Conclusion

Summary

The final chapter concludes by summarizing the book’s core argument before offering reflections on the implications my findings hold for ongoing decentralization reforms and the study of historical legacies.

The past few decades have seen rapid improvement in development outcomes around the world. Senegal halved the population living under $1.25 a day, achieved gender parity in primary education, and made great strides toward reducing child mortality. These gains – the result of high-profile objectives like the Millennium Development Goals (MDGs) and a broader push to expand basic service infrastructure – are often studied comparatively across countries. Mali and The Gambia were on track to meet the MDGs in 2015, for example, while Senegal lagged behind with neighboring Guinea and Guinea-Bissau. Yet as the preceding chapters have documented, it is just as critical to look within countries to assess how international agendas fare in practice. Following the confluence of decentralization, political liberalization, and the push to improve living standards, I suggest a novel source of subnational variation: progress may be made unevenly within single countries as a function of historically inherited and spatially uneven attributes.

The remaining pages of this book briefly summarize the argument before offering reflections on the implications my findings hold for ongoing decentralization reforms. I conclude with a discussion about the latent questions of determinacy and durability in the precolonial legacies I identify.

Summary of the Argument

This book has sought to explain why local governments engage in strikingly different redistributive patterns despite inhabiting a shared formal institutional framework. My central empirical claim is that areas of Senegal that were home to precolonial states engage in spatially broader strategies of representation and redistribution. I have argued that exposure to precolonial polities generated social institutions, or norms of appropriate behavior in the public sphere that are demarcated by group boundaries, and that these norms have persisted to the present, carried over time by rural social hierarchies. Social institutions work because they are deeply ingrained in how actors pursue political and social strategies, informing not only what they expect of others but how they understand the value of their social relations more broadly. My theory of institutional congruence stipulates that when formal institutions spatially overlap with informal, social ones inherited from the precolonial past, politicians’ behavior is reoriented toward community goals and relationships. When formal institutions are incongruent with local social institutions, we see more opportunism on the part of rural elites, specifically because their social worlds do not map onto their political ones.

Herein lies my answer to the first of two animating questions laid out at the start of this book: why are some communities able to come together to improve their collective lot while others are not? I have argued that some communities are endowed with informal social institutions that embed elites within local behavioral norms that reorient their behavior away from individual opportunism. Social institutions are carried into local governance through two mechanisms. First, precolonial states endowed communities with robust social identities that tie members together into a sense of shared fate. Shared identities imbue social institutions with meaning, but they inhere in the theory’s second mechanism of cross-village social networks. Dense network ties impact elite behavior through three channels: they circulate information, they transmit rewards for well-viewed behavior and sanctions for poor-viewed behavior, and they reinforce the internalization of social institutions as actors come to see abiding by these norms as intrinsically rewarding. When social institutions embed elites from across the community within such interconnected webs of obligations, elite behavior is reoriented toward prosocial group goals, but only when these networks are congruent with the bounds of the local state.

This suggests that state-building at the grassroots may be uneven at least in the near term. In the wake of institutional reforms across the Global South, I suggest that attention to how local elites negotiate new political spaces of local governance can shed as much – if not more – light on macro-institutional reforms initiated by the center as can be gained by focusing on the center’s own motivations. To the extent that one of the central claims of this study is that the performance of formal institutions is contingent on the impact of informal social institutions, I show how forces often conceptualized as impediments to the state-building project, such as precolonial, indigenous social organizations, may paradoxically serve to entrench the state locally by grafting it onto preconceived understandings of community. The implications of this are not only local, but seep upward to shape the strength of national political parties, the distribution of economic opportunity, and the democratic project itself.

The second question raised in this book’s opening pages asked when the past matters for contemporary politics and why. I have answered by arguing that long-defunct political institutions can remain politically consequential long after their formal structures have disappeared, but that the impact of these legacies may be contingent on how contemporary institutional delimitations put these historical inheritances into relief. My central claim, that precolonial legacies persist to powerfully shape contemporary understandings of political community, risks reading as a quaint claim about a distant and idealized past but embedded within these identities and the social relations that maintain them are deeply normative questions about which citizens are the most deserving. By examining when and why the precolonial past matters, my findings will hopefully serve as a reminder that institutions inherited from the past are intrinsically political creatures, amenable to reinvention and repurposing across time.

This holds key implications for the literature on African politics in its suggestion that scholars have underestimated salient local political identities by prioritizing ethnicity as the most potent political cleavage in sub-Saharan Africa. By identifying a powerful role for place-specific identities rooted in the past, the preceding chapters encourage us to take seriously the social identities that our respondents themselves invoke to make sense of their political lives. Certainly, these may revolve around ethnicity, but in rural West Africa they appear as much – if not more – likely to draw on far more localized histories. As I have shown, intra-ethnic tensions, around clan or caste, spoil prospects for local government cooperation as frequently as inter-ethnic dynamics themselves.

