Populism as political communication style
An empirical study of political parties’ discourse in Belgium
Jan Jagers and Stefaan Walgrave
Final version for publication in European Journal of Political Science, September 16th , 2005
Abstract: The scientific debate about populism has been revitalised by the recent rise of extreme-right parties
in Western Europe. Within the broad discussion about populism and its relationship with extreme-right, in
this article, we confine ourselves to three topics, a conceptual, an epistemological and an empirical issue. First,
taking a clear position in the ongoing definition struggle, we define populism primarily as a specific political
communication style. We conceive of populism as a political style essentially displaying proximity of the people,
while at the same time taking an anti-establishment stance and stressing the (ideal) homogeneity of the people
by excluding specific population segments. Second, we point out that defining populism as a style enables us to
turn it into a useful concept that has too often remained vague and blurred. Third, drawing on our
operational definition of populism, a comparative discourse analysis of the political party broadcasts of the
Belgian parties is carried out. Our quantitative analysis leads to a clear conclusion. In terms of the degree and
the kinds of populism embraced by the six political parties under scrutiny, the extreme-right party Vlaams
Blok behaves very differently from the other Belgian parties. Its messages are a copy-book example of
As in many European countries, an extreme-right wing party is thriving in Belgium1. An
offspring of the Flemish nationalist movement, Vlaams Blok was founded in 1978 when a
few hard-line-nationalists left the moderate Flemish nationalist party Volksunie. To a certain
extent, it exemplifies the upsurge of extreme-right parties all over Europe. Vlaams Blok
belongs to the so-called ‘new’ extreme-right party family. Unlike the ‘old’ extreme-right
parties these parties do not refer to the fascist tradition nor do they directly challenge
democracy. New extreme-right parties focus on immigrant and law-and-order themes in a
populist and antipolitical discourse (Ignazi, 1992). Vlaams Blok is the pariah of Belgian
politics and is contained in a so-called cordon sanitaire. Unlike in other European countries
with strong extreme-right parties (e.g. Austria, Italy, Denmark, etc.), all other parties have
solemnly agreed not to cooperate with Vlaams Blok under any circumstances and on any
political level. The party is accused of being racist, fascist, and undemocratic. The official
antiracist organisation sued the party and in November, 2004 it was condemned for racism.
1 Vlaams Blok only exists in the Flemish part of Belgium (60% of the Belgian population). It does not run
for office in the French-speaking part. In order not to complicate things we will always refer to Belgium
and the Belgian parties while actually only referring to the Flemish-speaking part of Belgium.
Consequently, it changed its name from Vlaams Blok to Vlaams Belang and refounded itself
with roughly the same programme and a similar structure2. Hostility towards Vlaams Blok
did not prevent it from increasing its votes’ share at every election since the beginning of the
1980s. Its breakthrough came with the 1991 general elections, later labelled Black Sunday:
the share of Flemish extreme-right votes more than tripled from 3% (1987) to 10% (1991).
In nine consecutive general and/or regional elections, the party managed to convince an ever
growing part of the Flemish population. At the most recent regional/European elections in
June 2004, Vlaams Blok even became the single largest party in the Flemish parliament with
24% of the vote. Its national upsurge is paralleled by local election results. In Antwerp,
Flanders’ major city, Vlaams Blok got 33% of all votes in 2000. As the Flemish party system
is strongly fragmented, it is in fact one of the most fragmented in Europe (Lane & Ersson,
1991; Anckar, 2000), unusual and large coalitions have to be formed to keep Vlaams Blok
out of office.
The spectacular expansion of Vlaams Blok has inspired many political scientists. However,
the supply side of the Vlaams Blok puzzle remains poorly understood. Mainly the voters’ or
demand side of the Vlaams Blok mystery has received much scholarly attention
(Swyngedouw, 1992; Elchardus, 1994; Maddens, 1998; Billiet & Swyngedouw, 1995; Billiet &
De Witte, 1995; Elchardus & Pelleriaux, 1998; Swyngedouw, 2001). Basically, like other
countries where extreme-right parties flourish, discussion has focused on whether Vlaams
Blok voters actually voted in favour of something - a party, a candidate, a programme, an
issue - or only against something - the other parties, the system, the establishment (Billiet &
De Witte, 1995). In other words: is a Vlaams Blok vote an ideological vote or just an
expression of protest/discontent (Van der Brug et al., 2000)? Many previous studies also
pinpointed the failure of the other parties to connect to the Vlaams Blok electorate and to
effectively handle the issues the Vlaams Blok ‘owns’ (immigration, crime…)(Walgrave &
Far too little research has been devoted to the supply side of the Vlaams Blok riddle.
Although there is an electoral breeding ground for extreme-right parties all over Europe,
strong extreme-right parties do not prevail in all European countries. A careful study of the
supply side might help us understand why this is the case. Some of these parties may be
doing a particularly good job, capitalising on the political opportunities at hand, while other
parties screw up and do not manage to seize the available chances. Regarding Vlaams Blok, it
seems the party is doing an excellent job (Coffé, 2005a, 2005b). It is professionally led and
thanks to its election results, has lots of money and an unusually clear political position. The
party has strong and verbally skilled politicians and a loyal constituency skilfully tied to the
2 Almost all Belgian (Flemish) parties changed names recently. However, throughout this paper we will
use the party names that were in use at the end of the period under study (1999-2001). This means that
we will not use the new name of the extreme-right Vlaams Blok – i.e.: Vlaams Belang - or use the new
name of the green Agalev, nowadays called Groen!.
party. In addition, Vlaams Blok, is a party that, until now, has hardly had any (open) internal
But, above all, Vlaams Blok appears to have developed a smart, omnipresent and
pronounced communication strategy, perfectly suited for striking antipolitical chords. In this
article we will focus on this political communication of the Vlaams Blok. The hypothesis we
want to test is that the Vlaams Blok’s external communication is characterised by an outspoken and all-
pervading populism. Testing this hypothesis requires comparative evidence. So as to be able to
categorise Vlaams Blok’s communication as populist, we need to contrast its communication
to that of other parties. We will therefore systematically compare Vlaams Blok’s televised
messages, via its political party broadcasts, with those of the other Belgian parties in the
same period (1999-2001).
