News discourse and readers’ comments: Expanding the range of citizenship positions?

Introduction

In a democracy, representations of agency and power are critical when citizens are called upon to express their views, values and knowledges about controversial policy plans through public consultation exercises. Do citizens’ voices acquire a particular newsworthiness in this kind of situation or do routine news production practices and criteria remain unchanged (Lewis et al., 2005)? In today’s world, the power of media political discourse is no longer restricted to meaning-making processes. Newspapers have the capability of engaging readers through interactive comment forums, a communicative space where citizens can, in principle, exercise power, creating and sustaining an identity as participating members in political processes (Dahlgren, 2009; Jones, 2006). What are the relations between the positions given to citizens in news coverage and the positions taken up by readers when commenting the news? How do readers make use of this potential citizen power?

In this article, we look at news reports on the 2007 Portuguese National Plan for Dams of High Hydro-electric Potential (NPDHHP) and at their subsequent comments. Dams have long been promoted throughout the world as symbols of modernity, prosperity, development and national prestige but have also been contested on a variety of grounds, especially by grassroots groups of dam-affected people and environmental non-governmental organisations (NGOs) (Kaika, 2006; McCully, 2001). Public consultation is normally a legal requirement and so was the case with NPDHHP, a plan that attracted lots of criticisms from different social actors. In processes involving public consultations, media coverage is an important element as it contributes to setting agendas, has the potential to shape who has a say and how, and may constrain the perceived possibilities of participation, either promoting or deterring political engagement (Dahlgren, 2009; Lester, 2010).

Building on constructionist approaches to environmental issues (Hansen, 2015), the sociology of news media, critical news discourse analysis and stancetaking sociolinguistic approaches, we aim to investigate whether readers use the participatory potential of the comments section within the boundaries of the ideological world offered to them by news discourse or whether they go beyond or change it in any way. We claim that critical discourse analysis (CDA) together with other strands of discourse analysis focusing on the semantic and interactional levels of discourse make crucial contributions to understanding these questions.

In the next sections, we present the political background of the case and then review research on media and policy-making, as well as on readers’ comments. After a presentation of the method, we move to a discourse analysis of news texts and readers’ comments.

Political background

The Portuguese government announced the NPDHHP in October 2007, proposing the construction of 10 large dams in the north and centre of the country to respond to the objectives of the national energy policy including ‘guaranteeing the security of energy supply’, ‘promoting competition in the interest of consumers’ and ‘ensuring that the whole energetic process is environmentally sound’ (Instituto da Água, IP, Direcção Geral de Energia e Geologia & Redes Energéticas Nacionais, 2007: 5). It argued that an increase in hydropower production capacity was needed to meet the target for renewable energy agreed with the European Union and evoked Portugal’s commitments to limit greenhouse gas emissions within the Kyoto Protocol.

In the last decade, concerns about energy security and climate change have renewed the popularity of hydropower, now presented as a clean or green low carbon energy source (Ahlers et al., 2015). However, this new hydropower discourse, and the subsequent come-back of large dams in developing and newly dominant economies, has been called into question in recent years. In Portugal, different social groups and bodies contested the need for the NPDHHP’s projected dams arguing for investment in energy efficiency instead, and stressed detrimental environmental, landscape-related and socio-cultural impacts. Importantly, the NPDHHP was subjected to a Strategic Environmental Assessment, which was supposed to assess ‘alternative visions and development perspectives (…), ensuring the global integration of biophysical, economic, social and political considerations’ (Decree-Law 232/2007, June 15). Since it was the first time this kind of instrument was applied, it could be expected that it would boost debate and public participation. Following public consultation, 10 dams were approved for construction with concessions for seven going to the largest electricity companies in Portugal and Spain: EDP, Iberdrola and Endesa. A new government came into power in November 2015, and cancelled two of the dam projects and suspended a third one. The remaining projects were at different stages of execution at the time of writing. Ever since it was planned, the construction of the Foz Tua dam led to most contestation from environmental NGOs, political parties and citizens. It saw several delays and changes resulting, in part, from initiatives taken by its opponents, including public petitions and campaigns (e.g. http://ultimoanodotua.pt). By June 2016, its construction was mostly finalised.

