Core vs. Periphery: Monetary and Fiscal Institutions
in the Ottoman Empire, 1600-1800

by Sevket Pamuk, Bogazici University, Istanbul

paper presented at the GEHN Workshop on Imperialism
Istanbul, September 11-12, 2005

I- Introduction
In comparison to the empires of the last two centuries, early modern empires had to work with much more primitive communications and transport technology. In part because of this, their capacity to collect taxes was also more limited. As a result, there were significant obstacles in the way of the early modern empires to govern large territories.
The Ottoman Empire also faced these limitations but managed to live with them for six centuries. In the first part of this study, I will examine the Ottoman methods and style of administration across the large empire. I will also examine the variations in Ottoman institutions in different regions. This discussion will emphasize Ottoman flexibility and pragmatism in the face of the limitations. The Ottomans chose to develop rather loose arrangements across many of the territories they governed. They accommodated local forces and traditions and chose to live with many of the local institutions. Such an approach tells us a good deal about how the Ottomans viewed the large entity that we now call “empire”. A good part of the examples I will provide in this section are taken from monetary institutions across the empire as I have investigated these in detail in another study.
The second part of the article focuses on another major limitation of the early modern states and empires, namely the limited capacity to tax. The political and administrative capacities of early modern governments often determined the limits on fiscal revenue. Without an effective administrative network for tax collection, states were often forced to share tax revenues with other groups. The ability of the Ottoman central government to collect taxes declined sharply during the seventeenth and eighteenth centuries because of the decline of state power and the rise of local groups in the provinces. As a result, I estimate that the annual tax revenues of the central government as a share of the total GDP of the empire remained below 3 percent during the seventeenth and eighteenth centuries. The amount of tax collections retained by various intermediaries including high level bureaucrats, financiers and provincial groups were definitely larger than the tax receipts of the central government during these centuries. In the 1820s, however, the central government began to undermine the powerful alliance between the high level bureaucrats and financiers in the capital and the notables in the provinces. As a result, it was able to exert greater control over the tax collection process. The state was able to increase the revenues collected at the center roughly from 2 to 3 percent of total production in the 1770s to 5 to 6 percent in the 1840s and to 11 percent on the eve of World War I. If greater share of the central government in economic resources can be taken as an indicator of modernization, these efforts can indeed be interpreted as the onset of the modern era for Ottoman state finances.
I also examine in the second part of the article taxation and resource flows inside the empire to determine whether there were net flows from the periphery to the core regions and capital city during the early modern centuries. Tribute payments, exchange rate manipulations in payments flows, direct requisitioning and other mechanisms of resource transfer as well as taxation and the direction of tax revenues will be examined in this second part in this order to assess the direction of new resource flows inside the empire.

    The Ottoman Empire stood at the crossroads of intercontinental trade, stretching from the Balkans and the Black Sea region through Anatolia, Syria, Mesopotamia and the Gulf to Egypt and most of the North African coast for six centuries until World War I. For most of the seventeenth and eighteenth centuries, its population exceeded 30 million (of which the European provinces accounted for half or more; Anatolia and Istanbul for 7 to 8 million, other Asian and North African provinces for another 7 to 8 million) but declined thereafter due to territorial losses.
For most of its six-century existence, the Ottoman Empire is best characterized as a bureaucratic, agrarian empire. Until recently, Ottoman historiography had depicted an empire in decline after the sixteenth century. I have argued elsewhere that the Ottoman state and society showed considerable ability to reorganize as a way of adapting to changing circumstances in Eurasia from the seventeenth through the nineteenth centuries. Beginning with the successful centralization drive in the second half of the fifteenth century, Ottoman economic institutions and policies were shaped to a large degree by the priorities and interests of a central bureaucracy. This central bureaucracy managed to contain the many challenges it faced with its pragmatism, flexibility and habit of negotiation to co-opt and incorporate into the state the social groups that rebelled against it. The Ottoman state also showed considerable flexibility to adapt not only its military technology but also its fiscal, financial and monetary institutions in response to the changing circumstances.
A comparison with the other two Muslim empires of Eurasia, the Safavids and the Mughals brings the Ottoman trajectory into sharper focus. The political economy of these three empires showed similar patterns of evolution during the sixteenth and seventeenth centuries. They all enjoyed a long period of stability, agricultural expansion and growing prosperity during the sixteenth century followed by severe fiscal and military difficulties and rising internal conflicts during the seventeenth century. The decline of central political institutions in all three of the empires was accompanied by the rise of provincial elites which had greater say on the evolution of regional economies. During the eighteenth century, both the Mughals and Safavids disintegrated under the pressure of tribal invasions. While the Mughals were taken over by the British, the Safavids were replaced by a regional dynasty, the Qajars. In contrast, the eighteenth century until the 1770s was a period of recovery, stability and economic expansion for the Ottoman Empire. Despite wars and internal conflict from the 1770s through the 1830s, the Ottomans managed to regroup and survive into the modern era with a strong central state and many of their central institutions intact.    
If pragmatism and flexibility refers to the willingness of actors not to be bound in their actions and in the institutions they adopt by specific and rigid rules based on custom, traditions, religion or past behavior, the Ottomans were familiar with these traits from the earliest period. Emerging in a highly heterogeneous region populated by Christians and Muslims, Turkish and Greek speakers, the Ottomans’ success in western Anatolia and later in the Balkans during the fourteenth and fifteenth centuries owed much to their willingness and ability to adapt to changing conditions, to utilize talent and accept allegiance from many sources, and to make many-sided appeals for support. They were thus able attract many followers not only as warriors fighting against the Christians but also Muslims and Christians fighting for the riches to be gained, the positions and power to be won. The Ottomans also displayed remarkable openness to technological innovation, to adapt firearms on a greater scale, more effectively and earlier than the neighboring states. Similarly, they exhibited considerable degree of flexibility and pragmatism while expanding the territories under their control. They were prepared to negotiate for the loyalty of local elites whenever the new state was unable to impose full control. They also proved to be quite adept at learning about and borrowing institutions from others. In short, the early Ottoman enterprise was not a religious state in the making, but rather a pragmatic one.
A leading historian of the Ottoman Empire described this approach in the following terms:
“In the early period of their expansion, the Ottomans pursued, in order to facilitate conquest, or to make the indigenous population favorably disposed, a policy called istimalet. It was intended to win over the population, peasants and townspeople, as well as military and clerics, by generous promises and concessions, sometimes going beyond the well known, toleranst stipulations of Islamic Law concerning non-Muslims who had submitted without resistance. Within this policy of istimalet, the Ottomans, especially durng the first transition period, maintained intact the laws and customs, the status and privileges, that had existed in the pre-conquest times, and what is more unusual, they incorporated the existing military and clerical groups into their own administrative system without discrimination, so that in many cases former pronoia-holders and seigneurs in the Balkans were left on their fiefs as Ottoman timar-holders.   
Institutional change did not apply equally to all areas of Ottoman economic life, however. Moreover, not all types of institutions were affected to the same degree by these changes. Because the central bureaucracy was able to retain its leading position in Ottoman society and politics, the influence of various social groups, not only of landowners but also of merchants, manufacturers and moneychangers, over economic matters, and more generally over the policies of the central government remained limited until the end of the empire. As a result, most of the pragmatism and flexibility was utilized by the central bureaucracy for the defense of the existing order and of its own position.
In contrast, institutional changes that may threaten the leading position of the central bureaucracy position were resisted more forcefully than others. Institutional change thus remained selective and many of the key institutions of the Ottoman order such as state ownership of land, urban guilds and restrictions on private capital accumulation remained intact until the nineteenth century. We will also argue that selective institutional change led to very different pace and patterns of change in the three factor markets. Capital markets, especially those related to affairs of the state showed considerable change in the early modern era. In contrast, changes in labor and land markets remained limited as these institutions were defended fiercely by the bureaucracy. The same argument may be stated in different terms: those in favor of greater changes in these factor markets, for example landowners, merchants and manufacturers were not strong enough to overcome the opposition of the bureaucracy and other political forces until the nineteenth century.

