The Law of the Land

Authored by: Colin Imber

The Ottoman World

Print publication date:  December  2011
Online publication date:  December  2011

Print ISBN: 9780415444927
eBook ISBN: 9780203142851
Adobe ISBN: 9781136498954




The economy of the Ottoman empire rested on peasant labour, and, in an area extending from the western Balkan peninsula to eastern Anatolia and into Syria and northern Iraq, the typical peasant tenement – or çift – was a farm worked usually, but not inevitably, by a single family. Most tenements belonged to a timar (fief), and it was the fief-holders who controlled the peasants’ access to the land. Between the fourteenth and the seventeenth centuries, most fief-holders were sipahis (cavalrymen), who held their fiefs in return for army service, but fiefs also supported vezirs, governors and other servants of the sultan during their time in office. From the seventeenth century, fiefs increasingly went to tax farmers and, in the eighteenth, following the introduction of lifetime tax farms in 1695, came to resemble private estates. However, neither peasant nor fief-holder owned the land: only what was above the land – trees, vines and buildings – could be held as private property. The issue was not who owned the land, 1 but who owned the revenues from it. Private proprietors owned the taxes from their estates as property which they bequeathed to their heirs; fief-holders owned the taxes from the fief for the period of their tenure.

 Add to shortlist  Cite

The Law of the Land

The economy of the Ottoman empire rested on peasant labour, and, in an area extending from the western Balkan peninsula to eastern Anatolia and into Syria and northern Iraq, the typical peasant tenement – or çift – was a farm worked usually, but not inevitably, by a single family. Most tenements belonged to a timar (fief), and it was the fief-holders who controlled the peasants’ access to the land. Between the fourteenth and the seventeenth centuries, most fief-holders were sipahis (cavalrymen), who held their fiefs in return for army service, but fiefs also supported vezirs, governors and other servants of the sultan during their time in office. From the seventeenth century, fiefs increasingly went to tax farmers and, in the eighteenth, following the introduction of lifetime tax farms in 1695, came to resemble private estates. However, neither peasant nor fief-holder owned the land: only what was above the land – trees, vines and buildings – could be held as private property. The issue was not who owned the land, 1 but who owned the revenues from it. Private proprietors owned the taxes from their estates as property which they bequeathed to their heirs; fief-holders owned the taxes from the fief for the period of their tenure.

This system of land tenure and taxation was an inheritance from pre-Ottoman regimes, with many local variations within the system. 2 Nonetheless, from the late fifteenth century, Ottoman administrators attempted to codify and, as far as possible, standardize the laws governing relations between peasant and fief-holder. The reign of Bayezid II (1481–1512), when there was a need to consolidate the conquests of his father Mehmed II (1451–81), saw the first major effort in this direction: the 1487 kanunname (law-book) for the district of Hüdavendigar around Bursa became the model for later codes. Shortly afterwards, a compiler amalgamated several rather disparate documents from the early to mid-fifteenth century to produce a kanunname misleadingly attributed to Mehmed II. 3 More importantly, a systematic kanunname , applicable throughout the empire, appeared in about 1499. 4 A new recension of this code appeared thirty or more years later, following the conquests of Selim I (1512–20) and Süleyman I (1520–66), 5 while the next major revision of the law came after the suppression of the celali rebellions in 1608, as part of an effort to restore order to the land.

The main concern of early kanunnames was taxation, the clauses on land tenure tending to deal with exceptional and even random cases. The 1670s, however, saw the promulgation of the ‘New kanunname ’, the most enduring achievement of the reforms of the Köprülü vezirs. This was a comprehensive land code, made up of extracts from earlier texts compiled systematically under headings. It remained in use until the appearance of the land code of 1858 during the tanzimat (reform) era. What follows is an account of the law as it developed to this point.

The Çift

The term çift means literally a ‘yoke’ and, in principle, comprised the area of land that a family with a yoke of oxen could cultivate within a year. In the late fifteenth century, the systematization of the laws of land and taxation created the need for a more precise definition, and this the compilers of the 1487 kanunname for Hüdavendigar supplied:

a çift is taken as consisting of 70 or 80 dönüms of good land, 100 dönüms of land of middling quality and 130 to 150 dönüms of poor quality land. A dönüm is an area of 40 normal paces lengthwise and breadthwise. 6

This definition, which took into account the size of the holding and the quality of the land, set the standard for the empire, reappearing in Bayezid II’s kanunname of c.1499 7 and in a number of local codes. Some later kanunnnames slightly modify this scheme, but it was the measurements fixed in 1487 that the compilers of the New kanunname were to adopt in the 1670s. 8

In some regions, however, old definitions persisted. In parts of central Anatolia in the mid-sixteenth century, the popular definition of a çift was not according to the precise measurements of the land, but rather according to the yield of the soil, while the system in use in Syria continued to measure land in terms of the work that a yoke of oxen could perform. 9 In these cases, the land surveyors adapted the term çift to refer to the type of tenement found in the locality rather than attempting to impose the standard. In south-east Anatolia and northern Iraq, some fief-holders continued to equate a çift of land with a yoke of oxen, but did so simply in order to increase the amount of tax that they could collect, by demanding that families with two or more yokes pay at a higher rate than those with only one. This was a practice which the land surveyors uncovered – and prohibited – in Diyarbekir in 1540, 10 extending the same prohibition later in the century to the adjoining province of Mosul. 11

