The Power of Labelling

How People are Categorized and Why It Matters

Rosalind Eyben, Joy Moncrieffe

• Publisher: Routledge
• Year: 2013
• Print ISBN: 9781844073955
• eBook ISBN: 9781136552519
• Edition: 1
• Topic: Economics
• Subtopic: Development Economics
• Index: Economics

The Power of Labelling illuminates a fundamental and intriguing dimension of social and political life. Striking cases from a range of policy contexts generate eyeopening analyses of labellings causes and consequences, uses and abuses, and of alternatives in thinking and relating.
DES GASPER, INSTITUTE OF SOCIAL STUDIES, THE HAGUE

The authors convincingly and often vividly explain how the unavoidable framings and labellings of the objects of policy secrete relations of power which can obscure as much as they reveal and often lead, in policy itself, to perverse outcomes. Their detail is riveting, their analyses persuasive, what they suggest realistic and deeply sensible. This immensely readable collection is indispensable for anyone who wants to think about how they think about 'development', and should be forced on all who dont.
GEOFFREY HAWTHORN, PROFESSOR OF INTERNATIONAL POLITICS, UNIVERSITY OF CAMBRIDGE

This is an essential book not only for those interested in understanding the development industry but also for development practitioners. It discusses key questions concerning the ways in which knowledge is generated by development agencies and reaffirms the importance of understanding who categorizes people, why and how.
R. L. STIRRAT, PROFESSOR OF SOCIAL ANTHROPOLOGY, UNIVERSITY OF SUSSEX

'Very important.'
Martin Kalungu-Banda, Oxfam GB

What does it mean to be part of the mass known as The Poor? What visions are conjured up in our minds when someone is labelled Muslim? What assumptions do we make about their needs, values and politics? How do we react individually and as a society? Who develops the labels, what power do they carry and how do such labels affect how people are treated?

This timely book tackles the critical and controversial issue of how people are labelled and categorized, and how their problems are framed and dealt with. Drawing on vast international experience and current theory, the authors examine how labels are constituted and applied by a variety of actors, including development policy makers, practitioners and researchers. The book exposes the intense and complex politics involved in processes of labelling, and highlights how the outcomes of labelling can undermine stated development goals. Importantly, one of the books principal objectives is to suggest how policy makers and professionals can tackle negative forms of labelling and encourage processes of counter-labelling, to enhance poverty reduction and human rights, and to tackle issues of race relations and global security. The Afterword encapsulates these ideas ands provides a good basis for reflection, further debate and action.

1

Labels, Welfare Regimes and Intermediation: Contesting Formal Power

Geof Wood¹

Introduction

In revisiting my earlier arguments about the ‘politics of development policy labelling’ (Wood, 1985b) for this volume, this chapter focuses upon the general relationship between processes of categorization and forms of intermediation. The argument starts with a review of the context for the original labelling thesis, and a summary of the ‘labelling as political manipulation’ argument that dominated that original paper. This leads into a brief overview of the relevant development and sociological discourses that followed those earlier arguments and thus an autocritique built around the limitations of the hegemonic, statist assumptions of authoritative labelling. This reflection sets up the basis for a revised argument which recognizes the greater significance of plurality and contestation in the labelling process as a way of understanding how formal power is either directly challenged or by-passed in societies where the exercise of informal, less bureaucratically configured power prevails.

This is the basis of deploying a comparative welfare regimes approach (Gough et al, 2004; Wood and Gough, 2006) to capture more systematically the variation in forms of intermediation, as informed by labelling, through which power is exercised and through which people have to pursue their livelihoods and well-being. The central feature of this welfare regimes framework is the relationship between rights, claims and correlative duties, and how these vary between different welfare regimes. It observes that scarcity is managed in different ways in different regimes through variations in the process of intermediation between rights and claims on the one hand, and correlative duties on the other. It also observes that the model of bureaucratic rationality characterized by authoritative labelling only applies successfully to societies where the state is sufficiently legitimate to perform both de-commodification and regulatory functions over the market, as well as community and household institutions. Within that notion of legitimacy is the widespread acceptance of the practices of bureaucratic rationality in classifying need and targeting resources to those needs.

