- Email: [email protected]
The application of the social tenure domain model (STDM) to family land in Trinidad and Tobago
Land Use Policy 28 (2011) 514–522
Contents lists available at ScienceDirect
Land Use Policy journal homepage: www.elsevier.com/locate/landusepol
The application of the social tenure domain model (STDM) to family land in Trinidad and Tobago Charisse Griffith-Charles ∗ The University of the West Indies, St. Augustine, Trinidad and Tobago
a r t i c l e
i n f o
Article history: Received 1 May 2009 Received in revised form 6 August 2010 Accepted 17 October 2010 Keywords: Land tenure Cadastre STDM Land administration Informal tenure
a b s t r a c t Family land is a form of communal tenure found in some of the countries of the English-speaking Caribbean inclusive of Trinidad and Tobago. It has been problematic to administer, is sometimes the source of land conflict and litigation, it has been seen as the cause of many land-related problems such as land degradation and fragmentation, and has therefore been targeted for eradication by land title registration programmes. Informal occupation of many years standing on state and private land is also widespread in Trinidad and Tobago. Recording the existence of these extant extralegal but legitimate rights would go a long way to improving the land administration and the security of tenure on these lands. The Land Administration Domain Model (LADM) is a logical and structured standard format for describing the land–human relationship, using an ontology that would allow for sharing of understanding, data and, eventually, software among jurisdictions. This standard is being developed, but has not as yet been finalised, by a technical committee of the ISO (The International Organization for Standardization). The Social Tenure Domain Model (STDM) is a refinement of the LADM, also being developed by ISO, related to describing more nuanced tenure and land administration components and arrangements that do not fit into the more general descriptions within the LADM, such as customary tenure and informal settlements. This paper charts a preliminary investigation into the applicability of the STDM descriptions to the particular land tenure situations of Trinidad and Tobago and to other countries in the Caribbean in which family land and other informal tenure forms are found. Describing these rights using the standard would then allow for the communication and understanding of the tenure situation and should support the eventual recordation of these rights. © 2010 Elsevier Ltd. All rights reserved.
Introduction Family land tenure derives from a cultural response to a shared history of colonisation and slavery in much of the Caribbean (Clarke, 1953; Smith, 1956; Besson, 1979; Griffith-Charles, 2006). It is defined to be a system wherein communal rights to land are derived from descent from a common antecedent. The origin of the tenure rights may be from initial purchase and subsequent unregistered inheritance or from original settlement. Rules of governance and membership may be unclear to or unacknowledged by the state and may also differ from family group to family group and from country to country within the Caribbean. This paper describes the extant tenure and registration situation in Trinidad and Tobago. It then examines the particular tenure situations that can be adequately accommodated by the Social Tenure Domain Model (STDM)
∗ Corresponding author. Tel.: +868 662 2002x2520; fax: +868 662 2002x3700. E-mail address: Charisse.Griffi[email protected] 0264-8377/$ – see front matter © 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.landusepol.2010.10.004
version of the Land Administration Domain Model (LADM), which is a standard being developed by a technical committee of the ISO (The International Organization for Standardization), and describes the particular components of the model that are applicable and what requires adaptation to the Trinidad and Tobago situation. It then concludes on whether the standard, when finalised, may be adopted in Trinidad and Tobago. As far as possible the current situation is described without proposing improvements or changes to the tenure structure that may not be implemented or necessary. Background Trinidad and Tobago is a small twin island state in the southern Caribbean of slightly over 5100 km2 in area and roughly 1.3 million in population making it a relatively densely populated country and thus increasing the necessity for a structured approach to land administration and management. More than 50% of the land is owned and managed by the state and this includes not only reserves and lands vested in state agencies but also lands
C. Griffith-Charles / Land Use Policy 28 (2011) 514–522
leased to private individuals and groups for residential, agricultural, commercial, industrial and other uses. While there is a nominally comprehensive cadastral index, and a land registry supported by land registration legislation, it is posited that 30% of parcels are unsurveyed and unregistered and some 47% of parcel occupants remain without adequate documentary security of tenure because of informal occupancy on state and private land or occupation of family land (Stanfield and Singer, 1993). Much of Tobago is said to be under family land ownership. The country has been proposing to perform a systematic titling programme for more than ten years and legal, institutional and technical infrastructure has been instituted to address this even though implementation has met with continuous delay. If systematic registration is performed, it may not be possible or advisable to individualise family land. When St. Lucia underwent systematic titling in the 1980s, with the intention, inter-alia, of individualising family land, the issue of family land proved to be insurmountable and other mechanisms were used to allow registration to continue (Stanfield, 1988). If Registration is not performed, some form of documentation of what exists is required to support security of tenure. There have been instances where land has been conveyed informally by family members or others without the knowledge of the family. Land in Tobago has become attractive as an investment especially to tourists and the escalating increase in value has proven to be an incentive to unscrupulous persons, even family members to fraudulently sell land to which they do not have rights. Residents on and owners of family land continue to demand some resolution to this issue of securing their rights. The Registrar General’s Department maintains both deed and title data in Trinidad and Tobago while the Surveys and Mapping Division maintains data on the spatial extent of parcels. There is no obvious link between the data held in both institutions for parcels held under the deed system while for parcels held under the title system, which number only some 15% of registered parcels, there is a reference number, noted on the survey plan that indicates where the original registration certificate is held in the registry. Since the title system is Torrens based there is no general map held by the land registry. For the deeds system, a survey plan is not legally required for a transaction to be registered but even when a survey plan is attached, the registration of the transaction is referenced sequentially and not by parcel. The land registration system can therefore be categorised as a centralised, primarily negative, state-supported, juridical and separated system based on fixed boundary surveys and sporadic adjudication, in the terminology collated by Bogaerts and Zevenbergen, 2001. Acknowledgement by the state of the differing tenure types and recordation of the rights held would help to provide security of tenure to the landholders in Trinidad and Tobago.
The STDM The STDM is an application of the more generic LADM which is a structured and standardised method of describing the land tenure relationships between land and humans. The STDM is a more flexible way of representing the actual tenure arrangements where the structured and individualised LADM is too general to describe the nuances of many existing informal and non-standard tenure relationships. The justification for modelling is twofold as stated by Lemmen and van Oosterom (2003); to provide a systematic way of describing the land–human tenure relationships and, if the model is actualised in software development, to allow interactions between different land registration systems (Lemmen and van Oosterom, 2003; van Oosterom and Lemmen, 2006). For the
515
Trinidad and Tobago context, attempting to portray the tenure in a model using the defined terminology would compel the examination of what exists to achieve a better understanding of the extant relationships, and, also, the structure would allow the possibility of interactions between the data from areas that would fit into the more generic core cadastral model and the data from family land and other informal areas that would not. Comprehensive and complete data would support land administration and land policy decision making. Researchers are currently applying the LADM and STDM models to particular jurisdictions such as Indonesia, Tanzania and Portugal to test its applicability in specific cases (Ary Sucaya, 2009; Mithofer, 2006; Hespanha et al., 2006). Each country or jurisdiction has to develop the application of the generic model to the particular situation as legislation and practice may restrict or expand the possible instances, attributes, authorising documents and other variables in the model. Testing will allow the model to be improved to accommodate alternative situations and will also assist individual countries. The standard utilises UML terminology and graphics to diagram the land administration relationships between human and land. Specific areas for attention Augustinus et al. (2006) look at specific problems related to Africa that the STDM is required to address and question whether the standard is flexible enough to accommodate the issues of land rights relevant to the poor in that environment. Some of the issues highlighted there may not arise in the particular case of Trinidad and Tobago because of the differing contexts. The specific issues that may be impacted by the use of STDM model in Trinidad and Tobago need to be identified and examined to determine the advisability of using the descriptions in the standard. Coverage Trinidad and Tobago has an established cadastre, illustrated by an index, comprising of an estimated 300,000 parcels which, because of the gaps, errors and lack of currency of the data, defies, in many instances, conclusive determination of legal status. The different types of tenure regimes, however, must be incorporated or linked into a comprehensive whole and this is one of the issues that the STDM must address. In Trinidad and Tobago individualised tenure held under either title or deed registration coexist with informal occupation of many years standing on state and private land and with family land ownership. There is, therefore, a continuum of tenure regimes from the state guaranteed registered title through the uncertain evidence of deed registration to the unrecognised legitimacy of family land and lastly to the uncertainty of informal occupation of varied duration. The different datasets in the information system would therefore need to be seamlessly merged into one spatial dataset especially since each tenure type is not physically grouped into a distinct area. This would effect comprehensive coverage of the country in the recordation or registration of tenure. Land right The nature of the land right on family land must also be able to be captured in the descriptions of the standard in an explicit fashion where these rules may never have been documented before. The rights governing use, restrictions on usage and responsibilities of the occupants in the family land scenario must be able to be described and the differences between different groups and their governance rules stated.