However defined, the term çift always denoted an undivided plot of land and was the unit which surveyors used to determine a peasant’s tax liability, fixing the rate according to whether he held a çift , half a çift or less. In mountainous regions where parcels of fertile land were too small and scattered to amalgamate into çifts, instead of calculating the tax per çift , the land surveyors calculated it per dönüm on each plot, at the standard rate of one akçe per two dönüms for good quality land and one akçe per three for land of poor quality. 12 In districts where, by contrast, there was a surplus of fertile land, the law permitted cultivators to farm additional plots, provided they continued to work the çift for which they were registered. When they did so, following a principle enunciated already in the 1487 kanunname for Hüdavendigar, they paid tax per dönüm on the extra land, in addition to the tax on their çift . 13 In defining the dimensions of a çift , the compilers of the 1487 kanunname also specified how much land the occupant could set aside as grazing for his animals and as space for threshing and storage, copying into the kanunname the words of the decree which they had received in answer to their query on this subject: ‘it is not forbidden for a peasant to leave a few dönüms of land fallow as pasture for the needs of a yoke of oxen and for a threshing floor’. 14 The compilers of the c.1499 kanunname copied this clause verbatim, and it reappears in a number of provincial codes. 15

Title To The Land


A new entrant to a çift acquired title – tapu – to the land by paying an entry fee to the fief-holder, the link between title and payment becoming so strong that the term tapu usually came to denote the fee rather than the title.

Not all property was subject to tapu . Buildings, vines and cultivated trees were private property. Fief-holders therefore awarded title to open fields, but not to buildings, vineyards or orchards. Remove the buildings, vines or trees, however, or run a plough through an orchard, and the soil reverted to its original status as tapu land. Nonetheless, some categories of open land were exempt from allocation by tapu . The 1487 kanunname forbids the appropriation of land used for common grazing: ‘since it is harmful to the public to sow, fence in or give by tapu meadows which serve as grazing land for the townsmen and villagers’ animals, these things have been forbidden’. 16 The same clause appears in the general . kanunname of c.1499 and in later provincial codes, the 1583 kanunname for Yeni Il applying the prohibition also to the land ‘in front of a person’s house’. 17 The question of how far the ‘front of a house’ could extend did not, however, receive an answer until the following century, when the jurisconsult (mufti) Mehmed Baha’i (d. 1654) and, following him, the New kanunname , ruled that it should occupy no more than half a dönüm . 18

The exemption from tapu also applied to the hassa. This was a çift reserved for the fief-holder himself, who, in principle, could not allocate it to anybody else. The basic rule appeared in the 1528 kanunname for Bolu: ‘ çifts recorded in the register as hassa may not be given by tapu ’. The rule, however, was not absolute. An additional clause gives the fief-holder the right to award title to the çift , but for no longer than the period of his tenure of the fief. 19 A similar rule in the 1525 kanunname for Sofia allocates natural water meadows and reed beds as hassa to the fief-holder, giving him the right to grant them by tapu during the period of his own tenure only. 20

The Level of tapu tax

The rate of tapu tax was the amount that ‘unprejudiced persons determined’, this formula applying universally until 1632, when a decree established that the fee payable for a çift which passed to a daughter, brother or sister by inheritance was the value of one year’s produce of the land. 21 The compilers of the 1716 kanunname for the Morea reduced this rate by fixing the fee at the value of the annual produce of the land less the value of the one-seventh tithe due to the fief-holder. Anyone other than a daughter, brother or sister of the deceased continued to pay what ‘unprejudiced persons’ estimated. 22 On the insecure borderlands of the empire, however, it was difficult to reach a consensus on the level of the fee, and it was this uncertainty that persuaded the surveyors in Bosnia, on the military frontier with the Habsburgs, to fix the rate of tapu tax in the 1530s, not by seeking expert opinion, but instead by holding an auction among bidders for the land. 23

In contrast to what was due for agricultural land, the tapu tax payable for houses was fixed. While houses themselves were private property, the land beneath them was not, and anyone building a house paid tapu tax for the site at a rate already in force in the first half of the fifteenth century and specified in the kanunname ‘of Mehmed II’: ‘ground tapu , for the site of a house: maximum, 50 akçes; below this, 40 or 30 akçes; if [the person] is poor, 20 or 10 akçes’. 24 The general kanunname of 1530–40 repeats the tariff, but adds that it does not apply if a person builds on land which he already holds by tapu ‘because it is as if he were building on private property’. 25 A later decree laid out the regulations in more detail. Before building a house, a person had to seek permission from the fief-holder. If he did not, the fief-holder could either demolish the house or grant tapu for the site at the standard rate. However, no tapu tax was due if the house was on the site of an old building which had fallen into disrepair. On some new houses, however, the decree imposed a new charge. If a person built on agricultural land, he became liable to a rent equivalent to the tax that he would have paid on the field if it had remained under cultivation, allowing the fief-holder to take an annual income from houses which – in most cases – would be on land which the builder already held by tapu and so exempt from tapu tax for the site. It was this law that was to appear in the New kanunname in the 1670s. 26

