to supply the missing premise(s). They rest on ‘the construction or validation of a premise from which further deductions can logically follow’ (Finlayson 2007, 558). Logos traditionally includes the naming of things and defining what a thing ‘is’: i.e. what is meant by a concept like privatization. By describing an issue in a particular way, logocentric arguments present ‘ways of conceiving of a phenomenon or an event as “like this” rather than “like that”’ (Finlayson 2007, 551).

The roles of affect and reason are at some point acknowledged by all authors mentioned above. Discourse theorist have turned to Spinoza, Lacan, and Freud to understand affect, but it is in rhetorical analysis that intersections of affect and reason can become the foci of analyses. Gottweis (2007) criticizes approaches to discourse and argumentation that reduce argumentation to the operation of logos and pay little attention to passion and emotions/affect. Acknowledging this, I explore how pathos, ethos, and logos are combined to persuade a public amid a social, financial, and political crisis, within the local, public political debate on privatization in Sweden.

Research strategy, method, and empirical material

The research strategy consists of four steps inspired by Glynos and Howarth (2007). First, I see the 1990s crisis (described in the next section) as a decisive socio-political dislocation and take that as a point of departure. Second, by surveying the public debate at the time and previous analyses, I locate significant debate fora and identify political matters on the contemporaneous agenda. The question of privatization of the Swedish public sector stands out as a polarized debate. During the years 1988–1993, media activity in the debate on privatization increases (Sunnercrantz 2017) and public opinion vacillates (Nilsson 1997). Previous analyses indicate that the larger media actors steer the debate. The scarcity of broad channels for political communication allows for exclusive control of the media debate, with socio-political and conceptual cohesion, monopolies, party-political editors, and power concentration. The most prominent differences are not between different fora, but between internal sections (such as editorials and culture sections) (Boréus 1994; Petersson and Carlberg 1990).

I subsequently limit the empirical selection to opinion pieces and culture sections in the leading daily newspaper (Dagens Nyheter, DN), one of the larger regional daily newspapers (Arbetet), and two periodicals launched in 1989: the neoliberal Nyliberalen and left-leaning Thélème (later TLM). I also trace other articles, reports, and books mentioned in these contributions. DN is the by far largest (of two) national daily newspapers. It is party independent but clearly (and later officially) liberal. To widen the scope outside of the capital, I include the smaller regional Arbetet, which with its social-democratic stance provides a political and geographical contrast to DN. Both journals Nyliberalen and TLM emerge as politicized alternatives to the opinion and culture sections of mass-media establishment and continually position themselves in relation to these. The two periodicals as well as mentioned reports and books provide fora for more in-depth developments of ideas and arguments. They function as a ‘backstage’ (Wodak 2009) to the public political debate.

Third, I uncover the discourse on privatization in the textual material and compile an empirical corpus consisting of contributions to the debate on privatization through archival research (Marttila 2015). I also do complimentary database searches based on lexical elements. The studied corpus consists of circa 225 articles in daily newspapers, 25 journal issues, and ten reports/books.² The fourth step proceeds to the empirical analysis. In reconstructing the discursive structures, I first code and map signifiers, enunciate positions, rhetorical tropes, political frontiers, chains of equivalence and difference (and so on), which provides the foundations for an analysis focused on the conceptual changes, subject positions, and political formations (see Sunnercrantz 2017). In the process, I discover patterns not accounted for in discourse theory, namely in the types of arguments employed. The turn to rhetoric is therefore a result of a retroductive procedure brought on by empirical findings. The appeals to either ethos, pathos, or logos, as described above, capture the strategies of persuasion used in the debate. While logos signifies the use of reason, pathos provides a methodological tool to operationalize affect.

No rhetorical situation is exclusive to one ideal type of rhetoric or the other, as Gottweis (2007) reminds us. Indeed, his suggestion to acknowledge combinations of etho-pathetic argumentation, or logo-pathetic constellations, and so on, is consistent with the empirical analysis. Moreover, one article does not correspond to only one, or a definite number of, arguments or modes of argumentation. Only a few of all analysed contributions to the debate are re-iterated in this article. Quotations given as examples here are representative of the forum and chosen year.

Affect and reason in the privatization debate

Several rhetorical strategies and appeals are used in the debate on privatization. While one or all elements of persuasion are used, certain tendencies of the different combinations of ethos-pathos, ethos-logos, and pathos-logos crystallize over time and fora in the debate. The presentation of the analysis mirrors these constellations and diachronic developments. Before embarking on the empirical analysis, I wish to highlight the context-specific historical origins and conditions of possibility for the debate on privatization. The following section concentrates on the 1990s crisis and the issue of privatization since it is precisely through a crisis that a stable hegemonic discourse may be dislocated, as it is confronted by events that it cannot explain, represent, or control (Glynos and Howarth 2007).

