Inquiry

ISSN: 0020-174X (Print) 1502-3923 (Online) Journal homepage: https://www.tandfonline.com/loi/sinq20

A Formal Recognition of Social Attachments:

Expanding Axel Honneth's Theory of Recognition

Bart van Leeuwen

To cite this article: Bart van Leeuwen (2007) A Formal Recognition of Social Attachments:
Expanding Axel Honneth's Theory of Recognition, Inquiry, 50:2, 180-205, DOI:
10.1080/00201740701239897

To link to this article: https://doi.org/10.1080/00201740701239897

Published online: 29 Mar 2007.

Submit your article to this journal

Article views: 575

Citing articles: 10 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=sinq20
Inquiry,
Vol. 50, No. 2, 180–205, April 2007

A Formal Recognition of Social

Attachments: Expanding Axel
Honneth’s Theory of Recognition

BART VAN LEEUWEN

University of Amsterdam, The Netherlands

(Received 18 November 2006)

ABSTRACT Axel Honneth draws a distinction between three types of recognition: (1)
love, (2) respect and (3) social esteem. In his The Struggle for Recognition, the
recognition of cultural particularity is situated in the third sphere. It will here be argued
that the logic of recognition of cultural identity also demands a non-evaluative
recognition, namely a respect for difference. Difference-respect is formal because it is a
recognition of the value of a particular culture not ‘‘for society’’ or ‘‘as such’’, but for
the social group involved. Yet, although it is formal, difference-respect cannot be
reduced to respect for personal autonomy and its preconditions, as Honneth wrongly
suggests in Redistribution or Recognition? It is argued here that difference-respect is
oriented towards another dimension of the person, namely social attachments. This kind
of respect entails a separate register of formal recognition with a corresponding concept
of personal identity and a parallel category of social disrespect. What morally justifies
difference-respect from a recognition-theoretic approach is the practical relation-to-self
that thus becomes possible, namely self-respect as a sense of belonging. The formal
conception of the good life that Honneth articulates should include the insight that this
sense of belonging is as much a necessary condition for the good life as is personal
autonomy.

Although the notion ‘‘recognition’’ is widely associated with theories and

practices of multiculturalism (or ‘‘politics of recognition’’), Axel Honneth’s
Kampf um Anerkennung has played only a marginal role in those debates.1
Whether one consults the most well-known philosophical database or the
indexes of major books and collections in the field of multiculturalism,

Correspondence Address: Bart van Leeuwen, University of Amsterdam, Department of

Philosophy, Nieuwe Doelenstraat 15, 1012 CP Amsterdam, The Netherlands. Email:
b.r.vanleeuwen@uva.nl
0020-174X Print/1502-3923 Online/07/020180–26 # 2007 Taylor & Francis
DOI: 10.1080/00201740701239897
 A Formal Recognition of Social Attachments 181

references to Honneth are scarce, indeed mostly absent.2 The fate of Charles
Taylor’s ‘‘The Politics of Recognition’’ stands in stark contrast to this: it is
one of the most widely cited articles in the field, notwithstanding the fact
that an actual theory of recognition is missing from it.3 One of the reasons
for this remarkable fact may be that, when writing his book, Honneth
actually had a different kind of social struggle in mind: his main question
was not so much how to understand and justify struggles for recognition by
cultural minorities in modern democratic states, but how to understand and
justify the struggle by lower social classes within capitalist society for social
esteem and equal respect.4
Only recently did Honneth pay particular attention to the so-called ‘‘new
social movements’’, namely in his discussion with Nancy Fraser in their joint
publication Redistribution or Recognition?5 For reasons I will explain later,
this intervention raises more questions than it answers. Moreover, it does
not, in the end, amount to a fundamental alteration of Honneth’s
categorization of recognition into three different spheres. I want to argue
that it would take a fundamental revision of his theory of recognition to
conceptualize a type of recognition that is able to accommodate the great
bulk of minority rights claims in modern pluralist societies. I will call this
type of recognition ‘‘difference-respect’’. This layer of respect is non-
evaluative in nature – and thus it is formal – because what is being
recognized is not so much the value of a particular culture as such, but
rather the value of it for the social group involved. This type of recognition
cannot be translated back into either of the recognition principles that
Honneth, in his current approach, applies to the social realm: it cannot be
reduced to the recognition of personal autonomy and its preconditions, nor
to the recognition of the value of particular achievements or abilities. To be
clear: I will not go into the – by all means important – question of whether
the redistribution of resources and wealth should be understood in terms of
recognition, as Honneth claims, or as a separate dimension of justice, as
Nancy Fraser argues. The thesis I will defend is that, in order to make
Honneth’s theory of recognition relevant to multiculturalism today, the
articulation of a fourth principle of recognition is urgently needed.

I. Honneth’s typology of recognition and his moral philosophy

The moral point of departure for Honneth’s theory of recognition is what he
refers to as a formal conception of the good life. What is elementary for the
good life in Honneth’s social philosophy is the possibility of individual self-
realization, that is, the ‘‘process of realizing, without coercion, one’s self-
chosen life-goals’’.6 The precondition for this is a positive self-relation,
namely a degree of trust in oneself and one’s abilities to set goals, to embark
upon particular life-plans and to pursue these successfully. What
characterizes the intersubjective character of Honneth’s theory is that this
182 Bart van Leeuwen

positive self-relation becomes possible only through patterns of recognition;

hence, these represent the intersubjective conditions that we must
presuppose in order to describe the general structures of the good life.
These two sets of strong relationships between ‘‘self-realization’’ and
‘‘personal identity’’7 and between ‘‘personal identity’’ and ‘‘recognition’’,
furnish the notion of ‘‘recognition’’ with a moral potential. For, if one
accepts these connections, social disrespect indirectly affects an individual’s
potential to exercise his own freedom in order to realize those values that he
deems to be constitutive of his own personality.
Honneth divides the concept of ‘‘recognition’’ into three categories: love,
which refers to an emotional concern for the well-being and needs of an
actual person; respect, which stands for the recognition of the equal moral
accountability of the legal person and is expressed in the moral and legal
right to personal autonomy; and esteem, which is the evaluation of
particular traits and abilities against the background of, mostly implicit,
standards of evaluation.
This tripartite division of types of recognition is phenomenologically
coupled with three different aspects of practical self-relation: self-confidence
refers to the personal ability to express one’s own needs and desires, as a
result of the awareness of being surrounded by a sphere of unconditional
love and emotional concern; self-respect is mediated by the sense of being a
bearer of equal rights, and thus being publicly recognized as a person with
an equal degree of moral accountability; and self-esteem is the positive self-
evaluation of one’s own particular capacities and achievements.
To understand the moral logic behind social conflicts, an understanding
of the concrete experiences of social disrespect is required. For that reason,
Honneth discusses the corresponding type of disrespect for each type of
recognition. These are, the violation of the body, the denial of rights and the
denigration of ways of life. These three types of degradation correspond to
three types of damaged identity, which together constitute a pathological
state of being that forms the basis for the formulation of moral obligations.8
People and institutions should avoid inflicting personal injuries which occur
as a consequence of indifference to the communicative conditions of
personal integrity. Moral duties follow from this intersubjective depen-
dency, although differently for each level.
What Honneth calls love – a category in which he includes both friendship
and parental love – is a type of recognition that is distinguished by moral
particularism, because this type of recognition cannot be extended to include
any number of people in a society. However, individuals should be legally
protected against the danger of physical abuse, a danger that is immanent to
the precarious balance of every emotional bond: the balance between a diffuse
longing for symbiosis and the necessary process of ego maintenance.
Respect refers to the principle of personal autonomy and its precondi-
tions. Autonomy in this context should be understood in terms of a right,
 A Formal Recognition of Social Attachments 183

that is, in terms of a socially accepted claim to self-determination which, in

both a moral and a legal sense, applies to all citizens. In the most minimal
sense, this principle refers to the right to shape one’s own life without being
hampered or obstructed by the state, organizations or other persons. To
respect someone as an autonomous person implies leaving certain decisions
to that individual himself and refraining from attempts to control these
decisions, for instance by means of force or manipulation. On a more
positive interpretation, respect for autonomy also includes rights that make
possible the exercise of autonomy.
Finally, Honneth links the notion of solidarity to the sphere of social
esteem. What is characteristic of solidarity is that people symmetrically
esteem each others’ contributions to a shared goal. Although this does not
imply a perfectly egalitarian appreciation for all members of society, it does
refer to a reciprocal evaluation in the light of a shared horizon of values that
makes it possible to experience each other’s uniqueness as meaningful for
shared praxis. Hence, this type of social esteem is heavily dependent on
collectively shared values and goals. To guarantee a certain amount of
solidarity, normative restrictions in this context should be oriented towards
the protection and formation of ‘‘community-generating value-horizons’’.9
This communitarian goal is qualified by two conditions: firstly, these value-
horizons should be sufficiently open and pluralistic to make it possible for
each member of society to be esteemed for his or her own particular
contributions; secondly, the principle of autonomy that structures legal
relations ought not to be violated.