Tenure On The Land

The Consequence of Neglect

No one enjoyed absolute security of tenure on the land. A person lost his çift if he failed to pay his taxes or if he left it uncultivated, the basic rule appearing already in the 1487 kanunname for Hüdavendigar:

if lands which are suitable for farming remain fallow when there is no hindrance [to cultivation], and if this would be harmful to the fief, then in order to alleviate the harm, it is permissible by custom to take them from the occupant and give them to someone else by tapu . 27

The general kanunname of c.1499 copies this clause, but replaces the vague stipulation that the occupant loses the land ‘if this would be harmful to the fief’ with a precise regulation. It is ‘land that remains uncultivated for three years’ that is subject to confiscation. This rule remained in force until the nineteenth century, but with some qualifications over time, the Hüdavendigar kanunname already making an exception for ‘mountainous and uncultivated lands or land subject to flooding, which cannot be farmed every year’. 28 This exemption became generally applicable, but, as the 1528 kanunname for Kütahya stipulates, a farmer might still lose the land if he failed more than once to sow crops in years when it was cultivable. 29

The three-year rule also applied to wasteland which a person intended to clear. If this lay outside the boundaries of a fief, he had an unrestricted right of occupancy until the moment of a new tax census, when the surveyor would register the land and allocate it to a fief. If, at the end of three years, the occupant still had not cleared the land, the fief-holder could reallocate it by tapu . If, on the other hand, the land due for clearance lay within the boundaries of a fief, the three-year period began from the moment of occupancy. If the occupant had cleared it within three years, he received title, but, if not, the fief-holder could reallocate it to someone else. If the occupant had cleared the land without the fief-holder’s permission, the fief-holder could remove him at any time, although the law enjoined – but did not oblige – him to grant title to the occupant as a reward for his efforts. The rules for occupancy of wasteland became more stringent in the late sixteenth century, with an addition to the standard three-year rule: ‘in mountainous lands … the “no obstacle” restriction does not apply’. In other words, the occupant of wasteland would lose title if he did not cultivate it within three years, whether or not he had a valid excuse for not doing so. 30

The peasant who faced eviction for neglecting his çift nonetheless had a residual claim to the land. In a refinement to the three-year rule, the 1528 kanunname for Aydın states: ‘if the holder of the land once again pays the tapu [tax] which an outsider would pay, the timar -holder should once again allocate [the çift ] to him and not to anybody else’. 31 This became a general although not a universal rule. A mid-sixteenth-century statute from Karaman requires the fief-holder to give due warning before expelling a person who has neglected his tenement for three years, but without, apparently, requiring him to pay tapu tax if he wishes to be reinstated. He simply has to resume cultivation. The statute further restricts the fief-holder’s powers by requiring any reallocation of the land to be ‘with the cognizance of the kadi’. 32 The 1569 kanunname for Silistre similarly enjoins the fief-holder to ‘warn’ the negligent farmer after three years, but does not explicitly require him to pay tapu tax if he wishes to keep his çift . 33

In the seventeenth century, the compilers of the New kanunname adopted the rule as it appeared in the kanunname for Aydın in 1528, their formulation implying that after three years of neglect a çift becomes automatically vacant, but that the previous – neglectful – occupant has the right of first refusal when the fief-holder wishes to reassign the land. 34 In answering the question as to whether a person loses his çift when he leaves it in the safekeeping of another, who then fails to till the soil for three years, they cite a law which appears in an early seventeenth-century collection of statutes attributed to the nişancı Celalzade (d. 1567): the fief-holder should reallocate the land, regardless of whether or not the original occupant had left it in the safekeeping of another person. 35 A related question concerned lands left uncultivated by a missing person. In 1595 a petition to the throne elicited the decree that, if there was no news of a person after three years, then the fief-holder should reallocate the land. This remained the law until 1613, when a new decree reduced the period to one year. Soon afterwards, a counterpetition complained that the one-year period was too short, and a new decree reinstated the three-year rule, which remained in place for inclusion in the New kanunname . 36

Double Taxation

Çifts often lay uncultivated when their occupants left them to work the land in other places, raising the problem of whether they should pay their taxes to the fief-holders with whom they were registered or to the fief-holders on whose land they now worked. This issue appears already in the kanunname ‘of Mehmed II’, which rules:

if a peasant possesses land, but leaves the land … uncultivated and instead sows on another fief-holder’s land, he should pay tithe twice: once for the land which he sows and once for the land which he has left uncultivated. But if the fief-holder has no land to give the peasant and he sows somewhere else, he should give the tithe, according to custom, only to the [fief-holder on whose] land he is settled. 37

The 1487 kanunname for Hüdavendgar reiterated this rule, 38 which then became general, and later kanunnames stipulated that the first fief-holder cannot take an extra tithe if the farmer cultivates a new plot of land only as extra labour after he has completed work on his original çift . This is the law that appears in the New kanunname . 39

In addition to a second tithe, anyone who abandoned his çift in order to farm in another place had to continue paying çift tax on his original tenement. However, the question remained of who paid the çift tax when a peasant left his village, leaving someone else to cultivate his land. Different laws, it seems, applied in different regions. In Diyarbekir and Mosul, it was the person who actively worked the land who paid both the tithe and the çift tax. 40 However, according to the 1569 kanunname for Silistre, the person in whose name the çift was registered remained liable for çift tax, even when another person farmed it and paid the tithes. 41

Abandoned Land

The law required peasants who abandoned their land in order to take up a trade to continue paying taxes to their fief-holders, a statute governing the payment due from a person who takes up the carrying trade appearing already in the kanunname ‘of Mehmed II’: ‘if a person who hires out pack-animals, or a carter who is a carter by trade, abandons his çift …, he should carry salt and other items for his fief-holder and pay him 50 akçes per year’. 42 The 50 akçes is evidently compensation for loss of the tithes and the corvée a substitute for the çift tax.