This is particularly important when studying institutional change, such as that brought on by decentralization. I challenge the assumption that certain political identities are always politically relevant by documenting how social identities latent at the grassroots reenter political life following institutional reform. In the wake of institutional reform, therefore, we should be attentive to the ways in which identities may shift to lose or gain salience not only because new institutional environments can create new identities, but because the very shape of institutional boundaries can valorize old identities in new ways. By focusing on the spatial interaction between identities and institutional boundaries, I offer a framework for understanding why identities may generate distinct behavior at some moments of time but not at others.

Three methodological implications follow. First, and most simply, this project calls attention to the necessity of studying institutional change where it takes place. Despite widespread interest in decentralization reforms, scholars have disproportionally kept their focus on the central state or, alternatively, on the village or project level, as seen in the literature on elite capture. As a new layer of governance, decentralized local governments face unique redistributive dilemmas that are often distinct from those faced at higher or lower levels of analysis. The interests of local actors demand concerted investigation and theorization.

Second, my findings show that we can gain unique insight into political behavior by taking seriously how individuals narrate their political lives. Even if the tales of precolonial grandeur that I have repeated in these pages verge on myth and stray from historical accuracy, the collective embrace of these narratives is a valuable form of data for understanding local political and social action explicitly because the act of their retelling valorizes and reproduces social institutions. To the extent that local actors recount these narratives from their own lived experiences, they shape how social and political objectives are understood in their communities and, by extensions, the appropriate policies to pursue them. In this way, the telling and retelling of local narratives is an explicitly political act and should be studied as such.

The third methodological lesson relates more narrowly to recent criticisms that the literature on historical legacies is apolitical and atemporal.¹ I strive to offer a corrective on both fronts. Much of the social sciences’ recent rediscovery of the past relies on the assumption that a historical cause, once unleashed, persists to the present in a uniform fashion. In contrast, I show that the effect of precolonial statehood in rural Senegal only appears following the exogenous institutional shock of the 1996 decentralization reforms. Embracing calls to “decompress” history, I look at when the precolonial past can explain distributional patterns to reveal that the social legacies of precolonial centralization may be spatially path-dependent over time, but that this does not necessitate persistent effects on the dependent variable.² In other words, I offer an explicitly temporal story: the long shadow of the precolonial past is at best a story of intermittent effects for local redistribution.

At the same time, I have sought to tell an explicitly political story by focusing on the mechanisms that keep social legacies rooted in the precolonial era alive within rural Senegal. Institutions always have distributional consequences, making them subject to enduring political contestation. This renders historical interactions all the more consequential, leaving the general lack of attention to how precolonial or colonial legacies play out over time puzzling. As discussed at length in Chapter 7, I find that colonial legacies have largely faded over time, while the impact of precolonial legacies on distributive outcomes has only reemerged following decentralization reforms. Although my argument is highly structural, it opens up room for the agency of the postcolonial state, which has increasingly exerted its own logics of rule at the expense of colonial legacies. Empirically, this serves as a reminder that we should (and can) test rather than assume the determinacy of the past.

Implications

Despite early pessimism about decentralization’s potential,³ evidence is emerging that the local state is an important site of contact between citizens and their government. Bratton et al. (2000), for example, note that many African citizens find it easier to contact their local councilors than distant state administrators. And even if the local state is just one actor among many, with decentralization “merely represent[ing] another remount in a long series of regime changes imposed from above by the state,” it has nonetheless become a critical arena where consequential decisions about citizen well-being are made.⁴ Despite this “silent revolution” in statecraft, however, this is only one of a handful of studies that explores the nature of political life under decentralization.⁵ What then does this study imply for these widely promoted yet understudied reforms?

The most immediate conclusion is that the political spaces generated by decentralization enable, create, and revive highly localized political interests. Local state elites are hardly mirrors of their central state counterparts and even as local actors actively engage with the central state, seeking projects, favors, and influence, these interactions do not define the totality of their political preferences. This serves as a reminder that local decision-makers are obliged to negotiate across villages when deciding how to distribute scarce projects and resources. Local government performance under decentralization is not reducible to the question of when local elites are able to control local positions of authority, as argued by the elite capture literature therefore, because this ignores the question of when the interests of local elites are more or less compatible in the first place. Centrally, the expansion in scope of what local agents control highlights the importance of interrogating when we expect local state actors to pursue their interests and, most importantly, when we think their own interests dovetail with those of the broader community.

