Social Capital as a source of Majority Sentiment

Abstract: Sociology textbooks present majorities as the counterparts of minorities. This reflects the influence of US sociology, fashioned in a country of immigration that lacks any Staatsvolk. Majority sentiment in situations of immigrant settlement elsewhere draws from many sources, including mass media reporting, the policies of central and local governments and personal experience of a decline in the sense of community with co-ethnics. The reported sense of loss can be explained as a decline in bonding capital. If majority members share goals with minority members, this can build bridging capital. Research into the preferences for association with co-ethnics could investigate the circumstances in which civic norms are given priority over ethnic norms.

Keywords: community, majority, preferences, social capital, Staatsvolk, Weber.

Majority sentiment is the dark matter of the sociological universe. Because it is amorphous, it is difficult to grasp; yet it exerts a profound influence upon the selection of problems for research, and therefore upon the character of sociological knowledge. Hitherto, the research worker has usually seen it from its mirror image, the sociology of minorities[1].

The conception of a minority as anything other than a numerical category dates from the end of World War I, from US President Woodrow Wilson’s insistence on the self-determination of peoples, and from the re-drawing of the map of Europe to implement the Treaty of Versailles. Borders were revised, transferring territories to other states, so that some members of the transferred population constituted national minorities. Some of the receiving states were obliged to enter into treaties under which the League of Nations became responsible for ensuring that their rights were protected. This gave rise to the conception of a minority as a political category, so that, for example, blacks in South Africa were on occasion accounted a minority even though they constituted a majority of the population.

Minorities needed protection, so minority rights were written into laws. Majorities did not need protection, so there was no need to consider whether there were any majority rights. Membership of a majority was a default condition, not a distinctive social position.

East and West of the Atlantic

Though the three greatest founders of sociological knowledge, Marx, Weber and Durkheim, were Europeans, much of the subject’s development has been the work of scholars in the USA, and particularly of scholars on the East Coast and in Chicago. US interest in sociology has reflected the policy concerns of the federal state and its population. Many US sociologists have taken their own country as a paradigm case, not least in the study of immigration and ethnic relations.

Looking at this field from a global perspective, it is important to appreciate that there is a major difference between societies founded upon immigration, like the USA and Australia, and those accustomed to emigration, like many West European countries. Both the United States and Australia developed with an immigrant conception of their national society. The indigenous peoples had been pushed aside and excluded from this conception as a disruptive influence. Only in Canada, where they have the status of ‘first nations’, has the conception of the national society not been biased in this way. In the sociology departments of universities of the US East coast, and in Chicago, relatively little attention was paid to the first nations of the USA.

In very many other countries there is a population consisting of people who, if pressed, will say that they are the Staatsvolk, the original inhabitants and therefore the owners of the territory[2]. Most of the time they are not pressed. They are not obliged to think about who they are, and do not want to engage in what many regard as a pointless exercise. Majority-minority relations are therefore different on the two sides of the Atlantic Ocean.

In most European countries, majority sentiment is inchoate and tacit. It is not easily investigated by survey methods because the shared sentiment is diffuse; in an opinion poll about people’s concerns, it would yield place to more specific and immediate priorities. So its importance is easily under-estimated.

Sociological knowledge about majority sentiment

Sociology has been sympathetic to the minority perspective. In its textbooks, majority sentiment is discussed as illustrating ethnocentrism, xenophobia, prejudice, discrimination and exclusion. In some recent research the study of stigmatization has been a focus of attention (Lamont and Mizrachi 2012). Approaches like these imply a moral criticism of majority sentiment, focusing on what are presented as its negative characteristics, neglecting the positive ones, and building an atmosphere of what Max Weber would have called pathos. This negative perspective has been central to the growth of sociology, once characterised by Werner Sombart as an oppositionswissenschaft. Yet if minorities have justifiable grievances, they are criticisms of majority action or inaction. If changes are to be encouraged, it is important to understand the sources and determinants of majority sentiment.

The negative assessment of the effects of majority power is exemplified by a ‘Theoretical essay on Established and Outsider Relations’ written by Norbert Elias to introduce a study, The Established and the Outsiders (1994). According to Elias, that study was an account of how in ‘Winston Parva’ an old-established group of residents closed ranks against those who had taken up residence more recently; they had stigmatised the newcomers as people of lesser human worth. The power differential between ‘the village’ and ‘the estate’ ‘was based on the high degree of cohesion of families who had known each other for two or three generations, in contrast to the newcomers who were strangers not only in relation to the old residents but also to each other’(1994: xviii). How acquaintance converted into cohesion, and thereby power, was not explained. It was said that in such relations stigmatisation was evident in the use of insulting expressions producing the humiliation ‘intended by their use’ [italics added] (1994: xxv).

In the book itself, Elias and Scotson reported that though the newcomers were regarded as socially inferior, this did not result in personal enmity or friction between the villagers and the incomers. Nor did it occasion difficulty in their occupational roles at their workplaces. A scale of socio-economic status was implicit in the way people in both the village and the estate sometimes thought about their relations with others. ‘Every woman in ‘the village’, if not every man, appeared to know the status and prestige rating of every family and every association in the community’ (1994: 56). The authors’ highlighting of customs of cooperation in the village conveyed an impression that the residents in the village got more enjoyment out of their lives than did the residents in the estate.

