Abstract
There is a long tradition of linguistic research on political discourse, but little attention has been paid to what the concept of political discourse itself encompasses. With this in mind, this article aims to understand what types of discourse are categorized as ‘political’ in linguistic research and what their characteristics are (form, type of actors, policy domains, geographical coverage). To this end, we conducted a systematic literature review of 164 scientific articles from the Scopus database. Overall, the findings show that political discourse is generally limited to the discourses of (institutionalized) political elites and most specifically to oral monological speeches. The review also highlights discrepancies regarding the geographical scope and the policy domains covered by the empirical analyses, more specifically a bias toward the Western world and issues related to external defense policies, justice and home affairs.
Keywords
Introduction
A major question of political discourse analysis is to determine what is political and what is not. In this respect, an analysis of the literature shows the wide variety of referents found under the label ‘political discourse’. For example, studies of political discourse analyze productions as diverse as monological speeches (Wodak and Boukala, 2015), parliamentary debates (Gruber, 2015), diplomatic condolences (Fenton-Smith, 2007), press articles (Musolff, 2017), campaign posters (Jones, 2014), tweets (Kreis, 2017), citizen forums (Perrez and Reuchamps, 2012, 2015) or even graffiti (Hanauer, 2011).
Under the influence of Van Dijk’s works on political discourse analysis, linguistic research on political discourse has mainly focused on speeches produced by political elites. This is in line with Van Dijk’s definition of political discourse, which is based on three dimensions: the actors, the political scope of the discourse and the context of communication (Van Dijk, 1997: 12–14). Following this definition, a discourse is considered ‘political’ when it is produced by a political actor carrying out a political action (e.g. to govern, legislate, protest or vote) in an institutional context of communication (e.g. parliamentary debates, public speeches, official addresses). This approach has thus left other forms of discourses, such as media discourse on political issues or (political) citizen discourse, out of the scope of political discourse analysis.
Other authors, such as Fairclough (2006), argue that the domain of politics should not be delimited ex ante but should instead be understood as ‘socially constructed’ (p. 33). Okulska and Cap (2010), for instance, point out that political discourse should be conceived as ‘socially oriented studies of polity and/or policies, located at the intersection of political/public discourse and political/social institutions, and studies conducted within the Critical Discourse Analysis paradigm with emphasis on politics’ (p. 4). As indicated by Chilton and Schäffner (2002: 6), the concept of political discourse can be understood in its institutional sense but it can also have a broader meaning and refer to any type of power negotiation at any level.
Against this background, the purpose of this article is to take stock of current academic knowledge on this issue by conducting a systematic review of empirical studies on political discourse over the last 20 years. Given the difficulty to define a priori what is political and what is not, a systematic review could help analyze and categorize the numerous empirical studies using this label. The article starts with a presentation of the method and scope of the systematic review. The next section presents overall trends in linguistic research on political discourse, along with four main streams of research on political discourse, namely studies focusing on (1) political elites (2) media discourses (3) actors from the civil society and (4) what we have termed ‘mixed’ discourses (i.e. when several categories of actors are jointly analyzed by the author(s)). The last part of this article is dedicated to conclusions and recommendations for further research.
Method and selection procedure
This essay presents the findings of a bibliometric analysis based on the ‘Preferred Reporting Items for Systematic Reviews and Meta-Analyses’ method (PRISMA) (Moher et al., 2009). The PRISMA method provides a transparent and reproducible review of the literature.
The review includes scientific articles published in English after 1997 and indexed in Scopus. For feasibility reasons, the review is limited to articles published in seven specialized journals. 1 The journals were selected on the basis of qualitative and quantitative criteria. 2 To identify ‘political discourse’ articles, the following keywords – covering the notion of ‘political discourse’ in a broad sense – were used: political discourse, political speeches, political corpus, political debates and public discourse. Finally, the period of analysis includes all the studies (1998–2018) that have been carried out since Van Dijk’s pioneering article was published in 1997. The search resulted in an initial list of 218 articles. Based on this preliminary list, two members of the research team independently carried out a first ‘screening’ of the articles based on the titles, abstracts and keywords. In the end, the review accounted for a total of 164 articles. Based on a common grid, the articles were coded by two – trained – research assistants. In addition, one author of the article also coded the article independently to ensure the quality of the coding, 3 hence making a total of two coders per article. If there was a doubt about the relevance of including an article in the sample or about the coding, a discussion was held among the authors until a consensus was reached.
