“Ideology and simultaneously more than mere ideology”: On Habermas’ reflections and hypotheses on a further structural transformation of the political public sphere

As someone who works interdisciplinarily between media and communication studies and sociology with an overall interest in critical social theory, Habermas’ “most successful” (Habermas, 2022a, p. 145) book is important in at least three respects: He develops a normative, critical concept of the public sphere and the formation of public opinion, which aims at democratizing domination, that is ties it to a process of unrestricted discussion of questions of general interest involving all those affected. The development of this possibility, but also the transformation and disintegration of the public sphere, is sociologically embedded, that is, considered in the light of changing socio-spatial frames of reference, mediatization, and political-economic developments (cf. Seeliger & Sevignani, 2022). Habermas, like few in critical social theory, is concerned with the organization and political economy of the media.¹ It was therefore a great honor that he not only contributed a commentary to texts edited by Martin Seeliger and me (first in German: Habermas, 2021, then also in English: Habermas, 2022a), but even wrote an independent text, which is now also available as a book together with smaller texts (Habermas, 2022b).

Against this background, I was very pleased to be invited to participate in this symposium on his new book. In this contribution, I will first, according to Habermas, briefly sketch the role of the public sphere in liberal-representative political systems and its transformation during the rise of digital communication. Then, I will point to a notable tension, immanent to Habermas new reflections, between, on the one hand, the normative goal of communicative learning and development and, on the other hand, his affirmation of “editorial tutelage.” This tension, I think, presses to repose the question of ideology again that Habermas has removed from and at best locates outside the public sphere. By making the gate-keeper paradigm of mass-media communication as a yardstick to evaluate the ongoing transformations, Habermas tends to misjudge the quality of digital semi-public spheres. Not the lack of generalization, I argue finally, but a different, emancipatory, form of generalization is the core problem of public opinion formation.

In his new reflections, Habermas recapitulates—in a concise but accessible form—his approach to critical theory as reconstructive critique (1), his “sociological translation” (Habermas, 1996, p. 315) of the political public sphere (2), and the effects brought about on this same public sphere by social media (3). This final point probably accounts for the great attention that the text has already received.

First, Habermas reconstructs, starting from the late 18th century, the rational but incompletely realized content of a modern subjectivity that sees itself as free, equal, and wanting to shape its future. This subjectivity could only have developed through democratic institutions, such as human rights, constitutions, and parliaments, which were historically of the same origin. To meet this rational content under the conditions of socio-economic heterogeneity and a lack of cultural background consensus, modern societies depend on the establishment of spheres of deliberation and public opinion-formation. In order to do justice to their own demands, these spheres must, as far as possible, have two characteristics, namely the equal participation of all those concerned and a specific discursive quality of communicative exchange, which consists of addressing relevant social problems and developing viable solutions for them.

Second, he outlines what he sees as an ideal two-track political system in which decision-making and deliberation, “strong” and “weak” publics (Fraser, 2014), are separated by political representation. In this system, a political liberal culture provides inputs that the mass media condense into a plural public opinion, which citizens then use to guide will-formation through their electoral choices. The direction of electoral decisions may only be modified—one could say post-democratized—by the inherent logic of political institutions and other direct influence on the legislative process to such an extent that the political decisions taken continue to be recognizable and perceivable as caused by the informed electoral decisions of citizens. Only if citizens can recognize themselves as the authors of the laws and rules of the political process will they abide by these rules and defend them.

In this way, a liberal political culture, which also recognizes electoral defeats and dissenting opinions as legitimate, is reproduced by the institutional structure of mass media and political system in a reinforcing way. Antagonisms that tear the community apart are thus hemmed into community-compatible agonisms (cf. also Mouffe, 2005), flanked by welfare state redistribution measures and state education support for the development of discursive competences of all citizens. The mass media are a crucial element of the politically liberal culture and a precondition for informed electoral decisions; they are supposed to be responsive to the problems of the citizens, to generate attention for these problems, and to condense and delimit positions through their editorial processing services, as well as to present them for discussion in a verified and comprehensible manner.