II. Esteem as solidarity?

Honneth is right to point out that a minimal sense of national solidarity is
vital to a social world in which personal and collective differences do not
give rise to serious social instability.10 When a society deteriorates to a
situation of anomy – a situation in which citizens experience a high level of
social disintegration and limited opportunities for positive social identifica-
tion – patterns of reciprocal recognition are at risk.11 But although there is a
link between solidarity and reciprocal esteem, there are limits to the attempt
to interpret a favourable judgement of value in terms of ‘‘solidarity’’. Both
from the standpoint of the one who gives this judgement and from the
perspective of the recipient, there is a crucial difference between genuine
esteem and being loyal to someone or a social group.
First of all, esteem cannot be presented as some sort of moral duty, even if
this duty is limited to a particular community, as Honneth sometimes
suggests.12 It is not a moral duty because an evaluative judgement depends
on something other than a ‘‘good will’’. It depends on particular standards
of evaluation and it is directed not at the desire to be recognized, but at
certain achievements or capacities that are perceived as truly meaningful,
184 Bart van Leeuwen

impressive or admirable. What is thus essential for esteeming someone (or a

group) is that this recognition is not a response, moral or otherwise, to a
desire to be recognized.13
Secondly, from the standpoint of the recipient of esteem there is also a
crucial difference between esteem and solidarity or help. The individual who
desires a positive judgement of worth does not desire that someone
positively evaluates him because he desires it. As Charles Taylor has rightly
pointed out, to be the recipient of a positive evaluation that results from a
sympathetic willingness to express one’s solidarity, can be condescending or
even ethnocentric. Although this kind of appreciation for the ‘‘other’’ might
be related to the insight that recognition is vital to someone’s practical self-
relation, the paradoxical effect is that the other does not feel that he or she is
being taken seriously at all.14
However, although esteem is external to direct personal control, it would
be wrong to conclude that Honneth’s third sphere of recognition – that of
social esteem – is thus beyond the moral universe. The fact that a judgement
of value depends largely on particular cultural standards of evaluation
implies a certain moral claim. To evaluate other cultural expressions simply
against the background of one’s own cultural ideals and convictions, will
often mean constituting these as deficient in one way or another. It is for
that reason that Taylor applies Gadamer’s concept of openness to
intercultural understanding and interpretation, namely in terms of a fusion
of horizons. Recently, Honneth followed Taylor in that regard.15
Yet in the third sphere of recognition, we should be cautious in our
optimism. A general moral appeal for ‘‘openness’’ towards the ‘‘other’’ is too
demanding. From the perspective of our central theme – namely recognition –
what can be a moral demand is that, in the case of an evaluation of
particularity (cultural or otherwise), the final judgement ought to be preceded
by a moment of genuine openness in the sense just referred to.
This trimmed-down interpretation of the third sphere seems more
appropriate than Honneth’s moral demand ‘‘to view one another in light
of values that allow the abilities and traits of the other to appear significant
for shared praxis’’.16 Not only is this type of evaluation too instrumental in
nature,17 but if we apply Honneth’s scheme to complex, modern societies,
his symmetrical esteem is overburdened by its necessary condition, namely a
national community of values. What this analysis overlooks is that in
modern societies there is a high degree of cultural complexity, of local
subcultural variation, and that these local patterns of value can be quite
distinct, although not completely detached, from larger cultural flows.18
My claim is that this overburdened character of the third sphere of
recognition is a consequence of the fact that Honneth’s tripartite division
does not leave enough room for a formal recognition of difference, that is: a
recognition of national, linguistic, cultural, ethnic or religious attachments
that is non-evaluative in nature.19
 A Formal Recognition of Social Attachments 185

III. A formal recognition of difference or difference-respect

Within Honneth’s framework, to accommodate social attachments in the
formal sphere of recognition, namely the sphere of respect and legal
relations, is problematic. Yet, if one wants to interpret the moral grammar
of claims to minority rights (and not just esteem), one has to make room for
this. The only way to include the recognition of social attachments in this
sphere would be to understand these either as ‘‘objects of choice’’ or as
‘‘conditions for autonomy’’. Neither of these solutions will prove viable.
The first solution would be counter-intuitive. Social attachments cannot
be interpreted as the object of choice. People do not ‘‘choose’’ their culture,
ethnicity, nationality or linguistic identity. A social attachment is not an
‘‘expensive taste’’, as Brian Barry wants to have it, for which government
support or recognition is fundamentally unjustified because people are
supposed to be responsible for it themselves.20 There is a huge gap between
cultivating an expensive hobby and protecting one’s cultural institutions,
between personal preferences such as ‘‘a preference for vanilla over
strawberry ice cream’’21 and the collective struggle to be able to speak
one’s mother tongue in the social institutions of the society in which one
lives. A preference for ice cream might also not be chosen in a literal sense,
but it does not violate one’s interpretive generosity to say that it is fair that
people are not compensated for the choices they make on the basis of these
kinds of individual preferences. However, this is not necessarily the case
with social attachments. The key to the difference is that in the case of a
preference for ice cream, we are dealing with a particular interpretation of
well-being, while in the case of social attachments the formal conditions for
the possibility of human well-being are at stake. That is why social
attachments, although not chosen, are a morally relevant feature of the
human condition. We will come back to this point.22
The second, more subtle way to include the recognition of social
attachments in the formal recognition sphere of respect is to consider
cultural embeddedness as a precondition for the equal exercise of individual
autonomy. Will Kymlicka is one of the most prominent proponents of this
thesis.23 This approach makes it possible to grant cultural rights to
minorities within liberal democratic states for the reason that the members
of these minorities, in order to exercise their liberal right to autonomy, need
a familiar culture as a context of choice, just as the members of the majority
do. Although he is not explicit about it, Honneth does seem to opt for a
solution in this direction. In Redistribution or Recognition?, he states that the
overwhelming majority of claims for recognition by social minorities are
‘‘essentially determined by the recognition principle of legal equality’’.24
Within Honneth’s tripartite division, this suggests that cultural rights
safeguard the conditions for personal autonomy, just as social and political
rights do in his theory.
186 Bart van Leeuwen

As a consequence, there seems to be no need for a radical change of his

original design. The only place where Honneth considers a fourth
recognition principle is with regard to esteem: in addition to the recognition
of the value of particular accomplishments in the light of the overarching
value horizon of a society, Honneth considers the esteem a minority culture
might receive independently of previously institutionalized value references,
i.e. when it is esteemed ‘‘absolutely’’ or ‘‘for its own sake’’.25 I deliberately
say that he ‘‘considers’’ this, because it is unclear whether he actually
proposes such a modification of his theory.26
There are three interconnected problems with Honneth’s solution of
incorporating claims of cultural minorities within his theoretical framework.
Firstly, by accommodating culture within the sphere of the autonomy
principle, Honneth actually has to reduce cultural ties to conditions for
autonomy. The central weakness of this approach is that Honneth becomes
just as vulnerable as Will Kymlicka is to the critique of instrumentalism.
Central to this critique is the observation that the reduction of culture to an
instrument for autonomy does not do justice to the real concerns of the
minorities involved. Minority groups are caught up in a struggle for
recognition not because they want to secure a familiar medium by which
they are capable of clarifying their own options, but because they derive
meaning and a sense of belonging from a unique set of traditions or a
particular cultural group. What is more, instrumentalism can easily result in
political and social pressures on minority groups to assimilate into the
dominant culture. What is characteristic of an instrument is that it is, in
principle, just one of many possible ways to achieve a certain goal. If other
more effective or efficient means that do not conflict with the final goal
(autonomy) present themselves, they may be applied. That people are
attached to their own culture and that a primary socialization cannot easily
be erased, as Kymlicka realizes, are in that case simply practical objections
that might be overcome by certain techniques.27 Moreover, the whole idea
of ‘‘contexts of choice’’ is inappropriate for the justification of minority
rights for immigrants, for whom there is simply not enough ‘‘minority
culture’’ available to sustain that kind of function.28
Secondly, as soon as one rejects the idea that the third sphere of
recognition is determined by solidarity-based esteem and should be
understood as esteem for particularity in a more general manner, the
distinction between two types of esteem – one in terms of the contributions
to national goals, the other in terms of ‘‘absolute’’ cultural value – becomes
superfluous. Although both are different types of evaluation of social
particularity (‘‘achievement’’ and ‘‘cultural value’’), the basic moral logic
remains the same. The demand for hermeneutical openness is, of course,
especially appropriate in the case of intercultural esteem, but generally,
esteem should always be based on a genuine evaluation, which implies by
 A Formal Recognition of Social Attachments 187

