Electronic copy available at: http://ssrn.com/abstract=2035539
Conceptualizing cultural groups and
cultural difference
The social mechanism approach
ROLAND PIERIK
Tilburg University,The Netherlands
ABSTRACT The aim of this article is to present a conceptualization of cultural
groups and cultural difference that provides a middle course between the Scylla of
essentialism and the Charybdis of reductionism. The method I employ is the social
mechanism approach. I argue that cultural groups and cultural difference should be
understood as the result of cognitive and social processes of categorization. I
describe two such processes in particular: categorization by others and selfcategorization.
Categorization by others is caused by processes of ascription: the
attribution by outsiders of certain characteristics, beliefs, and practices to individuals
who share a specific attribute. Self-categorization is caused by processes of
inscription and community-building: the adoption of certain beliefs and practices as
a result of socialization and enculturation. I therefore shift the focus from groups to
categories, and from categories to processes of categorization. I show that this
analytical distinction between categorization by others and self-categorization can
clarify an ambiguity in dominant debates in contemporary multiculturalism. I
conclude by indicating how injustices, commonly associated with multiculturalism,
can better be understood as socially generated injustices, and how government
should deal with these injustices.
KEYWORDS categorization ● constructivism ● multiculturalism ● social
categories ● essentialism
INTRODUCTION
In the last decades, cultural diversity has become a subject of major interest
in political theory. Although the normative importance of multiculturalism
A R T I C L E
Copyright © 2004 SAGE Publications (London,Thousand Oaks,CA and New Delhi) 1468-7968
Vol 4(4):523–544;047472
DOI:10.1177/1468796804047472
www.sagepublications.com
Electronic copy available at: http://ssrn.com/abstract=2035539
has been widely recognized, it is less clear how cultural difference can be
conceptualized while avoiding the pitfalls of essentializing cultural groups.1
The normative theory building in multiculturalism also encompasses the
descriptive debate: there seems to be an equilibrium between differential
conceptualizations of cultural groups and the normative claims made on
behalf of these groups. For example, the debate on Will Kymlicka’s Multicultural
Citizenship (1995) not only concerned his normative emphasis on
culture as the context of individual choice. Many critics also focused on his
conception of societal culture and cultural difference, in particular, his
‘categorical distinction’ between multinationality and polyethnicity (1995:
11–14).2
In this article, I seek to contribute to these normative debates in multiculturalism
by focusing on the underlying descriptions of cultural groups.3
Despite its centrality in normative multiculturalism, the term ‘group’ is
seldom analyzed in its own terms (Brubaker, 2002: 163). Cultural groups
are taken for granted as distinct entities, internally homogenous, externally
bounded, and seen as basic constituents of social life. Such a conceptualization
runs the risk of falling prey to the culturalistic fallacy:
The culturalistic fallacy may be said to be committed when one defines culture
as an ideational abstraction and then proceeds to convert or reify this ens
rationis into an independent ontological entity subject to its own laws of
development and conceived through itself alone. (Bidney, 1953: 51)
We can distinguish three aspects of the culturalistic fallacy. First, the
reification of culture: to regard something abstract as something material or
concrete. Second, the compartmentalization of culture: the tendency to view
cultures as discrete entities with sharp borders. Third, the essentializing of
culture: the tendency to see culture as an autonomous and immutable
entity, in which its individual members are regarded as only the passive
bearers of culture. The extreme essentialist and naturalized descriptions are
nowadays generally dismissed.4 Over time, we have seen a shifting emphasis
from ‘natural’ to ‘cultural’ descriptions of groups, phrased in terms of
‘blood’ via ‘race’ and ‘ethnicity’ to ‘culture.’5 It is generally accepted now
that culture is a socially constructed concept.
On the other hand, constructivism dissolves into reductionism when it
denies that culture is a real phenomenon in society, and merely sees it as a
‘narrative discourse,’ a ‘process,’ or as an ‘identity.’6 It might be true that
the essence of culture and cultural difference is better captured in a
narrative approach (cf. Preston, 1995; Somers and Gibson, 1994). However,
such an approach means that we cannot generalize about cultural difference
within our political institutions, and therefore undermines a normative
discussion of these issues in political theory (Bader, 2001).
The aim of this article is to present a conceptualization of cultural groups
and cultural difference that provides a middle course between the Scylla of
ETHNICITIES 4(4)524
525
essentialism and the Charybdis of reductionism. The method I employ is
the social mechanism approach. I argue that cultural groups and cultural
difference should be understood as the result of cognitive and social
processes of categorization. I describe two such processes in particular:
categorization by others and self-categorization. Categorization by others is
caused by processes of ascription: the attribution by outsiders of certain
characteristics, beliefs, and practices to individuals who share a specific
attribute. Self-categorization is caused by processes of inscription and
community-building: the adoption of certain beliefs and practices as a result
of socialization and enculturation. I therefore shift the focus from groups
to categories, and from categories to processes of categorization. I show
that this analytical distinction between categorization by others and selfcategorization
can clarify an ambiguity in dominant debates in contemporary
multiculturalism. I conclude by indicating how injustices, commonly
associated with multiculturalism, can better be understood as socially
generated injustices, and how government should deal with these injustices.
THE SOCIAL MECHANISM APPROACH
The social mechanism approach is a methodology that emerged in reaction
to the frustration about the unfruitfulness of the quest for universal laws in
social sciences.7 Mechanisms are explanatory devices halfway between
description or storytelling and universal laws of the form ‘if and only if A
then B.’ Social mechanisms are explanatory devices that seek to explain the
relations between (complexes of) interactions among individuals and
aggregate social regularities. The social mechanism approach is a qualitative
(instead of a quantitative) approach in which one seeks to find an explanation
that provides a plausible account of how one phenomenon can be
explained as a result of other(s) (Hedström and Swedberg, 1999: 7).
