Klein’s world of objects

One of Klein’s key contributions to psychoanalysis was object-relations theory, the idea that individual subjectivity develops through relations between internal and external objects. Internal objects, defined by John Bowlby, are ‘internal working models’. ‘Each individual builds working models of the world and of himself in it, with the aid of which he perceives events, forecasts the future, and constructs his plans’ (Bowlby Reference Bowlby1973, 203). External objects are the things the individual encounters in the world. Internal and external objects are constantly compared and altered in relation to each other through the individual’s experience and engagement in the world: Klein’s understanding of object relations is not simply about a series of adjustments between inner and outer worlds, but about the way in which each creates the other. Because of this, the self and the other are mutually constitutive, and relationships themselves are foundational of individual identity.

Klein argued that the process of constructing internal and external objects begins with the separation of aggression and love, and is then realized through their projection onto external objects. Thus, she discusses processes of splitting and projecting internal fears, fantasies, and emotions onto external objects that are encountered and reintrojected to form and reform internal objects. The processes begin with the new-born infant. At first, the infant cannot distinguish between itself and the outside world. It experiences inside and outside together, in bits, as wholly idealized good or aggressive part-objects. These engender either extreme feelings of love and security, or of hatred and destructiveness. The distinction between ‘inside’ and ‘outside’ remains unclear: it is as if the external objects are under the control of the baby, but also part of it. The baby experiences, attempts to control, and feels overwhelmed by good and bad objects all at once.

The tendency for the self (or ego) to split itself and its objects ‘occurs in part because the ego largely lacks cohesion at birth, and in part because it constitutes a defence against the primordial anxiety, and is therefore a means of preserving the ego’ (Klein Reference Klein1997a, 191). Projection allows a sense of distancing. ‘The fear of the destructive impulse seems to attach itself at once to an object – or rather it is experienced as the fear of an uncontrollable overpowering object’ (Klein Reference Klein1997b, 4). Destructive fantasies stemming from the innate death instinct are projected onto external objects that are then experienced as powerful and frightening. In this way, aggression is expelled and then re-encountered, coming now from external objects. At the same time, emotions of love and good, which arise from the life instinct, are projected onto external objects that are then encountered as loving. Klein argues that in this state that she calls ‘paranoid-schizoid’, splitting and projecting good and bad onto principally the mother leads the infant to experience ‘two’ mothers, one wholly good and one wholly bad, in a way that appears manageable (Klein Reference Klein1998a, 290–91).

Only towards the middle of the 1st year of life does the infant begin to put disaggregated objects together, to comprehend the ambiguity of whole objects, internal and external. The mother, for example, is gradually seen not as two mothers but one that is both loved and hated. With the realization of the ambiguity of objects, the child reaches a crucial stage in emotional development, which Klein called the ‘depressive position’, in which destructive fantasies in relation to vital good objects are acknowledged. Kristeva describes the depressive position as an acceptance of ‘psychic discomfort’. ‘Anxiety has not disappeared, as it is always present with Klein, but it chooses another domain: rather than splitting or fragmenting and rather than destroying and tearing into pieces, anxiety is tolerated as a source of pain relating to the Other and a source of guilt about having taken pleasure in hurting him’ (Kristeva Reference Kristeva2001, 89).

If the depressive position is acknowledged, the child can develop richer and more complex object relations, in which whole internal objects are reinforced by an engagement with the world. The acceptance of guilt, leading to a piecing together, repair, and introjection of ambiguous objects enables the child to develop more balanced, integrated relationships with the outside world, and an ongoing acceptance of ‘psychic discomfort’. Mature relationships in which the other is allowed to be complex and ambiguous thus enable and are enabled by an acceptance of the wholeness but also complexity of the self.

For Klein, this is not a ‘stage’ to be resolved once and for all, but a series of processes that can repeat throughout life.Footnote ⁴⁷ She traces the ways in which older children and adults continue to split apart internal objects and project them onto external objects as a way of protecting internal objects against their own feelings of aggression and destructiveness. ‘Bad’ objects can be hated more safely from a distance; ‘good’ objects can be idealized more plausibly. Thus, the world can sometimes appear to be populated by an assortment of good people and noble causes, alongside aggressive and hateful figures. Klein argues that such mechanisms are employed in order to keep intimate objects – those that are most closely bound up with the internal objects that constitute the self – safe from destructive feelings. ‘This division between love and hate towards people not too close to oneself also serves the purpose of keeping loved people more secure, both actually and in one’s mind. They are not only remote from one physically and thus inaccessible, but the division between the loving and hating attitude fosters the feeling that one can keep love unspoilt’ (Klein Reference Klein1998b, 330).