Third, according to Habermas, in the course of the digitalization of the media system, the way citizens use media is changing in such a way that the range of inputs is expanding while, at the same time, the intensity of reception is decreasing. For a minority of the audience (in the Federal Republic of Germany, at least), social media use forms echo chambers that tend to oppose each other, if not seal themselves off. Such publics relate to each other irreconcilably and no longer agonally. This hybridization of the media system also puts the classical mass media under pressure in the competition with social media for advertising revenues and the attention of citizens, with problematic consequences for the deliberative quality of plural opinion-forming.

Crucial for Habermas’ theoretical architecture, however, is the mixing of the private and the public in the newly created “semi-public sphere” (Habermas, 2022a, p. 166). The tense simultaneity of socio-material private and citizen interests (expressed, for example, in the framework of constitutional patriotism), which for Habermas could be successfully litigated through the political public sphere, tilts in the digital public spheres toward self-interest and to the detriment of the common good. People no longer discuss issues of fundamentally general interest, and they no longer address others as equal “opponents” within the framework of shared binding rules of discourse. This erodes the fragile liberal political culture and, as a consequence, the perception of political decisions as traceable to democratic principles that should confirm this kind of culture.

In view of this complex, coherent theoretical framework elaborated over a long period of time, one is always tempted as an interested reader or theorist to fall into the rut of impressive grand theory when agreeing with one or more aspects. Since the publication of Structural Transformation in 1962, however, the Habermasian theoretical framework has already been critically evaluated from a hegemony- and ideology-theoretical perspective (e.g., Eley, 1992; Fraser, 1992, 2007, 2014; Koivisto & Väliverronen, 1996; Negt & Kluge, 1993). These engagements posed a number of fundamental questions. The first of these addressed whether the bourgeois public sphere and the “unconstrained compulsion of the better argument” can really provide a model in which rule is democratically liquefied, or whether it cannot shed its class-specific taint because in it private interests must always be disregarded (ultimately even the ideal of communicative rationality remains a means of rule). A second issue was the extent to which civil society and a political-liberal culture can really be the starting and end point of social emancipation if, following Gramsci, domination is “softly” reproduced through the production of consensus (armored with state coercion). Finally, these critiques questioned whether the social differentiation achieved in capitalism between a private and a public sphere—and also between system, inclusive private governments in the economy, and life-world—can only be readjusted, but not established and set up differently. Pointedly, it has been argued against Habermas that the public sphere he envisages remains ideological. Following on from such problematizations, I will critically examine the extent to which the formation of public opinion conceptualized by Habermas can come into conflict with his own goal of developing people's communicative agency and competences, and what role social media and digital communication might play in this.

My point of departure are two passages in the new Reflections and Hypotheses which for me do not fit together without tension. The previous pre-digital media system, which is often described with the “gatekeeper” paradigm, is characterized by a one-to-many form of communication, a socially differentiated journalism that is institutionalized between the private and the public sphere and is carried by professionalized communicators or journalists who provide editorial ex-ante quality control of communication. For Habermas, this model does not imply a “disenfranchisement of media users; it merely describes a form of communication that can enable citizens to acquire the necessary knowledge and information so that each of them can form his or her own opinion about problems in need of political regulation” (Habermas, 2022a, p. 160; Sevignani, 2022b).

On the other hand, I think he is right to underline an ambivalence that accompanies the eroding gatekeeper paradigm and subsequent re-intermediation through social media platforms in what is now a “hybrid media systems” (cf. Chadwick, 2017). This system is now characterized by “mass self-communication” (Castells, 2009), that is, a blending of one-to-many and many-to-many communication, in which lay communication mixes with professional offerings, with the effect that the quality control of communication not only becomes questionable, but also tends to have to be done after publication by the recipients, who may also be authors. Habermas argues accordingly: “One effect is the self-empowerment of media users; the other is the price the latter pay for being released from the editorial tutelage of the old media as long as they have not yet learned to make good use of the new media. Just as printing made everyone a potential reader, today digitalisation is making everyone into a potential author. But how long did it take until everyone was able to read?” (Habermas, 2022a, p. 160).

