J A \ J U S T \ \ ' r T K A - \ 1

T H E L E I D E N M A N U S C R I P T O F T H EK I T A B A L - M U S T A . T . v T

Introduction

The Leiden copy of the tabular pharmacopoeia, the Kitab al-Musta'TnT by Yusuf b.Ishaq Ibn Baklarish al-Isra'ili (c. 5oo/rro6), is one of the oldest Oriental possessionsof Leiden University Library.'It is associated withJacobus Golius Q596-t667), whoacquired the manuscript while participating in a diplomatic mission to Morocco in16zz-4, before his dazzling scholarly and academic career had even started. In fi29the manuscript was incorporated in the Library, together with the other manuscriptswhich Golius had collected by that time, not only in North Africa, but above all inAleppo and Constantinople.

The sudden influx of more than zoo Oriental manuscript volumes'in the Llbrarymade it necessary to organize these materials. The University librarian decided tocreate an Oriental manuscripts section and to place the new arrivals together, witha separate system of class-marks. These are the well-known Cod. Or. numbers ofLeiden (Jniversity Llbrary, a system whlch is in use to this day.3 That momenr in16z9 can be regarded as the foundation of the Oriental collection within the LeidenLibrary. Oriental manuscripts had already been available in Leiden, albeit in smallnumbers, but it was the sheer bulk of Golius' acquisitions which made this novelapproach necessary. The Golius manuscripts were simply and roughly arrangedaccording to their size, and then numbered from Cod. Or. r onwards. Since thebinding of the Kitab al-Musta'TnT is of considerable size (3o.5 cm hlgh), it becamethe fifteenth entry in the inventory in order of size, and the volume consequentlyreceived the class-mark Or. r j, by which it has been known in scholarly literatureever since. The most important catalogues containing references to the manuscriptare DeJong and De Goeje's of r865 (pp.t+6--g), and Voorhoeve's Handlist, p. 243.4

r. The medical materials are given ln a rough alphabetical order by their names, divided into sections in thea l i I l , f I - j t - t 1 - 'rorlowrng oroeri di lJ, oa .Jtm. aat, na . waw. zay, na. ta , ya . kaJ. Idm, m|m, nun. stn. ayn.Ja , qaJ. ra , sad. ta .tha' , kha' , dhal, ghayn, shtn. Thts order is neither the Maghribi nor the Mashriqi order of abjad The sequenceof the lemmata within each letter (the first six simplicia given are: amir barrs, aqaqiya, aqhawan, abhal, anjara,anjaran; the last ones are shahm al-rubb, shahm al-tha'lab, shahm al-himar, shahm al-batt, shahn al-dajaj, shuhumnukhtallfu) does not fo1low this alphabetical order, nor does the order of the letters within each lemma.z. In fact the Golius collection now comprises 2rr inventory numbers, several of which consist of more thanone volume, and which together contain manv more texts.

3. At the t ime of writ ing (February zooO), the highesr number was Cod. Or. zo,5 r5.

4. Piet De Jong and Michael Jan De Goeje, Catalogus Codicum Orientalium Bibliothecae Academiae LugdunoBatauae, vol. 3, Leiden r 865 ; Petrus Voorhoeve, Handlist o;f Arabic Manuscripts in the Library of the Uniuersity ofLeiden and other Collections in The Netherlands, znd edition, The Hague r98o.

75

Fig . 28 .

Letter from Ahmad b. easimai-Andalusi in MarrakeshtoJacobus Golius in Safi,da ted iz Rab i ' t I ro33 A.H.(Friday z Februarv r6z4 A.D.).Manchester, John RylandsUniversity Library, MS Persran

9r3 , fo1 . r -59

J A N J U S T W I T K A M

Prouenance of the Leiden manuscript

.We are more than usually well informed about the Leiden manuscript's earlier pro-

venance and history. While in Morocco Golius corresponded with an old acouain-tance ofhis professor ofArabic in Leiden, Thomas Erp.rii", 1, 5g4-1624).Thi, l.;.;Musiim was Ahmad b. easim al-Andalusi (c. r5)o-after fi4o), an experiencedtraveller in partibus infdelium.s Jacobus Golius and Ahmad b. easi- .orr.irporrd.d.Golius was in Safi on the Atlantic coast, acting as an engineer attached to the Dutchdiplomatic mission, while his corresporrd.rri was in fuar.akesh in an infuentialposition in the Moroccan court.6 Hardly surprisingly, books are indeed among rhesubject of the correspondence and the Kitab al-Musta'tnl is mentioned several times.Two of Ahmad b. Qasim's letters to Golius are parricularly relevanr ro the earlierProvenance of the Leiden manuscript of the Kinb al-Musta'TnT and deserve to bequoted in full.7 on rz RabT' II ro33 (Friday z February 16z4) Ahmad b. easimwrote from Marrakech toJacobus Golius [Fig. zB]:s

