This volunte is dedicated to

Gherardo Gnoli

(r 937-20 I 2)

Already available tn The Idea of lran series

Birth of the Persian Entpire,Yol. IEdited by Vesta Sarkhosh Curtis (British Museum) and Sarah StewartLondon).ISBN; 978-1 -845 | | -062-5

The Age of the Parthians,Yol.2Edited by Vesta Sarkhosh Curtis (British Museum) and Sarah StewartLondon).ISBN: 978-l-845I I -406-0

The Sasanian Era,Yol.3Edited by Vesta Sarkhosh Curtis (British Museum) and Sarah StewartLondon).ISBN: 978-l-8451 I -690-3

The Rise of Islam, Vol. 4Edited by Vesta Sarkhosh Curtis (British Museum) and Sarah StewartLondon).ISBN: 978-l-845 I I -691-0

Early Islamic lrctn,Yol. 5

Edited by Edmund Herzig (University of Oxford) and Sarah StewartLondon).ISBN: 978-1-1 807 6-061 -2

(SOAS,

(soAS,

(soAS,

I

(soAS,

(soAS,

The Age of the Seljuqs

The ldea of lranVolume VI

Edited By

Edmund Herzig

(Faculty of Oriental Studies, University of Oxford)

and

Sarah Stewart

(Department of the Study of Religions, SOAS)

in association with The London Middle East Institute at SOASano

The Faculty of Oriental Studies, University of Oxford

Supported by the Soudavar Mernorial Foundation

Published in 2015 by I.B.Tauris & Co Ltd6 Salem Road, London W2 4BU175 Fifth Avenue, New YorkNY 10010

www.ibtauris-com

Distributed in the United States and Canada Exclusivelyby Palgrave Macmillan175 Fifth Avenue. New YorkNY 10010

Copyright O London Middle East Institute, 2015

The right of Edmund Heruig & Sarah Stewart to be identified as the editors

of this work has been asserted by them in accordance with the Copyright,

Designs and Patents Act 1988.

All rights reserved. Except for brief quotations in a review, this book, or any

part thereof, may not be reproduced, stored in or introduced into a retrieval

system, or transmitted, in any form or by any means, electronic, mechanical,

photocopying, recording or otherwise, without the prior written permission

of the publisher.

Every attempt has been made to gain permission for the use of the images inthis book. Any omissions will be rectified in future editions.

References to websites were correct at the time of writing.

The Idea of Iran Vol. 6

rsBN 978 | 7807694',7 9eISBN 978 0 85773 81 I 0

A full CIP record for this book is available from the British LibraryA full CIP record is available from the Library of Congress

Library of Congress Catalogue Card Number: available

Tlpeset by P. Fozooni

Printed and bound by CPI Group (UK) Ltd, Croydon, CRO 4YYfrom camera-ready copy edited and supplied by the editors

Contents

List of Illustrations

Acknowledgements

Foreword

Two Pattems of Acculturation to Islam:The Qarakhanids versus the Ghaznavids and SeljuqsIsndn Vdsdrv

Nizdm al-Mulk: A Maverick Yizier?Carole Hillenbrand

Ismaili-Seljuq Relations: Conflict and StalemateFarhad Daftary

What does the History of Isfahan tell us about Iranian Society during the 58Seljuq Period?David Durand-Guedy

Baghdad: A View from the Edge on the Seljuq EmpireVanessa Van Renterghem

Two Mirrors for Princes Fabricated at the Seljuq Court: 94Nizdm al-Mulk's Siyar al-mulik and al-Ghazall's Nasthat al-mulukAlexey Khismatulin

Stylistic Continuities in Classical Persian Poetry: 131Reflections on Manuchehri frorn Ddmehan and Amir Mo'ezziAsghar Seyed-Gohrab

Architecture and Politics: The North and South Dome Chambers of the 148Isfahan Jdmi'Robert Hillenbrand

Bibliography

vll

IX

29

4l

74

t74

6

Two Mirrors for Princes Fabricated at theSeljuq Court:

Nizdm al-Mulk's Siyar al-muluk and,al-Ghazah's Nasrhat al-muluk

Alexey Khismatulin(Institute of Oriental Manuscripts, St petersburg)

T vided into ons, dealing ith ther :'^:?",# ";':ii,?:i:, ;Tl

poet of the Seljuq dyrasty, who had the position and honorific title of ,Master

I. The Siyar al-mulukP rev i o Lts Tex t ual Sl u dies

There were at least five scholars involved i' studying, publishing ancttairslaii rig the .Sivar. a | - nt u I i lt in d i fferent countri es.

The text was published fbr rhe fir'st time in two parts - the per.sia' iext andrts annotated Fre'ch transration-by charles Scbefer, i' lggr and rg93respectively. In 1932, Lania al-RahTrn I(halkhalr acritrcal cdition of the text. Ni le llol.e edition by tl ntIiarrian scholar''Abbas Iqbal issued in the forni o s,

Two um.noRs FoR pRINCES FABRICATED AT THE sELruQ couRT 95

reader. Then, in 1949, the hrst Russian translation appeared, by BorisZakhoder. Eleven years later, Hubert Darke translated the text into English forthe first tirne, also preparing a critical edition of the Persian text. He then

published his second, revised translation. More recently, Iranian scholar Ja'farShi'ar published a critical edition of the text based on a different version.-

Today, one more critical edition of the text is being prepared for publication byanother lranian scholar Mahmld 'Abidi. To my knowledge, up to the present

time, the Siyar al-muluk has already been translated into at least I 1 languages,

and republished innumerable times in Iran, as the first political treatise writtenin Persian.

There are many different copies of the text, but two main versions can be

inferred from all of these copies. The first represents the texts descending ordirectly copied from the so-called Urumiya protocopy of 564 (1168). Thisprotocopy is stated to have been transcribed by order of Alp_Jamal al-Drn, who,

thanks to his position as AmTr Hajib (master chamberlain),' was a state figureof the highest rank (presumably, his name was IlqafshaVllfaqshat b. Qayrnaz

tjurl3 / jl-S ir,'..:'iiiJl /,'. ?'iirjl], master chambellain of the Seljuq sultan

Muhamrnad b. Mahmtd b. Muhammad b. Malikshah)a and could hardly have

ordered an unreliable copy. Until recently, these copies were used by scholars

as the basis of thi:ir work with the Siyar al-muluk. The oldest manuscript in this

group belonged to Charles Schefer, and had been transcribed in 690 (1291).

Preparing his first edition of the text, Hubert Darke was informed of the

existence of a manuscript in Muhammad Nakhjav6ni's collection held at the

Public Library of Tabriz. The manuscript had been copied in 673 (1274),whicnmakes it just 17 years older than Schefer's copy but a whole century younger

than the Urtmiya protocopy. This manuscript represents the second main

version of the Sryar al-mult1k. Darke used this single copy as the basis for the

revised edition ofhis translation. as did Shi'ar for his edition.

Both 'Abbas Iqbal and Zakhoder, on the one hand, and Darke and Shi'ar, on

the other, conducted serious textual research on the book; but they came to

diametrically opposed conclusions, differing on what are key characteristics ofany work - genre, authorship and date of writing. 'Abbas Iqbal and Zakhoder

concluded that some chapters had been completed and some had been added bythe first editor, a certain copyist named Muhammad MaghribT, after the murder

of Nizam al-Mulk, cornmitted by an Ismd'rh in 485 (1092). By contrast, Darke

and Shi'ar supposed that the whole text was written by Nizdm al-Mulk himselfbefore his murder, and that he was the author of almost the entire book- In his

revised edition, Darke even asserls that, with the new-found redaction, 'we

have a text closely descended frorn the author's authograph'.5

Having their own followers, both views still survive, supported by

additional argulnents tbr and against." They are reflected not only in every new

edition of the text but also in separate comtnents, reviews and research -though withor-rt ar-ry substantial progress in the final conclusions. Neither faction

96 THs ecE oF THE sELJUes

provides an answer to the centlal questions that would incontrovertibly provetheir statements: If the text was compiled by Nizarn al-Mulk, why are ilr.r" romany textual, historical, ideological and otlrer discrepancies? And if it was

arnmad Maghribi, who was he indeed, and why didcribe it to Nizam al-Mulk? And finally, how do itscorrelate with each other?

counterfeiting of the second part of the Nasrhat al-mulilk, ascribed(d. 505 [1 1 1 1]), that they deserve to be studied and compared in

Historical Context

At the end of 479 (1087), the first visit of Malikshah and Nizam al-Mulk to

spouses went into sLrch a decline that, upon receiving his daughter's complaint,Malikshah ordeled her to come back with her son to Isfahan; but she died soonafter arriving in the city. This chain of everts seems to have been one oi.therDarn reasons wliy, a little later, Malikshah nrade an ineversible decision tornove the Seljuq capital frorn Isfahan to Baghdad, arrd to turn al-Muqtadi out ot.Baghdad. Mo'eover, Turkan-Khafirn appears to have insistecl that Malikshahfbrce al-Muqtadr to recognize Abh'l-Fazi Ja'far as successor to the Abbasrdcalipl.rate, insteacl of one of al-MuqtadT's other sons. of course, these planswere in dilect cont.lict with those of Nizam al_Mull<.

Meanwhile. Tu'kan-Khaflui also gave birth to a son. The son was sive' therranre crl' Malrnrrrd arrd the honolific riile of Nasir :rl-Dunya va'l-bTn. He

Two unnoRs FoR PRINCES FABRICATED AT THE sEl..ll.'(.) ('( )l rl( |

became the youngest of Malikshah's sons who outlived their fatltcr lltc ollrt'ts

being Barkiyaruq, Muharnn.rad-Tapar and Sanjar. Accorcling lo rttctlit:r,:tl

historians, as an experienced politician and court intriguer (or pcrltal-rs, rttt tltc

cortrary, as a solicitous mother), Turkan-Khahln planned to tnakc MahnrLrtl thc

guaranteed sLlccessor to the empile of Malikshah while he was still a ol.rilcl. lior

this, a special type of document was required from the father. Nizarn al-Mulk,

who favoured Malikshah's elder son, Barkiydmq, as successor' opposed this

tum of events. Then Turkan-Khat[n brought over to her side one of her

favourites, Taj al-Mulk (Ab['l-Ghanayim Marzban Kiusraw al-Shrrazi) Bcing

responsible for the harem and tbr raising her young children, he began to hold

several important state positions, including the position of managing tlre private

property and treasure of Malikshah, and also that of head of the drvan ofwriting and records, or chancery. He also put his own protdg6 (Majd al-Mulk)

at the lread of the frvan for property and taxes, thus practically restricting

Nizam al-Mulk's power - except in areas related to the army, which was

completely devoted to the old vizier. Fearing an attack by the army, Taj al-

Mulk could not legally get rid of the vizier once and for all ir.r order to take his

place, but he and TurkAn-KhetUn did everl4hing to make MalikshAh take their

side, which is eventually what happened'

Judging/from the reports of historical chronicles, everyone was very well

aware of the conflict and rivalry between Taj al-Mulk, supported by Turkan-

I(hat[n, and Nizarn al-Mulk. In addition, Tej al-Mulk is stated by some

medieval scholars to have been an ally of one of the leading Isma'rlis in Iran,

Hasan b. al-Sabbah, and his agent in the govemment of Malikshah, thus having

lris own wide-ranging plans. 'Abd al-JalTl QazvTnT Raz1, a Shi'a author of the

sixth (twelfth) century, in his Ba'zt masalib al-navosib gives the following

characteristics of Taj al-Mulk:

Taj at-Mutk nicknanted Marzbdrt became the ferment of apostasy and

global sedition He was a favourite friend, sLtpporter and ally of

fHcrscrn-i] Sabbdh, and one o.f the. seven people who took an oath ofallegiance lo him at the beginning.'

-|our J JJt ) €t -Li ,Jt€- a;!9 >uJl at-vr.^- aStlr';y :jJ '(LJt 6u

,urf J-9r ,4 6 41 t).^i '- ;^ r-rT -,1t ,rt 3t-, [--.J

In the second halfof483 (1090), Hasan b. al-Sabbah captured the mountain

fortress of Alamht, and fortified it within a short tirne. Tlris seerns to have

served as a signal for the Isma'rhs' open uprising all over lran the followingyear (484), and laicl the foundatior.r for an independent state of the Isma'rirs to

arise in the fonn of fortress cities scattered around the mountainous regions ofthe country- In the Isrna'rlTs' own historiography, the capftrre of Alanrut is

considered the beginning of the Alarnht period, which contir.rued until the

Mongol invasion and the destruction of Isrna'rlT forlresses by the Mongols in

654 0256).

999B Tse ecg oF rHE sELJUes

The year 484 (1091) was marked by the second visit to Baghdad ofMalikshah, together with Nizam al-Mulk. The visit had no relation to the

Isma'rhs' uprisirig and the consolidation of the alliance between caliph and

sultan. Quite the contrary: Malikshah pointedly refused to acknowledge the

Caiiph al-MuqtadT, acting in a spirit of revenge. During the visit, he announced

that he would move the Seljuq capital from Isfahan to Baghdad, ordered Nizarnal-Mulk, Taj al-Mulk and other court dignitaries to start constructing Baghdadresidences for themselves And, finally, tried to push al-Muqtadr for a

recognition of Abl'l-Fazl Ja'far as the ]awful successor to the Abbasidcaliphate.