definition receptivity and openness, even when it concerns achievements by

individual members within our own cultural group.
Finally, the main problem with Honneth’s theory, including its recent
modifications, is that there is no place for a recognition of cultural
particularity that is independent of the evaluation of this particularity,
which is not at the same time being reduced to the principle of autonomy.
What kind of recognition of difference is missing? The recognition of
cultural difference does not necessarily imply the recognition of the
importance of the culture for us (or as such). For we can recognize that
the culture is important for the cultural group itself. What is being
recognized here is not so much the value of the culture itself, in a general
sense, as the value of it for a specific social group, that is, for the
practitioners of these cultural practices themselves.29 This is a formal
recognition of difference, because it is a recognition of particularity without
an evaluative moment: it is a recognition of the attachments or deep
identifications of the members of a social group with particular cultural
traditions. I call this kind of recognition difference-respect. Because social
attachments are not the objects of choice, but should be understood as a
context of identity that structures our choices and aspirations, this type of
respect is categorically different from the respect for personal autonomy.
Therefore, with this type of respect, a separate register of formal recognition
has to be introduced into the framework of Honneth’s theory, with a
corresponding concept of personal identity and a parallel category of social
disrespect. The justification for this expansion of Honneth’s theory of
recognition should be of the same type as for the other modes of
recognition, which means that both the negative and a corresponding
positive formulation should be articulated and that, in both instances, the
relation ought to be explored with the formal conditions of the good life.
Before we can do that, however, it is necessary first to define the
dimension of the person towards which this new sphere of recognition is
oriented. All three spheres of recognition in Honneth’s original scheme are
characterized not only by a unique mode of recognition, but also by a
particular aspect of the person to which they are directed: love has needs
and emotions as its object, respect has moral accountability and personal
autonomy, and social esteem has particular traits and abilities. If we want to
introduce a new register of recognition into this framework, the first
question that comes to mind is: what is being recognized here?

IV. Social attachments

The most basic demarcation in this regard is the one between individual
differences and social differences.30 The collective dimension of personal
identity includes things like religion, language and ethnicity. As Anthony
Appiah points out, such a list is heterogeneous: not everybody shares in all
188 Bart van Leeuwen

these categories and not all these categories have the same significance for
those who do share in them. To the individual dimension of personal
identity belong such traits as intelligence, charm, humour and tactfulness.
So the individual dimension does not limit itself to natural, congenital
differences, such as intelligence, but can also refer to certain virtues, such as
tactfulness or charm.
The criterion that determines the transition to the collective dimension of
the identity is not so much that many individuals share certain features and
traits. Within certain local parameters, many people may have a good sense
of humour, or be charming or tactful. In such a case, we could refer to them
as a logical category. But a logical category is not yet a social group. Those
who have these traits in common will not perceive themselves to be part of a
social association, such as the group of the ‘‘humorists’’ or the ‘‘charming’’.
We can speak of a social attachment or social group only if a process of
identification takes place; in other words, only if a group of people share not
merely particular traits, norms or values, but also this sharing itself.31 Only
if the sharing is shared and a common world is constituted does it become
possible for strong social bonds to develop, along with patterns of reciprocal
recognition as being members of the group. In that process, identification by
others as belonging to the group is just as essential as self-identification,
namely the understanding of oneself as belonging to a particular social
group or tradition. This second moment in the description of social
attachments is crucial, because otherwise there is a real risk that a particular
‘‘belonging’’ is unilaterally enforced upon people. That would not be a
social attachment.32
But even though self-identification should be central to our definition of
social attachments, these attachments cannot be understood as the result of
an individual decision or choice, contrary to, for instance, one’s membership
of a tennis club or a political party. An individual cannot simply choose to
grow up in a particular culture, or to be of a specific nationality or ethnic
descent. What is characteristic of a social attachment is precisely that, to a
certain degree, it always escapes individual understanding and personal
control. This does not imply that a social attachment to a cultural or
religious group is unavoidable, a fate, for one is always capable of re-
evaluating a personal attachment to a social group, no matter how
important this belonging has been for one’s own identity.33 Yet a social
group cannot be understood as the product of some social contract, nor can
belonging to it be understood as an individual decision.34 As Avishai
Margalit formulates it: ‘‘Belonging is generally determined by criteria that
are not the result of choice. People do not decide to belong to an
encompassing group. They belong because of what they are’’.35
That people do not so much choose their own social attachments should
be taken as an indication of the far-reaching importance of such
attachments for human identity. Social attachments can give orientation
 A Formal Recognition of Social Attachments 189

and meaning to a human life only because they are not the consequence of
an autonomous decision, but rather serve as a guideline for choice. The fact
is that social attachments often structure the standards of evaluation that
shape the background against which we evaluate our own behaviour and
choices and those of others. We do not choose these standards themselves,
but we apply them precisely in order to assess our own choices and desires.36
Social attachments go hand in hand with particular cultures, with a
relatively coherent set of practices and social institutions in which the norms
and values of some social group are expressed. Norms and values should be
understood not as autonomous ‘‘things’’ nor as mental ‘‘representations’’,
but as being embodied both in the particular physical dispositions and in the
language, the practices and the institutions of a culture or subculture.
‘‘Embodiment’’ means that these norms and values exist in this changing
cultural expression itself, and that they are transformed according to these
changes.
Finally, belonging to a social group is not directly dependent on personal
accomplishments. It is not necessary to prove oneself or to excel in
something in order to be considered a full member of the social group. It
certainly is possible, as Margalit points out, to be a prominent member of
the group as a consequence of specific accomplishments, but membership of
that group is not itself an achievement: ‘‘To be a good Irishman […] is a
matter of achievement. Being Irish is solely a matter of belonging’’.37
We are now able to define the nature of the social difference to which
difference-respect relates. These kinds of definitions are precarious, because
they tend to either exclude or include too much. Thus, it is best to speak of
necessary rather than sufficient conditions for social attachments. In any
case, difference-respect is concerned with social bonds that are not so much
characterized by a coincidental communality as by the identification of the
group itself as a group. That process of identification consists both of self-
identification and of identification by others. Moreover, the basis of this
affiliation is neither an individual choice nor a personal achievement.
Generally, what is at stake is the sharing of shared values that are expressed
in the shape of cultural or subcultural practices and institutions. Hence, a
formal recognition of difference – or difference-respect – is directed at a
social attachment that comes about through the fact that people consider
themselves and each other as part of a group on the basis of a shared good,
such as a language or culture, without this belonging being the direct result
of individual choices or certain achievements. Types of social attachment
that can figure as examples in this context are national, linguistic, cultural,
ethnic and religious attachments.
These typical examples of social attachments show that they should not
be thought of as exclusive; normally, they are combined. Most individuals
belong to several social groups that they identify with more or less.
Generally, one prefers certain types of belonging over others, and so these
190 Bart van Leeuwen

social attachments should be understood in terms of a hierarchy of

identifications. This hierarchy can change during a lifetime and certain
attachments can be more relevant in particular contexts than in others. For
instance, in some contexts ‘‘ethnic identity’’ may have no relevance to
someone, while in other contexts his or her ethnic allegiance may become
suddenly quite salient.38 Some types of attachments (such as those to one’s
mother tongue) are almost inevitable, while others (such as religious
attachments) can be completely absent.