Nevertheless, our implementation of the PRISMA method may show some limitations. First, our review only includes journal articles and does not consider studies that were published in other formats (books, chapters, etc.), leading to a potential publication bias. The latter, however, is counterbalanced by the size of our sample which includes a range of authors who have published the results of their research in journal articles as well as in books or book chapters. Another potential bias involves the focus on articles published only in English, although this is compensated for by the great number of countries, languages and various origins of the authors analyzed (i.e. almost 50 different countries are covered by the review).
Twenty years of research on political discourse
Overall, prior to 2008, an average of 3.5 articles focusing on political discourse were published every year. Starting from 2008 (and until 2018), the average rose up to 11.8 articles a year, indicating a growing interest for political discourse analysis. This increase is also associated with a progressive broadening of the geographical scope of studies. 4
Other major trends are observed concerning the policy domains, the type of actors and the form of analyzed discourses (see Table 1). First, three policy domains attracted most of scholars’ attention: international relations and defense 5 (21.82%), electoral campaigns (15.76%) and, starting from 2007, justice and home affairs (including immigration issues – 18.18%). Second, following Van Dijk’s works on political discourse analysis, a majority of articles (105; 64.02%) focus on discourses produced by political elites. In contrast, 8.53% (14) of the articles examine the (political) discourse produced by media actors, 7.31% (12) focus on civil society actors, while 20.12% (33) deal with the political discourses produced by several categories of actors at the same time (labeled ‘mixed discourses’ in this contribution). Moreover, our findings also emphasize that scholars favor the analysis of oral discourse (61.58%, 101) 6 over written discourse (23.17%, 38), whereas the combined analysis of oral and written forms of discourses occurs in 15.12% (25) of the articles.
Summary of the findings of the systematic review.
As a second step, we discuss four streams of research on political discourse in linguistic studies.
Political discourse produced by political elites
The first research stream relies on Van Dijk’s (1997) definition of political discourse and therefore brings together studies focusing on discourses produced by political elites in an institutional context with the aim of carrying out a political action. In total, 105 articles of 164 (64.02%) fall into this stream of research.
Within the category of political elites, some political actors have attracted more attention than others. 7 This is the case of members of the executive (President, Prime Minister, Ministers) and candidates for office (57; 54.28%). To a lesser extent, political parties and party leaders (16; 15.23%), members of parliaments (MPs) (13; 12.38%) and international actors, such as the United Nations or the European institutions (3; 2.85%) are also studied. Finally, some articles analyze at least two different types of political actors within the category ‘political elites’ (8; 7.61%).
The great majority of studies concentrate on oral discourse – though mostly based on written transcriptions (81; 77.14%) – while only 11 articles of our sample (10.47%) envisage discourse in its written form. Thirteen articles (12.38%) include both types of media. As such, the importance of oral discourse analysis is not surprising considering that the majority of the articles in our sample are based on the analysis of monological speeches (41 of 105, 39.04%), interviews in the media (Television and Radio; 17; 16.19%) or parliamentary debates (9; 8.57%). It is also worth noting that 26 articles (23.81%) combine the analysis of several genres of political discourse. There are also clear patterns regarding the geographical coverage and the policy domains put forward in these studies. Most studies focusing on political elites deal with Western countries (from Europe and North America) or with the European Union (72 of 105; 68.57%), and cover three policy domains: external relations and defense (23; 21.90%), election campaigns and referendums (22; 20.95%) and justice and home affairs (16; 15.23%).
Finally, if we consider the main objectives of the studies included in our sample, the following tendencies are observed: a major share of the articles aims at studying discursive strategies and questions related to the framing of political issues, events or actors (54 of 105; 51.42%). Major topics often associated with discursive and framing analyses are ideology or identity construction (see. Wang, 2017; Wodak and Boukala, 2015), populism and racism (Mayaffre and Scholz, 2017), or peace and war issues (Abid and Shakila, 2016; Anchimbe, 2008). Other studies concentrate to a large extent on specific linguistic characteristics of the discourses, such as Wang and Liu’s (2018) analysis of Trump discourses, or Roitman’s (2014) study of the use of pronouns by French presidential candidates. The rest of the studies deal with various topics, such as the analysis of conversation as discourse (see Cienki and Giansante, 2014); particular speech acts or features of speeches (e.g. political apologies or condolence messages; see for example, Murphy, 2015); argumentative studies (see. Fairclough and Fairclough, 2011) or metaphor studies 8 (see Lu and Ahrens, 2008).