Thus, by Ramadan 485 (October 1092), when Malikshah intended to visitBaghdad for the third and last time in order to put his plans into effect, Nizamal-Mulk, supported by the anny, became the main obstacle both for Malikshehwith Turkan-Khdtrin and, of course, for Taj al-Mulk, who also acted in Hasan b.

al-Sabbah's interests. Therefore, the murder of Nizam al-Mulk during this tripseems to have been inevitable. The Caliph al-Muqtadr, as one of the fewremaining political allies of Nizam al-Mulk, could have done nothing to protecthim, though he appears to have done everything necessary to protect himselfupon receiving Malikshah's final order to leave Baghdad within ten days.8

Malikshah was poisoned and died in Shawwal 485 (November 1092), just amonth after the murder of Nizam al-Mulk.

Upon his death the court of Malikshah fell apart. As a result, all militaryoperations planned by Nizam al-Mulk against the Isma'rhs were stopped.

Instead, intemecine wars began in the Seljuq state. Divide et impera! Thissituation was skilful1y exploited by the Ismd'ihs to strengthen their position.The former officials of Malikshah found themselves at a loose end and ran tohis sons and relatives, who had lined up against each other.

The Master of Poets, Ml'tzn, did not avoid this fate either though he wasstripped of neither his posts nor his titles. And perhaps he chose the onlycorrect survival tactics under such circumstances, although they were notwithout hypocrisy, or perhaps political prudence: first he praised all thepretenders to the throne, and then was forced to apologize it the qastdas

addressed to the winners of internecine wars for the ways in which he hadpraised their enemies. Such is the unenviable lot ofcourt poets.

Around 498 (1105), or a little earlier, changing addressees and cities,Mu'izn chose to serve in the circle of Sanjar, who from 490 (\097) arrangedhis administrative centre in Marv. Evidently, unhappy either with Sanjar'streatment of him or his own low status at court (we can only guess the reasons),

he wrote a qaslda for him. ln the qastda, he complained about his situation,saying that he had served for 30 years and that this was worthy of much moreresDect.

Tw<-r vinRoas FoR IRINCES FABRTCATED ATTnE sELrrle couRI

Thirtv yecrrs ofservice, after all, deserves ct certain re,epect,

thirtl,year5 of respect in sen'ice is not to be neglectecl-e

,.r.;e tQ ;i t, du. .jr !-r.l+6--:. BQ g*Ji rJ aJL" ,*- .>rr

ln Jumada al-akhira 498 (March 1105), Sanjar's brother Ghiyas al-DTn

Muhammad Tapar b. Malikshah cane to the throne. He was officiallyrecognized by the next Abbasid caliph, al-Mustazhir bi-Llah (r. 487-512

[1094-118]), as the sultan of the entire Seljuq Empire 'the Lord of the

Universe' (Khudavand-i 'alam) following the definition given in the Siyar al-muluk - with a residence in Isfahan. He reyived military operations against the

Isma'ilis, and in 500 (1107) captured two of their fortresses near Isfahan. In one

of these - Shahdiz - the head of the Isma'rhs in Iran, namely Ahmad 'Attash,

whose formal deputy was Hasan b. al-Sabbdh, had hidden for 12 years. Thisevent, even in the Isma'Tlis' own historiography, is undoubtedly considered a

major defeat, after which the Isma'rhs lost their influence in the region ofIsfahan. The sultan executed 'Attash and his son, sending their decapitated

heads to the Baghdad caliph. According to statenents by historians, in this

fortress, which was one of the two main stlongholds of the Isma'rhs in Iran (the

second one was the fortress of Alamtt, headed by Hasan b al-Sabbah). a huge

amount of wealth was gathered by Ahmad 'Attash. It is possible that sornethinglike the treasury of the Isma'Ths was found in it.r0

After these last events, Mrt'izn tried to secure himself a status under the

new supreme ruler that would have corresponded to his forrner status at

Malikshah's court. In order to do this, he fabricated tbe Siyar al-muluk-astrongly anti-IsmA'lh and even generally anti-Shi'ite text, in which the wholehistorical situation described above was reflected by him, perhaps, much tnore

recognisably and obviously for his contemporaries than for us He also wrote a

qasrda of praise in honour of Muhammad b. Malikshah's victory over the

Isma'ilis in 500 and added it to the compiled text, ascribed by him to Nizam al-Mulk. In fact, he used a rather innovative method for his forged compilation: he

combined the genre of official juridical documents with the genre of ordinarynarrative prose. To see how he did it, we have once again to look at bothversions ofthe text.

Historical Criticism

Apart from the references to the Urftriya protocopy of564 (1168), the earliest

records of the existence of the Siyar al-mulilk in independent sources belong tothe begiruring of the seventh (thirteenth) cenhrry - that is, a whole century later'

than the date that the text itself indicates. The one direct reference to itsexistence is confirmed by the direct borrowings from it in the second part of the

Nasthat al-mulfik. However, this part cannot be considered an independent

source (see below). Therefore, those who have studied the Siyar al-ntuluk frame

100 Tus,qcp oF THE sELJUes

the history of its appearance relying solely on the contents of the text in itsvarious versions. In this respect the foleword is the key part oftl.re text, and thecrucial question is: Who wrote it?

ForewordAs a rule, any rnedieval foreword writter.r by its author in classical style, andusually following a khutba ('opening address') would set out answers to a fewquestions concenring the main text: what rnotivated the author to start his work,who the author was, what form his writing would take - either collection (7a2,)and compilation (ta'I\fl, or composition (tasnlJ),tt and how he would tackle thetask he had set for himself in writing his work. In other words, such a pattemwas traditionally used by the author to clarifz four main questions: who, what,why and how. In many cases, however, forewords deviated from this standardand complete pattem for various reasons. One such reason might be that theauthor of the work had not himself written a foreword, leaving the door openfor a later writer to insert one.

Even today, a foreword written by someone other than the author of themain text can indicate that, for some reason, the main text was not intended forpublication, or at least not during its author's lifetirne. In such cases, the maintext may have been assembled by an editor frorn the author's privatecorrespondence, memoirs, diaries or offrcial documents, and it is the editor whocomposes the foreword to the work. This was true for a number of well-knownclassical Islamic texts. For example, the A1t farzand (the Ayytrho al-walad) andthe original part of the Nas\hat al-mtiluk, both by al-Ghazalt, were taken bymedieval publishers from his private correspondence. Naturally, these textstherefore have no folewords by al-Ghazali hirnself- just as their titles also arenot his own, but were attributed to these works by those who placed them in thepublic domain.

The Urumiya Version

According to the foreword written by a scribe of the private books depository tothe Siyar al-mulik, in either 479 (1087) or 484 (1091), Malikshah orderedNjzam al-Mulk, Taj al-Mulk and other dignitaries to prepare a report on thecurrent state of the Seljuq Empire and to indicate significant negativedevelopments and violations of revered traditions. This formar order, firstresponding to the questions 'why' and'what', was repeated in all copies of thetext in this version.

Of all the reports submitted for the highest examination, only the reporl byNizam al-Mulk pleased the sultan and was approved by hirn. But only after theaccession of Muhammad b. Malikshah in 4gg (1105) did the scribe decide topresent the text to him'forthe sake of the books depository'. He thereby triedto ensure himself a job, renewing his service under the new supreme sultan.This is the sole reason why the text was presented by the scribe- Although, in

f wo pttRRons FoR pRINCES FABRTcATED AT rHE sEr-JUo cou{r-!- l0l

Schefer-'s edition, this reason does not seem so obvious, and the addressee ts

completely absent, it is cleal fi'om tl're last senteuce tltat sor-ne tin're had passed

since the moment of writing:

A scribe ol the depositon) w'ilings ... has writlen this kilab fttr the strlce

of sen,ice to the cleposito4t and presented it. I/'Allah so desire,s, il ntay

be approved. Anlt podishoh and an1, ruler has no other choice bul .tohave and lcnow this copv-book (daftar), especialllt at the presenl tinte.''

.ryl j1 9 +:! tcr-j; c-iJt u€ jl .l15.ll .-. dili .91+Ut-U

P) U-t;t-iu,tJ::I) jllt(s,tJ l4t-9 9,J^t.'!)! €t':..r,-ioJ{J.j-r- (4ill tiil16)S;,J-t f eV.*e o;Q

Darke's first edition, based on othet copies of the Siyar al-muluk, gives much

more information, indicating the reason and the addressee clearly:

At present, a slave who wishes to report on hi.s status to the Lord of lhe

Universe-Ghiyas al-Dttnya va'l-Drn Muhammad b. Malikshah, malt

Allah exalt him - and renew his service, has wri.tten this da{tar for lhe

sake of the magniJicent book depository, may Allah fill it, and offered ilin service. If AIIAb so desires, it may be approved and accepted. Anypadishah and any ruler has no other choice but lo have and imov, thiscoptt-book, especialQ at the present time.t3

Ull JLi - ftr ur(1r.r- .i4 e-y J|: af€^-t3it' oJ{ o3e Jt- yt ;r -9

jl lj -rr) 4-l c$l>f o;li &-f ..bt rltr r, ,dit o)ipl ,oL:..(1,;,t -r.n:*;,--rJg

o,q*,-g ,nIt elJ;l .r.,;9T r;.1 c.Jnr J c--li ,oi'rt lrr", ,t3n* tS sxly c>42

oth P) g1l ;:;!r9 dFjtlr jl lt "fbl $31.v 3 dti)! erpJ .)tF J* t rT

.-l€yS,J-\ f eV tt/4In the final part of his foreword, which follows the table of contents, the scribemade an attempt, hrstly, to outline the number of chapters and their length, and

secondly, to answer the questions: How did a state document addressed by the

grand vizier to the sultan end up in the hands of a scribe, and why did nobodyknow about this previously? Since this part is repeated in bothbasic versions, itdeserves to be scrutinized separately.

Traditionally, scribes mention themselves only modestly in colophons ofmanuscripts, if at all. But the point here is not just the obtrusiveness of the

copyist and his intention to convince the addressee that he is carying out

almost the last wish of the grand vizier. In the Urumiya version, there is an

additional afterword in which all the explanations given by the scribe in his

foreword are reoeated on behalf of Nizam al-Mulk,

r02 THp ecE oF THE sElrues

The Nakhjavant Version

At frrst glance, Nakhjavani's copy, deprived ofthe afterword, appears to have a

more authentic foreword, stating at the very beginning that both tl.re forewordand the whole text, consisting of 50 chapters, were compiled by Nizarn al-Mulkhimself following Malikshah's order of 4'79 (108'7). However, the forewordwas combined with the Urumiya afterword and revised there in a way thatmisled some scholars, who prefer to consider the copy as more reliable, and todisregard the Ururniya version. For several reasons the NakhjavanT versronlooks less logical than that of the Urumiya:

a) Malikshah's order seems to have been fulhlled by no other dignitarythan Nizam al-Mulk.

b) The whole text was written by Nizam al-Mulk 'for the sake of the bookdepository' (as if the grand vizier were an ordinary scribe):

This slave [Nizam al-Mulk] ... has written this copy (nuskhat) forthe sake of the magnificent book depositol, may AllAh fill it, andoffered it in service If AllAh so desires, it may be approved and

,t4acceDled-

.r.,13T C-..U 42 r*; ,aitt tar.-c .r's e{ljr&) jl c.r+*i dll ... oJ:l

.:tF J*s J{T o{J:*t 6ritt ou;t

c) The sentence cited suggests that no part of the text had been seen andapproved by Maliksheh before, and that the whole book, consisting of50 chapters, had been submitted for the highest consideration for thefirst time.

d) Moreover, we encounter again the same final part of the Urumiyaforeword, located in the same place and provided with the samestatements. These are offered by an unnamed 'slave' to whom Nizamal-Mulk handed over the text - without any formal reason at all beforehis trip to Baghdad and his murder.

The Final Part of the Foreword

According to the statements given in the final part of the foreword in bothversions, Nizam al-Mulk had withdrawn his report, consisting of 39 shortchapters (fast) and already approved by Malikshah, and decided to add to it.After writing 11 more chapters and adding to those already written, the revisedreport acquired the form we recognize today (or close to it) - that is, the formof a proper book of 50 chapters. Nizam al-Mulk then presented the text not tohis addressee, but to the scribe, or slave, whom he named plainly in theUrumiya afterword as Muhammad Maghribr, or Muhammad the scribe ofMalikshah's private depository.'' Notwithstanding the fact that Malikshah wasaccompanied by Nizam al-Mulk on his last trip to Baghdad, the text did notreach him, because he was poisoned and died just a month after the murder of

Two wllRRoRs FoR pRTNCTES FABRICATED AT THE sELJUe couRT 103

Nizam al-Mulk. That is why the scribe, or slave, having done his work (eitherby copying or simply saving the text), careflrlly preserved it until the accesslonof anew Lord of the Universe (that is, anew supreme Seljuq ruler) i.e. for at

Ieast 13 years.