This should nuance how we think about the sharp subnational inequalities documented in this book. If decentralization was designed as a means to improve transparency and representation and to aid participatory development, then my suggestion that some areas are structurally predisposed to find these tasks easier implies that we will likely see a widening gap in outcomes for core donor agendas around basic service delivery and democratic consolidation. In the worst of cases, some local elites and the villages they represent report disengaging from their local governments altogether. Of course, this may lead a village or community to mobilize local social and economic capital to find solutions to their own problems that can produce acceptable outcomes over the short term. But if decentralization was designed with an eye to facilitate the development of more robust social contracts, we should temper our expectations about the broader process of democratic consolidation that decentralization reforms are often bound up within.

In some ways, this book risks shifting its focus too heavily to the question of why elites in historically centralized areas are endowed with more prosocial preferences, which may be read by some as suggesting that all good things go together. Delivering goods broadly across space is not necessarily the ideal criteria with which to evaluate local development outcomes; highly populated villages with overcrowded classrooms could suffer if local governments prioritize giving even small, isolated localities their own school. Nor, of course, is there something inherently objectionable about politicians rewarding those who vote for them as we see in some communities.⁶ Moreover, local social institutions around conflict resolution and equal voice may create “good” outcomes in some ways, but they carry with them profoundly antidemocratic aspects. On the flip side of social institutions’ ability to prioritize consensus and prevent political blockage of local government work, is the suppression of robust democratic contestation, which inherently relies on dissent and disagreement. In a like manner, the dense social ties or identities that serve to reproduce these institutions can be equally disadvantageous to minorities or those who do not otherwise conform. As I documented in Chapter 4, minorities and in-migrants “adopted” into congruent local governments do sometimes report weaker claims to voice in their communities.

As such, there is something intrinsically conservative in the argument laid out here. After all, the international community has been anxiously promoting participatory democracy initiatives for going on three decades in an effort to undermine the very elite and communal dynamics I identify.⁷ The logical consequences of my argument for this view are at once explosive but also fundamentally mundane. It indicates most clearly that there are limits to what outsiders (international and domestic alike) can achieve in the short term and that we should approach efforts to reengineer social norms with a certain wariness. To the extent that one of the most basic lessons in the preceding chapters is that top-down reforms are conditioned by local social relations, I echo Wibbels’ (2019, 21–23) recent suggestion that one reason many information and accountability experiments have fallen short of achieving their desired ends is because they fail to account for how social context intimately shapes the ways in which communities receive and act upon such initiatives. To the extent that such “social engineering” by development agencies and academics continues, we should be increasingly attentive to the ways in which it encounters local social realities that will shape the outcomes under study. If nothing else, this book has shown that large, international agendas, such as decentralization, never descend upon neutral terrain.

The argument further implies that the impacts of such social engineering are likely to vary subnationally to the extent that local communities possess distinct social networks and varying levels of cross-village solidarity. This could help resolve known noncumulative findings. Baldwin and Raffler (2019), for example, have recently observed that despite a proliferation of academic research on the subject, the impacts of the role of customary authorities in local governance and questions of development more broadly are not straightforward. Stronger traditional authorities seem to help project outcomes in some countries or regions, while inhibiting them elsewhere. As they note, outcomes seem to improve when traditional leaders are socially and geographically embedded.⁸ I hope to help us move away from debates over whether customary authorities and local elites are “good” or “bad” for development or democratic practice to recognize them as political actors responding to varied incentives. This then returns us to an animating question of this book: when are local elites more or less likely to see their social interests overlap with those of the local state?

Put otherwise, two issues are at stake: first, do our interventions target the scale at which individual and communities understand their social networks and obligations to be? And, if they are not, are we secondly willing (indeed, we may ask if it is ethically appropriate) to intervene wherever that may be? Should we conclude, for example, that we must simply find better boundaries and hunt for institutional congruence? I would caution against this as a viable solution because it risks creating infinitesimally small and non-sustainable political and economic units.⁹ Nor should we presume that cross-village social institutions are amenable to construction, at least in the short term. One potential reading of my argument, therefore, is that there are deeper social processes at work than those immediately responsive to outside intervention. Decentralization does not immediately generate cohesive communities, nor can the forms of institutional congruence I study here be created at will.

In contrast, we may have to content ourselves with Eckstein’s (1997, 15) observation that “incongruent authority patterns tend to change toward increased congruence.” Robust social institutions and social identities can take many forms, meaning that with time, local governments could become equally viable sources of identification as the identities inherited from the precolonial past documented here.¹⁰ This evolution will only happen, however, through politics, a necessarily unequal process that creates winners and losers. Proponents of decentralization would often be well-served to remember that decentralization shapes and redefines the fundamental ties that citizens hold with their states by design. That this generates short-term inequalities should not come as a surprise nor the fact that such reforms both “create and erode social resources.”¹¹ These processes are deeply political, but they are also deeply social; nothing, in other words, that can be easily, quickly, or painlessly adjusted from above.