V. Negative justification of difference-respect

In what follows I will articulate the moral logic behind a formal recognition
of social attachments, or behind difference-respect. Difference-respect is
formal because it is concerned not with a recognition of the value of a
culture, religion or language as such or for the wider social community (be it
the nation or even the human race), but with the value of a particular culture
for the members of the social group concerned. This is a non-evaluative
recognition of social attachments. The formal character of this type of
recognition opens up the possibility for the expansion of the sphere of
respect – a sphere that in Honneth’s theory of recognition is solely oriented
towards personal autonomy and its preconditions. In that case, ‘‘respect’’
can be understood as the formal recognition of both personal autonomy and
social attachments. In the first case, we will speak of autonomy-respect (the
first axis of respect), and in the second of difference-respect (the second axis
of respect). Autonomy-respect makes possible a sense of equal moral
accountability and responsibility, while difference-respect constitutes a sense
of social attachment and belonging. In both cases (or spheres) we are dealing
with types of self-respect.39
Both the negative and the positive justifications for difference-respect are
formal, but while the negative justification focuses on the damage to identity
that a denial of difference-respect brings about, the positive formulation
tries to articulate the contribution of difference-respect to the formation of a
healthy and intact sense of self. Because the moral importance of the
recognition of social attachments is revealed most clearly by the repercus-
sions of disrespect on the practical relation-to-self, we will start with the
negative justification for difference-respect (and we will deal with the
positive approach in the following section). A norm becomes visible only if
it is being violated.
The social or political denial of one’s social attachments, for instance by a
politics of forced assimilation, has an objective dimension as well as an
intersubjective dimension. The objective dimension of a lack of respect – be
it difference-respect or equality-respect – concerns the negative practical
consequences thereof, quite independently from the adverse effects on one’s
personal identity. The intersubjective dimension of such disrespect,
 A Formal Recognition of Social Attachments 191

Figure 1. The modified structure of relations of recognition.

however, concerns the impact on a person’s practical relation-to-self. For

instance, in the case of autonomy denial, the objective dimension can be that
certain citizens are unable to vote and thus influence the process of the
political definition of the common good, while in the case of forced cultural
assimilation the objective conditions under which the reproduction of a
particular culture is possible are undermined. These denials are morally
significant in and of themselves. However, what becomes especially relevant
from the moral perspective of the recognition-theoretic approach, is that
these types of disrespect interfere with the way in which a person
understands himself or herself and his or her relation to the world. The
thesis is that this denial of difference-respect may affect the practical
relation-to-self where the identification with a particular tradition, a
language or a cultural group is concerned. If respect can refer to the social
attachments of others – and there is a strong relation between recognition
and self-understanding – then a lack of respect for those ties implies the
possibility of a loss of self-respect, in other words, a decreasing ability to see
oneself as part of a particular cultural and social world.
Cultures can be interpreted as different modes of being human. What this
means is that institutions or citizens who deny a person’s attachments to a
particular culture will not only bring about frustration, but also constitute
a type of dehumanization. When we ridicule a person’s attachments to a
particular social group or tradition, it not only becomes more difficult for
this person to attach value to it himself or herself, with an erosion of self-
esteem as a consequence, but it can also set in motion the process of a
gradual loss of self-respect. Structural offence has the power to undermine a
sense of social belonging and thus to reduce someone to a ‘‘punctual’’ self,
to an ahistorical and asocial subject. Or if, by a politics of assimilation, no
room is offered for cultural expression, then what is being undermined
(intentionally or not) is someone’s self-understanding as being interrelated
192 Bart van Leeuwen

with a wider cultural and social association. The sense of belonging to an

individual-transcending social framework gives way here to a sense of
estrangement and social isolation.
Forced cultural assimilation, however, has recently made a comeback as a
putative solution to several of the problems of multicultural society.40 For
instance, Nathan Glazer argues that assimilation offers a solution to the
deplorable social-economic position of many black Americans in the United
States. Glazer insists that the historical project of assimilation has ‘‘failed’’,
however not in the sense of being a total fiasco, but in the sense of being
uncompleted.41 Assimilation is supposed to prevent social and economic
exclusion and marginalization. Arguments in its favour are often
accompanied by the complaint that multicultural policies are eroding the
welfare state.42 However, this thesis is contested to say the least.43 In any
case, what should be clear is that forced assimilation logically and
historically creates its own dissidents, abnormalities and outsiders, and
that this can have a detrimental effect on integration both in a cultural and in
a social-economic sense. The politics of assimilation is an instrument of
cultural nationalism to remove ethnocultural difference. It means that
cultural and linguistic minorities unilaterally and completely (that means in
all respects) adapt to the wider society in which they live, without the society
itself being transformed in any way during that process.44 This type of
politics creates its own exclusions, because it logically leads to a shrinking of
the socio-political bandwidth for acceptable ethnic and cultural variation.
For that reason, a fanatical politics of assimilation generally generates
feelings of aversion towards cultural otherness, resulting in intolerance and
disrespect.
The moral point is that the self-respect of minorities in that scenario will
be threatened in two ways. First of all, the social attachments that minorities
have developed with the wider society will be cut off. As a consequence of
the rejection of their cultural, ethnic, religious or linguistic particularity,
minority groups’ understanding that they, too, are part of the national
community will be undermined.
Secondly, forced assimilation infringes on self-respect in a different way.
Not only is the identification with the socio-political centre being cut off; the
social ties with one’s own culture and traditions are under pressure too.
Cultural, ethnic, religious or national ties exist in and through their
expression. The mandatory abandonment of cultural or religious practices
in order to enable complete absorption into a dominant culture, makes it
hard – although not impossible –45 to maintain a sense of connectedness
with a particular tradition that deviates from it. Often, such a politics of
assimilation is a particularism that parades as the universal ‘‘rights of man’’.
It tells the members of a minority that they ought to delegate their identity
to the private sphere and start thinking of themselves as ‘‘individuals’’. As a
consequence, those whose identity is already secure affirm their false sense
 A Formal Recognition of Social Attachments 193

of superiority, while those whose identity is precarious will feel that they
should choose between sentimental group-think and growing up. Hence,
self-respect as a sense of belonging is being threatened by a politics of forced
assimilation in a double sense: both a sense of national belonging and a
sense local belonging are at stake.
The fact that difference-respect is formal does not imply a blind recognition
of all group identifications. For a social group to qualify for difference-
respect, at least two conditions ought to be met. In the first place, the
particularity for which respect is being claimed should be of a certain kind,
namely what we have described before as ‘‘social attachments’’ (see 1 IV).
Secondly, difference-respect presupposes in this model that the collective
practices and institutions that a group claims to be attached to ought to be
weighed against the first axis of respect. In that case, the question is not
whether this group difference really is a social attachment, but whether
within this group a basic respect for personal autonomy and bodily integrity
is being granted. The point of this procedure is not to give the principle of
personal autonomy an unflinching priority over all other values, including
those values associated with social attachments, but to determine in each
particular case whether certain group-specific claims for recognition can be
granted without violating basic rights and needs of the members of a
minority group.
Cultural practices of minorities that are seeking recognition may seriously
harm minorities within minorities – such as women, children, gay men and
lesbians, religious dissenters and linguistic minorities within minorities – by
suppressing and subordinating these internal groups and by denying them
basic human rights.46 However, if we aim at maximal moral and legal
constraints on recognition of group identity, we run the risk of – under the
guise of universalism – imposing a thick, perfectionist liberal morality upon
everybody, thereby making impossible any meaningful pluralism either
nationally or globally. That is why it is wise in this context to focus on what
Veit Bader calls ‘‘threshold or satisfying theories instead of maximizing
theories’’.47 He defends minimalist moral and legal constraints on a politics
of difference.
Bader describes three legal areas within which certain group practices
sometimes clash with the basic rights of vulnerable minorities. These three
areas involve conflicts between the group nomos (customs, group laws) and:
(1) the core of modern criminal law; (2) the core of modern marriage and
divorce law; (3) minimal requirements of non-discrimination and equal
opportunity. Compared to the first area of conflict, the space for legitimate
group autonomy in the other two areas is somewhat broader: in some cases,
inequality (area 2) or even exclusion (area 3) should be legally tolerated.48 The
most important area of conflict for any threshold conception of personal
autonomy is the first one. Claims for group autonomy that conflict with the
core of modern criminal law ought to be rejected outright. As Bader puts it:
194 Bart van Leeuwen

Practices of slavery, caste, bondage, or of an unequal civil and political

status for ascriptive minorities, are incompatible with the most
minimalist interpretation of modern freedom and equality. Practices
like sati, domestic violence, stranger-rape, marital rape, sexual abuse,
genital mutilation, honor killing and forced collective suicide are
surely incompatible with the most minimalist interpretations of basic
rights to life and bodily integrity. Group autonomy should not be
allowed to legitimize these practices. Public opinion and liberal-
democratic polities must try to convince minorities to change these
practices, and jurisdictions must prosecute and punish perpetrators.49

Of course, this is just a rough outline of the minimal threshold. It leaves

unanswered a number of difficult questions, for instance, the question
whether all forms of female circumcision should be understood and
prosecuted as genital mutilation.50 The general point is to exclude practices
from being ‘‘recognized’’ when they conflict with fundamental human needs
and rights; in other words, when they are irreconcilable with a threshold
conception of respect for human life and bodily integrity.
This minimalist autonomy-test51 does not rule out meaningful pluralism,
because what remains beyond this formal test is an infinite variation
regarding the way individual life-projects are being shaped, the way in which
social groups give content to their social attachments and the way in which
these two axes are balanced. To be sure, the contents of particular
attachments are susceptible to a fuller evaluation – that is, when questions of
cultural merit are at stake – but this ought to take place within the moral
framework of the third sphere of recognition, namely on the basis of the
concept of ‘‘hermeneutical openness’’.