However, in societies where these principles of the welfare and developmental state do not obtain, then the relationship between rights, claims and correlative duties is not governed by bureaucratic and authoritative labelling. Thus we enter a range of situations that will be schematically outlined. A contrast is used between simple and dynamic reproduction in order to distinguish between situations of strong path dependency and thus simple reproduction through the domination of uncontested state categories of rights, and situations of weaker path dependency, characterized by plurality and/or contestation, entailing prospects for dynamic reproduction – positively or negatively in terms of the well-being of the powerless. The plurality of authoritative labelling refers to what elsewhere is termed ‘informal security regimes’, where the domain of policy and state implementation is more obviously obliged to compromise with the hierarchy of intermediary actors who de facto but not de jure command the relationship between rights, claims and correlative duties through forms of patronage and other informal practices that nevertheless entail the management of scarcity through the informal prioritization of needs. The ‘contestation’ end of that range is where the state is struggling but failing to establish the authoritativeness of its labelling over the rest of society – leading to various forms of contestation and subversion.

The context for earlier labelling theses

What was the point of departure for the original arguments?² They were several. Empirically, they were an instinctive counter-reaction to the practices of targeting, or even extreme targeting, which also required the convincing of other needy people outside the target that those so targeted were legitimately within it. Interestingly, in the late 1970s/early 1980s (partly as a continuation of Basic Needs discourses) it was the ‘progressive’, poverty-oriented like-minded Scandinavian, Dutch and Canadian donors along with international non-governmental organizations (INGOs) and domestic NGOs (DNGOs) who pursued targeting in mass poverty societies like Bangladesh. The Overseas Development Administration (known since 1997 as the UK Department for International Development (DFID)) was, at that time, out of the loop, still committed to non-targeted programme aid. While targeting the poorest appeared to be progressive (e.g. in rich, but unequal, western societies), in the context of mass poverty it could be understood as regressive in the sense of actually excluding the needy. That problem remains located in the contemporary micro-classifications of poverty (chronic, extreme, hard-core etc.) for policy focus, as illustrated in DFID’s current poverty-focused programmes in Bangladesh. Ideologically and politically, although Schaffer was my guru, we approached these issues via a tension between Schaffer’s critical Weberianism and my Marxism in the way the state should be analysed. However, both of us had written about ‘access’ (i.e. state–society relations at the interface of service provision and resource allocation) in the late 1970s, from our respective positions (i.e. for Schaffer, the mechanics of bureaucratic rationing via queues, interface and encounters; and for Wood in terms of the exercise of inequality, rooted in political economy, together with the social incompatibility of bureaucratic and peasant rationalities). Meanwhile I had been reading Althusser, Foucault and post-structuralists, as well as remembering Gramsci, Dahl and Lukes. So the theoretical convergence between Schaffer and Wood focused upon a frustration with the contemporary form of Marxian discourse about the state, which was silent on the actual processes of power amid the formal assertions that the ‘state acted’ in either fully captured or relatively autonomous ways that were necessarily consistent with the interests of prevailing dominant classes. So we were interested in the unasked questions about ‘how’ the state might serve the interests of some to the exclusion of others. Our entry point into this ‘how’ question was therefore the process of labelling, as a fundamental activity of exercising power. Althusser wrote about ideological apparatuses of the state. Foucault about hidden, unobserved power expressed through repeated, normalized technique. The post-structuralists nevertheless remained gloomy about agency, seeing it as always overridden. So our work was intended to reveal these hidden, insidious dimensions of power, where authoritative, ‘scientific’ technique is used to de-politicize an essentially political process of resource allocation and management of scarcity through the realization of conformity to labels that indicated the distribution of rights to entitlements. We settled upon this entry point as, in effect, the next step in our joint earlier interest in access. But in pursuit of this dimension of power, we thus entered a world of shadows, illusions and disguises. And arguably, all large organizations operate with these characteristics, which reflect internal power configurations as well as organizational power over others.