If decentralization granted an unforeseen advantage to communities with high degrees of congruence, while exacerbating intervillage inequality in others, it remains an open question as to how “stuck” these outcomes are. My evidence – from the first two decades of democratic decentralization – may describe what will prove to be a temporary outcome over the long run. We could easily imagine that the contestation and dissent that I document in many historically acephalous communities may generate stronger local governments over the longer term if it strengthens local party apparatuses or induces more responsive leadership in the years to come. Acephalous zones certainly do not lack some cultural requisites currently possessed by centralized areas.¹² To the contrary, villages I have visited in these zones are equally vibrant and they likewise produce dynamic leaders who wish to do well by their families and friends. But in the years during which research for this book was carried out, the social ties that bind elites in historically centralized areas have the primary benefit of time and all that it brings with it: intermarriage, histories of mutual aid, and the camaraderie that can arise out of cohabitation across generations.

Ultimately, I part ways with the increasingly prominent view in economics that suggest the past traps some communities in bad equilibria. This is particularly prominent in work on the long-run effects of cultural norms by scholars of political economy of development,¹³ which views historical path dependencies as leading countries or communities toward more or less productive equilibrium outcomes.¹⁴ These accounts bear the same risks of an earlier culturalist tradition in political science, which was similarly prone to the view that communities could be ensnared in patterns of belief and behavior. But this unsatisfactorily dooms political behavior to be “mechanistic and unvoluntarist” without recognizing that many of the most important questions we study see persistence without determinism as societies change slowly.¹⁵ To be clear, I view the long-run historical processes that create different configuration of local social institutions today as neither intentional nor something that we can create or reform from on high, but this should not be read as suggesting that these communities are not home to dynamic sociocultural changes that may move communities toward or away from institutional congruence even if formal institutions remain the same. By emphasizing the question of spatial overlap in lieu of invoking ideas of good or bad cultural equilibria, I argue we can explain why informal social institutions condition institutional performance and reform across radically different forms of government in a more productive fashion.

Indeed, the question of change is pervasive in many of the communities I have worked in. What will happen, one local government secretary mused, when the old men – who today take their horse cart out to visit old friends in nearby villages, spending the day discussing old times and contemporary problems alike before heading home at night – pass away if the youth are less likely to do this?¹⁶ Herein lies the very fragility of the social cohesion underpinning the dynamics of my theory. Because social institutions are bound up in social identities and social networks, shifts in either or both of these components may lead local actors to question the moral obligations generated by this shared social inheritance.

As the costs of distance are eased and time passes, networks are likely to broaden. This can expand social and political horizons as rural villagers gain new information that reshapes their sense of the possible.¹⁷ As “development” seeps further into the countryside, newly paved roads plying more and more taxis and minibuses, local elite social networks are likely to be increasingly challenged by new forms of claim-making by rural citizens. On the flip side, however, it is possible that even more communities will pull out and refresh histories that they can uniquely lay claim to. The claim to represent a deeper, “truer,” or more authentically “Senegalese” history – one that communities can embrace even as the country’s urban cores move full steam ahead – provides an attractive vehicle for rural communities to reclaim their status in the polity as their sons and daughters move away and as secular authority is questioned in new ways. Such dynamics are surely not to be limited to historically centralized areas; as Doquet (2006, 311) shows of Dogon communities in Mali, local officials in the commune of Dourou have actively deployed cross-village festivals and rituals, the success of which “testifies to an awareness by the population of communal unity. The gathering of villages on the same stage reinforces in their [local officials] eyes a new cultural identity.” How successful such efforts to reinvigorate or create these countervailing pressures will be remains to be seen, however.

In the end, I remain agnostic as to how permanent or fleeting the social institutions I have highlighted are likely to be. Group norms rely on habituated, shared spaces of action that keep shared history and norms alive. Decentralized governance – with its meetings, its consultative drafting of local government development plans, its attempts at improving local fiscal capacity – creates numerous venues and rituals that may alter and nuance the very meaning of these communities. In incongruent communities, we may be inclined to see this as a positive, while in congruent ones we may bemoan something lost. At least in the short term, however, the existence of a bounded understanding of community in the latter has certainly helped the local state take root in a crucial early period of state reform.

In the end, this book may be read most simply as a call to study institutional reform better by asking us to think seriously about questions of boundaries, scale, and spatial delimitation. What politics are engendered by any given mode of administrative statehood and what politics do these boundaries obscure? The outcomes of top-down reforms have been repeatedly shown to be contingent on social dynamics not legible from afar and about which we often go in knowingly ignorant. To the extent we seek to aid, to improve, or to render more accountable, we should seek first and foremost to understand how individuals understand their political obligations. By starting from the grassroots and looking up at the state, I hope the argument is read as a call to take seriously how actors perceive the nested boundaries of solidarity that are created by the political delimitations of their communities and the histories those boundaries inhabit.