Content uploaded by Tom De Herdt
Author content
All content in this area was uploaded by Tom De Herdt on Jul 08, 2014
Content may be subject to copyright.
REAL GOVERNANCE BEYOND THE ‘FAILED
STATE’: NEGOTIATING EDUCATION IN THE
DEMOCRATIC REPUBLIC OF THE CONGO
KRISTOF TITECA AND TOM DEHERDT*
ABSTRACT
In the Democratic Republic of Congo (DRC), the state administration has
retreated from much of the public domain. The specific case of the education
sector –a domain traditionally reserved for the state –shows how public services
continue to be provided, and how the Congolese state continues to survive and
transform itself. Although no overall power governs the system –there is no
overall regulatory authority –this does not mean that the education sector is
ungoverned. The state survives as an administrative framework whose role in pro-
viding public services has been redefined rather than evaporated. This article
describes the organization of the educational system as the direct result of an evol-
ving negotiation process between state and non-state actors. It shows how this
negotiated nature of statehood, and the power differentials between the various
actors, involve constant renegotiation. Instead of producing uniform results within
the education sector, this form of regulation depends on power configurations in
particular localities at particular times.
AN ARCHETYPICAL CHARACTERISTIC OF ‘FAILED STATES’is their retreat
from the public domain and more particularly their inability to provide
basic public services. These failed states are seen as a ‘vacuum of auth-
ority’in which the state is nothing more than ‘a mere geographical
expression, a black hole into which a failed polity has fallen’.
1
The
*Tom de Herdt (tom.deherdt@ua.ac.be) lectures at the Institute of Development Policy and
Management, University of Antwerp. Kristof Titeca (kristof.titeca@ua.ac.be) is a postdoc-
toral fellow of the Research Foundation –Flanders (FWO), based at the Institute of
Development Policy and Management, University of Antwerp. The research for this article
was part of an inter-university research project on ‘Post-conflict reconstruction in the DRC’,
funded by the Belgian Federal Science Policy, and of a research project funded by the UK
Department for International Development on primary education in the DRC. The authors
would like to thank everyone who participated in these projects. Special thanks go to Jean-
Pierre Mbwebwa, Johan Verhaghe, Jean-Pierre Olivier de Sardan, Marc Poncelet, Sonia
Mrsic-Garac, Inge Wagemakers, Géraldine André, two anonymous reviewers, and the
editors of this journal for their comments. The usual disclaimers apply.
1. Robert Rotberg, ‘Failed states, collapsed states, weak states: causes and indicators’in
Robert Rotberg (ed.), State Failure and State Weakness in a Time of Terror (World Peace
Foundation, Cambridge, MA, 2003), p. 21, cited in Pierre Englebert and Dennis Tull,
African Affairs, 00/00, 1–19 doi: 10.1093/afraf/adr005
© The Author 2011. Published by Oxford University Press on behalf of Royal African Society. All right s reserved
1
African Affairs Advance Access published February 25, 2011
at Universiteitsbibliotheek Antwerpen on March 31, 2011afraf.oxfordjournals.orgDownloaded from
Democratic Republic of Congo (DRC) is seen as a prime example of a
failed state: a ‘forsaken black hole characterized by calamity, chaos, con-
fusion, and a bizarre form of social cannibalism where society is its own
prey’.
2
In this ‘forsaken black hole’, public services seem almost utopian:
the state has to a large extent retreated from its basic functions, most
notably the provision of security,
3
but also other public policy domains.
Education is one of them: by the year 2000, public financing of education
in the DRC had virtually ceased. Yet, despite this collapse in public finan-
cing, the education sector continues to function. This article aims to show
how, even in the context of a so-called ‘failed state’, public services are still
being organized and provided. Our analysis focuses on the education
sector in the DRC. In response to state incapacity in certain (policy) areas,
empirical solutions have been negotiated through which public services
continue to be provided. The article offers empirical support to earlier
theoretical findings of authors like Shahar Hameiri, who criticize the failed
state concept for being a failed paradigm and instead argue for a descrip-
tion of the ‘real governance’at play in a particular sector.
4
The specific case of the education sector enables a demonstration of how
the Congolese state continues to survive and transform itself. As an admin-
istrative framework the state has never ceased to exist, and its role in pro-
viding public services has been redefined rather than having evaporated.
This article describes the basic organization of the educational system as
the direct result of an evolving negotiation process between state and non-
state actors. From this perspective, formal state-backed rules and agree-
ments only play an indirect role in governing the system, even if all actors
in the sector systematically refer to these rules and agreements.
The article draws on field research in 2008 and 2009 by the two
authors in several provinces, complemented by secondary information on
the sector as a whole. A broad range of actors within the education sector
were interviewed: donor officials, local and national government officials
‘Post-conflict reconstruction in Africa: flawed ideas about failed states’,International Security
32, 4 (2008), p. 125.
2. Theodore Trefon, Reinventing Order in the Congo: How people respond to state failure in
Kinshasa (Zed Books, London, 2004), p. 1.
3. Koen Vlassenroot and Timothy Raeymaekers, ‘The politics of rebellion and intervention
in Ituri: the emergence of a new political complex?’,African Affairs 103, 412 (2004), pp. 385–
412; Koen Vlassenroot and Timothy Raeymaekers, Conflict and Social Transformation in Eastern
DR Congo (Academia Press, Gent, 2004); Stephen Jackson, ‘Borderlands and the transform-
ation of war economies: lessons from the DR Congo’,Conflict, Security and Development 6,3
(2006), pp. 425–47.
4. Shahar Hameiri, ‘Failed states or a failed paradigm? State capacity and the limits of
institutionalism’,Journal of International Relations and Development 10, 2 (2007), pp. 122–49;
on real governance, see Jean-Pierre Olivier de Sardan, ‘Researching the practical norms of
real governance in Africa’(Discussion Paper 5, Africa, Power, and Politics Programme,
Overseas Development Institute, London, December 2008), p. 2.
2AFRICAN AFFAIRS
at Universiteitsbibliotheek Antwerpen on March 31, 2011afraf.oxfordjournals.orgDownloaded from
(including inspectors and other education officials –see details of the
different agencies below), teachers, head-teachers, and parents. We start
with an overview of the literature on real governance as different from –or
broader than –state governance. While different authors concur in high-
lighting this distinction, we propose to enrich the analysis of real govern-
ance as the result of a negotiation process by connecting it to the
literature on organizations as negotiated orders, sustained by legitimate
practices. We then introduce the case of the education sector in DRC.