Sweden on the brink

When privatizations of public sector services and enterprises become subject to public debate the Swedish welfare state is entering a severe crisis. With long-lasting effects on society, the 1990–1993 fiscal crisis overshadows even the 1930s crisis and serves as a point of reference in discussions on the 2008 North-Atlantic fiscal crisis (Krugman 2008; Hort 2014b). More than a crisis in and of the welfare state, it involves fiscal, social, and political change. It destabilizes the prevalent social-democratic hegemonic project as its dominance in political discourse is questioned, disputed, and disrupted. This presents an appropriate conjuncture for analysis.

Most research on the neoliberal turn in Sweden focuses on particular policies, reforms, or sectors of society (e.g. Larsson, Letell, and Thörn 2012; Sjöberg 2003). The neoliberalization of the social-democratic party is a well-discussed theme (see Ryner 2018; Wennerhag 2010; Pontusson 1987). The party’s return to power in 1982 is often identified as a turning point for fiscal policies and economic (in)equality in Sweden (Schnyder 2012; Therborn 2020; Pontusson 1987). Fiscal reforms recentre on devaluation and marketization, as well as measures against inflation, budgetary deficit, and wage costs (Hort 2014b; Harvey 2005; Ryner 2003; Blyth 2002). By 1985 bond-holding requirements for banks, restrictions on share purchases, and regulations of the credit market and commercial banks’ lending had been abolished. The ensuing credit explosion, inflation, indebted consumption, export losses, surge of stock market speculation, and classic housing bubble is often traced to these events (Jonung, Kiander, and Vartia 2009; Therborn 2020; Ryner 2003; Feldt 1994).

Sweden is on the brink of financial collapse by the start of the 1990s. Hort (2014b) aptly calls this the decennium horribile of the Swedish welfare state. Mass unemployment, austerity politics, and increased poverty characterize the search for budgetary balance. The Social Democrats abandon full employment as a policy objective in 1990 to focus on inflation control and international competitiveness. Inflation is seen as ‘public enemy number one’ (Blyth 2002, 147) and economists point to the social-democratic, and later right-wing, governments’ ‘cognitive locking’ (Hamilton and Rolander 1993) on fighting inflation rather than unemployment as fatal to the Swedish crisis (see also Jonung 2015). Triggered by the oil price rise in 1991, the housing bubble bursts; capital flight and internal bankruptcies ensue. In the election that year, a new populist right-wing party emerges, the Social Democrats receive record low support, and a centre-right coalition takes office. Thus, it is generally explained, neoliberalism replaced a long-standing social democratic hegemony in Sweden (Schnyder 2012; Harvey 2005; Hort 2014a; 2014b; Blyth 2002; Larsson, Letell, and Thörn 2012).

Overloaded budgets and fiscal pressure, rather than ideology and market-liberal ideas, is often seen as the cause of the Social Democrats’ decisions to deregulate capital markets, cut taxes, and down-size the welfare state. Such explanations fail to account for the active consent to neoliberalism among social-democratic actors (Ryner 2018). Although cuts in public welfare and state expenditure are argued to be the only available option, that does not necessarily mean that monetary matters cause the crisis or that this is a neutral, natural solution. Controlling the definitions of central concepts, as well as developing and implementing a dominant interpretation of the crisis is vital, according to Blyth’s ideational analysis (2002). Ryner’s (2018) analysis suggests that subtler, yet important, discursive factors triggered the neoliberalization of Swedish social democracy. Both Blyth (2002), Hort (2014a; 2014b), and Harvey (2005) point to the capitalist class’ emerging ideological warfare as a substantial factor behind Sweden’s neoliberal turn. Already in the 1970s, the national business association mobilizes to ‘increase factual, logical, and knowledge-searching elements in the public debate’ (Segerstedt 1988, 2). Think tanks, intellectuals, and publishing houses then engage to influence the public debate, the financial press, economists, and decision-makers (Blyth 2002; Maktutredningen 1990). A discourse coalition of neoliberal actors strive to institutionalize the inefficiencies of the welfare state and economic stagnation – and not the sudden market-conforming policies as the cause of the crisis (see Ståhl, Wickman, and Arvidsson 1993; cf. Korpi 1996; Blyth 2002; Sunnercrantz 2017).

Blyth, Ryner, Hort, and Harvey all argue we can only fully comprehend these political and institutional changes by understanding the ideas that agents use to interpret their situation. While this ideational dimension and discursive changes have been analysed (see Sunnercrantz 2017; Boréus 1994), my concern here is how certain forms of argument – rather than others – grip the audience and catch on in the political discourse.

The debate on privatization

In 1991, the right-wing takes office and initiates a far-reaching privatization programme of state enterprises and services, furthering previous social-democratic marketizations (Svensson 2001; Riksdagen 1991). Researchers generally treat privatization as a phenomenon of economic organization, welfare-state reformation, or system administration – a local, sectoral effect of wide-sweeping neoliberal transformation (Vickers and Yarrow 1988; Jordahl 2008; Rothstein and Ahlbäck Öberg 2010; Stolt, Blomqvist, and Winblad 2011). None explain how privatization as a political concept and project grips the public discourse.

