Question about Mamluks

A dialog from the H-Mideast-Medieval list, August 1998

Hi everyone,

Can someone give me some details as to how Mamluks were transported from the steppes to the main centers of the Islamic Empires?

Was there a center for their purchase in a specific spot?

How did they move, say if thier destinantion was Egypt, but cities along the way where at odds with the Fatimids?

Also, who would be able to buy a Mamluk? And for how much?

Sorry for all these questions, but I am in the seminal phase of a short story about this issue and would love as much info as possible. Please also excuse me if my questions sound silly!

Thanks,

Fawzan

Your query should be about the mamluks (lower case) instead of the Mamluks (upper case) since the period in question here stretches as far back as the Fatimids. However, here are some thoughts which I contribute in haste on the mamluk slave(for want of a better word) trade during the Ayyubid and early Mamluk periods:

The Mamluk army comprised various ethnicities one of which was Turkic. Dubbing the Bahri period (1250-1382) Turkic -as is the case in some Arabic writings- is therefore misleading and should be viewed as a misnomer.The lingua franca of the Mamluk sultanate during both the Bahri and Burgi(Circassian) periods was Qipchaq, a Turkic language. Turkic mamluks hailed from the Qipchaq Steppe. Their homeland since the early middle ages was the basin of the Irtish River, which separates Siberia from Kazakhstan. A group of them migrated south to the Syr Darya (Sayhun) basin in the 12th c., while another group relocated to eastern Europe. Both groups came under the direct rule of Gengiz Khan, especially after Batu Khan moved north and subjegated the Qipchaqs, Circassians, and Russians. The kingdom of Batu Khan extended from Khwarizm in the east to the outskirts of Constantinople in the west, and from the Russia in the north to the Caucasus in the south. His capital was Saray on the Volga.

It was during this time that mamluk slave trade burgeoned, with the emergence of an active class of slave merchants. The Ayyubid sultan al-Kamil saw in the Turkic mamluks a substitute for the Khwarizmians who by then had proven to be unreliable and corrupt.These newcomers were ensconced in his citadel on the Rawda Island on the Nile (Qalat al-Bahr), hence the name Bahri.

This explains why most of the Bahri Mamluks who ran the first Mamluk state were Qipchaq. Sultan Aybak (1250-57) was of a Turkish descent and was therefore known as al-Turkumani (Maqrizi, Suluk, 1:368). Sultan Qutuz (1259-60) was Mongol despite his claim to be the nephew of Khwarizm Shah (Ibn Taghri Bardi, Nujum, 7: 3, 85). Sultan Baybars I (1260-77) was born in the Qipchaq and brought to Cairo via the slave (mamluk) market in Sivas(Nujum, 7:96, 145). Sultan Qalawun (1280-90) was also a Qipchaq (Nujum, Suluk, 1: 663). It was a standard practice of sultans of Qipchaq descent to recruit or purchase mamluks of their Qipchaq ethnicity. Baybars was not satisfied with the large numbers of mamlluks who were brought to him in Cairo as prisoners of his wars in Asia Minor against the Seljuqs and Mongols. He dispatched slave merchants to the lands of the Tatar (Maqrizi, Suluk, 2:89; Ayni, Aqd al-Juman, 56: 11) to purchase mamluks to shore up his forces. Qalawun continued Baybars's legacy of purchasing mamluks directly from lands under Mongol rule (Ayni, 56: 724; Ibn al-Furat, Tarikh al-Duwal wa al-Muluk, 8: 97). However, he seems to have had a leaning towards the Circassian element, which he had housed in the towers (Abraj, pl. of Burj) of the Citadel (Qalat al-Jabal), hence the name Burji (Nujum, 7: 330; Suluk: 1:755-56;Maqrizi's Khitat, 3: 348). The Mamluk-Mongol wars yielded large numbers of women and children slaves who were sold at the various slave markets in Syria and Egypt (Baybars al-Dawadar, Zubdat al-Fikrah, 9: 268), especially during the reign of Sultan al-Nasir Muhammad (Zubdat, 9:384) prior to the detente. It was also during this period that the Genoese merchants took an active role in this trade and sought to secure large numbers of Tatar children to be sold in the Crimea. This prompted Tuqtuq Khan of the Qipchaqs to raid the slave markets in the Crimea in an attempt to stiffle the trade. The trade does not seem to have abated in the least (Zubdat, 9: 460).

