Navigation – Plan du site

AccueilTous les numéros109VariaA crazy machine or a strong “livi...

  • ENS Éditions
  • ENS de Lyon

A crazy machine or a strong “living apart together” relationship?

The role of metaphors in citizens’ perception of Belgian federalism
Julien Perrez et Min Reuchamps
Cet article est une traduction de :
Folle machine ou solide relation « living apart together » ?


Cet article propose une analyse quantitative et qualitative de l’emploi des métaphores dans le discours politique citoyen. Basée sur des données issues de focus groups organisés à la fois dans les communautés francophones et néerlandophones de Belgique et portant sur la perception citoyenne du fédéralisme, notre étude suggère que les citoyens ont fréquemment recours aux métaphores quand ils parlent de processus politiques complexes, notamment pour expliciter leur(s) perceptions de ces processus. Parmi les domaines conceptuels mobilisés, il apparaît en outre que le domaine de la famille joue un rôle prépondérant dans les perceptions citoyennes du fédéralisme belge.

Haut de page

Texte intégral

The authors wish to thank Tina Krenmayr, the anonymous reviewers and the editorial board, and in particular Chloé Gaboriaux, of the journal Mots. Les langages du politique for their constructive comments on an earlier version of this paper. The authors are responsible for any errors that remain. The research reported in this article was supported by the Fonds spéciaux pour la recherche of the Université catholique de Louvain and of the Université de Liège, Belgium, and by the Fonds de la Recherche Scientifique – FNRS under grant no. PDR-T.1036.15.

1. Scope and issues

1In the wake of Lakoff and Johnson’s seminal work , much attention has been devoted to the study of metaphors, not as rhetoric figures but as conceptual tools structuring complex realities. According to defenders of CMT, metaphors play a major role on our perception and categorization of abstract entities and make it possible to structure our comprehension of complex processes. They reflect different conceptions of our environment.

2While conceptual metaphors occur in every area of life, the political domain remains one area in which metaphors play a dominant role. As Semino puts it: “it is often claimed that the use of metaphor is particularly necessary in politics, since politics is an abstract and complex domain of experience, and metaphors can provide ways of simplifying complexities and making abstractions accessible.” The importance of metaphors in politics has also been stressed by Charteris-Black (2011, 28) who highlights their contribution to the construction of more accessible “mental representations of political issues” and suggests their power resides in their ability to “activate unconscious emotional associations [which] contributes to myth creation”. The need for more research on the political impact of metaphors has therefore often been emphasized: “if metaphor is at the heart of cognitive framing then it should be crucial to political study” .

3Accordingly, scholars in linguistics and in political science have moved towards investigating metaphors in the political domain . Yet, until now, the predominant focus has always been on discourses by the political elites (see for instance Charteris-Black 2011). What’s more, while analysing the metaphors used by the elites illustrates how they frame the political debate (L’Hôte 2010, 2012, Mercenier et al. 2015), citizens’ discourses on politics should also be taken into account. In fact, with a few exceptions (see for instance Cameron and Maslen 2010), such an investigation has been left out of the analysis: “while research on metaphors in political discourse has flourished in recent years, the focus on elite communication has left metaphor’s wider capacity as a reasoning tool for citizens underexplored” (Bougher 2012, 149). More specifically, research on citizens' discourse can lead to two kinds of insights: on the one hand, it makes it possible to assess to what extent metaphors produced by the political elite are integrated in the citizens’ political reasoning, but on the other hand, it also offers the opportunity to look at how citizens “generate their own metaphors (i.e., spontaneous metaphors) to make sense of the political environment” (Bougher 2012, 149).

4It is precisely the aim of this paper to take a bottom-up approach, based on focus group discussions, to analyse how citizens perceive the Belgian political system and more specifically to assess what role metaphors play on this perception. Divided Belgium offers a very good case to tackle this issue as her evolution from a unitary state to a federal state has sparked much political tension between and within the two main language groups or communities: Dutch-speakers/Flemish and French-speakers/Walloons (Perrez and Reuchamps 2012). Because of the saliency of this issue, especially during the 2007-2011 period when Belgium was in a long political crisis (Deschouwer 2012, Deschouwer and Reuchamps 2013), the question arises whether the metaphors are present in citizens’ discourse on the workings and the future of their country.

2. Data and method

2.1. Citizen corpora

5While it is quite usual to focus on political elite discourses, it is indeed less common to look at citizens’ discourse. As such, public citizens’ discourse hardly exists. Nonetheless, citizens often talk about politics. There are two possible ways to follow these talks: one is to use “recorded” authentic conversations of everyday life, for instance family discussions or online forums, another one is to organize focus groups. The relevance of such focus groups has been highlighted by De Cillia et al. (1999, 152-153):

the method of the ‘focus-group discussion’ (…) offers a very promising tool for ethnographic research in Critical Discourse Analysis. It enables one partially to study the recontextualization and transformation of specific political concepts and identity narratives which are expressed by politicians, taught in educational systems (e.g. by teachers and in schoolbooks), promoted in the mass-media, etc., and which are expressed in everyday situations and interactions.

6Though this method offers promising avenues for metaphor analysis, it has only rarely been implemented (see for example Cameron 2007, Cameron and Maslen 2010 for a few exceptions).

7To study metaphors in citizens' discourses, our research relies on data from eight focus groups composed of six to nine people and held after the 2007 federal elections in the midst of a political crisis: four in Liège (French-speaking Belgium) and four in Antwerp (Dutch-speaking Belgium). For over four hours, the participants, who came from various backgrounds and who held different political beliefs, discussed the future of Belgian federalism with fellow citizens in focus groups as well as with politicians and experts (Reuchamps 2011, Reuchamps and Perrez 2012). The discussions with two politicians (one Dutch-speaking and one French-speaking, from two different political parties) and two experts (university professor and journalist) were meant to offer the citizens different points of views on the issue at stake. First, the participants gathered for a first round of group discussions. Then, in a plenary session, they met the experts who gave a short presentation on Belgian federalism (history and political dynamics) and above all answered questions from the participants. Citizens gathered afterwards for a second round of discussions, before meeting the two politicians who gave a short speech (twenty minutes each) followed by a question and answer session. A third round of discussions between the participants in small groups brought the event to a close. The group discussions were recorded and transcribed in extenso. This resulted in two corpora of citizens’ discourse respectively in French (FR-corpus, 52,003 words) and Dutch (NL-corpus, 47,579 words).