    In this study I will examine monetary and fiscal institutions across the Ottoman Empire in the early modern era from this perspective of flexibility and pragmatism in order to assess the nature of the arrangements across the large empire and how the Ottomans dealt with the various limitations of their power and how they managed to keep this large entity together for six centuries. I begin below with a discussion of the economic institutions and economic policies.

II- Economic Institutions
Until the end of the sixteenth century, the rise of the Ottoman Empire was closely associated with military conquest. Military success, in turn, depended closely on the land tenure regime that supported a large, cavalry-based army. The Ottoman bureaucracy always took care to undertake detailed censuses of the new territories in order to assess their fiscal potential. Even after territorial expansion slowed down in the second half of the sixteenth century, agriculture continued to provide the economic livelihood for close to 90 percent of the population and key fiscal support for the Ottoman state. The durability of the empire, its achievements as well as limitations during the next three centuries can not be understood without studying its agrarian institutions.
The peasant family farm was the basic economic and fiscal unit in the countryside in most of Anatolia and the Balkans, the core areas of the empire where the relatively high land/labor ratios favored small holdings. The state established ownership over most of these lands early on and did not relinquish it until the second half of the nineteenth century. Hereditary usufruct of the land was given to peasant households which typically cultivated with a pair of oxen and family labor. Plagues in urban areas were frequent but large scale famines were rare in the Ottoman Empire thanks, in part, to the favorable land/labor ratios.
The state utilized the timar system to tax the rural population and support a large provincial army. Under this system, sipahis, state employees often chosen for their wartime valor, lived in the rural areas, collected taxes from the agricultural producers and spent the revenues locally on the training and equipment of a pre-determined number of soldiers as well as their own maintenance. The Ottoman central administration did not attempt to impose the timar regime in all of the conquered territories, however. In many of the more distant areas such as Eastern Anatolia, Iraq, Egypt, Yemen, Wallachia, Moldavia and the Maghrib, the Ottomans were eager to collect taxes but altered the existing land regimes either to a limited extent or none at all. The most important reason for this preference was the wish to avoid economic disruption and possible popular unrest. It was also not clear whether the central government had the fiscal, administrative and economic resources to establish a new regime in these areas.
The central government thus handled the task of establishing the land tenure cum fiscal regime for the expanding empire with a large degree of pragmatism. This approach was in fact quite similar to Ottoman practices in other areas. As a result, there emerged inside the empire zones with varying degrees of administrative control. At the core, were areas most closely administered by the capital with institutions most closely resembling those in the Istanbul region. With increasing distance from the capital, the institutions and administrative practices reflected the power balances between the capital and the local structures and forces. For example, the land regime and the fiscal practices in Ottoman Egypt remained closely linked to the demands of irrigated agriculture along the Nile valley.