A dividing line between protagonists and antagonists of privatization cuts across traditional divisions of left and right. Conflicting positions within and among parties, social movements, trade unions, trade associations, experts, and public intellectuals are assigned and asserted. Because the Social Democrats initiate marketization processes, anti-etatist proponents of privatization need only to negotiate the extent – not the presence – of such reforms. Once the primacy of private ownership is established as common-sensical, the legitimacy of privatization and cut-backs in the welfare state become a question of ‘how much’ rather than ‘whether or not’. Variegated rhetorical strategies frame privatization as either a moral, universal, and natural order – or as a technical and bureaucratic issue. The dislocation of the socio-political order brought on by the ongoing economic crisis enables this discursive shift. As the social-democratic hegemony is dislocated, the neoliberal project activates narratives of private property and market mechanisms to stabilize the discourse. The transformation of the economy is presented as a necessity: not just the right course of action but the only available option (Sunnercrantz 2017; 2021).

The mass-media discourse and the public opinion is radically divided on this issue. Public opinion largely opposes privatizations and supports state financing and delivery of welfare services at the time (Svallfors and Tyllström 2018). Support for privatization increase significantly in 1991, but backlashes in 1993 (Nilsson 1997). The public debate, in contrast, is hegemonized by proponents of privatization (Sunnercrantz 2017). Sweden stands out as an example of how far-reaching privatizations are implemented, despite the populations’ strong support for the traditional welfare-state model (Edlund and Johansson Sevä 2013; Svallfors and Tyllström 2018). Swedish society is thus divided hierarchically: between political spokespersons and the public they are meant to represent.

Demands for privatizations are prepared ahead of the crisis and distributed in semi-internal right-wing fora in the late 1980s (see Nordin 1988; Segerstedt 1988; Brunfelter et al. 1988). There is little initial concern voiced against privatizations in the public debate. A rare forewarning against privatizations of social services and welfare facilities, by philosopher von Wright, appears on the culture-debate section in DN on September 11, 1988. Neoliberals herald she social-democratic minister of finance as ‘the high priest of neoliberalism’, as told by the editor-in-chief of Nyliberalen, in the very first issue published in 1989. Several years after marketization and privatization reforms have come into effect, leftist intellectuals articulate a coherent critique of marketization politics. The structured ‘defence of’ (a) social insurances, (b) employment, and (c) the welfare state presented in three thematic issues of TLM in 1993, does not catch on or grip the discourse. Around 1992–1993, the opposition against privatization increases – as do the defence of privatization policies (Sunnercrantz 2017; 2021).

Now, we will take a closer look at the combinations of the rhetorical practices that assert the character and authority of the speaker (ethos); appeal to affect (pathos); and use logical reasoning (logos) to persuade the audience. The examples of modes of argumentation and persuasion reiterated here are typical for their respective time and forum. While the overarching categorization is thematic, each section is chronologically structured and mirrors a change over time where a logocentric mode of persuasion increasingly dominates the public debate.

Ethos-pathos

Affective investments of love, hatred, et cetera, are necessary to the creation of political identities. Attempts to engage the audience through emotional outbursts and passionate exclamations or appeals to the listeners’ sense of compassion count as pathos. A speaker’s way of engaging also shapes their ethos. Processes of investing hate in the state and the public sector are central to the neoliberal project. The debate book Private Property: on ownership and morality (1988), from the neoliberal think tank Timbro, illustrates this. The author, Nordin, describes the relationship between the state and the individual as that between perpetrator and victim: ‘each legislation, taxation, or other intrusion that strives towards more ambitious reforms within society is itself a criminal act’ (Nordin 1988, 96). ‘The individual’, ‘the human being’ and its ‘inviolability’ is pitted against ‘the state’, ‘the collective’ and ‘public authority’s wish to decide over the citizens’ lives and property’ (Nordin 1988). Affective inscriptions into this relation are reactivated in neoliberal rhetoric throughout the period. Most contributors to this sphere speak on behalf of ‘the individual’ ‘we’, ‘you’, and ‘I’ – constructing and ethos which is relatively unattached to any particular field or interest.

Appeals to an emotional or affective register diachronically decrease in mass media’s opinion sections. The article ‘Everything is negotiable in Malmö’, by Hedlén, in regional Arbetet on September 6, 1988, illustrates a typical etho-pathetic argument in the culture-debate section. It constructs an ethos primarily based on the social services managers’ local expertize and experiences. The author addresses local politicians and constituents. He relates an affective critique of the conservative party: ‘What worries me most is after all not the cutbacks, discontinuations, queues, and desertion. No, it is the fundamental socio-political attitude itself and that everything seems to be negotiable, even basic principles of legislation’. In redescribing market force as ‘harsh’ and ‘insensitive’ he creates a ‘they’ that is logocentric and insensitive; and an ‘us’ which is pathetic and sentient. These market forces then embody a ‘they’ which represent the cause of ‘our’ feelings of neglect, fear, and anxiety.³

Hedlén’s argument also affectively invests the signifier ‘our country’, Sweden, with pride, tradition, belonging, justice and solidarity – blaming the lack of its’ realization on conservative demands:

…  it has been built up over decades in our country […] ‘the Swedish model’, is what they want to throw away – and the Moderate [party] in Malmö don’t think that the Social Services Act, our foremost expression of justice and solidarity, in any way corresponds to their ideology and view of humanity.

