1We have observed an increasing interest among urban researchers across various domains of urban studies in examining the importance of language and narrative and leveraging the potential of discourse studies (DS) (Matthey et. al., 2023; Shirazi, 2023; Ameel, 2020). This aligns with the “linguistic turn” in philosophy and social sciences, which “called attention to language games that construct alternative realities, grammars that transform the perceptible into nonobvious meanings, and language as a form of action that generates radiating chains of connotations while undermining its own assumptions and assertions” (Edelman 1988: 103). As noted by Hajer (1993: 44), in contrast to the positivist tradition, where language held a neutral status, the linguistic turn problematized language “as a medium, a system of signification through which actors not simply describe but create the world” [Original Emphasis]. The positivist and rational approach to policymaking argues that policy solutions are universal truths that can be determined by policy professionals (Marston 2004b). The focus of this approach is on objective outcomes and grand narratives of truth, rationality, and progress, disregarding discursive processes that produce knowledge and construct policy environments. Consequently, it fails to analyse the role of power in negotiating, debating, and compromising policy agendas. In fact, the positivist approach treats language as a do-nothing domain, undermining the significance of discourse as a powerful element in generating political parameters and socio-cultural identities that set the stage for debating and legitimizing policy change.
2In response to the technical and technocratic approach to planning, an argumentative turn emerged in policy analysis that conceptualised policy analysis as a practical process of argumentation (Fischer and Forester 1993). The argumentative turn advocates for a multidimensional view of power and underscores the significance of policy language in policy-making, exploring the intricate relationship between agency, organizational structure, and decision-making. As the result of this awareness, researchers began integrating the study of language into urban research and analysis (Lees 2004), highlighting the role of discourse as a pivotal element in urban processes and policy change (Hastings 1999b). Consequently, scholars engaged in various areas of urban research such as urban geography, urban planning, urban policy, and housing studies acknowledged that “Language has the capacity to make politics, to create signs and symbols that shift power balances, to render events harmless or, on the contrary, to create political conflict” (Hajer and Versteeg 2005: 179). They recognised that “concepts and language are ways of not only representing the world but also of constructing environmental and social problems and their solutions” (Soini and Birkeland 2014: 221).
3Urban researchers have acknowledged the significant role of language in the policy arena, recognising its utility in pursuing organisational and political objectives (Ameel, 2017). They have thus noted that “a close examination of language in the form of utterances and texts can provide a richer and more nuanced understanding of the policy process than is possible from more traditional methods or techniques” (Jacobs 2006: 40). These urban researchers have become interested in understanding how language deployment encompasses social content and effects, promoting questions about “how language is used, why, by whom, in what circumstances and to what effect” (Hastings 2000: 132). Moreover, they have realised that language shapes urban policy, with argumentation in both oral and written form being central to the policy process (Majone 1992). In other words, as Weiss and Wodak (2003: 15) articulate, “texts are often sites of struggle in that they show traces of differing discourses and ideologies contending and struggling for dominance”. Ameel (2017, 2021) distinguishes between narratives “for, in and of planning.” Narratives for planning are narratives that planners can draw on in their practices, narratives in planning relate to the planning process as story-telling, and narratives of planning is about the storytelling that follows in the wake of planning practices.
4In 1999, a special issue was published, gathering selected papers from the conference “Discourse and Urban Change” held at the University of Glasgow, UK, in 1997. This issue centred on the role of language use in determining meaning in the urban policy process, aiming to contribute to the development of the linguistic turn within the study of urban policy and to explore urban issues from the perspective of discourse studies (Hastings 1999b). Since then, researchers’ interest in exploring urban issues from the perspective of discourse studies has steadily grown, resulting in a rich body of publications. These publications offer evidence of the centrality of language to urban processes and the significance of discourse studies to urban research. However, there has been no analytical study critically reviewing this body of literature.
5To address this knowledge gap, this paper offers a critical appraisal of the existing literature by investigating two key questions. First, which research areas of urban studies have utilised DS more in their research and how? Second, what types of discursive modes have been subject to study, and what kinds of materials and resources have been analysed? A critical appraisal of the existing literature will provide urban researchers with an overview of previous works and offer guidance for scholars who want to deploy DS in the future. Following a brief overview of DS, this paper presents and discusses research findings in two areas: the field of inquiry and the research corpus. The paper concludes by introducing nine key challenges of DS in urban research and suggesting directions for future research to address these challenges.
6Defining the term "discourse" comprehensively proves challenging due to its diverse connotations and implications. As Mills aptly states, “Discourse has perhaps the widest range of possible significations of any term in literary and cultural theory” (Mills 2004: 1). Fairclough (2016: 87) delineates three main senses in which discourse is generally employed: “(a) meaning-making as an element of the social process, (b) the language associated with a particular social field or practice (e.g. ‘political discourse’), (c) a way of construing aspects of the world associated with a particular social perspective (e.g. ‘neo-liberal discourse of globalization’)”. While discourse and text are often used interchangeably, as Resigl and Wodak (2016) clarify, texts represent only a part of discourses. Discourse is inherently multimodal, encompassing various modes such as images, writing, speech, layout, gestures, bodily movements, moving image, sound, and three-dimensional objects (Kress 2010: 79).