III- Economic Priorities and Policies
States in the Old World had to address a common range of economic problems during the late medieval and early modern periods. The most basic of these problems were related directly to the maintenance of the states themselves. The provisioning of the capital city, the armed forces and to a lesser extent of other urban areas, taxation, support and regulation of long distance trade and maintaining a steady supply of money were amongst the leading concerns of economic policy.
State economic polices did not pursue public interest in some abstract sense of the term. Instead, both the goals and design of economic policies as well as institutions related to their implementation were shaped by the social structure, the relationship between state and society, the interests of different social groups aligned with or represented by the state, and more generally, by the social and political influences acting on the state. To put it differently, social actors molded state policy. Interest and pressure groups and social classes sought to protect and promote their interests through the state. In some cases the influence of a particular social group was so strong that the state simply acted in their interest, became their state. In other cases, the state was in the hands of a bureaucracy which acted independently or was insulated from these social groups. In the longer term, of course, the interaction and causality worked in both directions. State economic policies had consequences for both the economy and the social groups. Similarly, economic changes had consequences to which the state had to respond.
To understand Ottoman economic policies or practices, it is thus necessary to examine the nature of the Ottoman state and its relations with different social groups. Until late in the fifteenth century, there existed considerable amount of tension in Ottoman society between the Turkish landed aristocracy of the provinces, who were deeply involved in the territorial conquests, and a bureaucracy at the center made up mostly of converted slaves (devshirme), with the balance of power often shifting between the two. The successful centralization drive of Mehmed II in second half of the fifteenth century moved the pendulum again, this time decisively. The landed aristocracy was defeated, state ownership was established over privately held lands, and power concentrated in the hands of the central bureaucracy. After this shift, the policies of the government in Istanbul began to reflect much more strongly the priorities of this bureaucracy. The influence of various social groups, not only of landowners but also of merchants and moneychangers, over the policies of the central government remained limited.
The central bureaucracy tried, above all, to create and reproduce a traditional order with the bureaucracy at the top. The provisioning of the urban areas, long distance trade and imports were all necessary for the stability of that social order. The state tolerated and even encouraged the activities of merchants, domestic manufacturers more or less independent of the guilds and moneychangers as long as they helped reproduce that traditional order. Despite the general trend towards decentralization of the empire during the seventeenth and eighteenth centuries, merchants and domestic producers who were the leading proponents and actual developers of mercantilist policies in Europe, never became powerful enough to exert sufficient pressure on the Ottoman government to change or even modify these traditional policies. Only in the provinces, locally powerful groups were able to exert increasing degrees of influence over the provincial administrators.
In the fifteenth and sixteenth centuries, the Ottoman government intervened frequently in local and long distance trade to regulate the markets and ensure the availability of goods for the military, palace, and more generally, the urban economy. In comparison to both Islamic law and the general practice in medieval Islamic states, the early Ottomans were definitely more interventionist in their approach. In economic and fiscal affairs as well as in many administrative practices, they often issued their own state laws (kanun) even if those came into conflict with the shariat. The practices they used such as the enforcement of regulations (hisba) in urban markets and price ceilings (narh) had their origins in Islamic tradition but the Ottomans relied more frequently on them. In addition, in the provisioning of the army and the urban economy, deliveries at fixed prices were required from merchants for some of the more important goods.
Interventions in the economy did not necessarily mean that the government succeeded in bringing about the desired outcomes, however. Pre-modern states did not have the capability to intervene in markets comprehensively and effectively. There existed serious limitations on the administrative resources, organization and capacity of the states in the late medieval and early modern periods. They did not have the capacity to intervene in markets comprehensively and effectively. The mixed success of government actions inevitably led the Ottoman authorities to recognize the limitations of their power. As a result, Ottoman governments moved away from a position of comprehensive interventionism as practiced during the reign of Mehmed II (1444 and 1451-1481) towards more selective interventionism in the later periods.
A more realistic assessment of the nature of Ottoman state interventionism in the economy is long overdue. When the biases of archival evidence and the limitations on the power and capabilities of the state are taken into account, Ottoman policy towards trade and the markets, is best characterized not as permanent and comprehensive interventionism, but as selective interventionism. In the later periods, interventions were used primarily for the provisioning of selected goods for the capital city and the army and during extraordinary periods when shortages reached crisis conditions.
The limitations of the government are most apparent in the case of money markets. In comparison to goods markets and long distance trade, it was more difficult for governments to control physical supplies of specie or coinage and regulate prices, that is, exchange and interest rates. Ottoman administrators were well aware that participants in the money markets, merchants, money changers and financiers were able to evade state rules and regulations more easily than those in the commodity markets. Observing the mixed success of government actions, they learned that interventionism in money markets did not always produce the desired results. For this reason, government interventions in money markets became more selective after the reign of Mehmed II and occurred mostly during extraordinary periods such as extreme monetary turbulence or wars. On the whole, Ottoman monetary practices exhibited a good deal of flexibility and pragmatism after the fifteenth century.
One of the most telling examples of Ottoman flexibility concerned the determination of exchange rates between different kinds coinage. In an environment of frequently recurring shortages of specie, the Ottoman administrators knew that it was essential to attract into the Ottoman lands and maintain in circulation as much coinage and bullion as possible. Their monetary practices were guided more by this concern than any other. They were also aware that the ratio between gold and silver as well as the value of different types of coins was subject to fluctuations. Under these conditions, a policy of fixed exchange rates between different coins would have driven the good or undervalued coins out of circulation through the workings of Gresham's Law. Instead, the government allowed the local markets to determine not only the exchange rate of the sultani, but those for all types of coins, Ottoman and foreign. Local court records show that the kadıs relied on these market rates to settle disputes between individuals. In addition, the government announced the official rates at which different coins, gold and silver, would be accepted as payment. Usually, these rates did not diverge significantly from the prevailing market rates for the same coins.  
Government policies towards foreign coinage provides another example of flexibility. From the earliest days, the authorities encouraged the circulation of foreign coinage and accepted them as payment. The government also exempted precious metals and foreign currency from import dues. In capitulations or privileges given to merchants of certain European states, the central government exempted them from all customs duties for the foreign coinage they brought. In addition, customs and mints officials were told not to demand that these coinage be surrendered to the authorities for the minting of Ottoman coinage. These privileges were eventually extended to the merchants of most European states during the sixteenth century.