Political discourse produced by media actors
This research stream focuses on discourses produced by journalists or editorial leaders (including op-ed articles). In total, this category includes 14 articles of 164 (8.53% of the sample).
With regard to the genres, forms of discourse, geographical coverage and policy domains covered by these studies, the following elements should be considered. First, there is a clear tendency to study written or online newspapers (12; 85.71%), with only one study focusing on an (oral) news broadcast (1; 7.14%). The last article of our sample focuses simultaneously on written newspapers, news broadcasts and face-to-face interviews (1; 7.14%). Second, when looking at the geographical coverage, two specificities can be highlighted: on the one hand, all three articles focusing on translation studies (see infra) cover at least one country from the Middle East and always examine translations between English and Arab. Consequently, these studies focus on several continents at the same time (3; 21.42%). Besides this observation, an important share of studies focuses on European countries (5; 35.71%), followed by Asia (3; 21.42%), the Middle East (2; 14.28%) and North America (1; 7.14%). Not a single article within this stream of research appears to deal with political media discourse in Africa, South America or Oceania.
In terms of analytical focus, the 14 articles of this stream can be sorted into three main categories. A first set of studies (6 of 14; 42.85%) deals with the framing of particular actors, institutions, countries or events within the discourse of media actors, such as the framing of immigrants in America (Otto, 1999), the concept of Europe in a French newspaper (Le, 2003), or on the representation of Ireland in German newspapers after the rejection of the Nice Treaty (Kelly-Holmes and O’Regan, 2004). It can also deal with the representation of specific actors (Iranian leader, Mohd Don and May, 2013) or events in the Israeli–Palestinian context (Livnat, 2011). A second collection of studies (5; 35.71%) considers questions related to identity building, as well as on the impact of ideologies on media discourse. For instance, Flowerdew and Leong (2007) focus on the construction of the notion of patriotism and cultural identity in postcolonial Hong Kong, Georgiou (2010) studies the orthography of toponyms in Cyprus, while Chiang and Duann’s (2007) analysis examines the impact of newspapers’ ideologies when covering the issue of severe acute respiratory syndrome (Chiang and Duann, 2007). The third category includes articles (3; 21.42%) analyzing the different factors influencing the translation of political texts, such as newspaper articles on Iran’s nuclear program (Aslani, 2016; Azodi and Salmani, 2015) or more globally, on the news representations in Western and Middle Eastern media sources (Bazzi, 2014).
Overall, within these studies, the conceptualization of ‘political discourse’ remains elusive. In fact, only two articles provide a detailed discussion of how they understand ‘political discourse’ and unsurprisingly, both favor a definition that is not limited to professional political elites. Azodi and Salmani (2015) indeed consider political discourse to be ‘a complex form of human activity which is based on the recognition that politics cannot be conducted without language’ (p. 183). Drawing from Schäffner (2004), the authors also point out that translation and translation studies are inherently political. Similarly, in his article investigating the orthography of place names in Cyprus, Georgiou (2010: 146) discusses the functions of political discourse, like reflecting the nationalist/culturalist position at hand, or fueling processes of unification/differentiation, language/culture maintenance and official recognition.
Political discourse produced by civil society actors
The articles focusing on discourse produced by actors from the civil society, in the broad sense of the term (i.e. citizens, social movements, experts, scholars and religious leaders), compose the third stream of research. In total, 12 articles of 164 (7.31%) analyzed the discourse produced by such actors. Overall, these studies envisage political discourse as a social construct (Fairclough, 2006). Similarly, following Pelinka’s (2007) argument that any issue can potentially be political, they also include discourses produced by actors from the civil society. Within this category, citizens in particular constitute the core of the research stream (8; 66.66%), while the rest of the articles considered various actors from civil society, such as religious leaders, scholars, interest groups or social movements (4; 33.33%).
