Building and Sustaining Freedom of Expression

The ways in which countries protect and promote freedom of expression differ greatly. However, within formally democratic constitutional systems free speech can be understood to have two basic aspects: non-censorship and diversity of voices (eg Lichtenberg 1990). Here, we examine how the approach to free speech in Sweden contains both aspects. Selected comparisons with the US First Amendment, and German broadcasting law, suggest the value in the Swedish approach but reveal challenges if free speech’s dual aspects are not clearly recognised. Articulating how non-censorship and diversity are part of free speech in Sweden might well prove to be significant, given challenges facing freedom of expression.

Free speech is commonly valued for multiple reasons: it is said to aid in discovering truth or developing knowledge, to serve people’s interest in self-development or autonomy, and to be necessary for democratic forms of self-government (Petäjä 2006). We concentrate here on the third reason. The democratic rationale for free speech is implicit in much journalism and media research, and it is the most frequently considered free speech rationale within law in general (Barendt 2005: 20) and in Sweden (Axberger 1984: 21; Bull 2006: 334).

The democratic rationale for free speech is central to Swedish constitutional provisions; for example, it underlies the “free exchange of opinion and availability of comprehensive information” described in the Freedom of the Press Act – that is, information “from all sides” as the original Swedish words state. This wording illustrates how the democratic rationale involves the idea of diversity. Diverse media content, involving a wide range of subjects, information, genres and viewpoints, allows the public to participate in free and varied debate or observe others engaged in such debate.

Aalberg and Curran 2012:9–12 summarise debates about whether ‘good’ political citizenship requires people be well-informed about contemporary events.

Clearly free speech involves the absence of censorship by the state – or, at least, careful evaluation of any direct attempts to restrict speech by law or executive action. But free speech means more. If the goals that free speech is said to serve are to be plausible, multiple voices are needed (eg Lichtenberg 1990; Gibbons 2012). This view of free speech means that media should be a diverse forum for debate and – through that diversity and the relative absence of censorship – a mechanism for scrutinising the exercise of power (as illustrated in Figure 1). Such accountability is an important implication of the democratic rationale for free speech. Democratic forms of government are commonly said to be aimed at holding power to account, in part through mediated speech. Public speech is not just directed by those with power, it is not only subject to prevailing social powers, but interrogates those with power.

Figure 1

This approach to free speech takes it beyond a bare liberty against government action. Free speech includes obligations for a certain “state of affairs” to be sought. Robert Alexy (2002) uses the concept of a state of affairs in relation to broadcasting freedom (and other constitutional rights in Germany), and it appears useful to adopt here. Free speech involves not just what the state should refrain from doing, but also what the state and others should do to support the environment for speech (Kenyon 2014; Kenyon 2016; Karppinen 2016). The changing and uncertain media environment may increase the challenges for state action to support diverse speech, but it does not remove the value of such efforts.

The state’s enabling function for democratic free speech is highlighted explicitly in Sweden, where state policy has a role in promoting and deepening democracy through safeguarding free speech (SOU 1975: 79, p.126; SOU 1995: 37, p.156) and to create prerequisites for plurality (Dir. 2015: 26, p.2). However, rhetoric about this aspect of free speech may be changing. For example, a 2015 official government report on media policy (SOU 2015: 94, p.37) describes the state’s relationship to free speech in ambiguous terms. It suggests: “Every aspect of state governance can … be interpreted as infringements of these fundamental freedoms” of speech and information, but it also reaffirms the state’s purpose in media policy is “to create as free a society as possible where diversity of media can secure the rights and interests of all citizens”.

Author’s translation, as for most other quotes from Swedish official documents.