‘Frightening clearances’ on public property are threatening the collective ‘us’ and ‘our’ ‘common property’ in this nationalistic narrative. In addressing both the local public and politicians, the article rallies the audience ‘to a strong defence of the values at stake’. The initial administrative ethos is replaced by that of the agitator through a pathos-centred concern for ‘all people’ in ‘protecting them from poverty’. No longer speaking on behalf of himself or his position, but on behalf of the collective, the passionate end statement urges ‘us’ to act in the coming election: ‘We have to take the warning signs seriously – we have the opportunity to do so on the 18th’. The article engages readers in this moment of identification with a ‘we’ against a differentiated ‘they’.

Patho-centric arguments are more common in alternative fora, especially the editorials of the neoliberal journal Nyliberalen. Messages to ‘hate the state’, ‘you own your life – violate infringing laws’, that ‘capitalism is freedom’ and ‘taxation is theft’, are printed on decals, t-shirts, and other paraphernalia available to order from Nyliberalen. Several articles in the 1st issue 1990 capture the passionate engagement of the editor, Varveus. The editorial ‘The welfare state creates immigrant-hate’, claims that ‘the welfare state creates hostilities or even hatred against immigrants and humiliate people by refusing to let them stand on their own two legs’. It encourages the reader to ‘hate the state – not the immigrants’. In a debate article in the same issue, the editor speaks directly to the reader and addresses active as well as potential members of the organization that publishes the journal:

Are You not content with simply clenching your fists in your trouser pocket? Do You really want to do something to force the development in the right direction and throw out the state from our lives? Now there is an option. Now there is the Freedom Front. So now it is really only up to You. We will meet on the battlefield!

This passionate rhetoric serves to arouse the audience and engage the sleeping masses. The use of capital ‘Y’ enforces the sense that the author is speaking directly to the reader. Rhetorical questions call upon the reader to sympathize with the authors’ standpoint and identify with the ‘we’. It is an affective call, not just to action, but to arms. ‘The state’ is reactivated as an affective signifier that symbolizes inimical intrusion, described a power to be forcefully ‘thrown out’. The editor constructs a narrative of indignation by speaking on behalf of a universal collective ‘us’ subjected to the impositions of this hateful state. He aligns himself with the reader against the state and the status quo. Such clear uses of pathos to mobilize the audience are frequent in this journal but scarcer in mass media.

In an opinion piece in DN, social-democrat Dahl encourages the leader of the People’s Party to ‘Break away from the government’, on April 3, 1993. The author constructs an affective appeal through a personal narrative of the ‘questions that Bengt Westerberg and I usually discuss and defend’, without mentioning the party that Westerberg represents. The article provides a passionately engaged call to action: to ‘stop’ the right wing from ‘deepening’ the crisis. It affectively invests negativity in right-wing attempts to ‘tear down’ welfare and equality; and a sense of concern as ‘we are deeply worried’ over ‘the extremely brutal development’ which the ‘systemic change and privatizations’ may lead to. A moment of subjective dis-identification: ‘But I/we are not system preservers-institution defenders!’, indicates that the old social-democratic ‘system’ is widely refuted.

The author asserts common points of identification and collaboration between the People’s Party and the Social Democrats, in parliament as well as amongst the public, through affective investments of hope and fear:

We – the Swedish people – have everything to gain from us working together to stop those who wish to use the deep economic crisis as an alibi to tear down welfare and equality. […] I confess that we Social Democrats think that work of renewal of this sort [welfare policies] is more important and more exciting than putting day care centres to companies. […] Which side are you on – Mr Westerberg? [italics in the original]

The use of cultural symbolism through reference to the protest song ‘Which side are you on’ functions as a point of identification and a call for the reader to take a stand. Inscribed with a feeling of division between oppressor and oppressed, the song recalls the labour movement and the coal miners’ struggle. It has the potential to grip the audience in a moment of identification with the tragic protagonist in the song. This reference links the mining companies to the contemporary right wing: both are antagonistic to the speaking ‘we’. As Rosenberg (2013) argues, such use of common symbols functions on the level of sensibility. Just as Hasta siempre carried Rosenberg back to her entrance into politics, ‘Which side are you on’ likely resonates in a social democrat, and perhaps a social-liberal like Westerberg. It forces the addressed ‘Mr Westerberg’ to take sides while narrowing that choice to one between oppressor and oppressed. This, and similar contributions, refer to privatization policies only in passing. Most concrete instances of privatizations are debated through a use of logos.