5' See Gerard'Wiegers, A Learned Muslim Acqudintance of Erpenius and colius: Ahmad b. Ktsim al-Andalustand Arabic studies in the Netherlands, Leiden r988; Ahmad ibn easim a1-Ha.1ari, Kifib Nasir al-dtn ,ala'l-qawmal-kai'trn (The supporter oJ Religion dgainst the kJidels),historical study, critical edition and annotated rranslationby P'S van Koningsveld, Q. al-Samarrai and G.A. wiegers. Madrid r997 (Fuentes Ardbico-Hispanas, zr).6. Ahmad b. Qasim's letrers are now kept in Leiden cod. or. rzzg, nos. z,r, 13, z4, ror, r r4. Not so long ago,the former companion volume of Leiden Cod. Or. rrzg was discovered byJan Schmidt in theJohn RylandsLibrary, Manchester, MS Persian no. 9r3. A film of that manuscript is in Leiden (A zogz). s..1.n Schmidt,'An

ostrich Egg for Golius; The Heyman Papers Preserved in the Leiden and Manchest"", (Jnrversrtl.Libraries and Early-Modern contacts between the Netherlands and the Middle East', in his The Joys oJPhilology: Studies in ottoman Literature, History and orientalism (t5oo_�tgzj),vol.2.. orientalists, Trauelers anrlMerchants in the ottoman Empire, Political Relatiotts between Europe and the porte,lstanbul 2oo2, pp. g-71, esp.PP' r9-2o. The Manchester manuscrtpt also contains other materials in the hand of Ahmaa U. cirri-. Ahmadb' Qasim ieft other traces during his peregrinations. In their historical study on Ahmad b. easim,s memoirs,van Koningsveld, a1-Samarrai and Wiegers (p. :t, tr. 83) mention r.,r.."l -".rurcripts in the BibliothdqueNationale de France in Paris which contain notes in Ahmad's hand (MSS ar. rrBr, 4rrg,4213 and a34g).Ahmad b' Qasim's somewhat nostalgic note in MS ar. r r8 r, fol. 99b, deserves to be quoted here in fu1l, if onl1,to enable us to dismiss the fantasies of the three Dutch scholars about Ahmad b. easim's way of writinq hiso w n n a m e :

-r+ e-U; ,^-l a)LJl; a,l-L r-dt 4lll ljrLet .:>.J! orl L:)! "l .,,.i( e:l ,..,i,c-,.,,15Sll tu .-,]t;_e _;tr i;r^ i.iiS'I

was in the town of Barrish [Paris] and I sar,v this book and I knew that it hacl been written t:.#t'":0,*n"r.';is in the Maghrib, may God return us to rt safe and sound, wa-l-Salam. Ahmad b. easim, poor [servant ofGod], the Andalusian, may God forgive him.'

The writer of these l ines was in France between 16ro and 1613. MS Pans ar. rrgr (containing al-Jazul i ,sDala'il al-Khayrat, dated r Ramadan rco7f t5g9 and copied by a certain Ahmad b. Ibrahim) was"then in thepossession of Etienne Hubert (a mutual friend of Ahmad b. Qasim and Thomas Erpenius), as becomes clearfrom Hubert's owner's note on the second flyleaf at the end of the volume: 'De

la Billiothecque de E. HubertLecteur du Rov en langue Arabicque si1 se perd il vous supplie luy rendre.'Hubert "pp".".rrly acquired themanuscript, still almost brand-new, when he himself was in Morocco, a few vears earlier.7. Golius'part ofthe correspondence is not known to have been preserved.8.John Rvlands Library, Manchester, MS persian no. 9r3, fol. rOq (rro. Sz)

76

l,.-

u,.*'f "llptrjn ? J:*6- brJ^., f. 1Ut, .dr;*:l\e o$, ^ l\

.-: '. J*]i-1i -* ," t ijl-ry iJl * ,, iJ\ .q5o,i 1; - L li .;ksl\, , ? .