In order to carry out such a comparison we need to develop a measurable concept of
populism. Scientific debate about populism has been revitalised by the recent rise of
extreme-right parties in Western-Europe (see among many others: Mény & Surel, 2002; Betz,
2004). Yet more globally, as political parties in general suffer from declined partisanship and
increased electoral turnover (Katz & Mair, 1994), they are searching for other means to
connect with voters and populism might be such a device. We will, in fact, define populism
as a communication style. The aim of this contribution, therefore, is not merely to assess
populism in the Belgian context, but to put forward an empirically measurable concept of
populism. As we will show, populism is a contested concept that has remained vague and
blurred too often. Our ambition is to develop a clear definition and, even more important,
show that this definition can be operationalised for quantitative content analysis. As such the
present study pursues conceptual (how must populism be defined?), epistemological (how
can populism be measured?) as well as empirical goals (is Vlaams Blok more populist than
POPULISM - A POLITICAL COMMUNICATION STYLE
Populism has become popular. Political scientists and political actors frequently refer to
populism to characterise certain political phenomena or brand competitors in a political
conflict. The use of populism causes confusion and debate about the definition of the
concept has not yet been settled (Canovan, 1981; Wievorka, 1993; Taguieff, 1995, 1998; de
Benoist, 2000; Taggart, 2000; Elchardus 2001; Mény & Surel, 2000, 2002; Mudde, 2004;
Abts, 2004). Following Taguieff (1998) and highlighting the most documented cases
throughout history, we can distinguish three successive waves of populism: agrarian
populism, Latin-American populism and new-right populism. Agrarian populism is to be
found in the Russian intellectual Narodniki in the second half of the 19th century, engaging
in an egalitarianist struggle on behalf of Russian peasants (Walicki, 1969). Also the American
People’s party, which pleaded against capitalism and in favour of agrarian socio-economic
interests around the turn of the century, is generally considered as an example of agrarian
populism (Worsley, 1969). The Latin-American variant of populism prospered in the 1940s
and 1950s with the authoritarian regimes of Péron in Argentina and Vargas in Brazil. These
nationalist, charismatic leaders pretended to be a direct emanation of the people and to
govern - no more than a servant – for the people against established interests (Hennessy, 1969).
New-right populism mobilises, from the 70s onwards, against traditional politics which is
reproached as being self-serving while systematically ignoring the real wishes of the people.
New-right populism typically focuses on issues such as immigration, taxes, crime, and
nationalism (Taggart, 2000). Three elements can be considered as common denominators of
these successive historical, and scholarly theoretical, shapes of populism. Populism (1) always
refers to the people and justifies its actions by appealing to and identifying with the people; (2)
it is rooted in anti-elite feelings; and (3) it considers the people as a monolithic group without
internal differences except for some very specific categories who are subject to an exclusion
Together, these three elements define populism. It is possible, though, to distinguish a ‘thin’
and a ‘thick’ concept of populism. We will use the thin definition, only relying on the first
element of merely making reference to the people, as an operational definition. The thick
definition comes close to the classic concept and consists of a combination of the three
elements and states that populism refers to the people, vents anti-establishment ideas and
simultaneously excludes certain population categories.
Referring to the people can hardly be considered as being a (new) ideology, let alone a
political movement. That is why we opt to define populism in the tradition of Canovan’s
‘politicians’ populism’, as a style rather than an ideology (see: Canovan, 1981, 1999;
Blommaert, 2001; Di Tella, 1997; Mudde, 2000, 2004; Elchardus, 2001; Deschouwer, 2001;
Pfahl-Traughber, 1994; Taguieff, 1998). We propose a thin definition of populism
considering it as a political communication style of political actors that refers to the people. These
political actors can be politicians and political parties, but also movement leaders, interest
group representatives and journalists. Populism, therefore, is a communication frame that
appeals to and identifies with the people, and pretends to speak in their name (Taggart, 2000;
Canovan, 1981). It is a master frame, a way to wrap up all kinds of issues. More concretely,
populism is a conspicuous exhibition of closeness to (ordinary) citizens. This self-
presentation can take different guises - using casual or colloquial language or adopting an
informal dress code – but the most important element of a political style is the content of
the discourse. Consequently, we understand thin populism as displaying closeness to the
people simply by talking about the people. By appealing implicitly to the people, a populist
communication style stresses the sovereignty of the people and the popular will. Political
actors speak about the people all the time. They frequently use words such as (the) people,
(the) public, (the) citizen(s), (the) voter(s), (the) taxpayer(s), (the) resident(s), (the)
consumer(s), (the) population… By referring to the people a political actor claims that he or
she cares about the people’s concerns, that he or she primarily wants to defend the interests
of the people, that he or she is not alienated from the public but knows what the people
really want. The implicit populist’s motto is: ‘I listen to you because I talk about you’.
In its thin conceptualisation, populism is totally stripped from all pejorative and authoritarian
connotations. Populism, thinly defined, has no political colour; it is colourless and can be of
the left and of the right. It is a normal political style adopted by all kinds of politicians from
all times. Populism is simply a strategy to mobilise support, it is a standard communication
technique to reach out to the constituency.
The reason for starting with such a thin conceptualisation of populism is twofold. First, we
contend that appealing to the people forms the essential core of populism. Without
reference to the people, populism is simply unthinkable. In all available definitions appealing
to the people is a minimal and necessary condition. While anti-elitism and exclusion can be
found amongst many other political discourses, it is reference to the people that most
fundamentally distinguishes populism from other types of discourse. It is no coincidence
that populism is derived from the Latin ‘populus’. The thin definition also dovetails with the
present use scholars make of the concept, applying populism to describe the behaviour of
mainstream political actors like Tony Blair or Bill Clinton (Kazin, 1995; Mair, 2002). Second,
this empty-shell, initial definition offers important analytical advantages, as we will see when
we tackle the empirical part of the paper. More concretely, our thin concept can be
employed as an operational device that helps us to select parts of a discourse to be further
scrutinised in search of thick populism. In other words: thin populism is a useful preselector
for thick populism. What, then, is thick populism?
As mentioned above, most academic definitions and historical examples of populism contain
two other constitutive elements: anti-establishment and homogeneity/exclusion. In addition
to referring to the people, these two distinction strategies form the concept of thick
populism. When political actors talk about the people and combine this with an explicit anti-
establishment position and with an exclusion of certain population categories, one can speak
of thick populism. Anti-elitism and exclusion, in a sense, fill in the empty shell of thin
populism and give the concept its more classic, restrictive meaning.