8A key question is, of course, the potential influence of the politicians and experts on the participants. Previous research (Reuchamps and Charlier 2011) showed that the participants’ preferences (measured via a pre- and a post-questionnaire), for some of them, evolved over the course of the event. But the nature of these changes varies. Some of the participants experienced a learning process: they did not know enough about Belgian politics to have their own opinion – their participation in these discussions helped them to form their own opinion. Others already had an opinion, but over the course of the discussions, it was revealed that this opinion was not in line with the current federal dynamics, which led them to change it. In these changes, the influence of the politicians and the experts is not direct, but rather their interventions allowed the participants to consolidate their own opinions.

9Beside the question of political impact, there is also the issue of linguistic impact; that is, to what extent the words (and more specifically the metaphors) used by the participants were inspired by the words used by the experts and politicians. First of all, experts and politicians had not been asked to use or not to use metaphors. No specific linguistic guidelines were given. A comparison between the interventions of the experts and politicians, on the one hand, and of the participants, on the other hand, does not reveal a strong circulation of metaphors. While some expressions such as “coquille vide” (empty shell) were repeated by the participants to refer to the federal state, no specific patterns of linguistic impacts could be identified. As we will see below, there was a stronger circulation of metaphors between participants than between them and the experts and politicians. The reason lies probably in the fact that there were many more interactions between participants.

2.2 Metaphor identification

10In order to assess to what extent the citizens used metaphors to talk about Belgian federalism, we applied a revised version of the Metaphor Identification Procedure (MIP) developed by the Pragglejazz Group (2007) according to the following four steps. We firstly read the entire corpora to come to a global understanding of their respective contents. From these two corpora we secondly automatically extracted the relevant data by performing a concordance search (i) for the lemmas directly referring to the target domain of Belgian federalism (see Table 1 for an overview of the target words included in this concordance search) and (ii) for lexical signals of cross-domain mappings, i.e. the words which might explicitly point to metaphorically used expressions (such as comme and (zo)als ‘like’, comparer and vergelijken ‘to compare’, symboliser and symboliseren ‘to symbolize’, etc.). This concordance search resulted respectively in 492 relevant occurrences in the FR-corpus and 495 relevant occurrences in the NL-corpus.

Table 1: lemmas referring to the target domain of Belgian federalism used for the automatic corpus extraction

Corpus Lemmas referring to the target domain of Belgian federalism
Corpus-FR Belg* ‘Belg*’, fédér* ‘feder*’, nation ‘nation’, pays ‘country’, Etat ‘state’, entité fédérée ‘federated entity’
Noun + belge ‘Belgian’, + communautaire ‘community’, + fédéral ‘federal’, + fédéré ‘federated’, + institutionnel ‘institutional’, + national ‘national’
Corpus-NL Belg* ‘Belgium’, feder* ‘feder*’, natie ‘nation’, land ‘country’, staat ‘state’, deelstaat ‘federated entity’
Noun +Belgisch ‘belgian’, +communautair ‘community’, + federaal ‘federal’, +institutioneel ‘institutional’, +nationaal ‘national’
  • 1 This difference is statistically significant (χ2 = 4.953, df = 1, p 0.05). However, since this ar (...)

11The relevant contexts were subsequently qualitatively analysed to find metaphorically used expressions. The analysed contexts consisted of 150 characters to the left and to the right of the target word. At this stage, it should be emphasized, that, since we were primarily interested in the way citizens talked about Belgian federalism, we did not assess the metaphorical potential of each lexical unit in these contexts, but only of the lexical units that were used to refer to Belgian federalism (unlike a typical MIP procedure). This implies, for instance, that we did not take the various metaphorical uses of the word ‘Belgium’ into account, unless it was used in relation to Belgian federalism. In order to find metaphor-related words, we relied on the MIP-methodology by comparing the meaning of the word in context with its more basic meaning (Pragglejazz Group 2007). As suggested by Steen et al. (2010), we did not take the historical dimension into account to establish the basic meanings of the analysed words. For this stage of the analysis, we respectively used Le Petit Robert de la langue française 2013 (online edition) as the reference dictionary for the FR-corpus, and the Van Dale: Groot Woordenboek der Nederlandse Taal as the reference dictionary for the NL-corpus. This second stage of analysis led to the identification of 99 metaphorical contexts in the FR-corpus (20.1%) and 73 in the NL-corpus (14.8%)1.

Table 2: metaphorical and non-metaphorical contexts in the FR-corpus and NL-corpus

FR-corpus NL-corpus
N % N %
Metaphorical contexts 99 20,1% 73 14,8%
Non-metaphorical contexts 393 79,9% 422 85,2%
Total 492 100% 495 100%

12This difficulty of deciding on the metaphorical character of some lexical units can be illustrated by examples (1) and (2) (see also Perrez and Reuchamps 2014). In example (1), the construction of Belgian federalism is referred to as an “usine à gaz” (lit. “gas storage plant”) to denote its complicated structure and lack of efficiency. In example (2), according to another citizen, this inefficiency is partly explained by “wafelijzerpolitiek” (lit. “waffle-iron policy”), referring to a mechanism implying that expenses which are initially predestined to one of the communities are compensated by symmetrical expenses in favour of the other community.

  • 2 The examples taken from our data are marked with an ID-number, composed of three parts. The first (...)