* ! * i : ' r

a *.1* *tL1*;e:.*ii*;**r3l**;Ll"-=,::b lx,: i":li * -"r-1,

r ; r : t i 5 l L i d i ! * - r " ; . ' - l a , . * s i L r . 5 : -= - t l * r r n "

. L = ";ii ,-,-L* ti;i J', *.,#-t r".-Jl-jl'i5lh. . ; i t . , . r - q , ' 1 0 , " ^

trrg;-:r*:,:;*e.1"-i i**f &b*rli-' +'u,,* -..' t. ---* *1 ;i l-r '.L'i f

'' . 1i# ;*,J*:* -**rySLt:.,-*-Lo o,*'1 * t***li ,. ' : iii ="-: 'l:.,U2 ,

, .r--*x.s,,,** . pt* ^^L, !" -a,., * j g&',-*b}-"t &' .L- 'b!-;u*'- I 5, *

, ai.L* - **,.-),* lo'-1L* rr-$'!": - : ;*" ^^ \^i, * l t U;ir-. ' ' ; .*

J A N J U S T W I T K A M

Alsenor jacob gule

;rJl p-l:,Jl i.Fii.r &-r,jlFjSYl &t & r)*llr 6)L:llr all r -\l

-H-clel+us oli r+r J-l; -.,yji+,s\i \i)lGjl;":illdiillu,*.,L{lltujrr r s^Jl u+-1..- .,ll,-.,uslr

l|iSll;o Llli-;l,,jJta.,JltJc."U<r t- Ll4j^.;i l-, -(;k l' '"€iJ-tul;' a1li-ti: Li-lroL-trjst 'ir irl il \-€-; li ..,p cl xh- L"s litiil d igr** crJr.,l ;Li n+'i,.*I-r ,r^,*J

I (iU,. 'i,LIl U+-i..- i-s,*r rJc 6ji-lj .9il ai,!,-, ':,ia: k # 4L ,J+J At^+_)! lpl5L

\itfi! Lj- J^c dr olll ..=J e)*ll 11l- t-rr.u J_rI Ayl ;rir',.., , cll j ,-!9 .,jL_=J ,,ti.. ,i 1i- - .+i;*. .J.c ,"rJ(r L LlJ Gt--- 3"r L Y | 4+s Cr.:-r ;l .i+:l yJ +tl I cJ 1 e r JiS ,.S.rl , - ,i i l^ e#\3ll_,dY '-orll e:v +15 rJ. d!: !6-l- [.:-i .rls r.Jt-j c;l j .-'iii-'i )J r'"hiyl :\ rx$l , i\j Jl r\

A 4l .9 re Jl ctlS .tLt .rl .,.i JKh G:ljJt clll A rJ).i .t ;)iJ olr.o +ll .,,JL\1 :ES J:l .;iJSiJ.j-., a-n crr; il 4J.,.j 4lJl t,i ;l ,i s.ur.. .-rl J"Jilr 6JrJr ,\:-.,.':, .,ll lr,alr €]J .,,1(i,r+lr r:.0 ;+ii- -illJ i+X: gX

eLc ;tll 6--.t iF. JiJ:c .JLill ,-+ alll I e U- "xsly J ,,i.,(l eJr\,rJl tJL

el-b^ll cl J- ' . ''i(J ai* lillJ .l* '"' 'J gg i-c: 6..11 ;11+:+ cr- dLilr j .,^+Jr . , 'r-,,.sJr ,,-J+r.r: U-19,rlJl

-Ei -l rr ' l

4+ 4lll ' ;L\

To Mr Jacob Cool

Praise be to God, and blessing and peace be upon His most noble messengers and on thosewhofollow then till the Day of Judgement.

This letter is to our honoured, intelligent, courteous, learned, Christian, Dutchfriend,Jacob Cool.

Your kind letter has reached us and we rejoiced at it uery much, and we haue understoodwhat you wrote and what was contained in the letter. In connection with what you saidabout t,he manuscript that we had copied of the book entitled al-Musta'T nr, I d;d my bestto do this in a perfect way, as will be clear to youfrom the translations in it, as I hauetranslated most of the names of the simple medicines into Spanish bi-l-a'jamiyya [sic],which should facilitate your comprehension o;f it, and I haue collated it with iiotir, ,opy,an additional one to the copy of ourfriend, the doctor. From these two we corrected yourcopy. I was induced to do this by two mdtters. The frst is the word of our prophet, prop b,upon him : 'Cod has mercy on whoeuer performs his craft to perfection."o Tlie ,rrond is thatI assumed that you wanted, mostly, to haue these included, and that it is only appropriate toindude something in it which is correct. As to what you said about informotion-irirrningthe land of origin of the author, I think this is Andalusia, but I was unable to ascertai, fio*