The interesting point here is this: A successful, emancipatory public sphere and public opinion-formation requires and promotes the development of communicative competences (here: reading and authorship), which are supposed to unfold in the course of media development. However, although he speaks of an “editorial tutelage” in the pre-digital media age, Habermas underestimates the problematic aspects of institutionalized (mass) communication. He sees “sluices” and “gate-keeping” in the process of public opinion-forming as positive and necessary because they increase the discursive quality or relieve the burden on citizens, thus enabling an overall rationalization of the lifeworld and thus representing a historical achievement. However, communication can not only dissolve domination but also solidify it—and this also has consequences for the development of communicative competences or the possibility of emancipatory public learning. Let me elaborate (for a more detailed line of argument cf. Sevignani, 2022b).

In the course of developing his theory, Habermas excluded problematic aspects of communication, that is, systematically distorted communications or “communication pathologies” (cf., e.g., Habermas, 1984, p. 252ff.) as well as “systematically distorted communication conditions” (Strecker, 2012, p. 179ff.) on the level of institutions in favor of the problem of communication suppressions—happening in the media system, for example, through strategic (advertising) communication (cf. Kempf, 2023). This leads him to a positive assessment of quality media, as long as they do not let their reporting be influenced by strategic (advertising) communication. In addition to what can reasonably be understood as communication suppression, however, other “filters” act on the condensation and provision of information and opinions within the gate-keeper paradigm with its “editorial tutelage” (cf. for instance the discussions in Mullen & Klaehn, 2010; Zollmann, 2019).

The emergence and permanent reproduction of ideological communication—not the suppression of communication—can be derived from a fundamental need of subjects to become capable of acting. Agency is defined here as “the human capacity to gain, in cooperation with others, control over each individual's own life conditions” (Holzkamp, 2013, p. 20), and this always has a communicative dimension (cf. Sevignani, 2019). As cultural beings, we are constantly interpreting the world around us. We do this by working with already created and thus found meanings that are interrelated and construct structures of meaning. I am referring to Critical Psychology, a school of thinking, which developed as an alternative to Freudo-Marxism in Germany and the Scandinavian countries and elaborated on the cultural-historical approach in (Soviet) psychology (e.g., Vygotsky, Leontev, and Luria) to distinguish between two categorial forms of agency: agency can be “restrictive” or “generalizing” (Holzkamp, 2013, p. 23f.). The former is the maintenance of agency by accepting prevailing conditions and making a deal with partial interests. Such restrictive agency externalizes the costs of maintaining one's own agency onto others, who thereby lose agency. Restrictive agency, while subjectively functional, is not sustainable in the long run. This is because the satisfaction of needs is still dependent on the will of others and is therefore fundamentally linked to the fear of a situation in which it is not granted by the powerful. The contradictions of restrictive agency also affect the subjective quality of action. Generalized communicative agency, on the contrary, means not only interpreting or decoding given (hegemonic) meanings, but also creating new structures of meaning, that is, the cultural conditions that determine the realm of what can become meaningful for a subject, in a generalizable way. This cannot be done in private and has a material dimension as it involves creating new or appropriating means of communication and organizing media to cooperatively change not only the material social conditions but also the cultural meanings of those conditions.