516
C. Griffith-Charles / Land Use Policy 28 (2011) 514–522
Rights holder The ability to describe who are the rights holders and what defines membership in the family land must be able to be included without necessarily providing a finite listing of the members. In some instances, acknowledged family members may have lived abroad for many years and may have lost contact with the other members but yet retain their rights. Rules for determining who the members of the family are may differ slightly from family to family and so flexibility is required for recording the authority by which membership is determined. Complete and comprehensive knowledge of the rights holders on family land is not usually available. St. Lucia addressed this issue in its systematic land title registration programme in the 1980s by titling the parcels in the name of ‘Heirs of X’ where ‘X’ was given as the last known antecedent to have formal title or a name commonly used to identify the family where a formal document did not exist. Another method used was to title with a list of all members available at the time and allowing rights to transact on the land to the first five persons listed on behalf of the group. Technical capacity Augustinus et al. (2006) also look at the advisability of having the purpose of the modelling be the construction of a solely digital database as opposed to retaining the ability to revert to a paper based environment. Two issues may arise from a solely digital database. One issue is whether there is an ability to reconstruct the databases in case of disaster in the centralised or decentralised location of the dataset and the other is having the ability to download to text and graphic documents in a readable and recognisable form by the community such as a certificate of title to land. The recoverability of the database after disaster can be addressed by storing the datasets online at remote locations and periodically revising the information. The example of the tsunami disaster in Aceh as compared with the hurricane in New Orleans and the differences in the subsequent recovery of the cadastral structure is instructive (Griffith-Charles, 2008a). Beyond the technical capacity to maintain the databases, there is a requirement for an institutional capacity to perform this function. The institutions at the level of the community must also be motivated to continue to perform this function even where the perception of security of tenure is heightened by the introduction of the system. Parcel definition The definition of the spatial unit is not available in many instances as parcels may not have been surveyed if occupied by settlement both for the informally occupied areas on state and private land and for the family land areas that cannot be related to a previously purchased parcel. The system must then be able to accept descriptions of extents of rights using indeterminate or fuzzy boundaries to define a unique parcel. In many instances formally registered parcels under the deed system have not been surveyed but the deed contains a verbal description by reference to adjoining owners. A related issue is that a family land holding may also consist of several scattered parcels that should be considered to be one unit for the purposes of recordation. Spatial and attribute precision The institution that maintains data on the spatial extent of the parcels in Trinidad and Tobago is separate from the entity that registers the rights. The parcels in the cadastre in Trinidad and Tobago
are not individually spatially referenced to the national framework. Therefore the requirements for coordinated points on the boundary of the individual parcels is not possible for the general cadastre and less so for the informal or family land areas. The topology of the surveyed parcels in the database is relatively high as surveyed parcels are referenced to surrounding parcels. Therefore, the parcel description in the standard for the formal areas as well as the informal areas would need to include the topology and not the coordinated location. Professional attitudes Accepting and adopting imprecision is necessary in such a standard that seeks to describe and hence acknowledge accruing rights. Both land surveyors and lawyers find difficulty in accepting information systems that are less precise and more flexible. The land surveyor seeks a standard of precise spatial information while the lawyer seeks a standard of precise attribute information on the nature of rights and specifics of the persons. While these standards are possible for the formal areas, the flexibility of imprecision needs to be built in to the STDM if it is to be effective for recording and managing the family land. The imprecision of recording the spatial extent of the parcel can be addressed by inputting calculated precisions depending on the method of survey or data acquisition. The imprecision of recording attribute information such as the number and spelling of the names of the members of the family must also be accommodated. Boundary points captured with non-survey grade GPS equipment or coordinated from remotely sensed imagery for defining areas of occupation must be able to be used. The standard should be simple but comprehensive enough to allay the fears of the professionals that the data contained are those that are salient for the purpose of securing rights to land. Process The process for updating of recordation must be a clear one especially if initial recordation, like the current title registration, continues to be sporadic and voluntary instead of systematic and compulsory. Decentralisation Decentralisation for the purposes of effective maintenance of the system is another crucial area to be addressed for the model. Lack of maintenance of the system soon causes an information system to lose much of its value. The family land has its own particular governance system for use and occupation of the land and it is this system that must be drawn on to maintain information on the status of use and rights. The system must therefore be able to be maintained by the family itself. This would encourage the maintenance of the systems and allow the state to have access to the information if the centralised databases need to be updated periodically. Applicability of the model The LADM and STDM provide a standard set of terminology and structures to model the relationships that already exist in land and those that can be conceived to exist and to be registered in some jurisdiction. Flexibility is also built in to accommodate other anticipated land-related objects and relationships that may be possible to register or record. Whether the standard covers all relationships that actually exist is what is required to be tested by individual countries or jurisdictions. If countries find structures that are not
C. Griffith-Charles / Land Use Policy 28 (2011) 514–522
captured in the model, the model is extensible to allow inclusion of other situations. The main objects in the LADM, which are described as the core classes, are the parcel related information (LA RecordedObject), the parcel (LA SpatialUnit), the person (LA Party) and the rights, responsibilities and restrictions (LA RRR) that link them (Lemmen and van Oosterom, 2003; van Oosterom and Lemmen, 2006). The STDM lists the latter three objects as the STDM SpatialUnit, STDM Party and SocialTenureRelationship to denote the fact that the attributes present in these objects differ from those legally prescribed in the jurisdiction but are those considered legitimate by the society (Lemmen and van Oosterom, 2003; van Oosterom and Lemmen, 2006; ISO, 2006). These STDM objects can be recorded as they exist instead of making them conform to the specifications required for legal registration. Changes to parcels, recorded objects, rights and persons through time are accommodated in the model by providing for descriptions of the time of beginning and ending of existence on an object or a right. Flexibility in the spatial extents of the rights is also accommodated. The LADM domain model is as shown in Fig. 1 which graphically depicts the abstract classes for the core objects of person, recorded object, parcel, and rights that are used in the standard. Fig. 1 also indicates the relationships between the objects in the LADM model. The notation indicates that each individual parcel can be associated with 1 or many rights, restrictions, responsibilities while each instance of party, whether a group or individual, human or abstract organisation or entity, is related to zero or more rights, restrictions, responsibilities as they apply to the one parcel. Coverage In the Trinidad and Tobago scenario the formal areas would contain the rights, restrictions and responsibilities that exist in
517
the English-based legal tenure system of the two registration regimes, title and deed, that exist in the country. Fig. 2 indicates the standard model as it is adapted to the situation in Trinidad and Tobago. The difference between the two types of formal registration would be captured in the attribute of the class LA Layer called LA T&TRegisterType to denote that it applies specifically to Trinidad and Tobago. The code list for this attribute would contain deed and title registration or Old Law and RPO (Real Property Ordinance) registration as they are more commonly known. This attribute can also accommodate the legal registration of land in Tobago or other decentralised registry or the recording of family land at community level. This would allow the different types of registration and recordation to be described by the same standard and therefore incorporated in the same dataset even though they may not be maintained by the same institutions. Currently there is only one centralised registry but documents may be submitted at the Tobago office to be taken to the central registry in Trinidad for authorisation and registration. The separation of the legal and spatial elements of the land registration system in Trinidad and Tobago can also be accommodated by the standard as shown in Fig. 2. The concept of the layer class can be used here to distinguish the different types of registration horizontally as well as the overlapping rights vertically so that they can eventually be spatially represented as distinct but coexistent registered or recorded entities. There is as yet no unique parcel number for parcels so the attribute Oid or object identification would first have to be assigned by the Surveys and Mapping Division based on the location of the parcel in the cadastral index at the largest available scale. Pending that implementation, the certificate of title volume and folio number can be used as the ID for titled land. The registered deed number for the first sale of a subdivided parcel can be used for deed registered parcels. Where family land parcels have a formal history and defined parcel, this deed reference method can also be used.