9. This line in Spanish, on rhe reverse side of the paper.ro. Quotation not found in the canonical collections.ir. This particular manuscript of al-Mas'udi's Muruj al-Dhahab cannot be identified. Golius had apparenrlvwanted to acquire it, but the copy was stolen, as is clear from Ahmad b. Qasim's letrer ro colio, of zgt.b.,rr.ir624. Therefore the identi f icat ion by Wiegers (A Learned Muslim Ac(luaintance,p.6J, n. r55) of the stolencopy of the Muraj al-Dhahab as mentioned in Ahmad b. Qasim's l.tt.. of z9 February rc24, withleiden Cod.or. tz7 seems impossible. His claim is silently corrected by van Koningsveld, al-Samarrai and Wiegers (n. 6above), P. $, n. r r r. Leiden Cod. or. rz7, a medieval manuscript which does indeed contain prrt (nalm.ly th.second book) of the Muruj al-Dhahab, is in so far remarkable that it has, among r.ver"l other simil", notes, ̂

T H E L E I D E N M A N U S C R I P T

which period. This may become clear from the book 'The Golden Meadouts' , since it u,dssaid at the beginning of the Kitab al-Trb that he presented it to so-and-so, and;from thatdate it is clear in which period the person to whom it was presented liued." On anotherday I will writefor you the answer to your Sheikh, the learned professor Erpenius, Godwilling, andfrom Him we ask to enlarge your blessings and to bring you back safely to yourcountry. Thus written in Marrakesh, may Cod protect it, on twelue Rabt' II of the yearone thousand and thirty-three, and in the Christian era on the second of February of theyear one thousand six hundred and twenty-four. Thus written by the seruant of the HighAuthority and the prisoner of his sins, Ahmad b. Qasim, may God treat him with kindness.

Preliminary remarks on the basis of this letter:

- allatT ansakhna 'that we had copied', and which was therefore a manuscript which

was apparently not copied by Ahmad b. Qasim himself. The owner of the firstoriginal was the medical doctor, whose name is not mentioned in either letter.-'translations into Spanish'. These were apparently made by Ahmad b. Qasim.- 'I have collated it with another copy, an additional one to the copy of our friend,the doctor.' This means that there was another, presumably unrelated, manuscriptcopy of the Kitab al-Musta'TnT around and available.- 'From these two we corrected your copy.' From this sentence, in which thegrammatical dual is clearly visible, we may conclude that at least three copies of theKitab al-Musta'TnT are involved: the copy belonging to the doctor, the other copy,and finally'your copy' (= Golius' copy).-'^ay become clear from the book

"The Golden Meadows"'. But the copy of that

manuscript, which may have contained valuable information on the provenance ofMS Leiden Or. r5, was stolen (as we read in Ahmad b. Qasim's letter to Golius of z9February 1624, quoted below), and that information must now be regarded as lost.-'the beginning of the Kitab al-TTb'.The meaning of this passage is unclear, unlessal-Trb is to be understood as al-Tibb, and then refers to the Kitab al-Musta'rnT.

On roJumada I ro33 (Thursday z9 February fi24) Ahmad b. Qasim wrote againfrom Marrakesh toJacob Golius in Sa{i [Fig. z9] :"

Alsefior jacob gulflamenco en saf

oLt .=J-c e\JlJ ;)-ll3 4ll i^-ll

c!\ls dJ+.r-rLi r=l lris ol L.)+ C_i .l:l C: $ C csJt l.o,-ll iJ,ill :sYl :sl;,]l 4J1 r^- ,r Lle$ lChC+ l l . ! , dJ ley . : \ :S i t ^ . " ' J ( : \ - L l o t ^J Jo l j r j , " , ( i a ' g \L r L i . ' r J . .+1JLq , " ' - l i _ i

birth note on its title page (added later) for a boy Abd al-Malik born on a Tuesday rzJune or r7 or r5 Rajab

(year not clear, but after 988/r58o, the date mentioned in a previously written birth note). May we conclude

from the double calendar notation that this manuscript was once in the possession of Moriscos such as Ahmad

b. Qasim, or were Christian dates used more often than one would think? In his first letter to Golius Ahmad b.

Qasim also refers to both calendars.

rz. Leiden Cod. Or. rzz8, no. 12.

Fig . 29 .

Letter from Ahmad b. Qasrmin Marrakesh toJacobus Golius

in Safi , dated roJumadd I ro33

A.H. (Thursday z9 February

r6z4 A.D.). Leiden, University

L ib rary , Cod. Or . rzz8 ,no .3z

J A N J U S T W I T K A M

e1+ rl ,",-ri, .rr,.-ll ..1l-jl l- .,s, :)[Jl or-o d eerrLt dhi 4L\-: .r+l .r ' , ]Jl ar; ct-.

