The Challenge of the Digital Public Sphere: Finnish Experiences of the Role of Social Media in Participatory Planning

Abstract In this paper, we examine social media-based participation and public deliberation in land-use planning. We use the Deweyan theory of the public, the Habermasian theory of the public sphere, and the recent theories of the digital public sphere as our framework, asking what should be the relation of public planners to the digital public sphere: should they try to manage self-organising participation, or should public opinion formation be free from the influence of public authorities? The empirical part of the study reflects on this question by investigating Finnish planners’ experiences of the role of social media in planning in the light of two recent surveys.


Introduction
Public participation has transformed urban planning in various countries during the last fifty years. The breakthrough of the internet during the last three decades, in turn, has transformed the nature of participation. This has been the case especially in Finland, the contextual locus of this study, and one of the leading countries in the world in utilising e-participation methods and tools as a part of a broader array of e-governance practices (United Nations, 2020).
Many planning scholars have argued that the internet, as a channel of participation, increases the amount of participation and enhances its quality (e.g. Fredericks & Foth, 2013;Kahila-Tani et al., 2019;Staffans et al., 2020). The internet facilitates the distribution of information and makes planning processes more transparent. Moreover, it provides an easy-access two-way channel, allowing not only the distribution of information coming from planners but also input from the public. Nonetheless, the internet is not only two-way but also a many-to-many channel that can host discussions and information exchange both vertically between parties such as planning organisations and the public, and horizontally between individual representatives of these and many other parties (cf. Hacker, 1996).
In what follows, we focus on the role of social media in planning. Social media is characterised precisely by its supportiveness of many-to-many communication. Unlike in traditional participation designed and organised by public officials, in social media the users themselves orchestrate the themes and contents to be discussed, thus posing a novel challenge to planners (Nummi, 2018b(Nummi, , 2019Williamson & Ruming, 2020;Sj€ oblom & Niitamo, 2020). We examine social media-based participation in the framework of the "digital public sphere" (see e.g. Sch€ afer, 2015;Parnes, 2016;Mazzoleni, 2015;Sousa et al., 2013), though acknowledging that the perspective of "the public sphere" is not the only relevant perspective on social media. While social media platforms provide planners with a unique window into people's private lives and the specific values and preferences of local communities (e.g. Afzalan & Evans-Cowley, 2015), the theoretical framework of the digital public sphere broadens the perspective beyond the personal and the local. Given that social media platforms can reach even global publics, they could be expected to enable deliberation not only on local matters but also matters of wider public interest.
While deliberating publics were discussed already in the first half of the 20th century by American pragmatists, especially by John Dewey (1927), the concept of the public sphere was made famous by the European philosopher and sociologist J€ urgen Habermas (1991Habermas ( , 1996, a major source of inspiration for the theories of communicative planning alongside Dewey and the pragmatist tradition. For Habermas, the public sphere was a sphere that emerged out of civil society during the early days of liberal capitalism in several European countries. It was also a sphere where citizens ideally left behind their positions as private people to deliberate on matters of public interest freely and disinterestedly, being thus able to mediate between the civil society and the state (Habermas, 1991). Unlike Habermas's abstract and philosophical theory of communicative action, his historically grounded concept of the public sphere has been only seldom discussed by the theorists of communicative planning. One reason for this might be that, unlike Dewey who focused especially on deliberating local publics, Habermas discusses the concept of the public sphere mainly in a nation-state scale.
Habermas's early works on the public sphere focus not only on the ideals of deliberation, but also on the systemic distortions of communication in late-capitalist societies. The discussion on the digital public sphere builds largely on Habermas's analysis of the origins of these distortions. Distortions are due to the increasing role of various gatekeepers of public discussion in the latecapitalist societies, for instance, commercialised mass media, but also the state and its administrative organs (Habermas, 1991). As the theorists of the digital public sphere have noted, the emergence of the internet a few decades ago re-introduced gatekeeper-free discussion fora that could potentially facilitate the realisation of various ideals present in Habermas's theory of the public spherehence the concept of the digital public sphere. Yet, the free and unconstrained nature of the digital public sphere paradoxically makes it also vulnerable to manipulation of opinion-formation processes, for instance through systematic spreading of misinformation.
Our theoretical research questions concern the dilemma that the theories of public sphere pose to participatory planning. On one hand, the Habermasian conception of the public sphere suggests that the state and its organsplanning organisations includedshould maintain a distance from the public sphere so that it can retain its character as an unconstrained space for deliberation. On the other hand, given that the public sphere is prone to various kinds of distortions even without the involvement of the public authorities, it makes sense to assume that public interventions are sometimes needed to prevent distortions (cf. Porwol et al., 2014). The Deweyan approach to public deliberation particularly suggests that public officials stand on the side of the public, and rather than being a source of distortions, they can eliminate the potential distortions in communication.