Elias and Scotson also stated that ‘As elsewhere people allowed themselves to be ranked lower than others because they could not prevent it (1994: 40). This may be the authors’ construction of the matter because they do not furnish any evidence about whether the residents of the estate were aware of this ranking, what they thought about it, and whether ranking should be seen as an exercise of power.

Much more important was their observation that ‘The people who settled and lived in Winston Parva had learned and stored away for reference in appropriate situations specific ways of arranging communal affairs and of dealing with communal problems. They could draw on these stored images as models indicating how to do and how not to do things in community affairs’ (Elias and Scotson 1965: 68). There is an implication that the villagers had more comprehensive models than those who lived on the estate, but in both of them there was a scheme of order that regulated social life.

This description of social life in Winston Parva may have been biased by the authors’ focus on the Outsider notion and by insufficient acquaintance with everyday life on the estate. Such a doubt is reinforced by Bernice Martin’s (1981: 53-78) account of growing up in the working-class culture of a Lancashire cotton town in the 1940s and 1950s; it contributed to her summary of a dozen studies of similar cultures. She stressed the extent to which it was a culture of control that operated by the observation of social boundaries. ‘Until very recently indeed, the price of relaxing any of the boundaries and controls was very high. It made the difference between respectability and degradation, between coping and debt, between survival and starvation, independence and the workhouse’(1981: 61). The social world was divided into categories that were ‘perceived as part of the natural order’(1981: 69).

Half a century later, the sociologist can interpret this order, and the models of ‘how to do and how not to do things’, as forms of social capital (Coleman 1990, Putnam 2000). No local community is static. There is always internal change, as one generation succeeds another, and as some individuals prosper while others fail. There will almost certainly be a scale of social status in which a low score on one social attribute can be traded off against a high score on another; the scale will be fluid, with some people rising and others falling. There will be changes that originate from events outside the community, requiring internal readjustments. These will entail trade-off, in which costs are accepted to the extent that they confer benefits. It is exchanges of this kind, the give-and-take of everyday life, that build social capital and generate the sense of order.

The concept of social capital is simple. Social relations are the relations between individuals, like John Doe and Richard Roe. These relations are conducted on the basis of different relationships, like father and son or seller and buyer; these are relations between roles, defined by rights and obligations. If John Doe and Richard Roe enter upon and maintain a new relationship, it is because they gain benefits from it, either material or immaterial, that exceed any costs. The sum of the net benefits from social relations within a group constitutes its social capital. Therefore some groups, like ‘the villagers’ in Winston Parva, may have had more social capital than others, like those who lived in the estate. This is another, in some respects more informative, way of grasping what Elias and Scotson meant when they referred to a difference in social cohesion. Some of the villagers were proud of their quality of life compared to life in the estate. To represent their attitudes as simply ‘stigmatisation’ is to ignore the most important feature of the difference between the two groups.

That difference was part of a bigger one that has long been known in sociology as the difference between community and association. The canonical statement is that of Max Weber[3]:

Communal relationships were exemplified by religious brotherhoods, erotic relationships, relations of personal loyalty, national communities, the esprit de corps of military units and by the family. The ‘purest cases of associative relationships’ were given as (a) free market exchange; (b) a voluntary association oriented to ‘specific ulterior interests’; and (c) a voluntary association that seeks only to serve a cause.

Weber did not make this explicit, but he appears to have believed that while a subjective feeling creates a social relationship, that relationship also creates the feeling. The two are reciprocal, each determining the other. To write of an ‘orientation’ is to suggest that there is a subjective feeling or an intention that pairs with the obligation; they are like two sides of one coin. If that is the case, the two sentences could equally well have read:

If the subjective feeling and the objective relationship are reciprocal, Weber’s statement can be simplified. When two persons interact, their conduct is governed by the norms of one or more social relationships. Gesellschaftlich relations are one-dimensional, governed by the norms of a single pair of roles. Gemeinschaftlich relations are multidimensional, governed by the norms of any pair of roles that the parties choose to play. Individuals do not normally choose to enter into communal relations; they are born into them, or find that they have entered them when they marry or, to use Weber’s example, have joined a military unit.

Associative relations are formed for specific purposes; parties to them play specific roles. If they start to interact on the basis of other roles as well, the relations between the persons in question are no longer strictly associative but are of a mixed character. Thus the contrast between community and association as polar opposites can be changed into a scale starting from one-dimensional relations up to a high point at which two parties identify with one another like identical twins.

Community living brings benefits and costs, both material and immaterial. Young adults may experience the pressure to conform as so great that they want to escape from it; Stadtluft macht frei ran the old saying. They maintain their membership of their communities of origin so long as they believe the costs of exit to be greater than the benefits of remaining within the group. If they choose to live elsewhere they may be able to retain a qualified membership.