It is thus clear that hydropower may constitute a ‘socio-technical controversy’ and that, like other socio-environmental issues, it ‘combines different interests, involves different regimes of justification and mixes scientific, political and ethical arguments’ (Aykut et al., 2012: 4).

Media discourse and policy-making

The media can play a key role in discursive legitimacy struggles by providing an arena where governments, institutions and non-state entities, such as interest groups, engage in self-justificatory practices. In this ideological space, they can operate more or less insulated from criticism and discredit, which is fundamental to preserve their freedom of action (Breeze, 2012). News media exercise significant power by steering legitimacy struggles and, concomitantly, public support for certain visions, actions or policies. The discourses of governments, political parties and other political actors invested with formal authority and power are often the primary focus of media attention. They also act as gatekeepers for much of the ‘factual’ information circulating in a context of participatory policy-making. This reinforces their particular power position vis-a-vis other actors.

Investigating media representations helps assessing who is an authorised participant in the debate, the hierarchisation and framing of political actors’ voices, and the social and persuasive functions of those choices, that is, how they work to reinforce the legitimacy of some actors and their discourses, while delegitimising others and thereby position the reader to regard them as more credible and trustworthy or less so (Fairclough, 1995).

By controlling the access and presence of civic actors and ordinary citizens in public debates, as well as the range of positions given to them, news media discourse is crucial in shaping the relative power position that both have in participatory policy-making processes. When recognised as legitimated sources or as authorities on the matter at hand, civic actors and ordinary citizens are given a better chance to exert influence over political institutions. Furthermore, news media discourse has the potential to shape and influence possibilities for public participation (Sjölander and Jönsson, 2012) by adopting a rhetoric that appeals to readers’ sense of political efficacy (Coleman et al., 2008) encouraging participation. It can represent the populations affected by public policies as agents of influence or change, as active and engaged rather than as individualised victims or subjects worth of consideration or commiseration (Gamson, 2001: 61).

Readers’ comments and citizenship positions

The implementation of Web 2.0 technologies in online newspapers, allowing readers to add their voice to journalistic discourse in commentary and debate spaces, creates a new channel for the visible (and public) discursive processing of news issues (Weber, 2014). Comments are normally published via an online form with minimal censorship and usually do not require commenters to use their real name.

Empirical research exploring the influence of journalistic content on comment practices shows that the most commented news stories are more likely to be about public affairs issues, especially politics (Boczkowski and Mitchelstein, 2012; Tenenboim and Cohen, 2015). It also shows that commenters tend to focus on the subject of the story, or on the social actors mentioned, or on the broader theme to which the article relates (Díaz-Noci et al., 2010).

The bulk of scholarly attention on readers’ comments is found in studies on participatory journalism and focuses on whether comments sections open up an opportunity for promoting public deliberation and/or on its effects in journalistic culture, values and practices (Domingo et al., 2008; Ihlebaek and Krumsvick, 2015; Silva, 2013). There is still much controversy about the deliberative potential of such conversations (Strandberg and Berg, 2013), especially when their quality is measured according to normative standards of truly deliberative discussions (Dahlberg, 2004). Scholars such as Ruiz et al. (2011) argue that readers’ comments are not fulfilling those ideals, while others claim that by including issues, perspectives and opinions not considered in newspapers’ discourse, these micro-forums enrich public debate on issues of collective interest, and their use by newspapers is in itself a recognition of the potential of readers to contribute to the discussion of public issues (Manosevitch and Walker, 2009: 6).

Little attention has been paid to the interaction between news discourse and readers’ comments, and to their conjoint action in the communicative construction of citizenship (Hausendorf and Bora, 2006). Underlying the analysis of this dynamic lays a double-sided concept of citizenship: citizenship as a social location that is represented by others different than citizens themselves (in this case, representations offered by news discourse), and citizenship as positioning or, in other words, as a complex game of social positions discursively constructed and displayed to others by citizens themselves in particular discursive situations (Fairclough et al., 2006; Phillips et al., 2012: 4–8). In both senses, that is, citizenship as an available discursive position resulting from social struggles, and citizenship as a rhetorical achievement (an outcome of interactional negotiation), the resulting citizen identity must be seen as a contingent, precarious, fluid and dynamic social and cultural phenomenon emerging in discourse contexts and therefore subject to power games.