;t ;l:.lE "+ jYl :r+.; +.,1; "+ gll ,. ''l(r crJj l-.,,L\l {-c.1i .rs rr+:ri*ll .i\rsll ,r\ ' j .sJ-c'JJSilI +LSll dl:+ :a;4j- 4jl* 3;Yl r- L.JJS. kl .tt J xl -++ttt ljst- + Jl a--JtLl:;,r Ol ,-'1lS-l _il U-^$l J"'J 4bi.- Lir;o drrc ctl J ,"'l<.t , " 'i(, ..!^lJl ri-c &L ., ,,ti.

J. -Ful .,-i 4*JFl .Al-r- .5s , , ''i(_r t)*JlJ ell+ ce .J ,",d_l &sSYl .JLJ .j-i yl a-,.f.4j- -ill; O+X_i J:)\j rLc' dr J3)l dil-+ ts

.,lJleli"ll.-:-

3-li;:^-=t!4 l l l ' ; L l

To Mr Jacob Cool, Dutchman in Saf '3

Praise be to Cod, Blessing and peace be to His messengers.After thanking God, the One, the (Jnique, the Eternal, Who is not born, and to Whomno lne is born, and to Whom there is no equdl.r4 Ihaue receiued your letter and I wasglad to see how well you can write, and I haue understood its content and meaning. Whatyou wrote about the 'Book, of the Colden Meadows' , namely that it had not arriued withyou and that the Arabs had taken itfrom its bearer, well this is their habit in this land andin this dfficult time.'s It gives me pleasure to receiue the news that it is your purpose toacquire Kitab al-Musta'TnT on the Art of Medicine. I am writing to you once more aboutthis because I think strongly that the copy , which is with its owner, the doctor, was writtenaboutfour hundred years ago. It is a maruellous book and it is uery much appreciated bythe Muslims. You mentioned that you had with you a text'6 on dstronomy lliterally:

'the

declination of the Sun'1. If it would be possible to translate this into Arabic or Spanish,and then send it to me, that would giue us pleasure. Creetings.Thus done in Marrakush, may God preserue it, on ten Jumada I of the year one-thousandand thirty-three.The Servant of the High Authority,Ahmadb. QasTm, may Godbe kind to him.'7

r 3 . This line in Spanish is written on the reverse side.

14. Qur'an rrz.

15. It is probable that another (incomplete) copy, of the second volume only, of the Muruj al-Dhahal,

in Maghribi script later came into Golius' possession anyhow. It is now in Leiden, Cod. Or. rz7. As I have

already said, it is hardly possible that this volume was the same as the copy of the Muruj al-Dhahab which .t .r.

stolen in Morocco while being brought to Golius.

16. Wiegers, A Learned Muslim Acqualntance, p. 65, translates hurufan mahfuzatan as'mnemonic words'. ,rtranslation which he connects with work bv Ahmad b. Qasim on al-Risala al-Zakntiyya (p. 66). Van

Koningsveld, al-Samarrai and Wiegers further elaborate on this in their historical study of r997 (pp. ++,s).r7. The letter has been fully translated by'W'iegers, A Learned Muslim Acquaintance, p. 65. Wieger's referetr...

to Ahmad b. Qasim's letters in manuscript Leiden Cod. Or. rzzS are confused, but his captions on pp. 8o, I j -a

are correct. This letter was previously edited by Reinhardt Dozv in Catalogus Codicum Orientalium Bibliotheta,Academiae Lugduno Batavae, vo1. r, Leiden r85r, pp. r6r-2, and later once more edited and translated (r:rt ,

Dutch) by Martijn Theodor Houtsma, Llit de Oostersche correspondentie uan Th. Erpenius, Jac. Colius en l,r't'Warner. Eene bijdrage tot de geschiedenis uan de beoefening der Oostersche letteren in Nederland (Yerhandelingen.1.rKon ink l i j keAkademievanWetenschappen,Afdee l ingLet te rkunde,vo l . r7 ,W.3) ,Amsterdamr887,no . -s .