Anxiety caused by aggression and envy continue to exist to a greater or lesser extent throughout life and Klein thought that everyone on occasion resorts to defensive mechanisms where guilt is denied and envy projected through ‘projective identification’. Segal describes this as a process in which the individual ‘gets rid of whatever is a painful or bad experience by trying to project it outside into the object, thereby increasing the badness of the object. And similarly, parts of the self felt as good and constructive may be projected into a good object, making it more ideal but stripping the ego of its potential’ (Segal Reference Segal2006, 95, emphasis added). The ego’s potential lies, for Klein, in its acknowledgement of anxiety or psychic discomfort. In extreme cases, when anxiety is overwhelming, individuals can resort to a paranoid-schizoid position in which splitting and projecting ideal (good and bad) objects become the only way they can deal with themselves in the world.

A defining characteristic of idealized objects is their symbolic role. Making symbols is an essential part of discovering and understanding the world. It necessarily involves some splitting and shaping of external objects to make them fit and reinforce preconceived ideas (or internal objects). An acceptance of anxiety and uncertainty allows us to approach this task with flexibility, to recognize when external objects do not ‘fit’, and to constantly reassess and recreate symbols. Problems arise when anxiety is overwhelming and symbolization becomes fixed or rigid. ‘A part of the ego becomes identified with the object and, as a consequence, the symbol is equated with the thing symbolized. The symbol does not represent the object, but is treated as though it was the object’ (Segal Reference Segal1997, 43, emphasis added). As Alford suggests, such rigid symbolism means that, ‘we fail to learn all we could about the world because we force it into a framework determined by our phantasies and needs, giving the world less of an opportunity to modify these phantasies’ (Alford Reference Alford1989, 149).

Applying the object-relations lens to norms produces two observations. First, norms as an expression of an ideal represent a form of rigid symbolism – a posited universalism enables engagements with external objects in a literal relation to internal objects by subjecting them to an already perfect internal structure. Thus, norms seem to order and rationalize the world, making it appear more predictable, manageable, and less frightening. Through rigid symbolization, the other becomes an expression of the idealized self. Second, the power of norms, which are social creations, is magnified because they are joint enterprises that collect individuals together in a common cause. This reinforces the sense of the universal: the organization can be an expression of the individual writ-large, a super, more powerful individual. In identifying with such an organization, the individual is affirmed in its sense of itself in the world, and of the wider certainty of its objects. The role of IOs in pulling together an ideal sense of the individual writ-large, of projecting onto the world clear-cut and certain categories, provides psychological underpinning to individuals that belong to or identify with it.Footnote ⁴⁸

Relating this to the identity and well-being of individuals, the construction of norms’ objects and the constraints on organizations that promote them – it could be argued that defensive splitting and projecting of very idealized good and bad onto external objects produces two effects. First, it limits the subject’s potential for engaging with the complexities and ambiguities of the world, and therefore of realizing more complex internal objects. This has profound implications for the IO that serves up certainty in this way. Such a simplified, rigid division of the world into idealized objects suggests an investment in an idealized image for such an organization, and an inability for it to recognize its own ambiguous wholeness. Second, it forces external objects into rigid categories that are over-determined by the subject’s internal objects. Such misrepresentations work to reinforce the power disparities found in relationships between large powerful IOs and the countries they work with. In the next section I will consider these effects through an exploration of the World Bank’s good governance programme in sub-Saharan Africa.