The case documents both the retreat of the state and the expansion of sta-
tehood to include an increasing array of actors. Then we discuss some
case studies of recent events to nuance the claim that the education sector
in DRC can best be seen as a negotiated order.
Producing public services between state and society
Much has been written about the weak, failed, or collapsed state in
Africa, and its weakened capacity to exercise authority and provide public
services.
5
The idea of a Weberian state providing the full array of public
services and goods in developing countries has proved to be simply
utopian. A variety of terms have been used to describe this situation:
‘weak state’,‘state collapse’,‘failed state’and ‘shadow state’.
6
Much
attention is given to what there is not; by applying fixed, ahistorical
Weberian definitions of statehood, particularly negative characteristics are
attributed to these African nations.
7
However, the fact that the state is
‘weak’or ‘failed’does not mean that a void exists in its place.
8
As Tobias
Hagmann and Didier Péclard argue, instead of analysing the state with
reference to the Weberian ideal-type, it is necessary to focus on a more
empirically grounded understanding of the state, or on ‘real governance’
as Olivier de Sardan calls it.
9
This may help us to understand the
5. Patrick Chabal and Jean-Pierre Daloz, Africa Works: Disorder as political instrument
(James Currey, Oxford, 1999).
6. Robert H. Jackson and Carl G. Rosberg, ‘Why Africa’s weak states persist: the empirical
and the juridical in statehood’,World Politics 35, 1 (1982), pp. 1–24; Jennifer Milliken and
Keith Krause, ‘State failure, state collapse and state reconstruction: concepts, lessons and
strategies’in J. Milliken (ed.), State Failure, Collapse and Reconstruction (Blackwell, Malden,
2003), pp. 1–24; William Zartman, Collapsed States: The disintegration and restoration of legiti-
mate authority (Lynne Rienner, Boulder, CO, 1995); Robert Rotberg, ‘Failed states in a
world of terror’,Foreign Affairs 81, 4 (2002), pp.127–40; William Reno, Warlord Politics and
African States (Lynne Rienner, Boulder, CO, 1998).
7. Jonathan Hill, ‘Beyond the other? A postcolonial critique of the failed state thesis’,
African Identities 3, 2 (2005), p. 139.
8. Thomas Bierschenk and Jean-Pierre Olivier de Sardan, ‘Local powers and a distant
state in rural Central African Republic’,Journal of Modern African Studies 35, 3 (1997),
pp. 441–68, p. 441.
9. Tobias Hagmann and Didier Péclard, ‘Negotiating statehood: dynamics of power and
domination in post-colonial Africa’,Development and Change 41, 4 (2010), pp. 539–62;
Olivier de Sardan, ‘Researching the practical norms of real governance in Africa’.
REAL GOVERNANCE BEYOND THE ‘FAILED STATE’3
at Universiteitsbibliotheek Antwerpen on March 31, 2011afraf.oxfordjournals.orgDownloaded from
‘different, shifting, and uncertain’processes through which the state
establishes itself.
10
In understanding statehood, attention must be given not only to state
institutions but to the whole spectrum of formal and informal actors in
the ‘field of power’around state institutions.
11
In essence, when analysing
governance in particular localities, one has to leave behind implicit state-
centric views of what constitutes governance, rather defining it as ‘an
emergent pattern or order of a social system, arising out of complex nego-
tiations and exchanges between “intermediate”social actors, groups,
forces, organizations, public and semi-public institutions in which state
organizations are only one –and not necessarily the most significant –
amongst many others seeking to steer or manage these relations’.
12
In this
context, Joel Migdal and Klaus Schlichte focus on ‘doing the state’,
which involves ‘the diverse, multiple actions of state actors as well as the
myriad responses and interactions between state officials and non-state
actors’.
13
These state practices are to be distinguished from the idea or the
image of the state as ‘a coherent, controlling organization in a territory,
which is a representation of the people bounded by that territory’.
14
Likewise, Christian Lund argues that in Africa,
it is difficult to ascribe exercised authority to the ‘state’as a coherent institution; rather,
public authority becomes the amalgamated result of the exercise of power by a variety of
local institutions and the imposition of external institutions, conjugated with the image of
a state. Hence the practice of governance varies from place to place, and even from field
to field.
15
The Weberian ideal-type of the state as a goal-oriented, centralizing, and
unitary actor which is distinct from society should be replaced, therefore,
by a view which pays attention to the different alliances and connections
with non-state actors who play an important role in the formation of
public authority.
16
Earlier, Achille Mbembe reworked Max Weber’s
10. Bruce Kapferer, ‘New formations of power, the oligarchic-corporate state, and anthro-
pological ideological discourse’,Anthropological Theory 5, 3 (2005), pp. 285–99, p. 286.
11. Joel S. Migdal and Klaus Schlichte, ‘Re-thinking the state’in K. Schlichte (ed.), The
Dynamics of States: The formation and crises of state domination (Ashgate, Burlington, VT,
2005), p. 15. See also Bierschenk and Olivier de Sardan, ‘Local powers and a distant state’;
Thomas Bierschenk and Jean-Pierre Olivier de Sardan, ‘Powers in the village: rural Benin
between democratisation and decentralisation’,Africa 73, 2 (2003), pp. 145–73.
12. Nikolas Rose, Powers of Freedom: Reframing political thought (Cambridge University
Press, Cambridge, 1999), p. 21 cited in Timothy Raeymaekers, Ken Menkhaus, and Koen
Vlassenroot, ‘State and non-state regulation in African protracted crises: governance without
government?’,Afrika Focus 21, 2 (2008), p. 9.
13. Migdal and Schlichte, ‘Re-thinking the state’, pp. 1–40.
14. Ibid., p. 15.
15. Christian Lund, ‘Twighlight institutions: public authority and local politics in Africa’,
Development and Change 37, 4 (2006), pp. 685–705, p. 686.
16. Joel Migdal, State in Society (Cambridge University Press, New York, NY, 2001),
pp. 13–15.
4AFRICAN AFFAIRS
at Universiteitsbibliotheek Antwerpen on March 31, 2011afraf.oxfordjournals.orgDownloaded from
concept of ‘discharge’to coin the term ‘indirect private government’–
where functions that are presumed ‘to be public, and [are based on] obli-
gations that flow from sovereignty, are increasingly performed by private
operators for private ends’.