Fig. 1. Relationships between core objects in the model (after ISO document ISO/CD 19152.0).
518
C. Griffith-Charles / Land Use Policy 28 (2011) 514–522
Fig. 2. The Trinidad and Tobago formal registration situation.
In Fig. 2 the graphic indicates how the information maintained by each agency is conceptually linked to information external to its own organisation. The linkage to the database on the parcel is shown here with the assumption that a direct connection can be achieved. The attributes of LA RRR would allow all shares in rights to sum to 1 and the period of time for which rights are to be given to be described. The LA Party would require a unique identification which, since the equivalent of the social security number is not issued or used for unique identification in Trinidad and Tobago, may be the individual’s birth certificate number, records of which are maintained at the same Registrar General’s Office or a national ID number which is issued on application after the age of 15 and is recorded and maintained at another institution. Again, these datasets are conceptually linked. Many of the attributes possible in the standard for each object are not necessary or possible in the Trinidad and Tobago context. Comprehensive coverage of the cadastral fabric of the country can therefore be obtained using the model.
The social tenure relationships can be defined in a list called a code list which would list all the instances possible in the particular jurisdiction. Most family land inheritance rules conform to cognatic principles but allowances can be made for other types of tenure situations that may arise on investigation. For Trinidad and Tobago this list could include:
Land right
Recording the various permutations of the rights situations would increase the complexities of the database and reduce the flexibility of the practice. It is important, however, to document those factors that would affect the rights and could potentially lead to conflict if omitted, such as rights by descent along the male or female lines. Some minor variations may be left out of the database as negotiation and discussion amongst the family members has and will continue to resolve issues of rights that arise. In the informally occupied, non-family areas, rights to be recorded can include the transitory legal status of adverse possession. The status of occupation by adverse possession need not be directly recorded but derived from an attribute of the class SocialTenureRelationship. The declared beginning time of occupation can
Fig. 3 describes the informal situation and is a view of the classes in the system that relate to the family land and other informal tenure relationships. The nature of the SocialTenureRelationship can be also based on shares to members that would sum to 1 as in the formal case or the rights can be held collectively as a group with no requirement to define the share held, as there is no ability to alienate or subdivide a share in the typical family land situation. There is usually a hierarchy in the family land structure with decision making at the top entrusted to someone; sometimes the eldest or most educated (Griffith-Charles, 2006) but this hierarchy may not be possible to translate into shares.
• familyLandtenure1—where rights are by cognatic descent • familyLandtenure2—where rights are by unilateral descent male • familyLandtenure3—where rights are by unilateral descent female • tenancy—month to month not legally requiring registration • shortLease—leases less than the legal requirement for registration • informalOccupationState—occupation without valid document • informalOccupationPrivate—occupation without valid document • informalSubdivision—subdivided parcel without subdivision development approval
C. Griffith-Charles / Land Use Policy 28 (2011) 514–522
519
Fig. 3. The Trinidad and Tobago informal recording situation.
be recorded and if eventually verified by a court to satisfy the legal requirement for possessory title by adverse possession, the title can be converted to the formal registration if desired. In the family land areas rights, restrictions and responsibilities would need to be documented to reflect the range of possibilities currently afforded members of different family groups. The range of these rights, restrictions and responsibilities may be recorded in code lists and entered as attributes of the group. These rights, responsibilities and restrictions may be public or private. Formal, public rights, restrictions and responsibilities such as land use rights, reservations, building height allowances, etc. are usually kept and amended at the Town and Country Planning Division and are applied to particular parcels upon individual application for development permission. Currently files are kept at a parcel level on development applications and determinations and these can be now kept using these standards to allow interaction of data between departments. Private rights and restrictions are textually stated in transaction deeds and are part of the registration. These can be extracted and registered individually in the standard format. These rights should be recorded for family land as comprehensively as possible even where they may be in conflict with public rights and restrictions. While for formal land the rights may be a subset of use, occupy, develop, transfer, and bequeath, for family land a subset, and various permutations of these subset rights are available. To transfer and bequeath are not possible, and use, occupation and development, while available, are subject to governance principles in the family. Family land rights, restrictions, and responsibility classes are therefore sub classes of the LA Rights, LA Restrictions, and LA Responsibility classes of the LADM and will have additional attributes. Code lists for rights supported by legislation and rights supported by family rules must be drawn up as indicated with some examples in Fig. 4. All types of exist-
ing rights can therefore be described and accommodated in the model. Right holder The STDM Party is an instance of LA Party and so inherits the attributes of the superclass, which are the partyId, the type and the role. The STDM Party class will be used to describe all informal occupants while the further subclass of STDM GroupParty will be used to describe family land groups. An individual member can be recorded but it is not necessary for all members to be individually recorded to still retain rights. A member will be recorded if a record of an overlapping use right for part of a parcel for example is required. The STDM GroupParty is a subclass of STDM Party and will also inherit but however, will have additional attributes such as a family name to identify the group, Smith Family, for example, which will be given an individual ID. Fig. 5 shows the relationship between the party objects in the STDM standard. There is therefore a way of describing all right holders. Parcel definition Fig. 6 shows the relationship between the instance of spatial unit or parcel in the family land or informal area and the superclass LA SpatialUnit in the spatial unit package. The attribute referencePoint allows the inclusion of an approximate centroid coordinate for the parcel, which may be taken off of an index, aerial photograph, satellite image, or acquired by handheld GPS. The extent of the parcel may also be described textually. Family land holdings of several individual parcels held under the same rights should be recorded using the parcel complex description as opposed to the formal situation where a person holding
520
C. Griffith-Charles / Land Use Policy 28 (2011) 514–522
Fig. 4. Possible rights, restrictions, responsibilities on family land.
several different parcels should register each individual parcel as a separate registration instance. The implication is that the same acknowledged family land right covers the entire holding comprising of several parcels. Spatial and attribute precision and professional attitudes
Fig. 5. Party package.