Although anti-establishmentness cannot be considered as an exclusive feature of populism –
many radical political movements in general are driven by anti-elitist attitudes and nurture
anti-elitist feelings - most populism scholars consider anti-elitism as a central feature of
populism (Canovan, 1981; Buzzi 1994; Wievorka, 1993; Taggart 2000; Mény & Surel 2000,
2002). Anti-elitism taps the vertical positioning of the political actor at stake. Anti-elitist or
anti-establishment discourse emphasises the distance and estrangement between the people
and the elites. Anti-elite populists side with the people against the elites who live in ivory
towers and only pursue their own interests. The enemy is external to the people, ‘up there’
and high above ordinary citizens. Since populism holds a very broad concept of politics, all
failures and problems are blamed on politics; they are caused by political incompetence,
unwillingness, and sabotage. This all-encompassing vision of politics corresponds to an
equally broad definition of the elites. Elites can be political elites (parties, government,
ministers, etc.), but also the media (media tycoons, journalists, etc.), the state (administration,
civil service), intellectuals (universities, writers, professors), or economic powers
(multinationals, employers, trade unions, capitalists). The more diffuse the anti-elitism, i.e.
directed against general and universal elite categories (e.g. all political parties), the more
fervent and radical it is. Note that, although not neutral or empty, this second constitutive
trait of populism can also be considered as a feature of a political style rather than implying a
The same applies to the third constitutive feature of populism. It relates to the horizontal
dimension. A typical element of populism is that the people are considered as a
homogeneous category (Taggart, 2000; Mudde, 2004). The people largely share the same
interests and have the same features. Yet, some isolated groups clearly do not share the
people’s ‘good’ characteristics. This enemy is not above but internal, within the people. The
true populist not only emphasises the unbridgeable gap between the people and the elites, he
also considers some groups’ values and behaviour to be irreconcilable with the people’s
general interest. Hence, some specific population segments are stigmatised and excluded
from ‘the people’; they are defined as being a threat to and a burden on society. Those
groups are blamed for all misfortune and accidents affecting the general population.
Consequently, these categories are scapegoated and must be fiercely dealt with, if not simply
removed from the territory of the people.
In sum we consider populism to be a communication style adopted by political actors. By
conceptualising populism as, minimally, rhetorically appealing to the people (thin concept)
combined with a vertical and horizontal differentiation (thick concept), we end up with four
types of populism that can be classified in four quadrants along a vertical and a horizontal
axis. The aim of the empirical section of this contribution is to test whether this
conceptualisation works in empirical research. Can we measure populism conceptualised as
such? Does the concept manage to differentiate Belgian parties from each other along the
lines of thin and thick populism? Can we empirically classify Belgian parties in the different
theoretical quadrants? In what follows, we will measure thin and thick populism separately,
relying on different measures. This allows us to test to what extent thin and thick populism
go hand in hand.
DATA AND DESIGN
To empirically assess populism and its dimensions, one needs a sample of political
communication. We opted to content-analyse Belgian parties’ political party broadcasts
(PPB), 10-minute programmes broadcast on channel 1 of the public TV-station VRT. The
advantage of this evidence is that it is foremost direct communication from the party
towards the population, without any intermediation. The PPB’s content is completely
controlled by the parties’ headquarters. The identical format of all broadcasts guarantees
similarity and comparability across parties.
For each of the six major Belgian-Flemish parties we randomly selected 20 PPBs broadcast
in the 1999-2001 period, adding up to 200 minutes per party or a total of 1200 minutes of
PPBs. In order to avoid specific political events affecting our results, we tried to maximise
dispersion of the selected PPBs over the whole three-year period (roughly seven broadcasts
per party per year). The analysed PPBs were not electoral party broadcasts but normal,
routine time shows broadcast on a regular basis (frequency depending on the parties’ vote
share). In the period under study two elections took place in Belgium: general/regional
elections in October 1999 and local elections in October 2000. In 1999 the CD&V-SP.A
government was caught in the middle of the so-called dioxin crisis and severely punished by
the electorate. For the first time in 50 years the christian-democrat CD&V ended up in the
opposition, together with the extreme-right Vlaams Blok. The victorious liberal party VLD
and, for the first time, the green party Agalev entered government. SP.A, the Flemish
socialist party, remained in government. The 1999 elections and their spectacular outcome
may have affected the content of the PPBs, but we ensured comparability between parties by
selecting PPBs in roughly the same period for each single party. We expect the differential
political position – government or opposition - of the six parties at stake to partially account
for differences in populism in their PPBs. The 2000 local elections yielded much less
eccentric outcomes and did not bring about major political change.
We tackled the 20 hours of PPB television drawing on our concept of thin populism. We
used thin populism as a heuristic device to select specific excerpts. A thorough search for
references to the people in general or to specific population categories yielded almost 1200
PPB excerpts. These thin, populist excerpts were typed out in full and painstakingly content-
analysed afterwards. This was done drawing on our thick conception of populism. We first
assessed their extent of anti-establishmentness, using a scaled distinction between specific
and diffuse anti-elite attitudes: the more diffuse the criticism of the elites, the stronger the
discourse’s anti-elitism, and vice versa. At the same time we also determined the target of the
excerpts’ anti-establishment rhetoric: politics, the state or the media. Second, evaluation of
the specific population categories mentioned in the excerpts was carefully estimated, based
on a very detailed codebook, and using a simple trichotomy: positive, neutral, and negative.
The more certain population categories are negatively treated and stigmatised in the PPBs,
the more we consider their parties to embrace an exclusive populism. More methodological
details can be found in the appendix.
MEASURING POPULISM IN POLITICAL COMMUNICATION
Thin populism: the people-index; referring to the people
To determine the degree of thin populism in the political communication of Belgian political
parties, we constructed a people-index. The index uses two different measures tapping the
amount of references to the people in the selected PPBs: proportion and intensity.
Proportion refers to the share (in characters, interspacing included) of the total PPBs that
could be characterised as thinly populist because they explicitly referred to the people in
general. For all parties, the total length of all excerpts containing references to the people
was related to the total estimated length of a standard PPB. TABLE 1 shows the figures.
TABLE 1: Proportion of thin populism in population-mentioning passages in PPBs of
each party (1999-2001)
The table shows that the extreme-right party Vlaams Blok relies far most on people-
mentioning discourse. As many as 37% of Vlaams Blok’s total PPB time contains references
to the population. However, Vlaams Blok is not the only party that frequently mentions the
people. All parties display their proximity of the people by referring to the public at large.