The Arab geographers called the Caucasus region Umm al-Lusun (mother of tongues), since almost every village had its own language. The preponderance of the Qipchaq ethnicity during the late Ayyubid and Bahri Maluk periods may account for the choice of Qipchaq, one of many regional lingua francas, as the lingua franca of the Mamluk court throughout Mamluk rule, including the second period when the Circassian element was at the helm.

Sabri Jarrar

There's a really good book on this subject in the Men-at-Arms series. It's called The Mamluks and shoud run you about $10-15. It has drawings of clothing as well as period artwork and goes into great detail about many aspects of Mamluk life, especially for such a tiny book. I don't know a lot about this topic myself, but this series is generally held to be excellently researched.

Ghazallah al-Badriyyah
http://www.mindspring.com/~whill/Ghazallah
Lauren Grover in Candia, NH
Dancer and Drummer with Qamar al-Franji

At 11:47 AM 8/4/98 -0400, Sabri Jarrar wrote:

I am grateful to Sabri Jarrar for his helpful notes on the identity of the Mamluks. A few comments and questions rise from it.

When he says,

> The lingua franca of the Mamluk sultanate during both the
> Bahri and Burgi(Circassian) periods was Qipchaq, a Turkic language.
> (SJ)

I feel sure he meant to say the house-language of the Mamluk military corps, and its court, were Qipchaq Turkish. It is for this reason that the Arabophone natives of the Sultanate in Egypt and Greater Syria referred to the Mamluks as Dawlat al-Atrak, the regime of the Turks.

The Qipchaqs seem to be the people referred to by the Byzantines as Cumans, and by the Russians as Polovtsi. While originally Turks, they seem to have carried off Slavic women and children in large numbers in Eastern Europe and the Volga Steppes, and thus increased their own numbers. Though in later times they seem all to have been Muslims, some of them appear to have converted to Byzantine Christianity before the Mongol Period. (JAW)

> al-Kamil saw in the Turkic mamluks a substitute for the Khwarizmians
> whoby then had proven to be unreliable and corrupt. (SJ)

While earlier Ayyubi Sultans had Turkish Mamluks, it is generally said that it was al-Salih Ayyub, last of the Ayyubi Sultans of Cairo, who greatly augmented his military slave corps, because he no longer trusted the hereditary cavalry of Turks and Kurds who beefed up Ayyubi armies in Egypt and Greater Syria.

Once the Mamluk element took control of the state at his death, they continued to recruit slaves from the Black Sea area, many of whom are described as Qipchaqs, and others as Tatars. However, there is a difficulty here, since the children of Muslims are not considered lawful candidates for slavery, and the `ulama of Egypt and Syria would have been unlikely to tolerate the enslaving of Muslim children. The same difficulty would operate with the children of the Golden Horde Mongols, the Tatars.

Mongol women and children came into Mamluk hands in the wars with the Il-Khan Mongols across the Euphrates river. There were no Mamluk wars in the Volga steppes or the Caucasus. On the contrary, the Mamluks had an alliance with the Golden Horde Mongols, who were Muslims (apparently acknowledging the `Abbasi Caliph in Cairo), against the Il Khans. (JAW)

> It was also during this period that the Genoese merchants took an
> active role in this trade and sought to secure large numbers of Tatar
> children to be sold in the Crimea. This prompted Tuqtuq Khan of the
> Qipchaqs to raid the slave markets in the Crimea in an attempt to
> stifle the trade. The trade does not seem to have abated in the least
> (Zubdat, 9: 460). (SJ)

Is it not likely that this was because the Tatars were not selling their own childen, but non-Muslim children? We know that a bit later the nominally Christian Circassians sold their excess children into slavery as a practical form of birth control/cash crop, knowing well that they would have wider horizons in Egypt than in the Caucasus.