VI. A positive justification of difference-respect

Although one could argue that, seen from a practical perspective, the
negative justification for difference-respect has priority over positive
formulations (which have a tendency to sound utopian at best, and
dangerously sentimental at worst), conceptually speaking it is impossible to
determine what humiliation is independently from an idea of the moral
good.52 Therefore, difference-respect in the shape of minority-rights or
special cultural facilities cannot merely be understood as a moral appeal to
‘‘avoid humiliation’’. Moreover, the negative justification suggests a passive
political approach to minorities, while a politics of difference often implies
more than simply abolishing offensive or humiliating expressions. Offering
certain public services in the minority language is one example.53 Hence, a
more affirmative commitment is at stake here, one that refers to a feature of
the human condition with a positive moral significance: social attachments.
 A Formal Recognition of Social Attachments 195

There are different ways in which a positive justification for difference-

respect could be formulated. In this regard, one should not give in to the
schematic choice between ‘‘liberalism’’, in which the justification is
grounded in cultural conditions for autonomy, versus a certain version of
‘‘communitarianism’’, in which this justification is grounded in the intrinsic
value of cultures.54 The justification of difference-respect that I defend here
differs from both positions. This is not because I deny that cultural
embeddedness makes individual freedom possible or that cultures might
have an intrinsic value, but because in both cases what is being ignored is the
actual motivation of members of social groups to protect and experience
their culture. In the first case this is because culture for the groups in
question is not simply a medium for choices, and in the second case it is
because a credible moral justification of difference-respect is not about the
value of a particular culture independent of the value this culture has for the
members of the cultural group itself. Whereas the first approach may lead to
a social or political justification of pressures to assimilate, the second
approach actually constitutes a way of ascribing rights to languages or
cultures against their own members. In that second case, the argument
underwrites not a right to culture, but an obligation to maintain it.55 Hence,
neither the reduction of social attachments to the transcendental conditions
for the possibility of autonomy nor intrinsic cultural value as such offers a
convincing moral ground for minority rights.
What morally justifies difference-respect from a recognition-theoretic
approach is the practical relation-to-self that thus becomes possible, namely
self-respect as a sense of being accepted and being part of a particular social
world. Difference-respect makes citizens positively at ease with their
particular identities. Such an implicit sense of inclusion is so elemental to
a person’s well-being that it should be available not only to the majority of
citizens, but to all citizens.
This fundamental sense of belonging is different in kind from the practical
self-relation that is being established by esteem, because esteem is dependent
on individual achievements and capacities. Hence, self-esteem refers to the
positive self-evaluation of personal abilities and achievements. Self-respect as
a sense of belonging, however, refers to what we have defined as social
attachments. The corresponding identity component is not so much a positive
self-evaluation, as a sense of being accepted by a particular social group.
How does it feel to be accepted somewhere? It is characteristic of
acceptance that one is able to move through social space in a generally
relaxed manner, without the feeling of unease that accompanies the sense
that one is being rejected.56 This sense of belonging or social acceptance
implies the safety of a quasi-unconditional recognition; namely the
recognition that one is part of a wider social group. This type of recognition
is quasi-unconditional for a member of the group, because a social
attachment is not dependent on personal merit or accomplishments.57
196 Bart van Leeuwen

Therefore, the sense of belonging is not threatened by the unavoidable risk

that an evaluation incorporates: the risk of being turned down, the risk of a
negative evaluation. This type of self-respect offers a more stable kind of
belonging than a solidarity that is based on individual achievements. Social
attachments are constituted by a kind of recognition that is not lost
overnight because one is suddenly unable, for some reason, to deliver certain
esteemed goods that ‘‘contribute to the realization of societal goals’’.58
How should we understand the formal nature of difference-respect? One can
make a distinction between two levels of government that are important for a
politics of recognition, namely the level of the state from which difference-
respect takes shape and the political subunit that requests recognition for its
identity. (Incidentally, sometimes these two levels are of a lower order, for
instance local government vis-à-vis internal ethnic communities.) The second
level of government is far from "neutral" with regards to the social attachments
that are at stake. After all, on this level recognition is being requested on the
basis of deep identifications with certain shared values – such as a particular
language and culture – that are not, or are only partially, in accordance with
the language and culture that are publicly supported.
But neither is the higher level of government value-free or culturally
neutral. Liberal politics cannot be reduced to practices of cooperation and
the political mediation of cultural differences. Every state is also culturally
and linguistically coloured.59 However, the fact that a state is not culturally
neutral does not imply that difference-respect is dependent on an evaluation
of the difference in question. After all, this type of recognition recognizes the
attachment to the difference, not the difference itself. In that sense, this type
of recognition is neutral. It is a version of the neutrality of justification.60
This does not entail that all the policies of a government should be based on
transcultural criteria, but only that the recognition of minority practices and
institutions should be unrelated to the assessment of the relative value of the
content of their culture. Neither are these measures in need of an
instrumental justification. A convincing justification of minority rights
refers to the intrinsic value of that particular culture for the members
themselves and not the intrinsic value of a culture ‘‘as such’’ or ‘‘absolutely’’.
This content-independent recognition of social attachments can be
reconciled with the norm of impartiality for the reason that it is not related
to a particular value hierarchy. So it does not involve favouring one
minority group over another. For this reason, a ‘‘politics of recognition’’ is
fully compatible with the principles of universal justice.61
To be clear, not all claims to recognition in multicultural or multinational
states can be reduced to difference-respect. Some struggles for recognition
are really about the re-evaluation of minority identities.62 However, the
politics of recognition cannot be interpreted as a mere demand for equal
esteem. Take, for instance, the situation in Canada. Quebec demands a
certain degree of self-government in order to safeguard its linguistic
 A Formal Recognition of Social Attachments 197

character. The recognition by the central Canadian government of this claim

would certainly not depend on the ‘‘appreciation’’ by government officials
of the relative value of this group difference itself (in this case, the French
language). If the federal government grants this right to linguistic
autonomy, what is recognized is the fact that the language is of importance
to the members of the social group itself, rather than that the preservation of
this language on Canadian soil is seen as valuable or important by the social
and political centre. This is the moral logic behind language rights for
minorities. That does not mean to say that it is logically excluded that,
besides this formal recognition of difference, a positive value is also
attributed to the institutionalization of this group identity. However, at the
same time it is perfectly conceivable that neither Canadian politicians nor
the Canadian population at large are very impressed by French language or
culture, but at the same time believe that there is a compelling moral
justification to recognize it. And that type of justification is based neither on
a positive evaluation – however much Quebec might welcome that – nor
simply on the maxim ‘‘do not humiliate your fellow men’’, but on the fact
that the French language matters to the Québecois themselves. And the
same applies to the way in which the recognition of Flemish is justified by
the Belgian government, or the recognition of the French and Italian
speaking minorities in Switzerland, or the recognition of Spanish by the
United States in Puerto Rico, to name just a few examples.
If a polity in this way takes the socio-cultural composition of a society into
account, this implies not only the practical possibility for minority groups to
express their own identity, even if it transgresses the private sphere, but also
that every citizen is able to understand himself as socially connected. This type
of recognition offers the intersubjective conditions for a feeling of security
through a sense of belonging that is independent of personal achievements or
accomplishments. Rather than contributing to the sense of being a person
with an equal moral responsibility (the first axis of respect), this opens up the
possibility of self-understanding as being part of a recognized social group
within the state, whether it concerns an immigrant community or a national
minority. In the last case (the case of multinational federations), the type of
belonging to the state is often very different from the first, namely, it is
mediated in a more profound way by the minority identity itself.63
As a final point I want to address a possible objection to this model. Is a
minority member’s sense of belonging not already sufficiently safeguarded
by patterns of recognition within the minority group? Hence, my belonging
to group X, so the argument goes, could not really be constitutively
threatened by the refusal of some person or group, situated outside X, to
recognize that I belong to it and that it matters to me. I want to argue that
this line of critique can be challenged on two different levels.
In the first place, it seems to presuppose what Richard Johnson has called
a ‘‘continental plates theory of culture’’, by which he means that cultural
198 Bart van Leeuwen