The corvée is an archaic feature, absent from later statutes, but the notion that 50 akçes was the rate of compensation due for loss of tithes lingered. In parts of early sixteenth-century Rumeli (the Ottoman Balkans), the çift -breaker’s tax payable for abandoning the land was 75 akçes, the sum comprising 50 akçes compensation for the loss of tithes plus 25 akçes for the loss of ispence (a tax of 25 akçes payable by Christian households). The rate was not, however, uniform. The 1569 kanunname for Silistre puts the çift -breaker’s tax at 72 akçes for a Muslim and 87 akçes for a Christian, fixing the rate by compounding, for Muslims, 50 akçes in lieu of tithes and 22 akçes in lieu of çift tax and, for Christians, 62 akçes in lieu of tithes and 25 akçes in lieu of ispence. 43 In neighbouring Sofia, by contrast, the rate in 1525 was only 6 akçes. 44 In Srem to the west of Belgrade it was higher, the differential rates of 120 and 80 akçes applying respectively to rich and poor rather than to Muslim and Christian. 45 In the mid-century the rate increased. A statute attributed to Celalzade raises the rate to 300 akçes for a whole çift , 150 akçes for half a çift , and 75 akçes for less than half, justifying the increase with a statement that the flight of peasants from the land to engage in trade or commerce had harmed the income of fiefs. 46

A decree of 1608, compelling fugitive peasants to return to their villages following the defeat of the celalis, confirmed the rate of the çift -breaker’s tax at 300, 150 and 75 akçes ‘according to the ancient law’, stipulating that, although no arrears were due, this was to be an annual tax. 47 This was the law that was to pass into the New kanunname , whose compilers also dispelled the misapprehension that çift -breaker’s tax was payable only by Christians: ‘concerning the law of the çift -breaker: Muslim peasants are also included’. 48

Poverty, Old Age and Infirmity

The double tithe and the çift -breaker’s tax applied to the able-bodied, but not when it was poverty, old age, illness or a natural disaster that had forced a peasant from the land. The 1487 kanunname for Hüdavendigar already states that it is ‘a blatant injustice’ to take çift tax from a person who has lost his land in this way. A statute in the general kanunname of c.1499 refined the law by ruling that, in these circumstances, the fief-holder should confiscate the abandoned çift and re-register the incapacitated peasant as landless, and so liable for a lower rate of tax. 49 The rule was probably universally applicable. In the province of Diyarbekir, however, peasants who had abandoned their çifts through poverty or natural disaster were liable both to the tax due from the landless and to ırgadiye, originally a corvée but often commuted for cash. This rule appears in a kanunname of 1540 50 and was copied into the later kanunname for Mosul. 51 More important, however, than the imposition of ırgadiye was the condition – absent from the kanunname of c.1499 – that the fief-holder could not remove the incapacitated peasant from his çift before a period of three years had elapsed. In other words, the normal rules for confiscating a çift after it had lain idle for three years applied also in this case.

Forced Resettlement

The aim of the çift-breaker’s tax was to ensure that the fief-holder did not suffer a loss of income when a peasant abandoned his çift . A set of regulations requiring the fief-holder to bring back and resettle fugitive peasants served the same purpose. The 1487 kanunname for Hüdavendigar already states that it is ‘ancient law to bring back and resettle local peasants who have dispersed’, adding that it was forbidden to remove peasants who had left their land for more than fifteen years. These, it is implied, and later kanunnames state more explicitly, should pay taxes to the holder of the fief where they now lived. In answering the question of what should happen to peasants who had settled in towns, the compilers of the kanunname quoted a decree issued in response to their query, which ruled that, after they had been resident for more than fifteen years, they were to be registered as townsmen. 52 The same rule reappears in the general kanunname of c.1499, which reduces the residence requirement for registration as a townsman to ten years, and adds that change of status was possible only ‘if [the settler] is not recorded in the register of peasants’. 53

These rules remained in force throughout the sixteenth century. The 1539 kanunname for Vize, for example, states that, if a peasant had been absent for more than ten years and his çift remained uncultivated, he should pay çift -breaker’s tax, but that if he had been absent for less than ten years, his fief-holder should bring him back ‘with the cognizance of the kadi’. 54 However, details of the regulations vary. One recension of the kanunname of 1530–40 gives a ten-year residence requirement for settlers in villages and a twenty-year period for settlers in towns before they could be registered as townsmen. This was unless they had settled in Istanbul: the code forbids the removal of migrants to the capital. 55 Another recension of this kanunname , by contrast, requires a delay of ‘ten or fifteen years’ before a peasant is safe from removal from a village and of fifteen years if he has settled in a town. A marginal note, however, emends this rule: ‘the son of a registered peasant is a peasant. He does not escape this status by being registered in a town’. 56 It is nevertheless questionable whether these rules were practical. They do not appear in seventeenth-century formulations of the law, suggesting that after 1600 they went into abeyance.

