Relatively little has changed in the post-independence era. Attempts at re-distributive reform through the establishment of land ceilings have been a feature of both the Pakistan and the Bangladesh periods. But whilst ostensibly designed to place land in the hands of the tiller, and to return water bodies to those who fish them, the reforms have largely been circumvented by the wealthy and powerful.

High population densities and increasing fragmentation of holdings mean, in any case, that the scope for re-distribution declines as time passes. Tenants' rights, including security of tenure, are enshrined in legislation. These are currently almost invariably ignored in practice, and may offer some scope for intervention.

Measures are also in place promising the landless access to government land created by alluvion, and to a range of water bodies. NGOs concerned with the land issue have tended, in recent years, to focus their attention on the different means by which these rights may, in practice, be secured.

If we go through the land policy and administration timeline, we will see that Hindu rulers of ancient India introduced land revenue systems. Sher Shah reforms introduced a regular system of land measurement, together with the assessment and collection of revenue. The British established an elaborate system of land surveys and registration based on the concept of net assets. This was designed to encourage the peasantry to settle in remote and marginal land, thus boosting revenue collection. The system continues, with some modifications, in the very different circumstances that now prevail.

In 1793, the Permanent Settlement Act vested rights to own land to a class of zamindars. Whilst intended to usher in the re-organisation of agriculture along capitalist lines, this had the actual effect of creating multiple-layers of sub-tenants.

In 1882, the Transfer of Property Act, the forerunner of the present registration procedures, was passed. From 1888 to 1940, a Cadastral Survey (CS) of undivided Bengal created the original record of land rights. This is often still accepted as evidence by modern courts.

In 1908, the Registration Act established land registers kept by the sub-registrar, an official under the Ministry of Law. These assess and collect "ad valorem" based registration fees, stamp duty and transfer tax, and provide deeds relating to the transfer of land. Ninety thousand cadastral maps, covering the whole of contemporary Bangladesh, were published in 1927. These are still considered to be the most reliable cartographic record by modern courts.

In 1946, the Tebhaga sharecroppers' movement campaigned for reforms in ratios and procedures governing division of produce. But nobody, now, really represents their interests, or carries the movement forward.

In 1947, Pakistan continued with a version of the net asset system but this declined in importance due to reduced frequency of settlements and poor maintenance of land records. In 1950, the zamindari system was abolished. Control of land was passed to the Revenue Department, which subsequently became the Ministry of Land.

The 1951 East Bengal State Acquisition and Tenancy Act (EBSATA) promoted the goal of retaining the agricultural character of land by giving cultivators first right of purchase and prohibiting other uses; but the large number of exceptions, and poor enforcement, diluted its impact. A land ceiling of 33.3 acres was imposed.

From 1950 to the early 1970s, a leftist movement targeted the landless poor and marginal farmers, but whilst land reform continued to excite the popular imagination, little was done by way of implementation. From 1956 to 62 a State Acquisition Survey was conducted, based on the CS blueprint, and in 1961, land ceiling was raised to 125 acres. In 1965 survey and revisional settlement operations commenced, but progress was very slow and by 1995 it had only been completed in 10% of all thanas.

In 1972, a land ceiling of 33.3 acres was re-established and various presidential orders provided for the distribution of khas land amongst the landless. It was expected that 2.5 million acres of excess land would be released, but in reality there was far less.

In 1976, a variety of land-related charges were consolidated into the Land Development Tax, which covers the whole country except CHT, but deficiencies in the record system mean that individual holdings cannot be checked, and switches to more heavily taxed non-agricultural uses frequently go unrecorded.

In 1984, the Land Reform Ordinance limited future land acquisitions to 21 acres whilst retaining present ceilings. Benami (ceiling avoiding) transfers to relations were outlawed, but again evasion was easy. Legal recognition to the rights of sharecroppers was given for the first time, and sharecropping was established as the only admissible form of tenancy contract.

In the late 1980s, the Muyeed Committee recommended that functions of Land Registration (sub-registrar) and record (Tehsil) be brought together in a single office at field level, but this was ignored.

In 1988, the cluster village program resettled landless people on state land, but only 800, with some 32,000 households, had been formed by 1996. In 1989, Board of Land Administration split into Land Appeals Board and Land Reforms Board to deal with the ever increasing volume of quasi-judicial appeals. In 1991, a land administration manual lay down detailed instructions regarding inspection and supervision of Union and Thana land offices.

In 1997, a new Agricultural Khas Land Management and Settlement Policy was introduced. In 1998, total khas land was found to be 0.75 million acres (or 3% of arable land area). But the actual amount remains unclear as a result of de facto private control arising from informal local settlements.

The present day administration of land splits into four different functions, divided between two ministries. The Directorate of Land Records and Surveys (DLRS) in the Ministry of Land conducts cadastral surveys, from which it produces mouza (revenue village) maps showing individual plots of land and khatian (individual land record certificates).

The Land Reform Board, also in the Ministry of Land, has a number of functions that it discharges through Upazilla Land Offices and Union Tehsil offices. It administers khas (public) land, and manages abandoned and vested property. It updates maps and land records between surveys, and sets and collects the Land Development Tax. It is also formally responsible for the implementation of land reform legislation and the implementation of tenant's rights.