Labelling as a fundamental social process

The acts of classification and taxonomy are rather fundamental to human behaviour and interaction. If we consider the world around us as constructed by concentric circles of increasing moral distance, then we increasingly rely upon our skills and memories of classification as our relationships move from inner to outer circles - that is from intimate kin and friends to strangers, from multi-dimensional to single-dimensional transactions, from gemeinschaft to gesellschaft. Of course, many things intrude into these processes of classification for personal survival: values, interests, preferences and learning from repeated interactions. Continuous adjustments to our taxonomies are made through symbolic processes of interaction. In this way, relationships can settle down to a pattern, and do not have to be derived from first principles each time, which would be too costly and insecure for functional interaction. To this social convenience of labelling as a proxy for unique and primary assessments must be added ‘power’. It is of course everywhere, when two or more persons interact. For interactions towards the outer circles, power is more institutionalized rather than the idiosyncratic outcome of personalities in interaction. But of course, even within the immediate family, age and gender provide non-idiosyncratic accounts of power. From this we can understand that the interesting question is not whether we label and categorize. We all do that, as asserted above. Rather, the interesting questions are which and whose labels prevail, and under what contextual conditions? These ‘which’, ‘whose’ and ‘what’ questions become more significant as we move to the outer circles, because these are the transactions more in the public than private domain. The public domain is one of institutionalized power within a wider framework of political economy, within which policies (through deliberation or default) are constructed to allocate resources and opportunities under conditions of overall scarcity. Such policies and their outcomes are an inextricable aspect of the power of labelling – the process of classifying needs and entitlements. And the interesting question here is whether that labelling is transparent and the result of open political competition, or whether it is hidden and arbitrarily imposed upon an unconvinced population. However, this is an extremely complex question to answer. Labelling in Development Policy (1985a) sought to answer this question both theoretically and ontologically as well as through case study application.

Labelling as political manipulation: Arguments from the 1980s

While, as argued above, labelling, categorization and classification is an intrinsic component of human agency, this is not the place to survey the entire breadth of labelling in all human interaction. Thus the interest here lies in a sub-set of the labelling process that pertains to prioritizing claims to welfare. Sometimes these claims will be understood as rights, and sometimes as effective demand. When understood as rights, the discourse of labelling will concern universal and moral concepts of need, deserving, targeting, inclusion/exclusion, prioritizing and queuing for access. When understood as effective demand, although an implicit list of similar qualifications may be deployed, there will also be the dimension of effectiveness of voice, meaningful threats of disloyalty, and realistic exit options that might harm resource controllers and service providers. There is, therefore, a tension between labelling as a hidden political process of technique, having recourse to ‘science’ for legitimacy and authoritativeness, and labelling as a negotiated, more obviously political process, reliant upon contingent settlement, always vulnerable to change. The central proposition is that the process of labelling is a relationship of power, in that the labels used by some sets of actors are more easily imposed upon a policy area, upon a situation, upon people as classification than those labels created and offered by others.

Labelling is a pervasive process, occurring at different levels and within different arenas of interaction. So, not just between the state and people in the society, but between people through constructions of social othering and identity creation. We are all labellers, and therefore we are all in turn labelled. Thus we abstract from the individual, the actuality, and then stereotype via the use of metaphor. All interaction requires labelling in the form of images, badges, stereotypes and metaphors which as signals guide perceptions and thus interactional behaviour. The power issue is expressed in terms of whether the individual controls the presentation of self-image, or receives and lives within the images imposed by others.

The original paper concluded, therefore, that the issue is not whether we label, but which labels are created, and whose labels prevail to define a whole situation or policy area, under what conditions and with what effects? Applied to the analysis of the state, and more particularly for this chapter an analysis of welfare state regimes, we have to ask how specific sets of labels become universalized and legitimized instead of some other set. How does one set become authoritative at the expense of other options and choices? This is the crucial insight into political process. Accepted or authoritative labelling is the entry point into understanding the political settlement that underpins stable social policy. This is to be contrasted to unsettled political circumstances when labelling is far more contested. Thus, authoritative labelling represents the conclusion or outcome of political settlement, when historic agreements have been reached between contending classes, ethnic and linguistic groups, genders and generations. While such agreements are not set in stone, their basic premises and assumptions are difficult to shift radically. Simple reproduction is more likely than dynamic, or extended reproduction. Thus the notion of political settlement reflects a situation of induced consensus where each potentially contending party and advocate of change also calculates the odds of achieving any significant improvement as remote and likely to put present, albeit inadequate, entitlements at risk. Thus were revolutions always contained. This is how political settlements can reflect highly unequal social and economic conditions, as in the UK.

The process whereby acceptance is gained is assisted by ‘politics appearing as technique’. This has been the contribution of Foucault. The authoritative labels of the state, and thus political settlements, are buttressed by the activities of science and the rationality assumed therein. Social sciences, especially in the forms of social policy and development studies, are essentially in the business of arranging people in different classifications and taxonomies for the purposes of data comparison to explain key variables in behaviour. Thus science, rationality and expertise appear as apolitical technique making the underlying assumptions about classification, arranging, grouping for the purposes of data comparison and policy justification unassailable in political debate. Grouping and classification is all about boundaries and thresholds, and where they are to be set for the purposes of attributing significance. While regression analysis offers more flexibility in terms of attributing significance to linear options, and thus more independence from the terms of the original question, it does not remove the arbitrariness of original category selection. A good example of politicized category selection has been Sida in Bangladesh when, in the late 1980s, it attempted to target the extreme poor (a different concept and label from chronic) by using <0.5 acres as the divider between included and excluded families for targeted benefits in a village ‘para’.³ But such an arbitrary snapshot approach bore little or no relation to the experience of being poor in those paras.