Pathos-logos

Logo-pathetic constellations combine the passionate exclamations and appeals of the affective dimension with nuanced rational discussion employing a logical, deductive consideration. A contribution to the culture-debate in DN on September 11, 1988, ‘The Welfare state and the future’, by philosopher von Wright does precisely this. Von Wright describes the development of society from the industrialization to contemporary challenges through deductive reasoning. He ‘experiences these developments as very worrying’, especially the new phenomena ‘privatization’, which ‘involves a gradual abandonment of the very idea of a welfare state’. In doing so he combines affective investments and appeals with logical arguments to convince the reader that his worries are, in fact, reasonable.

By 1992, privatization has become an established austerity measure under the right-wing government. The minister of finance, Wibble, contends that ‘The budget counteracts neoliberalism’ in DN’s opinion section, January 10, 1992. ‘I want to allow private doctors’ for ‘real freedom of choice and influence for the individual’; and ‘I want to allow private employment agencies’, et cetera, she argues. The article combines a strong use of ethos with affective signifiers such as ‘freedom’ and ‘the individual’, and economistic reasoning regarding budgetary balance.

A typical logo-pathetic argumentation ‘Freedom of choice in Vaxholm’ appears in the leftist periodical TLM no. 4, 1992. Journalists Alstadt and Malmqvist construct a quasi-logical argument:

The nursery loses municipal subsidies, the economy gets worse, and it becomes harder for the staff to do a good job, which leads to more children changing nursery. […] Deterioration into slum and segregation of the child-care system, schools, and elderly care may be the result of the new competitive situation.

The unspoken premise is that a competitive child-care system entails the creation of winners and losers. Privileged children consequently end up in ‘winner’ institutions and disadvantaged children in ‘loser’ institutions. The argument relies on the audience to make this connection between competition and segregation through a syllogism of competition-induced inequality. Yet, the feature also draws on feelings of fear, hate, despair, in addition to sympathy with the less fortunate.

The affective dimensions of policies are often discussed in TLM. In the first issue in 1993, the article ‘Why are there social insurances?’ by journalist Carlén weighs various arguments speaking for and against privatizations. He deconstructs these in a logical-deductive manner. He also appeals to readers’ sympathies and address recognizable fears through personal, everyday experiences:

I am on paternity leave and have the opportunity to experience my daughter’s exceptional progress. […] I have also so far been safe in the knowledge that if I get ill or get into an accident it does not necessarily entail a disaster for the whole family.

This constructs a familiar point of identification and an ethos of the family father. But this father–child relationship is threatened ‘by the recent attacks on the social insurances’. The author questions ‘to what extent are the reasons to smash this system grounded on facts?’ A reason-based focus on the factual relevance of privatization signifies a turn to logos. He re-describes the welfare state as a historical improvement-project initiated to replace the private ‘inefficient and inadequate older market-controlled social security systems’ – for the good of ‘all citizens’. In countering the ‘artificial’ critique of bureaucracy, the author invokes the ‘concerns of individuals’ and thus concedes to an individualistic perspective. Logo-centric constative utterances state matters as the ‘actually’ are: ‘it is very easy to blame bureaucracy, when in fact it has proved to be the most efficient and just way to manage social policy’. The current public system, he argues, is ‘profitable, efficient, and fair’. Yet, the following and concluding statement; ‘let us for goodness sake not repeat the mistakes of the past’, is passionate engagement by way of pathos.

Moral justifications of privatization are common to the neoliberal sphere. In a typical feature from Nyliberalen, regular writer Hinze attempts to aid fellow neoliberals to spread the ‘ideas of freedom’ in ‘Why aren’t all neoliberals anarchists?’, no. 4, 1993. By sharing reflections and experiences in ‘convincing others of the moral and practical superiority of neoliberalism’, the author mobilizes the reader in the fight for neoliberalism and privatization. He recommends that the reader does not ‘start by explaining how the police might be privatized’ but assumes that the reader agrees that the police should be privatized. The article asserts that a privatized society is the only reasonable solution. This use of logos incorporates de-contested arguments regarding the inefficiency of the state: ‘The logical follow-up question after having encountered the neoliberal philosophy for the first time is: “But if the state is so inefficient and politicians are so autocratic, then why don’t we get rid of it entirely?”’. The anonymous group of untrustworthy politicians is constructed as the enemy that hinders the realization of the neoliberal desire.

The appeal to reason is complemented, or even preceded, by the affective investments in ‘autocratic’ politicians. The author guides the reader through a logical sequence of questions and events, using fictional examples where security companies start a destructive war. His conclusion is passionate rather than rational: ‘then I would be the first to complain, just as I point out that what we have got above us is nothing else than a state. So, what do we have to lose?’ At this point, the state as a discursive signifier is already inscribed with feelings of hate and functions as an affective signifier to be opposed, fought, and overthrown.