Even in a tightly bonded community, people do not always like their neighbours. While a husband and wife may wish to have as little contact as possible with the people who live next door, situations can arise in which neighbours have to engage with each other, and perhaps co-operate. Communal relations were characterised by Max Gluckman (1965) as a web of crosscutting ties, such that two persons who were opponents on one issue had to take account of the possibility that there would be other issues on which they would want to be allies. His argument had something in common with the statement of Émile Durkheim that ‘the division of labour in society is in direct ratio to the moral or dynamic density of society’(1947: 257). Community living is based on high social density, which, in turn, is facilitated by the ability to readjust to changing circumstances. These processes produce what Hayek (1973: 37) called spontaneous order. He distinguished this from the order that results from deliberate organisation; differential power enters into both forms because every social relationship has a power dimension. Coleman advanced ‘the general point that organization brought into existence for one set of purposes can also aid others, thus constituting social capital that is available for use’ (1990: 312).

Majority sentiment is a product of community, not of association. It can be observed at different points on a scale, from a handful of persons, to a village, to members of an ethnic group, or to a nation. Just as Weber, on this interpretation, saw sentiment and social relationship as two sides of one coin, so majority sentiment interacts with social situations, taking different forms at different points in the scale.

The question of scale

The previous paragraphs have discussed majority sentiment at the national level. For sociological purposes it may be better first to consider how this sentiment may come to be generated and expressed at the local level. Many of the community studies that have been conducted in Britain have reported that only those born in the locality are regarded as truly ‘locals’[4]. The distance that is expressed towards ‘incomers’ may be a way in which people affirm their identification with a community, not a stigmatisation of them as people of lesser worth. In the most dangerous of situations, those of armed conflict, identification with a primary group is the key motivator and the guarantor of reciprocity, not the ‘secondary symbols’ of the national flag or the doctrines of national duty (see Shils 1950). Identification, whether it is with co-villagers, co-ethnics, or co-nationals, has common characteristics at all points in the scale.

Perhaps the most remarkable of the findings from research into what Tajfel (1981) called ‘minimal group discrimination’ was that so little stimulus was needed to cause young male subjects to identity as fellow members of a team when playing in an experimental game. Subjects were not supposed to know who their fellow team members were, and yet they were willing to make monetary sacrifices in the attempt to defeat the other team.

If it be accepted that there is some sort of scale of social identification running from very low-level to very high-level, how is it to be conceptualised? Insiders and outsiders have different perspectives. Majority sentiment is related to a proper name, a name unique to the group in question. It is concerned more with uniqueness than with position on any scale. Use of a proper name is a general feature of ‘we’ sentiment. It was exemplified in the choice of the title Navaho Means People for a book about one of the indigenous peoples of the US South-West.

‘We’ sentiment has to adjust when the ‘we’ come into relations with others for whom they are a ‘they’. The Navajo found that outsiders classified them as ‘Indians’ or ‘Native Americans’. Members of a minority may have their own ideas about how they should be identified. In the 1960s and 1970s a new generation of persons of Mexican origin in California engaged in a ‘battle of the name’. They sought to persuade others of Mexican origin, and members of the wider population, that they should be called Chicanos to demonstrate that they were a distinctive class of US citizens.

In the United States persons of Mexican origin may find themselves classified as ‘Hispanics’ along with the members of other social categories with whom they do not identify. Indeed, because members of certain ‘Hispanic’ categories may be of lower socio-economic status, other ‘Hispanics’ may try to avoid being associated with them. In Britain similarly, persons from very different parts of the Indian subcontinent find themselves classified as Asians. In the country of settlement, identification by religion may become more important than country of origin.

If self-identification has common characteristics whether they relate to a primary group or a large-scale one, one central characteristic will be the use of a scheme of order that regulates social life. Its expression may take an apparently nationalist form only because that is the possibly inadequate way in which others classify it (cf. Billig 1995: 39, 62). It may not be what some would consider pure nationalism. ‘Nationalism’ is an ordinary language word used to designate varied attitudes and kinds of behaviour; to move from everyday use and attempt socio-scientific analysis requires that new concepts be developed to unpack the complexities.

At the present time, outsiders often make use of the adjective ‘ethnic’ or the noun ‘ethnicity’ to identify social categories. One author, when writing about persons of French origin in Quebec, prefers to employ the proper name Québécois because ‘by using the term ethnic one is perceived as negating the legitimate right of a national community to self-determination’ (Juteau 2004: 98). She reacts against the approach that starts from the conception of a ‘dominant ethnie’ introduced by Anthony D. Smith (2004) and employed in the volume, Rethinking Ethnicity, in which her essay was published.

Smith’s argument that the United States’ image and use of the word ethnicity has distorted understanding of the social significance of differences of ethnic origin is one to be endorsed. Yet Smith and other contributors to the volume take the existence of ethnies for granted. They do not try to uncover the factors that, internally, sustain their existence. The attempt to ‘rethink ethnicity’ is misconceived because it starts from an idea rather than from an observation of behaviour that has to be explained. An approach from the analysis of social capital is a superior alternative because it copes with individual variation and the factors that result in the creation of groups, their maintenance and dissolution. It does not presume that any concept of ethnicity will have explanatory value.

This essay maintains that majority sentiment is tacit, inchoate, not explicit; often it is not adjusted to challenge – challenges are frequently rejected as misguided because majority members simply do not want to examine what they take for granted, namely their identification with others, neighbours, co-ethnics.