17
Béatrice Hibou in turn uses the term ‘priva-
tization’, in which there is an ‘extended use of private intermediaries for
an increasing number of functions previously devolving upon the state,
and redeployment by the state’.
18
Why is it that public authority is exercised by an amalgam of actors
rather than by the state institutions in the strict sense? A first explanation
looks at representatives of the state itself. The way in which Hibou
describes the phenomenon clearly implies that the agency in the privatiza-
tion process lies with the state: the state takes the decision to ‘contract
out’particular services to non-state actors. In weak state environments
perhaps more than elsewhere, representatives of the state have an interest
in transforming their authority based on sheer physical force into a
relationship based on recognition and obedience –what Max Weber con-
ceived of as the transformation of power into domination.
19
In ‘doing’the
state, its representatives find themselves in ‘an already established network
of powers and representations’
20
and are forced into accepting certain
practices. In other words, by allying with non-state actors, it becomes
easier to get things done. However, a second explanation argues that non-
state actors also have an interest in allying with the state and take up a
role in exercising public authority, something illustrated by Charles Tilly’s
concept of ‘brokered autonomy’. According to this concept, leaders of
what he calls ‘trust networks’‘yield resources and compliance to rulers in
return for significant autonomy within their own domains’.
21
In this case
the agency primarily lies with non-state actors, which govern particular
territories or domains, and with whom the state allies itself –or rather, is
forced to ally itself –to extend its rule over these territories or domains.
The term ‘negotiated state(ness)’
22
does not place agency on either
side: it underlines the ‘co-existence of multiple public authorities’linked
to ‘multiple parcels of authority’within the public sphere, each giving its
17. Max Weber, General Economic History (Collier Books, New York, NY, 1961),
pp. 54–63; Achille Mbembe, On the Postcolony (University of California Press, Berkeley,
CA, 2001) p. 80.
18. Béatrice Hibou, ‘From privatizing the economy to privatising the state’in B. Hibou
(ed.), Privatizing the State (Columbia University Press, New York, NY, 2004), p. 3.
19. Hagmann and Péclard, ‘Negotiating statehood’, p. 543.
20. Michel de Certeau, The Practice of Everyday Life (University of California Press,
Berkeley, CA, 1984), p. 18.
21. Charles Tilly, ‘Trust and rule’,Theory and Society 33, 1 (2004), pp. 1–30, p. 14.
22. As described, among others, by Ken Menkhaus, ‘Governance without government in
Somalia: spoilers, state building and the politics of coping’,International Security 31,3
(2006), pp. 74–106; Raeymaekers et al.,‘State and non-state regulation’; Hagmann and
Péclard, ‘Negotiating statehood’.
REAL GOVERNANCE BEYOND THE ‘FAILED STATE’5
at Universiteitsbibliotheek Antwerpen on March 31, 2011afraf.oxfordjournals.orgDownloaded from
own meaning to authority and political power.
23
The expression under-
lines the need for permanent negotiation between different ‘power
poles’
24
and indicates how provision of public service depends on various
negotiation processes with non-state authorities, to fulfil certain govern-
mental functions.
25
In other words, the central idea is that statehood is
being negotiated between the state and a number of non-state actors.
Clearly both the notion of the negotiated state and state formation
through state–society interactions are not only typical of ‘fragile’or con-
flict states; in all countries the state is negotiated with and through
society.
26
As Hagmann and Péclard emphasize, this negotiation does not occur
between equal parties, but among heterogeneous groups with different
assets and entitlements, both material and symbolic. Material assets
include money, weapons, access to land, and so on. They also point to the
role of symbolic assets, defining them as ‘the ability to draw on social and
cultural repertoires in order to give meaning to one’s actions’.
27
At this
point one can also connect the concept of negotiated statehood to earlier
literature on organizations as negotiated orders.
28
Drawing on Bourdieu,
among others, the central claim is that besides the individual attributes
social actors bring to the negotiation process, the outcome of that process
is ultimately determined by their symbolic power, the ‘power to define the
situation in which the interactions that comprise the negotiated order take
place …Symbolic power is typically deployed to further entrench the
reality that defines as valuable the practices that are the basis of legiti-
macy.’
29
This symbolic power is itself negotiated, as it is socially con-
structed by the parties involved in the process and ultimately dependent on
their agreement. The particular ‘negotiation context’, or what Hagman
and Péclard would define as the ‘negotiation arena’, results from previous
23. Lund, ‘Twilight institutions’, p. 694, cited in Raeymaekers et al.,‘State and non-state
regulation’, pp. 9–10. See also Karel Arnaut and Christian Højbjerg ‘Gouvernance et ethno-
graphie en temps de crise’,Politique africaine 111, pp. 5–21.
24. Bierschenk and Olivier de Sardan, ‘Local powers and a distant state’, p. 441.
25. Raeymaekers et al.,‘State and non-state regulation’, p. 17. Koen Vlassenroot and
Timothy Raeymaekers, ‘New political order in the DR Congo? The transformation of regu-
lation’,Afrika Focus 22, 2 (2008), pp. 39–52.
26. Tom de Herdt and Marc Poncelet (eds), ‘Enjeux et acteurs autour de la réduction des
frais scolaires en RDC’(unpublished DfID consultancy report, London, 2010), p. 23. See
also Hagmann and Péclard, ‘Negotiating statehood’; Migdal, State in Society; Migdal and
Schlichte, ‘Re-thinking the state’.
27. Hagman and Péclard, ‘Negotiated statehood’, p. 549.
28. Anselm Strauss, Negotiations: Varieties, contexts, processes, and social order (Jossey-Bass,
San Francisco, CA, 1978); Gary Allen Fine, ‘Negotiated orders and organizational cultures’,
Annual Review of Sociology 10 (1984), pp. 239–62.
29. Tim Hallett, ‘Symbolic power and organisational culture’,Sociological Theory 21,2
(2003), pp. 128–49, p. 133.
6AFRICAN AFFAIRS
at Universiteitsbibliotheek Antwerpen on March 31, 2011afraf.oxfordjournals.orgDownloaded from
interactions and directly determines the dynamics of statehood. Elements
like actors’individual attributes enter as background conditions.