7In broad terms, two approaches to discourse analysis prevail. The first views discourse as social action and interaction, leaning heavily on linguistic influence and favouring a texturally oriented analysis. The second regards discourse as a social contraction or reality, a form of knowledge. Influenced by social science and social theory, and strongly inspired by Michael Foucault, it pays little attention to the linguistic properties of texts and discourse (Fairclough 1995 2003). Within social theory, especially in the works of Foucault, discourse signifies varied ways in which knowledge and social practice are structured. Here, discourse doesn’t merely reflect social relations and entities; it constructs or constitutes them (Fairclough 1992). According to van Dijk, “there is still a gap between more linguistically oriented studies of text and talk and the various social and political approaches. The first often ignore concepts and theories in sociology and political science on power abuse and inequality, whereas the second seldom engages in detailed discourse analysis” (van Dijk 2015: 479).
8In the 1990s, a network of scholars established a new branch within DS (Wodak and Meyer 2016) known as Critical Discourse Analysis (CDA), or as we prefer to refer to it here, Critical Discourse Studies (CDS), following van Dijk (2008). Influenced by the “critical theory” of the Frankfurt School, which advocates that social theory should critique and seek to change society (Wodak and Meyer 2016), CDS “primarily studies the way social power abuse, dominance, and inequality are enacted, reproduced and resisted by text and talk in the social and political context” (van Dijk 2008: 85). CDS recognises a dialectical relationship between discourse and society: “Discourse is socially constitutive as well as socially shaped: it constitutes situations, objects of knowledge, and the social identities of and relationships between people and groups of people” (Fairclough et al. 2011: 357–358) [Original Emphasis]. Van Dijk (1995, 2015) outlines the key characteristics and objectives of CDA: it is problem- or issue-oriented in the sense that the employed approach is appropriate as long as it is capable of addressing existing social problems; it is explicitly critical towards texts and talks aligning with the tradition of critical studies in social sciences; it is interdisciplinary or multidisciplinary, focusing on the relationship between discourse and society; instead of merely describing discourse structures, it explains them as properties of social interaction and structure; it addresses all levels of discourse from grammar via style to interaction; it concentrates on relations of inequality, dominance, and power, exploring how they are reproduced and resisted, thereby tackling issues of class, gender, ethnicity, race, sexual orientation, etc.; it aims to uncover hidden or implicit aspects behind ideologies, studying strategies of manipulation, legitimisation, and manufacturing of consent, while challenging the abuse of power by elites and dominant classes; and finally, it formulates strategies to develop counter-power and counter-ideologies in resistance practices.
9Our study investigates two key questions: the main areas of inquiry that have utilised DS in their research (field of inquiry), and the key discursive modes and types of materials that have been subject to analysis (research corpus). To address these questions, many publications in urban research were analysed. Using the keywords of “urban” and “discourse”, a literature search was conducted in two search engines Google Scholar and Science Direct, to find relevant publications for analysis. Keywords were then expanded to cover relevant phrases such as urban studies, urban discourse, city, urban planning, urban design, urban policy, housing, and the like. A list of 189 items was identified for further investigation. The full text of all these publications were reviewed against the criteria, to identify cases that are theoretically informed by DS, or clearly and explicitly claim using DS as the methodological framework. After excluding irrelevant items, finally 126 publications (121 peer-reviewed articles, 4 book chapters, and one book) were selected for in-depth analysis. Full list of analysed publications is available as appendix.
10A specific timeframe was not designated for the retrieval of publications within this domain. Nevertheless, the chosen literature encompasses works from the 1990s onward, with the preponderance (104 out of 126) released post-2000. This pattern may stem in part from the digitization of academic materials, thereby enhancing accessibility. Additionally, it potentially reflects the escalating interest among urban scholars in employing DS for urban research over the past two decades. Based on the research questions, all the selected publications were carefully reviewed to extract data about two main thematic focuses of “field of inquiry” and “corpus of the analysis”: while the latter explores what type of materials and resources have been subject to analysis, the former investigates the main areas in urban research that have utilised DS in the research. There are two limitations to this analysis. Firstly, it solely encompasses publications in the English language. We acknowledge that numerous valuable resources have been published in languages other than English, and this exclusion poses a risk to the reliability of the study results and the comprehensiveness of the analysis corpus. Secondly, one may find publications that have not been included in this research. By expanding the search keywords, we managed to achieve a more comprehensive list of analysed publications, and by using different search engines this analysis reached a point where no new relevant articles were found.
11Our analysis indicates that DS has been extensively utilised in four major fields of urban research: urban planning and theory, urban policy, housing research, and environmental policy. Table 1 provides an overview of the four fields and sub-fields, and relevance of DS to these areas (Table 1).
Table 1: Fields of inquiry for DS in urban research
Field of Inquiry
|
Sub-theme/area
|
Relevance of DS
|
Urban planning and theory
|
Urban regeneration, urban renewal, and gentrification
|
Texts are sites of conflict, concrete realisations of discourses and discursive strategies.
Planning texts uncover power struggles underlying prevalent perceptions of urban issues.
Discursive activities of planners reveal distorted communication within relevant organisations and institutions.
Planning texts uphold and advance dominant ideologies and related power dynamics.
Planning texts disclose contested nature of planning issues, the construction and responses to planning questions, and the exercise of power.
Plans expose rationalities of their time and place.