IV- Monetary Zones across the Empire

    To further illustrate Ottoman pragmatism and flexibility and the variation in institutions across the empire, I will examine in this section how the Ottoman handled the task of establishing a monetary system across the empire as the Ottoman state expanded rapidly during the sixteenth century to cover a large area and population in three continents. Such a "big picture" perspective should also offer important insights into the history and evolution of Ottoman institutions and the very concept of empire, the nature of the entity and how the Ottomans themselves viewed it.
    For a long time it has been assumed that the use of money in the Balkans and Anatolia was limited to long distance trade and parts of the urban sector. Recent research has shown, however, that the urban population and some segments of the countryside were already part of the monetary economy by the end of the fifteenth century. Even more significantly, there occurred a substantial increase in the use of money during the sixteenth century, both because of the increased availability of specie and increasing commercialization of the rural economy. The evidence for this important development comes from a number of sources. First, recent research has pointed that population growth and urbanization during the sixteenth century were accompanied by the growth of economic linkages between the urban and rural areas. As a result, there emerged in the Balkans and Anatolia an intensive pattern of periodic markets and market fairs where peasants and larger landholders sold parts of their produce to urban residents. These markets also provided an important opportunity for the nomads to come into contact with both peasants and the urban population. Large sectors of the rural population came to use coinage, especially the small denominations of silver akçe and the copper mangir, through their participation in these markets.
Until the sixteenth century, Ottoman territories in Anatolia and the Balkans had a unified monetary system based on the gold sultani and the silver akçe. At the bottom of the hierarchy was the copper mangir or pul with nominal values and for small transactions. As the Ottoman state territorially expanded to become a full fledged empire, however, this simple system could not be continued. The newly conquered territories, each of which was subject to different economic forces and very different patterns of trade, already had well-established currency systems of their own. The Ottomans pursued a two-tiered approach to money and currency in these areas. They unified the gold coinage at the existing international standards, but allowed the creation of multiple currency zones in silver in view of the sharply different commercial relations and needs of the new provinces.
In gold, the sultani became the only Ottoman coin across the empire. This was due to both symbolic and economic reasons. With a single gold coin, the ultimate symbol of sovereignty, the Ottomans thus unified the empire from the Balkans to Egypt and the Maghrib. The standards of the sultani, its weight and fineness, were kept identical to those of the Venetian ducat that had become the accepted standard of payment in long distance trade across the Mediterranean and beyond. Whether Ottoman gold coinage was issued in a given territory or not depended upon its status; whether it was part of the empire proper or whether it was considered a province with some degree of autonomy. Hence, the sultani was issued regularly in Egypt, Algiers, Tunis as well as the Balkans and Anatolia. In contrast, it was never minted in the autonomous Danubian principalities of Wallachia and Moldovia. Similarly, the autonomous Crimean khanate was able to issue its own silver coinage but no gold coinage was minted there for either the khans or the Ottoman sultans.
In silver coinage used in daily transactions and to some extent in long distance trade, the central government chose to retain many of the local currencies in the newly conquered territories. The most important reason for this preference was the wish to avoid economic disruption and possible popular unrest. It was also not clear whether the central government had the fiscal, administrative and economic resources to unify the silver coinage of the empire. As a result, while the silver coinage minted in the new territories began to bear the name of the sultan, their designs and standards as well as the names of the currencies adhered to the pre-Ottoman forms and usages in many instances. Just as the akçe was the silver coin and unit of account in Anatolia and the Balkans, the medin or para became the silver coin and unit of account in Ottoman Egypt and the surrounding areas. The shahi, a larger silver coin served the same purpose in Ottoman Iraq and the square-shaped nasri in Tunis. Earlier styles and types of copper coinage were also continued in all regions across the empire.
Another important silver piece produced by the Ottomans was the lari used in the Gulf and on the Indian Ocean during the sixteenth century. The lari was an unusual type of money, a small rod of pure silver the size of the "pen of a goose feather", but twisted and folded in the middle so that the two ends met. At the head was a stamp with the inscriptions of the mint. Its origins went back to the region of Laristan, which was on the Hormuz-Shiraz caravan route on the Persian side of the Gulf. The laris began to be minted in the fourteenth century and became increasingly popular as a medium of payment in long distance commerce on the Indian Ocean. In the Gulf area, the laris were the leading form of payment for the goods arriving from the east. The laris did not circulate inland in large quantities, however.