1. Fr. (…) On va continuer à monter une usine à gaz (PBF, B8, 249-250)2
En. ‘(…) We’re setting a kludge (lit. gaz storage plant)’

2. Du. Ik denk dat als men het woord wafelijzerpolitiek ook zei als wij al een projectportefeuille op ons werk hebben waar een Europese subsidie moet verdeeld worden. (PBF, K6, 910-916).
En. ‘I think that we also use the word ‘wafelijzerpolitiek’ (lit. waffle-iron policy) when we already have a portfolio at our office where a European subvention has to be divided.’

13While both expressions appeal to vivid images, respectively an inefficient factory and a waffle iron, which can be related to the conceptual metaphor the state is a machine, both units are clearly lexicalized considering that they appear as monosemic units in the reference dictionaries and that their basic meanings precisely refer to a big, vague and inefficient system on the one hand and to a money transfer mechanism on the other. They should therefore be considered as non-metaphorical expressions.

2.3 Metaphor counting

14A last methodological remark concerns the guidelines we followed to count the metaphorical units. Since we are primarily interested in how citizens talk about Belgian federalism and not so much in the overall percentage of metaphorical units in our corpora, we adopted a different counting method to the one advocated by the MIP procedure, by focusing on metaphorical contexts rather than on lexical units. The following example illustrates this counting method:

3. Fr. c’est comme dans un ménage, on ne règle jamais les solutions une fois pour toutes. On se marie, ou en vit ensemble, peut importe, à 20 ans, puis on a des enfants, puis les enfants deviennent grands, puis le bonhomme fait sa crise de la quarantaine, puis on se dit que tout compte fait, on se dit que c’était quand même pas si mal et puis rien, et puis entre-temps, madame est ménopausée et puis... Puis elle a perdu son job, puis les enfants se sont mariés, voilà que la maison est trop grande... les situations évoluent et je ne pense pas qu’on va rêver d’avoir une situation immuable. J’arrête les figures et les fables. (PBF, B8, 1968-1977).

En. ‘it’s like in a couple, you can’t solve all problems once and for all. You get married, or you live together, whatever, at twenty, then you have kids, then the kids grow old, the husband goes through his midlife crisis, but then you realize it wasn’t that bad after all, and then nothing, and then in the meantime, his wife goes through the menopause and then…(…) and then she loses her job, then the kids get married, and then the house is too big… Situations evolve and I don’t think we dream of having a stable situation. I’m stopping with images and fables.’

15This passage is really interesting since this participant deliberately (as he himself admits it) produces a lot of family metaphors to talk about Belgian federalism. From an MIP perspective, from the second sentence on, every single word could be considered as an independent metaphorical unit (except perhaps for the last two sentences). However, when focusing on what is said about Belgian federalism, the speaker is ‘only’ suggesting that the problems will not get solved once and for all and that Belgium, like any other couple, will have ups and downs. So, basically, what is interesting for our analysis is that this citizen is comparing Belgium to a couple. That is why we counted this whole passage as one metaphor, though it would probably be more accurate to talk about metaphorical contexts than about metaphors. Similar examples were treated accordingly. We think this way of counting metaphorical contexts allows us to avoid any bias in favour of particular conceptual domains (in this example the conceptual domain of the family). The analyses of the two corpora were performed independently by both authors. Problematic cases were further discussed to come to an agreement. The results of this quantitative analysis are presented in the following section.

3. Results

3.1 General harvest

16Table 3 summarizes the overall results of our study. Accurate quantitative comparisons with previous studies (such as Krenmayr 2011) are difficult considering our diverging metaphor counting method and the fact that we did not consider the metaphorical nature of every single lexical unit from our corpus. However, the proportion of metaphors in the citizens' discourse appears to be fairly high, suggesting that citizens, when prompted to talk about Belgian federalism, tend to resort to metaphors to express their perception and comprehension of this abstract political concept.

Table 3: distribution of metaphors across the conceptual domains in the FR-corpus and NL-corpus

Source domain Number of metaphorical contexts Total
FR-corpus NL-corpus
construction 15 33 48
personification 16 11 27
machine 17 5 22
journey 16 6 22
family 12 4 16
disease 5 2 7
company 1 6 7
entity 1 4 5
instrument 4 1 5
container 4 1 5
food 4 0 4
theatre 2 0 2
laboratory 1 0 1
clothes 1 0 1
Total 99 73 172

17In the next sections we will focus on the use of metaphors by the citizens, by more particularly considering the relevant source domains on which they are respectively built. As advocated by Cameron (2007, 205):

researchers adopting a discourse approach to metaphor have to accept that it is not possible to come up with a limited set of categories into which each linguistic metaphor can be reliably placed. (…) A principled flexibility to the grouping of linguistic metaphors appears to be the most suitable approach with discourse data.

18Accordingly, we relied on a bottom-up approach to identify the relevant domains, placing the emerging metaphors in larger conceptual categories. Because we are interested in how metaphors are used by citizens in general to describe Belgian federalism, we will present the results of our quantitative analysis globally without making any further distinction between both citizen groups. We will, however, discuss the interlinguistic differences when these appear to be relevant.

3.2 Conceptual domains

19The most relevant conceptual domains emerging from the citizens' discourse are summarized in Table 3 above. These results show that when speaking metaphorically of Belgian federalism, the citizens frequently use construction metaphors, personifications, machine metaphors, journey metaphors, family metaphors, and to a lesser extent metaphors that relate to the disease and the company domains. The construction, machine and journey domains appear to be the most prevailing source domains in terms of which Belgian federalism is thought of, along with the cases of personifications.

20The construction domain is the most prominent domain in terms of which the citizens perceive Belgian federalism. As has been frequently pointed out in the literature (see among others Elvert 2006, Musolff 2004 or Schaffner 1996), this domain is particularly relevant to metaphorically account for political institutions and can easily be related to the primary conceptual metaphor organization is physical structure (Grady 1997). It is, for instance, very conventional to talk about the construction of a political structure or to perceive a state in terms of a building divided into various floors. In our data, the high frequency of metaphors referring to this domain can be explained by the frequent occurrence of conventional metaphorical extensions of the lexical units structure, structuur (‘structure’; see examples 4 and 5), niveau (‘level’; see example 6) and to a lesser extent construire (‘to construct’; see example 7) to describe one’s perception of Belgian federalism. Globally, this domain is more frequently used by the Dutch-speaking citizens than by the French-speaking citizens (see Table 3) to refer to Belgian federalism. One possible explanation for this difference is the more frequent use in Dutch of the word structure (28 times) than in French (only six times). That is why Dutch-speaking citizens are more likely to talk about federal structure, and French-speaking citizens are more likely to talk about federal system, even though they probably refer to the same entity.