' l

L:-d-sfu:^.^Jl*\.-**_?*rf , * ;J 1 *tsJI *r l i-

u ' /

r, L:*1;l-*llp.l ;t ye,lrr -s.1,, LJ f ;-,t-t*

ry, Jr+.*Il*rLr + * &'" *1*'1-:-,1S'G;arpLJl + ".*' rtJlJ-s lf * 4, qU 1r

,u-sb;tLidluCtu-\iV. t \L$"\t-ll / > +-

, ) \ \ L "I' i r f $tFl 197 e---- +j {,.\

{ ---- ' }

\ n l , '

i I r l r j r l t \ , 1 \ 1*,1 lg . $** , 3

'.irl,aJi.l ar' r-rs,:

-{-. ,Atl-&.} :

J A N J U S T \ v I T K A M

Preliminary remarks on the basis of this letter:

- The present owner of Kitab al-Musta'Tn7 is, again, said to be a medical doctor.- The book is said to be held in high esreem.- The doctor's manuscript is said to be four hundred years old.- Nothing is said about an incomplete or damaged manuscript.

The first two details can be taken at face value, but not so Ahmad b. Qasim's noteon the antiquity of the manuscript. If this applies to Leiden Cod. Or. r5 (which is byno means sure) it can only apply to the old part of the manuscript, which is certainlymedieval.

From the two letters of Ahmad b. Qasim we can gather that Golius'appetite forKitab al-Musta'TnT was whetted by the first letter. To this he may have answered thathe wished to acquire a copy of the work, and we can assume that his interest wasfurther aroused by the four-hundred-year-old manuscript mentioned in the secondletter.

Physical description of the Leiden manuscript

When leafing through the Leiden manuscript the reader is immediately struck by thefact that the physical constitution of the volume is diverse. At first glance it is evidentthat a medieval manuscript of the Kitab al-Musta'Tnt had become very incomplete inthe course of time, and that this incomplete copy (still containing ror folios of oldpaper) had been supplemented by 34 newly written paper leaves. Old and new leavestogether more or less make up the entire work, although only a full review of allexisting witnesses of the text can con{irm this. A clue for the dating of the repairedpart of the text is the watermark in the new paper: an ornamented and crowned pot,with a crescent on top.

Although Briquet and Heawood give many pots in watermarks dating fromthe latter part of the sixteenth and the early seventeenth centuries,'8 none of theseis similar to the watermark in Leiden Cod. Or. rJ, nor is the identification of thewatermark as given by

'Wiegers correct.'e In Heawood,'o as in Briquet, several of

the pots have peculiarities very similar to those in the watermark in Leiden Cod. Or.r5, but none is positively identi{iable with it. It shows, once more, that the referenceworks on watermarks can only be used as rough guides, not as precision instruments.The world of papermaking is infinitely more complex than can be described in suchbooks. Personally I only use them for the purpose of corroboration, or refutarion,and always in order to have an additional argument.

r8. Charles M. Briquet, Les jliyanes: dictionnaire historique des marques du papier dis leur apparition uers tzE.jusqu'en t6oo. A Facsimile of the ry07 Edition with Supplementary Material Contributed by a Number of Scholars,ed. Alan Stevenson, Amsterdam, r968.

19. 'Wiegers,

A Learned Muslim Acquaintance, p. 66, n. r59, where there is a reference to Heawood No. 3563.Heawood's No. 3563, however, is not dated 16zo, but 16oo. The letters PR in Heawood 3563 are also atvariance with what the watermark actually has : HL, as can been seen clearly in fol. r 3 5.zo. Edward Heawood, Watermarks Mainly o.f the tTth and t\th Centuries, Hilversum r95o.

8z

T H E L E I D E N M A N U S C R I P T

Wiegers was the {irst scholar to point out that the supplemented leaves in theLeiden manuscript of the Kinb al-Musta'TnT are in fact written by Ahmad b.

Qasim himself. For this he convincingly adduces palaeographical evidence, bv asimple juxtaposit ion of Ahmad b. Qasim's letter in Or. tzz8,no.32, to a randomsupplemented page in Or. t5." lf we compare this with the words by Ahmad b.

Qasim in his letter of z February fi24 (ansakhnahd,'we had it copied') it is clear thatthis cannot apply to MS Leiden Cod. Or. 15, which Ahmad b. Qasim did copy,but to another manuscript of the Kitab al-Musta'rnr, which must have been copiedby someone else. In the dme immedlately preceding Golius' acquisition of his copyof the Kitab al-Musta'TnT there must have been at least three manuscripts of the textaround.

It may also be assumed from this that the Kitab al-Musta'TnT was not a very rarework.