The class LA SourceDocument in the LADM is named the SocialTenureInventory in the STDM. Whereas the LA SourceDocument class would have instances of documents that would be evidence of formal rights to land, the SocialTenureInventory may contain instances of differing authority on rights such as an original deed or title of an antecedent where one exists. Alternative documentation such as affidavits of neighbours attesting to duration of occupation can be listed as possible evidence of rights. Fig. 7 shows the relationship between the documents that define the evidence of rights in the informal or family land situation. The SpatialUnitInventory class allows recordation of alternative survey measurements to a different precision than the formal situation. If the professionals are aware that the original documents of deed and survey plan to which they are accustomed are accommodated in the source documents as scanned documents then this should allay their fears of change. One of the advantages of the family land communal tenure system is its flexibility as derived through negotiation and practice. If extant rules of governance are documented then as changes occur to the policies and practice of the rules, the logical structure of relationships will change. The model has the ability to time stamp an
C. Griffith-Charles / Land Use Policy 28 (2011) 514–522
521
Fig. 6. Spatial unit package.
instance of occurrence so that changes can be made over time without losing the ability to examine the status at a previous date. This is done through listing the time stamp as one of the attributes. Process The process for the initial recordation must also be addressed as it would require a methodical adjudication process as would any systematic title registration process. Publicity and participation would ensure that the information obtained upon which the
decisions about rights would be made is the most comprehensive information possible. In the face of a dearth of formal documentation, oral testimony together with physical evidence of occupation would be the deciding factors as to whose land rights would be recognised. The final answers to be recorded must be those acceptable to the communities themselves guided by national government governance of the process through adjudication personnel with conflict resolution experience. A participatory process would also encourage acceptance of responsibility from the communities for the governance process of the management and updating of the database. It is significant that in a study conducted amongst family land rights holders only 31% felt that the system was beneficial and should be retained (Griffith-Charles, 2008b). The majority of 69% did not support the continuation of the tenure form. However, this perception was arrived at because of the hardships currently felt by this group where their rights are under threat from counter claims to their rights and fraudulent dealings that challenge their security. Whether the tenure form eventually dies over time, under the pressure of modernisation and urbanisation, is not as important as preserving the existing family land rights that the current holders possess. Decentralisation
Fig. 7. Document package.
If the separate institutions follow the standard descriptions and maintain the structure of the data then it should be possible for data to be collected and maintained by the separate institutions and still be compatible. Family leaders can maintain the data following the logical structure in analogue fashion or using software when this is eventually constructed. It may not be necessary for the decentralised data to be maintained current at all times in the central version of the database. The national database can hold data up to the level of the community or the family holding and the community or family can maintain data below this level as advocated by discourses on customary tenure by Ankersen and Barnes (2002) and Fitzpatrick (2005). These internally governed units are described as being a ‘tenurial shell’.
522
C. Griffith-Charles / Land Use Policy 28 (2011) 514–522
Conclusion The general descriptions of the LADM have been developed into a more flexible model in the STDM that has provided the possibility of adaptation in jurisdictions where a structured formal model cannot fit. In Trinidad and Tobago informal and communal areas exist alongside the formal registered cadastre. A preliminary investigation was made here into the parts of the LADM and STDM models that would require to be adapted for the particular situation in Trinidad and Tobago. The preliminary examination appears to indicate that the descriptions in the proposed standard can fit the situation on family land in Trinidad and Tobago if adopted. While the STDM can be further adapted to fit the particular context in Trinidad and Tobago by documenting the range of instances possible for each class, practical testing would need to be done to determine how the model will work in practice. This testing will involve documenting in the field the governance procedures followed by individual family groups and undergoing a systematic recordation of the family land holdings using this model. It is envisaged that a pilot project can advance this work a step further to implementation and also encourage support for its adoption. It must be borne in mind, however, that the recordation process is supposed to protect the rights that already exist and not to disadvantage land holders by negating unrecorded rights and right holders that currently exist. A problem that may arise is the creation of the perception that since recordation lists some stated rights then other rights not listed have been proscribed. The recordation of some rights should not negate the legitimacy of other rights previously enjoyed. The recordation of some individuals or groups should also not negate the legitimacy of the rights enjoyed by other individuals or groups not so recorded. Standards can be documented and applied but may not be adopted by the society unless the society subconsciously accepts, individually and collectively, that the benefits of adoption outweigh remaining with the status quo. There are standards, especially in the area of IT that have been widely adopted and have become de facto standards without being imposed. It remains to be seen therefore whether this standard of terminology and modelling will become the de facto standard for describing and modelling land administration systems of formal and informal tenure. Acknowledgements Acknowledgement is given to the Erasmus Mundus 10 Staff Exchange Programme for supporting a visit to the ITC (International Institute for Geo-information Science and Earth Observation) and this investigation and to Prof. Paul van der Molen, Prof. Jaap Zeven-
bergen, Dr. Harry Uitermark, Christian Lemmen, and Dr. Arbind Tuladhar for helpful discussions. References Ankersen, T., Barnes G., 2002. Inside the polygon: emerging community tenure systems and forest resource extraction. Available online: conservation.law.ufl.edu/ pdf/extraction.PDF. Ary Sucaya, I.K.G., 2009. Application and validation the land administration domain model in real life situation (a case study in Indonesia). M.Sc. Thesis, Urban and Regional Planning and Geo-information Management (PGM). International Institute for Geo-information Science and Earth Observation (ITC), Enschede, The Netherlands, 92 pp. Augustinus, C., Lemmen, C., van Oosterom, P., 2006. Social tenure domain model: requirements from the perspective of pro-poor land management. In: Proceedings of Promoting Land Administration and Good Governance. 5th FIG Regional Conference, Accra, Ghana, FIG. Besson, J., 1979. Symbolic aspects of land in The Caribbean: the tenure and transmission of land rights among Caribbean peasantries. In: Cross, M., Marks, A. (Eds.), Peasants, Plantations and Rural Communities in the Caribbean. Department of Sociology, University of Surrey/Department of Caribbean Studies, Royal Institute of Linguistics and Anthropology, Guildford/Leiden, pp. 86–116. Bogaerts, T., Zevenbergen, J., 2001. Cadastral systems—alternatives. Computers, Environment and Urban Systems 25 (4–5), 12. Clarke, E., 1953. Land Tenure and the family in four communities in Jamaica. Social and Economic Studies 1 (4), 81–118. Fitzpatrick, D., 2005. ‘Best practice’ options for the legal recognition of customary tenure. Development and Change 36 (3), 449–475. Griffith-Charles, C., 2008a. Disaster rehabilitation and reconstruction: the role of cadastral records. In: International Workshop: Disaster Risk Management, The Role of the Land Management Professionals. CASLE, St. Augustine, Trinidad and Tobago, 6 June. Griffith-Charles, C., 2008b. The implications of SDI development to the family land communal tenure structure in Trinidad and Tobago. In: Proceedings of GSDI 10: The Tenth Annual Conference for Geospatial Data Infrastructure. 25–29 February, 6 pp. Griffith-Charles, C., 2006. The Persistence of Family Land in the Caribbean in Proceedings of the XXVI International Congress of Latin American Studies Association (LASA), San Juan, Puerto Rico, 10 pp. Hespanha, J.P., van Oosterom, P., Zevenbergen, J., Dias, G.P., 2006. A modular standard for the cadastral domain: application to the Portuguese cadastre. Computers, Environment and Urban Systems 30 (2006), 23 pp. ISO, 2006. ISO/CD 19152.0. Geographic Information—Land Administration Domain Model (LADM) Committee Draft. Lemmen, C., van Oosterom, P., 2003. Further progress in the development of a core cadastral domain model. In: Proceedings of FIG Working Week. 13–17 April, Paris, France, 13 pp. Mithofer, K., 2006. Development of a GIS-based land registry for Tanzania. Centre for Geoinformatics. M.Sc. Thesis, Salzburg University, Salzburg, Austria, 93 pp. Smith, M., 1956. The transformation of land rights by transmission in carriacou. Social and Economic Studies 4 (1), 64–82. Stanfield, D., Singer, N., 1993. Land tenure and the management of land resources in Trinidad and Tobago. Land Tenure Center, University of Wisconsin, Madison, WI. Stanfield, D., 1988. The LRTP: evaluation of procedures and outcomes. Overview of land tenure patterns and effects of the land registration and titling project. Land Tenure Center, University of Wisconsin, Madison, WI, vol. 1, p. 27. van Oosterom, P., Lemmen, C., 2006. The core cadastral domain model: a tool for the development of distributed and interoperable cadastral systems. In: International Conference on Enhancing Land Registration and Cadastre for Economic Growth in India, New Delhi, India.