4. Fr. un état fédéral, une structure fédérale, ça peut être très différent, ça peut être une structure fédérale forte ou une structure fédérale résiduaire. (PBF, D8, 2336-2337)
En. ‘a federal state, a federal structure, it can be very different, it can be a strong federal structure or a residual federal structure.’

5. Du. Die solidariteit moet je buiten de Belgische structuur zien. (PBN, L 6, 1522-1523)
En. ‘You have to see this solidarity outside of the Belgian structure’.

6. Du. […] over het feit dat het Belgisch niveau overbodig is. (PBN, L 1, 1414)
En. ‘…about the idea that the Belgian level is superfluous’.

7. Fr. […] le fédéralisme tel qu’il a été construit progressivement (PBF, D 8, 3897-3898)
En. ‘…federalism, as it has progressively been constructed…’

21Perceiving a country as a person (the state is a person) has often been considered as one of the most conventional political conceptual metaphors (see for instance Lakoff 1996, 2004). It therefore comes as no surprise to see it emerge from our citizens' discourse. Among other things, comparing the state to a person makes it possible to understand its functioning in more concrete terms (see for instance 8).

8. Fr. et je pense que c’est ce qui est arrivé parce que quand la Flandre galérait en 1930, il était normal que la riche industrie sidérurgique wallonne alimente le pays (PBF, B8, 556-557).

22Another recurring domain is the machine domain. The high proportion of metaphors based on this conceptual domain can be explained by the frequent use of the conventional metaphorical extensions of the verbs fonctionner and werken (‘to function’, ‘to work’) to denote an organization fulfilling its function, this use accounting for more than 50% of the cases (see for instance examples 9 and 10). On the whole, when the citizens compare Belgian federalism to a machine (Belgian federalism is a machine), they tend to emphasize that the political system is too complex and not working properly (see examples 10 and 11). This image is even more present in the corpus of French-speaking citizens than in the corpus of Dutch-speaking citizens. This difference is almost the exact opposite as the one for the construction domain. This is indeed the same reason: to refer to the same object – federal system/structure – citizens in French more often use the machine domain while in Dutch it is the construction domain.

9. Fr. il faut distinguer le fédéralisme belge tel que nous le connaissons à l'heure actuelle, du fédéralisme tel qu'il a fonctionné dans les temps passé (PBF, D5, 2469-2471)
En. ‘one has to make a distinction between Belgian federalism as we know it today, and federalism as it has worked in the past’

10. Du. Maar ik denk dat je kunt concluderen dat het federalisme zoals het nu is dat het niet werkt (PBN, N4, 3318-3319)
En. ‘But I think that you can conclude that federalism as it is now, is not working.’

11. Fr. On a coupé le citoyen du fonctionnement d’une espèce de mécanisme, de machine folle lancée sur elle-même. (PBF, B8, 839-840)
En. ‘The citizen has been cut from the functioning of a sort of mechanism, of a crazy machine spinning around

23The metaphors structured around the notion of journey also constitute a large category, including various conventional metaphors either presenting Belgian federalism as an entity on a path (see 12), as a stop-over on a path (see 13) or as a destination (see 14 and 15). These metaphors can be understood in terms of the more encompassing conceptual metaphor politics is a journey, which has often been shown to play a significant role in the way we talk about politics (see for instance Arcimaviciene 2007, Charteris-Black 2011, Cibulskiene 2012). On the whole, these metaphors appear to be more frequent in the corpus of French-speaking citizens than in the corpus of Dutch-speaking citizens. This difference is politically interesting as it can be explained by the view held by several French-speaking citizens that federalism is only one step in Belgium’s evolution towards its separation, which they dread (Reuchamps 2009, 2013).

12. FR. Je crois qu’à partir du moment où le fédéralisme est évolutif, il ira de crises en crises. (PBF, D 8, 2561)

13. Fr. Il dit que la Belgique, c’est une étape qui est devenue non nécessaire (PBF, M, 780-781)
En. ‘He states that Belgium, it is a stop-over which has become unnecessary’

14. Du. doordat men maar in zeer kleine stapjes tot dat Belgisch federalisme is gekomen (PBN, K1, 230-231)
En. ‘because we came to that Belgian federalism in very small steps.’

15. Du. die discussie over de toekomst van het Belgisch federalisme, waar we naartoe moeten (PBN, M, 1219-1220)
En. ‘This discussion over the future of Belgian federalism, that is where we have to go to.’

24The fifth domain is the family domain. While not predominant in terms of occurrences, it appears qualitatively to be an important conceptual domain. Felstiner (1983) already mentioned the use of family metaphors in patriotic discourses but the relevance of family relationships for the way citizens commonly make sense of complex political processes has extensively been exposed by Lakoff (1996), in his account of how different parental models (strict father vs. nurturant parent) can shape our perception of given political issues. The conceptual influence of family relationships has been confirmed by several scholars, among whom Hayden (2003), who discusses the notion of "maternal politics" (as opposed to "strict father morality"), and Adams et al. (2008) who highlight the importance of family metaphors in life-narrative interviews.