As for the Qipchaqs, it is said that many of them had blue eyes and blond or brown hair. It has been suggested that these children were really captive Slav children, captured in raids or raised in slave-stockades, who were then taught Turkish and sold as valuable export commodities.

I wish someone could track down this suggestion. It is attractive, but needs substantiating.

John Alden Williams The College of William & Mary

Hi everyone,

Sabri's thoughts were very interesting and I thank him very much for them. My question though was a bit more specific about the method of transporting those slaves (mamluks) through other party lines and in what quantity.

Moreover, who sold them to the slave traders in the first place? was there a family tradition, say of selling one of the boys to aquire some wealth, or were the boys stolen?

Sabri points out that themamluk trade florished much later than I thought! I was under the impression that Nur El-Din Mahmud Zangi had a whole corp. of them an that the Ayyubids also had the same. My understanding is that the Zangids as a whole and the Seljuqs were children of mamluks.

does anyone have more detailed info about the trade itself?

fawzan

One should be careful not to confuse the Turkic military elements during the Abbasid period (e.g., Seljuqs and some of their Atabegs but not all of them) with those who were purchased as slaves specifically to be recruited in the armies. Were all the Turkic troops in Samarra originally slaves or were they directly recruited via the Abbasid governors of Khurasan and Transoxiana into the armies of the caliph? My understanding is that Kurdish atabegs like the Zengids were never enslaved.

Sabri Jarrar

You will find very useful information on the trade itself in the chronicles of the Mamluk period. Of particular interest are the biographical dictionaries where a mamluk's life and career are expounded.In many of the entries, we are told where the mamluk was purchased, who sold him, the itenerary of his trip to Cairo, Damascus, etc., the merchant who bought him and took him to the sultan or amir. Some of them even discuss why they were sold at the first place and how. Some of them sold themselves to Mamluk amirs, as was the case with Qawsun (sorry don't have the exact source but see Maqrizi's Khitat under Qawsun's architectural patronage).

Sabri Jarrar

I went back and reread Sabri Jarrar's mailing, which I seem to have filed without looking at closely. I think there are a few misconceptions which can be cleared up by referring to mamluks as a _generic_ term (mamluk=owned, i.e., a slave, and specifically a military slave) rather than to Mamluks with a capital M, as a dynasty.

Fawzan wrote:

> Hi everyone,
>
> Sabri's thoughts were very interesting and I thank him very much for
> them. My question though was a bit more specific about the method of
> transporting those slaves (mamluks) through other party lines and in
> what quantity.

I don't recall particular sources off the top of my head, but there is information for the earlier periods (which I'll come to in a moment); I would check Ibn Fadlan, for example, who mentions major centers of the trade in Transoxania; possible Abu Dulaf; possibly also the Hudud al-`Alam (all these are available in translation, btw).

> Moreover, who sold them to the slave traders in the first place? was
> there a family tradition, say of selling one of the boys to aquire some
> wealth, or were the boys stolen?

It appears that there was a family -- or rather a tribal -- tradition. Remember that not only money is involved here, but also a certain amount of prestige: military slaves could and did rise to very high positions. You may recall that the Ghaznavid line was founded by a mamluk commander.

> > Sabri points out that the mamluk trade florished much later than I
> thought! I was under the impression that Nur El-Din Mahmud Zangi had a
> whole corp. of them an that the Ayyubids also had the same. My
> understanding is that the Zangids as a whole and the Seljuqs were
> children of mamluks.

Sabri is a bit off-target here. Mamluks (generically speaking) were imported into the Abbasid empire at least from the early 9th century. It was because of the trouble they caused in Baghdad that the caliph al-Mu`tasim moved the capital to Samarra, it was a mamluk who murdered the caliph al-Mutawakkil, and during the Samarran period they were pretty much caliph-makers, as well as important patrons of building, literature and so on. They came primarily from Central Asia, by way of Transoxania. The trade to Baghdad from the east flourished for several centuries before it pretty much dried up when the Samanids stopped selling mamluks to their rivals the Buyids.