We discuss this dilemma in light of two recent surveys on the relation between Finnish planning officials to social media-based participation (Nummi, 2018a(Nummi, , 2019. We will investigate, in particular, what is the relation of the planners to the online public sphere and social mediabased planning discussions. We ask whether the planners recognise the online public sphere as one of deliberation on matters of public interest, and how do they see their own roles vis-a-vis such a public sphere.

Participatory Planning and its Problematic Relation to the Public Sphere
Public participation has been an oft-discussed topic especially in the context of communicative (e.g. Forester, 1989Forester, , 1993Sager, 1994) or collaborative planning theories (Healey 1992(Healey , 1997Booher & Innes, 2002;Innes & Booher, 1999. These theories state that the input from the public makes planning decisions more informed (see especially Healey, 1992Healey, , 1997Innes, 1998). Moreover, they argue that public participation makes planning more democratic, as it allows the public to monitor and participate in the making of normative decisions (e.g. Healey, 1997;Forester, 1989Forester, , 1993Sager, 1994). Finally, these theories suggest that through inclusive dialogue, planning discourses need not be only about competition between often conflicting interests, but that the participants could be expected to reflectively modify and change their views and interests (see especially Healey, 1992Healey, , 1997Sager, 1994).
J€ urgen Habermas's Theory of Communicative Action (Habermas, 1984(Habermas, , 1987 provided for many pioneers of communicative planning an idealised model of rational, agreement-oriented public deliberation (Forester, 1989(Forester, , 1993Healey, 1997;Sager, 1994). However, many of these "pioneers" relied also on the pragmatist tradition in philosophy, especially on John Dewey's thought which provides a more practice-oriented approach to the deliberative publics when compared to the Habermasian approach (e.g. Hoch, 1984a, 1984b, Healey, 2009). Nonetheless, these traditions have much in common, given especially that Habermas builds partly on Dewey. Both theories share the belief that the public is capable of making reasoned judgments about public affairs. In this, they differ from the mainstream sociological theories of democracy of the 19th and 20th centuries such as Weber's and Schumpeter's, where the stability of the complex industrialised societies is guaranteed by "democratic elitism," a model where the public chooses the elites who represent them in decision-making, and where the administrative elites implement the decisions, while public participation as such is minimised (Whipple, 2005).
However, there are also differences between Deweyan and Habermasian thought, many of which are relevant as regards deliberation in traditional and digital public spheres. Dewey's project has been credited for being more critical than Habermas's, at least in one important respect: Even though Dewey favoured cooperative approaches, he was not focused on the ideal of consensus-formation to the same extent as Habermas, but he emphasised also those moments where pre-established hegemonic consensuses are reflectively and creatively revised (Whipple, 2005). For Whipple (2005) this feature suggests that Dewey's thought aligns with contemporary agonistically-oriented critics of Habermas who value disagreement because of its potential in countering hegemonic political projects on a larger societal scale. Both Dewey's and Habermas's projects are critical towards the market domination in public discourse. However, Dewey in his time perhaps could not be as critical towards public administration as Habermas was later. Habermas brought to the fore the power-related systematic communication distortions in late-capitalist societies, where not only market institutions but also public administrationdeeply intertwined with the markets as it ismay compromise the rational quality of public deliberation (Antonio & Kellner, 1992). We are interested especially in Habermas's argument, thatunlike Dewey (1927) positedpublic administration may be a source of systematic communicative distortions in public deliberation. Habermas's Marxistinspired theory discussed late-capitalist societies, where the public has degenerated into masses that can be manipulated, not only through entertainment originating from commercialised mass media, but also via the manipulative force of political propaganda originating from the late-capitalist states. This theme is present especially in Habermas's The Structural Transformation of the Public Sphere (published originally in 1962 in German), where he claimed that the public sphere of early liberal capitalism declined soon after its emergence because of the deepening intertwinement of the state and the private sphere of economy. In particular, the distribution of economic power became increasingly uneven, and the powerful economic actors became directly involved in the public decision-making without the mediation of the public sphere (Habermas, 1991). The state and its organs became also increasingly involved in the private sphere of economy, and they increasingly aimed at buying public acceptance for their activities with marketing strategies and public relations work instead of subjugating these activities to rational-critical public scrutiny (Habermas, 1991).
Habermas's argumentation has given reasons for planning theorists such as Huxley (2000) to argue that a truly Habermasian communicative planning theory would not grant public planning organisations the role of managing participatory processes, but that critical participation should emerge out of the civil society without the potentially corruptive influence of the state's administrative organs. The problem is, however, that communication distortions can also originate from civil society, as Habermas (1991) shows. Dewey's (1927) insistence that the state and its officials should rather adopt an organising role in the public deliberative processes than maintain a distance from the public, is therefore also warranted.