25It is even more relevant when we look at our corpora. In fact, while in the other conceptual domains the metaphors do not reveal conflicting views on Belgian federalism, family-related metaphors by contrast show a high level of opposition (see example 3 above and examples 16 to 18 below). The metaphors that are used by the citizens to describe their opinion about federalism in Belgium refer to the conceptual metaphor belgian federalism is a love relationship. This metaphor stems from the more basic metaphor the state is a person (cf. discussion above). In the case of a federal country including two main communities, it seems to be quite natural to see the state as two persons interacting with one another. The metaphor belgian federalism is a love relationship does not, strictly speaking, appear as a family metaphor in the sense of the parental models discussed above. However, considering that the relationships between the main Belgian communities are framed in terms of couple, marriage and divorce in the citizen data and that these love relationships between two adults are central to family dynamics, this metaphor can be considered as a natural extension of the family domain.

26Quite interestingly, the political opinion behind the metaphor belgian federalism is a love relationship appears to vary to a large extent. Belgian federalism can be seen as a love relationship and the citizens holding this vision call for the unity of Belgium. Others see it in terms of a marriage of convenience and argue for a Belgian federalism that acknowledges the differences between Dutch-speakers and French-speakers, but also seeks to build a common future. A third group believes in a living apart together (LAT) relationship (see example 18) which means for them that two main language groups should remain within the same country but lead their own lives. A last group, on the contrary, sees Belgium as a forced marriage and therefore calls for its break up (see example in Table 4). From a political science point of view, it is an interesting finding: metaphors are thus definitely more than mere rhetoric figures as they bring about different political views.

16. Fr. Si l’on compare avec un ménage, certains ménages se marient avec contrat de mariage, d’autres pas (PBF, B1, 188-190)
En. ‘If we compare this to a couple, some couples get married under a wedding contract, others don’t…’

17. Fr. parce que je ne sais pas comment cohabitent la région wallonne et la région flamande avec le fédéralisme etc. (PBF, B6, 145-148)
En. ‘because I don’t know how the Flemish region and the Walloon region are living together under federalism’

18. Du. Normale partijen die een staatshervorming willen enzovoort die willen eigenlijke hetzelfde als we zo zeggen een ernstige LAT relatie in dit land. (PBN, M5, 3130-3131)
En. ‘Normal parties which want a state reform, they want in fact the same as let’s say a serious LAT relationship in this country.’

27This notion of love relationship between the different parts of the country seems to play a particularly significant role in the way citizens frame their understanding of the relations between the two main communities of the country. More than a simple stylistic device, this metaphor of the love relationship makes it possible to reflect different visions on Belgian federalism and, by so doing, offers a particularly salient conceptual reference point for the citizens to express their own perception of it. This is clearly illustrated by the following fragment (presented in Table 4) in which three citizens express their diverging views on Belgian federalism in terms of a marriage metaphor.

Table 4: discussion structured around the marriage metaphor to depict Belgian federalism (NL-corpus)

Dutch-speaking citizens English translation
L2: “het is vergelijken met dat huwelijk he. De Belgische staat is een gearrangeerd en geforceerd huwelijk geweest.” (2263-2266) L2: ‘Comparing to marriage, right? The Belgian state has been an arranged and forced marriage.’
(…) (…)
L6: “het is inderdaad een gearrangeerd huwelijk en het is gearrangeerd door de internationale gemeenschap” (2268-2269) (…) L6 : ‘It is indeed an arranged marriage and it has been arranged by the international community.’ (…)
L6 : “een gearrangeerd huwelijk kan ook ontbonden worden, zo moeilijk is dat allemaal niet. Het moet gewoon erkend worden door de internationale gemeenschap.” (2279-2280) L6: ‘an arranged marriage can also be dissolved, it’s not that difficult. It only has to be accepted by the international community’
L2 : “ja maar dat is getrouwd voor goede en kwade dagen en wij zijn nu in kwade dagen.” (2281-2282) L2: ‘yes, but it married for better or for worse and we are now in a bad patch’
L6 : “maar bij een gearrangeerd huwelijk is het niet in goede en kwade dagen vrijwillig, maar is het verplicht in kwade dagen. (…) ik hoop toch dat we zover zijn dat huwelijken niet meer verplicht zijn ofwel?” (2283-2287) L6: ‘but in an arranged marriage, it’s not voluntarily for better or for worse, but it’s forced in a bad patch. I hope we have come to a situation where marriages are no longer forced, haven’t we?
L1 : “Neen, maar je kan dan toch karakter tonen, karakter tonen.” (2288) L1: ‘No, but you can still show character’
L6 : “Als ons dat ieder jaar 10 miljard euro kost, vind ik dat toch... ” (2289) L6: ‘If it costs us 10 billion euro a year, I find that…’

28Apart from the marriage metaphor itself, this passage is a perfect example of how the use of a given metaphor by a citizen can frame the reactions of other citizens. This specific function of metaphors in communication has been described by Steen (2008: 230) as the “creation of a common ground of reference”, which he claims particularly occurs “when difficult or complex topics are to be dealt with between interlocutors.”

29Another frequent conceptual domain for deliberate metaphors is the disease domain (Belgian federalism is a disease). When presenting Belgian federalism as a disease, the citizens tend to emphasize that Belgium is an illness (see 19 and 20), an infectious excrescence (21), or a disease typical of rich people (22).

19. Du. …en dat dat is de ziekte van het federalisme. Ik heb dat niet voor niks daarstraks een noodzakelijk kwaad genoemd. (PBN, M1, 3069-3070)
En. ‘and that’s the disease of federalism. It’s not for nothing if I just named it a necessary evil’.

20. Du. Welke bevoegdheden op het nationale niveau, welke bevoegdheden op het regionale niveau? Het wordt helemaal een pest bij wijze van spreken (PBN, M, 2533-2534)
En. ‘which responsibilities at the national level? Which responsibilities at the regional level? It is becoming a plague so to speak’

21. Fr. On a été créer un espèce de furoncle qui s’appelait Belgique… (PBF, B8, 1097-1098)
En. ‘they created some kind of boil called Belgium…’

22. Fr. le fédéralisme est au départ le résultat d’un égocentrisme et d’une maladie de riches. (PBF, B8, 134-136)
En. ‘federalism is the result of egocentrism  and a rich people's disease

30A last conceptual domain used to produce metaphors about Belgian federalism is the company domain (Belgian federalism is a company). Metaphors based on this source domain turn out to be more frequently used by the Dutch-speaking citizens than by the French-speaking citizens. With this metaphor, the citizens tend to express their comprehension of how a state is working in general, and more specifically how politicians are running the country. These examples (23 and 24) also illustrate the growing importance of the economic paradigm in our understanding of political processes, as suggested by Koller (2009).