By focusing on the relations between the positions given to citizens in news discourse and the positions taken up by readers when commenting the news and on the participatory potential of this citizenship co-construction, we contribute to fill the mentioned research gap. We examine the possibility of readers’ comments functioning as a parallel space for debate and ‘virtual’ deliberation, and for negotiation of meanings and identities associated with citizenship: what being a citizen in current democracies means in terms of contributing to political decision-making, who and how takes up that position, and what is the scope of their expected influence.

Method and materials

The combination of CDA, a multilevel and multilayered approach to study text in context (Wodak, 2011), and strands of discourse analysis focusing on positioning and stancetaking (Du Bois, 2007; Thurlow and Mroczek, 2011) provides a valuable framework for analysing the interaction between news texts and readers’ comments.

Data were collected from the four (paid) Portuguese newspapers with highest circulation – Expresso, Público (quality papers), Jornal de Notícias (mid-market) and Correio da Manhã (popular paper) – starting 2 months before the public consultation on the NPDHHP (1 October–13 November 2007) and extending beyond the governmental approval of its final version (31 December 2007), in order to catch both expectations and reactions to the programme. We used the newspapers’ search engines with the terms ‘Plano Nacional de Barragens’, ‘Programa Nacional de Barragens’ (respectively National Dams Plan and National Dams Programme, which were used interchangeably) and ‘PNBEPH’ (Portuguese acronym of the programme). A first reading of the data led to selection of 23 articles (12 in Público, 7 in Correio da Manhã, 3 in Jornal de Notícias and 3 in Expresso) and of 121 readers’ comments (75 in Público and 46 in Correio da Manhã).

All newspapers chose to cover the issue using mainly the so-called hard news genre, focusing on ‘just the facts’ (Cotter, 2010: 145). CDA teaches us that each media text is both a mode of representation, constructing social reality and mediating the world, and a mode of interaction constitutive of the social identities of those involved in a communication act as well of their social relationships (Fairclough, 1995).

In a first approach to the semantics of news texts, we identified the main topics and controversies across the articles. We also mapped references to citizens, both explicit and implicit, and their place in news stories, and examined their linguistic formulation and potential to promote or hinder participation. The presence of signs indexing the journalistic construction of a discursive conflict between selected news sources across the corpus led us to consider the rhetorical dimension of news.

The rhetorical value of the headline is well demonstrated in the literature. Headlines are the most prominent elements of news reports, and together with the lead, they function as a summary of the report. Clearly signalling expressions of major topics, headlines define what is subjectively seen as most important in a news event (as represented by such reports) and point a preferred general meaning of the text to readers (Van Dijk, 1988). Through a detailed analysis of the framing of headlines (semantic and syntactic characteristics – Fowler, 1991; Van Leeuwen, 2008), we were able to account for the role and position attributed to social agents in the ongoing discursive conflict, and for their persuasive and social functions.

In the analysis of readers’ comments, we explored the interactional nature of stancetaking in terms of links between citizens’ positioning and news discourse, as well as in relation to other comments in the thread. We also investigated how positionings are persuasively communicated and their social functioning, that is, their inscription in or indexing of wider citizenship discourses or ideologies. Stance, a notion central to our analysis, can be defined as

a public act by a social actor, achieved dialogically through overt communicative means, through which social actors simultaneously evaluate objects, position subjects (themselves and others) and align with other subjects, with respect to any salient dimension of the sociocultural field. (Du Bois, 2007: 163)

Dialogicality makes its presence felt to the extent that a stancetaker’s words echo previous interlocutors and their stances or those of distal ones. Alignment, seen as the act of calibrating the relationship between two stances, and by implication between two stancetakers, takes place on a continuum with both positive and negative poles, and it is accomplished both explicitly and implicitly. The role of implicit stance alignment is especially important in the strategic management of intersubjectivity and interaction.