Another interesting difference between Habermas and Dewey is that whereas Habermas concentrates on post-conventional moral reasoning, which requires abstraction from the concrete and situated daily practices (Antonio & Kellner, 1992), planning theorists such as Healey (2009) have emphasised the relevance of Deweyan thought because of its emphasis on the situated community values. This dismissal of the moral relevance of situatedness seems to be typical of both Habermas's theory of communicative action (Habermas, 1984(Habermas, , 1987 and his early theory of the public sphere (Habermas 1991) where the lines between private and public spheres are sharp. Feminist theorists have argued that this has resulted in Habermas's failure to notice how, for instance, some of women's private, embodied and situated experiences have been publicly and politically relevant (Mattila, 2020). Dewey, in turn, has been criticised for being preoccupied with the small-scale, situated and embodied reasoning, and not being capable of looking beyond community-based values and visioning how broader-scale public deliberation might work in modern complex societies (Whipple, 2005). In this article, we build on the premise that planners should not only be sensitive to local values and traditions, but also recognise when these values and traditions need re-evaluation from a more abstract and universalising perspective, for example, to do justice to the rights of minority groups in local communities. Digital public spheres could be expected to make room for both ends of this spectrum in moral thoughtthe local-situated and the universalising one.
In his latest works, Habermas has partly moved towards the Deweyan position, and built a more nuanced conception of the interplay of personal and local discourses on the one hand, and public, abstract and universal rights-oriented discourses, on the other hand. He has also revised his view of the roles that political decision-makers and public officials have vis-a-vis the public sphere. In his Between Facts and Norms Habermas (1996) has sketched a picture of a public sphere as a system that hosts various types of discussion, some being informal and open to issues that have been traditionally considered "private," and some of them being more formal and restricted by rules that are meant to secure the impartiality of public decision-making (Habermas, 1996).
Thus, when the ideals and guiding principles of deliberative democracy are discussed, at issue should be the whole system of mutually interacting, different kinds of public discourses in the public sphere (Mansbridge et al., 2012). In deliberative systems, there would be space for, for instance, both laymen and expert-led discourses, self-organising and governmentmanaged discourses, and micro-scaled or local and macro-scale discourses. In this model, the lines between private and public do not need to be rigid but, for instance, certain issues pertaining to the private life of people could make their way into public discoursefor instance discourses on urban planning and development. The internet has already made the collection of information related to the private daily lives of people easy especially through Public Participation GIS (PPGIS) tools. These tools have been criticised, though, because they often allow planning organisations to paternalistically decide what kind of information they gather or what information they regard useful for public decision-making (Lapintie, 2017), instead of letting the public deliberate on the public significance of this information.
On many social media platforms, much of the contents consist of seemingly private matters and interests, but there are also many platforms where local publics search for shared interests and seek to influence local planning processes collectively (Afzalan & Evans-Cowley, 2015). What interests us in this article is how planners, who are traditionally considered to serve the public interest, evaluate the public relevance of the expressions of seemingly private matters and interests on social media. We also ask how planners themselves engage on social media, and whether they have observed and recognised deliberative processes that are oriented towards the public interest on a broader scale. Before turning to these questions, let us examine theoretically the relation between the theories of the digital public sphere and the possibilities the internet provides as a platform for deliberative discussion.

Digital Public Sphere and the Unfulfilled Promises of Public Participation
E-participation scholars have argued that the invention of the internet has reversed the structural transformation of the public sphere, revitalising the Habermasian public sphere (Papacharissi, 2008;Geiger, 2009) and expanding its scope (Dahlgren, 2005;Stromer-Galley & Wichowski, 2011). This novel type or part of the public sphere has been called for instance the "digital public sphere" (see e.g. Sch€ afer, 2015; Parnes, 2016;Mazzoleni, 2015), the "online public sphere" (Hindman, 2008) or the "virtual sphere" (Papacharissi, 2008). Given that the non-hierarchical and network-based structure of the internet enables it to host a variety of types of discourses, it is not surprising that the internet has been seen to resemble the Habermasian public sphere (Gimmler, 2001;Heng & de Moor, 2003;Dahlberg, 2007b).
The internet is also a structure that could be ideally independent of any single government or company. As such, the internet was originally expected to be able to resist commodification or colonisation by any single government or company (Gimmler, 2001). In reality, some companies that own the most popular social media platforms have power over users, and influence over the contents of the platforms through their moderation policies. Yet, internet-based platforms seem to be able, in principle, to advance certain features of Habermasian undistorted communication. For instance, the availability of space and time does not limit the discussions to the extent they do in face-to-face discussions (Geiger, 2009). Online discussions can be accessed from anywhere and anytime and usually at a relatively low cost, which makes the public sphere more egalitarian and inclusive than offline discussion fora (Donders et al., 2014).