'When Ahmad b. Qasim writes to Golius on zg February fi24 that the Kitab

al-Musta'tn|'is a marvellous book and... very much appreciated by the Muslims',this implies a certain popularity and availability of the text. If Ahmad b. Qasim,in that same letter to Golius, mentions four centuries as the age of the manuscript,and if he is writing about the manuscript which is now Leiden Cod. Or. r j, this canonly apply to the part of the text on old paper. Ahmad b. Qasim's estimate is notmeant as an accurate dating, but even if the four centuries are an exact indication ofthe manuscript's age in t624, the old part of the manuscript would date from around1224.22 I do not wish to exclude this as a possible estimate for the age of the old,original parts of the Leiden manuscript.

The division between the old part and the supplement on new paper is as follows:

New paper

folios r-r3r 6r 820-23

3 o44488 3

Old paper

folios r4-rsr 7r 924-29

3r -4345-4749-828+-96

zr. Wiegers, A Learned Muslim Acquaintance, p. 66. .Wiegers'remark

on Golius completing the manuscript

is wrong. A rather puzzling remark about Or. r5 was already made by Dozy, Catalogas, vo1. r, p. 16z,

n. r. It wouid imply that Or. 15 also contained copies of other books. This is not the case at presenr, nor,

in all probability, was it the case in Dozy's time, since the previous printed catalogue of the Leiden Library

(Catalogus Librorum tam Impressorum quam Manuscriptorum Bibliothecae Publicae Uniuersitatis Lugduno Batauae,

Leiden r716, p.443, no. 8o9) mentions the text as the only one in the bnding. In the descript ion of Or. r5

(DeJong and De Goeje, Catalogrzs, vo1. 3, no. 1339), p.249, n. r, we f ind the further specif icat ion that the

supplemented leaves were probably made Golii jussz, at Golius' request. Dozy, DeJong and De Goeje did not

realize that Ahmad b. Qasim was in fact the copyist himself, but they all knew Golius'handwriting too well to

commit the error which'W'iegers ascribes to them.

zz.This 'four

hundred years' may be based on knowledge of part of Leiden Cod. Or. r5, which is now lost.

This may be a coiophon or some owner's note. 'W'e

do not know. It is also possible that such information wassomehow preserved in the (stolen) copy ofthe Murnj al-Dhahab.

8 3

J A N J U S T W I T K A M

97I O I

r r 6t z6r z 8 - r 3 5

98 - roo

r o 2 - r r 5

I 1 7 - r 2 5

r 2 7

Total: 34 fol ios Total: ror fol ios

The manuscript as a whole now consists of r35 leaves, which were numbered in ther97os by a Leiden librarian, in pencil, as r-r35 in the upper margin of each rectopage. All references to the Leiden manuscript in this article follow rhis foliarion.Roughly three-quarters have been preserved on old paper and one quarter wassupplemented by Ahmad b. Qasim on new paper. The structure of the old paperoccasionally shows a pair of parallel chain lines (e.g. fbl. 9j) and is evidently ofmuch older date than the newly added leaves, the early thirteenth century havingbeen mentioned already as a possibility. The old paper is of a somewhat smaller size(26.7 x rg.2 crr' and sometimes less because of wear and tear) than the new paper(zg x r9.5 cm). The volume is now bound in a European-style full parchmentbinding, which may mean that the pages were only bound when Golius broughtthem back to Leiden. He did the same with other piles of paper which he collected,such as the Leiden Cod. Or. ra.

All the old leaves, anda number of the new ones, have been {ixed on guards of

Paper, with the result that no bfolium of the old part of the manuscripr has beenpreserved. If there were still bifolia intact when Golius acquired the manuscript,these must have been cut into single folia for the purpose of the resroration schemewhich was applied to the entire manuscript. These guards of paper of about r cmwidth have been made in such a way that they constitute quires and give the bookas a whole a quired structure. A conspectus of the quires of Leiden Cod. Or. r5 isa fo l lows: I ( r -8) , f t ( l - r+) , I I I ( r5-23) , IY (24-3r) , V (32-9) ,VI (4o-7) , VI I(+8-ss) , VI I I (s6-63) , IX (a+-zr) , X ( t " -g) , XI (8o-7) , XI I (88-e5) , XI I I(16-ro3) , XIV ( ro4-r r ) , XV $rz- tg) , XVI ( izo-7) , XVI I ( rz8-35) . There isan evident preference of Ahmad b. Qasim, who did the restoration work, to makequires of eight leaves. Bifolia of the newly supplied paper were kept inract as far aspossible: quires I and XVII consist entirely of the new paper and these thereforeconsist ofbifolia. The fact that these quires had to be fully replaced shows, again, thatthe most vulnerable parts of a book block are the beginning and the end.'3 All theother quires are of mixed composition and only rarely do we see bifolia in the newpaper in such mixed quires (fols. ro -r3, r r_nrz and zt-z only).