There are more ways in which the enabling function can be pursued than for a negative liberty. Not every state action to encourage free speech can be required by a free speech principle; rather, discretion exists for the diversity aspects of free speech in a manner that differs from non-censorship. With such discretion, a question arises as to who exercises it, and how (if at all) it is limited: who has responsibilities for encouraging diverse media speech? Possible actors include the state, market, civil society, journalists and editors. Of course, states are not absent from markets: regulating a commercial market while refraining from directly regulating speech is, in itself, a particular form of state influence on speech. State influence cannot be avoided; the question is whether a particular form of state action is justified, or indeed required, by free speech. While the US state, for example, might generally be understood as “restricted” in its role and European states as “more active”, the state plays “a significant role in shaping the media system in any society” (Hallin & Mancini 2004: 41, 49). (We largely leave aside questions about other scales of organisation, such as sub-national, regional and international organisations and the roles they might play.)

We are interested in matters of non-censorship and diversity as “negative” and “positive” aspects of democratic free speech, and ways in which they might be pursued by state and non-state actors. Such an inquiry is clearly broad. To begin exploring it here, we focus on some actions that might be taken by parts of the state. State actions linked to the idea of non-censorship include protection of media sources, rights of access to public information, and different procedural rules to protect speech in court processes, while notable state actions in support of diversity include public service media and press subsidies. But how do such measures relate to free speech, and what parts of the state would be involved in implementing them – the executive, parliament, courts, regulators? The analysis suggests that if free speech comes to be understood only in terms of non-censorship, there could be dangers in Sweden for the ability of parliament to sustain an architecture for diverse mediated speech. If that happens, courts could act to protect the diversity aspects of free speech as they have done, for example, under the constitutional protection of speech in Germany.

A degree of diversity existed in western media systems in the latter decades of the 20th century. It is not our argument that the historic level of diversity was ideal, but changes to media systems highlight diversity as warranting attention now. In doing so, the goal should be to support diverse public speech, not to support the business models of traditional media.

Several commonly discussed challenges to media systems serving democratic goals are laws that create liability and “chill” public interest speech, media ownership concentration, and commercial pressures that influence media content. Laws restricting speech have long been a concern, and clearly have continuing importance (eg Riga Joint Declaration 2015). But concerns with such laws may be somewhat less in Sweden because, in comparative terms, media speech in Sweden is relatively free of legal restriction (Bull 2006: 335–340). Before considering that, we outline the latter two points concerning media ownership and commercial pressures on content.

Concentration of media ownership is a longstanding concern in many countries. A perceived danger is of fewer owners leading to narrower public debate. As well as fairly stable and (to date) large public service broadcasters financed by licence fees, Swedish media contains many newspapers and commercial broadcasters. However, the multiple commercial media companies have just a handful of owners. In addition, the substantial decline in newspaper advertising has seen many newsrooms close or downsize (Ohlsson & Truedson 2015). Nearly 900 journalists lost their jobs during 2014 and 2015 (Granström 2015c) and news syndication has increased in local newspapers (Byström 2014). Thus, there are concerns about fewer voices and less scrutiny of power by journalists. While networked communications might be thought to change things dramatically, it is notable that a similar style of concern exists over Internet intermediaries because of their potential to influence the visibility or accessibility of online content. Adding to that concern, Internet media has not yet produced a sustainable economic model for producing journalistic content.

Another concern is the blurring of editorial and commercial content (Edström, Kenyon & Svensson 2016). Printed newspapers have seen reduced revenues from both traditional advertising and paid subscribers, which has led to new commercial deals. Major Swedish newspapers have started producing content that is presented like journalism, but paid for by advertisers (Edström 2015; Edström 2016; Piety 2016). The practice follows prominent news providers such as The Guardian and New York Times as well as newer companies like Buzzfeed (Berntsson 2013, Lu 2014). Production can be outsourced to partner-studios where commercial editors produce material for online and branded sections of newspapers. Revenues are not yet substantial but are growing (Hoelzel 2014). There are also new Swedish collaborations such as a telecommunications company, Telia, financing a Silicon Valley correspondent for the tabloid Aftonbladet (Granström 2015b) and an automotive company, Volvo, sponsoring an interview series with business leaders in the newspaper Svenska Dagbladet (Häger 2015). And, while editors do not appear to see problems with the development, journalists and self-regulatory bodies are more concerned (Granström 2015a; Edström 2015). A key concern is the lack of trust in news that may result when audiences can no longer identify who is behind stories (Austin & Newman 2015; Hernius & Rosenlind 2015).

