Shifting visions of property under competing political regimes: changing uses of Côte d ’ Ivoire ’ s 1998 Land Law*

Land law reform through registration and titling is often viewed as a techno-cratic, good-governance step toward building market economies and depoliticising land transactions. In actual practice, however, land registration and titling programmes can be highly partisan, bitterly contentious, and carried forward by political logics that diverge strongly from the market-enhancing vision. This paper uses evidence from Côte d ’ Ivoire to support and develop this claim. In Côte d ’ Ivoire after  , multiple, opposing political logics drove land law reform as it was pursued by successive governments representing rival coalitions of the national electorate. Between the mid-  s and  , different logics – alternatively privileging user rights, the ethnic land rights of autochthones, and ﬁ nally a state-building logic – prevailed in succession as national government crafted and then sought to implement the new  land law. The case underscores the extent to which deeply political questions are implicated in land registration and titling policies.


I N T R O D U C T I O N
Land law reform via registration and titling continues to be viewed by many as a technocratic, good-governance step toward building liberal market economies and depoliticising land transactions. In actual implementation, however, land registration and titling programmes have often been driven by very different priorities and dynamics. In the African countries of Côte d'Ivoire, Kenya, South Sudan, Uganda and others, recent land law reforms that promote formalisation of land rights have been used in partisan, starkly redistributive, and sometimes politically explosive ways. In these settings, land law reforms have often been as much about extending and consolidating political power as they are about limiting the state and freeing markets. In some situations, they have been as much an amplifier of social conflict as a solution to it. Léonard et al. (: ) argued that in the domain of land policy reform in Africa, political logics and control strategies often determine the course of change. This paper extends their argument, using the case of Côte d'Ivoire to show how regime change at the national level produced dramatic national-level shifts in the content, political purposes, and political uses of land law. Different logicsalternatively privileging user rights, the ethnic land rights of autochthones, and finally a statebuilding logicprevailed in succession from the mid-s to . Even with a new land law in  that aimed at land certification and titling, the politics around national strategies of implementation were at odds with liberal visions of market-building and individualisation.
In Côte d'Ivoire from the mid-s to , each change in the locus of political (partisan) control at the national level produced reversals in the thrust of land policy. Reversals were visible in the government's strategies to advance the land rights registration and titling agenda; in village delimitation and decentralisation strategies; and in legal adjudications of land law. This paper traces these shifts in national priorities, underscoring the deeply partisan thrust of land law reform in each era and in the larger state-building purposes that land law was harnessed to serve.
The analysis is organised chronologically. Part I is a brief overview of the pre-reform era, the s to the early s, when land policy under the country's first president, Félix Houphouët-Boigny, was a 'land-to-thetiller' (or user-rights affirming) policy that was designed to promote land-pioneering in the zone of the rapid expansion of export-crop  C A T H E R I N E B O O N E production. Houphouët's death in  set in motion a reconsideration of Ivoirian land policy.
Part II centres on mid-s pilot projects aimed at promoting the eventual registration (and perhaps titling) of landholdings. These took place amidst the national-level power struggles of -. Power struggles found expression in a new land law, Law -, which contained an array of potentially or inherently conflicting provisions for assigning land ownership rights on the basis of autochthony (i.e. by confirming customary land ownership claims), and then dissolving these away under a World Bank-inspired individualisation and titling plan. Opposition leader Laurent Gbagbo, with an electoral base in the Ivoirian centre-west, was the leading national champion of the autochthony-affirming provisions of Law -.
Part III tracks the efforts of the regime of Laurent Gbagbo, who came to power in , to realise the autochthony-affirming promise of Law -. Land activism under the Gbagbo government was aimed explicitly at replacing the Houphouëtist land regime with a land regime that asserted autochthonous communities' land ownership rights over those of migrants. The empowerment of autochthonous communities was to be confirmed through a sweeping political and administrative decentralisation.
Part IV shows that after the fall of the Gbagbo regime in , his successor, Alassane Ouattara, turned the tables once again in the land domain. The Ouattara government, with a power base in the Ivoirian north and among the northern diaspora of migrants in the southern Côte d'Ivoire, neutralised the anti-migrant thrust of the land law. In its first term (-), Ouattara's government used the provisions of the  land law to promote the recentralisation of power, and to extend and deepen the state's presence in war-torn localities along Côte d'Ivoire's turbulent borders in the far western regions of the country. In -, the government opened the doors to demand-driven registration and titling of land in a -mile radius around Abidjan, but did not invest political capital in pursuit of the land privatisation agenda.
The paper contributes to knowledge about land politics in general and Côte d'Ivoire in particular by offering a national-level, political analysis of land policy since . It fills a gap between two different literatures on Côte d'Ivoire since , one on politics and the other on land use. Political analysis in the s focused mostly on the national crise socio-politique that brought Gbagbo to power, crystallised in a  rebellion, divided the country into a government-controlled south and rebel-controlled north for five years (in a situation of 'neither peace nor war'), and culminated in disputed elections in  and Gbagbo's removal from power by rebel forces aided by the international community in .  The fundamental role of land issues and of the  land law in precipitating the crisis is a prominent theme in these studies. Yet there has been very little scholarly analysis of the Gbagbo government's efforts to actually deliver on the promise of Law - in the government-controlled south in the - period, or political analyses of land policy under Ouattara during his first term in office (-). Indeed, much political science commentary conveys the impression that the  land law and its political promises were swept away in the chaos of the crisis and war years. A separate stream of research on post- Côte d'Ivoire focuses on micro-level changes in land use and land contracting in the Ivorian south.  This work tracks bottom-up drivers of change in the export-crop producing smallholder sector. It treats the  land law as mostly non-implemented in the - period, with the important exception of the sharp uptake in land titling opportunities by members of the national urban elite in the region around the capital city of Abidjan.  The analysis here begins to fill these gaps. It draws upon local news reports, grey and secondary literature, Ministry of Agriculture internal documents from - and - that were consulted in Abidjan by the author. The record of land policy activism and its links to reforms of local governance structure during this period have not yet been described or analysed in the scholarly literature on Côte d'Ivoire. Approximately  unrecorded semi-structured interviews with actors in the Ivoirian land administration sector in  and  (in Abidjan and Bouaké) provided background information to supplement the document-based analysis.
The Ivoirian experience is significant for understanding contemporary land law reform in African countries (Boone ). It shows how land law reform initiatives aimed at land registration and titling can have domestic political drivers and effects that are only partially or even tangentially related to liberal visions of land privatisation, commodification and market-making. At the same time, it is a strong corrective to political science analyses that insist that partisan competition for state power in African countries is about ethnicity and patronage only, and void of real, programmatic stakes. In Côte d'Ivoire the opposite is the case: struggles over land law have been intertwined with partisan conflicts, electoral struggles, and actual battles over the national trajectory writ large.