A major characteristic of these studies is the diversity of genres and empirical data, which is indicative of the authors’ conception of what political discourse entails. More precisely, a first set of studies consider different types of citizen meetings, such as participatory democracy debates (Mondada, 2013) or Israeli–Palestinian ‘dialogue’ events (Zupnik, 2000). A second set of studies investigates written comments made by citizens on social media and newspaper websites, such as the analysis of users’ interaction on YouTube (Boyd, 2014) or online political discussion in UK and French Newspapers’ forum sections (Lewis, 2005). A third set of studies focuses on the production by citizens of musical and visual artifacts with political meaning. For example, Hanauer (2011) analyzes graffiti on the separation wall in the contested space of Abu Dis; other authors have looked into cartoons, posters, music videos or book covers produced by civil society actors (Angermüller, 2012; Arman, 2018; Lou, 2017; Way, 2016). Finally, a last set of studies focuses on more classical materials, such as the analysis of speeches by members of Protestant institutions in Northern Ireland (2007) or official documents of the military junta and religious leaders in Argentina (Bonnin, 2009).
Within this stream of research, none of the 12 articles offer a detailed exploration of the meaning of ‘political discourse’. Obviously, the actors producing the discourse (i.e. civil society actors) are not the core of the authors’ conceptualization. Indeed, in these studies, the discourse of civil society actors is only deemed political because its context and/or the content of the discourse are politicized.
In practice, some of the studies from this category bring together analyses of musical and visual artifacts. These artifacts are conceived as ‘political’ because they convey a political meaning. For example, Way (2016) has analyzed music video produced during Turkey’s June 2013 protests. Going in the same direction, Hanauer (2011) has studied graffiti in the context of the Palestinian–Israelian conflict and emphasized that ‘for Palestinians graffiti writing was “an intervention in a relationship of power” and a “critical component of a complex and diffuse attempt to overthrow hierarchy”’ (p.301). A last example is the study of Lou (2017) who has analyzed artifacts produced during the 2014 Hong Kong suffrage protests. The author ponders the ways in which cartoons representing ‘umbrellas’ could bring political meaning to a simple object. Other studies from this set focus on citizens’ comments on platforms like YouTube, or newspaper forums. These studies are regarded as epitomes of political discourse analyses because the citizens react to specific political events: Obama’s 2009 Inaugural Address for Boyd (2014) and online discussions of current affairs and political themes for Lewis (2005). The same rationale can be applied to the third group of studies in which civil society actors are politically institutionalized, because these are representative of larger social groups. The articles of Zupnik (2000), which reflect on discourse produced by Israeli and Palestinian representatives during dialogue meetings launched since the Intifadah, Bonnin (2011), analyzing the discourse of religious leaders and the military junta in Argentina and Stevenson et al. (2007) on the discourse of members of the Orange Order in Northern Ireland illustrate this situation. Finally, a last group of studies concentrates on new forms of public participation in decision-making processes. A typical example of this can be found in Mondada’s (2013) study, of a participatory democracy project related to urban planning in Lyon (France) in which the author states that urban democracy has been conceptualized at different levels, from abstract political approaches to empirical analyses of embodied situations in which social actors discuss, dissent, and co-construct different versions, revealing how urban space is the result of constant controversy and negotiation. (p. 41)
With respect to the policy domains, the geographical scope and the forms of analyzed discourse, several observations should be highlighted: first, the domain of justice and home affairs attracted most of the attention (5; 41.66%). The rest of the studies deal with various policy domains, including economic regulations, external relations and defense, social policy, education and culture. Second, discourse by civil society actors in the Western world (Europe – France in particular – and North America, 8 of 12; 66.66%) are the most frequently analyzed. In contrast, no journal article deals with civil society discourse in Africa or Oceania, while studies on Asia (1; 7.14%), South America (1; 7.14%) and the Middle East (2; 14.28%) are relatively limited. Third, regarding the form of the discourse, a great majority of articles rely on written discourse (6 of 12; 50%), while four articles concentrate on oral discourse (4; 33.33%) and two articles analyzed both oral and written forms of discourses (2; 16.66%). This finding suggests that, whereas the discourses of the political elites rely importantly on oral material, media and civil society, discourse analysts concentrate on alternative empirical materials. Indeed, civil society actors – and to a lesser extent media discourse – are less institutionalized and have fewer opportunities to take a public stance.