Core and Periphery in Ottoman Europe
The Balkans together with western and central Anatolia including the capital city and its environs constituted the core region of the Ottoman monetary system. The silver akçe was both the leading unit of account and the leading means of exchange in this region. For large transactions and hoarding purposes, the sultani was used together with European gold pieces.  The mint at the capital city, or "Kostantaniyye" as it was called on Ottoman coinage until the eighteenth century, was the leading mint of the region for both the akçe and the sultani. The Venetian gold ducat remained the most important foreign coin circulating in the Balkans and Anatolia during the sixteenth century. In the second half of the century, large European silver coins called groschen, most importantly the Spanish eight real piece and the Dutch lion thaler, also began to circulate in the Balkans and Anatolia.
Even though the Ottomans governed Hungary directly, in contrast to their practices in territories lying south of the Danube, they did not mint akçes or sultanis in Hungary during the sixteenth and seventeenth centuries. One important reason was the absence of silver and gold deposits in the Ottoman regions of Hungary although major gold mines were located in other parts of Hungary and these had supplied large areas of Europe ever since the late medieval period. Evidence from coin hoards indicates that akçes and sultanis produced in the Balkans circulated in Hungary, albeit in limited volume. Payments by the Ottoman state to the troops or suppliers was an important source of the akçes in Hungary. Circulating more widely were silver coinage of the neighboring states, most importantly the silver and gold coins of the independent Hungarian principalities, the small silver groats, zweirs, pfennings from the Habsburg domains, small silver coins from Poland such as the half-groats, and increasingly from the second half of the sixteenth and during the seventeenth century, the large thalers. Nonetheless, all taxes in Ottoman Hungary continued to be specified in akçes. In other words, while the akçe served as the unit of account, at least in governmental transactions, the coinage of the neighboring states served as the leading means of exchange and payments including taxes.
In contrast to the rest of the Balkans, the Danubian principalities were never fully incorporated into the Ottoman empire but became vassal states paying regular tribute, Wallachia beginning late in the fifteenth century and Moldovia in the early part of the sixteenth century. These principalities were mostly independent in their internal affairs and did not adopt Ottoman institutions such as the timar land tenure system. The fact that neither the Ottomans nor the local governments minted any coins, gold, silver or copper, in Wallachia or Moldavia is suggestive of both the extent and limits of the autonomy of these principalities.
During the sixteenth century, akçe was the leading unit of account in Wallachia but not in Moldavia. Evidence from coin hoards indicate that akçes minted in the Balkans and Istanbul circulated widely in these principalities during the sixteenth century, much more so than the Ottoman provinces of Hungary. Akçes and sultanis accounted for more than 86 percent of the value of all coins found in Wallachia dating until the 1580s, but only for 38 percent during the last two decades of the sixteenth century. The corresponding percentages were lower for Moldavia at 26 percent and 7 percent, respectively. In both cases, the balance was made up primarily by Hungarian coins in the earlier period and increasingly by the large silver coins from central and western Europe. In addition, Polish silver coins, especially the half groats, were significant in Moldavia at the end of the century. Circulation of gold coins remained limited.
The exact nature of the relationship between the Crimean Khanate and the Ottoman state, to what extent the Khan was a sovereign and heir to the political traditions of the steppe and to what extent he was a vassal of the Ottoman sultan, has long been debated. It is clear, however, that the Khanate enjoyed a unique relationship and status amongst all territories considered as part of the Ottoman Empire. After the incorporation of Crimea into the Ottoman lands in 1478, Caffa and part of the Crimean shore simply became another Ottoman province. The rest of the peninsula continued to be ruled by a hereditary family of khans who joined the Ottoman army as tributary, providing manpower during military campaigns. Although the Ottomans played a role in the choice of the khans, they usually accepted the selection made by the Crimean aristocracy.
The Crimean khans continued to display one of the most important symbols of steppe sovereignty, the Cengiz seal (tamga). They also retained the right to maintain diplomatic relations with Muscovy and Poland. Until the end of the seventeenth century, the Khanate received tributes of varying amounts directly from Muscovy, Poland and the Danubian Principalities. The relationship between the Crimean coinage and the Ottoman monetary system was thus unique. The Crimean khans minted their own silver coins bearing the seal of the Giray dynasty without the name of the Ottoman sultans. Yet, they were not sufficiently independent to mint their own gold coins, the ultimate symbol of sovereignty. The Ottoman gold sultani was never minted in Crimea either. During the century following the Ottoman conquest, the khans discontinued the use of adjectives such as "sultan" and various expressions of sovereignty used earlier and displayed only their names on the coins. They resumed the use of the term "sultan" and "khan" in the second half of the sixteenth century, however. The Crimean coinage was only loosely related to the Ottoman monetary system even though Crimea was considered part of the Ottoman empire in some senses of the term. It may be useful to compare the status of the Crimean Khanate in monetary affairs with those of the principalities of Wallachia and Moldavia which were also autonomous in their internal affairs. In the principalities, too, gold coins, Ottoman or otherwise, were not issued. Crimean autonomy in monetary affairs was greater, however, since the Khans issued silver and copper coinage with their own name but the principalities did not.
The basic coin and the leading unit of account in Crimea during the Ottoman period was a small silver coin called akçe, which was referred to in the Ottoman sources as Kefevi akçe (the akçe of Caffa) although akçes were also minted elsewhere in Crimea. The small akçe remained the basic coin of the Khanate until 1774, issueing coins with the names of the Crimean Khans. Copper coinage, with denominations of one akçe and its fractions were also minted for daily use. While larger coins such as the six-akçe piece were also minted in later periods, these could not meet the demands of trade and the economy. The fact that the Khanate did not issue gold coinage is fully consistent with its less than independent status during the Ottoman period. In the absence of gold, however, domestic silver and copper coinage could not meet the demands of the economy and trade. As a result, large European silver coins such as the Polish zloty as well as the internationally more prominent Spanish eight-real piece and the Dutch thaler circulated widely in the Khanate during the seventeenth century.
Even though the politically unique status and autonomy of the Crimean Khanate continued until the 1770s, with Ottoman military defeats Russian pressure on the Khanate increased steadily during the eighteenth century. The status of the Khanate changed in 1774 from autonomy within the empire to full independence and it was annexed by Russia nine years later. The special status of Crimea vis-a-vis the Ottoman monetary system, autonomous but not free of influence from Istanbul, also ended in 1774. Nonetheless, Crimean coinage and monetary practices during the nine year period of independence are very relevant and interesting for what they reveal about the nature and symbolism of the practices during the period of autonomy. In the early 1780s, just a few years before independence was ended by Russia, Şahin Giray, the last of the Crimean khans, began to mint in Caffa large gold coins bearing his name, the first for any Crimean ruler.
This survey of Ottoman monetary practices in both the core and frontier regions of Europe suggests that while the Ottomans paid attention to the fiscal and economic needs of each territory, they were quite careful about the political symbolism of their actions. More specifically, they followed certain rules regarding what types of Ottoman coins can be minted in territories with different administrative and political status. Whether Ottoman gold or silver coinage was issued in a given territory or not depended upon its status; whether it was part of the empire proper or whether it was considered a province with some degree of autonomy.