In the pre-reform era, from s to the early s, land policy under Félix Houphouët-Boigny was designed to promote land-pioneering in the region of rapid expansion of export-crop production in the country's centre-west and south-western forest zones. Above all, it enforced user rights in this zone (i.e. upholding the rights of whoever was using the land, even if the users had no ancestral claims to it). In doing so, land policy promoted the land pioneering interests of the ethnoregional groups who were represented by the Houphouët regime, i.e. the Baoulé of central Côte d'Ivoire and the northerners with whom they were allied within the ruling Parti Democratique de Côte d'Ivoire (PDCI). Baoulé and northern farmers and farmer-traders vigorously colonised the south-west forest zone with active state support and backing. To achieve this, Houphouët built a strongly authoritarian and centralised state which repressed indigenous groups (and their customary land claims) in the south-west. A voluminous literature documents and analyses this process.  Côte d'Ivoire was one of four francophone West African countries to begin implementing a Plan Foncier Rural (PFR) project in the late s. With active support from the World Bank and the French Cooperation, the PFR envisioned a process of definition of village territories and lands (terroirs villageois), registration of land rights, creation of the institutional framework necessary for the eventual conversion of customary rights into statutory land rights, and land use planning at the village level.  In western Côte d'Ivoire, land pressure was mounting, along with political tensions between autochthones and in-migrants over land rights. The general economic conjuncture contributed to rising social and political tension around land. The mid-s were a time of economic recession, shrinking employment and IFI-imposed austerity which choked off growth of the cities and sent the middle class into a tailspin. In response to recession and fiscal austerity, unemployed urban youth began to 'return to the land', but many who returned to the western regions found that in-migrants now occupied most of their families' land. One proximate goal of the PFR in the west, as least as far as the Ivoirian Ministry of Agriculture was concerned, was to release some of this pressure by identifying underutilised tracts of land that could be made available to unemployed youth. Around Soubré  S H I F T I N G V I S I O N S O F P R O P E R T Y terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0022278X18000198 in the south-west, plans to build a new dam on the Sassandra River created a separate need to survey local populations and landholdings which dovetailed with some of the proximate objectives of the PFR.
Between  and the end of , the larger political context also shifted. Côte d'Ivoire returned to multipartyism in , and in , Houphouët died and power passed to his hand-picked successor, Henry Konan Bédié. During this period, the PFR progressed in five pilot zones. About , ha of village land was mapped, with village boundaries and existing land rights traced out on the basis of local interviews: Béoumi (, ha), Korhogo (, ha), Abengourou (, ha), Soubré (, ha) and Daloa (, ha) (Chauveau et al. : ). In January , the PFR matured into the Programme National de Gestion des Terres Rurales (PNGTER), with a wider mandate and a greater emphasis on individual titling per se (see Stamm ).  The very process of 'village land surveys', the basic tool in the PFR toolkit, stoked tensions in the central and western parts of the Ivoirian forest zone. Surveys asked village elites and household heads to inventory their landholdings, and to specify who owned what, who had gained access to the land via whom, and who had rights to stay on or dispose of the land.  This unfolded in the context of intensifying partisan competition and political polarisation at both the local and the national level. Conflict was especially acute in the centre-west and west, where politicians from the ruling PDCI and their newly unleashed partisan rivals, organised in the centre-west by Laurent Gbabgo under the banner of the Front Populaire Ivoirien (FPI), instrumentalised indigene-stranger tensions to mobilise rival electoral constituencies. The political, economic and social climate deteriorated. In the  elections, as the PFR 'pilot' phase was coming to an end, , in-migrants (most of them Baoulé) were expelled by grassroots militia from the Gagnoa area in the Ivorian centre-west (Babo & Droz ). A large scholarly literature analyses this difficult history.  As land tensions and partisan struggle intensified, the World Bank pressured the Côte d'Ivoire government to move forward on a land policy aimed at registration and titling as part and parcel of 'second generation structural adjustment' reforms. Politicians and representatives from the regions most affected by land-related tensions (the centrewest, south-west and south-eastern coastal zone around Bonoua) became actively involved in formulating and 'sensibilising' populations around reform proposals, seeing promise and risk for themselves and their constituencies in the prospect of a major new land law. In December , the ruling party, the PDCI, introduced a new land law that would largely supersede the PFR and the PNTGER.  It aimed at titling and registration, and unleashed bitter controversy. In National Assembly debates, the ruling party was split by the proposal to confirm 'customary rights ceded to a third party' as property rights that could be registered and titled. PDCI delegates from the west joined the inflamed opposition party, Gbagbo's FPI, in protest.  The FPI saw this clause as a green-light to those who would dispossess autochthonous westerners of their land rights. Suspension of National Assembly deliberations gave the FPI time to compose a counter-proposal that (a) inserted a nationality clause into the law (reserving the right of property ownership to Ivoirian nationals) and (b) vested the power to confirm the initial allocation of land ownership rights in autochthonous elders at the village level. This satisfied the FPI and other members of parliament from the west who had opposed the first version of the law, and were intent on ensuring that guardians of autochthones' interests would have a strong hand in determining how the new law was implemented at the local level. Some observers remember that some Ivorian legislators were very concerned to block or pre-empt large-scale sales or concessions to foreign (non-African, corporate) investors. Meanwhile, the law received some support from deputies who wanted to create a path to registration and titling to facilitate private acquisition of land and investment in agriculture.  Law - is remembered as having passed unanimously, but if it did so, it passed in a very divided parliament. Some deputies apparently began to renounce it as soon as it was passed. One analyst reported that 'The Baoulé lobby opposed it' (Boni n.d.). If this is true, it is not surprising: the law dramatically weakened the position of in-migrants to the centre-west and south-west, be they Baoulé, Ivoirians from the north, or foreigners. Implementing decrees (décrets d'application) were promulgated on  October .
There was an upsurge in assertiveness of autochthones around land rights in the centre-west and far west. A dramatic expression was the November  expulsion of Ivoirien northerners (Lobi) and Burkinabe from Tabou.