23. Du. Hoe werkt een staat? Een beetje zoals een bedrijfsleider over zijn bedrijf. (PBN, K4, 1232-1233)
En. ‘How does a state work? A bit like a CEO with his company…’

24. Du. als je vergelijkt met een bedrijf, een bedrijf laat je ook niet leiden door de werkman of de kuisvrouw bij wijze van spreken (PBN, L3, 2327-2329)
En. ‘if you compare with a company, a company is not being run by a workman or a cleaning lady, so to speak.’

4. Discussion

31Our study, based on the analysis of focus group discussions on Belgian federalism, aimed to analyse citizens' discourse in order to assess to what extent citizens used metaphors to describe complex political relations and processes. To analyse metaphorical expressions in citizens' discourses, we applied a method based on a slightly diverging version of the MIP procedure. From the application of this methodology to our data, the following tendencies can be derived.

32First, our study shows that citizens produce a lot of metaphors when prompted to talk about Belgian federalism in the context of focus group discussions. This suggests that the role of metaphors is not limited to political discourse, i.e. elites’ discourses, but that the citizens actually think of political processes in metaphorical terms as well. This first observation confirms Bougher’s (2012) hypothesis that we can gain valuable political insights from the analysis of citizens' discourse. In this regard, focus group discussions seem to be a promising avenue to collect relevant citizens' discourse, especially for the study of metaphor.

33Our results also suggest that the citizens use various conceptual domains to make sense of Belgian federalism. Among this variety of conceptual domains, we notice that conceptual domains that have traditionally been shown to underlie political discourse, such as the construction domain, the family domain, the machine domain, the journey domain or personifications, also emerge from citizens' discourse, confirming their importance not only for the way we talk about politics but also for the way we think about politics.

34Focusing on citizens’ views on Belgian federalism, one notices that these domains are frequently mobilized to reflect negative images of federalism, often perceived negatively through these different metaphors, for instance as a complex structure, a machine not working properly, a forced marriage or a disease. However, these various domains are not simply different ways of saying the same thing. Choosing a particular source domain to depict Belgian federalism does have conceptual consequences. For instance, while speaking of Belgium in terms of a complicated structure or a deficient machine emphasizes the way the different layers of Belgian federalism have been put together and how these different political levels relate to one another (Caluwaerts, Reuchamps, 2015), comparing Belgium to a love relationship alludes to the links existing between the members of the two main language groups (see, to go further, a study of the impact of the metaphor of Tetris on the representation of Belgian federalism, Perrez, Reuchamps, 2015). For the study of discourses on Belgian federalism, a hot issue in this country, this is an important finding as putting emphasis on the structure calls for changes in terms of distribution of powers (between the different levels of government), while stressing the love relationship calls rather for changes in terms of inter-community relations (that is between Dutch-speakers and French-speakers).

35Further, one could claim that the family domain appears to be particularly relevant to the citizens’ understanding and framing of the Belgian political context. This suggestion comes from the observation that most of the family metaphors tend to frame whole discussion chunks (see for instance example 3 or Table 4). In this context, based on two main language groups often seen as constantly opposing one another, these family relationships materialize in metaphors related to the notion of love relationships, be it a LAT relationship, a couple living together or a marriage out of love. It is interesting that the diversity of types of love relationships makes it possible for the citizens to express a variety of perceptions of Belgian federalism. And they indeed do so.

36Another interesting insight emerging from the analysis of the conceptual domains mobilized by the citizens to produce metaphors is the similarity between the citizen groups under study. Both the French-speaking and the Dutch-speaking citizens tend to resort to the same conceptual domains to make sense of Belgian federalism. This does not necessarily mean that they have the same vision regarding its functions and future developments, but that the conceptual domains in terms of which they make sense of it show a high degree of overlap. Comparing the frequency of these conceptual domains nonetheless reveals some differences. For instance, the Dutch-speaking citizens tend to more frequently refer to the Belgian state in terms of a federal structure whereas the French-speaking citizens tend to refer to it in terms of a federal system. This difference implies potentially different representations of the state’s nature. Further work is needed to fully describe the differences in the way both communities perceive the state and how these differences can be accounted for.

37When considering our data, and more specifically the different metaphors we identified, one can be struck by the sometimes strongly varying communicative nature of the metaphors that emerged from the citizen data. While some metaphors seem to have been produced explicitly to convince one’s interlocutor or to explicit one’s point-of-view (see for instance example 3 above or table 4), other metaphors appear to be conventional ways of talking about politics. This distinction can be illustrated by the following examples (repeated here for the sake of convenience).

25. Fr. …le fédéralisme tel qu'il a été construit progressivement… (PBF, D8, 3897-3898)
En. ‘…federalism, as it has been constructed progressively…’

26. Fr. parce que pour moi la Belgique reste une espèce de grande famille, malgré tout. (PBF, D1, 2289-2290).
En. ‘because to me, Belgium remains a kind of large family, after all.’

38In both examples, Belgium is metaphorically accounted for, being respectively presented in terms of a physical structure in (25) and of a family in (26). While, on the conceptual level, one could claim that both metaphors are instances of frequent conceptual metaphors in political discourse (respectively organization is physical structure and the state is a family), they appear to have different communicative purposes. With the family metaphor in (26), this citizen is explicitly presenting his own conceptualization of Belgium. This is not the case in (25), where the construction metaphor is only indirectly expressed, through the use of a conventional metaphorical extension of the verb construire (‘to build’).