The proposal to conceptualize real governance as the outcome of a
process of negotiation between a variety of state and non-state actors has
two important consequences. First, there are certain implications for the
way in which we analyse the dynamics of the ‘negotiated order’; because
elements like actors’individual attributes only partially determine this
process, it is important to give due attention to the historicity of this
arena. Taking our inspiration from Lund, we focus therefore on ‘open
moments’to analyse these negotiation processes.
30
Open moments are
‘occasions when the social rules and structures are suddenly challenged
and the prerogatives and legitimacy of politico-legal institutions cease to
be taken for granted’.
31
Second, in theory, the ‘idea’of the state is that it constitutes the final
regulatory authority in a particular territory: it makes and enforces rules,
and in so doing, ‘the state generates domination in its designated space
through a uniform set of rules on how to behave …trumping any other
rules that might exist’.
32
Legislation is therefore a powerful instrument in
the creation of central authorities, as it ‘translates aspects of a governmen-
tal program into mechanisms that establish, constrain, or empower certain
agents or entities and set some of the key terms of their deliberations’.
33
However, the negotiated nature of statehood has a profound impact on
the ‘state qualities’of governance or the ability to ‘define and enforce col-
lectively binding decisions on members of society’.
34
Through the nego-
tiated character of statehood, and the power differential between the
various actors involved, this ability is no longer inherent in the state, but
is a result of ongoing negotiations. This does not produce uniform
results; rather, the outcomes depend on the power configurations in par-
ticular localities at particular times.
The evolution of the education sector in the DRC
Non-state actors have been present in the education sector since colonial
times, as the colonial state gave a near monopoly to the churches
(Catholic and Protestant) to organize the education sector.
35
Though
30. Christian Lund, ‘Struggles for land and political power: on the politicization of land
tenure and disputes in Niger’,Journal of Legal Pluralism 40 (1998), pp. 1–22.
31. Ibid., p. 2.
32. Migdal and Schlichte, ‘Re-thinking the state’,p.5.
33. Nikolas Rose and Peter Miller, ‘Political power beyond the state: problematics of
government’,British Journal of Sociology 43, 2 (1992), pp. 172–205, p. 193.
34. Lund, ‘Twilight institutions’, p. 685.
35. Patrick Boyle, ‘School wars: church, state, and the death of the Congo’,Journal of
Modern African Studies 33, 3 (1995), pp. 451–68.
REAL GOVERNANCE BEYOND THE ‘FAILED STATE’7
at Universiteitsbibliotheek Antwerpen on March 31, 2011afraf.oxfordjournals.orgDownloaded from
initially this arrangement was continued after independence, in 1974 it
was briefly interrupted by the Mobutu regime when, in an attempt to
found an ‘integral state’, it nationalized all religious schools. However,
without the support of the religious networks (particularly the religious
personnel and the financial resources they controlled) the entire education
system arrived at the point of collapse (also amid a generalized economic
crisis, resulting from a similar policy to ‘Zaïrianize’the economy). The
government therefore re-engaged the Catholic Church to collaborate in
the framework of a state-regulated education sector, which resulted in
1977 in the signature of the ‘Convention’, an agreement for management
of national schools between the state and the churches. Introducing the
text of the convention, which was distributed to all schools, the national
coordinator clarified the reasons behind it, and hence the shared ‘defi-
nition of the situation’that allowed the renewed involvement of the
churches in the governance of the education sector:
The objective pursued by the Churches was the same as the objective of the authorities as
well as that of the whole population: to obtain a change in attitude and mentality among the
school directors, the teachers and the pupils.
36
Following the convention, the public education sector, with the state as
an organizing power, was divided into five networks: the official network
or ‘non-conventionized’schools (non conventionnés) which are directly
managed by the state, and the four ‘conventionized’networks, in which
the schools are each managed by a different religious network: the
Catholic network (conventionné Catholique), the Kimbanguist network
(conventionné Kimbanguiste), the Protestant network (conventionné
Protestante), and the Islamic network (conventionné islamique).
37
Each of
these networks has its own administration. For example, within every
Catholic diocese there is a Catholic coordination office, which organizes
the Catholic education services within the diocese. The agents who work
in these offices are civil servants, who receive state salaries. The religious
networks also have their own inspection services. In this way the division
into five networks (four conventionnés and one non conventionné) has led to
a quintuple administration and a rather complex organization of the
public education sector. At present three-quarters of all primary school
pupils attend public religious or ‘conventionized’schools. Among them
the Catholic Church is the largest, with approximately 50 percent of all
36. Dibalu-Didi, cited in de Herdt and Poncelet, ‘Enjeux et acteurs’, p. 21 (italics in
original).
37. Private schools not financed by public authorities constitute about 10 percent of all
schools. See Tom de Herdt, Kristof Titeca, and Inge Wagemaekers, ‘Making investment in
education part of the peace dividend in the DRC’(unpublished paper written for UNESCO,
forthcoming 2011), p. 5.
8AFRICAN AFFAIRS
at Universiteitsbibliotheek Antwerpen on March 31, 2011afraf.oxfordjournals.orgDownloaded from
pupils. In the urban centres, the Catholic network is also disproportio-
nately present at the richer end of the market.
38
A more significant step in the retreat of the Congolese state from the
education sector dates back to the era of Zaïre’s structural adjustment
programme, which was marked by an implosion of the education budget.
Real expenditure per pupil dropped from US$159 in 1982 to $23 in 1987
and finally to around $4 in 2002 (Table 1). Teachers’salaries dropped
from $68 to $27 per month between 1982 and 1987, reaching an absolute
minimum of $12.90 in 2002 (Table 1). Further, the number of teachers
supported (paid) by the state was cut from 285,900 teachers in 1982 to
half of this number in 2002, although there has been a slow recovery
since that time.
39
Despite the collapse in public financing, between 1987 and 2007 the
total number of pupils doubled. Indeed, between 2002 and 2007 the
number of children attending school increased by 11 percent per year –
largely compensating for the much lower growth during the 1990s.
Moreover, in comparison with many other African or less-developed
countries, the DRC has a much lower percentage of people who have
never been to school (Table 2).
Indeed, according to these data, less than 17 percent of boys and
22 percent of girls have never gone to school, well below the sub-Saharan
African average (24 and 27 percent, respectively) and well below the
average of less-developed countries (27 and 33 percent respectively). To
be sure, the comparison with older generations suggests a significant wor-
sening of the situation, especially for boys. Still, if the DRC is ranked
among the lowest in terms of its economic performance, it certainly per-
forms better than average in terms of this indicator of education.