Moreover, it is an analytical approach: it does not intend to provide a
complete historical or anthropological account of social reality as it
actually happened – no theory can ever give such an account. Instead, it
intentionally aims to present an abstract and intelligibility account of the
relation between the explanans and explanandum (Hedström and
Swedberg, 1999: 13–14). Elster (1999: 1) gives the example of the effect of
an alcoholic environment on children. It is impossible to construct a general
law of what becomes of the child of an alcoholic. He or she might turn into
an alcoholic, or, quite the opposite, a teetotaler. Here the mechanism
approach can be helpful: we can we can think of very plausible explanations
for both reactions.8
Virtually no single phenomenon in the social sciences can be explained
in terms of general or universal laws – including the genesis of cultural
PIERIK ● CONCEPTUALIZING CULTURAL GROUPS
groups. Jon Elster and others have emphasized the usefulness of social
mechanisms for the explanation of such complex psychological and social
phenomena (Elster, 1999: 32). I will explain the genesis and existence of
cultural groups in terms of transformative mechanisms: mechanisms that
explain the emergence and existence of social phenomena on a macro scale
as the aggregated results of individual behavior on a micro scale (Hedström
and Swedberg, 1999: 21). In general, such social phenomena cannot be
explained in terms of one single mechanism, but, instead, need to be
explained by a combination of several elementary mechanisms (Hedström
and Swedberg, 1999: 21, 24; Elster, 1999: 6).9 In this article I will explain the
genesis of cultural groups as a result of (the interactions between) three
elementary mechanisms: inscription, ascription and community-building.
SOCIAL CATEGORIZATION, CULTURAL GROUPS, AND
CULTURAL DIFFERENCE
In this section I conceptualize the genesis and existence of cultural groups
and cultural difference in terms of cognitive and social processes of
categorization. Although one may not expect it at first sight, the act of
categorization and the use of categories play an important role in dayto-day
life. The world consists of an infinite number of objects and differences
and similarities between them, and without the ability to categorize,
‘we could not function at all, either in the physical world or in our social
and intellectual lives’ (Lakoff, 1987: 5–6).10 Jerome Bruner concludes that:
There is, perhaps, one universal truth about all forms of human cognition: the
ability to deal with knowledge is hugely exceeded by the potential knowledge
contained in man’s environment. To cope with this diversity, man’s perception,
his memory, and his thought processes early become governed by strategies
for protecting his limited capacities from the confusion of overloading.
(1979[1962]: 65)
Knowledge about one’s environment is the result of (mentally) organizing
it in categories.11 Animals, for example, are categorized in species:
mammals, reptiles, amphibians, etc. Survival in the forest may depend upon
a correct categorization of animals into classes such as ‘dangerous’ and
‘innocent.’ Categorization can be defined as the ordering or arrangement of
objects into groups or sets on the basis of their observable or inferred
relationships (Sokal, 1974: 1116). The purpose of categorization is to reduce
the infinity of possible differences to workable proportions, while maintaining
relevant discriminations between classes (Rosch, 1978: 29; Starr,
1992: 159). All kinds of things and situations can be and are actually
categorized and classified in this way. For the purpose of this article, I will
ETHNICITIES 4(4)526
527
concentrate on the categorization of persons – as distinguished from the
categorization of animals, material goods, situations, etc. A category, then,
is a set of individuals that are considered equivalent on a specific attribute.
An attribute is a feature that is (seen as) uniform among the individuals in
the category, for example, occupation, skin color, sex, etc. An attribute can
be invariable, e.g. eye color, date of birth, ethnic origin; or variable, e.g.
favorite film, shoe color, weight. It can be the result of a choice, e.g., the
membership of a football club; or unchosen, e.g., being part of a family.
During processes of categorization the emphasis shifts from individuals –
manifesting a multitude of characteristics – to categories – individuals
pigeonholed in distinguished classes (Minow, 1990: 53). Rogers Brubaker
concludes from research in cognitive anthropology and social psychology
that our overall mental architecture is such that we find such social
categories ‘easy to think’:
The evidence suggests that some common sense social categories – and
especially ethnic and racial categories – tend to be essentializing and
naturalizing. They are the vehicles of what has been called a ‘participants’
primordialism’ or a ‘psychological essentialism’. (Brubaker, 2002: 165–6)
Although categorization is a cognitive process, it is not necessarily an
idiosyncratic process. Individuals can invent innumerous categories, some
important for their personal or professional life, others totally random, such
as the subcategory of persons that I admire: those who ran a marathon and
read Derek Parfit’s Reasons and Persons from cover to cover. Other
categories are not idiosyncratic at all; instead, they are social categories,
used generally in society: women, Catholics, Muslims, Pakistanis, philosophers,
gays, and so forth.
Social categories obtain their own meaning by the stereotypical generalizations
about the persons within such a category: Canadians are polite,
women prefer motherhood above a career and blacks are lazy. Like
categories, such stereotypes are cognitive devices that, by relying on
categories, help us to make faster and more efficient perceptions, inference
and decision-making. The term ‘stereotype’ is used here in the nonpejorative
form and refers to the beliefs or expectations about the qualities
and characteristics of specific social categories (Nelson et al., 1996: 14).
These generalizations are assumed to apply to all members of the category.
Such stereotypes are very influential because of the human tendency to
infer strong interferences from surface similarities: ‘our thinking about
social categories gives disproportional strength to category differences
correlated with physical appearance’ (Rothbart and Taylor, 1992: 26–32,
quote on 26). Categorizations and generalizations are thus two sides of the
same coin: categories are known by their generalized labels, and generalizations
enable us to distinguish a specific social category from others. The
generalizations reinforce the differences between the different categories.
PIERIK ● CONCEPTUALIZING CULTURAL GROUPS
Based on this account of categories, I propose to define cultural groups in
terms of categories:
● A cultural group is a set of individuals who are categorized by an
individual attribute.
● In society, this category of individuals is recognized as distinguished
from people in general because of their actual or inferred
characteristics: beliefs, attitudes, practices, or modes of behavior.
● These beliefs, attitudes, practices, and modes of behavior are
generally attributed to individual members of the set.
The shared attribute is not only described in its capacity as a tool to
categorize individuals, but also as a tool to make generalizations about
these individuals within the category. This approach enables us to understand
cultural groups, as discussed in normative debates in multiculturalism,
within the terminology of social categorization. They can be seen as
sets of individuals that are considered equivalent on a specific attribute.
Consider Table 1 that gives a tentative list and can be supplemented with
other attributes, e.g. age or occupation and the related cultural groups.12
Notice that all citizens in a relevant society can be categorized according
to all the attributes as enumerated in the table. Moreover, it is not by
default the case that categories coagulate. For example, the well-known
category of ‘Wasps’ is more heterogeneous than generally assumed: not
everyone who has a western lifestyle is also European–American and
protestant.