Furthermore, it has been expected that social hierarchies, positions of power, or fear of discrimination do not influence deliberation as much as they do in face-to-face contexts, given that the participants can be for instance anonymised (Dahlberg, 2006;Stromer-Galley & Wichowski, 2011). Scholars such as Dahlberg (2006) and Stromer-Galley & Wichowski (2011) have also argued that internet-based fora might make the subjects more aware of their biases caused by their societal and cultural positions, as the participants in online discourses may be expected to encounter a larger variety of people and opinions than they would meet in their own neighbourhoods or workplaces. This has been expected to educate the users of the internet to listen to and understand viewpoints that differ from their own views (Dahlberg, 2006).
Nonetheless, the digital public sphere is not free from the practical constraints limiting these ideals. For instance, the accessibility of the digital public sphere is typically characterised by "digital divide," which means that there are differences in accessibility based, for instance, on people's income or geographic location (Hilbert, 2011). Furthermore, social hierarchies and positions of power are not abolished just by anonymising discussants, given that speakers are always influenced by the communicative traditions within which they have lived in the offline world, and which might have included for instance gendered or racialised practices and attitudes.
In addition to the fact that some of the promises of the internet have only been partially fulfilled, the internet often works completely against the ideals concerning public deliberation. This is especially because the internet is prone to communicative distortions, due to the unconstrained nature of the platformthe same feature that supports many desirable aspects of deliberation. Anonymous communication, for instance, does not necessarily advance a rational and neutral style of communication, but rather enables people to use disrespectful and aggressive language, and even to bully others without having to face the consequences of their behaviour (Moore et al., 2012). Anonymous discussants can use their anonymity as a shield when they attack the non-anonymous discussants based, for instance, on their gender or race (e.g. Schweitzer, 2014). Hence, the abstracted context of the internet does not always facilitate the emergence of Habermasian discourses oriented to the recognition of abstract and equal rights. Even though the Deweyan ideal of situated and embodied communication has been criticised for not being able to reach beyond the local and situated values, it still seems to have relevance for designing the institutional framework for the public discourses on planning even in those cases where not only local matters and values are at issue.
Attacks on gender and race are typical also for the phenomenon called trolling, that is, spreading of disinformation with an intention to manipulate the public and to create polarisation, conflicts, and chaos, eventually undermining the whole idea of rational deliberation (Dahlberg, 2006;Stromer-Galley & Wichowski, 2011). A related problem is that many social media platforms reduce the space left for deliberation-based opinion formation by relying on algorithms that are used to calculate what kinds of contents the users like to receive (Geiger, 2009;Stromer-Galley & Wichowski, 2011). Thus, the users do not always get exposed to the whole array of information and opinions, which might further polarisation (Stromer-Galley & Wichowski, 2011). However, it is also questionable whether polarisation could be avoided even if the users were exposed to diverse information and opposite opinions. Some scholars have presented evidence that this kind of exposure may even strengthen existing opinions (Bail et al., 2018;Nelimarkka et al., 2018).
The fragmentation of the public into like-minded sub-groups is the reason why Habermas himself has been sceptical about the possibilities of the internet to facilitate rational public opinion formation (Habermas, 2006). The question raised in recent agonistically-oriented planning-theoretical discourse is, by contrast, whether it is even desirable to have such platforms that advance Habermasian, agreement-oriented practices. Agonistically-oriented theorists argue that polarisation and emergence of competitive political alternatives to status quo are needed (e.g. Hillier, 2003;Pløger, 2004). They also criticise the tendency of Habermasian theory to prioritise argumentative use of language, which they think isdespite its seeming neutralityoften affirmative of the existing power constellations. Instead, they wish to promote passionate political language, which is assumed to be helpful in developing and expressing alternative political identities (Hillier, 2003;see also Sch€ afer, 2015). This being the case, the internet and social media platforms can be argued to be facilitative of "counterpublics" and thus also supportive of agonistic approaches (Dahlberg, 2007a(Dahlberg, , 2007b; see also Sch€ afer, 2015).
While the emergence of counterpublics might be a sign of a healthy political culture, for administrative culture counterpublics might pose a problem, given that administrationand planning as a part of thisis responsible for implementing political decisions rather than for making them. Therefore, public planning organisations might be expected to be geared towards micro-scaled and managed participation with procedural rules directed towards consensus-formation, rather than towards the chaotic and often disagreement-oriented macro-level (digital) public sphere (cf. Hendriks, 2006).

Finnish Planners on Participation in the Digital Public Sphere
In this section, we discuss the challenges that the digital public sphere poses for Finnish landuse planners. The context of public planning in Finland is characterised by publics strong on political decision-making and public administration (Puustinen et al., 2017). Political decisionmaking is based on a multi-party system, and the political culture has been described as consensus-oriented (Saukkonen, 2012). In this cultural and political context, public participation in general has not posed any major challenges to Finnish planners, and the polarisation of planning discourses has been rare thus far.