39The need of making this distinction between new metaphors and conventional metaphors has been stressed by Steen (2008), who proposed a three-dimensional model of metaphor analysis in discourse and communication. This discourse analytical framework relies on the distinction between different layers of metaphors at the linguistic (direct vs. indirect metaphors), conceptual (new vs. conventional metaphors) and communicative levels (deliberate vs. non-deliberate metaphors). In terms of Steen’s (2008) model of metaphor analysis, the family metaphor in example (26) would be considered as a direct, novel and deliberate metaphor, whereas the construction metaphor would be regarded as an instance of indirect, conventional and non-deliberate metaphor. While the relevance of this distinction between deliberate and non-deliberate metaphor has been questioned in the literature (see for instance Gibbs 2011), from the perspective of political discourse analysis, one could wonder to what extent making this distinction could point to diverging degrees of saliency in the citizens’ perceptions of the state’s interactions. In this case, for instance, one could claim that the family metaphor appears to be more entrenched in the citizens' political consciousness than the construction metaphor. Furthermore, making this distinction could help us understand (i) why political actors use metaphors – and which metaphors – in their discourse and (ii) how they actually perceive and conceptualize complex notions such as state structure and interactions. These questions should be addressed in future research.

5. Further work

40This paper on the use of metaphors by Belgian citizens constitutes a first step towards a more encompassing approach to the use of political metaphors by citizens and more especially to their political impact. In the literature, this political impact has often been taken for granted in metaphor analysis in political discourse. However, a more global understanding of what this political impact could consist of is still lacking from the current research agenda. As Koller (2009, 121) puts it:

metaphor helps construct particular aspects of reality and reproduce (or subvert) dominant schemas. The models of reality held and reinforced by groups with most power and discourse access then become quantitatively and qualitatively salient. In a cyclical fashion, discourse recipients are therefore more likely to encounter such hegemonic mental models underlying discourse, and given similar conditions of reception, such repeated exposure is likely to anchor and reinforce such models even further.

41To be able to account for how metaphors, through discourses, actively shape the political reality – or not -, it is important to look at the relationships between metaphorical discourses and their environment. Accordingly, two specific research avenues seem to particularly be at stake for the further developments of research on political metaphor.

42The first one focuses on the systematic study of the circularity of metaphors across various genres of political discourse (to put it simply from elite discourse to citizen discourse through media discourse) in order to assess to what extent a given metaphor circulates from one genre to another and which directions these metaphors follow. This research question raises several other issues, such as how do political metaphors reach the citizens? Do political metaphors always circulate in a top-down manner (from the elite or media discourse to the citizens) or are there alternative ways for the diffusion of metaphors? What role do the media play in this diffusion process? Do metaphors circulate within political groups (for instance among citizens)?

43A second avenue for future research is to look at how metaphors impact upon citizens. As Bougher (2012, 157) posits, metaphors offer “a cognitive mechanism that explains how citizens make sense of the political world by drawing from their non-political knowledge and experiences.” Metaphors therefore do not only reflect the perceived reality, but they also function as cues through which citizens come to understand political positions, and through which they shape their political behaviours. Further analysing how metaphors impact upon citizens can efficiently contribute to an overall understanding of what role and functions metaphors play in political discourse, and more globally in our everyday political interactions.

Haut de page


Arcimaviciene, Liudmila. 2007. "British politics today: Moral grounds as reflected in conceptual metaphor." Kalbotyra 57: 11-20.q

Bougher, Lori D. 2012. "The case for metaphor in political reasoning and cognition." Political Psychology 33 (1): 145-163.

Cameron, Lynne. 2003. Metaphor in Educational Discourse. London and New-York: Continuum.

Cameron, Lynne. 2007. "Patterns of metaphor use in reconciliation talk." Discourse & Society 18 (2): 197-222.

Caluwaerts Didier, Reuchamps Min, 2015, "Combining federalism with consociationalism. Is Belgian consociational federalism digging its own grave?", Ethnopolitics, no 14 (3), p. 277-295.

Cameron, Lynne, and Robert Maslen. 2010. "Using metaphor analysis to compare expert and public perceptions of the risk of terrorism." In Metaphor Analysis: Research Practice in Applied Linguistics, Social Sciences and the Humanities, ed. by Lynne Cameron, and Robert Maslen, 248-256. London: Equinox.

Charteris-Black, Jonathan. 2011. Politicians and Rhetoric: The Persuasive Power of Metaphor. London: Palgrave Macmillan.

Cibulskienė, Jurga. 2012. "The development of the journey metaphor in political discourse: Time-specific changes." Metaphor and the Social World 2 (2): 131-153.

De Cillia, Rudolf, Martin Reisigl, and Ruth Wodak. 1999. "The Discursive Construction of National Identities." Discourse & Society 10 (2): 149-173.

De Landtsheer, Christ'l. 2009. "Collecting political meaning from the count of metaphor." In Metaphor and Discourse, ed. by Andreas Musolff, and Jorg Zinken, 59-78. Houndmills: Palgrave Macmillan.

Deschouwer, Kris. 2012. The Politics of Belgium: Governing a Divided Society. London: Palgrave Macmillan.

Deschouwer, Kris, and Min Reuchamps. 2013. "The Belgian Federation at a Crossroad." Regional & Federal Studies 23 (3): 261-270.

Dikke Van Dale Online (2013). Utrecht: Van Dale Woordenboeken. URL:

Elvert, Jürgen. 2006. "How many rooms are there in the European house?" In L'Europe inachevée, ed. by Michel Dumoulin, and Geneviève Duchenne, 127-142. Brussels: Peter Lang.

Felstiner, Mary L. 1983. "Family Metaphors: The Language of an Independence Revolution." Comparative Studies in Society and History 25 (1): 154-180.

Gibbs, Raymond. 2011. "Are 'deliberate' metaphors really deliberate? A question of human consciousness and action." Metaphor and the Social World 1 (1): 26-52.

Grady, Joseph. 1997. Foundations of Meaning: Primary Metaphors and Primary Scenes. Ph.D. dissertation, University of California, Berkeley.