Cultural difference is thus not only embedded in individual cognition,
but also in the basic structure of society. We can distinguish idiosyncratic
categories from social categories in the sense that the latter are social
ETHNICITIES 4(4)528
Table 1 Attributes and their related social categories
Attribute Examples of related social categories
Comprehensive lifestyles Western,Amish,Native,Hutterites
Ethnicity European-Americans,African Americans,Asian-Americans,
Latinos (in the USA)
Religion Protestants,Catholics,Muslims,Hindus,Jews,Buddhists,
Atheists
Gender Males,females
Language English,French,indigenous languages (in Canada)
Sexual orientation heterosexuals,gays,lesbians,bisexuals
529
phenomena and the result of social processes: well-established and
frequently occurring patterns of behavior in society. The effects of such
social processes, e.g. the social construction of categorical differences
between cultural groups, cannot be reduced to discernable individuals and
their considerations and preferences. Instead, they must be understood as
the aggregated and accumulated result of social interactions guided by
cultural beliefs: norms and conventions emphasizing ‘accepted’ and
‘deviant’ behavior for individuals in general and sub-categories thereof.
These conventions and norms are passed on to new members of the society
by processes of socialization before the age of reason, and over time
become internalized. Categories and stereotypes can therefore be seen as
non-conscious or unconscious cognitive devices. Research in cognitive
social psychology has shown that individuals inevitably rely to some extent
on stereotypes when making judgments (Nelson et al., 1996).
Over time, the whole of crosscutting social categories has crystallized
into a social structure. The social structure is the way in which social life is
organized into predictable relationships and patterns of social interaction,
and includes distinctions between social categories (e.g. between men and
women, and categorizations along ethnic lines), the socially generated role
differentiations, and status differentiations between categories. The social
norms and conventions, patterns of behavior and interaction, social
categories, and social structures can be separated analytically; however,
they stand in a dialectical relation towards another (cf. Bader, 1995: 94–5).
Their relations are mutually reconfirming: norms direct patterns of
behavior, while the existence of these behavioral patterns reconfirms the
norm. Cultural beliefs structure cultural practices, and cultural practices
embody cultural beliefs. The social structure is the result of behavioral
patterns, but the structure endorses and strengthens the behavioral patterns
and the underlying norms in society.
NEITHER ESSENTIALIST NOR REDUCTIONIST
Cultural difference should thus be interpreted as the result of cognitive and
social processes. This approach, focusing on processes of categorization, is
neither essentialist, nor reductionist. It is not essentialist because cultural
difference is seen as the result of sociality, social cooperation, and interaction.
Describing cultural difference in terms of social categories is not ex
post identifying phenomena that were already (essentially) there, for
example as a natural fact. Cultural diversity can only be acknowledged after
social categories have been ‘constructed,’ mentally and socially. For as
much as culture is reified, this reification process does not take place in the
phase of theory building; instead, this reification is a social process: cultural
PIERIK ● CONCEPTUALIZING CULTURAL GROUPS
difference is the result of social processes of categorization that are
reconfirmed and reified over time (cf. Brubaker, 2002: 166).
Moreover, it does not compartmentalize a multicultural society. A multicultural
society is not neatly sub-divided into a certain number of sharply
bordered and mutually exclusive groups; it is not a ‘multichrome mosaic of
monochrome cultural blocks’ (Brubaker, 2002: 164). Instead, Table 1 shows
that cultural difference is a whole range of crosscutting and overlapping
cleavages. Describing cultural difference in terms of historical processes of
social categorization makes clear that the basic structure is not a univocal
essence. Instead, different societies are structured differently, and the basic
structure can change over time. Different countries have different historical
backgrounds, and thus different dominant categorization processes. The
Canadian history with Inuit peoples scattered around the country and
Anglophone and Francophone immigrants is very different from the US
history dominated by slavery and racial conflict, the British, dominated by
colonization, or the European countries in general and their (former) guest
workers. As a result, the multicultural composition of different countries is
dissimilar and their dominant multicultural debates differ in character.
Finally, the basic structure can change over time. Noel Ignatiev (1995)
describes in his How the Irish Became White how the first wave of catholic
Irish immigrants in the USA around 1840 were first seen as a distinct race –
comparable to African Americans – but were, over time, ‘accepted’ as whites.
On the other hand, the fact that cultural groups and cultural difference
are not ‘essences’ but socially constructed regularities does not make them
superficial or perishable either. Categories always emerge for a reason.
Some reasons are personal and idiosyncratic – my marathon/Parfit category
– others are generally shared in society. Certain attributes, the related social
categories, and the resulting generalizations and categorical distinctions are
deemed to be important, and are treated as such. Julie Novkov concludes
about the development of the concept of ‘race’ that: ‘the choice of
characteristics was historically contingent (as thinkers discussing the
genesis of race have shown) but once the set characters has become
embedded in the Western mind, they became reified as race’ (Novkov, 2002:
6). In such cases, labels, even arbitrary ones, are based on social convention,
are imbued with deep meaning (Rothbart and Taylor, 1992: 23). In
taxonomy, it is generally acknowledged that social categorizations are an
expression of a deeper social or symbolic structure in society (Starr, 1992:
157). Thus, the assertion that cultural difference is socially constructed does
not imply the reductionist position, claiming that culture is contingent,
easily changeable, or merely a theoretical concept (Bader, 2001: 254–5). I
emphasized the historical character of this social construction: over time,
culture and cultural diversity is ‘materialized:’ (subconsciously) internalized
by individuals and embedded in interactional patterns, the institutional
order, and the basic structure of society (Jenkins, 2000; see also Jenkins,
ETHNICITIES 4(4)530
531
1997: 56). Social categories are interwoven in the fabric of society and state.
We do not ordinarily think about nor act upon the categories of social life;
instead, we act and think within them (Starr, 1992: 155).
However, since social processes are so imperceptible, their results – the
socially constructed categories – appear to us as if they were always there
as ‘brute facts:’ independent entities with their own meaning in social interaction.
As Paul Starr argues:
Social classifications take on a life of their own apart from the claims initially
advanced with them. They become diffused and standardized, even on an
international scale. This diffusion may obscure their origins and make them
appear to be objective, natural, and self-evident. (Starr, 1992: 176)
For individuals, using these categories does not need to be a conscious
act. Instead, categories are embedded in habit and institutions and using
categories is internalized and usually subconscious. Since they are internalized,
social categories appear to be independent ‘things,’ disconnected from
the social processes that generated them. For example: racial categories in
the USA were once developed to distinguish slaves from (free) persons.13
Although the official racial classification was abandoned some 140 years
ago, such categories are still manifest. Martin Gilens concludes that racial
generalizations are still very much present in contemporary American
society. ‘In particular, the centuries-old stereotype of blacks as lazy remains
credible for large numbers of White Americans.’ This stereotype grew out
of, and was used to defend, slavery and it has been perpetuated over the
years by the continuing economic disparities between black and white
Americans (Gilens, 1999: 3).