The planning system in Finland, as laid out in the Land Use and Building Act (132/1999), is comprehensive and hierarchical, consisting of three tiers of plans (regional plans, local master plans and local detailed plans). The Act obliges public authorities to provide a possibility for public participation in all public land-use planning processes, but the requirements concerning the forms and the media of participation are flexible. Currently, however, the legislation concerning land-use is under comprehensive reform.
Many Finnish municipalities use the internet as a channel of participation, and the biggest cities especially host internet-based participation platforms which enable many-to-many communication. Nonetheless, many-to-many communication on planning issues takes place today most often on commercial platforms such as Facebook and Twitter. Internet-based fora are generally regarded as accessible for the Finns. Even though there is evidence of digital divide in Finland (Heponiemi et al. 2021; see also Sj€ oblom & Niitamo, 2020), Finland has a nationally coordinated and regionally implemented policy aimed at overcoming this divide (Finnish Ministry of Finances, 2019). This is related to Finnish Government's policy of advancing the digitalisation of public services, including public land-use planning (Government of Finland, 2019). This is the general context of our two recent surveys on existing practices, attitudes, goals and expectations related to the uses of social media in Finnish land-use planning. Survey 1 was carried out in 2016 with a purpose to map the ways in which social media is utilised in Finnish planning and to acquire knowledge of planners' experiences of and attitudes towards social media. Survey 2 was conducted in 2018 for the Finnish Ministry of the Environment to provide knowledge of the current usage of digital tools in planning, especially digital participation tools. The purpose of this knowledge was to inform the reform of the planning law. Although the focus of Survey 2 extended beyond social media-based participation, covering also those forms of participation where many-to-many communication was not present, we decided to use its results to complement Survey 1, because we expected that Survey 1 had a self-selection bias towards social media enthusiasts, while Survey 2 attracted also those respondents who had a more neutral view on social media.
In both surveys, data collection was based on open-access internet questionnaires. The questionnaire of Survey 1 was distributed by an author of this paper via professional email-lists and social media channels. It attracted 224 responses, of which 71% came from officials working in local planning offices (see Nummi, 2019). The questionnaire of Survey 2 was distributed to the municipalities by the Finnish Ministry of the Environment. It attracted 252 responses of which 82% came from local planning offices. For this article, we have used only the responses from local planning offices, given that the powers of land-use planning lie mainly at local level in Finland. The respondents came from municipalities of different sizes and locations across the country. Nonetheless, in both surveys big cities were better represented than small municipalities.
Both surveys included questions concerning respondents' backgrounds (job description, organisation type), their use of social media, and their attitudes towards public participation, digitalisation and social media. Furthermore, the surveys included structured questions about tools and channels through which e-participation took place and the goals that the tools and channels were thought to serve. The questionnaires ended with open questions which were focused on the experienced challenges of and expectations towards social media (Survey 1), and on the challenges and benefits of digital participation in planning in a broader sense (Survey 2).
Survey 2 showed that social media is the second most commonly used digital tool for e-participation after traditional web sites. Majority (60%) of planners use social media in their work, either sporadically (45%) or regularly (15%). According to the survey, the quantity of planners using social media in leisure time is a bit higher (76%). In total, 87% of the respondents used social media. This is more than the average in Finland, where 71% of employed persons used social media at least once a week in 2018 (Official Statistics of Finland [OSF], 2018). Thus, it is possible that the data of this survey is biased towards the views of social media users. Nonetheless, it is also possible that planners use social media more than the average, given that the share of social media users is higher in urban areas compared to rural areas, and individuals with higher education use social media more than those with basic education (OSF, 2018).
In Survey 2, almost half of the respondents (46%) report that social media-based discussions have been followed in their organisations, while almost a third (31%) report that plans have been discussed online during the planning process. The main goals for using social media in public planning organisations are information sharing (97% of respondents) and reaching stakeholder groups (85% of respondents). The results also show that planning organisations aim at openness and interactive use of social media (82% of respondents), though this goal has not been fully achieved as less than one third (28%) of the respondents reported dialogical ways of using e-participation tools in planning processes.
In both surveys, the majority of the respondents had positive views about both public participation and the role of social media in planning. However, it was typical that the respondents had either very positive or very negative experiences. In the following sections, we analyse the responses to the open questions in relation to questions and dilemmas arising from the theories of the (digital) public sphere.

Manageability of Participation in the Digital Public Sphere
Those who had positive views about social media reported that social media-based participation improves the quality of planning and enhances its legitimacy. The respondents thought that the best way to improve the quality of planning and advance public acceptance of plans was to gather information through social media especially in the early stages of planning processes: In the beginning of a planning process, you are able to collect a vast amount of useful information from the inhabitants of the area [ … ]. The inhabitants like to tell about their environment and things that they find important.