Ethos-logos

Informative rather than agitating contributions regularly feature in opinion-sections and culture-sections of the daily newspapers. Representatives of political parties, state bodies and authorities dominate the public debate. So does arguments in favour of privatization narrated in technical terms of administration, cost, and efficiency. Many debate contributions present a seemingly neutral and un-biased account of events and policy proposals – presenting demands as descriptions. Such matter-of-fact argumentation (logos) also constructs the speaker’s character and lends it a certain authority (ethos) exercized through the lack of passion (Finlayson 2007). Reason is thus performed through argumentation.

We find a typical example in Arbetet’s culture-debate section, on September 5, 1988. With less than a fortnight to the coming election a member of the local social democratic opposition, Wessling, demands that right-wing politicians ‘Step out into reality’. She implies that the municipal council is disassociated from the everyday reality of citizens – who would be better represented by the Social Democrats. The article criticizes the municipal commissioner for ‘selling off’ municipal property at ‘too low prices’. Rationalist arguments of cost-efficiency are combined with an ethos that speaks as and to local politicians, in the economic interest of taxpayers and residents. Three days later, the social-democratic minister for Civic Affairs, Holmberg, criticizes the oppositional liberal People’s Party (Fp) under the heading ‘Fp’s wobbliness threatens the welfare’, on September 8, 1988, in Arbetet’s culture-debate. He uses technical terms and internal jargon to describe the potential effects of right-wing policies on public administration. Similar contributions from politicians and municipal employees regularly appear in mass-media’s culture-debate sections, but few discuss cultural concerns. Then, as the left wing wins the local and regional election in 1988, the number of contributions debating privatization in Arbetet decrease.

Economist and social-democrat Södersten justifies privatizations as efficiency-measures, in several rational, informative, and seemingly neutral contributions to DN’s opinion-section. In ‘That’s why it’s so bad’, January 3, 1992, he concludes that ‘weak product development causes the fiscal crisis’. The formality of his statements obscures the political claims. Opponents of privatization concede to equally technical terms and logo-centric appeals. This is evident in the leftist periodical TLM. Here, efficiency-arguments are used to defend the welfare-state system, yet serve to instate economic, administrative, and managerial efficiency as central values. In ‘The escape from reality’, in TLM no. 4, 1992, Carlén emphasizes the ‘economic reasons for public financialization and production of services’. By 1993, logo-centric argumentations dominate the public debate. Contributions ‘tainted’ by passionate and political engagement and attempts to mobilize the public are scorned. In the opinion-piece ‘Scientific dishonesty’, historians Hedlund and Gerner argue that economist ‘Anders Åslund has left the researcher-role to become a purebred agitator’, in DN on January 28, 1993. Accordingly, a lack of pathos contributes to the creation of a convincing, reasonable, ethos.

The director of the newly instated Swedish Competition Authority speaks of county administration, regulations, and municipalities’ operational expenditures in the opinion-piece ‘Make nurseries into businesses’, in DN on August 16, 1993. A long chain of reasoning on public-childcare expenditure, explains that ‘elements of freedom of choice and pluralism’ provide ‘incentives for cost pressures’. Detailed descriptions of budgeting systems support the claim that ‘establishment of private nurseries will entail increased competition and likely lower costs’. Through deductive reasoning, the author argues against current regulations. He asserts that the reason ‘why so few municipalities hitherto have tried to improve childcare efficiency, by making e.g. nurseries subject to competition’ is municipal employees’ ‘poor knowledge of competition’. By undermining the ethos of these anonymous actors, he builds his own ethos – implying that he possesses the knowledge and thereby authority that they lack. As a clear use of logos through deductive reasoning it constructs an ethos based on knowledge and rationality. Yet, it also invests an affective sense of distrust and blame in municipal actors.

Many contributions from the social-democratic sphere argue for privatization in technical, administrative terms. The social-democratic parliament-group leader, Bergqvist, replies to an earlier critique from the liberal party’s group leader in DN’s opinion-section, March 17, 1993. The article ‘Show your true colours, Westerberg’ addresses the liberal politicians both directly and indirectly: ‘what shall we do about that unfreedom, Lars Leijonborg?’, ‘don’t give up, Lars’, and ‘how can the group leader write that […]’. The argument constructs an individualistic definition of freedom focused on ‘choice’. A politics for a ‘good society’, he argues, ‘must build on a combination’ of ‘the collective’ and ‘the individual’. He described collective decision-making as the means for increased individual freedom and ‘possibility to choose and participate in welfare’. Logos includes, in this case, naming and defining what privatization ‘is’, stating things as they are, and declaring what phenomena and perspective ‘must’ be seen as matters-of-course.