The sources of majority sentiment

In all societies humans grow up to regard the customs followed by their kin and neighbours as the proper way to behave. Many of those customs are associated with particular social classes and there is variation from one individual to another, but the relevant consideration is that the familiar is generally treated as in some degree normative. If they grow up in a milieu in which everyone is of the same colour and speaks the same language, many will regard that colour and that language also as normative. This proposition will apply, mutatis mutandis, whether the milieu is fashioned by members of the ethnic majority, by a population that on the national scale is an immigrant minority, or by a population of mixed composition.

Bernice Martin (1981: 141-152) has maintained that the principles that explain the working class culture of the Lancashire cotton towns can be applied to explain the new forms of expression displayed from the 1950s in the behaviour of urban youth. The ‘sense of group’ was the most distinctive trait of the ‘teddy boys’, the skinheads and the football cults. The boys who most wanted to break rules were at a loss in situations in which there were no rules to break. Social classes were seen as essentially antagonistic categories, but so were races, sexes and every kind of territorial unit. The significance of territory has persisted in the twentieth century. In London, there are reports of ‘post code identities’ among youth. In some neighbourhoods a strange young man is liable to be accosted and asked ‘Where are you from? What are you doing here?’ Groups form that are based on the codes used for the delivery of mail.

Sentiments are associated with territory at both the national and the local level. A study conducted in immigrant neighbourhoods in three Swiss cities has produced more detailed findings than those for Winston Parva though not inconsistent with them (Wimmer 2004). It found that in all three neighbourhoods residents cultivated a scheme of order that comprised virtues like cleanliness, punctuality and quiet, plus stable social relationships. From the perspective of the old-established residents, there had been ‘a loss of this order and even worse, an overall devaluation of order as a central value, because immigrants and younger Swiss were unwilling to conform to it’(1994: 9). These old-established residents may have been simply nostalgic, because the data showed that, to the contrary, ‘on the level of everyday life and daily interaction, established immigrants have developed a strong local orientation similar to the one of the non-immigrant population with a similar class background’(1994: 26).

The sociologist concluded that it was more important whether the courtyard was kept tidy and the rules of the building followed than whether a family was black or white or of Swiss or foreign origin. So that ‘while ethnic-national groups are taken for granted entities, they do not per se play a central role in describing and understanding the social world of our informants’(1994: 27). It was the scheme of order that determined the relative position of those assigned to an ethno-national category and this scheme provided for the possibly negative value of foreign origin to be traded off against the positive value of conformity to majority expectations.

A re-analysis of data collected in a large-scale survey in Germany noted that, because ethnic categories are taken for granted in German public discourse, it might be expected that ethnic categories would be salient in everyday life. The survey asked ‘Which groups are mostly responsible for problems in your neighbourhood? You may now name up to three groups’. It might have been expected that respondents would employ ethnic categories in their answers. However, only about 31 per cent of respondents named any group, and about 25 per cent of the actual references were to teenagers. The elderly, lower class people and drunkards were more frequently blamed for neighbourhood problems than Turks (the most frequently cited ethnic group). In regions undergoing economic decline there was a stronger tendency to specify groups as responsible for problems. An important conclusion of the re-analysis was that ‘the issues of people’s everyday lives differ from those that are expounded in public discourse’ (Schaeffer 2012: 169). The scheme of order identified in the Swiss and German studies represented the residents’ social capital, as described above.

Individuals will vary in their preferences for cultural uniformity, but some will experience a sense of loss when persons speaking other languages, wearing different costumes, and introducing what are seen as foreign institutions, appear in their neighbourhoods in significant number. They may believe that ‘when in Rome you should do as the Romans’.

In contemporary Britain there is an element of Staatsvolk sentiment within the ethnic majority. A study of London’s East End carried out in the mid-1990s reported that:

Majority sentiment comprehended both negative feelings about change to the neighbourhood and more neutral feelings about the incomers as individuals. (Dench, Gavron and Young 2006: 172, 177, 181, 208).

These statements were in line with Schaeffer`s conclusion that ‘the issues of people’s everyday lives’ differed from mass-media representations of the national scene. Other research has confirmed that experience-derived impressions of inter-group relations are more favourable than media-derived impressions of those relations; as a result, members of the public often believe that others are less well disposed towards incomers than they are themselves; this is a form of ‘pluralistic ignorance’.

Apart from the mass media and personal experience, majority sentiment about immigrants may be influenced by opinions about the policies of central and local governments. This was brought out strongly by a survey conducted in Willesden, an area of London (Miles and Phizacklea 1980). A sample of 149 persons was interviewed, half in a factory sample and half in a residential sample; half English, half West Indian. They were asked what they saw as the local problems facing ordinary people. The authors reported that the most striking characteristic of English responses was the sense of loss, of being left behind in a decaying area that had been taken over by the ‘coloureds’. Their complaints centred on the belief that they did not receive the right allocation of social housing, not upon interpersonal relations. There was more evidence of a general status struggle than of any struggle between black and white.