Shi'ar's statement that it is Malikshah whorr the unnamed siave addresses

as 'the Lord of the Universe''u in the Nakhjavam version does not hold water.If this had been so, the slave should have handed the text over to Malikshahwithin the month between the murder of Nizam al-Mulk and the death ofMalikshah. However, according to the slave, the political situation was

deteriorating sharply after the murder of Nizam al-Mulk, and did not allow himto present the text until 'justice and Islam have strengthened due to the Lord ofthe Universe'.

This parl ofthe foreword is the crucial one for finding the counterfeiter out.Undoubtedly, it was not written by Nizam al-Mulk in eitirer version. But itanswers the questions of how and by whom the text was compiled to reach itsnew addressee. Had this parl been absent, we could have taken the answers tothese questions from the afterword compiled on behalf of Nizam al-Muik in theUrumiya version. But in the Naldrjavam version, the same final statements,

wriften by the unnamed slave, form the only parl that clarifies the situatron.Without this part, we would never have known all these facts - just as we are

ignorant ofthe equivalent facts with respect to many other authors and books. Itwas common practice for an author, when he had composed a manuscript, togive it to his addressee or to a copyist and it was thus absolutely unnecessaryto describe such a commonplace procedure in the foreword to the manuscript.

This puzzle can be explained by the location of the final part after the tableof contents. It is impossible to see the substantial difference in size between thechapters ofthe text from the table of contents, which contains just the headingsof the 50 chapters without reference to their length. The difference becomes

clear only upon skimming over the main text. Upon examination, it is clear that

some of the chapters could onlybe referred to as 'shor1' (mukhtasar). Some ofthern consist literally of only hve to seven sentences, while others spread overfive to seven pages. In the Urumiya afterword, the phrase 'quiteshor?abbreviated' (bas mukhtasar), focusing on the same short chapters, rs

once again located in the right place, explaining their difference in length to the

addressee, who is supposed to have read the whole text.In short, all such clarihcations in the hnai part of the foreword show that the

scribe, or slave, had certainly read the whole text and knew it very well, and

that the foreword had been compiled by hini after completion of the book. Keen

to convince Muhammad b. Malikshah that the book had been written by Nizamal-Mulk, he is likely to have been afraid of being suspected of its fabrication, orat least to have felt uncomfortable not only because ofthe difference in length

between sorne chapters, but also because of the much more importantdifference in style among them. In other words, he was trying to explain the

t04 THE ,tct oF THE sEl.rues

lattel by drawing the addressee's attention to the former a t.nerely polernicaland literary device. This appears to be the only reason why he eurphasizes the

number'39 to lris addressee. Uponleading the headings in the table of contents,

this figure rnight have ren.rinded Muhammad b. Malikshah of a well-lsrown listof duties composed by the grand vizier, thus explaining the different style ofthe short chaptets. It was these short chaptels that wet'e apploved by Malikshah,after which the text became subjcct to change and addition, allegedly by Nizamal-Mulk, for whom the literary activity seems to have been extra-curricular, and

whose authorship needed to be insistently asserted by the scribe.

His insistence is clearly seen in the Un-rmiya version, where he asserted

Nizam al-Mulk's authorship tluee titnes: in the foreword on behalf of himself,in the afterword on behalf of Nizam al-Mulk, and in the qastdo, again on behalfofhimself. Such insistence gave a double effect. In the Nakhjavani version, the

foreword was rewritten on behalf of Nizam al-Mulk, the 'scribe' remained an

unidentified 'slave' in the final part of it, and.the afterword containing the

scribe's narre and position was entirely withdrawn from the text. Nor do we

find there the qastda in which the scribe's intention to submit the text in person

and to enswe himself a job at Mtrhammad b. Malikshah's court becomes

explicit. Therefore, the Nakhjavanl version can doubtless be considered

secondary to the Urumiya version, or as a revised edition of it intentionallymade by a medieval editor. He might conceivably have understood very clearlythe scribe's insistence on asserling Nizam al-Mulk's autl.rorship, ar.rd rnade the

necessary changes to the UrDmiya version. In my opinion, this is how the twoversions correlate with each other.

Hence, to understand why the scribe was worried about the differences inlength and style between various chapters, we need only attend to the forewordof the primary Urumiya version, noting a number of its details that are clearlydiscernible, but that have slipped out ofresearchers' field ofvision. The first ofthese concerns the form of the lesponse to Malikshdh's order.

The Form of the Reply

Malikshah's order does not even irnply that anyone should have submitted a

book in its modern sense to the sultan for approval. If it had, Malikshah wouldprobably have received a whole series ofbooks ofsimilar contents from severaldignitaries. His remark on Nizam al-Mulk's reply is quite cornplirnentary:

These chaptet's have been written exactly as I desired; there is nothingto add to them; I will mahe this book my guide and follow' its precepts.

l;lS.ft 9r .e-^.i ,St-| JtlJ.r-Jl-5i ,* Jr afri,t aJi trLz Qlait r^-o yt.c-et a^lP Ut J ) ?tf jr-f ?vl

(Translation: Hubert Darke I 7)

Two tr,ttnRoas Fot{ ptuNc'ES FABRTCATED A I TiJE. snL.rue c.(luR i i0_5

Dar-ke's first edition adds the followi'g phrase afte'Malikshal.r's w.rds:

Thet'eafter lte olyvctts t+,ent bt, this book, and ga,r,e his conntands crncly,role his n'ea6, ('cthd-ndnta) ac'c'ording to these chapters.ts

.:..!.,1 ,r^E t:I y C:,9: aaUtp j )t) t+t J9 f.g'L4! i/r gJAl sriThe small semantic shift rnade in the translation and the understanding of thewords fasl and, kitab led the first word to be understood as 'chapter' ancl thesecond as 'book', which is generally correct - but r.rot for the court langLrage oftbat period.

The tradition of Muslim scholars, when they come to the Majlis of theMalik of Islant, is to expottnd a decision consisting of fottr tl.tings: oproyer, contplintent, counsel and eliminalion ofneeds.

S-.:. .t t<;,aS 6c;I c-u4,r)Ut .<t";rye rl3a ,r)Lr c!& crrLe

..2V i;: tLe\+dt l;4 Vs :p lg.stSanjar's approval, cited by a rnedieval compiler of al-Ghazali's letters, reads:

Now we have a requesl for you lo w,rite down this decision, which cameffi'ont your lipsJ, with your own handwriting so that it can be read to us

attd we catt send copies of it evet.vwhere .-.'e

V 3 Jil3> d U J u ,f'-t\ ,-r._p !n ,vi1 6 ,lai O,t r5c-iT .r"ta:Jt r,Jgit

---";-..* jt bU uT a;-;l' other words, what was written by Nizam al-Mulk - and, judging by thestatement of the scribe in the urumiya version, by a number of other statefigules - was not hing more laconrc,which could be e texts submitted byother state figures s ofwhich farntdn_decrees could be be arrangld morecorrectly. And, of course, originally it carried no title - neither siyar al-mulfiknor Sivdsqt-ndnta.

sorne of the rs of the Sittar al_out that th thematically andheadings of ttvdza,at) _ that is,

an official , consisting ofa list of o shest rank ofofficiais w llf. *ritt"n ,n

106 THs ncE oF TFrE sELJUQS

the forrn of a diplorna ('ahd-nana), giving the sultan's written approval of theduties of the officer in question. Once the list of duties had received loyalapproval, it was not subject to any fuilher changes - least of all at the wish ofthe officer who had entered the sultan's service.

The same practice is shown in Muntajab al-Drn Juvayru's 'Aiabat al-katabctto have continued during the Seljuq peliod.t" The 'Atabat presents only theorders and decrees issued on beiralf of the sultan in order botl.r to approve theemployees' duties and to address whomever the duties might concern.Forhrnately, we have an excerpt taken from a contract of employment andwritten on behaif of an employee who had applied for the position of 'house

manager' (kadkhuda) of the ernir's son The excerpt is cited by Ab['l-FazlBayhaqT in hrs Tar1kh-i Bayhaqt. Notwithstanding their limited number, theheadings of the contract articles namedfasl and listed in the excerpt look verysimilar to the individnal headinss rnlhe Sivar al-multlh.

The Tartlch-i Boyhaqt The Siyar al-muh.1k

Concerning the palace pages (ghulams)and their master - a full article (/asf

(lti .,L4, - dLar rlLJ .j\- J\->,i 6t - ), J)

l2th fasl. On sending pages(ghulants) from the coufi onimporlant business(eL"-r- r, o5r Jt orV-;i ;).; ,lt)

Concerning the senior chamberlain and

other military leaders - an article(.,Lae -rCJ ovt;- ,1t1 Sts .=-w ,s- ls)

On going on a march, dismounting andgathering information about enemies -an article($Iat - .rL# ;1.;t

":-, drl )iJ Jg .:lr yry)

22nd fasl. On keeping fodderready at places of dismounting(1.\t. Jr -:'b JL;l) .>L Jajl)

On the salary for the army, theappointment and resignation of thedeputy of the military [rvan - an article2l

-sby o\: +!u .tLa-l1 ou! f-:J r,uC*.7urrr;/ l^t

23rd fasl. On clarity in theproperly status of all the army

1-3 rL+ Jyt ;-:t, u3a :.JJt)

The range of responsibilities of Nizam al-Mulk as the 'top manager' of theentire state was naturally much wider than that of the 'house manager' of theemir's son, and therefore consisted of 39 articles. As we have seen, these 39articles were approved by Malikshah, who issued decrees which were based onthem and which were confirmed by royal diploma. Like the sultan's orders anddecrees collected inthe 'Atabat, Nizam al-Mulk's contract of employment withMalikshah undoubtedly followed a set official pattern shared by similar

Two vrnRoRs FoR pRJNCES FABRICATED A-T TFIE sElruo c'ouRT 101

juridical documents. However, by no stretch of the imagination could it be

described as a 'book', 'litelary con.rposition', and so on. For ar.r experienced and

skilled secretary, dlawing up a contract of ernployment according to tl'ris patter:n

did not take much time. For example, the contract of employrnent of the enrir''s

son is shown by Bayhaqi to have taken no more than half a clay to wlite.Approximately the same span of tirre is implied by the sclibe both in the finalparl of the forewotd and in the afterword to the Sryar al-nrrluk, when he states

that 39 articles were written by Nizam al-Mulk 'extempore' (bar badlha).If we examine even just the three headings cited above from the Siyar al-

multlk, we discover to our surprise that they refer to the chapters that ale 'quiteshort' (though the twenty-third chapter has a few additions made by the sclibe),and we can easily detect their independent and formal style of wlitlng. Eachfirst describes a problem in governing, and then suggests how the problemshould be solved by the person in charge - namely, Nizam al-Mulk. Snggestingsolutions, these chapters all follow the same grammatical pattem, usingirnpersonal modal verbs: 'it should be that ... or it must be that.. .', 'it should ormust be done', 'it should or must be known', and so on- This patterr, thoughconverted into the passive voice, can also be seen clearly in H. Darke'stranslation:

Cltapter XII. On sending pages from the court upon important businessPages are frequently sent out from the court, some at the king's behest,

mostly not. They are apt to cause trouble to the people and extorl money

from them. [Sttpposing there is] a case inyolving a sum of two hundreddinars, a page goes out and takes five hundred as a perquisite; thiscauses extreme embaruassment and poverty to the people.

Pages should nol be sent unless there is an urgent matter, and iftheyare sent it should be only at the Sublinte Command; and they must begiven to understand the exact antounl due, and they are not to toke anyntore than this by v,ay of perqtrisite. Then everything will be in ordet':2

This passage presents one of the duties described in the genuine contract ofemployment of Nizam al-Mulk without any later additions made by the scribe.It is significant that the same grammatical pattern also exists in the contract ofemployment that was partially incorporated into the Tarrkh-i Bayhaql.

Malikshah's written apploval of Nizam al-Muik's contract of eriploymer.rtshould also have followed approximately the same fonnal style as tliatpresented by the Ghaznavid emir inthe Tdrtkh-i BayhaqT:

Then [the answers] in this fornt were stated. I copied [thent] mtder lhe

fust1l [pl. of fasl] of the contt"act of entployment (nuvaza'at), cmd the

amlr put a seal on them, and under it he wrote in his own handwriting:'The worthy Sir, ntay Allalt prolong His assistance to him, should relyon these answers which are writtetx by my order and approved with the

108 THe nce oF THE sElrues

seal, and show his abiliD, and honesh) lo counsel on each oJ'these ilemsin order to gain praise ond tnrst in himself, if Atlah so desires'.21

-hrq Oi JJ J tf ey;pl I cO*.,:;; ar-btJA &ar n S .crF jj oT-n.r-:

J ,t=*d l, jt"J4 64tf j# ,6JdL- ,i,r 3r:r ,.j*'aE azt9-" :aS:;;r. j;r-eJ1t*t ./lJll d/jl dq -P ;t ,-ts-y t-vWy c--1.ol-9.r;5rtce t ,U.3 53" Eg.