Through induction and references to a report from a right-wing think tank, the author argues that social-democratic municipalities privatized at lower costs than right-wing municipalities, because they ‘formulated contracts more precisely’ and did ‘their outmost to find more probable entrepreneurs’. Here, cost-efficiency arguments function as rational and logo-centric justifications to promote privatizations in social democratic hands. Rather than an affective appeal to freedom of the people – as might be expected from oppositional social democrats, the author appeals to reason: ‘the most important thing to do today to increase freedom in Sweden is to pursue a policy that will take us out of stagnation’. The word ‘economic’ is left unspoken and implicit before ‘stagnation’. In justifying that this ‘gets growth going and provides chances of work and education’ he also omits ‘economic’ before ‘growth’.

The author constructs an ethos by invoking his formal position as group-leader for the social-democratic party. By referring to personal experience, ‘having participated in the finance committee’s dramatic meetings’, he creates an authoritative and credible ethos. Finally, in defending privatized welfare he speaks on behalf of ‘We Social Democrats’ against the failed ‘Right-wing wave’: ‘It is soon time for a new era, for a new politics for work, justice – and freedom of choice’. This constructs a point of collective identification and an ethos tied to the social-democratic ‘we’. He also constructs a point of dis-identification with the right-wing adversary. Reason thus precedes and legitimizes the concluding affective investment in the social-democratic collective.

By 1993–1994, privatization has been institutionalized as a technical, administrative issue. It is discussed in terms of the instrumental welfare-state organization and ‘transfer systems’ in ‘Everyone was supposed to be included’, by Thorgren in TLM, no. 17/18t, 1993/1994. Technical and instrumental reasoning, as well as justifications through tradition, morals, and rational choice theory, appears in another example from the same issue: ‘The ticking bomb’, by Lappalainen. The author re-narrates marketization arguments, asserts that the public welfare system is relatively cheap and that ‘a market system runs the risk of being more expensive’. ‘The commercialization of health care will deprive us (i.e. society, citizens, politics) of control over the costs’, he argues while constructing several points of identification. In proving that a privatized system is inefficient, expensive, and doomed to fail, the author relies on facts, technicalities, and cost-efficiency arguments. Both proponents and opponents of privatization within the social-democratic sphere, and in mainstream opinion-sections, now use logo-centric arguments.

Concluding remarks

We cannot hope to fully understand political change without appreciating the practices that constitute political discourses. This includes how affect and reason are simultaneously appealed to, invested in, and performed in rhetorical practices of pathos and logos. This article has explored how the social democratic project lost its grip and how the neoliberal project in turn gripped the public discourse. The disruptive moment of the 1990s fiscal crisis loosens the grip of the existing social-democratic hegemony. A new neoliberal political project is instituted in its place. In moments of crisis people need to be persuaded to arrive at a mutual understanding, shared perception, and interpretation (Martin 2013). This is the purpose of the utterances in the debate on privatization.

It is clearly not an invitation to reasonable argumentation around privatization that carries the necessary force to rally actors to its cause. The proponents of privatization succeed because they first mobilize through affect. Neoliberal actors assess and questions the social-democratic project that still grips Sweden in the late 1980s (see Westholm 1988). By emphasizing the sense of crisis, they challenge the social-democratic world view with competing visions of the (privatized) future. In neoliberal fora, a strong appeal to pathos first engages and mobilizes readers. Philosophical, in-depth reasonings later work through logos to persuade them.

The neoliberal style of argumentation is effective because it amplifies the ‘pathetic’ dimension. An affective component is necessary for identifications to acquire long-term stability (Stavrakakis 2007). On the other hand, affectively invested rhetorical interventions can disrupt stable identifications – just as identifications with the social-democratic system are disrupted in the privatization debate. Passions and affects are crucial to disrupt the social-democratic consensus (see also Mouffe 2013; 2018).

Political spokespersons, experts, and intellectuals mobilize a variety of passions and affects by employing affective signifiers in the debate. Affective attachments take on the character of solidarity, pride, distrust, and hate. The social-democratic political grammar affectively invests the nation with fantasies of enjoyment, blaming the lack of their ultimate realization on the activity of certain enemies: the conservatives. Proponents of privatization invests the market with fantasies of enjoyment, blaming the lack of its realization on the existing system. The neoliberal project that hegemonizes the privatization debate does so primarily through affective appeals. The neoliberal discourse mobilizes a series of affective signifiers: ‘hate the state’, ‘capitalism is freedom’ which invest hate and distrust in the state, the system, and the public sector.

Through affective investments, ‘privatization’ functions as a signifier for various demands: freedom of choice, marketization, individual empowerment, et cetera, invested with positive enjoyment. The affective investments in individual freedom grips the public debate on privatization. It rallies around ‘the individual’ as a point of identification. The moment of subjective assent (Norval 2007) thus involves a process of identification with the individual and the promise of freedom, against the state. Although the emergent neoliberal ideology is individualistic, the rhetorical strategies centre on universal identification and common affects. The success of persuasion is less about the presence or absence of affect, than it is about the way affect is caught in the discursive network. In this case, affective investments in the state, the system, and the public sector, catch on in the discourse. The opposition of privatization is not as affectively invested and lacks a mobilization of passions. If passions constitute a motivating force causing agents to engage politically (Glynos and Howarth 2007), then we can at least state that neoliberal actors are motivated, and the neoliberal discourse constitutes the motivating force, at the time. The neoliberal political project provides a grammar for expressing emotion and channel affect into action. The neoliberal discourse captures a desire for political change, retaliation, and anger to be launched as a channel of political action. The left captures, instead, a desire for continuity that is not launched into action.