A more recent study of white working-class views in three urban neighbourhoods found that 72, 78 and 82 per cent of respondents agreed that in their areas ‘people from different backgrounds get on well together’. Yet ‘there was a sense that the government was not listening to [their] concerns’ and ‘the racialised commentary should be seen through the prism of neighbourhood loss, political disconnection and competition for scarce resources’. Residents ‘were outraged by the suggestion that their views could be considered racist’; ‘in all three areas [they] expressed a view that social capital was being eroded’ (Beider 2011: 7, 13, 47, 50-51)[5].

In such circumstances persons in both the majority and the minorities may feel aggrieved, and this feeling, more particularly on the part of majority members, may contribute to criminal attacks, often collectively undertaken, on minority members. General sentiment and the motivations of individuals have to be distinguished. Thus the study in London’s East End concluded that ‘conventional attempts to promote racial harmony by attacking ‘racism’ may not be helping. For such efforts often involve a misrepresentation of what it is that working-class respondents are trying to say. ‘When they try to express what they feel they are made to appear bitter, xenophobic and out of touch with mainstream national opinion’ (Dench, Gavron and Young 2006: 216-217). Working-class respondents sometimes complain that those they have elected to represent them do not represent their feelings in this matter[6].

It looks as if the ‘sense of loss’ reported by Kaufman (2004: 2) and by the Swiss and Willesden studies, expresses a loss of social capital. If social capital is built by the multiplication of social relationships, future research could well investigate popular preferences for association with co-ethnics in different situations. It could well focus on behaviour in situations in which individuals have to decide whether they will observe an ethnic norm or a civic norm. One reason for the paucity of such research may be a fear that this could prove a political minefield. Since members of minorities also have preferences for association with co-ethnics, and also have at times to choose between an ethnic norm and a civic norm, this ought not to be too sensitive an issue.

Public goods bring private benefits

Several studies have reported that co-operative action takes place more readily among socially homogeneous groups. For example, a study in Kampala, Uganda, reported that the residents of the poorer neighbourhoods had to cope with major problems of drainage, garbage removal, and personal security. Heavy rainfall caused severe flooding, and this was made more serious by the accumulation of refuse in the open drains. The city council failed to keep all the drains clear and to remove all the garbage. The cessation of funding for the community patrols that once served as a protection had meant that the constant threat of theft had reduced the quality of local life. Why then did the residents of many neighbourhoods not themselves organise to remove garbage and establish a neighbourhood watch?

The findings of the study were complex. While they supported the conclusion that ethnically heterogeneous communities have greater difficulty acting collectively, the sharing of ethnic origin did not provide a sufficient explanation. Underlying the social behaviour there appears to have been a universal norm of reciprocity. The subjects apparently found it easier to develop reciprocal relations with co-ethnics, and thereby to engage in collective action (Habyarimana, Humphreys, Posner, & Weinstein 2009: 124-129). An analysis of relative ethnic, linguistic and religious heterogeneity in states of the Organisation of Economic Cooperation and Development (OECD) found ethnic fractionalisation to be a significant predictor of economic performance (Patsiurko, Campbell & Hall 2012). It did not identify the causes of the observed association. Those who pioneered this mode of analysis traced the negative effects of ethnic fractionalisation to ethnic differences in preferences for alternative forms of public expenditure (Gisselquist 2013).

In a landmark contribution to the study of social capital, Robert Putnam compared two theories. The first, usually labelled the ‘contact hypothesis’, argues that ‘diversity fosters interethnic tolerance and social solidarity’(2007: 11). Yet Putnam thought it fair to observe that ‘most (though not all) empirical studies have tended instead to support the so-called ‘conflict theory’, which suggests that ... diversity fosters out-group distrust and in-group solidarity’ (2007: 12). Whether contact promotes tolerance or conflict depends upon the kind of relationship within which the contact occurs, and the attitudes and expectations people hold before entering into contact. These theories cannot therefore be properly tested by opinion surveys carried out at a single moment in time. Survey methods can uncover associations but not their causes.

Putnam argued that the touchstone of social capital is generalised reciprocity and that this rests upon trust in others (cf. Coleman 1990: 318-321). A major study in the Netherlands carried this argument further by examining the neighbourhood effects of ethnic, religious and economic diversity, plus Dutch-language proficiency, while allowing for individual characteristics of respondents (such as educational levels). These were assessed against three scales measuring different forms of social trust: ‘the first to the quality of contact, the second to trust in the neighbourhood and the third to trust between ethnic groups’ (Lancee and Dronkers 2011: 603).

Putnam concluded that in the USA ethnic heterogeneity was associated negatively with the quality of contact with neighbours. This finding was confirmed for the Netherlands, even after controlling for the effects of economic, religious and language differences. However, ‘we do not find an association between ethnically diverse neighbourhoods and trust in the neighbourhood, neither for the immigrants nor for the natives.’ Moreover, ‘ethnic diversity has a positive effect on the level of inter-ethnic trust of Dutch residents, but a negative effect on the quality of contact with neighbours for everybody’ (Lancee and Dronkers 2011: 615).

Diversity in the neighbourhood did not necessarily have, for Dutch residents, the same sort of effect as variations in individual characteristics. Thus religious diversity decreased the quality of contact with neighbours, trust in the neighbourhood and inter-ethnic trust. At the same time, a higher level of individual attendance at religious services increased the scores on all three indicators of trust. Another study in the Netherlands found that the theory of preferences and structural constraints can explain entry into inter-group contact among both immigrants and natives (Martinović 2013).