The Zangids were indeed ex-mamluks, as were the Ayyubids. The Saljuqs were not; they were steppe Turks who had never been part of the mamluk system. They had their own mamluks, however, many of whom held important positions in the military, and it was mamluks and ex- mamluks who were in large part responsible for the economic and administrative chaos which prevailed in the Saljuqs' Iraqi territories before and, especially, after the death in battle of the last of the Great Saljuq sultans in 1194.

> > does anyone have more detailed info about the trade itself? >

Not offhand; but I'll keep my eyes peeled, as some of the texts I'm dealing with should have information of this sort. You might look at Nizam al-Mulk's _Book of Government_, Chapter XXVII (On organizing the work of slaves, etc.), in which he describes the training of palace mamluks under the Samanids. There must also be a chapter on slaves in the Qabus-nama (I can't recall the title of Levy's translation; sorry). Also Muqaddasi's _Ahsan al-Taqasim_ (which has been, with some startling inaccuracies, translated into English). Probably most of the geographers have something about the major centers of the trade.

> fawzan

Best wishes,
Julie Meisami
Oxford

JAW's reply prompts me to add a postscript to my earlier reply to Fawzan:

JAW makes the important point that enslaving fellow-Muslims is unlawful. This applies as well to the trade in the east: the Turkic and other mamluks who were sold in this trade were not Muslims, though their upbringing presumably included conversion to, and education in the basic principles of, Islam. (This was presumably one of the reasons behind the translation into Persian of Tabari's History and his Tafsir in the late Samanid period: to provide ideologically correct texts for such education.) It should be noted also that the primary language of communication, both in the east and in Baghdad/Samarra, with these mamluks was Persian (there is evidence for this in Tabari). There were also at this time Slavic, Byzantine, and, if I'm not mistaken, Viking mamluks.

Julie Meisami
Oxford

Sabri Jarrar wrote:

> Date: Wed, 5 Aug 1998 17:27:26 +0100 (BST)
> From: Sabri Jarrar <sabri.jarrar@wolfson.oxford.ac.uk>

> You will find very useful information on the trade itself in the
> chronicles of the Mamluk period. Of particular interest are the
> biographical dictionaries where a mamluk's life and career are
> expounded.In many of the entries, we are told where the mamluk was
> purchased, who sold him, the itenerary of his trip to Cairo, Damascus,
> etc., the merchant who bought him and took him to the sultan or amir. Some
> of them even discuss why they were sold at the first place and how. Some
> of them sold themselves to Mamluk amirs, as was the case with Qawsun
> (sorry don't have the exact source but see Maqrizi's Khitat under Qawsun's
> architectural patronage).

This is true. If Fawzan Barrage has trouble locating al-Khitat, or reading its peculiar Mamluk Arabic, I have translated a good part of the account of Qawsun's career in the notes to an article on Urbanization and Monument Construction in Medieval Cairo, in Muqarnas Journal (Yale,1984).

This is quite a peculiar case. Qawsun was from the Golden Horde, free, and doubtless Muslim, in Cairo as a petty merchant. Sultan Nasir b. Qalawun who had a fondness for handsome youths met him, was taken with him, and persuaded him to sell himself to him as a Mamluk, thus making him eligible for a career among the elite of his empire. Such an act would have no legality whatever in Islamic law so far as I know. Apparently the `ulama of the period followed their usual practice of looking the other way when the ruler's private affairs were concerned. Otherwise it could only have been justified by Qawsun's attesting that he was not in fact a Muslim, and hence eligible to become a slave.

John Alden Williams
The College of William & Mary

Dear Julie,

My response to Fawzan's question specifically deals with the Mamluk dynasty. My reference to the efflorescence of the slave trade during the early Bahri period was not to imply that there have been no previous flourishings of such trade in the history of Islam. However, it would be hard to substantiate that the majority of the non-Arabic elements in the Abbasid armies were military slaves, or at least recruited as such.