I I I . L A N D P O L I C Y U N D E R T H E G B A G B O R E G I M E : S C A L E S T I P P E D D E C I S I V E L Y I N F A V O U R O F A U T O C H T H O N E S
The  land law overturned a status quo in which the government enforced the principle that working the land (mise en valeur of the land) was the source/origin of land use rights.  The new law was widely interpreted as mandating first registration of all unregistered rural land (approximately % of the national territory) in the name of autochthonous ownersthis was to happen throughout the national territory within a period of  years.  It reversed Houphouët's legacy of enforcing the primacy of user rights in the most dynamic zones of export-crop production in the centre-west, south-west and far west, and thus represented dramatic inversion of the rapports de force between in-migrants and autochthones in these regions. At the end of the -year registration period, all unregistered land was to become the property of the state.
Much commentary holds that Law - was not implemented or never implemented.  This statement is true in a narrow sense: by , only a miniscule number of land certificates had been issued for rural landholdings. Implementation more broadly understood, however, entailed concerted activism on the part of the Gbagbo regime to prepare the ground for implementing the  law, and to seize the possibilities for an autochthonous rights revival that the law seemed to authorise. Under Gbagbo, government policy aimed at transforming rapports de force between indigenes and migrants. Its thrust and intent were visible in three domains: (a) court cases; (b) the delimitation of village boundaries and the setting up of land rights-based political hierarchies within and among villages, and issuance of Certificats Fonciers (land certificates); and (c) efforts to pursue political and administrative decentralisation. Each is discussed below.
The scope of these efforts was limited in geographic and financial terms, largely due to the politico-military crisis of - and its aftermath, followed by the final defeat of the Gbagbo regime in . Even so, we see in these efforts a concerted attempt to use Law - to affirm autochthonous land rights and embed these in a larger political and legal order that gave primacy to the local.

Court cases
As Dagrou (: -, -) recounts systematically, the autochthony-affirming content of Law - was stridently asserted in a series of court cases between  and , notably from Daloa, a strongly pro-Gbagbo region. A June  court case in Cour d'Appel de Daloa ruling reads that 'given the law - holding that only les populations autochtones riveraines (in a locality) have use rights, we can deduce that such rights can never result from a simple mise en valeur, and above and beyond that, a non-Ivoirian foreigner who, under the terms of the law just cited, is not destined (n'a pas vocation) to become the owner of a parcel of rural land'.  In , the Tribunal de première instance at Abengourou ruled on the 'nullité absolue' of a sale contract that has not been notarised. The Cour Suprême followed up, issuing in December  an Arrêt confirming that no customary landholder has the right to cede property rights to a third party, even if the sale is notarised, further closing the door on legal recognition of past sales to non-autochthones. The Daloa court specified in June  that 'customary rights in no way proceed from the mise en valeur of the land'.
In April , a Sassandra ruling, confirmed in Daloa, went to the Cour Suprême which ruled that mise en valeur confers no customary use right (much less property right), and that the true property holder in the disputed case is the one who exercises customary rights 'sur la parcelle de ses ancêtres par voie de succession, comme la consacre la loi n. - du  december '. In June  the Daloa court proceeded to nullify notarised land sales and expel occupants from litigated parcels. In April  the very active Daloa court also ruled that informally documented sales (acte sous-seing privé, presumably what are called 'petits papiers' in the land literature) are 'absolument nulle'.