Economic diversity was associated with higher trust in the neighbourhood and higher inter-ethnic trust among both immigrant and native residents[7]. Two other authors suggest that economic differences can be ‘synergetic’. By this they apparently mean that ‘people with a different ethnic background are less likely to compete with one another. They may even be complementary: a consultant needs a bakery, and a renter profits from the owner renovating his house and making the street more attractive to live in’ (Lancee and Dronkers 2011: 601).

A recent study of social cohesion in local communities in Britain found no evidence that racial diversity had an eroding effect upon social interaction once allowance had been made for the association between racial diversity and economic deprivation. At the same time it reported a ‘puzzling’ finding that racial diversity had a direct negative effect on the perceptions of, and trust in, fellow neighbours (Letki 2008). That puzzle might be solved by further research into the conditions in which perceptions of neighbours are derived from the mass media rather than personal experience.

The findings of empirical research are in general compatible with the argument that multidimensional social relations embody more social capital than uni-dimensional relations because they facilitate reciprocity and the accumulation of social capital. They suggest that the sentiments that can make a Staatsvolk appear distinctive stem from the same source. Moreover, they point to possible ways in which such a mode of explanation may be developed.

Earlier it was noted that local communities frequently have to adapt to changes in their social environment. Their members have to engage in a process of trading off some of their customary satisfactions to gain new benefits, or to counter a decline in their circumstances. Social capital of the kind identified in the Winston Parva village has been called bonding capital, a reinforcement of the community’s homogeneity. Such reinforcement may hinder adaptation to a change in circumstances.

It would have brought diminishing benefit to the residents in what were described as ‘decaying’ neighbourhoods in Willesden. They might have done better to trade off the satisfactions of fellowship with co-ethnics against the benefits that might be secured by the development of alliances with the newcomers. This was reported as the strategy of the residents of Moss Side in Manchester when they were challenged by the city council’s plan for slum clearance. Anxiety about what new housing might be made available gave rise to ‘a series of housing movements, all characterized by a multi-racial membership and constituency and all seeking to secure interests which were not directly related to race or ethnicity’ (Ward 1979: 205). The development of new relationships that dissolve social boundaries is called bridging capital. In industrial societies adaptation requires both bonding and bridging capital.

Other regions

Comparison may throw further light on the special characteristics of the societies considered so far. For example, the changes to the border between Hungary and Romania seem to have engaged majority sentiment in important ways. The city of Cluj, previously in Hungary, was transferred to Romania in 1920, was back in Hungary during 1940-44, and then Romanian again thereafter. The authors of a recent study report a conversation in which a male Hungarian-Romanian recalled, ‘When I was a kid I got slapped once on a tram because I was speaking Hungarian’; to which a woman added ‘Me too!’ Other similar evidence shows that the use of a minority language in public settings can cause offence (Brubaker, Feischmidt, Fox and Grancea. 2006: 331 and 249-50). It is surely probable that whoever slapped these children was thinking ‘We are in Romania. People should speak the language of the Staatsvolk’.

In the mid-1980s the population of Fiji was recorded as including some 46 per cent of persons of Fijian origin (and of ultimately Polynesian origin) together with over 48 per cent persons of Indian origin. Powerful elements in the ethnically Fijian population were alarmed when the election of 1987 was won by an alliance that was perceived (or misrepresented) as a front for the Indian section of the population. A military coup overthrew a democratically elected government. Some ethnic Fijians have since sought to justify their political dominance by appealing to declarations of the rights of indigenous peoples (Salomon 2003). Can being indigenous be a source of majority rights?

The Federation of Malaya came into being in 1957 with a constitution that conferred privileges on Malays as an ethno-religio-linguistic social category. In 1963 it became the Federation of Malaysia, with Singapore as one of its component parts. Singapore was effectively expelled two years later, apparently because many Malay-Malaysians feared the power of Chinese-Malaysians within the Federation.

Malay-Malaysians believed that they were the Staatsvolk; they feared for their ability to survive as the kind of people they were. Unlike the ethnic Fijians, they could not appeal to any indigenous rights since it was the Orang Asli who were recognised as the earliest inhabitants of peninsula Malaysia.

Conclusion

The issues raised in this discussion suggest that while the study of majority sentiment is important, it is not adequately grasped if it is seen as a mirror image of minority sentiment. Majority sentiment can be traced to several sources, one of the most important being the sense of community as opposed to association. Even if the village in Winston Parva was a typical community, the estate was not an association; it too was a community, if one that displayed less cohesion. The opposition of community to association that has been traditional in sociology can be updated if community and association are treated as the end points of a scale. At one end are multidimensional social relations, with uni-dimensional relations at the other end.

Social capital is a resource that can be drawn upon by those who are parties to a web of social relationships. They can use it to promote ends of any kind. The sense of loss reported by ethnic majority members in neighbourhoods of immigrant settlement in European cities may reflect, not hostility to the newcomers, but a loss of social capital occasioned by the diminution of multidimensional relations. Such relations rest upon, and promote, reciprocity. One characteristic of bonding capital is the existence of a social boundary around the beneficiaries; the restriction of relations with those who do not participate to the same extent in this web of relationships is a necessary consequence.