Planning profession and planning language are mutually shaped.
|
Spatial/strategic planning
|
Urban development projects
|
Smart city
|
Place identity
|
Gated communities
|
Informal settlements
|
Public participation
|
Urban design
|
Urban policy
|
Urban governance
|
Normative policy frameworks and conventional structural approaches to policy-making are challenged.
Policy meanings and actions are contingent.
Real intentions of a policy are speculated.
Texts can be deployed to legitimise policy actions.
Power dynamics in the policy processes are explored.
Discursive conflicts shape the policy process, and policy processes are shaped by (and in turn shape) the real world.
Language can be used to promote policy outcomes.
|
Urban policy/policy procedure
|
Housing Research
|
Gentrification
|
Investigating housing question from a non-positivist epistemological standpoint.
Exploring forces involved in shaping housing research and policy practice.
Examine housing questions from non-traditional disciplinary perspectives, e.g. linguistics, psychology, and philosophy.
|
Housing policy
|
Homelessness
|
Sustainable housing
|
Housing planning and design
|
Social housing
|
Housing market
|
Environmental Policy
|
Urban water/urban infrastructure
|
Uncovering the role of language in politics.
Environmental arguments are linguistic constructions.
Environmental arguments are rooted in specific operational routines and accepted social norms.
Discourse studies democratise the processes of producing and naming the environment.
|
Environmental justice
|
Environmental policy and planning
|
Climate change
|
Green infrastructure
|
Urban agriculture
|
12In urban planning and theory, texts are perceived as “sites of conflict” (Tett and Wolfe 1991: 199), as concrete realisations of discourses and discursive strategies. Planning texts and manuscripts have been analysed to provide evidence for the mythical and rhetorical nature of “rationality” in planning and to explore power struggles underlying prevalent perceptions of urban issues. According to Fischer (2011), the discursive activities of planners, including their written and spoken words, serve as crucial sources of knowledge for understanding and analysing distorted communication within relevant organisations and institutions. Furthermore, discourse plays a pivotal role in the communicative turn in planning theory, which drawing on Habermas’s idea of communicative action, views planning practice as an intercommunicative process. Communicative planning presupposes the pre-existence of discourse communities with diverse meaning systems and knowledge forms, aiming to foster achievable levels of mutual understanding through respectful interdiscursive communicative action (Healey 1993).
13Another valuable application of DS in urban planning and theory lies in examining the central role of power in urban theory and urban planning. Debates have arisen regarding the effectiveness of two main approaches to power: the macro-politics of Habermas's communicative theory and the micro-politics of Foucault's actual reality (Flyvbjerg and Richardson 2002). DS investigates how key actors utilise policy and planning texts to uphold and advance dominant ideologies and related power dynamics. It sheds light on the contested nature of planning issues, the construction and responses to planning questions, and the exercise of power (Kumar and Pallathucheril 2004). Additionally, discourse in planning serves as a useful tool for understanding contemporary urban planning styles, because, unlike ideology, the notion of discourse implies a more flexible and less structured organization of ideas (Goodchild and Cole 2001), making it particularly suitable for cities with diverse and pluralistic characteristics.
14DS enables urban scholars to illustrate how plans and policies “realize the discourses, practices and rationalities of their time and place” (Matthews and Satsangi 2007: 506). Moreover, there exists a dialectical relationship between planning professions and the language practices of planners: “Planners must become sensitized to the administrative voices that inadvertently shape their existing language discourse, while at the same time recognize that their language practices are constitutive of their professional” (Tett and Wolfe 1991: 199). Through the use of DS, urban scholars can transcend the conventional approach of policy analysis of the city and delve deeper into the cultural constructions of the city, which policy often takes for granted, thus drawing attention to how sources of institutional power shape cultural constructions (Boyle and Rogerson 2001). In urban restructuring projects, influential stakeholders utilise discourse to frame urban restructuring as beneficial and normal, thereby influencing the creation of new place identities and legitimising the restructuring process and its associated social costs (Mele 2000). A discursive analysis of social practices and cultural images in urban development projects bridges objective and subjective factors, assuming a coherence between social and spatial arrangements (Zukin et al. 1998). Overall, the application of DS to urban planning and theory encompasses a wide variety of fields and themes, including urban regeneration, renewal and gentrification; spatial planning and strategic planning; urban development projects; smart city; place identity; gated communities; informal settlements; public participation; and urban design.
15In urban policy studies, DS encourages professionals and academics to challenge existing normative policy frameworks and question conventional structural approaches to policy-making. Since “policy-making is a constant discursive struggle”, as Fischer and Forester put it (1993: 1-2), DS presents alternative discourse types to challenge dominant policy discourses and critique the way policy problems are discursively structured and represented in social policy debates (Marston 2000). Furthermore, it offers empirical evidence to demonstrate the contingent nature of all policy meanings and actions (Marston 2004b). As Atkinson (2000: 230) argues, “Discourse analysis can help us to interrogate the very notion of ‘a policy’ as presented by the state, to question if this ‘policy’ is actually doing what the state portrays it as doing and to begin to speculate about the ‘real’ intentions of policy.” DS enables exploration of “the ways in which texts are deployed to legitimise policy action, how they are received and acted upon” (Jacobs 1999: 210).