Village boundaries, village hierarchies, and issuance of CFs
In , the PNGTER was extended for the - period. This programme, run by the Ministry of Agriculture and the prefectoral corps, took up the project of delimiting and demarcating village territories. Although Law - did not lay out procedures for village delimitation, delimitation was a precondition for the formation of village land committees (CVGFR) with authority to assign land rights within territorially demarcated units. For the Gbagbo government, the stakes in village delimitation were three-fold: formalising and officialising the primacy of autochthonous land rights; creating official village hierarchies which gave primacy to state-recognised autochthonous villages; and giving state recognition to 'naturalised' village origins and boundaries. Village and landholding mapping methodology was adapted from the procedures developed in the s under the PFR.  Issuance of land certificates followed from this and was underway in Gbagbo's last four years in power.
The first part of this sub-section describes this procedure, and the second analyses the patterns that emerged.

Gbagbo regime demarcation and registration procedure
The point-of-departure in delimiting village territories was the existing structure of territorial administration, in which the 'village territory' is the lowest-level administrative jurisdiction. Its political and administrative centre is the state-recognised village (village propre, or village érigé [erected] by the government), and the lowest-ranking agent of the territorial administration (subordinate to the sous-préfet) is the chief of the officially recognised village. The village territories demarcated in the PNGTER exercise measured between , and , ha in size (, ha = . square miles) and typically contain multiple, hierarchically ranked settlements. The officially recognised village is the seat of power with the village territory, and other settlements, often referred to as campements (camps) in Ivoirian parlance, are politically and administratively subordinate to (rattaché à) it. Administrative ideology and protocol is based on the idea that village territories are customary, historical or natural political homelands of some set of original inhabitants. In zones of in-migration, the campements or villages rattachés are settlements created by non-autochthonous 'outsiders' who settle with the permission of the indigenous local land authorities and the administrative chief. The in-migrants create their own, separate settlements within the village territory, at a distance from the villages propres. A land ownership hierarchy is thus embedded in officially recognised village hierarchies.
One of the stakes in the PNGTER delimitation and registration exercises was recognising or reforming pre-existing village hierarchies in the project zones. A campement or village rattaché that is granted proper village status (érigé en village) is 'liberated' from its former overlords, now holds land autonomously in its own right, and has its own village territory. Conversely, to lose village status and to be demoted to the status of a village rattaché is to be subjected to the authority of another group, and to have land access only with the permission of those who do control land with the village territory.
In the PNTGER project zones under the Gbagbo government, the state-led process of village delimitation, hierarchisation and land certification pivoted on the pervasive and long-maintained 'political- terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0022278X18000198 administrative distinction between 'autochthones', on the one hand, and 'allochtones de nationalité ivoirienne' and 'allogènes non-ivoiriens', on the other hand' (Chauveau & Colin ). The hierarchical structuring of villages and appointment of village chiefs was pre-decided, usually following the status quo but, as we shall see, not always. In advance of village surveys, sous-préfets organised the formation of Comités Villageois de Gestion de Foncier Rural (CVGFR, or Village Land Management Committees), presided over by the sous-préfets, who were to act as repositories of local knowledge about village histories, village boundaries, the allocation of customary land rights within village territories, and the granting of derived land rights to village outsiders. According to the PARFACI report (Varlet et al. : ), -% of the members of the CVGRF constituted in the  exercise were autochthones, including members of the chefferies (see also Colin : ). This is one indicator of the government's decision to throw its weight behind autochthones and a neocustomary vision of landownership.
With CVGFRs in place, an enquêteur appointed by the state was brought in to develop an official historique de la constitution du village using a item questionnaire, working in the presence of the CVGFR and other knowledgeable persons, and officialising the procès-verbal with the signatures of all CVGFR members. The next step entailed actually defining village boundaries, publicising the proposed boundaries in the village (and in neighbouring villages, the sous-préfecture office) to collect objections and endorsements, resolving disaccords, officialising village boundaries, placing boundary-markers and publishing an official map of the village limits.
Issuance of Certificats Fonciers and Certificats Fonciers Collectifs (CFCs) was to follow. CFCs were envisioned as instruments that would document extended family holdings, but would have a lifespan of only  years. Then the land would have to be divided up and registered and titled as individual private property.
The four project zones covered in , totalling approximately  villages, were in Abengourou, Daloa, Bonoua/Alépé and Soubré (see Figure ). The project areas were all politically accessible areas in southern Côte d'Ivoire (given the absence of the government in the north due to the political-military crisis) in which the FPI had either partisan control or a political foothold in a pro-FPI minority. The last column of Table I shows that two of the targeted zones, Bonoua/Alepe and Daloa, were areas of high partisan support for the FPI. Abengourou split its vote in  between the PDCI and the FPI, but voted for the FPI in . Soubré was a PDCI stronghold. Here, the PNTGER project built upon  the support base of a local pro-FPI autochthonous minority. Three of the four were PFR project zones, so groundwork for village delimitation and land adjudication had already been laid. Northern departments were absent from the list: the Gbagbo government could not access the rebel-controlled North until after ; it was a region hostile to the FPI and was also a region in which there was resistance to land certification.  The project zones were heterogeneous in terms of critical settlement history variables: the timing of in-migration, demographic balance between ethnic insiders and outsiders, and level of contestation over land rights, as summarised in Table I. In general, timing of in-migration is earlier in the east, having taken place in the colonial period, than it is in the west. In the west, the migratory in-flows were more recentthey happened in the post-colonial periodand the demographic weight of the in-migrants (ethnic outsiders) is higher. Levels of contestation over land rights were also much higher in the west than they were in the east (as discussed above).  A review of approximately half of the  village dossiers prepared by PNTGER in - offers evidence of the government's commitment to give pouvoir décisif aux autochtones in the Abengourou, Soubré and Daloa project zones.  In Bonoua/Alépé, the land-use rights of long-settled inmigrants were shored up. The village enquêtes confirm the heavy insistence on documenting first-comer rights, and confirming their primacy through the public process of documentation and recording (in writing) individuals' consent to the official record. Enquêtes clearly differentiated between land-owning villages and subordinate, non-owning villages (officialising status of 'village without land' or 'village installed within the territory of village X' in Abengourou and Bonoua/Alépé, and of village rattaché in Daloa and Soubré). In 'landless villages' and villages rattachés, male household heads signed (or gave fingerprints) to foreswear any claim to customary land rights of ownership. This officialised hierarchical relationships among villages and among residents of the village territories.