The reactions of ethnic Fijians to the growing power of the settlers of Indian origin, and those of ethnic Malays to the enterprise of settlers of Chinese origin, can similarly be interpreted as occasioned by the fear of a loss of social capital within a larger political union. A political union can be loosened when those in one of its component parts come to value their sense of community more than their association with others in the larger union[8].

The increasing body of research evidence showing that ethnic, linguistic and religious heterogeneity is negatively associated with economic performance may prompt policy-makers to investigate ways to promote the growth of bridging capital. The most promising strategy for the future lies in the devising of experimental studies that will separate the dimensions of interpersonal relations, particularly but not only in neighbourhoods of immigrant settlement. At the heart of the matter are questions with which sociologists have long been concerned, notably how social norms promote social cohesion and the quality of life. Analysis of the preferences of persons belonging to both ethnic majorities and minorities for the observance of civic as opposed to ethnic norms will be central. Measurements of these preferences could contribute to the analysis of the social capital of both majorities and minorities.

Biography

Michael Banton has been Professor Emeritus of Sociology in the University of Bristol since 1992. He taught social anthropology in the University of Edinburgh 1954-65; political science in Massachusetts Institute of Technology 1962-63; and sociology in the University of Bristol 1965-92. Michael was President of the Royal Anthropological Institute of Great Britain and Ireland 1987-89, and, from 1986 to 2001, a member of the UN Committee on the Elimination of Racial Discrimination (chairman, 1996-98).

References

• Beider, H. (2011) Community Cohesion: the views of white working-class communities. York: Joseph Rowntree Foundation
• Brubaker, R., Feischmidt M., Fox, J. & Grancea, L (2006) Nationalist Politics and Everyday Ethnicity in a Transylvanian Town. Princeton: Princeton University Press.
• Billig, M. (1995) Banal Nationalism. London: Sage.
• Coleman, J. S. (1990) Foundations of Social Theory. Cambridge MA: Harvard.
• Dench, G., Gavron, K.. & Young, M. (2006) The New East End. Kinship, Race and Conflict. London: Profile Books.
• Durkheim, É. (1947) The Division of Labor in Society. Glencoe, Ill.: The Free Press (first edition, Paris 1893).
• Elias, N. & Scotston, J. L. (1965/1994) The Established and the Outsiders. A Sociological Enquiry into Community Problems. 2^nd. Ed. 1994 London: Sage.
• Gisselquist, Rachel M. 2013 ‘Ethnic Divisions and Public Goods Provision, Revisited’. Ethnic and Racial Studies DOI:10.1080/01419870.2012.762106
• Gluckman, M. (1955) Custom and Conflict in Africa. Oxford: Blackwell.
• Habyarimana, J., Humphreys, M., Posner, D. N. & Weinstein J. M. (2009) Coethnicity. Diversity and the Dilemmas of Collective Action. New York: Russell Sage Foundation.
• Hayek, F. A. (1973) Law, Legislation and Liberty. Vol. 1: Rules and Order. London: Routledge.
• Jellinek, Georg 1921 Allgemeine Staatslehre. 3^rd ed. [1^st ed. 1900]. Berlin: Springer.
• Juteau, D. (2004) `Pures laines’ Québécois: the concealed ethnicity of dominant minorities. In Kaufmann E. P. (ed.) Rethinking Ethnicity. Majority Groups and Dominant Minorities. London: Routledge, 84-101.
• Jolley, Rachel, & Sunder Katwala 2012 Hopes and Fears: The British Future State of the Nation Report 2012. London: British Future.
• Kaufmann, E. P. (ed.) (2004) Rethinking Ethnicity. Majority Groups and Dominant Minorities. London: Routledge.
• Lamont, M. & Mizrachi, N. (2012) ‘Ordinary People doing extraordinary things: responses to stigmatization in comparative perspective’. Ethnic and Racial Studies 35(3): 365-381.
• Lancee, B. & Dronkers, J. (2011) ‘Ethnic, Religious and Economic Diversity in Dutch Neighbourhoods: Explaining Quality of Contact with Neighbours, Trust in the Neighbourhood and Inter-Ethnic Trust’. Journal of Ethnic and Migration Studies 37(4): 597-618.
• Lau, A. (1988) A Moment of Anguish. Singapore in Malaysia and the Politics of Disengagement. Singapore: Times Academic Press.
• Letki, N. (2008) ‘Does Diversity Erode Social Cohesion? Social Capital and Race in British Neighbourhoods’. Political Studies 56(1): 99-126.
• Martin, B. (1981) A Sociology of Contemporary Cultural Change. Oxford: Blackwell.
• Martinović, Borja (2013) ‘The Inter-Ethnic Contacts of Immigrants and Natives in the Netherlands: A Two-Sided Perspective’. Journal of Ethnic and Migration Studies 39(1): 69-85.
• Miles, R. & Phizacklea, A. (1980) Labour and Racism. London: Routledge.
• Nowak, Martin A. (2012) ‘Why We Help’, Scientific American 307(1): 20-25.
• O’Leary, Brendan (2001) ‘An iron law of nationalism and federation? A (neo-Diceyan) theory of the necessity of a federal Staatsvolk, and of consociational rescue’, Nations and Nationalism 7(3): 273-296.
• Patsiurko, N., Campbell J. L. & Hall J. A. (2012) ‘Measuring cultural diversity: ethnic, linguistic and religious fractionalization in the OECD’. Ethnic and Racial Studies 35(2): 195-217.
• Putnam, R. (2000) Bowling Alone: The Collapse and Revival of American Community’. New York: Simon & Schuster.
• Putnam, R. (2007) ‘E Pluribus Unum: Diversity and Community in the Twenty-first Century'. Scandinavian Political Studies 30: 137-174.
• Salomon, M. (2003) ‘Masking Inequality in the name of rights: the examination of Fiji’s state report under the International Convention on the Elimination of Racial Discrimination”, Asia-Pacific Journal on Human Rights and the Law 1: 105-138.
• Schaeffer, M. (2012) ‘Which groups are mostly responsible for problems in your neighbourhood?’ Ethnic and Racial Studies 36(1): 156-178. DOI: 10.1080/01419870.2011.644311.
• Shils, E. (1950) ‘Primary Groups in the American Army’. In Merton, R. K. & Lazarsfeld, P. F. (eds.) Continuities in Social Research: studies in the scope and method of `The American Soldier’. Glencoe: The Free Press, 16-37.
• Smith, A. D. (2004) ‘Ethnic cores and dominant ethnies’. In Kaufmann, E. P. (ed.) Rethinking Ethnicity. Majority Groups and Dominant Minorities. London: Routledge, 17-30.
• Tajfel, H. (1981) Human Groups and Social Categories. Cambridge: Cambridge University Press.
• Ward, R. (1979) ‘Where race didn’t divide. Some reflections on slum clearance in Moss Side’, in Miles, R. & Phizacklea, A. (eds) Racism and Political Action in Britain. London: Routledge, 204-222.
• Weber, M. (1968). Economy and Society. An Outline of Interpretive Sociology. ed. Roth, G. & Wittich, C. (translations by various authors). New York: Bedminster.
• Weber, M. (1972) Wirtschaft und Gesellschaft. Grundriss der verstehende Soziologie, 5th edn. (ed.) Winkelmann J. Tübingen: Mohr Siebeck.
• Wimmer, A. (2004) ‘Does ethnicity matter? Everyday group formation in three Swiss immigrant neighbourhoods’. Ethnic and Racial Studies 27(1): 1-36.