."ri'rt ,U::.it ,rrJrb:ct 9 )Let r-a.1ru ti

Thus, the short /asl-chapters in the Siyar al-ntultlk closely corlespond to thecontract /asl-articles cited in the TarTkh-i Bayhaqr in terms of their subject,form, brevity, grammatical pattern, the short span of time necessary to compilethem, and the sultan's response to them. Therefore, they certainly represent thepatteru of writing of contract documents. The pattern can also been seen clearlyin other shorl chapters of the ,Sr/dr al-muluk. Moleover, in these short chaptersthere are no historical examples or anecdotes, just as there is no indication ofthe sultan's duties, let alone criticisrn ofhis actions. AII ofthese appear, in oneform or another, in the long chapters. Following the pattern in these longchapters, we can separate the genuine parts of the contract of employment ofNizam al-Mulk from the later changes and additions made by the scribe. In myopinion, this separation should be understood as the main principle governingthe next editions of the Siyar al-muluk.

Working as a scribe in Malikshah's private depository, Maghribl had access

to state documents. Having at his disposal the contract of ernployment ofNizam al-Mulk, each of whose articles was dedicated to one of the numerousduties of the vizier, Maghdbr decided, so to speak, to enliven its impersonalstyle with examples and stories taken by him fiom various types of historicalliterature. This explains all the discrepancies that have been discovered in theSi1,ar al-muluk and mentioned by modern scholars.2a Having plenty of time athis disposal during internecine wars, MaghribT also decided to write at least I Imore chapters and add them to tle book. Naturally, he could not find historical'decorations' for all of the vizier's duties. Some chapters thus retained theiroriginal pattem, and remained 'quite short'.

Since the Siyar al-mulut originally represented the contract of employrnentof Nizam al-Mulk, Maghribr had to provide it with his own foreword. In theforeword, he had to answer four traditional questions: who, what, why andhow. To this end, he had to remind his addressee of a rather ordinary situationwhen Malikshah called a state council or arranged a reception. The onlypulpose of such a reminder was to assert Malikshah's preference for Nizam al-Mulk's point of view over the views of others. As is shown by the historicalcontext described above, neither the year 479 nor 484, assigned by Maghribl tothis event, correlates with the script proposed by him. In 479 there seems tohave been no obvious reason for Malikshah to concern himsclf with theproblem of improving his own government and to ask for various opinions andadvice. Everything was going quite well. The year 484 does not correspond

Two rutnnoes FoR pRINCES FABRIC'ATED AT THE sELiLro couRT 149

with tl.re preference given by hirn to Nizam al-Mulk over Taj al-Mulk sr-rpporteci

by Turkan-Khat[n. In other words, MaghribT could have assigned any otheldate to this event in order to explain the gap between Nizam al-Mulk'switl'rdrawal of the initial short version of the text and the submission of its final'improved' version. The need to disguise his own volurninons fabrication, ardthe combination of two contrasting pattems within the work, required a longerinterval to have elapsed between the two versions.

Worried about the clear difference in length, and especially in style.

between the text chapters, Maghribr attempted to prove Nizam al-Mulk'sauthorship by adding the afterword, with his statements repeated this tirne or.r

behalf of Nizam al-Mulk. Not satisfied with the compiled text, he decided also

to adorrr it with a postscript qasrda in praise of the Sultan Muhammad b.

Malikshah, in which he tried once again to attribute the text to Nizam al-Mulk

The QasldaUuforfunately, with the exception of Schefer, scholars have not published theqasrdatogether with the main text of the Siyar al-mulilk. If they had, the falseattlibution of the text to Nizarn al-Mulk would have been identified previously.In spite of tlre fact that the qastdo ts inseparable from the Siyar al-mult1k, ttsirnply vanished frorn researchers' view for more than 100 years. Since theqastda is quite informative, it is necessary to provc that it belongs to

Muhamnrad Maghribr. Of the 54 distichs (bayt) of the clasTda, the first 21 ale inpraise of the book, the following 26 are panegyrics for Muhammad b.

Malikshah, the next four are intended to describe the poet, and the hnal thr-ee

resume the praise of the Sultan. Some of the verses in praise of the book use

practically the same tums of phrase that were used by Muhammad Maghribr inthe afterword written by him on behalf of Nizam al-Mulk (see below).

Qastda in praise of the huppy SultanGhiyas al-Drn Muhammad b. MalikshahoU:;,-(l. ;,rt loz gr--rJt .r!i J,.el oUlJ CJr J) oJ=€i

I This book is a sea, fu1l of diverse pearls,or a garden to enchant the soul, flill ofdiverse fruits.

-,6;tFuF :, tr 16 y"\ c*.r-.

,t oFvF I y crr;rura f! r:

2 A garden, ifin the galden there is aplace for calculating [the harvest],a sea, ifin the sea there is a place forpearls.

-rl* er-f :jq 7ta ft .-^t\.

Dt Uf )y. ,?.4 ft -*

110 T--- ^ ^-I nt /{uE ut I nE JELJUVJ Two tl,ttRRoRs FoR pRINCES FABRIcATED AT THE sEr-lue couRl I 1 I

3 Each chapter in it is like a tree fgrown]from the foundation,its fruits are all wonderful, and its leaves

are all choice.

)14j jl r:+)) jz L$l J-a3 a. .l

)f M ,PJJ t-+!t 4.3 ,,,)C

4 It is a treasure full of marvels, and a

fount fulI of rarities,it is a scroll full ofnovelties, and acasket full of pearls.

.*+g J:'-V.e L c*'zK ip"

,61 --*tt J el{ I e.-*-.1)

5 Its roots are all rarities, and its shoots are

all beneficial.Its chapters are all meaningful, and itscomments are all homilies.

$, dAA iej S tttj +"r ;I.,l

Je e -i-r" *jl+ a.ca ,ri:Ia!

o There is seriousness [in it]. advice.wisdom, examples and narration,the tales in it are innumerable. and thestories in it are countless.

olllr9 JE IJ c*-(r3 b-3 r:*.ra

-F bt -H 3 cr-t{- r) J+1

7 Its style is flawless and as high as theHeavens,its meaning is like Venus, shining at

dawn.

ot*l F Jtr J ui.€4 Jl lrl,iJl

s,- K au.v o-$ _r it css

8 fThere are in it] the custom, traditionand life of crown-keeping shahs,

the arrangement of the kingdom and the

people, the determination of good and

evil.

)lJrU dLAlj etr, : t-u_l S ,Hl

,.9 f J-JiE I J[ 9 .(J" --,-;

9 You will clearly see in it descriptions offeasts and palaces,you will find hidden in it descriptions ofbattles and ofattack and retreat.

.1.6U \U -a e ! \:,AJ JLt q r) -!

9q5 c ei' cJi-, Ol.4i e\) rU

t0 The acquisition ofproperty and

belongings, the ordinances ofjustice and

religion,the organization oftroops and the

determination of time.

at-r3 rt: ;*_r; c<l,J Jb J=€*,"

Fy,t t-& 31i-"{ ,K r..ti

ll One can hnd in it the way of the wisherofevil and ofthe wisher ofgood,the means to attract beneht and the

methods to avoid loss.

ot1l>Sj S o!F{ ca-J.b r) u!

f q!: tgtt J c4i! i )L+-!a

12 Every word and every rneaning that are

contained in its chaptersis dearer than youth and sweeter than

sugar.

c*1 \:ai r;u6;t* I'oJ liJ ,'o

p )t j4F+ G\r tt ;f"

13 It is free from sarcasm and innovationand purged offancy,it is worthy as knowledge and as

necessary as raln.

t3a _rt o;5|3 c.et'j JV ; upW

-h O_f a:+-U3 jltr J^4-a c:r-Li

T4 From reading it, boredorn will notovercome the reader,25

everyone will gain insight who willhonour it with his sight.

J)1" 5 o-rig tS; p.tiry- y

;et y:;r;3/6 f y+ tf

l5 For every story [in it] there is an

argument from an ayat of the Qur''an,for every chapter [in it] there rs a

tradition from the words of theMessenger.

lJJ, ,F- .rTj c--i ; ts a-as .ra

r? 6- rlUJP )\,F ,f

l6 From any speech which you recall, it has

a srgn,

from any art whose name you mention.it has a trace.

d!5 -91Jtjl ,r'5 rU 6;,-.- ,,0 ,t

jt ,$t 6J,'U 6r,1 7 jt

1'7 A code of the customs of eminentluminaries,a list ofthe schedules ofcrown-bearingshahs.

j*v A€;1. sl+p1 dtr!

)J*tt otbu a..l;11.' c-"r4!

18 Everyone who reads this and puts it intopractlce,will eternally be successful, do not doubtthis.

y-t Jr.1t5:3q 3 tit3fl 4-l 6 ,rS -*

n JtSt g,,lq ,Jl)q JqSl

I9 A religious and righteous shahin-shahdeserves

a compilation-souvenir of Nizam, the

well-living.

19)l) ltr u4t $4a ;9>s,ul

eF eb )<)u".iJu

20 Never before has a shah seen anothervizier like this,never before has anyone created anotherbook like this.

;* ft GJ-t rJn,*'F*

ft e "iut.F ,-S F*

t2 THe A.c;p oF l-HE sELJUes J'wo vRRons F()t{ pRjNCES FABRjcATED AT THE sEL.rue c,outl il-l

2I This blessed copy-book ofthe vizier ofthe en-rperors,

may it be auspicious for a religious andrighteous shal.r!

bt31* ;9ut 5!" p: ,r"r

fts 3 ,t> ,Ji & J )L ore,'i

22 Sultan Ghiyas-i Drn-i Muharrmad, thatMuhamrnad 26 to whon-rGod gave the throne to triumph over.theenemies of religion.

aS;l .r* r@ OJ) nqi irUJ.

)ib Jt \SlJet J j)e ,SlE j^tr)

23 The shah, on the royal throne ir.r thepalace of religion,is a shahrn-shah ltke his grandfather anda shah like his father.27

'J-) 6V" )t ,s+i J-J' -r d-,,aLi

rt3q c-oLi3 Je rl-;ra C-=g-i!=ar-.1

24 The son will be a sultan, whose father isMalik [Malikshah],the father [Alp-Arslan] will be anemperor whose son is Malik.

"it" ;.rurt, ;4 6 rt.:ja dl-LL

r-i -i-l.:l.t *i, of ,-q t-r.t :ru

l) In order to sustain his victory constantly,divine fortunewill frequent the place, where he hasoccaslon to pass.

?tb,y*-e,tr"at 6\ 43>

tG >t t.t ,r E 1K.tSr4i

26 Conquest and victory are ever more inlove with his sword and anowthan is the heart of the pilgrim with thestone fof the Ka'ba].

Jt I t ti oTJ d-:=..o *: €-F J ev Jr a4I .rt rya-:r

27 If they write his name on a thorn and ongranlte,from the thorn a rose will gr.ow, andfrorn the granite - silver and goid.

9r ;U -u-r-.;r F rws :V-t

i s g.6Jr' jl J t41 S! tv ;t

28 Although there is a great distance fromheaven to earth.although is plenty ofdifference betweenthe sea and a Dool,2s

aju4.$-qc-oCr=l+f

r+l;'u Jj .,-q c-r F. )q)J

29 Consider the former (heaven) as earthnext to his fortune,and regard the latter (the sea) as a poolnext to his generosity.re

,JLt ,Sj ojz 3l t)3> \..? 4t tri

.*, :*:t OF Jl g-oir lq q \ ,J-J

30 As soon as the wind of the swor.d ofmajesty blew ir.r the world,as soon as the scent ofhisjustice andf'elvour found the dry and rnoist fthings],

d1.6a )) (---" Jl c-+ a+ )1, rj

i J S:-; c-ili Jl c*a, J,rp ut u

3l a partridge built a nest in the eyelasir of afalcon,and a gazelle went with a lion cub to thewatering place.

,rl=ji jq oy t tW SE

)f q\ )". r"= \ e3.,t gi S

)z As if he iiad inherited all three matters:knowledge and bravery from .Ah andjustice frorn 'Umat..

)g L' _p O\;u. J=a; J, eFJae )t )ltt,sl,e jl .-l+i J ,'to

33 O shadow of God of the world over allcreatures,

Sun ofemperors and the Sultan ofthesea and land,

out{- J r)16- d;lJt tu-tJ Gl

1 3 ,a irLL : it::-i #:f

34 allthe shahs are like a neclclace and vorubeing is the middle stone,3othe Universe is like a garden, and in it isthe presence of your justice.

aL't3 3;,r:"o g o-f-cr j? Ac.o OtaLi

Pi)t)bttiat4tFt#

35 The necklace will not be beautifulwithout a middle stone,the kingdom will not have elegance andan aura without iustice.

cJt"a nL-9 ,rr riu l1 o..tie ,rP S V) rl) !s, "l.ju l) .<1" J.