The intensity of the six parties’ thin populism is summarised in TABLE 2. Intensity refers to
the number of times ‘the people’ are mentioned in the citations included in the analysis. It
therefore measures to what extent the party’s discourse is permeated with population
TABLE 2: Intensity of thin populism in population-mentioning passages in PPBs of each
Again, Vlaams Blok champions the people. It talks about the people to a much larger extent
than the other parties. The green party Agalev catches the eye because of its low intensity. It
not only mentions the people less than other parties (TABLE 1), but when it does, the
passage is usually not infused with numerous references to the population (TABLE 2).
The compound people-index, is a simple multiplication of proportion and intensity. It
summarises the degree of thin populism in the PPBs’ discourse. The results are charted in
FIGURE 1: Thin populism: people-index (proportion*intensity) for each party (1999-2001)
Vlaams Blok dwarfs all other parties when it comes to mentioning the people. During the
scrutinised period (1999-2001), the christian-democrat CD&V, a traditional catch-all centre
party which dominated the Belgian government in the entire post-war period, was,
exceptionally, part of the opposition. This oppositional role probably explains its relatively
high score on the people-index. The same applies to the oppositional Flemish nationalist
VU-ID and its above average thin populism score. The lowest people-index is to be found in
the PPBs of the government parties: VLD (liberal party), SP.A (socialist party) and,
especially, Agalev (green party). Agalev’s low score is surprising: as a radical green party it
fundamentally challenges the prevailing political and economic system. However, in
rhetorical terms it obviously does not do so in the name of the people.
All this suggests that thin populism is a discursive strategy employed by opposition parties,
fuelling discontent in society and challenging the government by identifying itself and siding
with the people. Nevertheless, Vlaams Blok clearly stands out and surpasses normal
opposition rhetoric. Does the same apply to thick populism, to anti-elitism and exclusion?
Thick populism: the anti-establishment-index; against politics, the state and the media
We conceptualised anti-establishment as the vertical dimension of populism, identifying with
the people against the established elites. An all-encompassing anti-establishment measure
should take into account the different kinds of bad elites (politics, the state and the media)
and the degree of hostility towards these elites. In tables 3, 4 and 5 the degree of anti-state,
anti-politics and anti-media discourse is presented, taking into account the number of
excerpts and the scaled intensity of their ‘anti’-profiling. More technical details about these
scales and how they were operationalised can be found in the appendix.
TABLE 3: Anti-state discourse in population-mentioning passages in PPBs of each party
TABLE 4: Anti-politics discourse in population-mentioning passages in PPBs of each
TABLE 5: Anti-media discourse in population-mentioning passages in PPBs of each party
Anti-state and anti-media discourse is rare in PPBs, except in Vlaams Blok broadcasts.
Vlaams Blok’s PPBs contain many populist excerpts in which the party positions itself
against the state and takes up arms against the media. In terms of anti-politics, it is obvious
that all parties criticise other political actors to some extent, but Vlaams Blok’s criticism is
omnipresent in PPBs. Moreover, it is much more fervent and diffuse, targeting the whole
political system rather than specific politicians or political parties. The total anti-politics
count of Vlaams Blok alone (228), largely outweighs the total aggregated anti-politics score
of all other parties together (173). A simple addition of the three separate anti-establishment
scores per party yields an overall anti-establishment-index charted in FIGURE 2.
FIGURE 2: Anti-establishment-index (anti-state + anti-politics + anti-media) of each party
A deep cleavage cuts through the Belgian party system. Vlaams Blok distinguishes itself
prominently from the other parties by its outspoken and ubiquitous anti-establishment
discourse. This deep dividing line firmly splits the Belgian parties into two groups: Vlaams
Blok on one side and all other parties on the other side. Once more, both other opposition
parties, CD&V and VU-ID, display a slightly more cavilling discourse. However, they are
totally dwarfed by the Vlaams Blok’s pervasive and virulent anti-elitism. In terms of the first
aspect of thick populism, Vlaams Blok is an all-out populist party. What about the second,
horizontal dimension of thick populism, the exclusion of specific population categories? Is
Vlaams Blok extraordinary in this respect as well?
Thick populism: the exclusivity-index; excluding specific population categories
All parties have their traditional enemies — capitalists, workers, freethinkers, Catholics, etc.
Are these classic animosities still reflected in their political communication at the turn of the
20th century? Hardly. As TABLE 6 shows, most parties rarely discredit specific categories. In
contrast, they celebrate specific groups and show that they care for their problems. Most
political parties refer to particular groups in society overwhelmingly in a neutral context,
neither praising nor condemning them. Only one party stands out for its negative discourse:
Vlaams Blok. More than 70% of all negative evaluations in all PPBs of all parties were found
in the PPBs of this single party. The most positive parties are VU-ID and Agalev which
hardly broadcast any disapproving evaluations of societal groups and frequently court
particular population categories.
TABLE 6: Negative, neutral and positive evaluations of specific population categories in
population mentioning passages in PPBs of each party (1999-2001)
The figures in TABLE 6 can be summarised in so-called J-scores: the subtraction of the
positive minus the negative evaluations divided by the total amount of mentions (positive +
negative + neutral), yielding a relative measure between -1 and +1. FIGURE 3 contains the
results of this exercise.
FIGURE 3: Exclusivity-index: J-scores for evaluations of specific population categories of
each party (1999-2001)
Again, Vlaams Blok stands out. It is the only party with an unambiguous negative score
indicating that it systematically follows an exclusion strategy stigmatising and blaming groups
in society (in particular immigrants, asylum seekers and criminals). This is, by the way, the
exact argumentation given by the judges when they condemned the party for racism in 2004.
The party strongly embodies our second dimension of thick populism. Checking other
parties’ exclusivity-index teaches us that, this time, there is no clear government-opposition
pattern. Both other opposition parties, CD&V and VU-ID, embrace a discourse explicitly
including many groups. Among the government parties VLD, SP.A and Agalev, the green
party excels in positivism and woos specific population categories most.
Thin and thick populism combined
Thin and thick populism have been measured separately and can be combined. Drawing on
all PPB passages referring to the people (thin populism) how can we position the Belgian
parties in terms of thick populism’s two dimensions, anti-elitism and exclusivity? FIGURE 4
contains the three constructed indexes grasping thin and thick populism. It also graphically
illustrates the four quadrants of populism we discussed before. Thin populism (people-
index) is visualised by the size of the bubbles: the bigger the bubble the more the party
mentions ‘the people’ in its PPBs. Thick populism is visualised by the vertical and horizontal
position of a party’s bubble. The vertical dimension of the graph corresponds with the vertical
dimension of thick populism (anti-establishment-index): the higher up a party is situated in
the graph, the less this party adopts an anti-establishment discourse; the lower a party’s
position, the more it engages in attacking the state, politics and the media3. The horizontal
dimension of thick populism matches with the horizontal position in the graph (exclusivity-
index). The more a party is situated on the right side of in the chart, the more it is
characterised by a positive discourse towards groups in society; the more a party’s position is
on the left side, the more exclusionary its rhetoric.