Analysis of Gbagbo regime delimitation and registration
It is remarkable that in Soubré, this effect was achieved through inversion of the prevailing administrative status quo. In-migrants' status was demoted in order to elevate the autochthones. Administratively recognised, autonomous villages established in the s, mostly Baoulé, were demoted to the status of campements subordinated to officially recognised Bakwé villages. One marker of the in-migrants' reversal-offortune was the renaming of villages: Baoulé or northern-sounding village names replaced by Bakwé names. Because enquêtes historiques of villages were aimed at establishing autochthony or first-comer rights, settlement history was critical. Each village dossier contains answers to the  questions that made up the village history survey. Recorded answers are short and tend toward the formulaic. The official village histories locate the origins of almost all of the autochthonous villages deep in the pre-colonial periodgoing back to the s, s or s. Histories include accounts of relocations, especially for villages in the west (which are attributed to the search for better sites) and the date at which the settlement was érigé en village (i.e. officially granted 'village' status) by the colonial or Ivoirian administration.  Village histories record non-political and naturalised accounts of village limits, referring to a river, creek, bas-fonds, or the long history of entretien des pistes, roads or boundaries between villages. Mentions of disputes, boundary changes, administrative divisions, etc. do not appear. Agreement on village boundaries through this process was followed by investment in actually demarcating the villages. Approximately , geocoded bornes were set in place under the Gbagbo regime.  Official village histories tell the story of the land rights that the Gbagbo government sought to affirm in its implementation of Law -. They recounted 'usable histories' that supported the government's vision of Law -.
In December , a new Programme National de Sécurisation Foncier Rural (PNSFR) was launched to spearhead an ambitious expansion of the village delimitation and land registration programme, with World Bank and EU funding. It elaborated a plan for mass communication and sensibilisation about Law -. In , Desiré Zalo, then Director of the Direction du Foncier Rural (DFR), reported at an Abidjan conference that  villages (out of a total of , in Côte d'Ivoire) had been delimited so far, that  were in the process of delimitation, and that  were scheduled for delimitation in .  The final balance sheet for village delimitation and land certification under the Gbagbo government was as follows: approximately  village delimitations (covering . million hectares of territory) were completed,  and  CFs had been issued.  Of these CFs,  (% of the total) were issued in Abengourou and Agnibilikrou,  in Soubré (%) and  in Daoukro (%). The remaining  CFs were scattered around the south. Only  CFs (%) were issued in Gbagbo's home region.  Most of the certified parcels are small-to medium-sized by smallholder standards. Approximately % of the certified parcels in Abengourou and % of the parcels in Soubré were under  hectares.

S H I F T I N G V I S I O N S O F P R O P E R T Y
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0022278X18000198 A few CFs for very large holdings were issued to political dignitaries in the West.
The Abengourou region, the only region in which more than a few CFs were issued, was surely the easiest place to certify parcel ownership, given the generally low levels of land tension, long and uncontroversial settlement histories (see Table I), relatively wealthy landholders by Ivoirian standards (Gastellu , ), and the historically high levels of literacy and rural prosperity which characterise this region.  Chauveau & Colin () commented that land certification was easiest and progressed furthest where it was least needed, either to mitigate conflict or promote land transactions. Daloa and Soubré make this point by way of counter-example: conflict levels were high and land certification stalled.