A final advantage of this conceptualization is that it enables us to bridge
the distinction between the socially situated self and unencumbered self
(Sandel, 1984) or to distinguish ‘structure’ form ‘agency’ (Bader, 2001:
260–4). For the individuals involved, the ‘constructed’ character of the
social structure must be taken as a given. Membership of a social group has
the character of ‘thrownness:’ one finds oneself a member of a group, whose
existence and relations one experiences as always having been (cf. Young,
1989: 260). At the same time, a person’s agency implies that he or she can
identify with one or more of these cultural groups. For example, a woman
could choose to neglect her female identity because her lesbian identity or
black identity is much more important to her. Moreover, the crosscutting
character of the social structure can result in conflicting identities, e.g. being
a Catholic feminist. At the same time, someone’s agency enables him or her
to make choices to resolve these internal conflicts. As such, this approach
provides an exit option from the classic dichotomy between the socially
situated self and unencumbered self. Regardless of the structure they live
in, persons always have agency to influence their destiny, but, at the same
time, it is limited. Someone cannot escape being socially categorized and
PIERIK ● CONCEPTUALIZING CULTURAL GROUPS
labeled as a woman, or as a person of color, simply because these categorical
distinctions are embedded in the social structure. Gender inequality
results from the clash of stereotypes and norms associated with masculinity
and femininity. The distinctions between cultural groups are ‘produced’ and
‘organized’ by a history distinguishing between social categories, based on
labeling and stereotyping. In this sense, cultural groups work as a pre-reflective
background. This explains the elusive character of these distinctions
within social structures.
SELF-CATEGORIZATION AND CATEGORIZATION
BY OTHERS
Up to now, I have conceptualized cultural groups and cultural difference in
terms of cognitive and social processes of categorization. In the following
sections I will distinguish two different types thereof: self-categorization
and categorization by others.14 The description in the following sections is
an ideal-typical and analytic sketch of these separate processes as basic
mechanisms (cf. ‘The ‘Social mechanism approach’ earlier). Such a separate
presentation is an analytical separation of the indivisible, useful for heuristic
purposes. In ‘From basic mechanisms to an integrated explanation’ later, I
will give some examples of how actual social categories can be described in
terms of (combinations of) these social processes.
Categorization by others: Ascription
Categorization by others is a form of categorization that is caused by
processes of ascription. Ascription is the attribution by outsiders of certain
characteristics, behavior, and beliefs to a social collective of individuals who
share a certain attribute. As a result of processes of ascription, individuals
who share a certain attribute are categorized in specific categories. Those
who wear glasses are seen as intellectual; blond, pretty girls are seen as
dumb. Italians are assumed to be romantic and passionate, while the Dutch
are assumed to be rational and standoffish, etc. The emergence of social
groups can also be understood in terms of ascription. The best example is
the situation of African Americans. They do not necessarily share specific
common cultural beliefs, norms, or values. Instead, African American
people display the same range of conceptions of the good life as the rest of
society. There are left-wing African Americans and right-wing African
Americans; religious African Americans and atheist African Americans;
gay African Americans and heterosexual African Americans. Still, the
category of African Americans has a different position in a social structure
– especially in the USA – which can be explained by social processes of
ETHNICITIES 4(4)532
533
ascription. Americans often have different stereotypical images of different
ethnic groups. Asian Americans are regarded as productive, hard-working,
and obedient. African Americans are regarded as good at sports and music,
but less suitable and reliable as employees (Gilens, 1999: Chapter 6). These
categorical differences are the result of social processes: generations of
implicit and explicit generalizations, stereotyping, etc. These influence
chances of individual Asian Americans and African Americans, regardless
of their individual skills and attitudes. Processes of ascription definitely
influence an individual’s situation: all African Americans face certain
barriers in their everyday lives that members of other social categories do
not face. They do not necessarily identify themselves with the attribute, the
category, or the expected norms or behavior. Nevertheless, generally
speaking, they are categorized into these social categories.
Self-categorization: Inscription and community-building
Self-categorization is a form of categorization in which certain beliefs,
characteristics, and behavior are passed on to new members of social collectives
by socialization and enculturation before the age of reason. Sometimes
this process occurs imperceptibly and implicitly. This process could
be called inscription. Inscription is the process in which social norms and
expectations are transferred to children through socialization and enculturation
in the family, peer groups, and school. For the person involved,
inscription implies the internalizing of the shared beliefs on distinctions
between social categories. Shared community-building refers explicitly to
self-categorization of social collectives around a shared sense of belonging.
Examples are certain religious or linguistic communities within society, e.g.
Muslims or the Québécois. Such cultural communities share a conception
of the good life or a native language that distinguishes them from others.
Newborn members are introduced into and socialized within such
communities. This shared community-building is strongly dependent on
inscription: this is the way in which the concept of a good life is passed on
to the newborn members. Members of religious minority groups associate
with each other precisely because they share cultural beliefs, which differ
from the larger society. Members of the larger society have no interest in
or desire to act upon these beliefs, since they do not share and sometimes
even do not accept them.
From basic mechanisms to an integrated explanation
I have described the two basic mechanisms of categorization in their
ideal-typical appearance: self-categorization (inscription and communitybuilding)
and categorization by others (ascription). Analytically these
processes can be neatly separated, and such an analytical division is helpful
PIERIK ● CONCEPTUALIZING CULTURAL GROUPS
for our understanding of the basic mechanisms in themselves. After all, only
by abstracting from the complexities of the real world can we devise a
conceptual schema that can illuminate it. However, actual cultural groups
will never be the exclusive result of one of these processes. Cultural groups
as discussed in contemporary multiculturalism (see Table 1) are social
categories, resulting from combinations of these social processes: ascription,
inscription, and community-building. For a descriptive analysis, it is
not relevant which of the processes has been dominant for a specific cultural
group. Sometimes community-building dominates: certain communities
emphasize certain practices as essential to the culture and its survival:
rituals, religious rites, religious prescriptions, initiation rituals, etc. Sometimes
the ascription component is more forceful: for centuries it was
thought that women did not have the intellectual capacities to satisfy the
demands of citizenship – they did not have the right to vote – or to an
intellectual job – they were not admitted to university.