The Transfer Of Land


When a peasant died without heirs and there was more than one applicant for his çift , the question could arise as to whether an applicant from within his village had priority over an outsider. The general kanunname of 1530–40 gives precedence to the insider, 57 but this rule was unknown to the compilers of the 1583 kanunname for Yeni Il, who clearly believed the opposite: ‘on the question of lands which are subject to tapu , it is forbidden to discriminate between an insider and an outsider’. 58 In the seventeenth century the chief mufti Yahya (d. 1644), on the strength of a decree, settled the matter by permitting villagers to take back land which the fief-holder had allocated by tapu to an outsider, 59 and this was the rule that was to appear in the New kanunname :

if a person dies … leaving no heirs … his lands should be given to members of the village who are in need of lands and meadows and are applying for tapu . It is against the law to give them to an outsider.

Only if there were no applicants from the village could the land go to a stranger.


The sale of çifts raised problems. Unrestricted sales could upset the political order by allowing land to accumulate in the wrong hands, and sale itself presented a legal conundrum. Peasants did not own their çifts, and so, in principle, could not sell them. In practice they did, and the law had to accommodate the fact.

The sale of land became an issue during the reign of Bayezid II. A decree of 1501 rules that ‘sale and purchase of land are forbidden by custom (‘urf )’, 60 and a kanunname for Aydın from the same period similarly outlaws the practice. The prohibition, however, is only apparent, as the clause in the kanunname continues by laying down the conditions which the parties must observe if the sale is to be valid: ‘[The transaction] is permissible if, with the cognizance of the fief-holder, [the buyer] pays a sum for the right of residence. The fief-holder takes one-tenth of the price which [the vendor] receives’. 61 The same formula appears in several local codes and in some recensions of the general kanunname .

It solved two problems. By requiring the ‘cognizance’ of the fief-holder, it ensured that transfers of land did not slip out of control, and it solved the problem of how peasants could buy and sell land which they did not own by defining the object of the sale as the ‘right of residence’ on the land, rather than the land itself. The kanunnames also imply that title to the land passed to the buyer on payment of the price for the ‘right of residence’, and that the tithe which the fief-holder levied on the price was compensation for the loss of the tapu tax which he would have received had he granted title to the vacant çift .

Evidently, few people observed these regulations, and the issue arose again in about 1568, when, in reply to a questioner who stated that kadis in Rumeli were granting certificates confirming sales of land, the chief mufti Ebussuud declared that all certificates which described these transactions as ‘sale’ were void. He objected not to the transactions themselves but to the terminology: ‘it is absolutely contrary to the shari’a for kadis to define the giving and receiving [of land] by the peasants as “sale” and to issue certificates [accordingly]’. 62 This prohibition created the problem of how to define these transactions if they were not ‘sales’. Ebussuud’s solution was to invent a new terminology. In place of ‘sale’, he instructed kadis to use the term tefviz, meaning ‘consignment’. The usefulness of this word was that, while it carried the idea of ‘transfer’, it was not a technical term in the Islamic law of sale and was therefore available to describe transfers of çifts. Its adoption allowed Ebussuud to devise a new formula which he instructed kadis to use when registering these transactions: ‘X, with the permission of the fief-holder, received so-and-so many akçes from Y and transferred (tefviz) the use of his land to him. The fief-holder also gave the tapu to him’. 63 This formula avoided the term ‘sale’ and made it clear that what was being transferred was the exploitation of the land and not the soil itself.

The use of the term tefviz to denote a transfer of land continued in legal vocabulary into the seventeenth century and, through its adoption by the compilers of the New kanunname , into the nineteenth. It was not Ebussuud’s only innovation. He also abolished the tithe on the purchase price, instead making it a condition that the fief-holder grant the purchaser the tapu to the land:

if someone wishes to vacate the land which he occupies and, with the cognizance of the fief-holder, gives a sum of akçes for the right of residence and if, when he vacates it, the fief-holder gives that person [the land] by tapu , this is canonical and accepted. 64

In making transfers invalid unless the fief-holder specifically granted title, Ebussuud hoped to maintain control over the occupancy of land. He probably also intended the purchaser to pay tapu tax, with this payment replacing the tithe levied on transfers. This did not happen. Instead, it became the practice to pay the fief-holder for permission to transfer the land, raising the question of whether it was the ‘purchaser’ or the ‘vendor’ who paid – a problem which Yahya settled by confirming that it was the purchaser. 65 What he paid, however, was no longer understood as a tithe, the early seventeenth-century mufti Pir Mehmed (d.1611) stating outright: ‘in [the matter of giving] permission [to transfer land] and [granting] tapu , it is unheard of to take a tithe’. What an entrant to vacant land paid was tapu tax, while what a purchaser paid was a fee for permission to transfer the occupancy to himself, at a rate ‘that is deemed reasonable … to satisfy the fief-holder’. 66

The compilers of the New kanunname adopted these rules, adding insistently that tefviz was invalid without the explicit permission of the fief-holder.