16According to Richardson (Richardson 1996), discourse-based analysis and theorising of decision-making has gained popularity in urban policy analysis, typically falling into two camps: Habermasian and Foucauldian. While policy analysts have generally favoured Habermasian thinking, the Foucauldian approach is deemed more suitable for probing power dynamics in the policy process: “Communicative rationality may be posited as an idealized form of policy debate, but Foucault reaches deeper towards an understanding of the deployment of power in the real world” (Richardson 1996: 287). Planning for the real-world, Richardson argues, necessitates an understanding of the discursive conflicts that shape the policy process, and the Foucauldian approach facilitates such understanding. “Policy processes are shaped by (and in turn shape) the real world at several levels: sparked by real-world problems and shaped by discourses; resolving planning problems, and affecting the relations between discourses which shape the social world” (Richardson 1996: 288).
17Rydin (1998) distinguishes between the passive and active role of language in the policy process. A passive role of language in policy either emphasises justificatory marketing, wherein language is used to promote policy outcomes, or is utilised transparently by policy actors to communicate their opinions and attitudes. An active approach to language in policy either generates ideological effects through cultural practices, so that language shapes our understanding of policy outcomes, or establishes a dialectical relationship between the communicative actions of actors and discourse structures. Rydin concludes that “language can influence the policy process in a variety of ways: it can alter perceptions of interests and issues; it can define the object of policy attention; it can promote particular policy agendas; it can shape the nature of communication between actors; as we shall see, it can cement coalitions or differences between actors; and it can be diversionary, resulting in a form of symbolic politics” (Rydin 1998: 178). Our analysis suggests that DS has been widely used to investigate various aspects of policy-making such as urban governance, policy, and decision-making procedures.
18In the field of housing research, since the 1990s, DS has emerged as an alternative approach to the housing question, which had previously been dominated by positivism and evidence-based investigations (Marston 2002). Traditionally, housing policy literature did not delve into questions of epistemology and language; policy language was viewed as a neutral medium for presenting and discussing ideas (Jacobs and Manzi 1996). While housing policies have demonstrated themselves as evidence-based, the limitations of evidence-based and rational approaches have encouraged researchers to explore alternative approaches and look at the housing question from a different epistemological standpoint. Meanwhile, housing problems were often perceived as self-evident, limited to a few manageable challenges. This approach downplayed the role of power, dominant political agendas, policy-making, and lobbying agents (Jacobs et al. 2003). Overall, DS presents itself as a powerful form of social inquiry that facilitates the exploration of forces and changes shaping housing research and policy practice.
19According to Hastings (2000), DS offers three significant contributions to housing studies. Firstly, a discourse perspective marks an epistemological break with the previously positivist approach in hosing studies. Secondly, DS enables researchers to examine housing questions from non-traditional disciplinary perspectives, such as linguistics, psychology, and philosophy. Lastly, it opens up new empirical domains within the housing discipline, including written and verbal communication. By employing DS, scholars are able to bring into question established knowledge or practices within housing policy and practice, or focus on language to scrutinise the establishment and acceptance of orthodoxies. Our analysis indicates that scholars have extensively employed DS to investigate various aspects of the housing question, including housing policies (e.g. Marston 2004, Cole and Goodchild 2000), homelessness (e.g. Löfstrand and Juhila 2012, Arapoglou 2004), housing planning and design (Murdoch 2004), social housing (Watt 2008), and sustainable housing (Lovell 2004).
20DS has also been utilised in environmental policy debates. A discursive analysis of environmental policymaking “shows how environmental problems and a related set of subjects and objects are discursively produced and rendered governable” (Feindt and Oels 2005: 163). Within the environmental policy debate, DS has three specific strengths: it uncovers the role of language in politics, elucidates how language is embedded in practice, and sheds light on mechanisms and provide answers to “how questions” (Hajer and Versteeg 2005). DS can probe how environmental arguments are linguistic constructions, deriving their meanings from specific operational routines and accepted social norms.
21Overall, the study of environmental politics has been transformed by DS in five ways. Firstly, the environment is situated “inside” society and discursively re-produced. Secondly, DS promotes a reflexive understanding of “the political” that transforms the practice of policy analysis. Thirdly, DS explores the processes of subject and object. Fourthly, through DS, time and space are conceived as contested concepts. Lastly, DS democratises the processes of producing and naming the “environment” (Feindt and Oels 2005). Research employing DS to study various environmental aspects in urban research encompasses diverse areas and topics, including climate change, green infrastructure, environmental policy and planning, environmental justice, and urban agriculture.
22It is important to note that DS has found application in other areas of urban research such as urban tourism (Hassanli et al. 2022), urban health (Harris et al. 2019), and urban heritage (Waterton et al. 2006), to name but a few. Nonetheless, the four areas highlighted above remain the dominant fields in urban research.
23The corpus of study holds significant importance in DS. It refers to the materials subjected to analysis and various modes considered as discourse. The corpus can encompass a broad array of resources, including policy documents, newspaper articles, media coverage, films and videos, photographs, advertisements, maps, and many other types of materials. Table 2 provides an overview of the findings in terms of corpus type, definitions, approach, and examples.