This point sets up a key issue – the extent to which the authoritativeness of a label is undermined by lack of self-evident fit to the condition of the labelled. If we take the Sida case again: let us assume that 30 families live in a para of a Bangladeshi village. As nuclear entities, these families are ‘paribars’. But these nuclear entities are frequently grouped into immediate, perhaps extended, kinship groups as ‘baris’ – perhaps with dwellings surrounding and facing into a common courtyard where some activities are done together. While all these paribars may share poverty, for various reasons of multiple inheritance and debt circumstances they have variations in control over land at any one snapshot in time. But they know that the poorest family today was better off yesterday, and those who are coping today could be in a rapid downward trajectory tomorrow. In other words they are all in a livelihoods process, improving, coping or declining at any one point in time but always highly vulnerable to crisis round the corner. And each family knows this of each other. What sense then in trying to differentiate between them for targeting purposes on the basis of only one variable – land control? That control is so precarious for all. Can the label be imposed? Yes. But was it authoritative in the sense of being a self-evident and valid discriminator locally? No.

What has happened in this example? The agrarian economy has been understood too strongly in terms of land access and ownership as the prime determinant of livelihoods success. Thus landlessness becomes a key policy concept, as indeed it has been for three decades in the Bangladeshi discourse. From among the many roles, and thus including the many ways of earning a living, the land-owning variable has been plucked as indicating everything else about a family’s livelihoods prospects. Behavioural assumptions flow from this indicator. No account is taken of how a family may have entered or will exit from this condition. No attention to routes, in other words. Instead, the individual has been transformed into a client (i.e. the policy target) by being differentiated and disaggregated into components, and then identified with one component, with one principal label as the insight into the whole condition. The individual has thereby been transformed into an object, into a ‘case’ and de-linked from his/her own story. The greater the separation of the case from the story, the more the tendency away from self-evidence in terms of label applicability, and thus this separation is an index of power for the possessor of the case. Taking the Sida example of targeting in Bangladesh, the political significance of de-linking lies in severing the target families from their social context, breaking identities to kin, clan and neighbourhood, and re-establishing identity on the basis of the family’s relationship to or dependency upon an actual or potential category of state activity.

To those in power in unequal political economies, poverty is best conceptualized as behavioural rather than structural in order to separate the rich and powerful from responsibility for poverty through exploitative relations of production and exchange. This translates into behavioural rather than structural labels to designate the poverty problem and politically disorganize the poor through atomizing the causes of their condition. Such conceptualization underpins policy and strategy, directing it towards activity which is weakly linked or de-linked by this ideological representation to historical systems of unequal exchange. Thus the poor become labelled through other self-incriminating badges: beggars, street urchins, itinerants, refugees, slum-dwellers, lazy, incompetent and so on. In this way, we see the de-linking of individuals from their histories, e...

Table of contents

• Front Cover
• Title Page
• Copyright
• Contents
• About the Contributors
• Preface and Acknowledgements
• List of Acronyms and Abbreviations
• Introduction. Labelling, Power and Accountability: How and Why ‘Our’ Categories Matter
• 1 Labels, Welfare Regimes and Intermediation: Contesting Formal Power
• 2 Labelling People for Aid
• 3 The Politics of Representing ‘the Poor’
• 4 Disjunctures in Labelling Refugees and Oustees
• 5 When Labels Stigmatize: Encounters with ‘Street Children’ and ‘Restavecs’ in Haiti
• 6 Poverty as a Spectator Sport
• 7 ‘Muslim Women’ and ‘Moderate Muslims’: British Policy and the Strengthening of Religious Absolutist Control over Gender Development
• 8 Black Umbrellas: Labelling and Articulating Development in the Indonesian Mass Media
• 9 Labelling ‘Works’: The Language and Politics of Caste and Tribe in India
• 10 Exploring the Intersection of Racial Labels, Rainbow Citizenship and Citizens’ Rights in Post-Apartheid South Africa
• Afterword: Changing Practice
• Index