Table 1. Evolution of the budget of the Ministry of Primary, Secondary
and Professional Education, 1982–2006 (in constant US$, 2006)
40
Budget Number of
teachers
Teachers’monthly
salary
Million $US $/Pupil $/Capita
1982 $781 $159.67 $27.17 285,900 $68.35
1987 $97 $23.44 $2.88 196,300 $27.21
2002 $24 $4.45 $0.44 142,900 $12.90
2006 $112 $6.82 $0.93 214,200 $30.00
38. de Herdt et al.,‘Making investment’.
39. de Herdt and Poncelet, ‘Enjeux et acteurs’, p. 23. In 2008, the Social Security statistics
reported 226,800 teachers.
40. de Herdt and Poncelet, ‘Enjeux et acteurs’.
REAL GOVERNANCE BEYOND THE ‘FAILED STATE’9
at Universiteitsbibliotheek Antwerpen on March 31, 2011afraf.oxfordjournals.orgDownloaded from
This performance is partly due to the gradual replacement of the state
education budget with higher contributions paid by parents. Somewhat as
envisaged by the architects of structural adjustment (the World Bank and
IMF), the parents did indeed ‘take up their responsibility’from the
mid-1980s onwards, and their involvement increased significantly during
the early 1990s. In 1992, Congolese teachers went on a major strike
because the state was no longer paying –or rather, no longer able to pay –
their salaries. In order to avoid a ‘lost year’(année blanche) the Catholic
Church and the students’parents’association
42
offered to boost teachers’
salaries by creating a ‘salary top-up system’(prime de motivation).
Although this was seen as a temporary coping mechanism to compensate
for the lack of salaries, it soon became an institutionalized practice.
Everyone, including (or even especially) the state, started taking the
‘motivation system’(système de motivations) for granted.
43
Soon, a whole
range of other fees came into existence, which financed not only schools
but the education system right up to the highest national levels of the
administration.
44
This system is at the heart of the current Congolese
education sector. It is based on a distribution mechanism (ventilation)in
which money collected from parents filters up to the highest levels: the
fees are collected from parents at school level, from where the money
is passed on to the district level (for state schools this is the ‘division’,
for religious schools it is the Catholic, Protestant, or Kimbanguist
network offices). In turn these offices pass the money on to the various
state institutions responsible for running the ‘public’education sector.
This privatized financing system proved functional, even to the point of
Table 2. Percentage of people who never went to school
41
Aged 6–16
year%
Aged 17–27
year%
Aged 23–27
year%
Boys Girls Boys Girls Boys Girls
Less-developed Countries 22 26 20 32 24 39
Sub-Saharan Africa 24 27 21 33 25 39
DRC 16 22 6 20 7 21
41. UNESCO, ‘Deprivation and marginalisation in education’,<http://www.unesco.org/
en/efareport/dme/> ( 23 February 2010).
42. The Conference of the Bishops and the National Association of Parents of Students in
Congo (ANAPECO, Association Nationale des Parents d’Elèves du Congo).
43. Mrsic-Garac, ‘La “participation”’.
44. Johan Verhaghe, ‘School fee practice and policy in the DRC: frais de fonctionnement
or fonds de famille?’(Consultant’s report for UNICEF/USAID, December 2007).
10 AFRICAN AFFAIRS
at Universiteitsbibliotheek Antwerpen on March 31, 2011afraf.oxfordjournals.orgDownloaded from
funding an estimated 90 percent of all operating costs of the education
sector.
45
To summarize, by the end of the twentieth century, the education
sector in DRC could indeed be called ‘privatized’, both because most
schools were run by non-state actors and because most of the finance
came from parents themselves. Despite the extensive engagement of non-
state actors, however, education can still be considered a public service if
we take a ‘grounded’approach to public services, one which is empirical
rather than juridical or economical. As Olivier de Sardan argues, the prin-
cipal questions to be asked of public services are: (1) Are they considered
indispensable? (2) Do they correspond to the general interest? (3) Are
they a shared concern among the local population ‘who have high expec-
tations and consider that the state and political authorities have responsi-
bilities in these matters’?
46
This is also the case in the DRC: although the
state has to a large degree withdrawn from the education sector, the
various actors involved continue to look at the state as the primary actor
responsible for organizing and financing the sector. All actors involved –
churches, students, parents, teachers, and their trade unions –are clearly
dissatisfied with the high degree of private financing of the education
sector, and have been lobbying the state to increase its involvement. In
2004, the ‘Mbudi agreement’between the government and the teachers
signified a renewed engagement of the state by defining dramatically
increased salaries for teachers. As this engagement has not been
respected, teachers –supported by parents –have gone on strike at the
beginning of every school year, demanding that the state deliver on its
promise with regard to salaries. The churches have also taken similar
measures. As will be described later, in 2008 the Catholic network in
Kinshasa took concrete action to force the state to pay teachers’salaries.
In other words, strong expectations are held by all actors involved –
churches, students, parents, teachers and their trade unions, and the state
itself –that the state should fulfil its responsibilities in the education
sector.
The current situation is difficult to explain in terms of the assets the
individual actors contribute to the sector: not only does the ‘Convention’
of 1977 signal that one of the crucial assets to organize the sector (‘to
obtain a change in attitude and mentality among teachers, school direc-
tors and pupils’) is in the hands of the churches, but from the mid-1980s
onwards the state’s material contributions to the sector also evaporated,
45. World Bank, ‘Democratic Republic of Congo. Country Status Report on Education’
(Report No. 30860-ZR, 15 November 2004), p. 48.
46. Jean-Pierre Olivier de Sardan, ‘Local governance and public goods in Niger’(Working
Paper No. 10, Africa, Power, and Politics Programme, Overseas Development Institute,
London, July 2010), p. 3.
REAL GOVERNANCE BEYOND THE ‘FAILED STATE’11
at Universiteitsbibliotheek Antwerpen on March 31, 2011afraf.oxfordjournals.orgDownloaded from
and were soon deemed insufficient even to guarantee the most basic tasks
related to the function of keeping overall oversight and ‘organizing’the
sector. For example, inspectors lack the financial resources to visit schools
and educational authorities lack the coercive means to implement their
measures. However, this does not mean that the state is powerless:
according to Lund, it is the idea of the state that keeps the amalgam
together.