FIGURE 4: Thin and thick populism: people-index, anti-establishment-index and
exclusivity-index of each party (1999-2001)
3 We created an artificial zero point for the vertical anti-establishment dimension by taking the average of
the anti-establishment-index of all six parties. As we did not measure pro-establishment stances a
position in the upper half of the graph does not mean that this party develops a pro-establishment
discourse, but only that it engages less in anti-establishment statements than the average party. The
eccentric position of Vlaams Blok, however, distorts the whole scale somewhat and conceals the mutual
differences between the other parties.
FIGURE 4 makes it astonishingly clear that Vlaams Blok’s discourse most fundamentally
differs from the others parties’ communication. The Belgian extreme-right party is a
textbook example of thin and thick populism scoring high on all three indexes. The party
embraces an outspoken discursive thin populist style, abundantly referring to the people and
identifying with the public at large. In terms of thick populism, the party cultivates an
unmistakably exclusive and fervently anti-establishment variant of populism. Other Belgian
parties as well adopt a thin populist style to some extent (their bubbles have a certain size),
but their populism cannot be considered as exclusive, shutting out specific groups of the
population, nor is it permeated with anti-elite statements (their bubbles are all positioned in a
quadrant opposing the Vlaams Blok’s position). The conclusion can only be that, in terms of
political communication, Vlaams Blok fundamentally differs from the other parties,
confirming the hypothesis we set off with.
CONCLUSION AND DISCUSSION
This paper intended to make an empirical, a conceptual, and an epistemological point. In
terms of the empirical point, not much more needs to be said. The outcome of the empirical
analysis speaks for itself: the Belgian extreme-right party Vlaams Blok is a text-book case of
populism. Regarding the conceptual goal of the study, we proposed a clear, succinct,
multidimensional and multilayered definition of populism enabling us to broaden the scope
of populism research to analyses comprising not only ‘established’ populist parties but all
political parties. The distinction between thin and thick populism is useful. Whether
considered as merely an operational device to select communication for further scrutiny or
as the true core of populism, the concept of thin populism helps us to understand and
operationalise the elusive concept of populism. Combined with both dimensions of thick
populism, thin populism offers a rich understanding of populism and its different guises.
This brings us to our epistemological goal. We wanted to develop a concept that is
measurable in quantitative empirical research. We hope to have shown that populism, as we
defined it, can be reliably, empirically gauged by drawing on a careful analysis of the political
discourse of political actors. Thin and thick populism can be measured. In fact, we claim that
the same kind of content analysis can easily be undertaken by drawing on the
communication of other political actors. Not only parties, but individual politicians,
movement leaders, journalists and TV-makers too can embrace a populist discourse and
refer to ‘the people’ all the time (whether or not attacking the establishment and stigmatising
specific categories). The proposed conceptualisation of populism, in addition, is not bound
to the Belgian context but suitable for comparative research.
Let us ponder further on the four-fold typology of populism. We can coin a name for our
four types of populism: complete populism (lower-left quadrant); excluding populism (upper-
left); anti-elitist populism (lower-right) and empty populism (upper-right). If one looks at
FIGURE 4 our populism typology is structured theoretically from the upper-right quadrant,
complete populism, diagonally to the lower-left quadrant, empty populism. The closer a
political actor comes to the lower-left corner of the graph, the more its political discourse
resembles the thick and classic definition, or ideal type, of populism. ‘Normal’ political
discourse is predominantly situated in the upper-right quadrant (empty populism). This
positioning of parties, however, only makes sense if they display at least a certain degree of
thin populism (symbolized by the size of the bubble). Thin populism is a minimal
precondition for thick populism. If discourse does not refer to the population yet fiercely
criticises the establishment and at the same time stigmatises popular categories, it cannot be
considered as populism since the required appeal to the people is missing (the size of the
bubble will be small or even non-existent). FIGURE 5 visualises these ideas.
FIGURE 5: Types of populism
Our results suggest that all three defining aspects of populism are associated. Thin and thick
populism, at least in the Belgian context in 1999-2001, go hand in hand: parties that often
refer to the people also attack the establishment and exclude groups. Both dimensions of
thick populism are associated as well; anti-elitism and exclusiveness, or their absence,
characterize the same parties. On the one hand, this association strengthens our belief that
we are really dealing with a consistent populist type of political communication. On the other
hand, the Belgian findings challenge the notion that we genuinely have a typology containing
four full types of populism: among the Belgian parties only two types seem to exist and two
quadrants remained empty. This questions the empirical multidimensionality of populism.
However, we do believe that all quadrants are in principal possible and that adopting the
corresponding type of populism is plausible. We would need more evidence on more
countries to substantiate this claim, but we have some anecdotical Belgian evidence. In terms
of the (absent) anti-elitist populism - anti-elitism without exclusion - we deem this might
have been the position of the green party Agalev before it took office in 1999. Also the
VLD’s discourse before 1999 – the party was in the opposition - was characterized by a
fierce anti-elite stance inspired by the present Prime Minister Guy Verhofstadt’s so-called
Citizens’ Manifests (Blommaert, 2001). The reason why, in 1999-2001, the CD&V did not
capitalise on the anti-establishment dimension, might have been caused by the party’s recent
banishment to the opposition. It still employed a government-style discourse after more than
fifty years of office. In terms of the excluding populism– absence of anti-elitism combined
with exclusion - we did not find any in our 1999-2001 analysis either. However, recent
switches in the stances and rhetoric of the two main parties, VLD and CD&V, towards a
stronger law-and-order position, towards a more demanding and critical position vis-à-vis
immigration, and towards a more Flemish nationalist and anti-Belgian stance might have
entailed, in terms of our typology, a shift from empty populism to excluding populism.
However, all these counter examples cannot prevent the empty vs. complete populism
dimension from remaining dominant and cases falling into one of these two quadrants from
being encountered most frequently in reality.
The empirical results and the above plausibility discussion suggest that the type of populism
adopted by political actors depends on their position in the political game. It is, for example,
a government or opposition position that can partially account for populism or its absence.