Decentralisation under the Gbagbo regime
The Gbagbo regime's land agenda was integrated into a broader vision of state restructuring that would promote localism, autochtonie and ethnic prerogative. The government undertook to remould chefferie law to create a chieftaincy hierarchy that would be comprised of already existing, first-level chefs de village (members of the CVGFR), complemented by second-level chefs de tribu, third-level chefs de canton and fourth-level chefs de province (at the departmental level). Gbagbo's Minister of the Interior, Désiré Tagro, undertook consultations 'with kings and traditional chiefs' to promote the official effort to 'bringing back the old structures' (remet en scène l'ancienne organisation).  An elaborate plan for reorganising territorial administration furthered this vision of restoring local prerogative and valorising communal initiative. In , a 'strategy for decentralisation and development (aménagement) of the national territory' had been adopted by the Côte d'Ivoire Donors' Roundtable. It formalised the goal of total 'communalisation' of the countryi.e. conversion of all sous-préfectures to communes, so that each would have its own elected government. This idea lay dormant until , when the Gbagbo government introduced a decentralisation strategy to be enacted through a series of laws and decrees in , , ,  and  that went even further than the  proposals in enlarging and deepening the scope of decentralisation (Traoré ). A  law (implemented in ) suppressed the old 'communautés rurales' and created new political jurisdictionsincluding villes, departments, districts and regionsand a new administrative hierarchy. As Traoré (: ) describes it, this 'marked the massive creation of new communes and new departments … In effect, by decree, the number of departments increased from  to  and the number of communes in the country went from  to ,, thus achieving the almost complete communalisation of the national territory.' Communessome of them large villageswere given financial autonomy and a mandate to organise collective life, promote local development, modernise the countryside [le monde rural], and manage land and other local resources.  This 'émittement territorial' (breaking up of the national territory into ever smaller units) was supposed to create , jobs for new agents municipaux to be hired by local elected politicians (Traoré : ).
Possibilities for effective realisation of Gbagbo's vision of a new state architectureone in which the local power-holders were strengthened, ethnic institutions were valorised, and resources and prerogative were devolved to the local levelwere undermined by a broader crisis of state and economy after .  Gbagbo lost the disputed October  national elections to Ouattara. In April  Gbagbo was finally dislodged from power through military action by the Ouattara-aligned Forces Nouvelles rebellion assisted by international action. 

I V . L A N D P O L I T I C S U N D E R O U A T T A R A : T A B L E S T U R N E D
Under the Ouattara government, the tables were turned. Ouattara's election in  and installation in office in April  represented a clear victory for those defending the rights of in-migrants in southern Côte d'Ivoire, and clear renunciation of the autochthony-promoting vision of Gbagbo and the FPI.  The new government acted decisively to recentralise state power and to use possibilities inherent in the  land law to achieve a new set of objectives.
Chauveau & Colin () describe a brief initial impulse on the part of the Ouattara government to discard the  land law, or to revise it completely. In , the government announced that Law - still stood, but with two reforms.  One prolonged the period of application of Law - to . The second revised the nationality code to make naturalisation easier. The FPI (absent from the legislature due to a boycott of legislative elections) and elements of its strong regional constituency in the west protested the nationality law in strongest terms. Laurent Fologo, an FPI firebrand, excoriated the Ouattara government  S H I F T I N G V I S I O N S O F P R O P E R T Y terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0022278X18000198 for undertaking to 'by the wave of a magic wand [through naturalisation law], expropriate the land of Ivioriens'.  Implementation of the  land tenure law remained a priority of Ouattara's backers in the EU and World Bank, surely because they subscribed to the general consensus that land problems had played a key role in fomenting devastating civil conflict in Côte d'Ivoire, and because they saw registration and titling as clearing the way for largescale investment in agriculture. Chauveau & Colin (: -) criticised the donors for maintaining 'a naive faith in land titles' and seeing the land cadastre 'as the holy grail'. Donors pledged about % of the estimated cost of delimiting all of Côte d'Ivoire's , villages, setting up CVGFRs, and issuing the estimated , CFs that would be necessary to complete the task. As part of this effort, the EUand World Bank-funded PRSFR was resurrected in  as the Programme d'Appui à la Relance des Filières Agricoles de Côte d'Ivoire (PARFACI). This relance called for the improvement of the legal framework, training of actors involved in implementing the [land] law, securisation of village lands by delimitation of terroirs villageois, the creation of a digital cadastre (Système d'Information Foncière)  and the relaunching of the implementation (mise en oeuvre) of the  land law (Leroy et al. ).
Yet during Ouattara's first mandate (-), the Ivoirian government did not embrace these priorities. Ambitions for general land certification and private land titling were restrained. Much landrelated conflict still simmered in the rural areas. In the West, where , people were killed in land-related violence in -, the flames of civil war had barely subsided. Land-related violence continued throughout ,  and . Activism to implement (or indeed, to revoke) Law - could have stirred up conflict that would be difficult to contain in southern Côte d'Ivoire. In northern Côte d'Ivoire, to promote Law - would have been to antagonise core loyalists.  The overriding objectives of the Ouattara government in - were to tamp down civil unrest, protect its partisans in the rural and urban areas, and maintain control over the national territory.
Under the first Ouattara government, the political possibilities inherent in Law - were used most concertedly to promote state-building objectives, rather than the donor's objectives of land titling and promotion of land markets. The government's priorities were visible in three domains. First, in the domain of land certification, the government developed the legal/bureaucratic machinery (disposatif legal) required to respond to individual demands for land certification and titling, rather than pushing for state-led titling (except in two local zones). Second, the main thrust of the village delimitation process was redefined. Under Gbagbo, the process had been led by PNTGER and the Ministry of Agriculture, and was aimed at allocating and securing autochthones' land rights. Under Ouattara it was placed under the control of the Ministry of the Interior and harnessed to the 'redeployment of the state' and rebuilding of the territorial administration in war-torn western Côte d'Ivoire.  Third, in the domain of decentralisation, Gbagbo strove to create a decentralised system anchored at the grassroots in local autochthonous communities and CVGFRs. The Ouattara government undertook a decisive recentralisation of authority.