Contents lists available at ScienceDirect
Land Use Policy journal homepage: www.elsevier.com/locate/landusepol
The application of the social tenure domain model (STDM) to family land in Trinidad and Tobago Charisse Griffith-Charles ∗ The University of the West Indies, St. Augustine, Trinidad and Tobago
a r t i c l e
i n f o
Article history: Received 1 May 2009 Received in revised form 6 August 2010 Accepted 17 October 2010 Keywords: Land tenure Cadastre STDM Land administration Informal tenure
a b s t r a c t Family land is a form of communal tenure found in some of the countries of the English-speaking Caribbean inclusive of Trinidad and Tobago. It has been problematic to administer, is sometimes the source of land conflict and litigation, it has been seen as the cause of many land-related problems such as land degradation and fragmentation, and has therefore been targeted for eradication by land title registration programmes. Informal occupation of many years standing on state and private land is also widespread in Trinidad and Tobago. Recording the existence of these extant extralegal but legitimate rights would go a long way to improving the land administration and the security of tenure on these lands. The Land Administration Domain Model (LADM) is a logical and structured standard format for describing the land–human relationship, using an ontology that would allow for sharing of understanding, data and, eventually, software among jurisdictions. This standard is being developed, but has not as yet been finalised, by a technical committee of the ISO (The International Organization for Standardization). The Social Tenure Domain Model (STDM) is a refinement of the LADM, also being developed by ISO, related to describing more nuanced tenure and land administration components and arrangements that do not fit into the more general descriptions within the LADM, such as customary tenure and informal settlements. This paper charts a preliminary investigation into the applicability of the STDM descriptions to the particular land tenure situations of Trinidad and Tobago and to other countries in the Caribbean in which family land and other informal tenure forms are found. Describing these rights using the standard would then allow for the communication and understanding of the tenure situation and should support the eventual recordation of these rights. © 2010 Elsevier Ltd. All rights reserved.
Introduction Family land tenure derives from a cultural response to a shared history of colonisation and slavery in much of the Caribbean (Clarke, 1953; Smith, 1956; Besson, 1979; Griffith-Charles, 2006). It is defined to be a system wherein communal rights to land are derived from descent from a common antecedent. The origin of the tenure rights may be from initial purchase and subsequent unregistered inheritance or from original settlement. Rules of governance and membership may be unclear to or unacknowledged by the state and may also differ from family group to family group and from country to country within the Caribbean. This paper describes the extant tenure and registration situation in Trinidad and Tobago. It then examines the particular tenure situations that can be adequately accommodated by the Social Tenure Domain Model (STDM)
∗ Corresponding author. Tel.: +868 662 2002x2520; fax: +868 662 2002x3700. E-mail address: Charisse.Griffi[email protected] 0264-8377/$ – see front matter © 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.landusepol.2010.10.004
version of the Land Administration Domain Model (LADM), which is a standard being developed by a technical committee of the ISO (The International Organization for Standardization), and describes the particular components of the model that are applicable and what requires adaptation to the Trinidad and Tobago situation. It then concludes on whether the standard, when finalised, may be adopted in Trinidad and Tobago. As far as possible the current situation is described without proposing improvements or changes to the tenure structure that may not be implemented or necessary. Background Trinidad and Tobago is a small twin island state in the southern Caribbean of slightly over 5100 km2 in area and roughly 1.3 million in population making it a relatively densely populated country and thus increasing the necessity for a structured approach to land administration and management. More than 50% of the land is owned and managed by the state and this includes not only reserves and lands vested in state agencies but also lands
C. Griffith-Charles / Land Use Policy 28 (2011) 514–522
leased to private individuals and groups for residential, agricultural, commercial, industrial and other uses. While there is a nominally comprehensive cadastral index, and a land registry supported by land registration legislation, it is posited that 30% of parcels are unsurveyed and unregistered and some 47% of parcel occupants remain without adequate documentary security of tenure because of informal occupancy on state and private land or occupation of family land (Stanfield and Singer, 1993). Much of Tobago is said to be under family land ownership. The country has been proposing to perform a systematic titling programme for more than ten years and legal, institutional and technical infrastructure has been instituted to address this even though implementation has met with continuous delay. If systematic registration is performed, it may not be possible or advisable to individualise family land. When St. Lucia underwent systematic titling in the 1980s, with the intention, inter-alia, of individualising family land, the issue of family land proved to be insurmountable and other mechanisms were used to allow registration to continue (Stanfield, 1988). If Registration is not performed, some form of documentation of what exists is required to support security of tenure. There have been instances where land has been conveyed informally by family members or others without the knowledge of the family. Land in Tobago has become attractive as an investment especially to tourists and the escalating increase in value has proven to be an incentive to unscrupulous persons, even family members to fraudulently sell land to which they do not have rights. Residents on and owners of family land continue to demand some resolution to this issue of securing their rights. The Registrar General’s Department maintains both deed and title data in Trinidad and Tobago while the Surveys and Mapping Division maintains data on the spatial extent of parcels. There is no obvious link between the data held in both institutions for parcels held under the deed system while for parcels held under the title system, which number only some 15% of registered parcels, there is a reference number, noted on the survey plan that indicates where the original registration certificate is held in the registry. Since the title system is Torrens based there is no general map held by the land registry. For the deeds system, a survey plan is not legally required for a transaction to be registered but even when a survey plan is attached, the registration of the transaction is referenced sequentially and not by parcel. The land registration system can therefore be categorised as a centralised, primarily negative, state-supported, juridical and separated system based on fixed boundary surveys and sporadic adjudication, in the terminology collated by Bogaerts and Zevenbergen, 2001. Acknowledgement by the state of the differing tenure types and recordation of the rights held would help to provide security of tenure to the landholders in Trinidad and Tobago.
The STDM The STDM is an application of the more generic LADM which is a structured and standardised method of describing the land tenure relationships between land and humans. The STDM is a more flexible way of representing the actual tenure arrangements where the structured and individualised LADM is too general to describe the nuances of many existing informal and non-standard tenure relationships. The justification for modelling is twofold as stated by Lemmen and van Oosterom (2003); to provide a systematic way of describing the land–human tenure relationships and, if the model is actualised in software development, to allow interactions between different land registration systems (Lemmen and van Oosterom, 2003; van Oosterom and Lemmen, 2006). For the
515
Trinidad and Tobago context, attempting to portray the tenure in a model using the defined terminology would compel the examination of what exists to achieve a better understanding of the extant relationships, and, also, the structure would allow the possibility of interactions between the data from areas that would fit into the more generic core cadastral model and the data from family land and other informal areas that would not. Comprehensive and complete data would support land administration and land policy decision making. Researchers are currently applying the LADM and STDM models to particular jurisdictions such as Indonesia, Tanzania and Portugal to test its applicability in specific cases (Ary Sucaya, 2009; Mithofer, 2006; Hespanha et al., 2006). Each country or jurisdiction has to develop the application of the generic model to the particular situation as legislation and practice may restrict or expand the possible instances, attributes, authorising documents and other variables in the model. Testing will allow the model to be improved to accommodate alternative situations and will also assist individual countries. The standard utilises UML terminology and graphics to diagram the land administration relationships between human and land. Specific areas for attention Augustinus et al. (2006) look at specific problems related to Africa that the STDM is required to address and question whether the standard is flexible enough to accommodate the issues of land rights relevant to the poor in that environment. Some of the issues highlighted there may not arise in the particular case of Trinidad and Tobago because of the differing contexts. The specific issues that may be impacted by the use of STDM model in Trinidad and Tobago need to be identified and examined to determine the advisability of using the descriptions in the standard. Coverage Trinidad and Tobago has an established cadastre, illustrated by an index, comprising of an estimated 300,000 parcels which, because of the gaps, errors and lack of currency of the data, defies, in many instances, conclusive determination of legal status. The different types of tenure regimes, however, must be incorporated or linked into a comprehensive whole and this is one of the issues that the STDM must address. In Trinidad and Tobago individualised tenure held under either title or deed registration coexist with informal occupation of many years standing on state and private land and with family land ownership. There is, therefore, a continuum of tenure regimes from the state guaranteed registered title through the uncertain evidence of deed registration to the unrecognised legitimacy of family land and lastly to the uncertainty of informal occupation of varied duration. The different datasets in the information system would therefore need to be seamlessly merged into one spatial dataset especially since each tenure type is not physically grouped into a distinct area. This would effect comprehensive coverage of the country in the recordation or registration of tenure. Land right The nature of the land right on family land must also be able to be captured in the descriptions of the standard in an explicit fashion where these rules may never have been documented before. The rights governing use, restrictions on usage and responsibilities of the occupants in the family land scenario must be able to be described and the differences between different groups and their governance rules stated.