At the same time we can see that the three processes interact. African
Americans in the USA, for example, were described as an example of the
process of ascription. But there are also inscriptive tendencies: that the
racial stereotypes also affect the self-image of blacks (Gilens, 1999). A
striking example is the emergence of GangstaRap, a term describing a
certain type of music and related scene, mainly performed by African
Americans. The distinguishing feature of GangstaRap is the use of images
of criminal behavior. Both the music and the performing of it play with the
images of African Americans as criminals, gang members, and ghettodwellers.
In this way, black rappers like Snoop-doggy-dog co-opt an image
or stereotype for their own purposes, taking control of the image, and
transforming it into a positive image and a source of pride. By satirizing and
exaggerating the negative images of African Americans, GangstaRap turns
an ascribed negative image into an inscribed token of pride. We can also
find more positive interpretations of the black self-imaging (Jenkins, 2000:
13), for example the clothing brand ‘FUBU’ (For Us By Us). This company
was set up by couple of young blacks, selling jeans made by young black
designers and marketing those items that are popular within black culture.
After finding it difficult to market their clothing through traditional
advertising channels, FUBU succeeded in promoting its products by using
black celebrities and hip hop artists. However, shared community-building
does not seem to be the main distinguishing characteristic for blacks as a
category.15 Although some Afrocentrists claim blacks to be a cultural
community, shared around the value of Kwanza or other aspects of the
African heritage, this seem to be a marginal movement. The social category
of Muslims is clearly associated with community-building, based on
processes of inscription. The shared ideas of the good life that distinguish
this cultural community form from others are passed from one generation
to the next through socialization. However, especially in the aftermath of
ETHNICITIES 4(4)534
535
the September 11 events, we can encounter clear ascriptive processes with
evident negative results for individual Muslims.
To conclude: so far I have used the social mechanism approach to explain
the genesis of cultural groups and cultural difference. I have given a qualitative
and analytical description of three processes of social categorization
as basic mechanisms: ascription, inscription, and community-building.
Analytically these processes can be neatly separated. However, actual
cultural groups are the result of interactions between ascription, inscription
and community-building; no actual cultural group in society is exclusively
caused by one of them alone. For a descriptive analysis it does not really
matter which of these processes is dominant. The distinction between them
therefore indicates a difference of emphasis, rather than of kind. However,
the distinction between self-categorization and categorization by others is
relevant for a normative analysis.
NORMATIVE DEBATES IN MULTICULTURALISM
We can use this description to clarify an ambiguity in contemporary normative
debates on cultural difference. Although the term ‘multiculturalism’
suggests a singularity in the debate, we can clearly distinguish two different
kinds of normative claims (Kymlicka, 2002: 330–1). Some social categories,
let’s call them social groups, mainly (not solely) prefer to integrate in
mainstream society, but feel wrongly excluded because of their sex, sexual
preference, skin color, etc. Other categories, let’s call these cultural
communities, differ significantly from mainstream society because of their
different conception of the good life (e.g. religious minorities such as
the Muslims in western societies) or the language they speak (e.g. the
Québécois in Canada). They mainly (not solely) prefer to maintain the
distinction between them and mainstream society to preserve their distinct
collective identity: religion and its practices, a distinct language, etc. To
oversimplify, we could say the former claims are mainly discussed in
debates in the USA (e.g. by Iris Marion Young and Nancy Fraser), whereas
the second category of claims dominate the Canadian debate (e.g. Will
Kymlicka and Charles Taylor).
Although both claims are put forward as multicultural claims, they seem
to contradict: social groups prefer to integrate in the wider society, while
cultural communities prefer to preserve their distinct collective identity.
This section discusses them separately and shows that the distinction
between categorization by others and self-categorization enables us to deal
with these apparently contradictory claims.16
PIERIK ● CONCEPTUALIZING CULTURAL GROUPS
Distinguishing social groups from cultural communities
Although the distinction is not strictly categorical, the two normative claims
can be associated with distinct social categories and distinct processes of
categorization. The claim of integration is mainly voiced by social groups,
which are mainly the result of processes of ascription. The claim of
preservation of distinction is mainly voiced by cultural communities for
which inscription and community-building are dominant.
Cultural communities are different from social groups in three respects:
identity, social structure, and interdependence. The identity criterion refers
to the collective awareness of the members of a cultural community as a
distinct social entity; they perceive and define themselves as a community
that shares a common identity. The criterion of social structure refers to the
stabilized and organized relations between members. Moreover, a system
of role and status differentiations exists and relations within such cultural
communities are regulated by shared social norms and values that prescribe
beliefs, attitudes and conduct in matters relevant to the group. The interdependent
criterion implies that members should be positively interdependent
in some way (Turner, 1987: 19–21). The transition of a social category
to a cultural community implies a ‘virtually organic moment’ that occurs
after the creation of a shared character ‘strong enough to define a meaningful
aspect of each individual member’s social identity’ (Mitnick, 2000: 191).
As such, culture ‘materializes’ and becomes incorporated in cultural
practices (Bader, 1995: 95; 2001: 257).
This implies that we have to distinguish ‘looser’ social groups from more
‘close knit’ cultural communities. Indeed, social groups like ‘women’ or
‘African Americans’ cannot be described as communities in this understanding:
women as a category do not have a shared social structure; not all
African Americans are positively interdependent upon one another.
Processes of ascription do not result in cultural communities, since the
persons involved do not share distinctive cultural beliefs or cultural
practices; their main shared interest is to fight hegemonic norms. As Bader
argues, ‘people which are exploited, dominated, excluded or marginalized
on the basis of ascribed criteria protest against these practices without
developing a common cultural identity or a new common culture’ (2001:
261). Of course, in society we find associations which manifest themselves
in the name of these cultural groups: women are organized in feminist
groups; Some Afrocentrists emphasize Kwanza as a shared identityproviding
aspect of the African heritage. However, it is important in our
conceptualization of cultural groups to distinguish social groups from these
self-organized interest groups (cf. Brubaker, 2002: 166–7). Linguistic and
religious communities, on the other hand satisfy the three conditions of
cultural communities.