Sabri

Unless I've missed something, it doesn't seem to me that anyone has mentioned the fundamental works of David Ayalon: e.g., <<Studies on the Mamluks of Egypt (1250-1517)>> (London : Variorum Reprints, 1977), <<Islam and the abode of war : military slaves and Islamic adversaries>> (Aldershot, Great Britain ; Brookfield, Vt., USA : Variorum, 1994), etc. Furthermore, I have vague memories (don't have my library here) of a risala or two, precisely on the slave trade, edited and published (in Arabic) in <<Nawadir al-Makhtutat>> bi-tahqiq Abd al-Salam Harun (2 vols).

It's all way past my time (so to speak), but I hope there's something to use there regarding the mechanics, logistics, and human interest aspects of mamlukiyya.

Department of Near Eastern Languages and Cultures
The Ohio State University
203 B & Z Building / 1735 Neil Avenue
Columbus OH USA 43210
Office: 614-292-9255 (Department) /614-292-4399 (Private)
FAX: 614-292-1262 / Home: 614-258-8866

Sabri Jarrar wrote:

> Date: Wed, 5 Aug 1998 17:18:01 +0100 (BST)
> From: Sabri Jarrar <sabri.jarrar@wolfson.oxford.ac.uk>
>
> One should be careful not to confuse the Turkic military elements during
> the Abbasid period (e.g., Seljuqs and some of their Atabegs but not all of
> them) with those who were purchased as slaves specifically to be recruited
> in the armies. Were all the Turkic troops in Samarra originally slaves or
> were they directly recruited via the Abbasid governors of Khurasan and
> Transoxiana into the armies of the caliph? My understanding is that
> Kurdish atabegs like the Zengids were never enslaved.
>
> Sabri Jarrar

Quite right. My understanding is that the Turkic troops in Samarra were, originally, purchased slaves, though some may later have been freed. Atabegs were probably not slaves. I'm not sure about (all of) the Zangids, and would have to check.

Julie Meisami

Dear Julie,

While Ibn Wasil (Mufarrij al-Kurub) describes Qasim al-Dawla Aq Sunqur, the founder of the Zengid line, as a mamluk of Sultan Alp Arsalan, both Ibn al-Athir (al-Tarikh al-Bahir) and Abu Shama (al-Rawdatayn) maintain that he was a crony of Malikshah, and that they grew up in the same house as chums. Ibn Wasil also discusses this strong bond when he says that Aq SUnqur was raised with the children of Alp Arsalan and his companions. Sulayman al-Saigh (Tarikh al-Musil, 1: 165) mentioned that Aq Sunqur had a well-known father by the name Abdullah, which denotes a Muslim pedigree.

Sabri Jarrar

> Dear Julie,
>
> My response to Fawzan's question specifically deals with the Mamluk
> dynasty. My reference to the efflorescence of the slave trade during the
> early Bahri period was not to imply that there have been no previous
> flourishings of such trade in the history of Islam. However, it would be
> hard to substantiate that the majority of the non-Arabic elements in the
> Abbasid armies were military slaves, or at least recruited as such.
>
> Sabri

Quite right, I think; there are clearly others more knowledgeable on this subject than I, but they seem to be on vacation at the moment...

It was not, specifically, of the Abbasid _armies_ that I was thinking. The mamluks served primarily as caliphal troops, and specifically as the caliphal bodyguard (though this doesn't mean they didn't participate in campaigns further afield). The Samarra site contains, among other things, what appears to be an extensive barracks-housing complex for these mamluks (see the forthcoming proceedings of the Samarra Workshop held in Oxford a few years ago; and don't ask me _when_ it's forthcoming!). The Abbasid armies, as such, were a very mixed bag; the Turkish mamluks were, in some sense, an elite (along with the Khurasanis, with whom they were in constant dispute, it seems, especially when it came to making and breaking caliphs). But in the greater scheme of things (political power-plays etc.) the mutattawi`a etc. don't really count.

Julie Meisami, Oxford

Dear Fawzan,

I completely agree with Michael Zwettler and John Alden Williams that the best starting point for a research on slave soldiers are David Ayalon's works. In addition to those mentioned by Zwettler, I would like to mention also his L'esclavage du mamelouk (Jerusalem 1951) and The Mamluks of the Seljuks: Islam's Military Might at the Crossroads (Journal of the Royal Asiatic Society, 1996). Of some interest will also be Speros Vryonis's Seljuk Gulams and Ottoman Devshirmes (Der Islam, 1965), Osman Ismail's Mu'tasim and the Turks (Bulletin of the School of Oriental and African Studies, 1966), and Daniel Pipes's Slave soldiers and Islam (New Haven 1981).