(Survey 2, Digital participation in different stages of planning process) During the process of gathering information, planners typically had a position of a gatekeeper. When using internet platforms tailored for the purposes of their organisation and administered by their organisation, they could manage the input coming from the public. The respondents generally appreciated the possibility of designing the questions asked of the public and of steering the discussion towards topics they regarded as relevant. In this respect, their views can be associated with democratic elitism (see e.g. Whipple, 2005).
Map-based online questionnaires, in particular, were regarded as useful, given that planners could save time as they did not need to map the comments themselves. The respondents also held that maps force the participants "to look beyond their own quarter" (Survey 2). Some cities are known to utilise this feature of map-based participation to halt NIMBYism. However, mapbased online participation tools do not always fulfil the criteria of social media, as they do not necessarily allow many-to-many communication, a feature that would enable the participants to discuss and elaborate, for instance on their alternatives to NIMBYism, collaboratively and building a shared understanding.
Nonetheless, many of those respondents who had positive attitudes towards e-participation in the early phase of planning noted that when the discussion shifts to concrete draft plans, online participation becomes less constructive and its consequences more difficult to manage: Once you have the draft plans, opposition emerges, and the feedback you get might be very biased/ skewed. Usually, it is just opposition started by few people, but the influence of their feedback may have a disproportioned role for instance when the plan is assessed in the planning board [a board consisting of political decision-makers]. This in turn may steer planning into a direction that is unfavourable for the community as a whole (including the future inhabitants).
(Survey 2, Digital participation in different stages of planning process) This response reflects the fact that planners in Finland think that they are servants of the public interest (Puustinen et al., 2017), and that their plans can represent the public interest if the plans are based on a comprehensive gathering of knowledge at early stages of the planning process. However, they see as a threat to the quality of plans the tendency of the politicians to abandon this supposedly neutral knowledge base if it is challenged by alternative views that gain popularity on social media.
In both surveys, the respondents stated that managing participatory processes is time consuming especially in the context of social media. The lack of human resources was an important reason for planners to abstain from interactive use of social media even if they might have wanted to become engaged with social media-based discussions.
One of the features that gave planners reasons to question the relevance of social media for planning was the alleged dominance of aggressively promoted private preferences and opinions. While the presence of private preferences in public platforms was accepted per se, many respondents doubted whether people on social media platforms are willing or able to defend the public relevance of their interests and preferences in a manner that would be even remotely reminiscent of Deweyan or Habermasian public deliberation. Therefore the "likings" and other personal or sectional interests presented in the online discussions were not generally regarded as valid information for public planning: Urban planning should not be about personal likings. This is to say that the attitude towards social media is highly negative.
(Survey 1, experienced challenges) I don't expect much [from social media]. In my experience, social media-based discussion is dominated by embittered individuals who are just bashing the city [ … ] and the politicians for everything they have done, or of things that they haven't done, without having any knowledge on the background or the consequences [ … ].
(Survey 1, expectations towards social media) I don't think it is important that people have a possibility to influence [planning] through social media. Lay-men do not know much about planning, and it is the most loud-mouthed ones shouting there, and probably at some point (if we take social media seriously) construction companies will begin to try to influence the public opinion, and trolling becomes a commonplace. (Survey 2, feedback for the law reform)

Anonymity and Misinformation
Some respondents regarded the possibility for anonymous communication as a major problem in social media-based participation, arguing that social media-based discussion cannot replace face-to-face communication where discussants generally respect the reciprocity of communication and are willing to defend their arguments and to expose them to criticism. These respondents implicitly supported the Deweyan ideal of discourse where the discussants are physically present and have social bonds to each other. Anonymity was seen to have the consequence that people can throw in unfounded claims to support their own agendas: if people are allowed to participate anonymously [ … ] it is possible that the loudest ones succeed in mobilising others [ … ] against planning projects by spreading misinformation (Survey 2, experienced challenges).
In Survey 2, some respondents also noted that due to the rapid pace of discussions, planners and planning organisations have a hard time in correcting misinformation on social media. Some respondents held that distracting behaviour and irrational inputs jeopardise the whole idea of participation as a rational and constructive process, hindering "normal" people from participating in planning.

Aggressive Styles of Communication
According to some respondents, planners are reluctant to use social media in their work because of the aggressive style of online communication, which advances polarisation and to which planners do not wish to expose themselves, particularly if they use their personal profiles as they engage in social media-based discourses: An impediment [for planners' use of social media] is the risk that they might end up in the firing line, I fully understand why planners do not want to use their personal profiles to discuss in social media, because all the slanders become personal.