Political discourse from a mixed perspective
The last research stream concentrates on articles that propose a joint analysis of the political discourse of at least two different types of actors (i.e. political elites or actors from the media or from the civil society). Overall, 33 of 164 articles (20.12%) have adopted this approach.
At first sight, this research stream seems to be the most encompassing one, as it focuses at least on two different categories of actors at the same time. Yet, in practice, the bibliometric analysis reveals an interesting, yet unsurprising, tendency: the analysis of political elites remains central within this research stream. Indeed, political elites are studied in combination with other actors (i.e. civil society or media actors) in 32 articles of 33 (96.96%). In particular, political elites are studied in combination with civil society actors in 16 articles (50%), with media actors in 10 articles (31.25%) and with both civil society and media actors in six cases (18.75%). The only exception is the article by Sidiropoulou (2013) that focuses on journalistic (media) discourse and academic (political) discourse.
This stream of research also presents a great variety of analytical foci. On the one hand, a majority of studies concentrate on discursive strategies and questions related to the framing of a political issues or events (10; 30.30%) and on the use of metaphors (4; 12.12%). On the other hand, an interesting – but not really unexpected – finding is the strong focus on studies taking the analysis of interactions between actors as their starting point (6; 18.18%). Thornborrow and Fitzgerald (2013), for example, have focused on the frameworks for interaction in a UK political radio phone-in between politicians and listeners. Similarly, Lehti (2013) investigated conversation-like interactions between politicians and citizens on political blogs. One advantage of adopting a broader understanding of political discourse is indeed to analyze citizen-political elites’ interactions. Another group of studies focus on apologies by Israeli public and political figures (2; 6.06%) (Kampf, 2008, 2009). A last collection of studies (3; 9.09%) concentrates on linguistic characteristics of discourses, such as the use of pronouns (Cramer, 2010) or the use of dialects (Hernández-Campoy and Cutillas-Espinosa, 2010). The remaining articles (8; 24.24%) were brought together in one broad category ‘Miscellaneous’ as they are really diverse in scope. As an illustration, the category includes studies focusing on translations, argumentation, testimonies, or dealing with specific analysis foci, such as the use of the notion of ‘risk’ in discourse or the impact of e-mail on the routine of radio (political) broadcast.
An interesting question regarding this stream of research is how the authors conceived the notion of ‘political discourse’. In the systematic review, seven articles (of 33) provide a more detailed exploration of concepts related to political discourse. They adopt a broad definition of political discourse, with Chilton and Schäffner (2002), Schäffner (2004) and Chilton (2008) as the most cited authors. For instance, Durović and Silaški (2010) used the term political discourse as an ‘umbrella term encompassing both types of discourse’, that is, political discourses used by politicians and political media discourses (p. 242). For his part, Gavriely-Nuri (2008), studying the use of metaphors in Israeli political discourse, defined political discourse as ‘a discourse relating to political issues. That is, this discourse refers to a specific subject and content rather than to the identity of the figures taking part in its construction’ (2008: 18). With this type of framing, scholars can include various actors in their analyses, including politicians as well as military officers, journalists or civilians. What these studies have in common is their conception of political discourse as being not limited to political elites. Instead, they emphasize the prevailing role of language in the business of politics (Chilton, 2008).
As has been observed for the other categories of studies, the research composing the fourth stream mainly focuses on discourse produced in Europe (18; 54.54%) and to a lesser extent in the Middle East (4; 12.11%, concentrating mainly on Israel, except for one study) and North America (3; 9.09%). Only two studies focus either on South America, Oceania or on several continents (2; 6.06%). Finally, the less studied continents are Africa and Asia, with one study on each (1; 3.03%). In terms of policy fields covered, the tendency is the same as for the other research streams: external relations and defense on the one hand (7; 21.21%), and justice and home affairs on the other (7; 21.21%) are the two most studied policy fields, with less attention than for other research streams given to electoral politics than for other research streams (3; 9.09%). Finally, regarding the form of the political discourses, there are more articles focusing on oral discourses (15; 45.46%) than on written ones (9; 27.27%) or on both oral and written discourses (9; 27.27%).