Of course, this reduction of the variety of cultural groups into two
ETHNICITIES 4(4)536
537
categories might not do justice to the rich diversity of actual cultural differences.
But this is also not my primary aim in this section. Instead of focusing
on the social groups and cultural communities themselves, in the normative
debate I will emphasize the two relevant social processes generating them.17
How do processes of categorization by others and self-categorization affect
the situation of individuals?
A normative analysis of ascription
Processes of ascription result in social groups that would like to integrate
in mainstream society, but feel wrongly excluded because of their sex,
sexual preference, skin color, or other ‘ascribed characteristics that historically
served as markers of inferiority and exclusion’ (Young, 2001: 4). The
color of one’s skin or one’s gender does not in itself determine someone’s
life inescapably, in contrast to the inability to see, which does. What does
distinguish being black from being white is the ‘web of socially generated
expectations’ in society. Institutional patterns of cultural value and labels
of deviance constitute some social categories as ‘inferior, excluded, wholly
other or simply invisible’ (Fraser, 2002: 24). Specific attributes such as
gender, ethnicity, sexual preference, etc., generate social categories whose
membership is not chosen deliberately.
As individuals, persons experience the effects of processes of ascription
because they are not judged on their individual character and behavior, but
on their membership in a category. The problem with such stereotypes is
that they encourage overgeneralized expectations: the ascription of specific
qualities and characteristics only by virtue of his or her category membership.
Earlier, I discussed the categorical distinctions between Asian
Americans and African Americans. Let me elaborate here the example of
women on the labor market. The ascribed part of gender inequality on the
labor market can be explained in terms of statistical discrimination
(Robeyns, 2001: 93). Statistical discrimination of women occurs when
employers base their decisions on information about categories (here men
and women), instead of using individuated information about specific applicants.
If such an employer relies on categorical information, he or she will
assume that a specific candidate has the average characteristics (and hence
productivity) of the individuals in her category (women). When employers
believe that women are less productive than men – in our societies women
usually bear more responsibility for housekeeping and the care of children
and the elderly – it is rational for an employer to discriminate against an
individual woman. This is true even if this particular woman does not share
the characteristics with the other women, for example because she prioritizes
a career above children. Still, the characteristics of the average woman
are ascribed to this person. She is not treated as a person with individual
PIERIK ● CONCEPTUALIZING CULTURAL GROUPS
characteristics, plans and preferences. On the contrary, she is treated as the
average woman, locked into her category. These generalized expectations
affect all members of a category, regardless of their individual character or
preferences.
As a category, they feel wrongly excluded by mainstream culture and
argue that society ignores diversity, because it is exclusively or mainly based
on hegemonic norms, related to the financially independent heterosexual
white male. Members of these categories would prefer to integrate in the
larger society and inclusion in society’s mainstream institutions, but want
to renegotiate the basic rules. Gays and lesbians cannot participate as full
members in a societal culture that sees them as inferior, and treats them as
less worthy of concern or respect. Therefore they propose a range of
policies and rights that undermine hegemonic norms and attack the status
hierarchy resulting from these norms. For example, Janet RadcliffeRichards
argues for the gender division of labor that:
If a group is kept out of something for long enough, it is overwhelmingly likely
that activities of that sort will develop in a way unsuited to the excluded group.
We know for certain that women have been kept out of many kinds of work,
and this means that the work is quite likely to be unsuited to them. The most
obvious example of this is the incompatibility of most work with the bearing
and raising of children; I am firmly convinced that if women had been fully
involved in the running of society from the start they would have found a way
of arranging work and children to fit each other. Men have had no such
motivations, and we can see the results. (Radcliffe-Richards, 1980: 113–14, as
quoted in Kymlicka, 2002: 381)
Historic examples of such policies are the abolishment of the slavery
system and opening the vote for women. Actual discussions concern
affirmative action, opening up marriage for same-sex couples, or fighting
barriers that formally or practically hamper the combination of work and
parenthood.
A normative analysis of inscription and community-building
Processes of community-building result in cultural communities that differ
significantly from mainstream society on one issue or a related set of issues,
e.g. conception of the good life (religious minorities, e.g. Muslims in
western societies, ethno-religious groups such as the Amish), or native
language (Québécois in Canada). They mainly (not solely) prefer to
maintain that difference between them and mainstream society to maintain
their distinct religion and its practices, or the distinct language. Members
associate to celebrate their shared beliefs and practices. Since these beliefs
and practices are not shared by members of the larger society, cultural
communities seek protection of their distinct cultural heritage and cultural
practices via cultural rights.
ETHNICITIES 4(4)538
539
How should government deal with these preferences of minority
cultures? The default liberal answer is that government should be neutral
to different ideas about the good life, and should provide a neutral framework
within which different and potentially conflicting conceptions of the
good can be pursued (Kymlicka, 1989: 883; Hampton, 1996: 173). This
neutrality is usually interpreted as a doctrine of restraint: government
should take a ‘hands-off’ approach towards (collective expressions of)
these differential conceptions of the good life, for example religious and
linguistic groups. However, critiques argue that the basic structure of
society is not a neutral framework but, due to nation-building processes,
is ‘hegemonically biased.’ The fact that in the western world weekends do
not include Friday shows which days are considered to be the ‘normal’
days for attending religious gatherings. Muslims who have a full-time job
and want to visit the mosque for their customary Friday afternoon prayer
have to make special arrangements to do so. The fact that English is the
dominant language in Canada is understandable, but does affect Francophones
who want to preserve the French-speaking Québécois
community.
The fact that hegemonic subcultures had an important moulding effect
on the basic structure of any society cannot be changed retrospectively. But
it should affect the way we interpret neutrality. The idea of the basic
structure of society as a ‘neutral framework within which different and
potentially conflicting conceptions of the good can be pursued’ is overtly
optimistic, because it is oversimplified. Members of hegemonic subcultures
are in an advantaged situation because their practices are ingrained in the
basic structure during the nation building process. The Christian feast days
are firmly anchored in the Western calendar, unlike the Islamic. For government
to be neutral implies having an eye for cultural differences and being
aware of the ‘hegemonic bias’ in its basic structure. Joseph Carens
concludes that in these situations a hands-off approach is inadequate and
should be replaced by an approach of ‘comparable support:’
Now being fair does not mean that every cultural claim and identity will be
given equal weight but rather that each will be given appropriate weight under
the circumstances within the framework of a commitment to equal respect for
all. History matters, numbers matter, the relative importance of the claims to
the claimants matters, and so do many other considerations. (2000: 12–3)
Such an approach would emphasize evenhanded liberal policies, for
example exemptions from laws which penalize or burden minority cultural
practices, or assistance to cultural communities to do those things the
majority can do unassisted.