V- Examples of Payments Flows Across the Empire
The Balkans
During the early period of expansion in the fourteenth and fifteenth centuries, the Ottomans demanded one-time or annual tributes from vassal states, often paid in gold ducats. A Venetian survey of the receipts of the Ottoman state in the second half of the fifteenth century listed the following sums paid annually by the vassal countries : Bosnia and Herzegovina, 18,000 ducats; Wallachia, 17,000, Moldavia, 6,000; Trebizond, 3,000; Kaffa, 3,000; Amasra and Sinop, 14,000. The tributes paid by the Venetian posssessions in the Morea and in Albania were not included in this list.
For the vassal principalities of Wallachia and Moldavia which were governed by local princes until late in the seventeenth century, annual tribute payments were the most important form of specie outflows during the sixteenth and seventeenth centuries and these were usually balanced by their trade surpluses. The tribute payments had remained small until the end of the fifteenth century, below 10,000 Venetian ducats for each principality but began to exceed 50,000 ducats around the middle of the sixteenth century and 100,000 ducats in the second half of the century for Wallachia and averaged close to 35,000 ducats per year for Moldavia. In addition, the principalities sent to Istanbul foodstuffs and various raw materials at official prices established  by the Ottoman government. These commodities played an important role in the provisioning of the capital, army and the palace.
Another important source of monetary or payments flows across the Balkans was the military campaigns. Part of the provisioning needs of a campaign was typically financed by ordinary or extraordinary taxes collected mostly in kind but some in cash in the regions through which the army moved. The army which often exceeded 100,000 soldiers during the sixteenth century also purchased some of its supplies. For this purpose, large sums were often sent from the imperial treasury. In addition, the soldiers continued to receive their salaries during the campaigns and these often reached large sums. These large sums were then injected into the local economies. The payments flows increased proportionally when the campaigns lasted longer and large numbers of troops were kept at the frontier for extended periods. Total sums spent in one of these campaigns often reached millions of Venetian ducats.
Caroline Finkel provides a detailed account of Ottoman spending during the unusually long military campaign against the Habsburgs in Hungary at the turn of the seventeenth century. The imperial treasury paid out a total of 380 million akçes or the equivalent of 3.2 million gold ducats during the 11 month period beginning in July 1599 and 310 million akçes or approximately 2.5 million gold ducats for a two year period beginning in July 1602. Records of the central government show that 67 percent of these payments were made by gold coin, 23 percent in silver akçes and approximately 10 percent in large European silver coins. Approximately 70 percent of these expenditures were payments of wages to the troops.

Egypt and the Holy Cities
The Ottomans did not change the land tenure system or the momentary system in Egypt, one of the largest and richest provinces of the empire in the sixteenth century but demanded a significant amount in annual tribute payments from Cairo to the imperial treasury in Istanbul. During the sixteenth century, this amount fluctuated around 400,000 to 500,000 gold pieces, or 16 to 20 million paras at the prevailing rate of exchange of 40 paras per sultani. This was a huge sum by the standards of the sixteenth century and it was a major addition to the annual receipts of the imperial treasury, even at the zenith of its power.   Stanford Shaw provides a colorful account of the early evolution of this payment :
"In the early period of Ottoman rule ... the yearly remittances were set at 500,000 gold pieces. ... After the appointment of Hosrev Paşa as vali (governor) of Egypt, the annual remittances were raised to 700,000 gold pieces or 28 million paras a year, at his own request and in the year 1535-36 he sent to Istanbul more than one million gold pieces. When it arrived in Istanbul, however, the Sultan (Süleyman I) refused to accept it saying that it was too much and expressing the fear that it had been taken tyranically from the poor. Hosrev Paşa had hoped to impress his master by the attention of his collections, and he replied that he had been able to collect that much by special efforts in the border regions of Egypt. But the Sultan ordered that the money collected in such a way could be spent only for the water cisterns of the Muslims in the Porte and the Holy Cities, and that thereafter the remittance should be no more than 500,000 gold pieces, every year."

    This pattern of payment and specie flows changed substantially during the seventeenth century. The annual remittance to Istanbul began to be sent mostly in silver. This shift suggests that gold flows from the south slowed down after the sixteenth century, if not earlier. As political and administrative disorders caused revenues in Egypt to decline and Istanbul's control over Egypt disappeared, however, for many years until the last quarter of the seventeenth century, nothing at all was sent to the capital.   Gold coinage from Cairo regained its prominence in Istanbul in the late seventeenth and early eighteenth century suggesting that gold flows from the south to Cairo resumed around this time if not earlier. During the first half of the eighteenth century annual payments fluctuated between 8 and 30 million paras, averaging 18 million paras or 135 thousand gold pieces per year. In addition, in the eighteenth century the annual remittances sent on account of the sultan from Egypt to the Holy Cities in Hijaz rose from about half million paras to 10 million paras.

    In addition, Cairo shipped sugar, rice, coffee and other commodities to the capital city every year. It is likely, however, that these shipments were also discontinued after the sixteenth century.

    One issue of almost permanent concern for Istanbul was the lower standards or lower silver content of the para of Egypt vis-a-vis the para and kuruş of Istanbul. Since the exchange rate between the two units remained fixed, the divergence in the respective silver contents led to an outflow of silver from the Istanbul region to Egypt. Another reason for the concern of the central government was the annual remittances sent from Cairo to Istanbul. The annual remittance from Cairo to Istanbul was often made in paras of Cairo and converted to gold sultanis at the official rate of exchange. The decline in the standards of the para thus meant lower actual receipts for the treasury in Istanbul.
Despite Istanbul's efforts, however, the difference between the silver content of the coins minted at Istanbul and Cairo persisted. Even though this gap occasionally reached 20 or even 30 percent, the para of Cairo remained linked to the kuruş and the para of Istanbul on a long term basis. As the latter lost about 40 percent of their silver content from the 1720s to the 1760s, the Egyptian unit followed it downwards. The exchange rates of the two units against the benchmark ducat also show that the two remained well linked. This linkage was severed in the 1760s, however, as the economic and fiscal crises in Egypt led to a sharp decline in the silver content of the para to about the half of those of corresponding coins minted at Istanbul.