516
C. Griffith-Charles / Land Use Policy 28 (2011) 514–522
Rights holder The ability to describe who are the rights holders and what defines membership in the family land must be able to be included without necessarily providing a finite listing of the members. In some instances, acknowledged family members may have lived abroad for many years and may have lost contact with the other members but yet retain their rights. Rules for determining who the members of the family are may differ slightly from family to family and so flexibility is required for recording the authority by which membership is determined. Complete and comprehensive knowledge of the rights holders on family land is not usually available. St. Lucia addressed this issue in its systematic land title registration programme in the 1980s by titling the parcels in the name of ‘Heirs of X’ where ‘X’ was given as the last known antecedent to have formal title or a name commonly used to identify the family where a formal document did not exist. Another method used was to title with a list of all members available at the time and allowing rights to transact on the land to the first five persons listed on behalf of the group. Technical capacity Augustinus et al. (2006) also look at the advisability of having the purpose of the modelling be the construction of a solely digital database as opposed to retaining the ability to revert to a paper based environment. Two issues may arise from a solely digital database. One issue is whether there is an ability to reconstruct the databases in case of disaster in the centralised or decentralised location of the dataset and the other is having the ability to download to text and graphic documents in a readable and recognisable form by the community such as a certificate of title to land. The recoverability of the database after disaster can be addressed by storing the datasets online at remote locations and periodically revising the information. The example of the tsunami disaster in Aceh as compared with the hurricane in New Orleans and the differences in the subsequent recovery of the cadastral structure is instructive (Griffith-Charles, 2008a). Beyond the technical capacity to maintain the databases, there is a requirement for an institutional capacity to perform this function. The institutions at the level of the community must also be motivated to continue to perform this function even where the perception of security of tenure is heightened by the introduction of the system. Parcel definition The definition of the spatial unit is not available in many instances as parcels may not have been surveyed if occupied by settlement both for the informally occupied areas on state and private land and for the family land areas that cannot be related to a previously purchased parcel. The system must then be able to accept descriptions of extents of rights using indeterminate or fuzzy boundaries to define a unique parcel. In many instances formally registered parcels under the deed system have not been surveyed but the deed contains a verbal description by reference to adjoining owners. A related issue is that a family land holding may also consist of several scattered parcels that should be considered to be one unit for the purposes of recordation. Spatial and attribute precision The institution that maintains data on the spatial extent of the parcels in Trinidad and Tobago is separate from the entity that registers the rights. The parcels in the cadastre in Trinidad and Tobago
are not individually spatially referenced to the national framework. Therefore the requirements for coordinated points on the boundary of the individual parcels is not possible for the general cadastre and less so for the informal or family land areas. The topology of the surveyed parcels in the database is relatively high as surveyed parcels are referenced to surrounding parcels. Therefore, the parcel description in the standard for the formal areas as well as the informal areas would need to include the topology and not the coordinated location. Professional attitudes Accepting and adopting imprecision is necessary in such a standard that seeks to describe and hence acknowledge accruing rights. Both land surveyors and lawyers find difficulty in accepting information systems that are less precise and more flexible. The land surveyor seeks a standard of precise spatial information while the lawyer seeks a standard of precise attribute information on the nature of rights and specifics of the persons. While these standards are possible for the formal areas, the flexibility of imprecision needs to be built in to the STDM if it is to be effective for recording and managing the family land. The imprecision of recording the spatial extent of the parcel can be addressed by inputting calculated precisions depending on the method of survey or data acquisition. The imprecision of recording attribute information such as the number and spelling of the names of the members of the family must also be accommodated. Boundary points captured with non-survey grade GPS equipment or coordinated from remotely sensed imagery for defining areas of occupation must be able to be used. The standard should be simple but comprehensive enough to allay the fears of the professionals that the data contained are those that are salient for the purpose of securing rights to land. Process The process for updating of recordation must be a clear one especially if initial recordation, like the current title registration, continues to be sporadic and voluntary instead of systematic and compulsory. Decentralisation Decentralisation for the purposes of effective maintenance of the system is another crucial area to be addressed for the model. Lack of maintenance of the system soon causes an information system to lose much of its value. The family land has its own particular governance system for use and occupation of the land and it is this system that must be drawn on to maintain information on the status of use and rights. The system must therefore be able to be maintained by the family itself. This would encourage the maintenance of the systems and allow the state to have access to the information if the centralised databases need to be updated periodically. Applicability of the model The LADM and STDM provide a standard set of terminology and structures to model the relationships that already exist in land and those that can be conceived to exist and to be registered in some jurisdiction. Flexibility is also built in to accommodate other anticipated land-related objects and relationships that may be possible to register or record. Whether the standard covers all relationships that actually exist is what is required to be tested by individual countries or jurisdictions. If countries find structures that are not
C. Griffith-Charles / Land Use Policy 28 (2011) 514–522
captured in the model, the model is extensible to allow inclusion of other situations. The main objects in the LADM, which are described as the core classes, are the parcel related information (LA RecordedObject), the parcel (LA SpatialUnit), the person (LA Party) and the rights, responsibilities and restrictions (LA RRR) that link them (Lemmen and van Oosterom, 2003; van Oosterom and Lemmen, 2006). The STDM lists the latter three objects as the STDM SpatialUnit, STDM Party and SocialTenureRelationship to denote the fact that the attributes present in these objects differ from those legally prescribed in the jurisdiction but are those considered legitimate by the society (Lemmen and van Oosterom, 2003; van Oosterom and Lemmen, 2006; ISO, 2006). These STDM objects can be recorded as they exist instead of making them conform to the specifications required for legal registration. Changes to parcels, recorded objects, rights and persons through time are accommodated in the model by providing for descriptions of the time of beginning and ending of existence on an object or a right. Flexibility in the spatial extents of the rights is also accommodated. The LADM domain model is as shown in Fig. 1 which graphically depicts the abstract classes for the core objects of person, recorded object, parcel, and rights that are used in the standard. Fig. 1 also indicates the relationships between the objects in the LADM model. The notation indicates that each individual parcel can be associated with 1 or many rights, restrictions, responsibilities while each instance of party, whether a group or individual, human or abstract organisation or entity, is related to zero or more rights, restrictions, responsibilities as they apply to the one parcel. Coverage In the Trinidad and Tobago scenario the formal areas would contain the rights, restrictions and responsibilities that exist in
517
the English-based legal tenure system of the two registration regimes, title and deed, that exist in the country. Fig. 2 indicates the standard model as it is adapted to the situation in Trinidad and Tobago. The difference between the two types of formal registration would be captured in the attribute of the class LA Layer called LA T&TRegisterType to denote that it applies specifically to Trinidad and Tobago. The code list for this attribute would contain deed and title registration or Old Law and RPO (Real Property Ordinance) registration as they are more commonly known. This attribute can also accommodate the legal registration of land in Tobago or other decentralised registry or the recording of family land at community level. This would allow the different types of registration and recordation to be described by the same standard and therefore incorporated in the same dataset even though they may not be maintained by the same institutions. Currently there is only one centralised registry but documents may be submitted at the Tobago office to be taken to the central registry in Trinidad for authorisation and registration. The separation of the legal and spatial elements of the land registration system in Trinidad and Tobago can also be accommodated by the standard as shown in Fig. 2. The concept of the layer class can be used here to distinguish the different types of registration horizontally as well as the overlapping rights vertically so that they can eventually be spatially represented as distinct but coexistent registered or recorded entities. There is as yet no unique parcel number for parcels so the attribute Oid or object identification would first have to be assigned by the Surveys and Mapping Division based on the location of the parcel in the cadastral index at the largest available scale. Pending that implementation, the certificate of title volume and folio number can be used as the ID for titled land. The registered deed number for the first sale of a subdivided parcel can be used for deed registered parcels. Where family land parcels have a formal history and defined parcel, this deed reference method can also be used.