(Survey 1, experienced challenges) Citizens react fiercely and emotionally to plans that harm their own living conditions, for instance, to an apartment house that blocks their views. If planners defended their plans with their own names and faces, their physical safety could be jeopardised. Even if things would not go this far, it is possible that you can no longer walk in the streets without being disturbed.

Planners' Dual Role: Inside and outside the Public Sphere
Some respondents held that public planning organisations should maintain a critical distance from social media platforms as fora for public opinion formation, while still monitoring the opinion formation there. This is the way in which early Habermas (1991) describes the ideal relation between the public sphere and formal public decision-making. However, these respondents typically also had doubts about the quality of opinions expressed on social media platforms, which is why they thought in a Deweyan manner that they needed to contribute to the discussions by providing the arguments that could correct for instance biased representations of public opinion: E-participation gives you a random sample of the views of those citizens who are most active and interested. It does not amount to much more than providing an overall idea of what is the general atmosphere in the opinion formation. This might differ very much from the solution that is the most rational and most consistent with the interest of the whole municipality. But it helps you in building the right arguments to support the best solution if you know the general atmosphere.
(Survey 2, usefulness of digital participation in planning process) Some respondents wanted to maintain a distance from social media-based planning discussions because they did not wish to confuse their roles as public officials and private people or citizens: I don't stand unreservedly behind all those things that I do at work, so there would be a risk of confusion, and I would not represent the view of the [planning] department. I don't want to end up in that situation.
(Survey 1, experienced challenges) Rather than promoting a Deweyan view of the public officials as unquestionable representatives of the public interest, this respondent's comment aligns with the Habermasian analysis of public administration as a potential source of communication distortions. The comment might suggest that planners as public officials oftentimes just need to adapt to the decisions that the politicians make, perhaps under the influence of powerful economic actors. As citizens in the public sphere, the same planners might think that these decisions do not represent the public interest (c.f. Mattila et al., 2012;Kangasoja & Mattila, 2018).
Nonetheless, there were also respondents who stated that they have not experienced conflicts between their roles as public officials, private people, and citizens, and that they could stand behind their work in any role, knowing that their organisations would support their contributions to social media-based discussions if needed. However, according to Survey 2, only a minority (15%) of public planning organisations encourage their employees to use their personal profiles as they discuss planning projects on social media. Some of the respondents reported that their organisations had shared accounts on Facebook and Twitter through which individual planners can contribute to discussions. In some cases, communication on social media was taken care of by organisations' communication specialists, who typically did not experience the problem of role confusion as they were not personally engaged in planning work.

Discussion
Even though many respondents held that social media has, or could have, positive impacts on planning, it is striking that none of the respondents mentioned the potential of social media in hosting Deweyan or Habermasian deliberative processes where local communities' interests, let alone broader public interests, could emerge. Quite the contrary, those who commented on the quality of discussions on social media held that it was not the best arguments but the loudest voicesusually targeted against some specific plansthat typically dominated discussion fora. One of the main problems of social media-based discourses was reported to be that discussions were dominated by sectional interests and personal "likings," which were presented with an implicit or explicit claim that they should be taken into account in public decision-making, even though their generalisability was not backed up with arguments or tested in discourses. Planners did not dismiss personal "likings" altogether, however, given that many of the respondents reported that they appreciate social media as a channel through which planning organisations can acquire information covering not only objective qualities of the environment but also various types of local and experiential knowledge and subjective views that people had of their surroundingsincluding information about what people like and what they do not like in their surroundings.
We were left wondering whether the respondents thought that personal likes or dislikes have value for public decision-making when there is quantitative knowledge about them, but this was not explicitly stated in the responses. As already discussed, Habermasian-inspired communicative planning theorists criticise the aggregative, technocratically-oriented and elitist approach where only the quantity of preferences, interests or opinions count. They argue that instead of an aggregative approach, public planning and decision-making should rely on discussion-based decision-making, where people have a possibility to elaborate on their preferences and interests, and formulate new, shared objectives during the process (Healey, 1997; see also Forester 1989Forester , 1993. Nonetheless, we know from previous research that Finnish planners appreciate some social media platforms and like to contribute to the discussions on these platforms. Planners in Helsinki, the capital of Finland, often refer to discussions on the Facebook group "Lis€ a€ a kaupunkia Helsinkiin" (More city to Helsinki), which is focused on discussion on growth and densification in Helsinki (Sj€ oblom & Niitamo, 2020). It has attracted thousands of participants who wish to halt NIMBYist arguments and to promote public interest on a city-wide level, considering not only the interests of the current residents but also the future residents of Helsinki.