In capitalist societies, social conditions are specific as they are intrinsically antagonistic conditions in which the happiness of the strong correlates with the suffering of the weak (Boltanski & Chiapello, 2007, p. 360ff.). If this is so, some can only gain agency at the expense of others. The ideological arises in order to achieve social integration despite these conditions of disrupted commonality, that is, to secure the subjects’ agency on the one hand, but at the same time also the conflict-generating conditions that restrict them. This is, among others, made possible by building a fundamental “competence/incompetence structure” (Haug, 1993, p. 70, cf. Rehmann, 2013, p. 254ff.) into the social interplay of publics. This structure divides citizens and editorial offices and is held together by an over-reaching social imaginary of a “bourgeois public sphere” that is inclusive, domination-free and promises to pacify antagonistic relations. In the editorial offices of the mass media, meanings are constructed by specialists. These instances then provide orientation distinguish by proxy, for example, true from false, legitimate from illegitimate and valuable from worthless. Professional rules, a journalistic habitus and a “journalistic ideology” (Deuze, 2005), distinguish these instances from activists and PR agents. They claim to perform the “bourgeois public sphere” and civic publicity that, as integrative values, function as a kind of “empty signifiers” (cf. Laclau, 2007, 36ff.). By aligning to this counter-factual imaginary, journalists and editorial offices also interpret the community-founding communicative values, such as impartiality, objectivity, general relevance, in an influential way.

The mass media and their editorial offices are not neutral terrains but, to make a concept of critical state theory fruitful for the analysis of the “density and resilience” (Poulantzas) of institutions of the public sphere, should be understood as a “material compression of power relations” (cf. Jessop, 2008, p. 122ff.). Thus, depending on resources and dominant meaning structures, certain world and problem interpretations are not considered legitimate and represented in the media at all, other interpretations are marginally represented, and still others are considered legitimate without question, but only within an interpretive framework that itself favors one position over another (e.g., demanding wage increases is only plausible if profit rates rise). Integration and generalization in the process of hegemony formation can succeed by absorbing and simultaneously subordinating experiences or by refraining from antagonistic private interests. Negt and Kluge (1993) had already objected to Habermas’ original conception of the bourgeois public sphere that a central communicative mechanism of domination in the process of public opinion-formation consists in the refraining from private interests and that it is precisely through this that the communicative construction of an imaginary, but counterfactual, community becomes possible despite the reproduction of antagonistic social relations in capitalist societies. Media systems vary historically and culturally with regard to the proportion of antagonistic and integrating publics and media (cf. Hallin & Mancini, 2014). In Germany, for example, public broadcasting already functions as a counterweight to interest-driven, “divisive” reporting by antagonistic media through its organizational structure.

The general problem, however, is a false, that is, ideological. Generalization is driven by editorial processes, which is made possible by the surrender of communicative competences of social self-association to the editorial tutelage. In this way, subjects gain agency, but at the same time they acquire capacities of communicative subalternity. The scope gained is that of a restrictive agency; it allows them to be informed in everyday life and thus to have at least partial control over the social contexts that affect their own interests. At the same time, however, this goes hand-in-hand with the loss of communicative competences to not only check this information for truth and accuracy, but also to participate in the communicative construction of reality. Under conditions of inequality in antagonistic social relations, such communicative competences cannot develop sufficiently (cf. Bohman, 1990, p. 107). In short, my argument is, being a reader only enables a form of domination that is based on a deficient development of communicative competences. The institutionalized separation of the reception role and the authorial role is and was more problematic than Habermas thinks. Mechanisms of the ideological public sphere indicate that communication and its claims to validity or participation in public spheres are not repressed, but damaged by the mechanisms of compression and compromise formation, because consensus and understanding themselves contribute to shielding problematic antagonistic social relations from being discursively and practically challenged.