24Regarding the corpus of the analysis, we can distinguish between the number (quantity) and type (mode) of analysed materials. Our analysis reveals that in a significant number of studies, discourse analysis is confined to a single published material, primarily written texts and official documents such as legislative papers, policy documents, and official statements. For instance, Atkinson (1999) analyses one single official manual to investigate how the concepts of partnership and empowerment have been discursively constructed in the official discourse of British urban regeneration, and demonstrates how the dominant discourse of the document simultaneously facilitates, structures, and constrains community participation in urban regeneration. Similarly, Hastings (1998) analyses an urban policy document titled “New Life for Urban Scotland” to explore how residents of disadvantaged neighbourhoods are overlooked as potential agents in shaping the future of their communities. Watt and Jacobs (2000) examine one policy report, “Bringing Britain together: a national strategy for neighbourhood renewal,” published by the UK Government’s Social Exclusion Unit, to analyse how the terms “poor neighbourhoods” and “housing problems” are represented in the report.
Table 2. Monomodal and multimodal corpus of the analysis in urban research
|
Corpus of the Analysis
|
Corpus
|
Monomodal Corpus
|
Multimodal Corpus
|
Definition
|
The corpus of analysis is limited to one type of documents
|
The corpus of analysis encompasses a range of different types of documents
|
Approach
|
Single document of a single type
|
Multiple documents of a single type
|
Multiple documents of multiple types
|
Examples (selective)
|
A single official manual (Atkinson, 1999); one urban policy document (Hastings (1998); A policy report (Watt and Jacobs, 2000)
|
Two local government texts (Stenson and Watt, 1999); 10 policy documents (Damurski and Oleksy 2018); 222 newspaper articles (Munro, 2018
|
annual reports, interviews, policy texts, media reports, press releases, and ministerial correspondence (Marston, 2004b, 2002, 2000); an extensive array of web-based documentary resources, including news stories, industry and survey reports, magazine articles, expert reviews, tourism advertisements, editorials, and travel dairies (Li et al., 2018); marketing materials, website animations and a video (Matthews and Satsangi, 2007)
|
25There are also many studies that utilise multiple numbers of materials, yet the corpus is confined to a single type of document. For example, to investigate power relations amongst participants involved in a multi-sectoral urban regeneration partnership, Hastings (1999a) analyses two sets of interviews conducted in the early stages of the partnership and then one year later with representatives of the local authority and the private sector. Stenson and Watt (1999) solely focus on two local government texts related to Wycombe District in south-east England to explore the extent to which local authorities advocate neo-liberal political rationalities, which signify the demise of the “social”. Richardson and Jensen (Jensen and Richardson 2001, Richardson and Jensen 2000, 2003) scrutinise a small number of official policy documents to examine dialectical relations of space in European spatial planning and explore the emergence of new spatial discourses. Damurski and Oleksy analyse 10 policy documents to investigate the importance of communicative and participatory paradigms in contemporary European territorial policy (Damurski and Oleksy 2018). To identify the storylines shaping sociotechnical imaginaries of urban carbon neutrality among the 17 founding members of the Carbon Neutral Cities Alliance, Tozer and Klenk (2018) analyse climate governance texts produced by founding members. Munro (2018) analyses 222 newspaper articles to discern the discourses used in reporting house prices and housing markets in UK newspaper coverage.
26To attain a more profound insight into the subject at hand, certain researchers broaden the scope of their corpus and scrutinise a wider array of materials and documents. For instance, Marston (2004b, 2002, 2000) uses archival materials including annual reports, interviews, policy texts, media reports, press releases, and ministerial correspondence, to investigate housing policy change in Queensland, Australia. In examining the efficacy of the strategic plan of the City of Lahti in Finland, Vaara and colleagues (2010), supplement their analyse of official documents, interviews, and various media texts, with personal diaries of the participants and personal emails of some decision makers. Investigating how constructed perceptions of owners and tenants, lenders and landlords influence the policy agenda and the rationalisations that legitimise policy choices in the UK, Hunter and Nixon (1999), in addition to studying parliamentary speeches and newspaper articles, analyse court records and interview with judges. Huovinen et al. (2017) study diaries obtained from tram passengers to observe the changing and varied identities of four Helsinki neighbourhoods along the route of Tram Number Eight, elucidating how place identities can be illustrated as a cultural, discursive map. Li et al. (2018) examine an extensive array of web-based documentary resources, including news stories, industry and survey reports, magazine articles, expert reviews, tourism advertisements, editorials, and travel dairies.
27In certain studies, the corpus of the analysis extends beyond textual materials to encompass non-textual resources such as videos and audio. For example, Taylor (1999) undertakes an in-depth examination of a video regarding stock transfer distributed to tenants, besides benefitting from other sources like official documents, interviews, and leaflets, to scrutinise the Scottish Homes stock transfer programme. Matthews and Satsangi (2007) analyse the Leith Docks Development Framework (LDDF) and pertinent marketing materials including website animations and a video produced by the landowner, in order to explore power dynamics between the landowner and developer in the redevelopment of Leith Docks in Edinburgh, complemented by semi-structured interviews with all key stakeholders.
28In line with the growing presence of social media in news reporting, event documentation, and the dissemination of personal viewpoints, social platforms have become part of the corpus analysis. Budge (Budge 2019), alongside policy documents, analyses Instagram posts to examine popular commentary surrounding city-making processes via both mainstream media and social netwroking platforms. Likewise, Williamson and Ruming (2017) analyse Twitter content generated by the Save Bronte community group to demonstrate their concerns regarding a proposed redevelopment initiative in Bronte, Sydney. Table 3 offers an overview of the common corpus types utilised in urban research, based on primarily textual and non-textual resources.