There is an overall shift in argumentative style (understood as a function of pathos, ethos, and logos) that takes place in the late 1980s to early 1990s in the public political debate. Affective appeals and patho-centric persuasion give way to technical logo-centric contributions in public opinion-sections. A few pathetic arguments are uttered in opposition to rational appeals to reason, while others combine pathos and logos. A tendency to belittle and depreciate passionate and politicizing contributions are evident in all examined fora. Whether regarded as inappropriate, irrelevant, or illegitimate, affective appeals and passionate engagements are pushed out of the public political debate.

Proponents of privatization are better prepared for this turn than opponents of privatization politics. Because the neoliberal discourse has already mobilized on affect, the mainstream debate’s development into a space of reasons serve to assert the neoliberal discourse and minimize passionate contestations. The character of contributions to the leftist periodical mimics the mainstream debate. The journal develops into a one-sided dialogue with the mass-medialized debate, possibly to gain legitimacy and entry to that sphere. Ignoring the affective dimension locks the opposition of privatization in a rationalist framework unable to grip the discourse (cf. Mouffe 2018). Moreover, it locks the public debate in place as a neoliberal space of reason (cf. Norval 2007). The opposition ends up in a system defence that fails to grip the debate – even if it concurs with public opinion.

The arguments available to the broader public through mass-media are by 1990 predominantly logocentric and favour privatization. This results in a consensus – not just in privatization politics – but in forms of political argumentation: on how to participate in the public debate. It is, in that sense, not affect that threatens democratic society – but the exclusion of passionate, affective engagements that leads to an erosion of pluralism in public political discourse. Opinion sections function as internal fora for political spokespersons, business, and civil servants who use logo-centric arguments. Passionate engagements are banished but find other outlets of expression, especially in neoliberal fora. Mainstream debate sections are dominated by an exclusive, rational expert-discourse promoting privatization. Such arguments, however logical, factual, and reasonable, do not really engage, persuade, or mobilize the audience. Opinion polls show that the public majority is not persuaded by pro-privatization rhetoric but are still captivated by the welfare-state project. The hierarchical division between persons in power on the opinion sections and the public poses a democratic problem. Nevertheless, the Social Democrats abandon that project at this time. Several contributions seem to cover up this change of course by invigorated collective identifications (‘we’, ‘social democrats’, ‘our country’). Such identifications are constructed, as the analysis shows, through affective investments of sympathy with the less fortunate against an oppressive other.

Attempts at rallying a defence for the welfare state appear as scattered elements, where some support and others oppose privatizations. Looking back on the neoliberal analysis of the social-democratic hegemony in 1988 (see Westholm 1988; Segerstedt 1988) we can see that the tables are turned. It is no longer the labour movement but neoliberalism which stands for faith in the future. The neoliberal project is carried by demands for freedom and private property; but it is legitimized by a broader discursive emphasis on cost-efficiency and a naturalization of privatization as a viable solution to perceived problems in the welfare-state system. As demonstrated in this article, a turn to rhetoric allows for a nuanced analysis of the connections, interrelations, and constitutions of affect and reason in discourse. The reinvigoration of rhetorical analysis disrupts lingering divisions between affect and reason, while still analytically distinguishing the two.

Acknowledgements

This article benefitted from helpful anonymous reviews and editor Lars Tønder’s comments. I would especially like to thank reviewer #2 who understood the text better than I did myself, and whose comments greatly aided in the theoretical and analytical sharpening of the article.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Liv Sunnercrantz

Liv Sunnercrantz is a post-doctoral fellow in Sociology at the Department of Media and Social Sciences, University of Stavanger since 2019. Her work falls mainly in the areas of post-foundational discourse theory, the sociology of knowledge and the sociology of intellectuals. She also studies hegemony and populism, and has co-founded the research group on Populism, Anti-Gender & Democracy at the University of Stavanger.

Notes

1 All translations from Swedish to English are mine.

2 For a detailed account of the empirical selection, method, and selected fora, see Sunnercrantz (2017).

3 Later contributions in DN’s opinion-section define the market as insentient and rationally calculating. See e.g. Alserud’s, ‘The lid is already on in municipalities’, November 19, 1991; ‘The economists’ esprit de corps silences critique’ by Boethius and Rothstein on March 1, 1993; and ‘All public production should be open to competition’, by Lemmel, Lindmark, Schartau, and Tunhammar, June 29, 1993.