We examined the links between each comment section and associated news text regarding overall themes, stance orientation, argumentative alignment and ‘participation framework’ (Goffman, 1981). ¹ The emergence of citizens’ positionings was analysed at the local interactional level (Kleinke, 2010) on the basis of discursive work done by commenters concerning intersubjective positioning, that is, their self and other social positioning concerning rights, duties and expectations as members of a larger political community. The focus was on interactional strategies (to impart/change knowledge, opinions, emotions and power relations) used in stancetaking. These practices are indexed either by surface resources (lexical and syntactical variation, for example, evaluative words, personal pronouns, modal verbs and auxiliaries) or implicit ones (presuppositions, implications, concessions, contrasts, mitigation, polarisation, etc.) (White, 2003).

Analysis

In this section, we explore journalistic representations and communicated citizenships through an analysis of news texts and readers’ comments.

Citizens in news discourse

Two research questions guided our analysis of news texts: How did newspapers represent political actors’ discourses and what were the social and persuasive functions of those choices? How was citizen participation in the consultation process implicitly or explicitly signified? A detailed analysis of news headlines was combined with an analysis of stories’ main topics and references to citizens. Due to space limitations, we privilege the former here and do not discuss differences between newspapers.

We have grouped the topics in three major thematic classes, which summarise how the press defined the NPDHHP’s situation: the NPDHHP as verbal struggle (11 reports), as matter of fact (7 reports) and as matter of enactions (5 reports). Under the representation of the NPDHHP as a verbal struggle, the decision to build more dams emerged as a controversial policy issue with two clear poles, against and in favour. Citizen voices and their verbal activities were included in the lead of one story with a local angle (‘Dam will drown Tua line’, Correio da Manhã, 08/10/2007). Following standard news writing conventions (Cotter, 2010: 120), its lead conveyed the statements of a person who felt negatively affected by the dams plan and who could thus illustrate its implications. Described as ‘a member of the public’ (‘um popular’ – slightly derogatory expression) from Mirandela, this ‘60 year old’ is attributed direct quotations (e.g. ‘the government does not know yet what the people from Trás-os-Montes are capable of when they are angry’). His words are qualified as ‘pointing the finger to politics’ and as ‘accusations’. There were two other references, brought to the discussion via environmental organisations (through direct and indirect quotes). In a co-text of criticisms and accusations, citizens were described as a collective (‘public opinion’; ‘population’) that was ‘off guard’ or unprepared, was ‘caught by’ political manoeuvres and badly informed by the government about the dams’ environmental impact. Citizens were thus constructed as subjects of commiseration.

When represented as a matter of fact, the NPDHHP was defined not as a future reality but as a present and ongoing one. What the government said took centre stage. The presence of citizens and their actions could be inferred at the bottom of two stories that referred to the government’s plan to open ‘a public discussion’. Citizens were also indirectly mentioned in the only story about the consultation process through information attributed to a public official. Their participation in the consultation process was represented as an event passively experienced by unspecified official agents and as abstracted positive or negative reactions to the construction of specific dams, as in ‘Construction of Tua dam received 28 unfavourable submissions’, Expresso, 07/12/2007 (emphasis in headlines is ours).

Representations of the NPDHHP as a matter of enactions suggested that it was amenable to change, a policy decision under pressure to be halted or changed. Surprisingly, these news accounts did not include any reference to citizens or to their attempts to influence other citizens or relevant authorities to modify the plan.

The identification of the voices mentioned in headlines was our next step. Twenty-one headlines included an explicit reference to social actors. Seven included public authorities and six the Green Party. In the other eight headlines, the press referred to NGOs/environmental groups (two headlines); mayors (two headlines); EDP – a major player in the energy industry (two headlines); and a hydrobiology ‘expert’ (one headline). A range of relevant social actors to the debate on this policy proposal, such as social movements, local communities and, more generally, directly affected populations, were not included in the headlines.

Subsequently, we analysed how actors were described. Newspapers only individualised government-related actors (Manuel Pinho – Minister of Economy, José Sócrates – Prime Minister), thus picturing leaders as strong individuals while other voices were mostly collectivised (‘the Greens’, ‘EDP’). Professional groups and NGOs were also categorised in terms of their social goals (‘GEOTA’s environmentalists’; ‘environmental organizations’).