Fig. 1. Relationships between core objects in the model (after ISO document ISO/CD 19152.0).
518
C. Griffith-Charles / Land Use Policy 28 (2011) 514–522
Fig. 2. The Trinidad and Tobago formal registration situation.
In Fig. 2 the graphic indicates how the information maintained by each agency is conceptually linked to information external to its own organisation. The linkage to the database on the parcel is shown here with the assumption that a direct connection can be achieved. The attributes of LA RRR would allow all shares in rights to sum to 1 and the period of time for which rights are to be given to be described. The LA Party would require a unique identification which, since the equivalent of the social security number is not issued or used for unique identification in Trinidad and Tobago, may be the individual’s birth certificate number, records of which are maintained at the same Registrar General’s Office or a national ID number which is issued on application after the age of 15 and is recorded and maintained at another institution. Again, these datasets are conceptually linked. Many of the attributes possible in the standard for each object are not necessary or possible in the Trinidad and Tobago context. Comprehensive coverage of the cadastral fabric of the country can therefore be obtained using the model.
The social tenure relationships can be defined in a list called a code list which would list all the instances possible in the particular jurisdiction. Most family land inheritance rules conform to cognatic principles but allowances can be made for other types of tenure situations that may arise on investigation. For Trinidad and Tobago this list could include:
Land right
Recording the various permutations of the rights situations would increase the complexities of the database and reduce the flexibility of the practice. It is important, however, to document those factors that would affect the rights and could potentially lead to conflict if omitted, such as rights by descent along the male or female lines. Some minor variations may be left out of the database as negotiation and discussion amongst the family members has and will continue to resolve issues of rights that arise. In the informally occupied, non-family areas, rights to be recorded can include the transitory legal status of adverse possession. The status of occupation by adverse possession need not be directly recorded but derived from an attribute of the class SocialTenureRelationship. The declared beginning time of occupation can
Fig. 3 describes the informal situation and is a view of the classes in the system that relate to the family land and other informal tenure relationships. The nature of the SocialTenureRelationship can be also based on shares to members that would sum to 1 as in the formal case or the rights can be held collectively as a group with no requirement to define the share held, as there is no ability to alienate or subdivide a share in the typical family land situation. There is usually a hierarchy in the family land structure with decision making at the top entrusted to someone; sometimes the eldest or most educated (Griffith-Charles, 2006) but this hierarchy may not be possible to translate into shares.
• familyLandtenure1—where rights are by cognatic descent • familyLandtenure2—where rights are by unilateral descent male • familyLandtenure3—where rights are by unilateral descent female • tenancy—month to month not legally requiring registration • shortLease—leases less than the legal requirement for registration • informalOccupationState—occupation without valid document • informalOccupationPrivate—occupation without valid document • informalSubdivision—subdivided parcel without subdivision development approval
C. Griffith-Charles / Land Use Policy 28 (2011) 514–522
519
Fig. 3. The Trinidad and Tobago informal recording situation.
be recorded and if eventually verified by a court to satisfy the legal requirement for possessory title by adverse possession, the title can be converted to the formal registration if desired. In the family land areas rights, restrictions and responsibilities would need to be documented to reflect the range of possibilities currently afforded members of different family groups. The range of these rights, restrictions and responsibilities may be recorded in code lists and entered as attributes of the group. These rights, responsibilities and restrictions may be public or private. Formal, public rights, restrictions and responsibilities such as land use rights, reservations, building height allowances, etc. are usually kept and amended at the Town and Country Planning Division and are applied to particular parcels upon individual application for development permission. Currently files are kept at a parcel level on development applications and determinations and these can be now kept using these standards to allow interaction of data between departments. Private rights and restrictions are textually stated in transaction deeds and are part of the registration. These can be extracted and registered individually in the standard format. These rights should be recorded for family land as comprehensively as possible even where they may be in conflict with public rights and restrictions. While for formal land the rights may be a subset of use, occupy, develop, transfer, and bequeath, for family land a subset, and various permutations of these subset rights are available. To transfer and bequeath are not possible, and use, occupation and development, while available, are subject to governance principles in the family. Family land rights, restrictions, and responsibility classes are therefore sub classes of the LA Rights, LA Restrictions, and LA Responsibility classes of the LADM and will have additional attributes. Code lists for rights supported by legislation and rights supported by family rules must be drawn up as indicated with some examples in Fig. 4. All types of exist-
ing rights can therefore be described and accommodated in the model. Right holder The STDM Party is an instance of LA Party and so inherits the attributes of the superclass, which are the partyId, the type and the role. The STDM Party class will be used to describe all informal occupants while the further subclass of STDM GroupParty will be used to describe family land groups. An individual member can be recorded but it is not necessary for all members to be individually recorded to still retain rights. A member will be recorded if a record of an overlapping use right for part of a parcel for example is required. The STDM GroupParty is a subclass of STDM Party and will also inherit but however, will have additional attributes such as a family name to identify the group, Smith Family, for example, which will be given an individual ID. Fig. 5 shows the relationship between the party objects in the STDM standard. There is therefore a way of describing all right holders. Parcel definition Fig. 6 shows the relationship between the instance of spatial unit or parcel in the family land or informal area and the superclass LA SpatialUnit in the spatial unit package. The attribute referencePoint allows the inclusion of an approximate centroid coordinate for the parcel, which may be taken off of an index, aerial photograph, satellite image, or acquired by handheld GPS. The extent of the parcel may also be described textually. Family land holdings of several individual parcels held under the same rights should be recorded using the parcel complex description as opposed to the formal situation where a person holding
520
C. Griffith-Charles / Land Use Policy 28 (2011) 514–522
Fig. 4. Possible rights, restrictions, responsibilities on family land.
several different parcels should register each individual parcel as a separate registration instance. The implication is that the same acknowledged family land right covers the entire holding comprising of several parcels. Spatial and attribute precision and professional attitudes
Fig. 5. Party package.