The Land Appeals Board (again in the Ministry of Land), is the highest revenue court in the land, serving as the final arbiter in matters of khas land, changes in records, plot demarcation and taxation which cannot be resolved at lower levels. As such, it represents the final link in a chain running upwards from the Assistant Commissioner (Land) and the Nirbahi Officer at the Upazilla, through the Additional Deputy Collector (Revenue) and the Deputy Revenue Collector at the District.

Finally, the Department of Land Registration in the Ministry of Law, Justice and Parliamentary Affairs records land mutations arising through sale, inheritance or other forms of transfer, and reports changes to the Ministry of Land as well as collects the Immovable Property Transfer Tax.

The land survey process is referred to as land settlement and is administered by the Directorate of Land Records and Surveys. At headquarters there is a diara settlement officer who oversees surveys in riverine areas and major urban centres where frequent changes of ownership take place.

The Amin and two chairmen are responsible for drawing revised mouza map showing changes in area, location and characteristics of land followed by demarcation of boundaries. These are temporary junior staff. Insecurity and low pay affect their morale, performance, accuracy and reliability. Besides, they have to depend on the local elite for board and lodging during season and are thus open to their influence.

Display of notices and beating of drums summons owners, neighbours and interested parties to khanapuri at which each claimant presents his case and the Amin fills up the columns of the khatian form giving plot number, khatian number, classification of land that affects land revenue, area, crops grown, name of owner, agricultural practices, and this khatian also officially contains information on tenancy since 1984 Land Reform. Again poorly paid field workers are susceptible to bribery here. Besides, in practise, tenancy is rarely recorded because of pressure from the rich.

Tehsildar, assisted by a clerk, hears each owner, listens to any disputes and, if satisfied, attests the khatian by signing it in red. Otherwise a re-survey may be ordered. Sixty khatians may be attested in a day, but there are particular backlogs at this stage. It may take two years to clear the work of one field season.

Where objections arise, cases are heard by ASO with decisions recorded in violet. Under these circumstances, these mid level staffs have few chances for promotion, and extra field allowances that used to be provided have been stopped. Naturally this encourages corruption.

ZSO and ASO hear appeals at Upazilla level and some are referred to District level where decisions marked in black. Again long delays are caused by shortage of suitably qualified staff to hear appeals.

Map correction, amalgamation and splitting up of jamas (interests) are done by the permanent surveyors and their supporting staffs. As documents are about to be dispatched for printing, powerful local people often intervene to lobby for changes.

Formerly, both khatians and maps were printed centrally at DLRS presses. Zonal offices now produce khatians, which has speeded up the process, but maps continue to be printed centrally and compositors names are now printed on khatians which has significantly reduced tampering at this stage. Methods are antiquated and equipment is obsolete. Besides, newly promoted, inexperienced officials are usually given responsibility for complex tasks. Khatians and maps arriving heavily exceed the capacity to process, causing increasing backlog (estimates of size of which vary widely).

Once completed, copies of the ROR are passed to DC, thana, and union land offices for land management, with originals retained in the district office, under lock and key. The records are then updated as a consequence of sale, and transfer is done through mutation process. Tehsil registers are not freely open to inspection, but for payment of a small fee, landowners are formally entitled to a certified copy of the ROR and mouza map. In this situation, local officials are unable to keep records updated. If they could, there would be no need for revisional settlement. In practice a substantial bribe must be paid to access registers, effectively restricting access to the better off.

In the case of the land transfer process through sale the reality is that some transfers occur on an entirely unofficial basis, perhaps when land is mortgaged, but this is becoming less common. Some buyers may not check the AC records first and even if they do, these may well not be up to date. The deed writers and Sub-Registrar collude to ensure that this step only proceeds if a bribe is paid first, whilst the buyer and seller may also collude to reduce the amount of Immovable Property Transfer Tax (IPTT), which is levied at 10% of the sale value. There is no requirement to check the legality of the transaction and it is not uncommon for the same plot to be "sold" to several different buyers, although this is much more frequent in urban areas. This is supposed to be issued within a month, but frequently takes a year and the payment of a bribe. The AC (Land) generally does not update the record unless he is paid a bribe to do so.

The diversity of ways in which land records may be updated and the problems associated with each, give rise to numerous disputes in which the rich and powerful inevitably enjoy the upper hand. Where a decision relating to the recording of land title is disputed, the appeals process starts from Tehsildar and then movers progressively upwards until the appellants, and other interested parties, either accept the judgement given, or lack the resources to proceed further.

Many issues are dealt with in informal local shalish and so never reach the Land Appeals Board. Where suits do enter the formal system, the cost is considerable and suits can take 15-20 years to resolve, with different parties often each in possession of documentation from different official bodies. Only the rich and well connected are able to climb all the way to the top.

Where a settlement survey is under operation, appeals passing beyond the Tehsildar are supposed to be heard by the ASO and if not resolved there then pass on directly to the District level. All civil court proceedings relating to land should formally be suspended when a settlement is in process, but this would lead to even further delays and is thus generally ignored, adding further confusion by having two channels in operation at the same time.