We then considered the roles that actors were made to play in headlines. Actors were topicalised in initial positions and endowed with an active semantic role in 15 out of 23 headlines where they were represented as an active, dynamic force in a verbal process. But the semantic nature of the verbal process that was attributed to them varied. Governmental actors were made to play an agentive role in two kinds of verbal processes: as protagonists bringing forward a standpoint (e.g. ‘José Sócrates defends new national dams plan’, Público, 04/10/2007) and as information providers (e.g. ‘Government announces location of ten new dams’, Correio da Manhã, 04/10/2007). The Green Party was attributed an agentive role as seeking influence in the international arena in three headlines (e.g. ‘Greens appeal to UNESCO to prevent construction of Tua dam’, Jornal de Notícias, Público, Expresso, 04/11/2007), in providing information in one headline (‘Greens say they will do everything to have debate in São Bento [Parliament] on national dams plan’, Público, 20/12/2007), and in attacking the Socialist (governing) party in one headline (‘Greens accuse PS of helping government evade debate on National Dams Plan’, Expresso, 20/12/2007). Environmental NGOs were portrayed in an agentive role in their opposition to and criticism of the plan (e.g. ‘GEOTA environmentalists against dams plan for energy production’, Público, 04/11/2007). EDP was given an agentive role in attempting to influence the government (‘EDP asks for preference for five dams’, Público, 12/12/2007), whereas the ‘biology expert’ and the mayor from Castelo Rodrigo got agentive roles in their attacks of the plan (‘Hydrobiology expert contests “explosion” of dams in northeast’, Público, 04/10/2007; ‘Castelo Rodrigo mayor insists on a dam in Coa’, Público, 06/10/2007).

This systematic description of headlines shows that the press gave priority not to the government’s discourse, but to the discourses of other elites. Does this weightier presence mean that the press represented those actors and their discourse as more powerful or legitimate than the government’s? The answer is no. First, because environmental NGOs/groups/experts and mayors, when given an agentive role, were mostly collectivised and attributed only a confrontational stance, which contributed to portray them/their discourse as a homogenous blocking force that had nothing to offer but resistance or refusal of the dams plan. This foregrounding functioned to discredit them, as in the mind of some readers they may sound too extreme, and depoliticise their counter discourse, because it made it seem pointless since devoid of solutions. In fact, their alternative counter-argumentation in favour of a different policy – to invest more in energy efficiency and energy measures – mentioned in the body of two news stories, was never included in the headlines, thus indirectly contributing to reinforce the power of the government’s discourse. Second, because within the temporal line of the news coverage, the journalistic choice of framing governmental voices as information providers meant that their dialogicality was not recognised, or at least was backgrounded. While it can be argued that in this kind of political situation governmental officials are expected to be active in providing information on policy plans, the journalistic holding of their propositions as factual givens, instead of factual reasons adduced to justify their standpoints, positioned readers to view them as highly warrantable and thus credible. In contrast, the fact that NGOs/green groups/experts and mayors were never given an agentive role as information providers reinforced the partisan nature of their discourse and may thus have worked to disqualify them by making their discourse seem less credible, reliable and legitimate.

From the above, we conclude that the press created a ladder of power positions, where official actors and their (mostly) semiotic actions were at the top of the political pyramid, their legitimate opponents and their discourses were deprived of symbolical efficacy, and citizens were prevented from entering the legitimate political order of discourse as citizens, that is, as agents actively engaged in the public sphere, in civic politics or institutionalised politics, let alone partaking in decision-making within the political domain.

Citizen positionings in readers’ comments

To what extent did readers, when commenting news pieces, act within the terms circumscribed by newspaper discourse? Did this interaction enlarge the scope of citizenship communicated in the public debate on the NPDHHP?