As far the trade itself, the Black Sea, the main pool from which Egyptian Mamluks draw their recruits, was an Italian lake until the fall of Costantinople to the Ottomans in 1453, and the Genoese kept some settlements there even after that date. Thus, studies on the Republics of Genoa and Venice are likely to yield some information on the trade. A great American scholar of Venice was Frederick C. Lane, and a hefty study of the Republic of Genoa was published one or two years ago in the US (unfortunately I don't remember the name of the author).

It is usually held that the Circassians and the other peoples living on the Eastern coast of the Black Sea sold their children to the slave-dealers. The Circassian, Abkhaz, Abaza and Ubykh tribes were divided into clans whose members were held to be relatives of the head of the clan (much as it happened in the Scottish clans), so it is likely that most of the times it was Circassian chieftains and noblemen who actually sold the children of their lower-class relatives or of their slaves, and not theirs own.

Someone wrote that Egyptian and Syrian 'ulama would have been likely to object to the practice of enslaving fellow Muslims, had anyone tried to do that. I just would like to remind that the Ottoman practice of the devshirme too didn't abide by the Islamic law, but as far as I know no Ottoman jurist objected to the enslaving of peaceful Christian subjects.

Giorgio Rota

To whom it may concern:

Re: a father named Abdallah

The fact that a Turkish soldier or officer has a father named `Abdallah is a pretty good indication that he is indeed a mamluk. This is indeed a nice way of saying that fulan's dad was a pagan. There are literally 100s of examples of mamluks whose names were followed by b. `Abdallah. No Muslim pedigree is implied here. Thus, its combination with Aqsunqur only goes to strengthen the suggestion that he was himself originally a mamluk. Also it was not uncommon for selected mamluks to be brought up together with princes.

Finally someone has mentioned the relevance of the late Ayalon's work for this discussion. I might add that his Esclavage is certainly worth examining in this context.

with best wishes,

Reuven Amitai
Dept. of Islamic and Middle Eastern Studies
Hebrew University of Jerusalem

Giorgio Rota wrote,

>Someone wrote that Egyptian and Syrian 'ulama would have been likely to
>object
>to the practice of enslaving fellow Muslims, had anyone tried to do that.
>I just would like to remind that the Ottoman practice of the devshirme too
>didn't abide by the Islamic law, but as far as I know no Ottoman juristy
>objected to the enslaving of peaceful Christian subjects.

It was I who wrote that. And yes, the devshirme always appears to be an anomaly. The Ottomans certainly did have mamluks--e.g. Circassian mamluks, up until at least the late 19th century, apparently, and earlier than that slaveboys taken in warfare, which was always lawful.

But here is the interesting question, and I wish the Ottomanists would get involved here. I have been assured that devshirme, or boy-levy draftees, were never slaves juridically, and are never referred to as slaves (mamluk), in Ottoman dcuments. So Niyazi Berkes, who was interested in this question, as well as the late Cengiz Orhonlu, an expert on the Ottoman archives, informed me at any rate. While treated exactly as slaves would have been, they were in fact simply draftees from the Orthodox Christian peasant population, which in earlier times seems rarely to have objected. Such boys were often in a position to be of great benefit to their Christian families, like Sokullu Mehmet, vazir of Sulayman and Selim II, who made his brother the first independent Serbian Orthodox patriarch. The word qapu-qul, by which they were designated, doesn't mean slave (again, I was told), and was reserved for the products of the devshirme. If they had been actually slaves, they would have been called quli (again, I was told).

Orientalists have called the quls slaves. The proof as always has to lie in the original documents. If in fact the quls were not juridically `abd or mamluk, then they were not slaves in fact.