47
As argued above, all actors involved expect the state to provide
this particular service. In other words, the ‘idea’of the state as the final
actor responsible for this particular service is not being challenged, which
gives the state a high degree of symbolic power in the negotiation
arena: no-one challenges it as the central actor producing a shared defi-
nition of the situation –and, in particular, of the role to be played by the
other actors.
How this symbolic power works out in practice, and how the combi-
nation of individual assets and symbolic power leads to change in the
sector, will be studied through an analysis of specific events. We first
discuss the financing of buildings for the administration of the education
sector in the province of Bandundu. Second, we discuss the introduction
of free education in that same province. A third case is built around the
so-called ‘solidarity fund’, which was introduced by the Catholic network
in the capital of Kinshasa.
Construction of buildings in Bandundu: power and negotiation areas in the
education sector
As argued above, the state has lacked sufficient public means to pay tea-
chers’salaries since the 1980s. What little funds existed were spent
entirely on current expenditures, to the neglect of capital investment, for
which the state counted on the international networks of religious organ-
izations. Nevertheless, to our surprise, when we visited the school admin-
istration in the city of Kikwit, Bandundu, we were invited into a
brand-new building. Over the years, the province of Bandundu has devel-
oped a particular tradition of ‘self-financing’(via parents) of various edu-
cational buildings. This happened for the first time in 2004, when the
former educational province of Bandundu was subdivided into three new
provinces (Bandundu 1, 2, and 3). The new educational province
Bandundu 2 did not have any buildings for its different educational ser-
vices, so the main actors in the sector (the head of the educational pro-
vince, the chief inspector, the heads of the various religious networks, and
parents’representatives) gathered and asked for parental contributions to
construct buildings. Primary and secondary school students were each
47. Lund, ‘Twilight institutions’.
12 AFRICAN AFFAIRS
at Universiteitsbibliotheek Antwerpen on March 31, 2011afraf.oxfordjournals.orgDownloaded from
asked for US$0.1 (FC100) during two academic years (2004–5, 2005–6),
as well as some contributions in kind. This system proved a success: con-
struction works for the building were started in March 2005 and finished
in June 2006. In other words, as the national government was not provid-
ing the necessary funding for construction, the local actors organized
themselves to finance the buildings. This successful action was sub-
sequently repeated several times in the district.
However, this system was not always successful in other areas. The sub-
division of Masimanimba faced a similar problem: no buildings for the
inspection service. The chief inspector and ‘regional administrator’
(administrateur de terroire, the highest public authority in the district) there-
fore sought contributions from the parents in the subdivision. In 2007,
they obliged the schools to contribute to this, and the inspectors made
sure this happened. The Catholic network, however, refused to contrib-
ute, allegedly because they were not consulted beforehand, and because
they were not present in the management committee handling the funds.
As the Catholic network is the biggest network in the area (60 percent of
the public schools are Catholic), the project could not succeed: no
Catholic schools were allowed to participate. According to the Catholic
coordinator, the state in turn blocked the Catholic schools from partici-
pating in the final exams for the final years of primary and secondary
schools. The situation was not solved for some months, until a new
‘regional administrator’arrived in the area. This administrator relaunched
the initiative, but this time following the principles which had been estab-
lished by the construction of the building in Kikwit in 2004: the manage-
ment committee was composed of all major actors in the sector, and
therefore also representatives of the Catholic network. The joint commit-
tee then took the initiative to finance the buildings and to manage the
whole process. As an inspector summarized the arrangement: ‘the incor-
poration of all these actors is legally not obliged, but it is nevertheless
necessary’.
48
Since then, these principles have been followed for all con-
structions in the district.
As we understand it, the above example teaches us several things. First,
the state does not automatically succeed in establishing itself as a central
authority. As it cannot claim to be the supreme rule enforcer, its public
authority must be negotiated. Without the consent of powerful non-state
actors, it is not able to implement its decisions. In this particular instance,
without the material assets of the Catholic Church, the project was bound
to fail, despite its public support and promotion. Yet non-state actors
have agreed to invest in state infrastructure, thereby achieving something
the central state itself was apparently unable to do. Second, once the
48. Interview, school inspecter, Kikwit, 16 June 2009.
REAL GOVERNANCE BEYOND THE ‘FAILED STATE’13
at Universiteitsbibliotheek Antwerpen on March 31, 2011afraf.oxfordjournals.orgDownloaded from
principles of engagement of all stakeholders were established, they
defined as it were the appropriate way to engage in collective action. As
argued by Tim Hallett, once this arrangement was put into practice, these
practices also influence future interactions, as they provide a model for an
appropriate arrangement. In this way, previous arrangements become a
source of symbolic power.
49
In the case of the building of Masimanimba,
the Catholic Church did not challenge the authority of the state as such –
it did not question either the state as an object of political practice or its
legitimacy to take these initiatives –but it did question the way in which
this was done, as these particular state actors did not sufficiently recognize
how deep the mediation by these non-state actors is ‘entrenched in the
procedures, policies and habits’of statehood.
50
Third, it is to be empha-
sized that the case of Bandundu was rather exceptional when compared
to other provinces. The negotiated order that made possible a non-
negligible investment in state infrastructure is certainly locally specific.
The locally specific character of these arrangements becomes even clearer
in the next section.
De facto decentralization of public authority
If real governance is negotiated between local state and non-state actors,
the image of the state as the ultimate regulatory authority remains just an
idea. But the tension between the state as an image and practical state-
hood does not arise only from the involvement of non-state actors in the
production of public services; it can already be observed within the state
administration itself. We can aptly describe the Congolese state adminis-
tration with Mbembe’s words, as a proliferation of ‘more or less auton-
omous pockets in the heart of what was, until recently, a system’or a de
facto decentralization.
51
In this section, we will describe how local state
actors enjoy considerable autonomy in defining their own priorities, but
how this also creates a number of practical problems and a lack of
coherence.
In 2004, the Congolese constitution explicitly declared that ‘education
in primary public schools is free and compulsory’(Article 46). However,
the national state did not implement this measure. Nor was it able to do
so. In practice, the national government issued an (unconstitutional)
inter-ministerial decree that limited school costs to a maximum of
FC1020 for all public schools in October 2007.
52
Worse, in a letter dated
49. Hallett, ‘Symbolic power and organizational culture’, p. 133.
50. Ward Berenschot, ‘Everyday mediation: the politics of public service delivery in
Gujarat, India’,Development and Change 41, 5 (2010), pp. 883–905.