JO Your age has become an omament to theeras ofemperors.Your order turned away the dangerarising from the rebels' ways.

otJ;-.) e)tt )t.P ra t Wp;,sr:<, t+ )t 5J j e.t

.JI From the excess ofart, which youalways show artfully,as if the Universe shrank to the world ofart.

{{ .r.a.SJl ;^jt l6,t- ,t, ),

j"..iJ!o Lsr4i, o44 gF

38 You are reasonable and pure ofsoul, ifin the soul for reasonmight be a space and is a wor.thy place.

r) Ji, or- f ,su= o4 up;', cSV r.5r j ol.(r lj.;(*

39 The entire vault of Heaven girds up itsloins for the service to the one.who girded up his loins for the service toyou.

.<l!,r.i Fi .-l^t _rdr.r+

-t"to* J i L4Jt .33t a-5a

40 And for him who opens his lips [it:tongue] to compliment you,the Wolshipped one will open a

thousand gates of happiness.

:t31 9i G;Ui d.i dtJj 3r al u,S;ig

J) )fj^ ())r^rj ;*-t;K >St

4l Your enemy will not be equal to you onthe day ofbattle,if[even all] the people ofthe day ofresunection assemble on the held lofbattlel.

3r.y1 9\t'-VyygeJir r}S,JlJr.l il-* f- ,95 f

42 Your arrow like the arrow of death willfly from the soul [it: from the liver],if your enemy makes a shield fromhundreds ofplates.

k ,, tr,ast St i _F oiJ- J5,JlJl ,t- ;t 3s;r":: f

43 A soul that [was] against you has

departed the world,"without a doubt, in the other world itwill reside in Hell.

tl3r,1 rltgt)t u!1 3; ul)l;r1 6,3b

e *' .r;,tlOW,-rt*.(.:a

44 Your insignificant slaves in worth andrankare Isfandryar, and Rustam son ofDastan, and agedZdl.

oV S l& cr="SoK-r. l*^.a

D Jtj f ,fU.r gr_1 3 ltr-Jrirl

45 No one can escape from fate and

destiny,your sword has become like fate, andyour command like destiny,

,f-f 'rt tss r."s 1r .r,ryl

's Ofi ! ,eg t: r-23 ;tn i €

46 Since after this you will take from themischief-makersall the gold there is, like the gold dug upin the districts.

,Jlrri,jl .lE*l Cfjt .i+ d;fjl

tf i of i 1+.4 r:-o 6r.r.ro

47 For the persuasion, promise and reliance(on Allah) that you have,God will give you everything that hasappeared in your thoughts.

ea lj rA 6,fys ws rtc=l yl

*-,, aJ-t li 4?r,'l g,-lJt Lo!

48 This long-standing slave, calligrapher,and panegyristhas the rights for thirty years of serviceand more.

aF CbS 9t-h;9 ,=c-.r3 o.r: y-t

,*i/ d!.- .,J crJt 63re. tltt

Il4 Tur e.cr oF- rrrE sEl.rues Two Ir,tnRons FoR pRINCES FABRIC'ATED AT Tr-rE SELJUo couRT 115

The fact that the addressee of the qasrda is the sultan already says a lot. It is

unlikely that the 'man in the street' would dare to write - let alone bring to thesultan - a panegyric, knowing that there was a whole staff of state poets in theservice of the great Seljuqs. In other words, the dedication of the qastda to thesultan in itselfprovides a strong indication that the author was a colrrt poet, and

rvas used to mixing with high-ranking offrcials. Therefore, it would not havesurprised anyone if he appeared before the sultan with a message from the lategrand vizier.

All of this is also confrrmed by the 48th distich of the qastda, whele itsauthor announces that he has been in state service for over 30 years, working as

a calligrapher and a panegyrist; and verses 49 50 once more confirm thestatement in the foreword: the author wants to be worthy of his fotlner status,

lost by him due to the collapse of Malikshah's govemment.

49 His ver:se will follow verse, and hisaffairs will be in order.if the shah mercifully throws a glance inthe direction ofthe slave

lU r-"_r ;V;tS I arQ t"; ,-iuSn

b 6,-*S o"+ st-r 8 f

50 If he gives him back his name, tladition,and respect,

for the slave to revive and becomewinged and feathered.

"ta> _7t ;! jl u./) pJ, eu rrl

,y J! ;.r-iu i ,tF o;lj) o*. v

5l For the firmament to be with an axis,and for the earth to be with stability,for the day to be lit up with the sun, and

the night with the brightness of themoon.

)tj \,-tw ry ll.tr l1 gra U

P tb tt ? J )y )t,fst btv

52 May everything that the sun shines upon,pass into your possession!

May everything that the body of theearth brings forth, be your treasure !

ou!'i c.!u )V q -p rq 3, .i,

tJiv lf r!) 4" .1l t\ i {

53 May fortune and victory be your helpersin your home!May triumph and success accompanyyou on your travels!

-p )t yl-J cJ9: :! 3; Jt1

i- tr JFtf o).ai )u j c.iq

May your day and night be auspicious,and may every'thing be distant from youthat does damage to your magnificenceand hieh rank!

)! j) tj J Q9 )b )u oljre

tt 41 ,tI; tj oV: ))q e f

t16 THB,qce orr rHE sELJUes

There is, however, not one pelson with the nisba of MaghribT among the high-ranking Seljuq officials listed by 'Abbas Iqbal in l'is Vazaral. And it wouldlook rather strange for a Seljuq official to be known as Magtu'ibT (lit. 'fi'orn the

west'). So I would claim that the name of Muharnmad Maghribl conceals the

person of Muhatrmad Mu'izn l.hshabhli (,v-,,,*t+ dj! rs4)-the tnost

farrous court poet ofthe Seljuq dynasty

Most significantly,Ml'izn did not intend to hide his name uttdet a false

nisba in the afterword to the Siyar al-multlk. There was a fateful coincidencethat subsequently influenced all future study and understanding of this forgedcornpilation. Just one lower dot was added by a rnedieval copyist in the

manuscript that belonged to Charles Schefer and examined by me de vlszt, and

Mttzzl (t-rF) tumed into Maghribi (ce.y.) - that is, from a very well-knownperson to a person who is absolutely unidentified!

The fact that nearly one half of the qastda is dedicated to the book (21

distichs of 54), in addition to his foreword and afterword, shows his directparticipation in the writing of the Siyar al-mulik. Essentially, he praises his

own work. And even the absence of the qastda in the majority of copies of the

Siyar al-muluft has absolutely uo effect on the conclusion that the qastda waswritten by him. Let us irnagine that we had encounteredthe qasrda somewhere

separate frorn the text of the book. We would immediately ask ourselves: Whatbook by Nizam al-Mulk was under discussion, and who was the author of the

0q,stao !

The qastda was not included in the d:an of Mr'tzzl edited by 'Abbas Iqbal,which surprises me, because, just one year apart, he first published the drvan byMu'izzr (in 1940) and then the Siyar al-muluk (in 1941), and could easily have

attributed it to the poet. But the fact that the qasrda was written by Mu'izzr, and

not by anybody else, may seen'r even more convincing if, besides the arguments

given above, we compare its style with tl,e style of the clasrdas included in his

atvan'.

a) I have conected two distichs (28 and 29) of the qasrda inthe Siyar al-mulilk based on the text of another qastda whtch was written byMt/izz for a son of Nizam al-Mulk

- Mu'ayyid al-Mulk (killed 494

[1 ro1]).r'?b) The expression 'thirty years of service' (4lt* L# &.ri), used in the 48th

distich of this qastda, is encountered again in the 44tli distich of the

qaslda by Mrt'izn dedicated to the Sanjar' (see n. 9, above). It is notencountered among the works of any other Persian poets, as it applies toa fact from the personal biography of Mu'izzr. To compare:

48. This long-standing slave, calligrapher, and panegyris has the

rights Jbr thirty years of service and more.

44. Thirty years of set'vice, a/ier all, desemes a certain respecl,

thi.rty years of respect in sen,ice is not lo be neglected.

'l'wo t"trnnoRs FoR pRlNc'ES FABRIcATED AT THE sELJUe c'ouRl iiT

T)re plose text of the Srrral ul-tnuliiL and the qastclu Ihat goes with it add newfacts to tl-re bioglaphy of Mt'izn. From the 48th distich of the tlasrdcr, we learn

that, iu addition to his position as panegyrist - or more exactly, Master ofPoets-Mu'izz sirnultaneor,rsly held the position of calligraphel From the

forervord and afterword to the text in the primaly UrDmiya version, it follo',vs

that he was also a sclibe, but very far from an ordinaly sclibe. The 'privatebooks' (crB .sLti6), or, more precisely, the 'writings', inentioned in the

afteru,ord on behalf of Nizarn al-Mulk, should be understood as the 'privatecorrespondence' addressed directly to Malikshah rr Besides, according toNizamT 'ArfrzT Sarnarqandl (lived in the sixth [twelfth] centuly), Mtt'izzl at onepoint was favoured with the status of naftm ('conftdant' and 'drinkingcompanion') of Malikshah, and thus entered the circle of the sultan's intimatecourtiers.34 After the collapse of Malikshdh's court and the beginning ofintemecine wars, Mn'izzl l.rad undoubtedly lost this status.

The status of nadlm appears to have been the safest one, since radfus seen-r

to have been the most informed persons at the Seljuq court, whose multipleintrigues often ended in a lethal outcome for some of their participants.Perhaps, due to the absence ofdefinitive preferences in the verses composed byMt'izn for the candidates to the Seljuq throne during intemecine wars, he

might initially have been deprived of the status under Sanjar. In fear for his life,he might have tried to gain it back under his brother, the Sultan Muhamrnad b.Malikshah. This explains why he first composed the qosrda in praise of Sarrjar,

where he corlplains about respect lost to him, and then wrote the qasrda rnpraise of Muhammad b. Malikshah, where in verses 49-50 he once morecomplains of his hurniliated stahls using the same expression, 'thirty years ofservice'.

The Date of the Qaslda's Composition

Qastdas are known to have been written by medieval court poets, usually forsome memorable occasion, in order to record it in history, as well as the eventsconnected with it. Iu our case, tire date of the qasrda's composition may be

named quite unambiguously. The 22nd distich, beginning the praise of thesultan, talks of his victory over a certain enemy; the 36th distich sketches theenemy in more detail, rnentioning rebels and the danger from thern - whichwas, however, elin.rinated by Muhammad b. Malikshah; the 43rd distich talks ofthe soul of the enemy, which'will reside in Hell'; the 45th and 46th distichsinform us of the corunand of Muhammad b. Malikshah, and of the confiscationof all the gold frorn the rebels.

The only historical event tbat fits the details given by the poet nthis clastda

was the victory of Muhamrnad b. Malikshdh over the Isrna'rlT fofiress ofShahdiz, and the capture and execution of the lread of the Isma'TlTs in lran. Wemay safely conclude that the qaslda was con.rposed to celebrate this victory.Evidently, by this time - 500 (1 107) - the final text of the Sivar al-ntuh1k, along

118 T,,- ^ /.-I NE /\UE TJI I HT JTLJIJV)

with the qostdo, had forrned at last. Thus Mu'izzr had 13 to 15 years at his

disposal to er.r.rbellish Nizam al-Mulk's original contract of employrtent.Apart from selfish reasons, there seems to be only one possible and less

reprehensible motive for Mu'izzi's counterfeiting- that is, the possibility thathe had been asked personally by Nizam al-Mulk to compile and then 'improve'the text on his behalf, as is the practice of modem speech-writers. But thisclaim seems unprovable. Since Mu'izzr continued his service at Sanjar's court,the text with the qostda probably rerrained in the personal archive of the poetwithout ever being formally presented during the poet's lifetirne, and appeared

on the medieval book market only after his death. We may be sure that Mu'izzrdid not send the qaslda alone to his addressee, because ifhe had submitted the

qasida then he would also have had to present the text of the Siar al-mub1k,

which is under discussion in the qaslda-and this event would have been

recorded in the historical accounts of contemporaries. But there is no otherknown evidence, or even a hint of the existence, of the .lryar al-muluk in theworks of historians or contemporaries of the Seljuq period, apart from the

solitary direct reference in the second part of the Nasrhat al-muluk. This is the

theme of the final section of this chapter.

II. The Bipartite I'{asthat al-muluk and Its Relation tothe Siyar al-muluk

Previous Textual Studies

The second part of the Nasrhat al-muluk (henceforth NM2) consists of sevenchapters. Taken together, they are three times longer than the so-called'foreword', or the first part (henceforth NMI), which was certainly written byal-Ghazali. The Nasthat al-muh1k was compiled after the Siyar al-mulift. Thisis why its first publisher, Jalal al-Drn Huma'T, suggested lhat aI-GhazalI used

the td'ly'of Nizam al-Mulk for his own work. In other words, he borrowed 14

accounts from it and incorporated them into NM2, once even referring directlyto the Sryar al-muluk:

Throughout /&e NasThat al-rnuluk, which consists of two parts - lhe

foreword and set,en chapters- l6 stories, precepts and coutsels are laken

.front the Siyar al-mulflk. Two of them are in the frst part, and the rest ofthem are in the second one. Especially those given in the second part arequite obviousllt and clearly taken directly from the Siyar al-mullk, since inmany places the phrasing of the two books is identical, coinciding word-for-word.ls

Hurna'T's viewpoint on the correlation between the two texts remains the

predominant one today for scholars who prefer to considet'NM2 as an authentic

contribution by al-Ghazah.