Another interesting finding was that planners thought that poor quality social media-based discussions meant that they could ignore such discussions. This may be related to institutional resistance (cf. Macintosh et al., 2009), and to a fear of entering a sphere where participation is difficult to plan and manage. Nonetheless, social media influences planning even if planners and planning organisations would bypass it. For example, in Helsinki, some major planning projects have recently been cancelled at a late stage as a result of fierce social media-based opposition (e.g. Hirvola & M€ antysalo, 2019). There is evidence that planners are better able to advance their projects if they do not settle with informing the public but engage in a dialogue with the public throughout the planning process (Schweitzer, 2014).
In contrast to many contemporary, agonistically-oriented planning theorists' views, the respondents to our questionnaires did not see progressive potential in counterpublics arising from social media platforms. This might be due to the fact that the Finnish planning institution has been so far capable of carrying out relatively fair and transparent planning processes and producing good-quality outcomes. Nonetheless, there are reasons to expect that counterpublics will increasingly challenge the position of public planners in the future.
Even though participation through social media is unpredictable and often unmanageable, we argue that planners and planning organisations should not disregard social media. They should adopt a Deweyan attitude and take part in the public discussion, for instance by monitoring the discussions going on in the relatively uncontrollable macro-level digital public sphere, and contribute to discussions by correcting misinformation, misunderstandings and misrepresentations of public opinion and public interest in such an environment (cf. Porwol et al., 2014). In this way, they can maintain their organisations' capacity to plan in a turbulent environment where social media riots are known to be able to halt planning projects even at a very late stage. At the same time, planning organisations' own online and offline micro-scale participation fora with specific rules for communication have their merits. They can contribute to the quality of public input in planning especially when they are brought into contact with the broader environment of self-organising discussions in social media. Public planning organisations should have strategies for building such communication architectures that bring together the discussions in the "free" digital public sphere and on their own managed online and offline participation platforms.

Conclusions
The results of our surveys suggest that Finnish planners do not believe that the digital public sphere emerging from social media platforms forms as such a public sphere that can give a rise to rationally motivated public interests or public opinions. Even though the respondents of our surveys appreciated the fact that social media has broadened and diversified the group of people that participates in planning, they emphasised the relevance of the input coming from channels designed for the purposes of planning organisations rather than the input coming from social media channels. Social media-based discussions were not valued as highly as the platforms designed and managed by planning organisations were. Even though social media-based discussions provided important information for planners, they were seen not to produce reliable presentations of the public opinion or public interest, but rather a collection of private or sectional views, the public relevance of which was only seldom backed up with arguments. If this observation made by the respondents is correct, then there are reasons to argue that the discussion on these views should be continued on such online and offline fora where facilitators require arguments and justifications from the discussants, thus helping the public to sort out which private or sectional interests have public relevance.
Whereas the respondents seemed to think that the poor quality of discussions entitled them to dismiss the opinions emerging from the discussions on social media and retreat into "democratic elitism" (c.f. Whipple, 2005), recent research literature suggests that planning institutions should not dismiss the opinion formation in the digital public sphere. The digital public sphere seems to create new possibilities for counterpublics to mobilise people against the plans. The more the minority voices are being ignored by planning organisations, the likelier it becomes that plans will be challenged by self-organising counterpublicsbe they progressive publics wishing to advance goals such as sustainable development or populist groups that just wish to oppose public authorities. If planners and planning organisations wish to avoid conflicts and promote constructive forms of participation, open dialogue with the public throughout planning processes is needed in all relevant channels, social media included (cf. Schweitzer, 2014).
Nonetheless, this requires that planning organisations support planners' active presence on social media. According to our survey, many Finnish planners now engage in work-related social media discussions after office hours. Planning organisations should coordinate their social media engagement on the organisational level and provide their employers with resources for the use of social media as well as education on its use. In addition, one important prerequisite of the constructive online presence of planners is that planning organisations, including the political decision-makers, are committed to advancing the public interest in planning, instead of letting for instance the powerful economic actors steer their work. Only in this way can planners openly, critically, and constructively discuss their plans on social media, where their roles as deliberating citizens and public officials always tend to get intertwined.

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

Funding
The work of Hanna Mattila has been supported by the Academy of Finland project TRANAPLAN [grant no. 13333366] and a grant from Kunnallisalan kehitt€ amiss€ a€ ati€ o. The work of Pilvi Nummi has been supported by the European Commission through the H2020 project Finest Twins [grant No. 856602]. The second survey has been financed by the Ministry of the Environment in Finland.

Notes on Contributors
Hanna Mattila (Dr.Sc., tech.) is Associate Professor at the Department of Architecture, Design and Media Technology at Aalborg University and Visiting Professor at the Department of Built Environment, Aalto University. Her research interests include planning theory, planning systems and law, and urban design.
Pilvi Nummi (architect, PhD) is Postdoctoral Researcher at the Department of Built Environment, Aalto University and FinEst Centre for Smart Cities, TalTech. Her research addresses digital participation in urban planning and she currently works with user-centered development of new 3D tools for communicative planning of urban green areas.