Demand-driven certification
Throughout almost all of the national territory, the Ouattara regime adopted a laissez-faire approach to land certification, waiting for demands for land registration to percolate up from commercially minded actors and investors. Demand-driven land certification was highly localised and concentrated in zones of new inward investment. Large-scale investors pushed some land certification. Foreign corporate investors covered the costs of issuing Certificats Fonciers Collectifs (CFCs) to a few villages adjacent to land concessions destined for palm oil and sugar plantations, in order to clarify boundaries. In Aboisso, the Ministry of Agriculture promoted the certification of outgrowers' plantations on palm oil schemes (with costs of the CF deducted from producers' income by the millers).  There was a similar case in Tabou. Yet the main drivers on the demand-side were what Tarrouth & Colin () call cadres acquéreurs, or the acquiring salaried class or more generally, the national urban elite. Of a total of  CFs covering a total of , hectares issued between  and April ,   (%), accounting for a land area of , ha, were for parcels within an - km radius from Abidjanin Agboville, Lagunes, Alépé. The average parcel size in Agboville was  ha. Much of the investment was going into the creation of hévéa (rubber) plantations.

State-led certification
State-led land certification continued in two local areas. Most was concentrated in EU-and WB-funded project zones in Abengourou and in  S H I F T I N G V I S I O N S O F P R O P E R T Y terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0022278X18000198 the Soubré dam project zone. In Abengourou, an EU-funded registration project aimed at issuing , CFs in sous-prefectures delimited by PNTGER in . By April ,  CFs had been issued in Abengourou and Agnibilekrou (with average parcel sizes of  and  ha, respectively), with costs of certification covered by the international donors. The Abidjan hinterland (the zone of demand-led certification) and Abengourou-Agnibilekou (the zone of state-led certification) accounted for over three-quarters of all CFs issued as of April .
There was a secondary zone of land-certification in the conflict epicentre around Duékoué in the Far West. The government's land certification project was financed by the African Development Bank (ADB). By April  about  CFs had been issued in Duékoué and its surrounding villages. By September , about  applications (dossiers) were in various stages of processing.  The Duékoué projects were surely aimed at stabilisation and re-establishing order in this zone.
Critics of the government's slow process overall on the certification front argued that the lengthy ( steps from the beginning of the certification process to the actual issuance of a land title) and costly registration process discouraged smallholders from seeking certification (Varlet et al. : , ). There was some urgency to the matter: under Law - and its  extension, land not certificated by  will be considered 'sans maître' or ownerless and will become property of the state.

Village delimitation in the far west
During Ouattara's first mandate, the goal of using village delimitation to institutionalise the primacy of autochthonous land and political rights was dropped. Processes of village delimitation and land registration were disassociated.  Under Ouattara from -, security and state-building logics drove the agenda. Village delimitation was employed as a tool rebuilding the territorial administration in the wartorn West.
In , a government ordinance specified procedures for delimiting village territories.  Shift from the term terroir to the term territoire signalled a shift away from the land-use and tenure concerns of the PFR and Gbagbo periods, and to a preoccupation with administrativeterritorial control. The Ministry of the Interior replaced the Ministry of Agriculture as the lead agency. With financing from the ADB, the government undertook village delimitation in the Far West. A large part of the effort was focused on the critical border-crossing sous-préfecture of Toulepleu, where  village territories were delimited during Ouattara's first mandate.
Village delimitation was accompanied by the appointment of village chiefs. An urgent priority was to empower local chiefs who could arbitrate on land conflicts (backed up by the prefectoral corps) between war refugees and those who had occupied the land in their absence.  Village demarcation and the appointment of chiefs were measures aimed at re-establishing order, securing 'local cohesion' and encouraging the return of the refugees. 

Recentralisation under Ouattara
A September  ordinance reorganising the territorial administration was followed in March  by the suppression of  communes,  thus reversing Gbagbo's policy of communalisation totale. ' [The government] believes that  communes were created for political reasons and that they are not economically viable. The remaining  communes will stay.'  Villages were reattached to existing communes. Notre Voie, the FPI newspaper, reported bitterly that a Gbagbo regime 'policy that had been greeted enthusiastically by populations, offering them chance to 'take their own destiny in hand', had been 'killed in embryo … Under Ouattara, Côte d'Ivoire has turned its back on decentralisation.'  Districts with their own governors, another product of Gbagbo's restructuring of the national political and administrative structure, were also eliminated. In a  meeting with the prefectoral corps held at the presidential palace in Abidjan, Ouattara announced that he was not convinced that the district level of government was needed. The press announced 'the death of the districts'.  According to press reports, this was welcomed by the prefects who had viewed the district governors as compromising their mandate.  A new chefferie law, passed in , retained the four-tiered chiefly hierarchy in some regions, but not others. In Gagnoa, all of the hierarchical structure disappeared, leaving only the chefs de village. The Varlet report (Varlet et al. : ) explained that the goal of the government was to 'stabilise the chefferie'. Chiefs remain representatives of the state at the local level (village and canton), nominated by sous-préfets (sometimes after a local election), appointed officially by the Ministry of the Interior, and removable only with the permission of the administration.  As for the village-level land committees, the CVGFR, the conventional wisdom in the administrative services responsible for land certification in  was that these were not permanent local institutions. 'They will die along with the CFs' (i.e. when the CF is transformed into a land title).
Ouattara's first term in office ended in October . During this time, the implementation of land policy served a state logic of extending territorial control more than the market-making logic envisioned by the IFIs, or the smallholder securisation logic that is of most concern to NGOs. A great deal of land conflict continued to simmer in the localities of Côte d'Ivoire, including conflicts between communities over 'autochthonous rights', between communities and the state, between investors and communities, and between autochthonous landholders and those claiming to hold ceded land rights. 