Table 3. Common corpus types used in urban research
Common Corpus Types
|
(Primarily) textual corpus
|
Official documents; city council meetings; interviews; city plans; urban development projects; media reports; journal articles; meetings minutes; court records; petitions; campaign documents; diaries; emails; leaflets; newspaper articles
|
(Primarily) non-textual corpus
|
Films and videos; participant observation; marketing and advertising documents; audio tapes of meetings and interviews; animation; social media (Instagram, Twitter, Facebook); advertisements; travel dairies; site visits; maps
|
29In this section, we will discuss findings of our analysis from two perspectives: the field of inquiry and the corpus of analysis. Our examination revealed that DS has been embraced by urban scholars across various domains of urban research. While certain areas such as urban planning, housing research, urban policy studies, and environmental policy studies predominantly feature DS, there is a burgeoning interest in its application across a spectrum of urban studies fields. This increasing interest stems from the belief that DS provides urban scholars with a robust methodological framework to delve into the discursive dynamics of urban questions. For instance, Arapoglou (2004: 105) posits that DS “can provide the methodological tools to explore the dialectical relationship between discursive and social practices of the actors involved in pluralistic forms of governance” (Arapoglou 2004: 105). According to Jacobs (1999: 212), “Discourse analysis provides the social scientist with a supplementary method that can reveal the different ways that terms and concepts are drawn upon to justify urban policy intervention”. Matthews and Satsangi (2007: 509) suggest that techniques of DS “provide a useful set of methodological tools for analysing planning documents to uncover the power and economic realities that belie major developments and policy directions”. Other scholars use DS “as a vehicle of policy and project analysis” (Michalec et al. 2019: 4), “as a theoretical and methodological perspective” (Huovinen et al. 2017: 20), and “as an alternative” way of analysis (Weil 2019: 1284). These diverse applications underscore the robustness and reliability of DS as a framework for investigating urban questions, suggesting significant potential for broadening the disciplinary scope of research that utilises DS.
30From the point of view of the corpus of the analysis, our findings suggests that in many instances, the corpus of the study has been restricted to a single document or multiple resources of a single type, primarily comprising text-based materials. This is referred to as “monomodality of corpus”. Indeed, scholars have long recognised the significance of incorporating diverse document types to enrich the corpus. One consequence of this diversification has been the expansion beyond the text-centric discourse to encompass non-verbal materials such as images, videos, and audio, thus creating a multimodal corpus. In the remainder of this section, we will demonstrate how scholars in the field of DS and urban research have advocated for the diversification of the corpus.
31To diversify the corpus of analysis, three preliminary strategies have been sproposed. Firstly, scholars advocate for the integration of complementary methods such as ethnographic techniques, conversation analysis, participant and non-participant observation, and interviews. This approach aims to prevent the confinement of research solely to published policy documents. According to Fairclough, interviews can enrich the corpus of the data and elicit participants’ interpretation of the topic under question (Fairclough 1992: 227). In the field of urban studies, “Interviews with policy actors represent an interpretive account of discursive processes involved in text production and text interpretation… The written policy texts are concrete outcomes of discourse practice, while the interview data provides an interpretive account of the socio-cognitive and socio-historical practices that were involved in producing and interpreting the policy changes” (Marston 2004b: 7).
32The second approach to diversifying the corpus of analysis involves extending beyond text-based resources and incorporating non-verbal materials. Urban researchers have critiqued the excessive emphasis on textual analysis and underrating the non-verbal dimension of discourse (Jacobs 1999). As Marston (2002) articulates, prioritising the textual dimension of social practices results in inadequate capturing of non-verbal aspects of communication and cultural meaning, such as bodily gestures, dressing codes, and vocal intonations. Furthermore, since linguistic methods predominantly centre on words, a text-oriented analysis may have limited applicability in studying planning discourses, which encompass a multitude of discourses, involve numerous actors, and are shaped by multiple social constructs influencing each other (Kumar and Pallathucheril 2004). In this sense, non-verbal materials integral to the produced discourse and carrying significant meanings are often disregarded and underestimated.
33In DS, the imperative for diversifying the corpus has been explored under the concept of “multimodality”. As articulated by Reisigl and Wodak (2016), texts represent only one facet of discourse, and different types of visualized, written, and oral texts, which could be assigned as genres, objectify linguistic actions. The notion of multimodality and multimodal discourse was introduced in a more articulated way through the ground-breaking works of Kress and Van Leeuwen in “Reading Images” (1996) and “Multimodal Discourse” (2001). In these works, Kress and Van Leeuwen elucidate a “grammar of visual design” to illustrate the significant semantic dimension inherent in pictorial structures, while emphasising that communication arises from a social foundation. Multimodality of discourse resonates with the term ‘transmedial narratology’ that underlines the importance of non-textual modes of narrative Ryan, 2016; Ryan et. al., 2004). Transmedial narratology advocates going beyond language-based narratives to narrative media such as film, drama, comics, and video games, and thus refers to the migration of narrative content across various media. It argues that narrative is not limited to literary text, but encompasses various media such as text, image, and sound, in various formats such as books, films, and internet.