diversity is understood in terms of ‘‘whole ways of life’’ that are ‘‘humped

about by the same sort of people’’ and that only occasionally bump up
against other ‘‘great slabs of meaning humped about by others’’.64 However,
the fact that modern societies are characterized by a high level of cultural
complexity means that many different social affiliations exist in a dynamic
‘‘multitude of crosscutting social and cultural cleavages’’.65 Hence, instead
of the reassuring image of internally happy groups moving en bloc through
society, most modern citizens are constantly exposed to different value
systems and cultural affiliations, both in interpersonal contacts and via
modern information and communication technology.66 It is here that
intercultural tensions and thus difference-respect becomes very relevant.
Secondly, and more importantly, recognition in the context of a politics of
recognition should not be understood as a purely symbolic or verbal
affirmation of identity, but as being embodied in a concrete institutional
form. There is more at stake for minorities than friendly gestures by
members of the majority. In the case of national minorities this is obvious,
for their demands for recognition concern things like self-government of
education, language and culture, or even of naturalization, immigration,
healthcare and the police. But also, in the case of immigrant multi-
culturalism, claims for recognition have a public and often institutional
dimension, for instance encouraging a more inclusive curriculum in schools,
promoting equal group representation in government bodies, revising work
schedules or dress codes so as to accommodate the religious holidays and
practices of immigrant groups, launching cultural development initiatives
such as the funding of ethnic-cultural organizations or ethnic studies
programmes, providing certain services to immigrants in their mother
tongue in order for them to be able to have access to vital public services
(such as healthcare), or granting double nationality. All these objective
measures can be interpreted as an embodiment of difference-respect, in the
sense that they involve the public recognition of the importance of social
attachments for the minorities in question and thereby contribute to their
sense of belonging, both to the minority culture and to the nation. The
alternative is minority marginalization with all its repercussions for the
integrity and the stability of society at large.

VII. Conclusions
Difference-respect is formal not only because it implies no substantial
evaluation of particular social attachments, but also because it does not
prescribe how citizens should feel. The internal life of the beneficiary of
recognition is relevant only to the moral justification of a politics of
recognition. Difference-respect is justified by what it makes possible for the
self-understanding of citizens. It does not follow that the state obliges people
to develop self-respect because the state recognizes them. And just as there is
 A Formal Recognition of Social Attachments 199

no duty to develop self-respect, there cannot be a right to it either. There can

be no right to a positive self-understanding any more than there can be a
right to health. However, just as there is a right to basic healthcare, it is
sound to argue for a right to be recognized as a person with an equal moral
accountability and with particular social attachments, be they of a national,
linguistic, ethnic or religious kind.67
The point I have tried to make in this article is that the formal conception of
the good life that Honneth articulates should include the insight that
necessary conditions for the good life include not only autonomy, but also the
sense of being socially accepted by the communities in which one lives. The
sense of belonging, however implicit and natural a feeling it may be to many
of us (especially privileged groups), is experienced by most people as crucial to
their well-being. This type of self-respect is not the consequence of a choice, an
autonomous act, nor can it be reduced to a condition for individual freedom.
It is precisely the fact that social attachments always seem to escape us to
some degree that makes it possible that these constitute meaning and
orientation to our lives. This does not imply that social attachments nail us
down to the territory of some tribe. The point is to recognize both dimensions
of the person – freedom and attachments – without reducing one to the other.
Where exactly the optimal juncture of these axes lies in concrete political
and social contexts has not been examined here in detail. (The question
remains whether this is even a task for philosophy.) It is clear that political
mediation should play an important role in that regard.68 The goal of this
article, however, has been to articulate the moral logic of a recognition of
social attachments within the formal sphere of respect, a sphere that
Honneth exclusively allocates to respect for equal personal autonomy and
its preconditions. In this regard, Honneth can be situated in an intellectual
tradition that starts with Kant’s practical philosophy and his aversion to
heteronomy in the moral realm. Even modern liberal thinkers with a well-
developed sense of cultural dependency, such as Will Kymlicka, reduce
respect for alterity to respect for autonomy. It is this type of moral monism
that has been questioned here. As Charles Taylor puts it: ‘‘Why this rage for
reduction? Why do people think in the first place that it might be plausible
that all our ethical commitments might be derived from a single principle?
Alternatively put, what seems so strange about moral diversity?’’69 70

Notes
1. Axel Honneth (1992) Kampf um Anerkennung (Frankfurt am Main: Suhrkamp)
Translated as: Axel Honneth The Struggle for Recognition (Trans.) J. Anderson
(Cambridge: Polity, 1995).
2. The Philosopher’s Index database (1940–2006/06) refers to a total of only 4 publications
(minus those of Honneth) with the following combinations of terms: ‘‘Honneth’’ and
‘‘multiculturalism’’; ‘‘Honneth’’ and ‘‘politics of identity’’; and ‘‘Honneth’’ and ‘‘politics
of recognition’’.
200 Bart van Leeuwen

3. Charles Taylor (1994) ‘‘The Politics of Recognition’’ in Amy Gutmann (Ed.)

Multiculturalism (Princeton: Princeton University Press) pp. 25–73. The Philosopher’s
Index gives 35 hits just with the combination of the terms ‘‘multiculturalism’’ and
‘‘Taylor’’ (excluding namesakes).
4. Axel Honneth ‘‘Ethiek van erkenning’’ interview by P. van den Berg and G. Oenen Krisis
Vol. 77 (2000) pp. 24–36.
5. Nancy Fraser and Axel Honneth (2003) Redistribution or Recognition? (London: Verso).
6. Honneth The Struggle for Recognition p. 174.
7. The notion ‘‘personal identity’’ denotes a person’s practical relation-to-self, that is, the
way in which a person evaluates and understands him- or herself in everyday thinking
and acting. (Ibid., pp. 133, 163)
8. Axel Honneth (1997) ‘‘Anerkennung und moralische Verpflichtung’’ Zeitschrift für
Philosophische Forschung Vol 51. pp. 25–41.
9. Honneth The Struggle for Recognition p. 177.
10. Axel Honneth (1994) Desintegration (Frankfurt am Main: Fischer Taschenbuch).
11. Jaak Billiet and Hans De Witte (1995) ‘‘Attitudinal dispositions to vote for a ‘new’
extreme right-wing party’’ European Journal of Political Research Vol. 27 pp. 181–202.
12. Honneth ‘‘Anerkennung und moralische Verpflichtung’’ pp. 37–39.
13. Arnold Burms (1990) ‘‘Helping and Appreciating’’ in Sander Griffioen (Ed.) What Right
does Ethics have? (Amsterdam: V.U. Press) pp. 67–77.
14. In that sense it is ironic that Honneth refers to Taylor’s ‘‘The Politics of Recognition’’ as
an illustration of a communitarianism that ‘‘die Anerkennungskategorie heute benutzt,
um solche Formen einer Wertschätzung von fremden Lebensweisen zu charakterisieren,
wie sie sich typischerweise im Horizont gesellschaftlicher Solidarität herausbilden’’.
Honneth ‘‘Anerkennung und moralische Verpflichtung’’ p. 26, italics mine. Of all
people, it is Taylor who fiercely criticized the idea that ‘‘an appreciation of unfamiliar
ways of life’’ ought to be understood as a type of ‘‘solidarity’’. After all, ‘‘then […] the
act of declaring another culture’s creations to be of worth and the act of declaring
oneself on their side, even if their creations aren’t all that impressive, become
indistinguishable. The difference is only in the packaging. Yet the first is normally
understood as a genuine expression of respect, the second often as unsufferable
patronizing […] A favorable judgment on demand is nonsense […] Then the question is
no more one of respect, but of taking sides, of solidarity’’. Taylor ‘‘The Politics of
Recognition’’ pp. 69–70, italics mine.
15. Fraser and Honneth Redistribution or Recognition? pp. 168–169.
16. Honneth The Struggle for Recognition p. 129.
17. For a critique of Honneth’s unjustified restriction of esteem to the ‘‘usefulness for
society’’ in his The Struggle for Recognition, see Arno Laitinen (2002) ‘‘Interpersonal
recognition’’ Inquiry Vol. 45 pp. 463–478, 471–472; Hans-Christoph Schmidt am Busch
(2004) ‘‘Marktwirtschaft und Anerkennung’’ in Christoph Halbig and Michael Quante
(Eds.) Axel Honneth (Münster: Lit Verlag) pp. 93–97.
18. For a similar critique: Rainer Forst Contexts of Justice (2002) (Trans.) J.M.M. Farell
(Berkeley: University of California Press) p. 282; Martin Fuchs (1999) Kampf um
Differenz (Frankfurt am Main: Suhrkamp) pp. 322 ff. Incidentally, this cultural
variation should not be understood as a ‘‘mosaic’’, as a collection of internally
homogeneous and hard-edged pieces that are neatly positioned against one another.
Instead, cultures are internally mixed, often overlapping, while cultural and subcultural
boundaries are by nature more often blurred than clear. See Ulf Hannerz (1992) Cultural
Complexity (New York: Colombia University Press).
19. At first sight, this critique resembles Nancy Fraser’s break with identity-based notions of
recognition (Nancy Fraser (2002) ‘‘Recognition without Ethics?’’ in Scott Lash and
Mike Featherstone (Eds.) Recognition & Difference (London: Sage) pp. 21–42; Fraser
 A Formal Recognition of Social Attachments 201