PIERIK ● CONCEPTUALIZING CULTURAL GROUPS
CONCLUSION
In this article I used the social mechanism approach to explain the genesis
of cultural groups and cultural difference. I have given a qualitative and
analytical description of three processes of social categorization as basic
mechanisms: ascription, inscription and community-building. I thus shifted
the emphasis from cultural groups and cultural difference to the social
processes that generate and maintain these phenomena. Moreover, I
distinguished two such processes – categorization by others and selfcategorization
– and showed that this separation can clarify an ambiguity
in dominant debates in contemporary multiculturalism.
Compared with other theories in the field, this approach has some
advantages. For one thing, as argued earlier, the approach is neither
essentialist nor reductionist. It conceives cultural groups as results of social
processes, instead of seeing them as reified ‘things.’ It shifts the focus from
cultural groups themselves to social practices generating cultural difference,
namely cognitive and social processes of categorization. Moreover,
this approach offers a single theoretical basis for discussing a broad variety
of sources of cultural diversity. Not only are the standard issues in multiculturalism
dealt with – linguistic and religious groups – but also cultural
difference related to, for example, ethnicity, gender, and, sexual preference.
Therefore, it offers a more subtle description of cultural diversity in
contemporary Western liberal democracies than, for example, Kymlicka’s
distinction between multinationality and polyethnicity.
Acknowledgements
For comments on earlier versions of the paper, I thank Gerard Delanty, Yolande
Jansen, Will Kymlicka, Ingrid Robeyns, Wibren van der Burg, and two anonymous
referees for this journal. I acknowledge the financial support of the Netherlands
Organization for Scientific Research.
Notes
1 Although the normative relevance is not accepted unanimously, see Barry
(2001). For critiques of Barry’s position, see Ethnicities 2(2) and the contributions
to Kelly (2002) and Pierik (2002), among others.
2 In the words of Iris Marion Young: ‘. . . two categories which are opposing and
mutually exclusive’ (1997: 50–1). On Kymlicka’s conception of societal culture,
see Carens (2000: 52–87).
3 For example, Seyla Benhabib argues that Kymlicka’s work is based on ‘a poor
man’s anthropology’ (1999: 53–4).
4 The anthropological work of Franz Boas in the early twentieth century is
generally seen as the turning point from a biological to cultural interpretation
of ‘race.’ See Novkov, 2002: 3.
ETHNICITIES 4(4)540
541
5 See Jenkins, 1997: 16–24, for a parallel shift in social anthropology from tribes
to ethnic groups.
6 Bader sees this radical constructivism as an abstract overreaction against essentialism
and cultural determinism (2001: 257–64).
7 For an overview of the social mechanism approach see Elster, 1989: 3–10; 1999:
1–47 and Hedström and Swedberg, 1999.
8 Many examples of social mechanisms have become standard elements in the
canon of social science. For example, the prisoner dilemma describes a social
mechanism that unintentionally generates suboptimal results as a result of
rational behavior of self-interested agents. The market mechanism describes
prices, demand and supply in terms of individual economic behavior and the
equilibrium between individual production, distribution and consumption. The
seminal example of a social mechanism is a self-fulfilling prophecy: a forecast
that might be true or untrue, but by the very fact that it is publicly stated, has
the effect of bringing about what it claims, thereby making it more ‘true’ than
it would have been without the public articulation. For example, if people think
that a bank will go bankrupt, they will massively withdraw their money, and
therefore a bank will indeed fail (Merton, 1968: 476–7).
9 Sometimes these basic mechanisms are mutually reinforcing, sometimes they
are counteracting.
10 The (mental) processes of categorizing and classification are studied within an
academic sub-field called taxonomy, bringing together biologists, linguists, and
psychologists. For introductory texts, see: Sokal, 1974, Douglas and Hull, 1992,
and Rosch, 1978.
11 ‘A basic task for all organisms (indeed, one mark of living things) is a segmentation
of the environment into classifications by means of which nonidentical
stimuli can be treated as equivalent’ (Rosch and Lloyd, 1978: 1).
12 Conditional upon one’s interpretation of the relevance of different social
categories.
13 Eric Mitnick gives an overview of categories of human beings that have been
deemed ‘non-persons’ (2000: 195). For a discussion of the origin of racial
classification, see Stuurman, 2000. For a historic overview of the legal constitution
of race in the United States, see Haney López, 1996.
14 A similar distinction between ‘self-identification’ and ‘categorization by others’
can be found in Jenkins (2000). Jenkins pays more attention to power issues
(1997: 52–73; 2000).
15 One could claim that a religious shared community-building was an important
cause for the emergence of the civil rights movement in the 1960s. However,
this has lost its importance nowadays.
16 For an account of the relation between liberal egalitarianism and multiculturalism
as two complementary interpretations of the ideal of equality, see
(Pierik, 2004).
17 Rogers Brubaker (2002) proposes a similar shift of emphasis from ‘groups’ to
‘groupness.’
PIERIK ● CONCEPTUALIZING CULTURAL GROUPS
References
Bader, Veit (1995) Rassismus, Ethnizität, Bürgerschaft: Soziologische und
Philosophische Überlegungen. Münster: Westfälisches Dampfboot.
Bader, Veit (2001) ‘Culture and Identity. Contesting Constructivism’. Ethnicities
1(2): 251–73.
Barry, Brian (2001) Culture and Equality: An Egalitarian Critique of Multiculturalism.
Cambridge, MA: Harvard University Press.
Benhabib, Seyla (1999) ‘“Nous” et “les Autres”: The politics of Complex Cultural
Dialogue in a Global Civilization’, in C. Joppke and S. Lukes (eds) Multicultural
Questions, pp. 44–62. Oxford: Oxford University Press.
Bidney, David (1953) ‘The Concept of Culture and Some Cultural Fallacies’, in
D. Bidney, Theoretical Anthropology, pp. 23–53. New York: Columbia
University Press.
Brubaker, Rogers (2002) ‘Ethnicity without Groups’, European Journal of
Sociology XLIII(2): 163–89.
Bruner, Jerome (1979[1962]) On Knowing: Essays for the Left Hand. Cambridge,
MA: Harvard University Press.