Two tqnRons FOR pRlN['ES IfABRTCATED AT Tt-il] sELJUo cour{T I 19

In my opinion, it is impossible to irnagine the Argument of Islarn (Hqjat al-

Islarn) - a mtjtahid of respectable age, ah'eady known in all the comels of the

Islar.nic world-copying excerpts frorn the compilation of tl.re Gland Yizrer,even if this eminent state figure was his fellow countryman or fi-iend. Appliedto the situation today, it would rrean that a religious figure of high rank(cardinal, archbishop or imam), already advanced in years and working on a

religious treatise, had copied something from a report by the prirne minister ofhis country. This seems quite ridiculous to me. It seems most probable that boththe Siyar al-muluk and NM2 were compiled by the same person. Until now,such a possibility has not even been considered.

Besides Humd'i, the problem of the authenticity of NM2 was also discussedin detail by Patricia Crone in her article published 25 years ago, and recently byNasrullah Pourjav6dy.36 Their final conclusions were based rnainly onscrupulous textual analysis of NM2 in comparison with NMI and otherauthentic works by al-Ghazdh. Taking into account their conclusions as well as

the findings of the first section of this chapter, the present research aims to givea few additional arguments in support of their point of view, mostly in terms ofthe historical factors that led to the compilation of NMI and the addition to it ofNM2. To this end, one should look more closely at the historical circumstancesand date of composition of the Nasrhat al-multik.

Historical Context

The historical circumstances related to the writing of the Nasthat al-muluk arereflected in the first three of the letters addressed by al-Ghazab to state andreligious figures of his time. The letters were collected by a medievalcompiler-evidently a descendant of the imam- into a volume published by'AbbEs Iqbal- He su_pposed that the volume was compiled 'not long after al-Ghazdli's lifetime'.'' However, judging from a citation taken by the compilerfrom al-Lami-vya - the famous qasrda by Ibn al-ValdT (d. 749 [t349])rs - thistook place more than two centuries after al-Ghazdh's death. The late date rs

indirectly confirmed by the compiler, to whom the Nasrhat al-mulilk wasalready known under this title as a work in its own right.

The three letters present a brief chronology of the events of 503 (l 109) -one-and-a-halfyears before al-Ghazdh's death, after he stopped working at theNrshdblr Nizamiyya, asking to resign- Accompanied by the compiler'scomments on them, the letters do furnish sevelal irnportant details: the datewhen the Nasrhal was written, the reason for its composition, its addressee, andthe location where the text may have been edited.

The first letter is al-GhazdlT's letter of excuse in which he explained why he

could not visit Sanjar, in order to clarify the situation relating to his 30-year-oldnotebook, entitled al-Mankhtll min ta'hq ol-usfil. This notebook was forged byhis ill-wishers, and he himself was accused of dishonouring the Sunnites ingeneral and the Hanafites in particr,rlar. However, Sanjar sent a messenger

t20 THE nce oF TFIE sEr,tues

ordering al-Ghazah to come to the headquarters located in the settlement ofTurugh, near modem Mashhad. Al-Ghazdtr did so, and tnade a speech at the

reception. Sanjar asked him to write the speech down in his own handwriting

and give it to him. Thus, another letter to Sanjar appeared which is essentially a

lecord by al-GhazalT ofhis speech at the receptiotl.

Upon receiving the text of the speech, Sanjar sent the author some game

which he had hunted and killed himself as a sign of his attention and favour.

Al-GhazalT appreciated this attention, and compiled the Nasthat al-multlk in

reply. At the end of the text, he gave a brief explanation on the chronology ofevents connected with al-Mankhil min ta'ltq al-usul.In this explanation, he

inforrned Sanjar also how his ill-wishers had first tried to falsif, his

autobiographi cal al-Munqiz min al-zalal and the Mishkat al-anvdr, and then to

gain his written permission on the backs of these two works to make authorized

copies (asl) of them with his help. This 'explanatory note' or 'covering letter'

attached to the Nasrhal can be considered as the third letter by al-GhazalT to

Sanjar.Just as the qastda in the Siyar al-mtrlilk slipped fiom researchers' field of

vision, the letter has never been associated or published with the Nasrhat al-

muluk as the author's supplement to it. But the letter does appear to belong with

the Nasthat, and to indicate that the Nasthat was part of officialcorrespondence, and could not have been voluminous before beginning its

independent life. Since the letter also shows that counterfeiting a book during

the Seljuq era was not an unusual practice, it deserves to be cited here.

It happened that in 499 tll06l thewriter of these words' al-Ghazal4 after

he had spent fwelve years in seclusiott .'. in his cell, undertoolc to go loNisabilr and to occt'tpy himself in dffising knowledge and propagating the

Shart'at, since languor and exhaustion had penetrated into the motter ofknowledge. Then the hearts of the venerable ones atnong the lords of the

hearts and the people of inner vision [epithets for lhose with special

spiritttal qualitiesJ canne to his aid: exhortations came in dreams attd

wakefulness that such a ntotion would be the starting-point for acts ofcharity and the reason for knowledge and the Shart'at to revive-3e

After consent was given and leaching became lively, and knowledge'

seekers began convergingf'om all over theworld, envious people arose out

of envy- They could find no acceptable invective other than to make a

falsification (albts). They changed several phrases in the book al-Munqiz

ntin al-zalal and the book Mishkat al-anvdr, introducing phrases ofmrbelief, and senl them to nte so that I wrote my perntission on the backs ofthe books. The Worshipped Most Glorious and High, by His mercy and

generosity, ittspired nte to shtdy them and realize theit'falsificatiott.Then this incident becatne known to the head of Khurasan. He arrested

the forger (nuzawir) and finally sent him away Jrom Nisabur. Then the

forger went to the headquarters of the Malik of Isldnt and loosened his

Two vlnnoRs FoR pRINCES FABRTcATED AT rFrE sELJUO ('ounr 121

tongue oJ itnecliye, butt'vas exltausted b.l,this. Then lte took o nole|;ook lhatI had [compiledJ in nry childhood, on the bacli of which I had written al-Mankhtll ntin la'ltq al-usill Thirty vears before this, some people oltt ofsn1,y had alread,v oclded to it several phrases of invective against lhe InrantADU l70ntla.Then sonte religious people uttered contplintents concerning tne, explailringthe situation to the Malik of Islant and exaggerating it so ntuch that theMalik o/ Islam said: 'We *-ish to see hint, listen to him and be blessed b1t hisprayer.'Then, as he directed, we immediatelywent to the shrine [of InnmRiza, i.e. the city of MashhadJ. Then the Malik of Islan sent the supremejudge Mahtnild to the shrine, who being one of the elites of the Hazrat and astale counsellor, v,as in fact a bilateral assistant [i.e. assisting both theHanafites and the Shaf itesJ nominally and substantially, with the ntessageof a desire to meet. Then we had to go to the headquarters o/'Turt1gh, to thethrone of Malik, and pray for him. Then there -yvas an order to con/irm withour own handwriting what had happened. In compliance with the order,what had happened in fact was conJ'irmed.ao

Thus, the place of action and the circumstances that prompted al-Ghazah towrite these letters, and also their contents, certainly show that t\)e Nctsthat waswritten by the author in gratitude for the hunting trophy sent to him by Sanjar.And the addressee of the text was Saniar. and not his brother Muhammad b.Malikshah.

The Size of the Text

The historical circumstances also provide evidence that the genuine Nasthat al-multik should have been shorl enough to be compiled without any considerabledelay upon receipt ofSanjar's hunting trophy - say, between two and five days.This goal could have been reached only by means of corrpilation of thernaterials taken from the books already written by al-Ghazdh, without carryingout any further research, which would have taken considerable time. NM2accounts for three-quarters of the combined text, and its contents consist ofmaterials taken from a variety of sources, most of which had never been usedby al-Ghazah before.

The bipartite Naslhat al-mulilk resembles a book in its full sense, thoughwithout a traditional foreword. All of the reasons given above for its absence,therefore, can also be applied to this case. I doubt that al-Ghazah even gave atitle to lris short epistle to Sanjar. Most probabiy, it was titled later, since itstitle, usually translated as 'The Counsel for Kings', can also be understood as'The Counsel of Kings'. The latter interpretation exactly corresponds with thecontents of NM2.

There is further indirect evidence that tluows more light on the size of thetext and the identity of its addressee. In NMl, al-Ghazah used an expression

122 T--- . ^-I r1I, 1\Uts UI I Nts )ELJ UVJ

that under other circumstances would be self-evident and unremarkable. He

wrote:Wen the stm rises, tell sonteone to read this kitab lo you anC to read itagain everyt Friday until it stalts irt )tour ntemory.*'

J t- Jil3t j! a+rT f 9 tJt\jz f I e6 4-t tJ ,s& \,rJ,.rr-I 1. ,-tUifi A9a

.'f,bl )U,

Of course, there is nothing unusual in the private correspondence received bymedieval sultans being read to them by third parlies. But by Sanjar's own

admission, in a letter published by 'Abbes Iqbal, he was absolutely illiterate -he could neither read nor write: a fact which everyone knew about.a2 Hence, itwould be disrespectful on al-GhazalT's part to recommend Sanjar to remember

the text by heart, when it was increased by three-quarlers due to NM2, and withthe endless stream of names that appear in it. The second parl is clearly not

suitable for memorizing.

Pro c edures for Offi cial Correspondence

The fact that the addressee of the Nasthat al-muluk was one of the highest

officials of the Seljuq state determined the subsequent destiny of the work. As

we have seen, each Seljuq ruler had a private depository where the

correspondence addressed directly to him was copied and collected' As a

private epistle, therefore, the Nasthat should certainly have passed through the

usual bureaucratic procedure, coming under the responsibility of a scribe ofSanjar's private correspondence, who should then have copied the text for the

depository. After al-Ghazaf's death, the text might have been augmented withNM2 at the same place-

Textual Analysis

The only substantial argument in favour of NM2 having been written by al-

Ghazall' which Huma'T gives in his textual analysis, is the mention of the

bipartite Nasthat in the works of medieval literary critics, written almost a

centuy after al-Ghazali's death-together with the appearance of its A-rabic

translation under the tille al-Tibr al-masbulcfi-Nasthat al-muluk as early as the

second half of the sixth (twelfth) century. At the satne time, Huma'r provides us

with 11 arguments against al-Ghazah's authorship of NM2. All these

arguments have been scrutinized and confirmed by Crone, and were recently

discussed by Pourjavady. In addition to that, Crone has examined the

discrepancies between the two parts in terms of the religious and ideological

backgrounds of their authors.

I can add to these arguments that, at the beginning of the aslrl of the Ktmiya-yi sa'adat and the kitabs of the Ihya' 'ulum al-dfu, al-Ghazah always mentions

the headings of the chapters and their numbers, while in NM2 the author does

not say a single word about the seven following chapters.

Tr.ito vrnRons FOR pRINCES FABRICATED AT THE sELJUo couRT I23

Besides, if NMI consists almost entirely of excerpts trom the l{tmhta andthe lhya', then why in the seven chapters of NM2 does the author tesort toother sources? Here it should be noted that, irr his later years, al-Ghazah did nottake inforrration from new sources. On the contraly, for each subsecluerrt bookhe used the works already written by him, each time arranging the materialdiffer-ently depending on the topic to be discussed, while Huma'T in his researchsingles out at least 1l works that were not written by the imam, but were usedby the unknown author to compile NM2.

As the textual analysis of the content of HMI shows, al-Ghazalt's conceptwas rather simple. For his illiterate addressee, he used an image of the Quranrctree and revealed it from a religious point of view: 'Hast thou not seen howGod has struck a similitude? A good word is as a good tree its roots are firm,and its branches are in Heaven' lQ 14:241. The 'good word' (kalima tayyiba)in the verse is commonly understood as a formula of the Islamic creed (al-shahada). To fully reveal the image, al-Ghazah look recourse to the workscomposed by him before. The main of them was his Ktmiya-yi sa'adat.

--j:---INContents of NMI-.--.-

The Kuniya-yi sa'adat The Ihya' 'ulum al-dtn

The ForewordOn the Principles of theCreed, which are theRoots of Faith

Ten principles-roots

The lst Chapter('Attaining Orthodoxbelief ) of the lst Pillar

The Branches of the Treeof Faith

Ten branches

The l0th Chapter ('OnHaving the Subjects andRuling') of the 2nd Pillar'

The Springs which waterthe Tree of Faith

The beginning ofthe lOthChapter of the 2nd Pillar

The First Spring

Ten examples

The 3rd Topic: 'OnSpiritual KaowledgeThis World'

theof

The 6th and Ttlr booksof the third volurne(rub)

The Second Spdng

Five anecdotes

The lOth Chapter ('On theRemembrance of Death')of the 4th Pillar

The 7th book ofthird volume (rab )The 10th book offourth volume (nrb')

the

the

Despite the obvious tirne limit, he has shaped this simple image completely andhas made it courprehensible to anyone, with no need to add anything to it,especially in the fortr of the seven following chapters of NM2.