Let us come back to the specific position of the Belgian Vlaams Blok to refine and elaborate
this idea. Vlaams Blok is not just an ordinary opposition party; it is a party that, via the
cordon sanitaire, is totally excluded from participation in political decision-making at every
political level. As a deeply-rooted and eternal opposition party it is not surprising that
Vlaams Blok is deeply populist. But is VB populist because it is excluded or is it excluded
because it is populist? We have no systematic evidence, but it appears that the party first
adopted populism and only afterwards was caught in a cordon sanitaire. From its foundation
in 1978 onwards the party fostered a fierce anti-elitist stance and soon developed an
exclusive discourse in particular shutting out foreigners. Remarkably, the party seemed to
have become only fully populist after its first major electoral success. Mentioning the people
and contending to speak in the people’s name, indeed, only makes sense when a considerable
share of the people support the party in the ballot box. This implies that parties not only
embrace populism because of strategic reasons – it may be a fruitful strategy for opposition
parties - but because of ideological reasons. The more parties fundamentally oppose the
political system in which they operate, the larger the chance that they will embrace a populist
style. In the case of the Vlaams Blok it is the party’s extreme nationalism - the party wants to
break up the Belgian state and set up a separate Flemish state - that explains its populism.
This also implies that unless the Vlaams Blok gets what it wants, i.e. the abolishment of
Belgium, it will probably sustain its populist and corrosive discourse, even when it is in a
position to take office. This may distinguish Vlaams Blok from most other extreme-right
parties in Western-Europe.
Let us, at the end of our endeavour, sketch two tracks for further research. First, we
established that Vlaams Blok is very different from the other Belgian parties. However, this
does not mean that we substantiated that the populist communication of Vlaams Blok can
explain its electoral success. We only carried out an analysis of the message sent out by Vlaams
Blok. To test whether this message makes a difference and affects voters, we need evidence
about the reception of the populist discourse among voters. What effect does populism have
on voters? Do they recognise the populist frame? Do they distinguish the different types of
populism? Does an alignment between the populist frame and the voters’ own attitudes
encourage them to vote for a populist party? Does a populist discourse only attract voters
who already felt deserted by the political elites and who previously supported exclusionist
policies; or, does a populist rhetoric create these feelings and produce the breeding ground for
its own success in a kind of self-propelling process? These key questions remain unanswered
and are interesting topics for further research. The case of Vlaams Blok, though, suggests
that populism might play a role. Election surveys have shown over and over again that
Vlaams Blok has the most loyal supporters of all Belgian parties: once a Vlaams Blok voter,
always a Vlaams Blok voter. It is as if voting for Vlaams Blok is crossing Caesar’s Rubicon:
they throw the dice and there is no way back. Our analysis shows why this might be the case:
the rhetorical distance between Vlaams Blok and the other parties is enormous; its populist
discourse turns it into a different party. Voting for Vlaams Blok is symbolically turning your
back on the traditional parties and quitting the political system. Such a major move cannot
be easily undone. Vlaams Blok’s aggressive populist communication strategy constantly
nurtures the distance between itself and the other parties.
Second and finally, we call for research that explores the relationship between populism as a
communication style, the path we followed in the present study, and populism as an
ideology. Indeed, the scholarly debate on populism is often linked to the very essence of
democracy; many students consider populism to be a threat to democracy (see for example:
Mair, 2002). Defined as an ideology populism is a particular view of democracy, a democracy
theory. Populist ideology reduces democracy to its core: the principle of the sovereignty of
the people. Populism takes ‘government by the people, for the people’ literally and rejects all
checks and balances on the popular will (see for example: Kitschelt, 1995). Other
constitutive elements of democracy - the rule of law, the division of power or respect for the
rights of minorities - are rejected because they confine the people’s sovereignty. The
question then becomes whether populism as style and populism as ideology can be
disconnected or not. If politicians cultivate a rhetorical populism, like Vlaams Blok in
Belgium, do they inevitably promote an ideological populism too? In fact, as we already
mentioned, talking a lot about the people implicitly suggests that those people and their will
are the cornerstone of democracy; it is underlining the sovereignty of the people. In addition,
by stressing their direct ties with the electorate, even without exercising an anti-establishment
and exclusionary variant of populism, political leaders may reinforce public distrust towards
the institutions of representative democracy (parliament, government, political parties, etc.).
By doing so they nurture the idea that all problems would be easily solved if only the political
will was present. Moreover, referring to ‘the people’ as a homogeneous mass might create an
illusion of unity. The imagined community of ‘the people’ is in fact divided by conflicting
interests. Instead of ‘doing good for everyone’, politics is the art of making collective choices
with inevitable winners and losers. Traditional parties who are tempted, in order to counter
the extreme-right populist challenger, to emulate the populist discourse might worsen things
instead of undercutting the extreme-right’s success.
1. Selection excerpts
Excerpts were selected when they contained terms that refer to the population (population as a whole or population categories).
Every excerpt was typed out literally. The length depended on the context. Sentences were excluded that did not make sense
without context. Every excerpt contains only words from one person, taking into account that the off-screen voice(s) is (are)
considered to be one and the same person. Most of the time, the beginning of a new excerpt coincided with a switch of images.
The reference to ‘the people’ had to cover the people in political terms, meaning the political entity. Any reference that did not meet
this criterion was not selected.
Furthermore, a population category is defined as a group of people having in common a constant feature that is of electoral interest in the given
rhetorical context. This electoral component originated in the literature, which says that populism is always involved with mobilising
people or voters. A few examples can clarify this definition. First of all, the common characteristic could not be entirely
ephemeral. For instance, ‘the people who are smoking a cigarette outside’ could not be generalised to ‘the smokers’ and thus this
excerpt was not taken into account. The second feature of the definition can be illustrated by a similar example. Suppose a certain
party in one of its PPBs reports about a meeting and before the meeting starts someone says ‘we have to wait because the
smokers are still outside’. This is a practical remark without taking a political position about smoking, so it was not included in our
2. Encoding excerpts
2.1 Encoding references
In the coding scheme, there was room for maximum five referential terms per excerpt. For each reference to the people it was
noted whether it concerned a reference to the whole population or to a population category. It was also noted how many times
the word literally appeared and how many times it was referred to by indirect terms (personal and possessive pronouns). This is
done in order to weigh the cases, with the purpose of giving a correct figure of how much the excerpt is pervaded with references
to the people. The list of references was put together inductively: during the analysis of PPB’s we decided whether a certain
mention had to be taken into account or not. The encoding process forced us to make a number of methodological choices,
explained briefly below.