But even assuming that they were, the fact that complaisant Ottoman muftis were willing to justify the practice in the case of the ruler's household may again only reflect an extension of the general rule that `ulama could be extraordinarily negligent of the rulers' private practices (e.g. wine-bibbing and sodomy), forbidden though these might be, rather than jeopardize the precious partnership they enjoyed with him.

The principle still holds true, that Muslims were generally not considered candidates for slavery, whether or not a part of the dhimmi population (which never included Romanians, Armenians, Arab or Syriac Christians in the Ottoman state) was.

John Alden Williams
The College of William & Mary

Dear Colleagues,

I have been following the recent discussion of the mamluk institution with great interest. The publishing industry permitting, I hope to have my study of the Samarran Turkish military/community available in the near future.

I wanted to add the following to the reading list:

Paul, Jurgen. The State and the Military: the Samanid Case
Papers on Inner Asia, no. 26
Indiana Univ, Research Inst. for Inner Asian Stds Bloomington, IN, 1994

/blocquote>

Paul discusses modern scholarship on the institution and offers valuable material on one case study of the use of slave soldiers.

Al-Suli, Muhammad ibn Yahya. Kitab al-awraq.
St Petersburg, 1998. Edited by Anas Khalidov.

Forgive the abbreviated citation - I don't read Russian. This is a critical edition of the St. Petersburg mss. cited, among other places, by the editors of al-Tabari's Ta'rikh [Leiden]. It contains the section of the Awraq from the caliphate of al-Wathiq (r. 227-232/842-847) to that of al-Muhtadi (r. 255-256/869-870), so, in effect, much of the Samarra period.

The edition contains the Arabic, a Russian translation, notes, indices of names, places and verse, and a short comment by Professor Khalidov. This is a welcome arrival to say the least.

It is being distributed, I believe, by Harrassowitz. I have only sampled the book and have already come upon several bits of evidence of great value to my own work.

Matthew Gordon
Dept of History
Miami University
Oxford, OH 45056
USA

Date: Thu, 13 Aug 1998 07:18:03 +0300
From: Amy Singer <asinger@post.tau.ac.il>

Just to start things off,

Why would the Ottomans called boys levied in the devshirme mamluk when they used Turkish? They were called kul. The form kuli/quli that JAW cites is the same word in Turkish with a possessive ending. mamluk was a term that continued to be used in Egypt.

On this see Jane Hathaway's new book. Also on the kuls see Metin Kunt, The Sultan's Servants

The boys were all called b. Abdallah cf. Reuven Amitai's comment some days ago.

Moreover, the assessment that Christian boys and their families either did or didn't benefit from their levy can never be categorized either completely negatively or positively. Certainly one can point to examples of those who made it and who left the material evidence (endowed structures) to indicate they retained ties (and seemingly affections) for their places of birth. However, there were hundreds taken as children. One has, I think, to understand that after some period of Ottoman rule in an area, that Christian families may have lived with the expectation that some of their boys would be removed. Some may have been able to make the most of it, others not.

Perhaps it was the case that some of these men were formally manumitted at a later stage of life. I had always learned that they were effectively slaves when drafted. You might, however, try posting your query on H-TURK.

Your last comment about Arab Christians not being included in the dhimmis under the Ottomans is unclear. Of course they were. But they were not devshirme candidates.

A. Singer
Department of Middle East & African History
Tel Aviv University
69978 Ramat Aviv, Israel
Tel. 972-3-640-6764 Fax 972-3-640-6934

Date: Sun, 16 Aug 1998 17:33:49 -0600
From: Warren Schultz <wschultz@wppost.depaul.edu>

Colleagues,

Sorry for this rather tardy addendum to the recent Mamluk discussion, but list members might find the following information useful. There is an extensive online Mamluk Bibliography available at the following URL:

www.lib.uchicago.edu/LibInfo/SourcesBySubject/MiddleEast/MamBib.html

This bibliography represents an on-going project to compile all research and discussion, scholarly and popular, germane to the Mamluk Sultanate of Egypt and Syria. The WWW version of the bibliography includes introductory material to the project and a Search Page that allows the user to search by author and title.

Warren Schultz
DePaul University
Chicago, IL