34According to van Leeuwen, much of the work conducted in DS remains predominantly monomodal, focusing only on written and spoken language (van Leeuwen 2012). Kress contends that to understand the “boundedness of language” and its uses, we should understand the effect of other modes of communication co-present in any given text (Kress 2000). In fact, meaning does not solely reside within language; rather, it is distributed across various modes. This broadens the multidisciplinary scope of DS, as many other established disciplines such as media and visual studies converge with the predominantly text-oriented tradition of linguistics. The concept of mode refers to a socially shaped and culturally given semiotic resource to convey meanings, and may encompass a variety of modes including image, writing, layout, music, gesture, speech, moving, moving image, sound-track, and three-dimensional objects (Kress 2010). Modes are defined as “semiotic resources which allow the simultaneous realisation of discourses and types of (inter)action” (Kress and Leeuwen 2001: 21). They present material resources employed in recognisable and consistent ways to articulate discourse. Hence, multimodality refers to the “use of several semiotic modes in the design of a semiotic product or event, together with the particular way in which these modes are combined” (Kress and Leeuwen 2001: 20). In other words, multimodal discourses mirror, constitute, or even mask social reality, and social reality is constructed and (re)produced through the orchestration of multimodal discourses.
35The multimodality of discourse carries significant social and political implications. Power dynamics, truth regimes, knowledge construction, and control mechanisms are all constructed and exercised through different modes of discourse. The multimodality of discourse is governed by social, cultural and institutional norms that vary from one context to another. The selection of appropriate modes and their combination reflects socio-cultural acceptability, institutional legitimacy, and instrumental capabilities within a given context. In recent times, the multimodality of discourse has played a pivotal role in the power-control dialectics. The advent of the digital communication revolution has broadened the array of emerging discourse modes accessible to the public, challenging hegemonic access to power by the powerful and the advantaged. This shift has potentially favoured marginalised groups, enabling them to mobilise, protest, and resist. This reconfiguration of power, as Kress notes (2010), signifies a transition from “vertical” or hierarchical power structures to “horizontal” and more participative ones. Alternative, non-verbal modes of communication, such as photography, have the capacity to bring silenced discourses into the public debate, amplifying marginalised voices. In this sense, “multimodal discourse is a vehicle for subversive and alternative worldviews, while verbal discourse often favours the status quo as well as dominant interests and positions” (Jancsary et al. 2016: 189).
36To acknowledge the multimodality of discourse, some scholars have introduced Multimodal Critical Discourse Analysis (MCDA) that links the multimodality of discourse to DS (Djonov and Zhao 2014, Machin 2013). It shows how authors utilise various modes, such as images, photographs, diagrams, and graphics, to construct meaning. It posits that the selection of visual elements not only represents but also construct the world; visual communication shapes and maintains a society’s ideologies, at the same time creates, maintains, and legitimises certain types of social practices. In this sense, visual communication, as well as language, both shape and are shaped by society (Machin and Mayr 2012).
37The third strategy employed to diversify the research corpus involves opening up the analysis to actors and elements beyond the institutional body of policy-makers. This entails engaging with individuals who become the object of policy discourse and subject to dominant discourses in their everyday lives (Marston 2002). Watt (2008) contends that DS in urban studies has remained mainly within the limits of the discourses of policymakers, hereby neglecting the voice of non-government interest groups who challenge prevailing urban policies. This observation is noteworthy, considering that as non-governmental groups, NGOs, social campaigns, and local organisations also contribute to discourse production. Their conflicts and tensions with the official or governmental discourse production influence their discursive practices.
38Considering the proposed strategies aimed at diversifying the analysis corpus, researchers may encounter two critical challenges when dealing with the analysis corpus. Firstly, researchers must navigate critical decisions concerning the selection of the corpus and inclusion of the most appropriate resources in the body of analysis. As Jacobs suggests, “there is a danger that important material is neglected or that too much significance is made of specific words or texts whilst economic and political factors are overlooked” (Jacobs 2004: 820). Goodchild and Cole (2001: 104) raise concerns about a tendency towards fragmentation, wherein the focus is on the meaning and usage of words, without specifying which words warrant analysis: “Given the diversity of words in any policy context, discourse analysis risks degenerating into a confusing series of episodic narratives that cannot be put together”. In fact, researchers should make crucial decisions across two domains: determining the type of materials to include in the analysis and selecting which materials within each category merit analysis. While the former pertains to the multimodality of the corpus, the latter concerns the volume and quantity of materials.
39The second challenge lies in the predominant focus of DS in urban research on the end output, such as official documents and reports, while overlooking the procedural dimension of discourse production. In this sense, what is missing is understanding the process and dynamics of discourse production, power relations and competition between involved actors, the evolution of discourse over time, and the struggle involved in formulating elements of final production. In other words, the temporal aspect is neglected and underestimated. One approach to addressing the temporal dimension of discourse production has been analysing drafts pf policy document and exploring wording, presentation, and argumentation changes. This type of comparative study can reveal how discourse evolves over time, and how key elements of documents adapts to reflect shifting power dynamics, compromises, and competitions among different stakeholders. Another method entails conducting repeated data collection throughout the study and examining the transformation of opinions and sentiments on a certain subject over time.