An extensive analysis of all the comments showed several tendencies. We will start by summarising them and then move to a detailed analysis of illustrative examples. First, we found nearly no debate of journalism itself, suggesting that most readers do not position themselves as critical actors or active citizens towards the newspapers. Only a small share of comments (3) made an explicit reference to the quality of the news article, either to correct factual information or to criticise the spelling. The points of contention were the actors, positions and facts reported on, rather than the journalist or the newspaper and their authority, credibility and trustworthiness. This both echoes findings of previous studies (Borton, 2013; Freund, 2011) and differs from them (Neurauter-Kessels, 2011). We can infer that the fact that newspapers often reproduced the government’s discourse and did not create openings for, let alone promote, participation in the public consultation was not met with a critical reading by commenters.

Second, the ‘participation framework’ in our data does not resemble the classic dyadic scheme definable in terms of exchanges between two participants (or user to user). Participants aimed their comments at different social targets: (1) news actors, which were absent participants in the context (Van Dijk, 2008) but constituted the comments’ main referents and most frequent target: although they did not belong to the pragmatic context, they were indirectly, socially addressed; (2) other posters; and (3) actual, potential, imagined or ideal readers envisioned by the poster. In contrast with previous studies, our dataset shows that posters are not simply making unilateral or ‘declarative’ comments, but engage in intersubjective positioning and social interaction (Langlotz and Locher, 2012). Although the majority of posts remained isolated observations, posters did take up various discursive and social roles, had different addressed/intended targets and invested in negotiating relationships with others. It thus seems reductionist to claim that commenters use online platforms just to ‘amplify’ their voice.

Third, we found a coherence link between each news text and the respective comments, as well as within each comment set, which was manifested through various linguistic features. ² The stancetaking process within this semantic co-text and participation framework was not a linear or univocal process. It was often like a dance, made of composite or fluctuating moves. As the analysis below will show, alignments and dealignments may be accomplished simultaneously, both directly/explicitly and indirectly/explicitly. Furthermore, there are comment sets in which posters did not start a discussion or debate, that is, that are not organised around a point of contention. This means that dissent and disagreement, typically associated with this kind of online interaction, does not preclude the occurrence of instances of mutual reinforcement or cooperation, suggesting that emotions/feelings of solidarity or communality might also be relevant motivations for expressing opinions in this kind of online forums.

Fourth, stancetaking may be realised through various kinds of speech acts. For instance, expressions of dissent/disagreement or dealignment were operationalised through different means, by foregrounding a conflictual stance, mitigating confrontation or foregrounding cooperation. This implies that questions of style and of rhetoric, and not only of content, should be relevant for discussions around the democratic deliberative nature of such discussions.

Fifth, a distinction must be drawn between text function or discourse doings – what is this commenter doing in this post? interactional function – how does discourse affect actual or intended targets? and social function – how does it implicate particular identities, roles and relationships between participants? Most remarkable in our sample were the interactional strategies used to engage other readers and alternative perspectives. Commentators showed commitment to the interaction itself, negotiated potentially conflictive passages and engaged in alliance building. Those moves indicate awareness of the collective character of the discussion – and of the risk of facing opposition and criticisms – but also imply respect for others and the need of having their views recognised by them. We found instances of intergroup rudeness (Kleinke and Bös, 2015) that included the use of identity markers based on regional or ideological belongings with an insulting or offensive effect. But those instances of flaming were not dominant.

These semantic, interactive and stancetaking properties of our dataset reveal the complexity of a detailed analysis of stancetaking in readers’ comments. Conceiving, as we do, citizenship as a process and as mode of engaging others (Asen, 2004) – whether in the immediate context or in a more distal one – means that it is more relevant to analyse the ‘how’ than the ‘what’ of discourse. Below we look at selected cases in detail.

We focus on a set of comments where posters explicitly positioned themselves and ‘intended’ readers as part of a larger community of interests (or citizenry) that is being acted upon and is reacting in response, that is, as part of a political community. Two kinds of citizen subject positions are present: (1) the bounded citizen, the most common positioning, which expresses citizenship in the form of the claim to the right to have a say on decision-making but presupposes acceptance of a passive position towards decision-makers: it may involve defensive resistance and critical vigilance but betrays a hierarchical view of political authority and power – this positioning works in tandem or aligns with the kind of citizenship communicated by the news stories; (b) the agentive citizen, a positioning that foregrounds the possibility of intervening in the shaping of policies, and encourages a sense of collective agency and a proactive stance in policy-making – this positioning dealigns with, and in some ways is an alternative to the one advanced by the news texts.