Hayden, Sara. 2003. "Family Metaphors and the Nation: Promoting a Politics of Care through the Million Mom March." Quarterly Journal of Speech 89(3): 196-215.

Koller, Veronika. 2009. "Missions and empires: Religious and political metaphors in corporate discourse." In Metaphor and Discourse, ed. by Andreas Musolff, and Jorg Zinken, 116-134. Houndmills: Palgrave Macmillan.

Krenmayr, Tina. 2011. Metaphor in Newspapers, PhD, Utrecht: LOT dissertation series, 276.

Lakoff, George. 1996. Moral politics: How Liberals and Conservatives Think. Chicago: University of Chicago Press.

Lakoff, George. 2004. Don't Think of an Elephant: Know Your Values and Frame the Debate. Vermont: Chelsea Green Publishing.

Lakoff, George, and Mark Johnson. 1980. Metaphors We Live By. Chicago: University of Chicago Press.

Le Petit Robert de la Langue Française (2014). Paris: Dictionnaires Le Robert [Online edition]. URL:

L'Hôte, Emilie. 2010. "New Labour and globalization: Globalist discourse with a twist?" Discourse & Society 21 (4): 355-376.

L'Hôte, Emilie. 2012. "Breaking up Britain? Métaphores et discours sur la dévolution au Royaume-Uni." In Les relations communautaires en Belgique : Approches politiques et linguistiques, ed. by Julien Perrez, and Min Reuchamps, 161-189. Louvain-la-Neuve: Academia-L'Harmattan.

McAdams, Dan P., Michelle Albaugh, Emily Farber, Jennifer Daniels, Regina L. Logan, and Brad Olson. 2008. "Family Metaphors and Moral Intuitions: How Conservatives and Liberals Narrate Their Lives". Journal of Personality and Social Psychological 95 (4): 978-990.

Mercenier, Heidi, Julien Perrez, and Min Reuchamps. 2015. "Constructing Quebec and Wallonia: how political parties speak about their region." In Minority Nations in Multinational Federations: A comparative study of Quebec and Wallonia, edited by Min Reuchamps, 49-81. Abingdon & New York: Routledge.

Musolff, Andreas. 2004. Metaphor and Political Discourse: Analogical Reasoning in Debates about Europe. London: Palgrave-Macmillan.

Musolff, Andreas, Christina Schaeffner, and Michael Townson (eds). 1996. Conceiving of Europe: Diversity in Unity. Aldershot: Ashgate.

Perrez, Julien, and Min Reuchamps. 2012. Les relations communautaires en Belgique : Approches politiques et linguistiques, Science politique. Louvain-la-Neuve: Academia-L'harmattan.

Perrez, Julien and Min Reuchamps. 2014. "Deliberate metaphors in political discourse: the case of citizen discourse", 25: 7-41.

— 2015, «"The “Belgian Tetris”. Assessing the political impact of metaphors on citizens’ perception of and attitude towards Belgian federalism", Metaphor and the social world, no 5 (2), p. 223-244.

Pragglejazz Group. 2007. "MIP: A method for identifying metaphorically used words in discourse." Metaphor & Symbol 22 (1): 1-39.

Reuchamps, Min. 2009. "Le fédéralisme et ses deux sens - négatifs - en Belgique francophone." Revue de la Faculté de droit de l’Université de Liège 54 (4): 527-540.

Reuchamps, Min. 2011. L'avenir du fédéralisme en Belgique et au Canada. Quand les citoyens en parlent. Bruxelles: P.I.E.-Peter Lang.

Reuchamps, Min. 2013. "The Future of Belgian Federalism Through the Eyes of the Citizens." Regional & Federal Studies 23 (3): 353-368.

Reuchamps, Min, and Julien Perrez. 2012. "Métaphores conceptuelles dans les discours citoyens en Belgique." In Les relations communautaires en Belgique : Approches politiques et linguistiques, edited by Julien Perrez and Min Reuchamps, 133-158. Louvain-la-Neuve: Academia-L'harmattan, p. 133-158.

Schäffner, Christina. 1996. "Building a European House? Or at Two Speeds into a Dead End? Metaphors in the Debate on the United Europe." In Conceiving of Europe: Diversity in Unity, ed. by Andreas Musolff, Christina Schaeffner, and Michael Townson, 31-59. Aldershot: Dartmouth.

Semino, Elena. 2008. Metaphor in Discourse. Cambridge: Cambridge University Press.

Steen, Gerard J. 2008. "The paradox of metaphor: Why we need a three-dimensional model for metaphor." Metaphor & Symbol 23 (4): 213-241.

Steen, Gerard J., Aletta G. Dorst, J. Berenike Herrmann, Anna Kaal, Tina Krennmayr, and Trijntje Pasma. (2010). A Method for Linguistic Metaphor Identification: From MIP to MIPVU. Amsterdam: John Benjamins.

Haut de page


1 This difference is statistically significant (χ2 = 4.953, df = 1, p < 0.05). However, since this article does not address the possible cultural and linguistic differences in metaphor usage, we do not dare to draw any conclusion from this observation. It should be interesting to address this question more specifically in future research.

2 The examples taken from our data are marked with an ID-number, composed of three parts. The first part points to the corpus, the label ‘PBF’ referring to the FR-corpus, the label ‘PBN’ referring to the NL-corpus. The second part is the ID of the participant and the third part points to the lines the passage is referring to in the respective corpora. In our examples, the relevant metaphorically used expressions are italicized. The lexical units pointing to potential metaphors are underlined.

Haut de page

Pour citer cet article

Référence électronique

Julien Perrez et Min Reuchamps, « A crazy machine or a strong “living apart together” relationship? », Mots. Les langages du politique [En ligne], 109 | 2015, mis en ligne le 30 novembre 2017, consulté le 20 juin 2021. URL : ; DOI :

Haut de page


Julien Perrez

Université de Liège

Min Reuchamps

Université catholique de Louvain

Haut de page

Droits d’auteur

© ENS Éditions

Haut de page
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search