Good governance: chopping Africa into bits

Good governance is an agenda of best political practice designed to promote economic development. The World Bank employs this norm in its work with developing countries, using it to define what has gone wrong with the role and work of the state, and to devise ways to make it work better. The concept was first developed in the Bank’s 1989 report, Sub-Saharan Africa: From Crisis to Sustainable Growth, as a response to the continent’s particularly poor development record throughout the 1980s (Maldonado Reference Maldonado2010, 4). Governance was presented as a major shift in the Bank’s approach to development, one that replaced its previous structural adjustment paradigm rooted in a more simplistic idea that states are bad and should be cut back, to one which argued that states are vital for development, that some states are good and others are bad, and that it is possible to turn bad states into good states.

In this section I do three things. First, drawing on its own policy literature and that of supporting organizations, I describe the World Bank’s understanding of good governance. Second, I discuss some of the critical literature on good governance that highlights the ways in which the Bank simplifies the political context within which it works. And finally, I bring object-relations theory into the discussion to explore the ways in which such simplifications are predicated on three discursive and ideational splits that create separate and idealized good and bad objects.

Governance has become the lever by which the World Bank identifies development failure and prescribes remedies for it. ‘The clamor for greater government effectiveness has reached crisis proportions in many developing countries where the state has failed to deliver even such fundamental public goods as property rights, roads, and basic health and education’ (World Bank 1997, 21). States with low capacity need to get the ‘fundamentals’ right, ‘without which sustainable, shared, poverty-reducing development is impossible’ (World Bank 1997, 23). The fundamentals include: ‘establishing a foundation of law, maintaining a non-distortionary policy environment, including macroeconomic stability, investing in basic social services and infrastructure, protecting the vulnerable and protecting the environment’ (World Bank 1997). The key priorities of public policy reform that dominate the good governance agenda are ‘reducing corruption, strengthening civil society, and reducing poverty’ (World Bank 2001).

Good governance has been a successful norm in terms of its uptake by a wide range of donors, including Western states, multilateral donors, and international non-governmental organizations (NGOs) (Abrahamsen Reference Abrahamsen2000, 48). Unlike the Bank’s previous structural adjustment policies, which remained contested, good governance has succeeded in normative terms by becoming the unquestioned prescription to promote development in poor countries. Such a shared agenda necessarily expresses a sense of a universal good, and the good governance approach does this by presenting the business of governance as a neutral, essentially technical exercise involving a set of institutions and practices that can be applied anywhere in the world. ‘A crucial part of the justification of the Bank’s activities is that economic relations are natural, hence aiding their development is not imposing its own conception of the good but rather assisting in what is a natural course of development’ (Williams Reference Williams1999, 81).

Together, donors attempt to transfuse these principles into outlier states, bringing them into the international fold through socialization. For example, Britain’s Department for International Development (DfID), an enthusiastic promoter of the good governance agenda, argues that good governance policies are designed to bring about ‘capable’ and ‘accountable’ governance, and that ‘this is largely contingent on getting the right kind of politics’ (DfID 2009, 5). And, as US Secretary of State, Hillary Clinton, argued on a visit to Nigeria: ‘the most immediate source of the disconnect between Nigeria’s wealth and its poverty is a failure of governance at the local, state, and federal level… Without good governance, no amount of oil or no amount of aid, no amount of effort can guarantee Nigeria’s success. But with good governance, nothing can stop Nigeria’ (Clinton Reference Clinton2009). Africa, as one of the most substantial targets of development funding and policy, has been subjected to good governance reforms for the best part of two decades because, from the perspective of donors, ‘institutions in Africa have not yet progressed sufficiently to make a significant contribution to development… [they] are not yet strong enough to permit countries to continue pursuing appropriate policies’ (Nsouli Reference Nsouli2001).

The longevity and vigour of the good governance programme in Africa has spawned an extensive critical literature that explores the effects of transplanting a supposedly universal norm onto African contexts (Williams and Young Reference Williams and Young1994; Williams 1993, Reference Williams1999; Abrahamsen Reference Abrahamsen2000; Weiss Reference Wiess2000; Kelsall Reference Kelsall2002; Mkandawire Reference Mkandawire2002; Harrison 2001, Reference Harrison2005, for example). In particular, this literature deals with the ways in which the World Bank and other donors have tended to simplify, ignore, or overlook the political complexities of African polities. Kelsall discusses how reforms, which appear to ‘take place in an air-conditioned, administrative bubble, isolated from the political atmosphere’, potentially disrupt deeply entrenched interests and political relationships (Kelsall Reference Kelsall2002, 605). In African political communities and wider societies, Harrison argues, ‘individuals are not isolated and utilitarian, but members of ethnically-defined networks of power and mutual support’ (Reference Harrison2005, 253) and as such are characterized by a ‘“rhizomatic” form of power that pervades society’ (Harrison Reference Harrison2005, 252). The simplifications of the good governance analysis and prescription come nowhere close to getting to grips with these complexities. Instead, it has been suggested that they better represent preoccupations of the donors, and in particular donors’ desire to formulate a ‘liberal universalism’; a ‘liberal project’ that seeks to remake Africa and Africans on supposedly neutral liberal lines (Williams and Young Reference Williams and Young1994; Williams Reference Williams1999).