Beyond these concerns, changes in states’ ability to promote public speech due to regional and global trade agreements are also noteworthy. One example is the current EU-US negotiations for a trade and investment agreement – the Transatlantic Trade and Investment Partnership (TTIP). Audiovisual services and the cultural sector have so far been excluded, at least as officially revealed (eg European Commission 2014). But the threats to state action in areas relevant to future communication technologies may be very substantial. Equally, state aid rules within the European Union impose hurdles for state action to promote diverse public speech through supporting public service broadcasting (Gibbons & Humphreys 2012).

The above changes are relevant to the context of the challenges, but they do not remove the value in seeking to sustain free speech’s dual aspects of non-censorship and diversity, something to which Sweden has taken an unusual and noteworthy approach.

The Swedish state plays a significant role in shaping the media system as a fundamental element of democracy. Sweden can be seen within a Nordic tradition of “media welfare states” (Syvertsen et al 2014). The approach has been characterized as a democratic corporatist model (Hallin & Mancini 2004) with the state expected to take steps “to ensure that freedom of expression and freedom of the press are formally and in reality protected by legislation … [and] to create and maintain an information and press system that will accommodate many and diverse voices” (ibid: 160 quoting Gustafsson 1980: 104). The state’s role is not at all intended to amount to state control: democratic media is expected to have independence from state and commercial interests. In Sweden, “news reporting and commentary”

The Swedish term ‘nyhetsförmedling’ does not translate directly into English; encompassing ideas such as news reporting, commentary, news coverage and news gathering.

One of the questions we began with was who has responsibilities for encouraging diverse media speech. In Sweden, the state is clearly one of the responsible actors even if there may be a shift in some public rhetoric about free speech and the state. We have considered how structural positive aspects of free speech have been pursued, in part, by parliamentary and government action in three areas: broadcasting content obligations; public service funding; and press subsidies. We would suggest the basis for action in each instance has been quite consistent with the democratic rationale underlying Swedish free speech, but it has not always been set out in free speech terms. It has instead been a matter for general political processes, on which there has been remarkable agreement to date compared with many other western countries.

However, arguments against those state actions have gained some recognition. And, as media and legal situations change – with greater media commercialisation and further integration of Sweden’s legal system into Europe – there could be real value in considering more fully and explicitly how free speech includes structural positive aspects. In Sweden, public service broadcasting is said to be vital for “free opinion formation”. This parallels the German concept of free individual and public opinion formation, which is the recognised purpose of Germany’s constitutional protection of free speech and broadcasting (ZDF Treaty decision 2014). However, the future of Swedish public service media is a matter for political and public debate, while in Germany, public service broadcasting is also a matter of free speech law in which the Federal Constitutional Court has a decisive role. That court has long emphasised the central position of public service broadcasting in promoting free speech. Unlike Germany, Swedish courts have not yet addressed these issues to any substantial degree. Even though the role of Swedish courts is changing, it may take some time for such arguments to find traction judicially.

Similarly, Swedish press subsidies seek to support “diverse, comprehensive debate” but do not appear to be described explicitly in terms of free speech. The scheme can be seen to arise from free speech’s democratic basis, but it is not overtly understood to be required by it. As with broadcasting, subsidies are a matter for parliament. In part, this reflects longstanding Swedish understanding of the roles of parliament and courts, but the roles are changing somewhat with European integration (eg Valguarnera 2015). As Ingela Wadbring and Mart Ots commented in dissenting in the 2013 press subsidies review discussed above, “the report should have found a basis for subsidies on a more principled and comprehensive democratic basis” (SOU 2013: 66, p.451). Free speech may offer just that principled and comprehensive democratic basis. In the changing situation of media and law in Sweden, it may also offer a particularly useful avenue.

To that end, it is notable that another Swedish cultural support system does more clearly reference free speech. The Swedish Arts Council has a legislative remit that includes safeguarding freedom of expression.