What follows are illustrative cases of each category, preceded by a short summary of our analysis of the comment set or sets in which they occur: identification of main topics, main orientation (positive or negative), main argumentative alignments, targets of contributions and discursive doings.

Conclusion

We investigated the relationship between the positions attributed to citizens in news media and the positions that readers took up when commenting news texts, as well as the potential role of this discursive interaction in fostering public engagement in consultation processes. At stake was the participatory potential of each discourse and especially of their ‘joint activity’. The data came from the coverage of a hydroelectric power plan in Portugal.

We have shown, perhaps unsurprisingly, that rather than debating the quality of the governmental proposal and opening it up to a society-wide scrutiny and critical questioning, the press offered their readers a narrow and simplistic map of the political process where only elite actors mattered and where citizen participation was reduced to reified objectivations or individual emotional stances, a representation that downgrades any sense of agency and signals lack of power. Consequently, the potential relevance of citizen participation in the public consultation (as well as in other forms of public debate) was undermined. Whereas it could ideally contribute to enhancing democratisation in policy-making, the press foreclosed debate and the participation of citizens at large.

Within the parameters of online user-generated comments and the political ordering set up by the news, readers redefined the role of citizens within the political process by acting upon that order as concerned members of a common political unity engaged in discussing policy issues, voicing stances, expressing emotions and interacting with each other, building oppositional and solidarity alliances among themselves, political actors and larger audiences.

In this situated struggle, citizenship discourses were used as resources for stancetaking and intersubjective positioning with different rhetorical purposes. They were used to assign oneself and others strong affective bonds based on common negative expectations or political mistrust towards the actions of governments or state actors, shared responsibilities (keeping a watch on governments’ actions), purposes (the progress of the community), duties (respecting the common good and general will), affiliations (nationhood), shared rights (claiming for justice and fair treatment), and shared causes (the country’s best interest). In this sense, citizenship discourses worked as an implicit normative framework structuring the stancetaking process, citizen positionings and how these actors were to govern themselves and each other regarding political decision-makers.

We have distinguished the citizenship styles of the bounded citizen and the agentive citizen, which involve different power positions. Bounded citizens attribute stances to themselves and others that cast them as outsiders in policy decision-making, thus reinforcing traditional power hierarchies of democratic citizenships. Agentive citizens’ stancetaking constitutes them and others into agents who act together in the name of the country’s best interests in order to change policy decisions, thus creating a scene of political participation on their own terms. The first style was the most common one in our data, meaning that in their local rehearsal of power positions, commenters exercised their citizenship within a conventionalised political democratic arena sustained by traditional citizenship regimes. In this sense, the political ordering of news discourse does seem to influence readers’ participatory acts.

Regarding the public consultation on the dams policy plan, we can conclude that the ways in which readers’ comments related to news discourse expanded the public debate by widening the public agenda and the possibilities of participation for interested citizens, notwithstanding the apparent irrelevance of the issue in both discourses. Whether this in fact had real or indirect effects on other forms of collective action to influence the NPDHHP and over policy outcomes remains an unanswered question.

In sum, this article helps filling research gaps regarding the interaction between news discourse and readers’ comments (Koteyko et al., 2013; Laslo et al., 2011) in the communicative construction of citizenship and its participatory potentials. As outlined here, CDA and interaction-oriented stance research offer a way of conceptualising the relationship between news discourse and readers’ comments not in terms of a deterministic unilateral influence, but as a process of resonances, appropriations and enactments, which is multivocal by nature, thus opening a new range of discursive possibilities. Our proposal to focus on the semantic and interactional dimensions of both discourses, and on their social functioning, showed their intimate connections at the levels of both knowledge production and stance alignments and orientations. This means that scholars interested in the participatory potential of readers’ comments would benefit from taking into account the news discourse that originates them, as well as the work readers do to engage their audiences, something that has been neglected in the literature due to its specific focus on commenter-to-commenter interaction. The persuasive nature of comments, the local power struggles and alliances at stake, and their potential to fuel citizenship-related sentiments seem to be worthy of scholars’ attention.