Inheritance by Sons

Çifts passed directly and unconditionally from father to son. The same rule applied, even when the son was a minor. A clause in the 1487 kanunname for Hüdavendgar outlined the procedure to be followed:

tapu [tax] for an orphan is a rejected and prohibited innovation. His father’s land is treated as if it were his inherited property. If the land descending from his father remains uncultivated, it should be given to another person, and if, on reaching maturity, the orphan demands it back, it is once again assigned to the orphan. 67

Despite the law, the practice of reassigning orphans’ land by tapu continued. A clause in the Aydın kanunname repeats the prohibition, requiring tapu -holders to vacate the land when the orphan reclaims it, and to seek reimbursement for the tapu tax from the fief-holder. 68 The same clause reappears in the general kanunname of 1530–40, which also places a limit of ten years from puberty during which the orphan must claim his inheritance, 69 a rule that was to remain in place for inclusion in the New kanunname . 70

If the deceased left several sons, these could inherit his çift as a jointly held tenement. The 1487 kanunname ruled that, even if one of the sons appeared in the register as the holder of the çift and the others as landless, they in fact held the çift jointly, and should divide the tax equitably among themselves. 71 Later kanunnames repeat this ruling, but do not address the problem of whether the brothers held the title to the land collectively, or whether each one held title to his own portion. This became an issue when one of the brothers died. If they held the title collectively, the surviving brothers would automatically acquire the deceased’s share of the land, and before the early sixteenth century this was the usual pattern. However, in about 1516, the grand vezir Hersekzade Ahmed Paşa ruled that, when brothers held a çift jointly, each one should hold title separately for his own portion. Consequently, if one of them died, the survivors could take over his portion only on payment of the tapu tax. Nonetheless, they retained a prior claim to the land. The fief-holder could give the land to an outsider only if the brothers could not pay or renounced it voluntarily. 72

Hersekzade’s ruling did not gain immediate currency. A recension of the general kanunname , copied in 1565, repeats the law allowing brothers to hold title collectively, but a marginal note attributed to Celalzade comments: ‘this is wrong. The fief-holder seeks tapu [tax] from his brothers’ – bringing the law into line with Hersekzade’s amendment. 73 Nonetheless, uncertainty continued. In the first decade of the seventeenth century, Pir Mehmed ruled that, when two brothers held property jointly, a fief-holder could not demand that the survivor pay tapu tax for his deceased brother’s share. Elsewhere, on the strength of a decree of 1608, he states the opposite, suggesting that the issue had again become contentious, until finally a decree settled the issue in favour of Hersekzade’s original law. 74

If one of the brothers occupying a çift died leaving a son, the son inherited his share of the land without paying tapu tax, a rule which appears already in the kanunname of Aydın from the reign of Bayezid II. 75

Inheritance by Brothers and Grandsons

If the holder of a çift died without leaving a son, certain of the deceased’s relatives, provided they paid the tapu tax, could inherit the land. Before 1568, the fief-holder should first of all offer the land to his brother – the term ‘brother’, Hersekzade explained, applying only to full brothers or to half-brothers sharing a father. 76 This remained the rule into the seventeenth century, the compilers of the New kanunname stating that, ‘in matters of tapu ’, the term ‘brother’ applied without distinction to full brothers or to brothers sharing the same father. Half-brothers sharing a mother were legally not brothers. 77

The rules governing inheritance by grandsons changed during the sixteenth century. Hersekzade stated that, provided his father was no longer alive, a grandson could inherit a çift from his grandfather without payment, a rule that appeared in the general kanunname of 1530–40. However, Celalzade’s marginal note in the kanunname states bluntly: ‘land does not descend from grandfather to grandson’, adding that the rule applying to brothers applies also to grandsons: if the grandson pays tapu tax, the fief-holder should not give the land to anybody else. 78 It was Celalzade’s ruling that was to survive into the seventeenth century, when it was incorporated into the New kanunname with the proviso that a grandson must claim the land within a time limit of ten years from his grandfather’s death. 79

Inheritance by Daughters

Before 1568, women could not legally occupy land in their own right. Nonetheless, the 1487 kanunname for Hüdavendigar and later codes stated that a fief-holder could not expel a woman provided she kept her land under cultivation and paid her taxes. 80 The assumption was that any woman in this position would be a widow, but in emending the relevant clause, which, in one recension of the kanunname of c.1499, opens with the phrase ‘if a woman on her husband’s land does not leave [the fields] uncultivated …’, 81 to read ‘if a woman, by some means or other, occupies land …’, the redactors of the kanunname of 1530–40 accepted that widowhood was not the only way for women to acquire land. 82

Nonetheless, the administration tried to curtail women’s access to the land. The kanunname of 1530–40 copies a clause from a group of western Anatolian kanunnames of 1528 forbidding women from acquiring çifts by paying tapu tax. This prohibition, as the 1541 kanunnname for Çemişgezek makes clear, 83 applied to both widows and daughters. A further restriction followed in 1551, when a decree forbade daughters from acquiring shares in çifts by becoming partners with their brothers. 84 The law allowed one exception to this rule. To the clause in the kanunname of 1530–40 barring daughters from inheriting land, Celalzade added this comment:

if the deceased person cleared the land with his own axe and, in short, expended cash and effort, and if he died leaving no son or brother, but leaving a daughter, it is now commanded that it should be given to his daughter. 85

The new law – probably dating from the1550s – stipulated nonetheless that, in order to acquire the land, the daughter should pay tapu tax.