and Honneth Redistribution or Recognition?). However, her approach – the status model
of recognition – has two major drawbacks. Firstly, given her division in two moral
categories ‘‘status’’ and ‘‘class’’, it becomes conceptually difficult to make fine-grained
distinctions within the status model between different types of recognition with distinct
normative claims and specific justificatory burdens. Secondly, Fraser’s normative point
of departure, parity of participation, ultimately lacks normative conviction because it
has a tendency to sidestep the internal perspective of the agent of recognition-claims. See
Christopher Zurn (2003) ‘‘Identity or status?’’ Constellations Vol. 4 pp. 519–537.
. Another recent attempt to articulate a non-evaluative recognition of difference in
terms of equal status has been made by Peter Jones (2006) ‘‘Equality, recognition and
difference’’ Critical Review of International Social and Political Philosophy Vol. 9 pp. 23–
46. Although some of his observations are similar to mine, the problems are roughly the
same as with Fraser’s approach. The lack of conceptual differentiation is especially
striking. He does not differentiate sufficiently between the recognition of a person’s
autonomy and a person’s social attachments. As a result, it is unclear what the difference
is between his model and the traditional idea of respect for equal autonomy – an idea that
after all can also be explained with Jones’ formula: ‘‘what matters to you should matter to
me because it matters to you’’ (p. 30). This is potentially detrimental to the idea of a
politics of recognition (see beginning 1 III of this article). A related problem is that he is
unable to conceptualise conflicts between cultural rights and basic civil and human rights.
20. Brian Barry (2002) Culture and Equality (Cambridge: Polity Press) p. 40.
21. Ibid., p. 36.
22. See 1 IV. In fact, see also 1 V and 1 VI.
23. Amongst others: Will Kymlicka (1989) Liberalism, Community and Culture (Oxford:
Clarendon Press); Will Kymlicka (1995) Multicultural Citizenship (Oxford: Clarendon
Press).
24. Fraser and Honneth Redistribution or Recognition? p. 169.
25. Ibid., p. 167.
26. Ibid., p. 169.
27. See my ‘‘Social attachments as conditions for the condition of the good life? A critique
of Will Kymlicka’s moral monism’’ Philosophy & Social Criticism 32 (2006) pp. 401–428.
Although others have criticized Kymlicka as well for being too instrumentalist (I refer to
the relevant literature in my article), none of these commentaries provides a clear
conceptual distinction between: 1) the instrumental value of language or culture; 2) the
intrinsic value of language or culture as such; and 3) the intrinsic value of language or
culture for the group members involved. As a consequence, it often remains unclear
whether the alternative moral justification for minority rights is based upon value type 2
or 3. In the case of type 2, one actually ascribes rights to cultures themselves. As a
consequence, one creates the possibility of a duty for the group members involved to
maintain their own culture. See also footnote 55.
28. Tariq Modood (2000) ‘‘Anti-Essentialism, Multiculturalism, and the ‘Recognition’ of
Religious Groups’’ in Will Kymlicka and Wayne Norman (Eds.) Citizenship in Diverse
Societies (Oxford: Oxford University Press) pp. 175–195, 186–187. This is why it is
misleading to apply the ‘‘equality principle’’ to cultural rights in the case of immigrants,
because immigrants cannot claim equality with regard to the expression of their own
cultural identity in relation to the native population. There is no moral symmetry
between immigrant groups and the native population with regard to cultural rights,
although there is symmetry with regard to civil, political and social rights. For a sensible
approach to these kinds of moral asymmetries, see Joseph Carens (2000) Culture,
Citizenship, and Community (Oxford: Oxford University Press).
29. Cf. Lawrence Blum (1998) ‘‘Recognition, value, and equality’’ Constellations Vol. 5
pp. 51–68, especially 55 ff.
202 Bart van Leeuwen

30. Kwame Anthony Appiah (1994) ‘‘Identity, Authenticity, Survival’’ in Gutmann (Ed.)
Multiculturalism pp. 149–163, 151.
31. ‘‘Convergence is what happens when our values are shared. But what is required for
common meanings is that this shared value be part of the common world, that this
sharing be shared’’. Charles Taylor (1985) Philosophy and the Human Sciences
(Cambridge: Cambridge University Press) p. 39, italics mine.
32. It would constitute the kind of social affiliation that Avigail Eisenberg refers to as an
‘‘involuntary association’’, something very different from ‘‘nonvoluntary associations’’
and ‘‘voluntary associations’’. Avigail Eisenberg (1995) Reconstructing Political
Pluralism (New York: State University of New York Press) pp. 177 ff. An involuntary
association is an association that the individual actively rejects but, at the same time,
cannot avoid because it is linked to a characteristic that he or she possesses
nonvoluntarily, such as being black, and this characteristic influences how this person
is treated by others (p. 180). Hence he or she is identified by others as a member of a
certain group, but self-identification is lacking. In the case of a nonvoluntary association,
however, the affiliation is characterized by active commitments to the values of the
group, although the affiliations are not chosen in this case either (p. 179). Here both
other-identification and self-identification are present, as is the case with my conception
of social attachments. Voluntary associations are characterized by the fact that a person
makes an autonomous choice to be part of them, depending on whether or not the group
values in question seem reasonable or appealing – as is the case, for instance, with
membership to a political party.
33. Kymlicka Liberalism, Community and Culture ch. 4; Charles Taylor (1985) Human
Agency and Language (Cambridge: Cambridge University Press) especially on ‘‘radical
re-evaluation’’, pp. 40 ff.
34. Iris Young (1990) Justice and the Politics of Difference (Princeton: Princeton University
Press) p. 46.
35. Avishai Margalit (1996) The Decent Society (Cambridge: Harvard University Press)
p. 140.
36. Charles Taylor refers to this aspect of human agency as ‘‘strong evaluation’’. See Taylor
Human Agency and Language part 1; Charles Taylor (1989) Sources of the Self
(Cambridge: Harvard University Press) part 1.
37. Margalit The Decent Society p. 140.
38. Eugeen Roosens (1989) Creating Ethnicity (London: Sage) pp. 15–17
39. Why call a sense of belonging ‘‘self-respect’’? Because a sense of belonging qualifies as a
particular relation-to-self: it incorporates the sense that you are part of a larger social
unit. It is the sense of yourself as ‘‘attached’’, just as traditional self-respect (1st axis)
comprises the sense of yourself as ‘‘morally accountable’’.
40. Indicative of this trend is Christian Joppke and Ewa Morawska (Eds.) (2003) Toward
Assimilation and Citizenship (New York: Palgrave).
41. Nathan Glazer (1997) We Are All Multiculturalists Now (Cambridge: Harvard
University Press) ch. 6, ‘‘Where Assimilation Failed’’.
42. Two prominent writers in this regard are Barry, Culture and Equality and Richard Rorty
(1998) Achieving Our Country (Cambridge: Harvard University Press).
43. Recent empirical research covering 21 countries shows, for example, that ‘‘there is no
evidence […] of a systematic tendency for multicultural policies to weaken the welfare
state […] Indeed, in the case of immigrant groups and national minorities, the countries
with the strongest multicultural policies fared best in terms of growth in social spending
and greater redistribution, providing a hint that perhaps multicultural policies may
actually ease any tension between diversity and redistribution’’. Keith Banting and Will
Kymlicka (2005) ‘‘Multiculturalism and the welfare state’’ Canadian Diversity Vol. 4
pp. 103–106, 105; Keith Banting and Will Kymlicka (2004) ‘‘Do Multiculturalism
 A Formal Recognition of Social Attachments 203