40Acknowledging the temporal dimension of discourse, or what we may term as discourse transformation, stems from the understanding that key terminologies utilised in policy documents are essentially value-laden: “The way language is used in government documents is value-laden and embedded with cultural, ideological and political objectives” (Oakley 2007: 281). This implies that these terms can undergo shifting interpretations or be ascribed different meanings to accommodate policy changes that align with the expectations of specific groups and stakeholders. A genealogy of discourse unveils such shifts and elucidates the rationale behind them. As Cole and Goodchild (2000: 352) articulate, discourse must “be analysed, at least at first, through genealogy, a study of its development, comparing different meanings at different times. Discourse expresses the sense that a body of ideas is embedded in a historical context, and therefore a comparison of the use of the terms in different time periods of policy formulation and implementation can generate some useful insights”.
41In the wake of the “linguistic turn” in philosophy and social sciences that problematised language as a form of action that generates meanings and connotations, and in line with the argumentative turn in planning and policy studies, which underscored the role of language and argumentation in planning and policy-making, urban scholars have embraced DS as both a theoretical framework and methodological approach to analyse urban questions across various disciplines of urban studies. To shed light on the growing attention of urban scholars to DS, we conducted an extensive analysis of pertinent publications to probe two key questions: firstly, which domains of urban studies have availed themselves of DS and for what reasons (field of inquiry), and secondly, what types of materials constitute the body of data analysis (corpus of analysis).
42Our analysis suggests that while DS has been widely employed to investigate urban questions in fields such as urban planning and theory, urban policy, housing research, and environmental policy, other areas have also benefitted from what DS offers to the researchers. This interest stems from the belief that DS provides scholars with a robust theoretical framework, a methodological tool, a supplementary analytical method, and an alternative way to explore the discursive aspects of urban questions. The corpus of analysis, our study finds out, has predominantly been restricted to singular items or multiple items of the same type, primarily comprising text-based materials. This monomodality of corpus and monopoly of text-centred discourse, despite being a routine approach, stand in contrast to the growing recognition that discourse is inherently multimodal, encompassing various non-verbal modes such as imagery, music, gestures, motion, moving image, soundtrack, and three-dimensional objects. Diversifying the corpus, in terms of expanding the size and typology of the corpus, enables researchers to attain a more holistic understanding of the inherently multimodal nature of discourse production.
43Drawing upon the insights gleaned from this analysis, I will bring the paper to a close by delineating nine key challenges faced by Discourse Studies (DS) in urban research, which resonate with the principal themes of the study – namely, the field of inquiry and the corpus of analysis. Furthermore, I will propose avenues for future research to navigate these challenges effectively.
441. The utility of DS in urban research has seen growth in specific research domains; however, the potential and capacity of the application of DS in many other areas of urban research have not been properly explored. There is a need to broaden the scope of research to encompass areas where DS has been less frequently employed.
452. There is a risk of overlooking relevant materials or exhibiting bias in the selection of materials for analysis, favouring those that align with our methodology and arguments. Careful consideration must be given to selecting the analysis corpus to avoid biased discussions and oversimplified generalisations.
463. The analysis corpus has predominantly consisted of single items or a limited number of items of a similar nature. A prevalent trend in conducting DS in urban research has been to focus on published policy documents and legislative papers. Diversification of the analysis corpus is essential to incorporate a larger variety of materials and achieve a deeper understanding of the study subject.
474. The analysis corpus has primarily centred on textual sources; neglecting other modes of discourse. This monomodality of the corpus has resulted in a partial and insufficient understating of urban questions and the multidimensionality of discourse production. Recognising the multimodality of discourse is crucial, and if relevant, the analysis corpus should be diversified to include non-verbal modes of communication.
485. Embracing multimodality implies considering the multisensory aspect of discourse, including body language, vocal properties, and gestures, s part of the analysis corpus. Researchers must enhance their proficiency in DS to incorporate non-textual methods effectively. Utilising new techniques and tools of inquiry is imperative for investigating these non-verbal dimensions of discourse.
496. Social media platforms have emerged as significant repositories of both verbal and non-verbal information regarding urban issues. These platforms should be integrated into the analysis corpus, with appropriate tools employed for analysis.
507. Restricting the analysis corpus to textual sources may create a misconception that research is confined to office-based activities dealing solely with published documents, detached from lived experiences. Embracing the multimodality of the corpus allows for engagement with real-world problems and offers insights into lived experiences.
518. The analysis corpus has largely been shaped by the voices of the powerful and governing agencies, often overlooking the perspectives of marginalized groups. DS in urban research should move beyond dominant discourse and acknowledge the voices of the marginalized to address power imbalances.
529. The analysis corpus primarily comprises and product such as policy documents and reports, leading to an oversight of the procedural dimension of discourse production. To address this, we first need to introduce the time element into the study and analyse the evolution of discourse over time, a process that can be referred to as the genealogy of discourse. One approach to achieving this is by examining changes in policy documents in terms of wording, terminologies, and arguments, which can reveal the dynamics of power relations among involved agencies, institutions, and stakeholders. Additionally, certain elements of the corpus, such as interviews, can be repeated throughout the research to document changes in participants’ perceptions regarding the urban question under investigation.
53In conclusion, the future of DS in urban research hinges on our ability to acknowledge these challenges and develop strategies and techniques to address them effectively.