    The Holy Places in Hijaz and the pilgrimage to Mekke gave rise every year to one of the largest payments and specie flows within the Ottoman Empire. The financing of the caravans including provisioning, payments to tribal leaders en route for security and funds carried by tens of thousands, and in some years close to 100,000 pilgrims gave rise to large flows of gold and silver from Egypt, Syria and Anatolia to the Hijaz every year. Even more importantly, the governments in Istanbul and Egypt and the various official, semi-official and private foundations sent large sums every year to support the Holy Cities. In her detailed study, Suraiya Faroqhi estimates that official remittances from Cairo and Istanbul were roughly equal in magnitude, fluctuating mostly between 50,000 and 100,000 gold sultanis per year from each during the sixteenth and first half of the seventeenth centuries.  In addition, the annual revenues of many small and large pious foundations (vakif) in Anatolia and some of the largest foundations in Egypt and were set aside for the Hijaz. Total remittances by the foundations roughly equalled the amounts sent by the governments in Istanbul and Cairo. From Egypt, some of these net revenues were sent in kind, as cereals. Faroqhi thus estimates that a total of 300,000 to 400,000 sultanis were sent to the Hijaz every year from Istanbul, Anatolia and Egypt combined in addition to the payments and specie flows arising from the pilgrimage caravans themselves. The funds in cash were sent in gold whenever available, because gold was the preferred specie in the Hijaz.   Annual payments flows to the Hijaz were not as large as government spending on military campaigns in the Balkans which exceeded 600 thousand gold pieces excluding payments to the soldiers and reached 2 to 3 million gold pieces in one year including payments to the soldiers; but they were huge by any other standard.   The size of the flows to the Hijaz, from both official and private sources also gives a good indication of the importance attached to the Holy Places by the Ottoman government and society at large.

The Maghrib
Northwest Africa stretching from Tripolitania in the east to Algeria in the west was the scene of a major struggle between the Ottoman and the Spanish empires during the sixteenth century. Algeria, Tunisia and Tripolitania were eventually incorporated under Ottoman rule thanks to the military power of the corsair leaders. The Ottoman government then appointed governors to each of these territories. However, they were actually ruled by the corsair leaders and the leaders of the janissaries who rose through the ranks after being recruited from Anatolia as ordinary soldiers.
Until the eighteenth century these local governments limited their rule to the coast and urban areas. They collected taxes from the peasants and nomadic tribes in the interior, but beyond that, did not intervene in the internal affairs of the rural population. They directed their efforts to pirating which prospered because of the weakness of the European navies in the Mediterranean.
The Ottoman government regarded these territories as distant provinces with continuing links to Istanbul. The provinces, however, behaved more like autonomous states. The autonomy was made easier by their remoteness and the increasing weakness of the Ottoman state. Nonetheless, the rulers of these provinces were reluctant to severe all ties to Istanbul and declare complete independence. In addition to religious and political reasons of legitimacy, they needed to continue recruiting soldiers from Anatolia. For that, the good will and permission of the Istanbul government was essential.
Since the Maghrib did not possess any silver or gold mines, the availability of specie and mint output depended primarily on external trade balances and revenues from corsairing. The Mediterranean trade was the most important but the territories also had access to sub-Saharan gold through the caravan trade with that region. The gold output of the mints in Algiers, Tunis and to a lesser extent Tripoli depended, above all, on these inflows of gold from the south.
While Egypt was expected to send an annual remittance to Istanbul, the same did not apply to the Maghrib, Tripoli, Tunis and Algiers. Aside from occasional gifts to the sultan and the influential in the capital, regular payments to the state treasury were not expected from these provinces. The Ottoman government contented itself with demonstrations of submission from this distant province without contesting the autonomy its rulers enjoyed.

The “Convergence” of Ottoman Currencies, 1720-1850 
As mentioned earlier, there is a large amount of evidence indicating that Istanbul followed monetary standards in Cairo very closely during the eighteenth century. In contrast, no evidence has so far been located either at the Ottoman archives at Istanbul or at the mint archives in Tunis to suggest that the central government at Istanbul tried to control the course of the currency at Tunis. The circumstances behind the depreciation of the riyal of Tunis in the eighteenth century are not yet well understood. If the experience of the kuruş at Istanbul is any guide, fiscal causes probably played an important role, but additional research on the monetary and fiscal conditions in Tunis are needed before that question can be answered more satisfactorily.
In contrast, the gold coins issued by the Tunis mint during the first half of the eighteenth century continued to follow the design and the standards of the Ottoman sultanis even though the latter were discontinued at Istanbul and Cairo in the 1690s. The sultanis and half-sultanis minted at Tunis had lower gold content than their seventeenth century counterparts in the eastern Mediterranean, however. The standards of the sultanis of Tunis may have been following the zeri-mahbub gold pieces minted in Istanbul and Cairo at that time. In any case, after mid-century, the gold pieces of Tunis began to be referred to as the zeri-mahbub as well.   The volume of these issues were limited, however, especially in relation to the gold coins minted at Algiers.   They were used in the Mediterranean trade and carried to the eastern Mediterranean by the European merchants.

    During the crisis of the seventeenth century the ties between Istanbul and the currencies in different parts of the empire, Egypt, Tripoli, Tunis and Algiers had weakened substantially if not dissolved altogether. In contrast, the eighteenth century was a period of recovery and stronger linkages between the center and periphery of the Ottoman Empire. With the establishment of the new kuruş and centralization of mint activity in the core regions of the empire, the imperial mint in Istanbul was reasonably successful in supplying silver coinage a large geographical area from the Balkans to Anatolia, as well as Syria and Iraq. There also occurred growing interaction between the silver currencies of Egypt, Tripoli, Tunis, Crimea and Algiers and that of Istanbul during this period. These linkages were strongest for Cairo and Tripoli but weaker for Tunis, Crimea and Algiers. This picture based on money and currencies may appear paradoxical because eighteenth century is generally regarded by historians as a period of increasing de-centralization of the empire.
Another important development involving Istanbul and these provinces took place in the early decades of the nineteenth century when Cairo (1834), Istanbul (1844) and Tunis (1847) all undertook virtually identical monetary reforms, adopting the bimetallic system with fixed exchange rates between gold and silver coinage, and at the same time, abandoning the debasement of silver currencies as a means of raising fiscal revenue. It was not so much the interaction between these governments that brought about this shift. Instead, it was primarily the rapid increase in trade with Europe, the growing interaction with European merchants and governments as well as their advice and pressure that led governments in Cairo and Tunis as well as Istanbul to embrace those monetary institutions that conformed to the requirements of international trade.
The decision to abandon debasements as a means of raising fiscal revenue without the elimination of the budget deficits, however, proved to be very costly in the long term for all three governments. All three began to borrow in the European financial markets during the 1850s in order to meet their short term budgetary needs. By the middle of the 1870s, with their annual debt payments far in excess of their ability to pay, all of them were forced to declare moratoriums on their outstanding debt. The establishment of the European Public Debt Administration in Istanbul (1881), and even more  dramatically, the occupation of Tunis (1881) and Egypt (1882) by the European powers were directly linked to these moratoriums.