51. Mbembe, On the Postcolony, p. 80.
52. Inter-ministerial decree No. MINEPSP/CABMIN/0311/2007, 11 October 2007.
14 AFRICAN AFFAIRS
at Universiteitsbibliotheek Antwerpen on March 31, 2011afraf.oxfordjournals.orgDownloaded from
August 2008, the Minister added further categories of costs.
53
In the
meantime, the government has negotiated an agreement with DfID,
which has volunteered to pay the insurance of all primary school pupils.
54
However, the Ministerial letter remains silent about this, and conse-
quently some government agents continue to collect insurance fees in
schools. It should be added that, according to the Framework Law signed
in 1986, it is a provincial prerogative to declare the relevant school costs.
In other words, in theory the Framework Law allows the provinces rather
than the national government the final authority to set costs,and thus pro-
vides them with a legal basis for ignoring the inter-ministerial decrees.
Several provinces have even unilaterally declared some form of free edu-
cation. This was the case in Bandundu, for example, where the provincial
government at the beginning of the academic year 2007–8 declared uni-
versal education for the first year of primary school. However, it did not
have the means to finance this measure. In other words, the measure was
introduced unilaterally by one actor in the educational system, the provin-
cial government, but not by the other actors, such as the national govern-
ment or the schools, which continued charging fees to the students.
Students still had to pay a range of fees such as the national enrolment fee
(‘minerval’) and the ‘bulletin’(school report) fee, costs defined at the
national level. They also had to pay costs specific to each school, such as
construction costs, for example. The religious networks, too, continued
their usual practices: since the provincial government did not provide
financial compensation, they continued charging network-specific fees for
the first year of primary school. Their legal reference for this is the
Convention of 1977. For example, when the Catholic network in
Bandundu (Kenge Nord) decided to build its provincial headquarters, it
asked all students in its network, including the first years, to contribute.
Moreover, at the beginning of the academic year 2008–9 (the second year
of the measure), the Catholic network ignored the initiative by continuing
to charge fees from the schools –something which only stopped after an
intervention from the state authorities, but a full six months after the
beginning of the school year. In other words, the practical difficulties in
implementing particular measures are the result of the fragmentation of
the education sector into different units or ‘fiefdoms’, both in spatial and
sectoral terms. There is not only a great distance between the ‘centre’and
the ‘periphery’, but also between state actors and the religious networks.
It is clear from this example that sometimes there is indeed no ‘nego-
tiated order’–or, in a more benevolent assessment, that the situation is
53. Ministerial Letter No. MINEPSP/CABMIN/006/2008, 7 August 2008.
54. Every pupil has to pay ‘insurance’against accidents to the National Assurance
Company, SONAS (Société Nationale d’Assurances).
REAL GOVERNANCE BEYOND THE ‘FAILED STATE’15
at Universiteitsbibliotheek Antwerpen on March 31, 2011afraf.oxfordjournals.orgDownloaded from
only converging towards such an order. True, all actors refer to the same
image of the state, but in practice they (and even the same actors) do
quite contradictory things. It may be the case that actors refer to a
common definition of the situation, but this does not prevent the exist-
ence of what Hallett called multiple legitimate practices within the same
negotiation context.
55
More specifically, Hallett connects the existence of
multiple legitimate practices to the existence of multiple audiences, or
groups of people who are witness to a relevant practice.
56
As the audi-
ences of different state representatives in the education sector vary from
international donors to local-level church organizations and parents, and
as the sources of finance and individual-level symbolic capital are as het-
erogeneous, it would indeed be difficult to expect the state regulatory fra-
mework to function as a coherent system. Though symbolic power as we
defined it is shared by several actors, it is not necessarily shared by all in
the whole administration. While, for example, the district- and national-
level politicians may introduced free education as a measure to satisfy
their electorates, church leaders or national-level civil servants have com-
pletely different legitimate practices and ‘audiences’. In fact, achieving
such coherence would itself have to be the outcome of a specially
mounted negotiation effort; incoherence would be the default.
Multiple legitimate practices within non-state actors
Afinal illustration of the dynamics in real governance comes from
Kinshasa, where the Catholic school network unilaterally took the initiat-
ive to install the so-called ‘solidarity fund’(fonds de solidarité). This fund
was introduced in 2007 by the head of the Catholic school network, Mgr
Monsengwo, the Archbishop of Kinshasa. The initiative introduced set
fees for all Catholic schools in the capital for the year 2007–8: $65 for the
primary schools and $95 for the secondary schools. The money would be
managed at the central level and allegedly serve as a veritable ‘solidarity’
fund for the poorer schools, whereby the wealth of the richer schools
would be redistributed to help the others. At the same time, the
Archdiocese would have a say in the way in which the money would be
spent. In particular, the collected money was explicitly not meant to
finance the teachers’salaries. The bishop aimed to force the state to take
up its responsibilities by obliging it to start paying salaries as promised.
57
55. Hallett, ‘Symbolic power and organizational culture’, p. 135.
56. Ibid.
57. The decree which introduced this measure to the schools is very clear on this: ‘It is
time to correct the system, by avoiding a situation in which the salaries of the public sector
teachers weigh upon parents’shoulders. It is, indeed, inadmissible that we ask parents, the
majority of whom do not receive a salary, to support the burden of teachers’salaries which
16 AFRICAN AFFAIRS
at Universiteitsbibliotheek Antwerpen on March 31, 2011afraf.oxfordjournals.orgDownloaded from
Finally, by introducing set school fees, the Archbishop wanted to abolish
inequalities in the school system, where some schools asked higher fees
than others.
This fund challenged the authority of the state at three levels. First, in
theory the state has the final authority to decide on school fees: every
year, the Ministry of Education issues a circular on authorized fees. This
measure was a clear negation of this authority. Second, the initiative did
not match with the current efforts made by the government and some
international donors to reduce school fees in the DRC. Indeed, the soli-
darity fund went radically against this, by strongly increasing fees for the
majority: in most schools, the parents paid less than one sixth of the pro-
posed equivalent of $65. Third, as highlighted above, the initiative was
intended to force the state to increase teachers’salaries. The fact that the
parents were no longer allowed to pay the salaries was expected to create
a social movement, forcing the state to pay. However, this did not turn
out to be the case: the teachers instead starting acting against the schools,
not against the state.