The class LA SourceDocument in the LADM is named the SocialTenureInventory in the STDM. Whereas the LA SourceDocument class would have instances of documents that would be evidence of formal rights to land, the SocialTenureInventory may contain instances of differing authority on rights such as an original deed or title of an antecedent where one exists. Alternative documentation such as affidavits of neighbours attesting to duration of occupation can be listed as possible evidence of rights. Fig. 7 shows the relationship between the documents that define the evidence of rights in the informal or family land situation. The SpatialUnitInventory class allows recordation of alternative survey measurements to a different precision than the formal situation. If the professionals are aware that the original documents of deed and survey plan to which they are accustomed are accommodated in the source documents as scanned documents then this should allay their fears of change. One of the advantages of the family land communal tenure system is its flexibility as derived through negotiation and practice. If extant rules of governance are documented then as changes occur to the policies and practice of the rules, the logical structure of relationships will change. The model has the ability to time stamp an
C. Griffith-Charles / Land Use Policy 28 (2011) 514–522
521
Fig. 6. Spatial unit package.
instance of occurrence so that changes can be made over time without losing the ability to examine the status at a previous date. This is done through listing the time stamp as one of the attributes. Process The process for the initial recordation must also be addressed as it would require a methodical adjudication process as would any systematic title registration process. Publicity and participation would ensure that the information obtained upon which the
decisions about rights would be made is the most comprehensive information possible. In the face of a dearth of formal documentation, oral testimony together with physical evidence of occupation would be the deciding factors as to whose land rights would be recognised. The final answers to be recorded must be those acceptable to the communities themselves guided by national government governance of the process through adjudication personnel with conflict resolution experience. A participatory process would also encourage acceptance of responsibility from the communities for the governance process of the management and updating of the database. It is significant that in a study conducted amongst family land rights holders only 31% felt that the system was beneficial and should be retained (Griffith-Charles, 2008b). The majority of 69% did not support the continuation of the tenure form. However, this perception was arrived at because of the hardships currently felt by this group where their rights are under threat from counter claims to their rights and fraudulent dealings that challenge their security. Whether the tenure form eventually dies over time, under the pressure of modernisation and urbanisation, is not as important as preserving the existing family land rights that the current holders possess. Decentralisation
Fig. 7. Document package.
If the separate institutions follow the standard descriptions and maintain the structure of the data then it should be possible for data to be collected and maintained by the separate institutions and still be compatible. Family leaders can maintain the data following the logical structure in analogue fashion or using software when this is eventually constructed. It may not be necessary for the decentralised data to be maintained current at all times in the central version of the database. The national database can hold data up to the level of the community or the family holding and the community or family can maintain data below this level as advocated by discourses on customary tenure by Ankersen and Barnes (2002) and Fitzpatrick (2005). These internally governed units are described as being a ‘tenurial shell’.
522
C. Griffith-Charles / Land Use Policy 28 (2011) 514–522
Conclusion The general descriptions of the LADM have been developed into a more flexible model in the STDM that has provided the possibility of adaptation in jurisdictions where a structured formal model cannot fit. In Trinidad and Tobago informal and communal areas exist alongside the formal registered cadastre. A preliminary investigation was made here into the parts of the LADM and STDM models that would require to be adapted for the particular situation in Trinidad and Tobago. The preliminary examination appears to indicate that the descriptions in the proposed standard can fit the situation on family land in Trinidad and Tobago if adopted. While the STDM can be further adapted to fit the particular context in Trinidad and Tobago by documenting the range of instances possible for each class, practical testing would need to be done to determine how the model will work in practice. This testing will involve documenting in the field the governance procedures followed by individual family groups and undergoing a systematic recordation of the family land holdings using this model. It is envisaged that a pilot project can advance this work a step further to implementation and also encourage support for its adoption. It must be borne in mind, however, that the recordation process is supposed to protect the rights that already exist and not to disadvantage land holders by negating unrecorded rights and right holders that currently exist. A problem that may arise is the creation of the perception that since recordation lists some stated rights then other rights not listed have been proscribed. The recordation of some rights should not negate the legitimacy of other rights previously enjoyed. The recordation of some individuals or groups should also not negate the legitimacy of the rights enjoyed by other individuals or groups not so recorded. Standards can be documented and applied but may not be adopted by the society unless the society subconsciously accepts, individually and collectively, that the benefits of adoption outweigh remaining with the status quo. There are standards, especially in the area of IT that have been widely adopted and have become de facto standards without being imposed. It remains to be seen therefore whether this standard of terminology and modelling will become the de facto standard for describing and modelling land administration systems of formal and informal tenure. Acknowledgements Acknowledgement is given to the Erasmus Mundus 10 Staff Exchange Programme for supporting a visit to the ITC (International Institute for Geo-information Science and Earth Observation) and this investigation and to Prof. Paul van der Molen, Prof. Jaap Zeven-
bergen, Dr. Harry Uitermark, Christian Lemmen, and Dr. Arbind Tuladhar for helpful discussions. References Ankersen, T., Barnes G., 2002. Inside the polygon: emerging community tenure systems and forest resource extraction. Available online: conservation.law.ufl.edu/ pdf/extraction.PDF. Ary Sucaya, I.K.G., 2009. Application and validation the land administration domain model in real life situation (a case study in Indonesia). M.Sc. Thesis, Urban and Regional Planning and Geo-information Management (PGM). International Institute for Geo-information Science and Earth Observation (ITC), Enschede, The Netherlands, 92 pp. Augustinus, C., Lemmen, C., van Oosterom, P., 2006. Social tenure domain model: requirements from the perspective of pro-poor land management. In: Proceedings of Promoting Land Administration and Good Governance. 5th FIG Regional Conference, Accra, Ghana, FIG. Besson, J., 1979. Symbolic aspects of land in The Caribbean: the tenure and transmission of land rights among Caribbean peasantries. In: Cross, M., Marks, A. (Eds.), Peasants, Plantations and Rural Communities in the Caribbean. Department of Sociology, University of Surrey/Department of Caribbean Studies, Royal Institute of Linguistics and Anthropology, Guildford/Leiden, pp. 86–116. Bogaerts, T., Zevenbergen, J., 2001. Cadastral systems—alternatives. Computers, Environment and Urban Systems 25 (4–5), 12. Clarke, E., 1953. Land Tenure and the family in four communities in Jamaica. Social and Economic Studies 1 (4), 81–118. Fitzpatrick, D., 2005. ‘Best practice’ options for the legal recognition of customary tenure. Development and Change 36 (3), 449–475. Griffith-Charles, C., 2008a. Disaster rehabilitation and reconstruction: the role of cadastral records. In: International Workshop: Disaster Risk Management, The Role of the Land Management Professionals. CASLE, St. Augustine, Trinidad and Tobago, 6 June. Griffith-Charles, C., 2008b. The implications of SDI development to the family land communal tenure structure in Trinidad and Tobago. In: Proceedings of GSDI 10: The Tenth Annual Conference for Geospatial Data Infrastructure. 25–29 February, 6 pp. Griffith-Charles, C., 2006. The Persistence of Family Land in the Caribbean in Proceedings of the XXVI International Congress of Latin American Studies Association (LASA), San Juan, Puerto Rico, 10 pp. Hespanha, J.P., van Oosterom, P., Zevenbergen, J., Dias, G.P., 2006. A modular standard for the cadastral domain: application to the Portuguese cadastre. Computers, Environment and Urban Systems 30 (2006), 23 pp. ISO, 2006. ISO/CD 19152.0. Geographic Information—Land Administration Domain Model (LADM) Committee Draft. Lemmen, C., van Oosterom, P., 2003. Further progress in the development of a core cadastral domain model. In: Proceedings of FIG Working Week. 13–17 April, Paris, France, 13 pp. Mithofer, K., 2006. Development of a GIS-based land registry for Tanzania. Centre for Geoinformatics. M.Sc. Thesis, Salzburg University, Salzburg, Austria, 93 pp. Smith, M., 1956. The transformation of land rights by transmission in carriacou. Social and Economic Studies 4 (1), 64–82. Stanfield, D., Singer, N., 1993. Land tenure and the management of land resources in Trinidad and Tobago. Land Tenure Center, University of Wisconsin, Madison, WI. Stanfield, D., 1988. The LRTP: evaluation of procedures and outcomes. Overview of land tenure patterns and effects of the land registration and titling project. Land Tenure Center, University of Wisconsin, Madison, WI, vol. 1, p. 27. van Oosterom, P., Lemmen, C., 2006. The core cadastral domain model: a tool for the development of distributed and interoperable cadastral systems. In: International Conference on Enhancing Land Registration and Cadastre for Economic Growth in India, New Delhi, India.