A major change followed in 1568, when, for the first time, a decree allowed ‘the lands of a deceased person who leaves no son to be given to his daughters who are requesting them, for the tapu [tax] which disinterested Muslims determine’. 86 This was to raise two problems. First, the decree placed no time limit on daughters’ claims, with the result that some demanded their inheritance many years after their father’s death, when the land had already been reallocated several times. In 1596, a decree following a petition put a limit of ten years from the death of the father. 87 The second question was whether, and on what terms, the daughter’s own children could inherit from their mother, the assumption from 1568 being that a son could inherit without paying tapu tax and that a daughter could inherit on payment of the fee. In 1604, however, a decree stipulated that land could descend to sons only on payment of tapu tax, raising the question of whether this applied also to daughters. The answer came in an order issued at some time before 1620: ‘the deceased [mother]’s vacant land descends by tapu , [and] only to her son’. 88

Inheritance by Sisters, Fathers and Mothers

In 1602, a decree extended the right of inheritance to a deceased man’s sister, provided that she paid tapu tax and was living in his place of residence. In the following year a petitioner asked for the residence requirement to be lifted, eliciting a decree which modified the original law by restricting inheritance to sisters living ‘in the region’. 89 In 1609 another decree, again in answer to a petition, extended the right of inheritance to the deceased’s parents:

it has been decreed that if the deceased leaves no son, daughter, or brother with whom he shares a father, but leaves a father and a mother, all his lands and meadows should be given to his father for the tapu tax which disinterested Muslims determine. If he leaves no father, they should all be given to his mother. 90

Between 1568 and 1609, therefore, a series of decrees had extended the right to inherit land to family members other than sons and brothers. The changes had occurred piecemeal, but nonetheless followed a series of consistent principles. Only sons could inherit without paying tapu tax. Males took precedence over females in the same degree of relationship, and descent was patrilineal. The changes had created a hierarchy among the heirs and, with it, the need for a concise statement of the rules. The task of providing one fell to the nişancı Okçuzade, who, in 1622, at the request of the mufti Yahya, summarized the laws of inheritance. It was Okçuzade’s summary which appeared in the New kanunname :

the deceased’s vacant lands are given by tapu to his daughter; if he has no [daughter], to his brother with whom he shares a father; if he has no [brother with whom he shares a father], to his sister resident in the [same] place; if he has no [sister resident in the same place], to his father; if he has no [father], to his mother … A deceased man’s land descends without tapu only to his son. 91


The laws of land tenure served two purposes. The first was to keep revenues flowing by ensuring that the land remained under continuous cultivation; the second, by controlling access to the land, was to prevent the growth of large private estates which would have diverted the revenues to private owners. The fundamental principles of the law remained intact between the fifteenth and nineteenth centuries, the major change being the extension, after 1568, of the right to inherit land to family members other than sons and brothers, and to women in particular. It was evidently demography that forced the change. Before 1568, the population had been growing and there was no shortage of men to till the soil. However, during the wars and rebellions of the late sixteenth and early seventeenth centuries, there was a flight of men from the land, and it was to ensure that the deserted çifts would not lie idle that the administration legalized their descent to the daughters, sisters and parents of the deceased.

Postcript 1

Who Owned the Land?

The question of who owned the land itself did not arise until the 1540s. Following the annexation of central Hungary as an Ottoman province in 1541, the kazasker of Rumeli, Ebussuud, provided a statement on the laws of land tenure that were to apply in the newly acquired territory. His statement not only gives a summary of the rules of land tenure, but attempts to do so within a framework of Islamic law which treats land as a commodity that can be held as private property. Ebussuud therefore faced the problem of deciding who actually owned the land. His solution was to fix ownership in the treasury, using the formula ‘the real substance of the land is the Treasury of the Muslims’. 92 In Islamic legal theory, the ‘Treasury of the Muslims’ is the joint property of the Muslim community. The compilers of the New kanunname in the 1670s adopted Ebussuud’s statement as the basic document on land law. Elsewhere, however, they make the contradictory statement: ‘ miri land’ – that is, land which was not controlled by a trust or by a private owner – ‘is the sultan’s’. 93 However, both they and Ebussuud are careful to avoid using any technical term which would imply that either the treasury or the sultan held the land as private property. They indicate ‘ownership’ by the simple use of the Turkish possessive genitive, which is legally non-specific. In reality, the question of ownership of the land was irrelevant. The important issue was who controlled the land or, more specifically, its revenues.

Postscript 2

From the Corpus Juris Civilis to the New Kanunname

Although the Ottoman land regime developed over the centuries to meet changing needs, its basic principles remained intact, and these, it is clear, were an inheritance from pre-Ottoman times whose distant origins are plain.

In the fifth century ce, Roman jurists faced the problem of describing a type of contract which transferred land to be enjoyed in perpetuity by a tenant and his heir or assignee, provided only that he continue to pay rent to the owner. When the jurists were unable to decide whether this was a contract of hire or of sale, the Emperor Zeno established it as an independent contract under the name emphyteusis. 94 It was a contract whose terms conform exactly to the conditions of peasant tenure in the Ottoman empire.

In the following century, the Emperor Justinian refined the law by decreeing that, in the absence of any agreement to the contrary, the owner could eject the emphyteutical tenant if he failed to pay the rent for three consecutive years. 95 The same rule was to apply to Ottoman peasants. Similarly, the distinction in Ottoman law between fruit-trees, vines and buildings, which could be held as private property, and the soil itself, which could not, seems also to date from the time of Justinian. It was this emperor who decreed that an emphyteutical tenant could, provided he had received permission from the owner, sell any improvements which he had made to the land. 96 Since it is impractical to sell improvements to the soil itself, this rule could apply only to things above the soil, most obviously to fruit-trees, vines and buildings. Since Justinian still requires the tenant to obtain the permission of the owner of the land before selling the improvements, these are not yet in the fullest sense the tenant’s private property as they were later to become in Ottoman law. Nonetheless, the separation in Ottoman law between the soil itself and things above the soil clearly has its origin in Justinian’s ruling.