For these reasons, the urban commission for educational services –the
highest relevant authority in Kinshasa –declared the solidarity fund
illegal. However, the ruling had no real effect and the decision was not
even communicated to any other actors, such as the schools, the civil
servants in the education sector, or the Catholic network itself. This
shows how –just as in the case of the school buildings in Bandundu –
order must be negotiated if it is to be realized; it cannot be unilaterally
decreed by the state.
Neither can order be unilaterally decreed by the churches. While the
initiative of the Archdiocese was not in any way imperilled by the state, it
was eventually overruled from within the church itself. Indeed, within the
church, the initiative was quite heavily contested, if not in words at least
in practice. On the one hand, the well-to-do elitist schools refused to par-
ticipate, as $65 per pupil was not sufficient for them to cover their oper-
ational costs. On the other hand, schools in popular neighborhoods
risked losing many pupils who were unable to pay the exorbitant cost.
And, of course, in all cases teachers refused to work unless they were paid
their usual ‘motivation costs’. Eventually, most schools did not contribute
to the solidarity fund. Even if many charged the $65, most of
them managed the money themselves, inter alia to pay their teachers’
motivation fee.
are a state responsibility, whatever the name given to this parental intervention: premium,
intervention, salary, etc.’(our translation, emphasis in original). Decree of Mgr
Monsengwo, 19 September 2008, p. 1.
REAL GOVERNANCE BEYOND THE ‘FAILED STATE’17
at Universiteitsbibliotheek Antwerpen on March 31, 2011afraf.oxfordjournals.orgDownloaded from
The case illustrates again that, in practical arrangements, it is very diffi-
cult for any single actor to muster sufficient capital to create order. Just as
in the case of school buildings in Bandundu Province, it is clear that the
state’s regulatory framework is barely helpful in disciplining non-state
actors; the latter can draw on their own legitimate practices, which only
partly overlap with those of the state. The state has thus lost its monopoly
on producing real governance, but the case of the solidarity fund also
demonstrates that monopoly is not wielded by the Catholic Church,
either: the symbolic power that keeps a church together as a unified actor
is also, necessarily, the result of a negotiation process bridging multiple
legitimate practices. In other words, to the extent that (if at all) the
Catholic Church can be called a unified actor, it is a negotiated actor.
Conclusions
This article has shown how the retreat of the state from the public sector
has not led to a vacuum, but rather to the filling of this public arena by a
range of other actors, who play a crucial role in the provision of public
services. Non-state actors, and particularly religious networks, stepped in
to provide educational services, and continued organizing the sector. The
resulting mode of governance was negotiated between an amalgam of
state- and non-state actors. The state (and its different representatives) is
only one of various actors which provide public services and construct
governance in a certain locality. In this situation, the locus and pro-
duction of such services lies not only with the state, but in the negotiation
between this broader coalition of actors. As real governance functions
through alliances with non-state actors, regulation similarly happens
through negotiation with non-state actors. In this negotiation, relation-
ships are not equal, as certain actors have more leverage than others. This
power is not a manifestation of formal rules and authority, but a reflection
of the de facto organization of the education sector. In other words, the
state has lost its position as final regulatory authority; instead, this role is
negotiated between various actors rather than being claimed by a supreme
authority (the state). The state alone is unable to construct a normative
(legal) framework, as its ‘technologies of rule’are no longer respected. In
this situation, ‘stateness’is therefore not something fixed, but rather
something which ‘waxes and wanes’–state institutions are never defini-
tively formed, but are in a constant process of formation.
58
This became very clear during the ‘negotiation arenas’which were
analysed. The first example (the construction of the education buildings in
Bandundu) showed how the state is incapable of organizing the sector –and
58. Lund, ‘Twilight institutions’, pp. 686 and 697.
18 AFRICAN AFFAIRS
at Universiteitsbibliotheek Antwerpen on March 31, 2011afraf.oxfordjournals.orgDownloaded from
more specifically of financing its buildings –without the explicit involvement
of the Catholic network. In other words, all ‘poles of power’need to be
involved for the public sector to function. Moreover, this helped us to
understand the power relationships in the education sector: although the
state still wields a degree of symbolic power, it is not the lead actor in this
particular public sector. Instead it is dependent on non-state actors particu-
larly for material and organizational resources. The second example docu-
ments the bewildering incoherence in actions or decisions within the state
itself. We argue that this incoherence is directly related to the heterogeneity
of audiences with which the state and non-state representatives at different
levels are allying to ‘do’the state. The third example suggests that in the
case of non-state actors, too, internal coherence is not a given one can take
for granted but something that must be negotiated.
It is important to realize that this ‘balance of forces’is unequal, and
that the negotiation process is not inclusive. Rules do not stem from the
law (that is, from the state as the regulatory authority), but rather from
the pragmatic negotiations between the different poles of public authority.
In other words, the extent to which actors ‘carry out the will of another is
always conditional on the particular balance of force, energy and meaning
at any time and at any point’.
59
However, the absence of an overall power to govern the system –a
regulatory authority –does not mean that the education sector is ungov-
erned. In several non-trivial ways, this mode of governance also functions,
as testified by enrolment data. Through the negotiation between the
different entities, the education sector remains governed, but through
‘intrinsically messy, contradictory, illiberal, and constantly renegotiated
deals’.
60
The outcome of these negotiations varies over time and between
places. To refer again to Weber’s distinction between ‘power’and ‘domi-
nation’: the power of the state has gradually been replaced by the power
of other actors. Domination is a more complicated matter. Almost all
actors refer to the image of the state as provider of the ultimate regulatory
framework. This idea keeps together the amalgam of actors which exerts
real governance and determines the practice of statehood. The different
cases we discussed allowed us to nuance these ideas still further. In par-
ticular, the recognition of the state as the provider of the ultimate regulat-
ory framework does not imply a consensus on what can be considered
legitimate practices. It is more helpful to think of a negotiation arena in
terms of a set of multiple, but partly overlapping legitimate practices. In
this way we can understand the internal incoherence of state representa-
tives’actions, but also the incoherent actions within the Catholic Church.
59. Rose and Miller, ‘Political power beyond the state’, p. 193.
60. Raeymaekers et al.,‘State and non-state regulation’.
REAL GOVERNANCE BEYOND THE ‘FAILED STATE’19
at Universiteitsbibliotheek Antwerpen on March 31, 2011afraf.oxfordjournals.orgDownloaded from