C O N C L U S I O N
The adoption of a new land law in Côte d'Ivoire in  has been interpreted as a decisive commitment to building a true land market. The argument here is that in fact, for the last two decades, the goals of freeing markets and facilitating transactions among market actors have often been tangential to the domestic political drivers and to the most immediate effects of reform. The Ivoirian case underscores the extent to which the politics of land titling and registration implicates, and is even defined by, political questions, rather than purely technocratic or narrowly market-focused ones.
Under Gbagbo and during Ouattara's first term in office, land law reforms, in both content and implementation, were instruments in larger political struggles for control of the national territory and the national trajectory. Land law reform revolved around questions of citizenship rights (autochthony vs. user rights), state structure (decentralisation and political status of the local community) and state sovereignty (state control over population and territory), rather than politically neutral market-making as imagined by some international promoters of land law reform. The Ivoirian experience is a useful point of reference for understanding land politics in cases such as Kenya and Uganda, where many of the same tensions and issues animate national and subnational struggles over land policy and land rights.
A counter-trend is visible across a zone of a radius of about  km from Abidjan, where registration and titling is driven by private investors who want to acquire land, mostly for hévéa plantations and mostly about  ha in size (as described above). Much of the land in this zone was extracted from the customary tenure domain in the - period for the creation of foreign and state plantations.  Colin was describing this process when he observed that Law - was accelerating processes of class formation in the agrarian sector 'through selective implementation of the law … on a case-by-case basis in order to secure land transfers benefitting national elites (through purchase) or national or foreign agro-industrial firms (through long-term leases on certified and then titled land) (Colin : ). It does indeed square with the market-promoting vision embraced by the liberal promoters of land law reform. The question going forward is whether this will remain an enclave phenomenon, or whether it will diffuse into zones that are now under neo-customary forms of tenure. It remains possible that larger political-economic changes or land conflicts will swamp and subvert processes of privatisation and commodification.
A more immediate question has to do with the future of Law - itself. Under current law, holdings not certified by  become the property of the state. Many insiders in Côte d'Ivoire believe that this will never happen, that the law will be amended, and that the current provisions are only there to pressure smallholders to seek CFs. One could cite the disciplinary tone of the député (MP) in Divo, N'Guessan Denis, who told people in  to 'immatriculate your lands before , because if you don't, they will become the property of the state'.  Under Ouattara's second term, which began in November , government policy has been to accelerate the land certification process. Even so, it is possible that  will arrive, and that the majority of small-scale landholders in Côte d'Ivoire will find themselves in the same legal position as in , before the land law reform process began. Yet in contrast to , the government's control over most land across the national territory will be extended, strengthened and more firmly rooted in law.  State actors will be able to make longterm leases with investors throughout most of the national territory. In this situation, the security of most small-scale users' land rights will be determined by the political character of the regime in power, interactions between national and local actors and agendas, and the more general balance of state-society relations at the time. These larger political forces have gone far in determining the character of land law and policy to date. . Zalo (: ) refers to a different time frame in offering a balance sheet. He describes the PFR (and the PNGTER, presumably) as rolled out in three phases in nine zones over the - period: 'Over the course of this period, ,, ha were delimited, , parcels were marked out (levées), and  villages were covered.' . See Chauveau et al. (). . See note . . Law - was not the direct or intended result of the PFR or PNGTER. Chauveau (: ) explains that although the PFR had been seen as a step in the process of formulating new land legislation, (a) it did not include a plan to require formal registration of land certificates, and (b) it was still underway, having just scaled up from seven to nine project zones, in . Even so, in , the methodologies and procedures of the PFR and PNGTER were integrated into official procedures for land certification. This was formalised under Ouattara in decree no. -.
. Stamm () reported that the draft legislation contained a clause that would have revoked the user rights of those who could not prove that their land had been 'brought into use' (mise en valeur), but that this provision was withdrawn early on.
. Some observers believe that some members of the national assembly were eager to acquire titles and invest.
 . In fact the law provided for the possibility that an Ivoirian could receive a certificate and title for land in a region in which he or she was a non-autochthone. Foreigners were not barred from receiving land certificates, but titles to this land are established in the name of the state (Chauveau ; Boone ).
. Zalo (: ); Dagrou (: ); Colin (). . See also Boni (n.d.). . See Varlet et al. (: , ). Law - did not specify how village boundaries were to be delimited. So on this, the PNTGER programme operated in a vide juridique (juridical void) that was not filled until the Ouattara regime issued decree -. On the PFR process, see Colin et al. . There is silence around the history of French wars of pacification and forced regrouping and resettlement of populations in the West. See Dozon (). Enqueteurs recorded the respondents' information about when the sites were érigé en village. Answers from Daloa included, for example, 'By the Commandant de Cercle in ', and in Soubré or Bonoua, 'By sous-préfet or préfet so-and-so in  …'.