Digging beneath the political analyses of good governance, in a way that seeks to complement rather than replace them, I now turn to the ways in which the good governance agenda can be understood within the context of object relations. I argue that good governance produces discursive and ideational splits in three distinct planes: between the donor and the recipient; between the recipient state and society; and between individual actors within the recipient state. In all cases, splitting results in the creation of apparently wholly good and wholly bad objects.

First, the good governance norm reifies a split between a principal, the donor ‘who wants “good policies”’ and the agents, the recipient regimes that are ‘deemed to be self-serving and corrupt’ (Mkandawire Reference Mkandawire2002, 15). This perception was the basis of a regime of conditionalities whereby recipient regimes were expected to shift economic and political policy in order to secure continuing aid (Mkandawire Reference Mkandawire2002). Mkandawire argues that in a whole range of donor approaches to Africa’s governance problems, this differentiation between donor and recipient is constantly reinforced: ‘In the culturalist view, African states… were depicted as hopelessly and incurably steeped in Africa’s debilitating culture… From an economics point of view, African institutions were infested with those who had captured state policies to serve their narrow interests [while] a more technocratic view questioned the analytic capacity and the bureaucratic acumen of the African state’ (Mkandawire Reference Mkandawire2002). Whether it was Africa’s ‘debilitating culture’, its self-interested elites, or the weak analytic capacity of its bureaucrats, African countries as the objects of development reform are decayed, corrupt, and incapable.

In such rigid symbolic thinking African states and political actors become straightforwardly bad, fitted into a category that reinforces the subject’s contrasting goodness. In a speech he made in Nigeria on good governance and the ways in which the World Bank was supporting transformation in Africa, then president Paul Wolfowitz related a discussion he had had with his daughter about how often ordinary Africans greeted him in public places.

Although most World Bank officials would adopt a less overtly self-congratulatory tone, Wolfowitz’s story distills an impression of ‘good works’ that oozes out of the organization. All of this conveys the sense of a very clean cut between donor and recipient, overlooking a long, and often fraught, history of relationships between African and Western actors.

Second, the good governance agenda creates a split between ‘good’ society and individuals, and ‘bad’ government. This is rooted in ‘an artificial dichotomy … between “state” and “market”’ (Weiss Reference Wiess2000, 803) where the market encompasses civil society, NGOs, and forms of public participation that can hold the state in check. Civil society is the primary focus, its organizations characterized by World Bank officials as ‘institutions that exert pressure and control as “watchdogs” on state institutions in the area of governance and development’ (Matovu Reference Matovu2002). In this depiction, civil society carries all the virtues – ‘participation, transparency, accountability, equity, effectiveness and efficiency, strategic vision, and good management’ (Matovu Reference Matovu2002) – while government carries all the vices – ‘corruption, lack of transparency and accountability, poorly designed decentralization, inadequate service delivery, and social exclusion’ (Parker Reference Parker2001). On one side, ‘the poor, needy, and powerless’; on the other, ‘the rich, replete and dominant’ (Parker Reference Parker2001). Once again, this approach denies the complexity outlined, for example, in Bayart’s discussion of Africa’s ‘rhizome states’ whose roots reach down into society (Reference Bayart2009).