In the context of commercial digital platforms, we find similar tendencies as in the pre-digital commercial mass media: unequal attention is reinforced rather than balanced out; attention becomes even more of a commodity because one can not only place advertisements or indirectly create a coverage of one's favor with incentives for journalists, but can also directly place high-reach “sponsored stories”; and possibly a new quality of communicative acceleration with a lower depth of reception—a click is not yet a reception, a like is not yet an argument. However, I would like to point out two important differences: All kinds of media are now accessible on commercial platforms, so the public sphere tends to lose the character of being a space that emerges unavailable between different actors but is increasingly privately curated by digital corporations and thus indirectly controlled. This privileged position enables platforms’ surveillance-based business models and algorithmic recommendation systems based on the evaluation of previous interactions and the accumulated knowledge of users’ preferences. Algorithmic recommendations therefore combine and simultaneously consider the business interests of the platforms and the current preferences of the users. The latter are constantly monitored, with the aim of linking to these preferences and directing attention and (purchase) actions into exploitable, profitable paths (cf. Sevignani, 2022c).

In my view, this means three things for the public sphere. First, to the extent that communication can be channeled even more effectively towards valorization, this denotes a further depoliticization of the public sphere. Second, if similar opinions and interests can be found more easily due to the simple searchability of digital media and the mentioned recommendation systems, this creates opportunities for a partial generalization of experiences and interests in identity-forming sub-publics. This process can, if the partial publics do not turn against each other, have a democratizing effect (cf. Fraser, 1992). The many debates on digital media and the formation of social movements (e.g., Porta, 2022) also tie in with this possibility. Finally, there is the possibility that there will be no effective generalization of experiences and interests if private opinions are networked but, driven by the possibility of finally also being authorized as an author, close themselves off from communicative irritation and thus from opinion formation.

The thesis of a doubling of experiences, as developed in the culture industry analyses of the older Frankfurt School and which Negt and Kluge and Habermas himself in his habilitation thesis took up, could still be relevant in this context. With the important restriction that the effect of doubling experiences is now actively performed and not only receptive. This means that the examination of journalistic “quality” and “truth” is now only undertaken privately in the mode of restrictive agency and with the acceptance of antagonistic social relations rather than in an intersubjective process of understanding with its potentially emancipatory epistemic and praxeological effects.

It is striking that Habermas does not focus on the second option of a possible democratizing function of the collection of private experiences and their discussion that transcends their immediate private character in identity-forming partial publics and short-circuits the formation of digital partial publics with the third problem of networked private opinions. He remains stuck in a scheme in which privatization on the one hand and fragmentation with threatening disintegration of the public sphere on the other are irreversibly bound together. The new platform mediators of the public sphere have (so far at least) no journalistic pretensions: there are no editorial offices where discussions take place and selections are made according to journalistic criteria such as topicality, objectivity, balance, transparency, and careful research. For Habermas, in an editorially unfiltered digital public sphere, citizens do not expect things of general concern to be discussed from different perspectives; rather, they feel encouraged to seek attention for their private experiences and opinions. This can lead to self-affirming and non-discussing closed sub-publics, which Habermas calls semi-public spheres.

As I understand Habermas, the civic refraining from private interests guarantees the unifying element in the public sphere; if this is omitted, the democratic function of the public sphere is threatened for him. Lack of generalization—but not different forms of generalization—is thus the core problem of the semi-public sphere. When he speaks of an educational process in which those who communicate digitally should learn to be authors, just as they once learned to be readers, he presumably means that they should learn to refrain from their private interests and communicate as citizens. But how is the purification of the participatory public sphere from private interests and experiences to be achieved without limiting participation problematic control through editorship? But should we strive for this at all, or should we rather leave the Habermasian framework here? After all, according to what has been said above, an ideological function that secures power lies in the exclusion of private interests and private experiences from the public sphere. Against this background, the threat of mixing private with public is relativized, as is the fear of a fragmentation of the public sphere. The model of the bourgeois public sphere with its problems of communicative competence, lack of experience and aloofness, as well as securing power through consensus and compromise is therefore not a suitable yardstick for evaluating digital transformation. Of course, the question then arises—with all urgency—as to what forms of non-ideological communicative generalization and public learning might look like in digital media environments.