VI- Taxation and Fiscal Deficits to the End of the Empire
In the longer term, however, the more important obstacle against the ability of the Ottoman state to hold the large empire together lay not in the direction of resource flows but in the capacity of the Ottoman state to collect taxes. The political and administrative capacities of early modern governments often determined the limits on fiscal revenue. Without an effective administrative network for tax collection, governments were often forced to share tax revenues with other groups.
During the sixteenth century when the central government was more powerful and more effective in the provinces, the state had utilized the prebendal timar system to tax the rural population and support a large provincial army. Under this system, sipahis, state employees often chosen for their wartime valor, lived in the rural areas, collected taxes from the agricultural producers and spent the revenues locally on the training and equipment of a pre-determined number of soldiers as well as their own maintenance. As a large part of the tax collections never reached the treasury, the share of tax revenues in the total GDP of the empire remained below 3 percent during the sixteenth century.
The effectiveness of the timar based cavalry army began to decline towards the end of the sixteenth century. As a result, the central government abandoned the timar system in favor of tax-farming in order to receive a larger share of the tax revenues at the center and use them to maintain a larger, permanent army. Despite this shift, the central government was unable to increase its tax receipts, however. The ability of the Ottoman central government to collect taxes declined sharply during the seventeenth and eighteenth centuries because of the decline of state power and the rise of local groups in the provinces.
Unable to check the growing power of the provincial notables, the Ottoman state was forced to share the tax collections with various intermediaries. I estimate that the annual tax revenues of the central government as a share of the total GDP of the empire remained below 3 percent during the seventeenth and eighteenth centuries. The amounts retained by the high level bureaucrats, large financiers in the capital and the local notables in the provinces must have easily exceeded the amounts annually reaching Istanbul. They may have been as high as twice the revenues (3 percent of GDP) entering the treasury. As a result, state finances came under increasing pressure in the seventeenth century and again from the 1770s onwards, especially during periods of war. Fiscally motivated debasements were used rather frequently during these periods. New instruments for public borrowing began to emerge during the eighteenth century in response to these fiscal pressures.
In the early part of the nineteenth century, the center, supported by the new technologies, was able to re-assert its power over the provinces. After the central government began to undermine the power of the provincial notables in the 1820s and 1830s, many of the long term tax collection contracts were pulled back to the center and their revenues began to be collected once again by centrally appointed tax farmers. The malikane or the life-term tax-farming system was phased out in the 1840s as part of a larger package of administrative and economic reforms. With the same package of reforms the central government also attempted to eliminate short term tax farmers. This last step failed, however, due to the administrative limitations of the central government. Short-term tax-farming continued until World War I. After the centralizing reforms of the nineteenth century, Istanbul was able to increase the ratio of tax revenues to GDP from about 3 percent to more than 5 percent around mid-nineteenth century and to more than 10 percent in the early part of the twentieth century. Yet, the costs of keeping this large empire together continued to rise at a faster pace and outstrip revenues during the nineteenth century. State finances remained under pressure until the end of the empire. State borrowing in the European financial markets led to a default in the 1870s and partial control of state finances by European creditors until World War I.

VII- Conclusion
Faced with limited capacity to impose their institutions across a very large empire, the Ottomans chose to live with rather loose arrangements in many of the territories, especially in many of the more distant territories they governed. Detailed examples taken from Ottoman monetary institutions showed that they accommodated local forces and traditions and chose to live with many of the local institutions.
Our detailed but not comprehensive survey of the resource flows inside this empire during the early modern era suggests that while there were some net flows from the periphery to the core, most notably in the case of sixteenth century Egypt and to a lesser extent in Wallahia and Moldavia during the same period, the amounts of these resource transfers remained very very small in relation to the size of the economies involved. In fact, the net resource transfer was in the other direction, from the center to the periphery in some cases.
In the longer term, however, the more important obstacle against the ability of the Ottoman state to hold the large empire together lay not in the direction of resource flows but in the capacity of the Ottoman state to collect taxes. The political and administrative capacities of early modern governments often determined the limits on fiscal revenue. Without an effective administrative network for tax collection, states were often forced to share tax revenues with other groups. The ability of the Ottoman central government to collect taxes declined sharply during the seventeenth and eighteenth centuries because of the decline of state power and the rise of local groups in the provinces. As a result, I estimate that the annual tax revenues of the central government as a share of the total GDP of the empire remained below 3 percent during the seventeenth and eighteenth centuries. The amount of tax collections retained by various intermediaries including high level bureaucrats, financiers and provincial groups were definitely larger than the tax receipts of the central government during these centuries.
After the centralizing reforms of the nineteenth century and with the help of the modern technologies, military and otherwise, the center was able to impose its will more strongly on the rest of the empire and increase steadily the tax revenues reaching the treasury. Yet, the costs of keeping this large empire together continued to rise at a faster pace and outstrip revenues during the nineteenth century. State finances remained under pressure until the end of the empire.