Notes

1. Majority sentiment can exist whenever a population is divided into a majority and one or more minorities. There can be a religious, linguistic or political majority. This essay discusses the sentiment of ethnic majorities that can be distinguished when there is immigration by persons different in national or ethnic origin. It is concerned with the sentiment that differentiates the majority.

2. Staatsvolk is used here in a sense similar to that of O’Leary (2001: 285), for whom it designates ‘a national or ethnic people, who are demographically and electorally dominant’. This is different from its earlier use (Jellinek 1921: 406) to designate the entire population of a state.

3. This is surely the best translation possible of a badly worded pair of sentences, the first of which begins: ‘“ Vergemeinschaftung” soll eine soziale Beziehung heißen, wenn und soweit...’ Its first word means ‘the making of’; grammatically, it is the subject of the sentence, but this is not the meaning that the sentence conveys.

4. For example, the author once lived in a village in South Wales. Someone once asked him ‘Have you lived here long?’ He replied ‘Twenty years’, and waited to hear whether that was considered long, but got no answer. It was marginal. One section of the village population consisted of people who had been born in the locality, as had their parents. For some of them, twenty years was not ‘long’.

5. Sunder Katwala, commenting on an opinion poll prior to the 2012 Olympics, noted that while the findings highlighted a national anxiety about immigration, when people thought about their local areas, there was less concern. While 30 per cent placed immigration first when thinking about tensions facing British society as a whole, only 19 per cent chose it as the most divisive issue in their own area (Observer 13 January 2013; see also Jolley and Katwala, 2012).

6. During the British general election of 2010, the Prime Minister, Gordon Brown, visited Rochdale in Lancashire. He was introduced to a life-long Labour supporter who voiced some complaints, and, among other things, said ‘You can’t say anything about these immigrants. All these eastern Europeans what are coming in – where are they flocking from?’ After he returned to his car, Brown complained ‘You should never have put me with that woman ... She was just a sort of bigoted woman’. He still had a microphone attached to his lapel and the recording company released the audio to the press. There was a media frenzy that seriously damaged the Labour cause (Rawnsley 2010: 730-732). The Labour Party might have won more votes in that, and other, elections, had it presented itself as representing the ethnic majority over against the ethnic minorities.

7. By encouraging reciprocity, trust promotes cooperation and the growth of social capital. This has far reaching implications, extending even to response to global climate change and to organic evolution (Nowak 2012).

8. In 2014 there is to be a referendum in Scotland on independence. The campaign in favour of the maintenance of the present union is to argue that ‘we are stronger together’. The Scottish nationalist position resembles that of the Québécois presented by Juteau; it maintains that the scale of identification from primary group to global community is segmented. The unionist position is that the higher up this scale individuals co-identify, the greater the social capital.