Conclusion and avenues for further research
The objective of the present article was to review articles focusing on political discourse over the last 20 years. Our findings confirm that, in practice, political discourse is generally limited to the discourses of (institutionalized) political elites and most specifically to oral monological speeches. As a consequence, the discourses produced by non-professional actors appear to be underrepresented. In the current political context, where citizen actions are developing new forms of interaction and participation both off- and online, where civil society pleads for transformations and where, in general, participatory democracy is gaining ground, it seems appropriate to expand the spectrum of political discourse analysis to these kinds of actors.
Studying different types of political discourses could contribute to a more fine-grained understanding of what constitutes political discourse. Indeed, while the various types of discourse produced by political, media and civil society actors have political implications, conditions and consequences, additional research should be conducted to analyze if the discourse produced by these actors constitutes a political genre in itself. 9 On this matter, in a first study (Perrez et al., 2019a), we have reflected on the contours of the political genre and its textual registers by comparing formal linguistic characteristics of three sub-genres of political discourse (parliamentary debates, televised debates with political elites and citizen forums) on a similar political issue (i.e. federalism in Belgium). We have showed that there was very little linguistic homogeneity between these sub-genres, suggesting that they could be considered different textual registers. The variation between the different types of discourses appears to be explained by factors related to the context in which these discourses were produced, rather than by a common core of linguistic features that would distinguish political discourses from other homogeneous genres. To put it differently, whatever the political conditions or consequences of the discourse, it seems that the communication context matters most to account for the variation in the textual registers. Consequently, further research including other types of political corpora is needed to refine our understanding of the different textual registers that constitute political discourse. This would also allow us to broaden our knowledge of the relationship between the type of actors producing discourse with political consequences, the communicative context in which the discourse is produced and their textual registers.
Taking this one step further, studying the discourses of various actors on similar political issues could also open a promising ground for further – comparative – research (Perrez et al., 2019b). Comparing the discourses of political elites, media actors and citizens on a similar political issue would help assess and measure the distribution and circulation of frames and arguments between actors. Considering the increase of protest movements toward the traditional political institutions, as well as the rise of nationalist movements in Europe and beyond, it is particularly relevant to undertake research focusing on the circulation of frames both from a top-down (i.e. from the political elites down to civil society actors) and a bottom-up (i.e. from civil society actors to political elites) perspective.
Our systematic review has also highlighted discrepancies regarding the political (sub-)genres, the geographical scope and the policy domains covered by the empirical analyses. The review pointed to a slow but positive increase of studies covering, among others, continents such as Oceania, South America and Africa. In addition, since 2010, there has also been a slight but growing tendency to rely on digital content, such as blog discussions, YouTube comments, tweets, music video lyrics or direct democracy meetings. This increase should be encouraged in the future, as these studies allow researchers to fill in knowledge gaps and to further our understanding of what political discourse means.
Last but not least, we could investigate internal differences within political discourse genres. Variation can be observed across continents, policy domains, sub-genres, the form of the discourse and the various actors that speak the language of politics. Fostering comparative research would be a promising ground in this respect, as this would allow us to systematically highlight convergences and divergences between the analyzed political discourses (whether in terms of framing or on formal linguistic characteristics). In addition, it would be interesting to systematically analyze and compare the political actions of an actor (e.g. the EU) both from the inside (i.e. inbound perspective, political discourse within the EU) and the outside (i.e. outbound perspective, political discourse from outside the EU) on a similar issue.
Supplemental Material
Appendix – Supplemental material for Twenty years of research on political discourse: A systematic review and directions for future research
Supplemental material, Appendix for Twenty years of research on political discourse: A systematic review and directions for future research by François Randour, Julien Perrez and Min Reuchamps in Discourse & Society
Footnotes
Acknowledgements
We would like to thank the participants of this conference and also the editor of Discourse & Society for their valuable comments on earlier versions of this research. We would like also to express our gratitude to our colleague, Laura Pascolo, and research assistants, Sophie Fafchamps and Christina Piot, who helped with data collection and coding, as well as our colleague, Mathilde Mergeai, for the linguistic revision of this paper. Remaining inaccuracies are of course our sole responsibility.
Authors’ note
Previous versions of this article have been presented at several occasions, notably at the Aflico Jet 2018 conference on Corpora and Representativeness (3–4 May 2018, Université Paris-Nanterre).
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Fonds de la Recherche Scientifique – FNRS: [Grant Number PDR-T.1036.15].
Supplemental material
Supplemental material for this article is available online.
Notes
Author biographies
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