As Abrahamsen points out, this discourse ‘constructs a binary opposition between an alien state intervention, which is associated with past development failures, and an “indigenous capitalism”, which represents the basis for future development successes’ (Reference Abrahamsen2000, 50). She argues that the World Bank, having flattened and idealized the ‘innate’ capitalist instincts of ordinary Africans, constructs ‘an essential unity of purpose … between the development apparatus and the “ordinary people”, in that they all oppose the state and seek to reduce it’ (Abrahamsen Reference Abrahamsen2000, 52). Civil society’s virtuous actors – who are represented as ‘undifferentiated and harmonious’ with ‘no classes, no races, no genders, [or] ethnic groups of oppressors’ (Abrahamsen Reference Abrahamsen2000, 56) – confront the inertia, corruption, and vested interests of government. ‘Citizens demand quality services, while city managers lack the resources and morale to perform’ (Matovu 2002). All the energy, initiative, and qualities of good governance are on the side of civil society, while government is, at best, tired and cash-strapped, and at worst ‘using communities to achieve selfish political and economic gains’ (Matovu 2002).

Third, the good governance agenda splits up actors within states, identifying and supporting ‘champions’ or ‘change teams’ (Harrison Reference Harrison2005, 247) in their purported struggle with ‘predatory local elites’ (Mkandawire Reference Mkandawire2002, 18). DfID has been particularly attached to this approach that has been called ‘Drivers for Change’ (DfID 2004) and is now known as ‘political economy analysis’ (DfID 2009). The approach aims to ‘get beneath the formal structures to reveal the underlying interests, incentives and institutions that enable or frustrate change’ (DfID 2009, 1) and increasingly ‘is being used to identify opportunities for leveraging policy change and supporting reform’ (DfID 2009, 5–6). It builds on a World Bank analysis that suggests that donors must get past the evidence of failure, to explore ‘the associated institutional set-up and governance arrangements’ and beneath these, the sources of inertia (Fritz et al. Reference Fritz, Kaiser and Levy2009). Political reform can be ‘driven’ by a handful of enlightened individuals often educated in the West who, with donor support, can combat inertia and bad governance and realize substantial policy reform (Harris Reference Harris2013). DfID, in advocating this approach, encourages its officials to ‘drill down’ to the ‘drivers which constrain or support progressive change’ (DfID 2009, 13).

This approach fits into broader Bank analyses that advocate simplified state structures, in which political and complex bureaucratic mechanisms are cut up into discrete units. Within state institutions, incentives are used to get individuals to do the right thing; and between state and society, transparency and freedom of information enables citizens to check and monitor individual representatives and officials. A further favourite theme is that of decentralization, explored in a 2002 Development Report, Building Institutions for Markets, that argued that decentralized institutions could ‘limit the state’s capacity for arbitrary action’ (World Bank 2002, 2). In this document, the theme of corruption ‘a tax which distorts the choice between activities and lowers the returns to public and private investments’ (World Bank 2002, 9) is addressed. We are told that complex regulation and closed trade, alongside centralized government, weak electoral frameworks, and a lack of information all feed corruption. Here is the strong sense that complexity, ambiguity, and the sense of opacity together erode governance. The World Bank instead advocates a pared down, simplified, and unambiguous state structure.

Such simplifications and flattenings make the programme of development appear more straightforward, but they cannot in fact make it more achievable. Harris, for example, notes that in the case of Sierra Leone, which has had intense involvement from DfID, there is little evidence that individual drivers for change working in government ministries are doing much more than exhausting themselves (Harris Reference Harris2013). A more open and flexible attempt to see what African societies are like would reveal a more complex picture in which state–society relations are thickly interwoven (Chabal and Daloz Reference Chabal and Daloz1999; Mbembe Reference Mbembe2001). And yet the Bank and other donors persist in their characterizations of individual Africans as virtuous entrepreneurs or victims who are ‘just like us’ (Gallagher Reference Gallagher2011).

The Bank’s approach does particular violence to relationships that are constantly deconstructed and simplified. Historical, social, and political complexity appear to lend too much muddiness to the straightforward way it wants to see the world; and so, the Bank chops the world up into bits through the norm of good governance. I have argued that this is possible because the norm is constructed as ideal, so disallowing room for uncertainty and ambiguity. The way in which this norm is viewed as both good and self-evident or ‘natural’ allows only one way to structure the world. This is an example of rigid symbolism, one in which the Bank creates stark and idealized objects that reflect a sense of the Bank itself and its relationships as ideal.