See http://www.kulturradet.se/en/In-English.

As to the possible role of courts, Markku Suksi has recently summarised similarities in Nordic constitutional practices, suggesting there is “a way of thinking” that combines “a cautious form of judicial review and constitutionalism with popular sovereignty and (limited) supremacy of Parliament” (2014: 88 quoting Husa 2002: 185, emphasis in original). Tempering this judicial caution is the constitutional weight of TF and YGL, and the way in which free speech for publications is recognised as central to the country’s constitutional ethos. As Bull has noted (2014: 19), it is as if without press freedom, including access to public documents, Sweden would no longer be Sweden. Even in the nineteenth century, it was thought “that no other constitutional right than ‘the right to print’” (ibid: 18) needed protection, nor was it worthwhile protecting other rights “without the right to print freely” (ibid). Free media would safeguard against all misuse of power. Here, one could cite Germany’s Federal Constitutional Court, which itself quoted a famous US judicial aphorism to state that free speech is “the matrix, the indispensable condition of nearly every other form of freedom” (7 BVerfGE 198, 1958, p.208). There are clear parallels to the importance of media free speech in Sweden, and changes can be seen in Swedish courts’ attitudes. Courts appear to be gradually gaining a role to uphold constitutional rules against parliament. Even so, the reserved role of courts to date is particularly notable. In Sweden, the state is understood to have an obligation to act in support of free speech, but so far it is largely other parts of the state than courts. The value and importance of free speech is understood in a manner equivalent to, for example, Germany – its purpose is to aid “free opinion formation” in both countries and it is a matrix of each country’s constitutional order. But how free speech protection has been pursued in Sweden is remarkably different, and its long term durability is open to question.

To date, Swedish debates about state support for diverse media have not resulted in very substantial reductions in public funding. But that position may change. An expected consequence of lower funding would be less diverse and critical media content, and perhaps reduced public knowledge about public events. Extensive research shows differences in media content between commercially-driven systems and more diverse systems with substantial public media; in addition, research suggests a correlation between strong public service media and public understanding of contemporary events and politics (see discussion in Kenyon 2014). If funding declines, research suggests there is a plausible risk of reduced content diversity and public understanding.

This has a notable implication here: the weight that the Swedish constitutional system places on the political process becomes questionable. Politics will begin to operate in a different context – one with less effective media in terms of content diversity and public knowledge. The degree to which parliament is connected with public concerns, and the degree to which parliamentary processes are communicated publicly, could be expected to lessen. The media would no longer be achieving even its limited historical levels of scrutinising power and publicising political and state actions.

The centrality of diverse public speech to constitutional processes, the way in which it can be seen as necessary structural element within a democratic constitutional system, is an important element of free speech in Sweden. If free speech can be articulated publicly in terms of both non-censorship and diversity, perhaps the Swedish parliament will be able to sustain the structural aspects of democratic free speech. But if parliament proves unable to do that, attention could be paid to how Germany has approached the matter. Closely parallel ideas underlie the Federal Constitutional Court’s support for free speech through public service broadcasting. This structural aspect of free speech is understood to be too important – and too consequential for the operation of democracy – to be left entirely to parliament. The court fears that, without judicial action, democratic legitimacy would be undermined through a political process reshaped by commercially-driven media (eg 57 BVerfGE 295, 1981). The situation is seen very differently than in a traditional Nordic approach, and the example challenges belief in the efficacy of democratic parliaments, at least in relation to free public speech.

All this suggests that, without greater consideration of the free speech arguments implicit in the diversity-pursuing measures outlined above, the Swedish achievement of its particular form of democratic corporatist media may change into a more commercially-driven version of public speech. Then the democratic basis for Swedish free speech would be substantially undermined. Developing the free speech arguments may aid parliamentary processes uphold important structural aspects of free speech – as parliament remains the preferred Swedish actor for such matters – but it would also bring into focus larger questions about the limits of the parliamentary process alone in building a viable system of freedom of expression for the future.