Policies Erode the Welfare State?’’ in Philippe Van Parijs (Ed.) (2004) Cultural Diversity
versus Economic Solidarity (Brussels: De Boeck) pp. 227–284.
44. ‘‘Integration’’, on the contrary, means adaptation, for instance, to the national language
and fundamental laws, as well as the possibility of public expression of a minority identity.
45. It is an historical fact that forced assimilation did not always diminish the strong
identification of minorities with their particular cultural background. In certain
circumstances, especially in the case of territorial concentration, an ethnic revival after
a period of enforced uniformity is possible. The risk here is that the political suppression
of social identities reifies a sense of belonging and, as a consequence, might make it close
to impossible for different social groups to live together. See James Tully (1995) Strange
Multiplicity (Cambridge: Cambridge University Press) p. 197.
46. Avigail Eisenberg and Jeff Spinner-Halev (Eds.) (2005) Minorities within Minorities
(Cambridge: Cambridge University Press); Joshua Cohen et al (Eds.) (1999) Is
Multiculturalism Bad for Women? (Princeton: Princeton University Press); Ayelet
Shachar (2001) Multicultural Jurisdictions (Cambridge: Cambridge University Press).
47. Veit Bader (2005) ‘‘Associative Democracy and Minorities within Minorities’’ in
Eisenberg and Spinner-Halev (Eds.) Minorities within Minorities pp. 319–339, 328.
48. I will give two examples. Concerning area 2: Although polygamy is in fact legally
forbidden in all Western liberal democratic states, and although it creates a certain
asymmetry between the spouses, it is not obvious why it should be prohibited. How does
this fit with the general principle that adults should in principle be able to enter into
whatever contracts or personal relationships they choose? See Carens Culture,
Citizenship, and Community p. 155. Moreover, there are advantages to legal recognition
of this practice, such as: a) access to the rights associated with the status of legal spouse;
b) the possibility of legal encouragement to reform, for instance by making marriage
conditional on general consent (of all partners) and minimum age and; c) institutional
affirmation of at least a formal equality between the sexes by allowing both polygyny
(more than one wife) and polyandry (more than one husband). Cf. Jacob Levy (2000)
The Multiculturalism of Fear (Oxford: Oxford University Press) p. 57–59.
. Concerning area 3: Exclusion and discrimination – however morally wrong – have to
be legally tolerated in some cases, depending on the goals of the association and the
availability of realistic exit options. Meaningful associational autonomy for religions
implies the legal toleration of exclusion (for instance of women from priesthood) by
religious organizations, in matters directly connected with core beliefs and practices
(Ibid., p. 331). However, Bader rightly stresses that the state or public opinion can and
should interfere in these cases in other ways, for instance, with public criticism,
campaigns or by cutting subsidies. Cf. Amy Gutmann (2003) Identity in Democracy
(Princeton: Princeton University Press) ch. 2. Moreover, ‘‘the further the distance from
religious core organisations and core activities, the weaker the shield of ‘‘Free Exercise’’
should work, the more legitimate is the legal imposition of non-discrimination and equal
opportunity legislation’’ (Ibid., p. 332).
49. Ibid., p. 328.
50. Ibid., p. 328. Carens rejects infibulation in all cases, but he accepts the possibility of less
radical forms of circumcision in the case of adult women who voluntarily choose to
undergo this ritual. He also argues for accepting symbolic types of circumcision for
children (a small incision) that causes no more harm then male circumcision. Carens
Culture, Citizenship and Community pp. 145–153.
51. The adjective ‘‘minimalist’’ indicates that we are not talking of the list of rights and
recourses that are, traditionally, part and parcel of a full-blown conception of respect for
personal autonomy. Bader speaks of a ‘‘minimal morality’’ in this context. Cf. Veit
Bader (2007) Secularism or Priority for Liberal Democracy? (Amsterdam: Amsterdam
University Press).
204 Bart van Leeuwen

52. See Honneth’s critical review of Margalit’s The Decent Society: ‘‘A society without
humiliation?’’ European Journal of Philosophy Vol. 5 (1997) pp. 306–323. Another more
recent example of the negative justification for a politics of difference is Jacob Levy’s The
Multiculturalism of Fear. This book offers a very different perspective from my own
negative justification, on the basis of the recognition-theoretic approach. Where I
concentrate on the moral consequences of the potentially adverse effects on identity that
occur as a consequence of disrespect, Levy tends to concentrate more on the risk of social
conflicts and intercommunual violence that might emerge if certain groups within a society
are systematically downgraded and marginalized. Hence, the multiculturalism of fear. I do
not believe, however, that fear is either a legitimate or a convincing ground for the
justification of minority rights. For one thing, it gives the impression that ‘‘those pushy
minorities’’ simply get their way because otherwise they will wreak havoc on society.
53. See Will Kymlicka and Alan Patten (Eds.) (2003) Language Rights and Political Theory
(Oxford: Oxford University Press).
54. Honneth reduces, for example, the possibilities for the justification of a politics of
difference to these choices. Axel Honneth (1999) ‘‘Negative freedom and cultural
belonging’’ Social Research Vol. 66 pp. 1063–1077.
55. Daniel Weinstock ‘‘The Antinomy of Language Policy’’ in Kymlicka and Patten (Eds.)
Language Rights and Political Theory pp. 250–270, 1 2. To give just one example of a
famous argument that runs this risk: Charles Taylor refers in his work to the political
importance of cultural survival. Charles Taylor ‘‘Can liberalism be communitarian?’’
Critical Review Vol. 8 (1994) pp. 257–262; Taylor ‘‘The Politics of Recognition’’. But the
implication of this politics of cultural survival is ambiguous, because it could contain a
claim that the culture ‘‘itself’’ is of value and for that reason ‘‘survival through indefinite
future generations’’ should be safeguarded (ibid., p. 41). That strategy ignores the point
of view of these future generations themselves, and thereby ‘‘overlooks the possibility of
misrecognition occurring through one generation imprisoning the next in a false vision
of itself […] A culture’s self-understanding can change, and efforts to force a
contemporary understanding on the future seem hardly immune to the problems
around imposed identities Taylor seeks to avoid’’. Andy Lamey (1999) ‘‘Francophonia
for ever’’ The Times Literary Supplement Vol. 23 pp. 12–15, 14.
56. Cf. Avishai Margalit (2001) ‘‘The Crooked Timber of Nationalism’’ in Mark Lilla et al
(Eds.) The Legacy of Isaiah Berlin (New York: New York Review Books) pp. 147–159.
57. Cf. 1 IV. For this aspect of social attachments, see also the description of ‘encompassing
groups’, especially characterization number 5, in Margalit The Decent Society pp. 139–
140.
58. Honneth The Struggle for Recognition p. 122, cf. pp. 123, 126.
59. Will Kymlicka (1997) States, Nations and Cultures (Assen: Van Gorcum).
60. Will Kymlicka (2002) Contemporary Political Philosophy (Oxford: Oxford University
Press) pp. 217–219, 344.
61. Cf. Anna Galeotti (2002) Toleration as Recognition (Cambridge: Cambridge University
Press).
62. Gail Presbey (2003) ‘‘The struggle for recognition in the philosophy of Axel Honneth,
applied to the current South African situation and its call for an African renaissance’’
Philosophy & Social Criticism Vol. 29 pp. 537–561.
63. Charles Taylor (1994) Reconciling the Solitudes (Montreal: McGill-Queen’s University
Press) p. 183.
64. Richard Johnson (1986) ‘‘The Story So Far’’ in David Punter (Ed.), Introduction to
Contemporary Cultural Studies (London: Longman) pp. 277–313, 303.
65. Gerd Baumann (1997) ‘‘Dominant and Demotic Discourses of Culture’’ in Phina
Werbner and Tariq Modood (Eds.) Debating Cultural Hybridity (London: Zed Books)
pp. 209–225, 210.
 A Formal Recognition of Social Attachments 205

66. Some groups (e.g. the Amish) do live in almost complete isolation from the rest of
society. But these groups are unique and can be considered an exception to the rule.
67. I am indebted to Veit Bader for the analogy with health and healthcare.
68. One could argue that there cannot be a ‘‘politics of recognition’’ without a ‘‘recognition
of politics’’: Kwame Anthony Appiah (2005) The Ethics of Identity (Princeton: Princeton
University Press) p. 101. See also Jürgen Habermas ‘‘Struggles for Recognition in the
Democratic Constitutional State’’ in Gutmann (Ed.) Multiculturalism pp. 107–148;
James Tully (2000) ‘‘Struggles over recognition and distribution’’ Constellations Vol. 7
pp. 469–482.
69. Taylor ‘‘Can liberalism be communitarian?’’ p. 262.
70. This article is a substantially revised version of Bart van Leeuwen (2001) ‘‘Erkenning,
identiteit en verschil’’ Tijdschrift voor Filosofie 63 pp. 751–784. I thank Veit Bader, and
two anonymous referees for their helpful comments. This research has been funded by
the Netherlands Organisation for Scientific Research.