Furthermore, it seems incredible that a Shaf ite imam, and especially al-GhazalT, would acquit a counterfeiter of falsifying a letter, as is done by theauthor of NM2 in its fourth chapter. The chapter relates how an educatedperson from Iraq became bankrupt and decided to solve his financiat problems

t24 Tue ncg oF THE sELJUes

by counterfeiting a letter fron'r Yahya b. Khelid al-BarmakT to 'Abd Allah b

Melik al-Khuza'I, a favourite of the Caliph Har[n al-RashTd.ar

A Portrait of the Author of NM2

The above evidence provides us with tbe following traits of the author of NM2.Some of these have already been identified by Crone. To sum up: a) in terms ofage-'a younger contemporary of al-Ghazah'; b) in terms of professional

background-'notan'alimatall,stilllessathinker; notareligiousscholar'; c)

in terms of national identity - 'an Iranian, more precisely an Iranian

nationalist'; d) in terms of confessional priorities - Zoroastrian; e) in terms offonnal religious stahls -'a Sunni, possibly a Hanafi'; f) in terms ofoccupation'presumably a secretary' (dabtr).aa

One can add a few features to his social status as demonstrated explicitly bythe text of NM2: g) he was an educated lranian who, at least, liked Persian and

Arabic poetry, and perhaps also wrote verses himselfi h) he memorized a not

inconsiderable number of short Arabic expressions taken from the Qur'6n and

traditions (ahAdTs) in order to illustrate his speech; i) his use of the forms ofaddless ('O Sultan of Islam', 'you', 'O brother') canbe treated as an indication

of an informal relationship between the addresser and his addressee. The latter

was undoubtedly a Seljuq ruler of high rank, since the former praises this

dynasly as the current one.

This portrait provides us with suffrcient reason to consider the author ofNM2 as a nadlm, since the same characteristics were described by Kaykav[s b.

Iskandar (lived in the fifth feleventh] century) inhts Qabtls-ndnta in stating the

qualities necessary in a nadtm. Among other things, such a person must be well

acquainted with secretarial duties (dabrri); and even if he is not a poet, he must

be proficient in Arabic and Persian poetry and able to cite a verse at the

appropriate moment; he must have memorized many stories and tales; he must

know how to play backgammon and chess; he must know the Qur'an, its formal

exegesis, some Islamic law and Muslim tradition in order to answer a mler'squestion when necess,ary; he must have read a great deal about the deeds of the

kings G iy ar - i mu I t1 k)."'Hence, the views reflected in the mosaic pattem of NM2 are not al-

Ghazdlr's own, but those of the nadlm, who has transformed the imam from a

strict Shafi'ite into a religious liberal. According to Nizamr- 'Aruzr SamarqandT,

'all Seljuq kin were fond of poetry' ('ri'ljr; r:uj'Uu:l 4'^ dJ+l*r,Ji;, and some ofthem preferred to have only poets as nadms.*o As demonstrated above in

relation to the Siar al-mulilk, Master of Poets Muhammad Mu'izzT had

undoubtedly lost his status of nadtm drre to the collapse of Malikshah's court.

Then he composed two qasldas in praise of Sanjar and Muhammad b.

Malikshah, where be twice complained about respect lost to him. He even

fabricated the Siyar al-muluk and ascribed it to Nizem al-Mulk in order to

Two turmRoRs FoR pRINCES FABRTCAIED AT THE sELJUo c'ouRl 125

recover lris job, and tbus retum to his lost positions and statr.rs. Obviously, hesucceeded in regaining all ofthese at Sanjar's court.

It is noteworthy that al-Ghazall's ill-wishers were accused by the compilerof his letters of misleading a 'certain Maghribi', inciting him to spread negariveinventions about al-Ghazali arnong the state figures at Sanjar's court.aT Sincethe cornpiler wrote it more than two centuries after al-GhazalT's death, he (or asource he used) could have turned Mu'izn into Maghribr. As the ill-wishersrelied on this person, he cerlainly belonged to Sanjar's close circle.

The Reason for the Compilation of NM2 and lts Addition to NMIHuma'T's critical edition of the Persian text of the Nasthat ol-nuiluk is based onseven manuscripts, three of which contain only NMl. The Arabic translation ofthe work, which was also included by Huma'T in his publication, does notcontain NM2 either. The separate existence of NMI under its own title in someof the preserved Persian manuscripts and under different titles in the Arabictranslation - along with the text of NM2 in Arabic under another title as a workof its own, and the bipartite Nasthat al-mulilk with its Arabic translation of a/-Tibr al-masbtik fi-Naslhat al-muluk - has already been described by Crone, andrecently discussed by Pourjavady. It led them both to the conclusion that NM2,an originally Persian compilation by an unknown author, was added to NMI inthe second half of the sixth (twelfth) century. Pourjavady seems inclined tosuppose that it was coincidentally made by an early medieval scribe, since'similar complements made by scribes are also seen in the writings of otherauthors'.48 On the contrary, Crone asserts that NM2, as a,practicalcomplement', was intentionally added by a medieval forger to NMl, andattributed to al-Ghazah shortly after his death.ae As it seems hardly possiblethal al-Ghazali's name was used coincidentally - and unintentionally-herassertion is more convincing, though she does not explain what the word'practical' implies in the case in question.

It does not seem unreasonable to suppose that the writing of NM2 waspreceded by a certain historical event that prompted the author of NM2 tocompile it, and that might be reflected in it. Its contents give a generalimpression that the addressee of NM2 faced a problem, and that NM2 wasintended to help him solve the problem. Its first chapter is mostly devoted toarguing that the behaviour of the former Zoroastrian rulers be taken as a modelto follow. The second chapter distinguishes between worthy and unworthyviziers, and identifies the kin of Nizam al-Mulk as the worlhiest of all, to bepreferred for the moment.so The very short third chapter, which concerrrssecretaries (dabtran) and their art, is obviously aimed at irnplicitly praising theauthor himself and his skill, by dernonstrating the breadth of his knowledge andhow worthy he is. The next chapter, filled with a variety of stories, describesthe meanings of himmat - understood by the author as generosity. lavishness,and the leniency of rulers and viziers towards their subjects (contrary to its

t26 THp,,qce oF THE SELruQS

understanding by al-GhazeL). Remarkably, it enjoins any ruler without these

qnalities to take lessons frour his viziers and nadtms. The fifth chapter plesents

quite a long list of sayings and maxims related to wisdom. Unlike other

chapters containing brief authorial remarks and explanations, tbis one seems to

include none of the author.'s own words The sixth chapter deals with the

subject ofintelligence, which was ofgreat value for Sasanians and other rulers

in making decisions. Finally, the seventh chapter is wholly dedicated to

women. It has a remarkable concluding section that sets out the author's

recommendations on choosing a wife and giving a daughter in marriage-

The author of NM2 thus offers the following advice: rule as the just

Zoroastrian rulers did; prefer Nizam al-Mulk's kin when appointing. viziers;

appreciate this author; be lavish in rewarding your servants (first of all, this

author); be a wise ruler; make decisions using your own mind and intelligence;

act contrary to women's advice; and 'take precautions in choosing a wife and

giving a daughter in marriage, especially a grown-up daughter, and so avoid

falling into disgrace and embarrassment''''Two events in the later period of Seljuq history seem to correspond with the

allusions made by the author of NM2. In 518 (l124), Sanjar gave his daughter

Mah-i Mulk-Khdtun in marriage to his nephew Mahmfd, a son of his late

brother Ghiy6s al-Drn Muhammad, and issued a document appointing Mahmud

as a successor to the throne. This was done at the insistence of Taj al-Dtn-

Kiatun, Sanjar's mother and Mahmfd's grandmother. A year earlier, Mahmud

had executed his prime minister Shams al-Mulk (Rabi'al-awwal 517 [May1123D, the fifth son of Nizam al-Mulk, by o"rder of his powerful uncle, and left

himself without a prime minister for a year.'"

Thus, the story of Malikshah, Turkan-Khattin with her young son' and the

murder of Nizam al-Mulk (committed due to the direct participation of his

fierce rival Taj al-Mulk, who, as we have seen, was the ally of one of the

leading Isma.rhs in Iran, Hasan b. al-Sabbah) repeated itself, albeit with some

variations of detail. It would not be surprising, therefore, to encounter one more

state figure who provoked all these actions and put some of them into effect. He

is stated by many historians to have been the most treacherous, villainous and

greedy vizier throughout the whole of Seljuq history. His name was Qavdm al-

Din Abt'l-Qdsim DargazTni (executed in 527 l1l33l - It was he who arranged a

conspiracy to discredit Shams al-Mulk in the eyes of Mahmrld and Sanjar, so as

to overthrow him and later occupy his position. It was he who, several years

before 518 (1124), was sent by Mahmhd to Sanjar's headquarlers as an ofhcialrepresentative or ambassador, with the sole purpose of ensuring Mahmhd's and,

of course, his own future. As stated by the late Jalal al-Din Muhaddis lJtmavt,

the editor of the Ba'zt masalib al-navdsib, 'to overthrow their inner rivals,

some of the Seljuq sultans, viziers and emirs used to involve the Isma'Tlis and

ask for their assistance'.tt His statement fits both Taj al-Mulk and Abl'l-QasimDargaztri.. The latter had a reputation for advancing his personal interests by

Two vttR.Rons FoR eRTNCES FABRTcATED AT TLIE. sELJUe couRt 121

inducing others to do lris dirry work, and tnauy of his potential rivals did nottake the risk of standing in his way.

In fear for his life, it is possible that the author of NM2 added his text toNMl, concealing his authorship because the practical advice given in NM2contradicted DargazTnr's intentions. This would explain why he left his textwithout a formal foreword. On the other hand, Crone is absolutely right insaying that the author of NM2 'rnakes no attempt to impersonate al-Ghazah bywliting in his style'.sa In my opinion, this discrepancy was cleally intendecl toshow that there was no relation between the two works except the commonaddressee, and to distinguish NM2 from NMI by the inclusion of somepractical advice under the cilcurnstances outlined above.

Conclusions

To countelfeit a book during the Seljuq era, when the battle between theSeljuqs and the Isma'rhs was being waged with varied success on all fronts,was not an unusual practice. Nor was it unusual to drag a prominent scholar orstatesman into the political bickering against his will, both during his lifetimeand - by editing his works and notes on them with selfrsh, ideological, andother partial motives - after his death. Therefore, the prose text of the Siyar al-muluk fabricated by Mu'izzr, as well as his qastda that goes with it, not onlyadd new facts to his biography and work, but also give us the evidence requiredto consider hini in the context ofhis times. In this case, one can be confident inthe attribution of the fabrication to Mu'izzr. The case of the second part of theNasrhat al-mululc is less clear cut, but here too one can discern the hand ofMu'izn. The principal reason for assuming that NM2 too is the work ofMtt'izn is its relation to the Siyar ql-muluk, since Mu'izzr seerrs to have beenthe only one who had the Siyar al-muluk at his disposal, and who could thusborrow from it and even refer to it directly in NM2. All of these facts can alsobe used in evidence against him, to unmask him as a counterfeiter.Nevertheless, these facts remain, in my opinion, insufficient to identi! thecounterfeiter with certainty. If NM2 was written and added to NMl by Ml'izzr,tlren he must have compiled it between 517 and 518 (1123 and lI24), not longbefore his death, when he was around 80 years old. If so, then thecounterfeiting was committed by him no more than l5 years after the date whenNM1 was submitted by al,-Ghazall to Sanjar, and the compiled text wastranslated into Arabic after the death of Mrjizn. On the other hand, thecircumstances reflected in al-GhazalT's letters with regard to the attempts atcounterfeiting some of his texts finally led also to the composition of the firstpart of the NasThal al-nruluk- and presumably also made al-Ghazdh-, or moreIikely his students, prepare a copy from his authentic work, so that today wehave two versions ofthe text.