It is possible also that the Ottomans inherited two further rules of late Roman law. First, when an Ottoman peasant sold his right to occupy a çift , his fief-holder levied a charge, fixed officially at one-tenth of the price realized. This seems to reflect a late Roman practice: Justinian attempted to curb abuses by landowners by limiting the sum which they could collect when an emphyteutical tenant transferred his land to another party, to one-fiftieth of its price or valuation. 97 Second, Ottoman law required a new entrant to a çift to pay an entry fee to the fief-holder. From the late fourth century ce, Roman law required a tenant who obtained land on the imperial domain by emphyteutical right to furnish sureties against the land being abandoned. 98 By Ottoman times, this payment had lost its character as a security, but the principle that occupancy of the land was conditional on the payment of an entry fee remained in place.

The Roman contract of emphyteusis, with some variation in its precise form, survived in the Byzantine empire and its successor states down to the Palaiologan era. 99 It is this contract that the Ottomans inherited and which was to become the basic form of peasant tenure in the Ottoman empire. It was the contract, in its Ottoman guise, that Ebussuud was to describe in summary in the 1540s and which the New kanunname of the 1670s was to regulate in detail, more than a millennium after Justinian had laid out the basic rules in the Corpus Juris Civilis.


See Postscript 1, p. 53.

See Postscript 2, p. 54.

Barkan 1943: no. 106; Kraelitz-Greifenhorst [1921–2] 1972.

Facsimile of Koyunoğlu MS, in Akgündüz 1990–96: II, 78–110.

Heyd 1973: 7–43.

Barkan 1943: 6.

Akgündüz 1990–96: II, 66.

MTM 1913: 99.

Barkan 1943: 220

Ibid. : 133.

Ibid. : 175.

Ibid. : 169.

Ibid. : 2.

Ibid. : 3.

Akgündüz 1990–96: II, 166.

Barkan 1943: 3.

Ibid. : 78.

MTM 1913: 86.

Barkan 1943: 30.

Ibid. : 254.

MTM 1913: 68.

Barkan 1943: 331.

Ibid. : 399, 401.

Ibid. : 389; Kraelitz-Greifenhorst [1921–2] 1972: 22, 36.

Tuncer 1963: 105.

MTM 1913: 83–4.

Barkan 1943: 3.

Ibid. : 25.

Imber 2008.

Barkan 1943: 7.

Ibid. : 47.

Ibid. : 47.

MTM 1913: 61.

Ibid. : 91–2.

Barkan 1943: 391–2; Kraelitz-Greifenhorst [1921–2] 1972: 25–6.

Barkan 1943: 2.

MTM 1913: 112.

Barkan 1943: 132, 174.

Ibid. : 282.

Ibid. : 390; Kraelitz-Greifenhorst [1921–2] 1972: 24, 38.

Barkan 1943: 288.

Akgündüz 1990–96: VI, 653.

Barkan 1943, no.89/34: 312.

Akgündüz 1990–96: VII, 330.

MTM 1913: 111.

Ibid. : 112.

Akgündüz 1990–96: II, 48.

Barkan 1943: 132.

Ibid. : 174.

Ibid. : 3.

Akgündüz 1990–96: II, 66.

Barkan 1943: 234.

Akgündüz 1990–96: IV, 394.

Ibid. : IV, 96–7.

Ibid. : III, 76.

Barkan 1943: 78.

MTM 1913: 79.

S¸ahin and Emecen 1994: 45.

Akgündüz 1990–96: II, 158.

MTM 1913: 51.

Barkan 1943: 299.

MTM 1913: 94.

Ibid. : 84.

Barkan 1943: 3.

Ibid. : 7.

Akgündüz 1990–96: III, 98.

MTM 1913: 61.

Barkan 1943: 3.

Akgündüz 1990–96: II, 128–9.

Ibid. : III, 78.

MTM 1913: 70.

Akgündüz 1990–96: II, 156.

Ibid. : II, 129.

MTM 1913: 60.

Akgündüz 1990–96: III, 98.

MTM 1913: 62.

Barkan 1943: 3.

Arif 1911: 51.

Akgündüz 1990–96: III, 99.

Barkan 1943: 189.

Ibid. : 46; Beldiceanu and Beldiceanu-Steinherr 1968: 47; MTM 1913: 68.

Akgündüz 1990–96: III, 98.

MTM 1913: 65.

Ibid. : 62.

Ibid. : 70–71.

Ibid. : 73.

Ibid. : 59.

Barkan 1943: 276.

MTM 1913: 84.

Justinian 1987: 115.

Corpus [1932] 1973: XIII, 135.

Ibid. : XIII, 135–6.

Ibid. : XIII, 136.

Ibid. : III, 224.

Weiss 1978; Zakythinos 1932: 185–7.

Search for more...
Back to top

Use of cookies on this website

We are using cookies to provide statistics that help us give you the best experience of our site. You can find out more in our Privacy Policy. By continuing to use the site you are agreeing to our use of cookies.