Carens, Joseph (2000) Culture, Citizenship and Community: A Contextual Exploration
of Justice as Evenhandedness. Oxford: Oxford University Press.
Douglas, Mary and David Hull, eds (1992) How Classification Works: Nelson
Goodman Among the Social Sciences. Edinburgh: Edinburgh University Press.
Elster, Jon (1989) Nuts and Bolts for the Social Sciences. Cambridge: Cambridge
University Press.
Elster, Jon (1999) Alchemies of the Mind: Rationality and the Emotions. Cambridge:
Cambridge University Press.
Fraser, Nancy (2002) Social Justice in the Age of Identity Politics: Redistribution,
Recognition, and Participation. New York.
Gilens, Martin (1999) Why Americans Hate Welfare: Race, Media, and the Politics
of Antipoverty Policy. Chicago: Chicago University Press.
Hampton, Jean (1996) Political Philosophy. Boulder, CO: Westview Press.
Haney López, Ian F (1996) White by Law: The Legal Constitution of Race. New
York: New York University Press.
Hedström, Peter and Richard Swedberg, eds (1999) Social Mechanisms: An
Analytical Approach to Social Theory. Cambridge: Cambridge University Press.
Ignatiev, Noel (1995) How the Irish Became White. New York: Routledge.
Jenkins, Richard (1997) Rethinking Ethnicity: Arguments and Explorations.
London: Sage.
Jenkins, Richard (2000) ‘Identity, Social Process and Epistemology’. Current
Sociology 48(4): 7–25.
Kelly, Paul, ed. (2002) Multiculturalism Reconsidered: Culture and Equality and its
Critics. Cambridge: Polity Press.
Kymlicka, Will (1989) ‘Liberal Individualism and Liberal Neutrality’, Ethics 99:
883–905.
Kymlicka, Will (1995) Multicultural Citizenship: A Liberal Theory of Minority
Rights. Oxford: Oxford University Press.
Kymlicka, Will (2002) Contemporary Political Philosophy: An Introduction (2nd
edition). Oxford: Oxford University Press.
ETHNICITIES 4(4)542
543
Lakoff, George (1987) Women, Fire, and Dangerous Things. Chicago: University of
Chicago Press.
Merton, Robert K. (1968) Social Theory and Social Structure. New York: Free Press.
Minow, Martha (1990) Making All the Difference. Ithaca, London: Cornell
University Press.
Mitnick, Eric (2000) ‘Constitutive Rights’, Oxford Journal of Legal Studies 20(2):
185–204.
Nelson, Thomas, Michele Acker, and Melvin Manis (1996) ‘Irrepressible Stereotypes’,
Journal of Experimental Social Psychology 32: 13–38.
Novkov, Julie (2002) ‘Deconstructing Whiteness? Miscegenation and the Institutionalization
of Race’, paper presented at the Annual Meeting of the Law and
Society Association, May 30 – June 2, Vancouver, Canada.
Pierik, Roland (2002) ‘Review of Brian Barry’s “Culture and Equality”’, Political
Theory 30(5): 752–60.
Pierik, Roland (2004) ‘The Ideal of Equality in Political Philosophy’, in W. van der
Burg and S. Taekema (eds) The Importance of Ideals: Debating their Relevance
in Law, Morality, and Politics, pp. 173–96. Brussels, New York: P.I.E.-Peter Lang.
Preston, Larry (1995) ‘Theorizing Difference: Voices from the Margins’, The
American Political Science Review 89(4): 941–53.
Radcliffe Richards, Janet (1980) The Skeptical Feminist: A Philosophical Enquiry.
London: Routledge and Kegan Paul.
Robeyns, Ingrid (2001) ‘Will a Basic Income do Justice to Women?’, Analyse und
Kritik 23(1): 88–105.
Rosch, Eleanor (1978) ‘Principles of Categorization’, in E. Rosch and B.B. Lloyd
(eds) Cognition and Categorization, pp. 28–48. Hillsdale, NJ: Lawrence Erlbaum.
Rosch, Eleanor, and Barbara B. Lloyd (1978) ‘Introduction’, in E. Rosch and B.B.
Lloyd (eds) Cognition and Categorization, pp. 1–3. Hillsdale, NJ: Lawrence
Erlbaum.
Rothbart, Myron, and Majorie Taylor (1992) ‘Category Labels and Social Reality:
Do We View Social Categories As Natural Kinds?’ in G. Semin and K. Fiedler
(eds) Language, Interaction, and Social Cognition, pp. 11–36. London: Sage
Sandel, Michael (1984) ‘The Unencumbered Self and the Procedural Republic’,
Political Theory 12: 81–96.
Sokal, R.R. (1974) ‘Classification: Purpose, Principles, Progress, Prospects’, Science
185: 1115–23.
Somers, Margaret, and Gloria Gibson (1994) ‘Reclaiming the Epistemological
Other: Narrative and the Social Construction of Identity’, in C. Calhoun (ed.)
Social Theory and the Politics of Identity, pp. 37–99. Cambridge, MA: Blackwell.
Starr, Paul (1992) ‘Social Categories and Claims in the Liberal State’, in edited
by M. Douglas and D. Hull (eds) How Classification Works: Nelson Goodman
among the Social Sciences, pp. 154–79. Edinburgh: Edinburgh University
Press.
Stuurman, Siep (2000) ‘François Bernier and the Invention of Racial Classification’,
History Workshop Journal 50: 1–21.
Turner, John C. (1987) Rediscovering the Social Group: A Self-Categorization
Theory. New York: Blackwell.
Young, Iris Marion (1989) ‘Polity and Group Difference: A Critique of the Ideal of
Universal Citizenship’, Ethics 99: 250–74.
PIERIK ● CONCEPTUALIZING CULTURAL GROUPS
Young, Iris Marion (1997) ‘A Multicultural Continuum: A Critique of Will
Kymlicka’s Ethnic-Nation Dichotomy’, Constellations 4(4): 48–53.
Young, Iris Marion (2001) ‘Equality of Whom? Social Groups and Judgments of
Injustice’, Journal of Political Philosophy 9(1): 1–18.
ETHNICITIES 4(4)544
ROLAND PIERIK is Assistant Professor of Political Theory at Tilburg
University, The Netherlands. Address: Tilburg University Law School, PO
Box 9O153, 5OOO LE Tilburg,The Netherlands. [email: r.pierik@uvt.nl]