128 TsE Rce oF THE sEr,JUes

Notes:

1. This section is a substantial developurent ofthe main ideas published in rny earlierarlicle on this topic: Alexey Khismaturlin, 'To Forge a Book in the Medieval Ages:Nezdnr al-Molk's SDrra- ol-Molulr (SiyAsol-1r,Urro)' Jttutnal o;[ Per,sianate Studies 1

(2008), pp. 30-662. See the bibliographical data for the editions of C. Schefer, B. Zakhoder, H. Darke

and J. Shi'ar below, in the notes-

3. C. Rieu, Catalogue of the Persian Manuscripls in the British Museunr, 3 vols

L(London, I 879-83), II, p. 446.On Jamal al-Drn Ilfaqshat/Ilqafshat b. Qaymdz see 'Abd al-Jahl QazvTnr PrAn, Naqz(Ba'zt masalib al-nawasib), ed. Jalal al-DTn Muhaddis Umavi (Tehran, 1358

[1979]), p 425 (fn.2 by Urmavi); and Edward G. Blowne, 'Account of a rare, if notunique, manuscript History of the Seljuqs in the Schefer Collection lately acquiredby the Bibliothdque Nationale in Paris', -rRlS 14:3, (1902), pp. 567-610, and 14: 4,(1902),pp.849 87(seeinparticularpp.ST0-l).Ataround564(1168),thereseemsto have been no other AmIr Hajib with the honorific title Jamal al-Drn.Nizdm al-Mulk, Ab[ 'AlT Hasan b. 'Ali TusI, The Book of Got,ernnrent or Rules forKirtgs. The Siyasat-ndma or Siyar al-nuliik o/ Nizam al-Mulk, transl. from thePersian by H. Darke (London, l9'7B,2nd edn), p. xii.See, for example, Marta Simidchieva,'Sidsat-ndne Revisited: The Question ofAuthenticity', in Proceedings of the Second European Conference on IranianStudies, ed. Bert G. Fragner et al. (Rome, 1995), pp. 657-74; and 'Kingship andLegitimacy in Nizarn al-Mulk's Siydsal-nama, Fifth/Eleventh Cenhrry', in B.Griindler and L. Marlow, eds, Wrilers and Rttlers: Perspeclives on TheirRelationship f'om Abbasid to Safavid Times (Wiesbaden, 2004), pp.97-I3lQazvrni F:an, Naqz (Ba'zr masalib al-nawasib), p. 119Abti'l-Hasan Shahvan, 'Ravebit-i Seljuqiyan va 'Abbasiyan (Malikshah va al-Muqtadr)' tn P azhtT his h- na nt a -y i ta r tkh, 3 : 14 ( I 3 8 8 [20 1 0] ), pp. 57 -7 8.Mu'izz7, Dwan-i Amlr al-shu'ara-yi Muhamntad b 'Abd a|-Malik-i Ntshoburtmutakhallis ba Mu'izzl, ed. 'Abbas Iqbal Ashtiyani (Tehran, 1319 [940]), p. 715.The verse corresponds to number 16,412 rl:''Abbds lqbal's numeration.'Abbds Iqbal Ashtiyani, Vazdrat dar 'ahd-i salAfin-i buzurg-i Seljuqt az tarlkh-itashktl-i ln silsila ta marg-i sulton-i Sonjar (432-552), ed. Muhammad Taqr Danish-Pazh[h va Yahya Dhukd' (Tehran, 1338 !959]), pp. 158 9

For more details, see A. Khismatulin, 'Vidy musul'manskoi nar.rchnoi litelatury vX-XV vekakh: sochineniya (tasn-t) i kompilyatsii (ta'l-ti' ('The Forms of IslamicScholarly Literature in the 10th l5th Centuries: Compositions (tasn-tfl andCompilations (ta'llfl'), in Rahntat-nama (St Petersburg, 2008), pp. 41043-Siasset nameh. Traild de gouvernement compose pour le sultan Seldjoukide Melik-chah par le vizir Nizant oul-Moulk, texte persan et traduction francaise, ed. CharlesSchefer (Paris, 1891, 1893), I, pp 12.Abtr'Ali Hasan b 'AlT T[si Nizam al-Mulk, Siyar al-muliik (Siyasat-nama), edH. Darke (Tehran, 1340 [1962]), pp a-5

14. AbU 'Ali Hasan b 'Ali Tiisi Nizam al-Mulk, Siydsat-ndma (Siyar ol-muluk), edJa'far Shi'ar (Tehran, 1385 [2006], 13th edn), p. 2.

15. Schefer, Siasset nanreh, I, pp. 21 0-l l; Darke, Siyar al-tnultlk, p. 307 .

16. Shi'ar, Siyasot-nama (Siyar al-ntuluk), p 5, fn. l.l7 Zakhoder's translation is approximately the same.

Two tr,trnRoRs FoR pRTNCES IABRTcATED A l r HE sELJUo couR.T 129

18. Darke, Siyar r,1unu,oO,p.4. Abh -Alr Hasan b.'AlIThsr Niz5nr al-Mulk,The Boolr

of Governmenl or Rules for Kings: TIte Si1ato,-ro*n or Sh,ar al-muluk of Nizam ol-Mulk,transl. fiorn the Persian by H Darke (London, 1960). p. 2. Sioset-nome Kniguo provlenii wazit'o XJ stoletiva Nizatn al-ntulka, transl. from the Persian, withintloduction and comueutary by B. N Zak-hoder (Moscow/Leningrad. 1949), p B.

19. Al-Ghazalr, Mokatlb-i fclrsl--vi Ghazall ba nam-i Faza'il al-anarn min rasa'il Huj.iaial-lslam, ed. 'Abbas Iqbal Ashtiyani (Tehran, 1363 [1984],2nd edn), pp 10, 11

(corresponding to the citations).20. Muntajab al-Din Juvayni.'Atobot ol-katoba, ed 'Abbas Iqbal Ashtiyani (Tehran,

r329I9501).21. Muhammad b Husayn Bayhaqi, Abu'l-Fazl, Tartkh-i Ba1,fuoq7, ed. K}alTl Khatib

Rahbar, 3 vols (Tel.rran, 1386 [2007]), ilI, p. 98722 Darke, Th e B o o k of Govern m en t (London, I 97 8, 2nd edn), p 7 4.

23. Bayhaqr, Tartkh-i Bayhaqr, II1, p 987.24 See, for example, 'Ah Abn'l-Hasanl, 'Mundarajdt-i "ztdd-| shi'ay"-yi Siyasat-nantct

az dn-i khwaja-yi Nizam al-Mulk nrst', Falsafa, kalam va 'itfin,29 3l (1378

[999]), pp. 150 9, 12847, 140-60, respectively.25. This verse repeats in the same words the text in the aftelword given on behalf of

Nizdm al-Mulk: 'For from reading this book, boredom will not overcome [thereader]'.

26. Aplay on the literal meaning of the sultari's title: Ghiyes-i Drn neans the 'succourofreligion', so the whole phrase literally means, 'Sultan Succour-of-the-Religion-ofMuhammad; that Muhammad to whom God gave the throne to triumph over the

enemies of religion'.27. Grandfather, i.e. Alp-Arslan; father, i.e. Maliksheh.28. In this bal I have slightly corrected the text as given by Schefer, according to the

text I will discuss below'first line: .tJ/ u J4 s-l .a a-f ) + f *.tj li rr{ s4 c-* :- t ^a fSecOndline: r-3UAj u-".?\} le)a1-)J +.r:Ugil14\J y)4.D.Sclrefer declined to trauslate this and the subsequent bayt:'Je n'ai pas pu trouver unsens raisonnable aux quatre vers dont j'ai omis de donner ici Ia traduction' (see

Schefer, Siassel namelt, II, p.310).29. The same goes for this bayl:

first line: cLt 'sj l* Jl cJJ) * I !I +srt+ ,s-t Jt c)t: (+ ! JJjl;second line: *t rJ bn 9l a q L,t),lJ +p p Jj7 3t ra s.a t l.'1 y-9

30, That is, a necklace where the central place is occupied by a large jewel or a pearl31. Here there is a very important reference to the death of an enemy of Muhammad b.

Malikshah. This referince makes it possible to give a precise date for thecomposition of Ihe qaslda with greater confidence, as well as for the completion ofcompilation ofthe entire text ofthe Siyar al-ntuluk.

32.Mts'izzr, Dlvdn, p. 252. The numeration of the verses corresponds to 6093 4 in'Abbas lqbal's numeration

33. This translation of the word ,tfiass (crl:) is confirmed by the Seljuq officialconespondence which has already been published by many scholars. See, forexarnple, Hedbert Horst, Die Stoatsverwaltung der Grosselguqen und Horazntshdlts(I 038 1 23 l) (Wiesbaden, 1964), pp. 19-24.

34. 'Umar b. 'AlT Ahmad b. NizamT '.nrlrzr Samarqanfr, Chahar maqala, ed M, Mu'tn(Tehran, 1382 [2003], l2th edn), pp.71-2.

35. Al-Ghazah, Nasthat al-nuh7k, ed. Jalal al-DTn Huma'l (Tehran, l36l [1982]), p. 93(of the foreword by Huma'r).

5

'7.

8.

9.

l0

l1

t2

13

130 T,,- , ^-I htr AUr Ur | flL JtsLJUVJ

36 Patricia Crone,'Did al-Ghazali Write a Mirror for Princes? On the Authorsbip ofNasthat al-multlk', JSAI I0 (1987). pp 161-91; Nasr:ullah Pourjavady, Dr7ntujoddid pazhuhishha'i dor baroli Muharnntcrcl-i Ghozalt yo Falihr-i Rozr(Tellran, 1381 [2002]), pp. 413-24.

31 ,\l-GhazalI, Faza'il, p 3 (of the foreword by 'Abbas Iqbal).38 Al-Ghazeh, Fozq'il, p. 3 (of the rr-rain text). The citation reads: ;l*i 4+.ri

the full version is: J+ g--.l j" fllr^)l g) ury;:yS ;2\ lv ;'1.39 The situation surrounding the departure for Nrshdblr on the insistence of vizier:

Faklr al-Mulk is described by al-Ghazah in almost the same language in his a/-Munqiz min al-zaldl. See al-Ghazah, al-l+hrnqiz ntin al-zalal, ed. Jauul Saliba atdKemil 'A1yed (Beirut, 1967,71h edn), p 12l-2

40. Al-Ghazeh, Faza'il,pp. 1 1-l 2

41 Al-Ghazali, Nasrhot, p. 4. Cf. Ghazalt's Book of Coun,sel for Kings (Nasrhat al-ntultlk), transl. F R. C Bagley (Oxfotd, 196a), p. 5.

42.'Abbas lqba\, Vazarat, p. 316.43. Al-Ghazeh, Nasthat, pp. 2l l-15 (Chapter 4)44. Crone, 'Did al-Ghazall Write a Mirror for Princes?', pp. 178, 180, 187, 189,

respectively.45 Kaykar.us b. Iskandar b. Qabls b VashmgTr', Qobtls-rtdntcr, ed Ghularn-Husayn

Ynsufi (Tehran, 1 385 [2006], 14th edn), pp. 2034.46. Nizemr 'Arun, Chaltar maqala,pp 71-2..47. Al-Ghazall, Faza'il, p. 3. The word combination looks rather strange (o..1t*.,t 5 *!).

In the initial version, this could have been written and understood as (b ././,).Otherwise, we have to admit that those who entered Sanjar's service from thewestern territories of the Seljuq Empire with its capital in Isfahan could bear thenisba of Maghribi.

48 Pourjavady, Du mujaddid, p. 415.49. Crone, 'Did al-GhazalT Write a Miror for Princes?' pp. 190-1.50 Al-Ghazalt, Nasrltat, p. 184; cf. Bagley, Ghazall's Book of Counsel for Kings, p.

1l l.51. Al-GhazalT, Nasthat, p. 285; cf. Bagley, Ghazah's Book of Counsel for Kings, p

172.52 'Abbas Iqbdl, Vazarat, pp 187,266-853 Jalal al-Din Muhaddis UrmavT, Ta'hqot-i Naqz,2 vols (Tehran, 1358 [1979]), l,

p 285.54. Crone, 'Did al-Ghazafi Write a Minor forPrinces?', p 190.

7

Sfylistic Contiieuities in Classicai FersianPoetry: Reflections on Manuchehri from

Damghan and Amir Mo'ezzl

Asghar Seyed-Gohrab(Leiden University)

he stylistic developments in classical Persian poetry between the tenth

and sixteenth centuries are commonly classified on the basis of poeticstyles and the literary activity in courtly centres ovet a wide

geographical area, although ideologies and political factors also played an

important part in this development The triparlite classification based on

stylistic criteria is as follows:

a) Sobk-e Khorasani or 'the style of Kliorasan', an 'inimitably sirrple'style mnning from the dawn of Persian poetry to the middle of the

twelfth cennlry in the eastern parts of Persia. I The salient features of this

style are the harmonious use of imagety, which is usually limited to one

couplet (mora'at nozir), and infrequent borrowing from the Arabiclanguage.

b) Sabk-e 'eraqi or 'the style of 'Eraq', indicating the shift of literaryactivities liom the eastern parts of medieval Persian territories to the

west during the second period, from the middle of the twelfth century tothe sixteenth century. During this period, the imagery and metaphors

become sophisticated, and there is an increasing use of extended

metaphors, sometimes running over several couplels. Learned allttsions

and the infusion of riystical themes made the poetry convoluted and

interpretable at various levels.c) The third period is called sablc-e hendi ('tlie Indian style'), extending

from the beginning ofthe sixteenth century to the eighteenth century inPersia, but still maintained in other areas such as Afghanistan, central

Asia and India. Poets of this period deviate from the harmonious use ofimagery, use allegory, and dwell on mystical and philosophical themes.r

This tripartite division of Persian poetry based on styie appeared in Persian

literary criticism in the nineteenth century, in literary circies in Kliorasan, to

which the poet Sabuhi, the father of Malek al-Sho'ara Mohammad-Taqi Bahar,