2.1.1 Based on the verbatim term, the reference was put in the corresponding category to avoid a never ending list of
referential terms. Content takes priority over verbatim terms.
2.1.2 The terms referring to the people could be split up in two kinds: those which refer to the population (group) as an
undividable unity, mostly preceded by a definite article (e.g. the voter, the people, the consumer) and those which do not.
Since populists are inclined to mobilise as broadly as possible, the first kind of referential terms can be regarded as the
most solid indicator of populism.
2.1.3 Terms like ‘public opinion’, ‘(political) participation’ and ‘democracy’ are also noted as (indirect) references to the
people if they allude to ‘the will of the people’.
2.1.4 Nouns used as adjectives to point at a population category were not selected as a reference. E.g. “The Belgian Flor Van
Noppen is 42 years old and self-employed entrepreneur.”
2.1.5 The selected references had to reach beyond the private life of the speaker. The reason is the limited mobilising effect
of terms that do not go beyond the private life. E.g. “My father is a doctor” vs. “Doctors want to be respected by their
2.2 Evaluating references
Interpreting attitudes is a delicate matter. We realise that our data were not meant to be objective, but that we are working with
pre-eminently coloured information: political propaganda. Notwithstanding, we have ventured onto thin ice, very cautiously and
aware of the existing risks.
Simply talking about something is paying attention to it and, implicitly, shows most of the time a positive attitude towards the
population category in question. Although there is no explicit positive term in the following example, it seems clear that neutral
references are often the expression of a positive attitude.
Example: necessary but insufficient condition
“Workers and small businessmen, Vlaams Blok is the party of the (ordinary) people.” (off-screen voice in NOS
Because of this vagueness and ambiguity we adopted a stricter criterion for deviating from the ‘neutral’ default interpretation. For
methodological reasons of uniformity we adopt the following standard: there has to be given a positive fact about the mentioned
population category before we encoded this reference as positive.
Example: sufficient condition
“Another example: we notice that among our young people there is a lot of creative talent. Today they do not get the
chance or opportunity to rehearse, or rent a place at a reasonable price.” (Sara Boogers in GROM 30/5/2000)
“I have a Moroccan friend and it is very important to me that everyone accepts this.” (citizen in GROM 8/10/2001)
“When I came back from Oxford, I was a pleasantly surprised to see so many young people active in the Socialist Party
today.” (Frank Vandenbroucke in SOM 23/2/1999)
Only explicit positive references are encoded as positive references because this is the only way to guarantee that border line cases
are judged consistently. However, this implies that some implicitly positive references are in fact evaluated as neutral. Anyhow, the
criterion applies for all quotes of all parties so, comparatively, no party’s relative stance is changed.
Example: implicitly positive but neutral evaluation because of strict criterion
“We urgently have to pay attention to the content of education and focus on the pupil again. Something has to be
done about young people and make kids feel better at school again.” (Marleen Vanderpoorten in LIBRADO
“A prospective policy has an eye for what is of interest to young people. Creating opportunities for young people is
one of the biggest challenges for our ruling elite.” (off-screen voice in LIBRADO 16/5/2000)
2.3 Encoding anti-establishment statements
For each of the different kinds of elites (politics, the state and the media) we designed a similar scale to measure the degree of
hostility (from specific to diffuse). The most diffuse form of hostility was identical for all the three elite groups. This is the case
when ‘the system’ as such is criticised and attacked. The lower values on the scale, i.e. the specific ‘anti’-attitudes, are part of every
normal political discourse. Criticising a specific government, for example, is not an exceptional political activity.
Anti-state remarks focus on the failure of the state to render the required service (public transport, guarantee the right to justice or
safety, the administration, semi-political institutions). Let us take a closer look at our anti-state scale (1-5).
1. Single failure: one specific fact is criticised, without pointing at the public service as a whole. E.g. “Yesterday, the
(public) train was 10 minutes late.”
2. Systematic failure: criticism is more substantial and concerns the functioning of the public service as a whole. E.g.
“(Public) trains are always at least 10 minutes late.”
3. Public service should be abolished: public services or institutions serve no purpose and will be abolished as soon as
possible. E.g. “Public transport should be privatised.”
4. All public services are criticised at once. E.g. “We have a lot of competent officials, but in general public services are
5. The system: e.g. when the judiciary is seen as the accomplice of the denounced ruling regime.
Our main concern here was to differentiate quotes which play the man and not the ball (or the other way around). Let us take a
closer look at our scale values.
1. Policy measure or present situation: a certain measure or present situation is criticised. In most of the cases, it concerns a
present situation against which a political party wants to take action. E.g. “It cannot be that teachers are burdened with
tasks that have nothing to do with their reaching task” (teacher in VNOS 20/12/2000)
2. Policy: to direct criticism against the government policy, without playing the man. Pointedly, criticism is focussed on the
content of the policy. E.g. “The signal given by the government is almost equal to advertising in favour of narcotics. In my opinion,
that is one step too far.” (Ingrid Van Kessel in CDO 13/2/2001)
3. Politician: the name of a certain politician is stated in a negative context. E.g. “Minister Miet Smet manipulated the
employment figures” (Pierre Chevalier in LIBRADO 21/3/1999)
4. Party: negative attention is focussed on one single party. E.g. “In Antwerp, Groen! runs the health and environment
department, and nevertheless they voted for an incineration of household rubbish and dioxin emissions in a residential
area.” (off-screen in NOS 14/12/1999)
5. Group of parties: criticism is targeted at more than one party or e.g. at the government. In the case of opposition rhetoric,
one must ask whether it is the policy or the government itself that is the main subject of criticism. “In my opinion, the
government makes a bad impression by governing in a totally irresponsible way.” (Geert Bourgeois, VNOS 13/11/2001)
6. All parties are criticised, except the own party. E.g. “While the so called democratic parties were working hard on removing
subsidies from a troublesome opposition party, in Schaerbeek a Flemish worker was beaten to death by a group of young
immigrants.” (off-screen NOS 12/1/1999)
7. The system: E.g. “(…) And so it is clear that the Flemish majority in this country is once again ignored. Within the
Belgian framework, democracy is no longer possible.” (Joris Van Hauthem in NOS 1/9/1999)
As shown by the results of the analysis, anti-media statements are rare in comparison to the criticisms of the other two elite
groups. On our scale only value three, four and five were present in the encoded dataset. From specific to diffuse, we
differentiated between following media targets of criticism:
3. Newspaper/magazine/TV-channel: e.g. the Flemish public TV-channel VRT is criticised for being politically biased.
4. Group of media: e.g. TV-channels VRT and VTM (commercial) are blamed for being too soft on certain media savvy
5. All media (the media): e.g. the media as a whole are reproached of having no attention for the election campaign of a
K. Abts, 2004, Het Populistisch Appel. Voorbij de populaire communicatiestijl en de
ordinaire democratiekritiek. Tijdschrift voor Sociologie, 25, 451-476.
K. Anckar, 2000, Size and Party System Fragmentation, Party Politics, 6, 3, 305-328.
H.-G. Betz, 2004, La droite populiste en Europe. Extrême et démocrate? Paris : Corlet.
J. Billiet & H. De Witte, 1995, Attitudinal dispositions to vote for a ‘new’ extreme right-wing
party: The case of ‘Vlaams Blok’, European Journal of Political Research, 27, 181-202.
J. Billiet & M. Swyngedouw, 1995, De keuze tussen groen en extreem-rechts. Sporen van een
nieuwe breuklijn, Res Publica, 35, 2, 247-262.
J. Blommaert, 2001, Ik stel vast: politiek taalgebruik, politieke vernieuwing en verrechtsing, Berchem:
P. Buzzi, 1994, Le front national: entre national-populisme et idéologie d’extrême droite.
Notes et études documentaires, 4996, 15-36.
M. Canovan, 1981, Populism. London: London Junction books.
M. Canovan, 1999, Trust the people! Populism and the two faces of democracy. Political
Studies, 47, 1, 2-16.
H. Coffé, 2005a, Do individual factors explain the different success of the two Belgian
extreme right parties? Acta Politica, 40, 74-93.
H. Coffé, 2005b, Extreem-rechts in Vlaanderen en Wallonië. Het verschil, Roeselare: Roularta
A. de Benoist, 2000, Wie is er bang van het populisme?, Telos, 96, 6-8.
T.S. Di Tella, 1997, Populism into the twenty-first century, Government and Opposition, 32, 2,
K. Deschouwer, 2001, De zorgeloze consensus. De statuten van het Vlaams Blok en de
partijentheorie, Tijdschrift voor Sociologie, 22, 1, 63-87.
M. Elchardus, 1994, Gekaapte Deugden, Samenleving en Politiek, 1, 1, 20-27.
M. Elchardus & K. Pelleriaux, 1998, De polis verdeeld – hoe de kiezers links en rechts
herdefiniëren, In: Marc Swyngedouw, Jaak Billiet, Ann Carton en Roeland Beerten
(red.), De (on)redelijke kiezer, Leuven: ACCO, 183-210.
M. Elchardus, 2001, Moet populisme omarmd worden als politieke vernieuwing?, Socialisme
en democratie, 58, 226-258.
A. Hennessy, 1969, Latin-Amerika. G. Ionescu & E. Gellner, Populism. Its meanings and national
characteristics. London: Weidenfeld and Nicholson, 26-61.
P. Ignazi, 1992, The silent counter-revolution. Hypotheses on the emergence of extreme
right-wing parties in Europe, European Journal of Political research, 22, 3-34.
R.S. Katz & P. Mair, 1994, How parties organize: change and adaptation in party organizations in
western democracies, London: Sage
M. Kazin, 1995, The populist persuasion: an American history, New York: Basic Books.
H. Kitschelt & A.J. McGann, 1995, Radical right in Western-Europe. A comparative analysis. Ann
Harbor: University of Michigan Press.
J-E. Lane & S. Ersson, 1991, Politics and society in Western Europe, London: Sage.
M. Lazar, 1997, Du populisme à gauche: les cas français et italien. Vingtième siècle – Revue
d’Histoire, 56, 121-131.
B. Maddens, 1998, De partijkeuze en de beleidsthema’s, In: Marc Swyngedouw, Jaak Billiet,
Ann Carton en Roeland Beerten (red.), De (on)redelijke kiezer, Leuven: ACCO, 69-91.
P. Mair, 2002, Populist Democracy vs. Party Democracy. Y. Mény & Y. Surel, Democracies and
the populist challenge, Basingstoke: Palgrave, 81-100.
Y. Mény & Y. Surel, 2000, Par le peuple, pour le peuple, Paris: Fayard.
Y. Mény & Y. Surel, 2002, Democracies and the populist challenge, Basingstoke: Palgrave.
C. Mudde, 2000, The ideology of the extreme right, Manchester: Manchester University Press
C. Mudde, 2004, The Populist Zeitgeist. Government and Opposition, 39, 542-563
W. Müller & K. Strøm, 1999, Policy, Office or Votes? How political parties in Western Europe make
hard decisions, Cambridge: Cambridge University Press.
A. Pfahl-Traughber, 1994, Volkes Stimme? Rechtspopulismus in Europa, Bonn: Dietz
M. Swyngedouw, 2001, The subjective cognitive and affective map of extreme right voters:
using open-ended questions in exit-polls. Electoral Studies, 20, 2, 217-241.
M. Swyngedouw, 1992, Waar voor je waarden. De opkomst van Vlaams Blok en Groen! in de jaren
’80, Leuven: ISPO.
P. Taggart, 2000, Populism. Buckingham: Open University Press.
P.-A. Taguieff, 1995, Political science confronts populism: from a conceptual mirage to a
real problem. Telos, 103, 9-44.
P.-A. Taguieff, 1998, Populismes et antipopulismes: le choc des argumentations, Mots – Les
langages du Politique, 55, 5-26.
W. Van der Brug, M. Fennema & J. Tillie, 2000, Anti-Immigrant Parties in Europe:
Ideological or Protest vote? European Journal of Political Research, 37, 77-102.
S. Walgrave & K. Deswert, 2004, The making of the (issues of the) Vlaams Blok. Political
Communication, 21, 4, 479-500.
A. Walicki, 1969, Russia. G. Ionescu & E. Gellner, Populism. Its meanings and national
characteristics. London: Weidenfeld and Nicholson, 62-96.
M. Wievorka, 1993, La démocratie à l’épreuve: nationalisme, populisme et ethnicité. Paris: La
P. Worsley, 1969, The concept of populism. G. Ionescu & E. Gellner, Populism. Its meanings
and national characteristics. London: Weidenfeld and Nicholson, 212-250.