23. One might even wonder whether the traditional Islamic horizontal storage of books may have been morelikely to aggravate this type of damage, which u'ould be less likely to occur in, or would be less detrimental to,books which are stored vertically.

84

T H E L E I D E N M A N U S C R I P T

Maintaining the order of the leaues

That several owners of the manuscript have tried to maintain the correct order of

the leaves becomes clear by their numbering of the leaves or Pages. Apart from the

modern foliation, the Leiden manuscript shows five older systems of foliation or

pagination, only one of which was made on both the old and the new paPer. This

i,ttl foli"tion was probably made by Ahmad b. Qasim. He did not number the folios

of the introductoi y part, since the correct order of the leaves there is secured by a

system of catchworJs. The double-page openings with the jadawil, the {irst being

folr. ,rb-, 3^, arenumbered by Ahmad b. Qasim from the {irst one onwards. These

numbers ft-lzz) are writren, almost without a break, in the uPPer right-hand corner

of each verso (i.e. right-hand) page, from fol. rzb onwards. Sometimes one can see

correcrions in mistakenly written numbers. The purpose of this numbering was to

counr the double openings of the text. 'Where

a double opening has text which may

be disregarded, this page has been crossed out and the opening is not numbered [fois.

4b-tai: Fig.3o]. This system of numbering is therefore related to the restoration

work on the Leiden manuscriPt.

The other numbering sysrems are older, since they can only be seen on the leaves

of old paper. Their relative age could not be established, but there may be a relation-

ship between some of them:

- A foliation with roman numerals on fols. r4a, rsa, r7a, r9a, at the bottom of

eachrecto page, sometimes somewhat centred. It shows the numbefs 3,4,6 and 8,

respectively. Nor. that numb ers r, 2, 5 and 7 are missing, but for these the old paper

*r, ,.plr.ed by rhe new, meaning that the foliation was added when the old paPer

part of the manuscript was (more) complete.: A pagination with roman numerals (z-r r) at the bottom of the page in the section

.o.rrir,l.rg of fols. z5a-zgb. There is no numeral visible on fol. z4a. Fols. z4 and z5

,r. "pp"..rrtly the leaves that were on top of the pile and have suffered most from

-.", *d tear. The pagination had been added when these leaves were on toP of the

bundle ofpaper.- A system of foliation in large ghuban numerals, written with a thick pen in the

upper margin of a recto page. The {irst of these that can be observed is on fol. z8a,

foliowed by a f.ag-ent of a numeral whlch is on fol. z9a; neither can be identi{ied

with certainty. The tens seem to be written in reverse order: 'o3'for '3o' (fol ' 43a),

'o4' for '4o' (fol. 5za),'os'for

'50' (fol. ooa). To this series ofghubarT numerals belong

the numbers within rhe section consisting of fols. 4za-65a [Flg. 3 r]. After fol. 65a this

thick numbering suddenly stops. Not all of these ghuban numerals are very legible,

but remna.rr, oi most numerals can still be seen. Several lacunae in the numbering

can be observed as well (If '4o'is on fol. 5za and

'5o' on fol. 6oa, either two leaves

are missing or the scribe has skipped two numbers by mistake). To find out whether

these lacunae in the numbering coincide with lacunae in the text is the editor's task.- Another system of foliation in large ghuban numerals, now written with a thinner

pen in the upper margin of each recto page. Between fols. 59a and 63a there is an

L.r.rl"p of the two ghuban foliation systems. The thinner one there is possibly meant

to be a correction to the thicker one. The entire system of these thin ghubarl numerals

Fig . 3o .The {irst ieaf of old (original)

paper of the Leiden MS of the

Kitab al-Musta'rni.-lhe text or1

rh is rec to page o f the o ld par t .

being onlv halfofa double page

with information (?), could not

be used by Ahmad b. Qasrma1-Andalusi in his resroration

programme, and was therefore

crossed out. Later, he supplied

the text in his own hand on fol.

zza . Le iden, Un ivers i t i L ib rar l ' .

Cod. Or. r5, fol. r4a

F O L L O W w I N G O V E R L E A F

F i g . : r .

The long passage in Spanish,

probably writ ten bi 'Ahmad b.

Qasim a1-Andalusi. The passage

reads: 'que

virtud pueden

criar las vmagines pues 11evan

a quien las haze y a quien las

adora al ynfierno y si fueren de

oro o plara no se pueden sen ir

dellas antes deven echarlas en

el fuego'. On the basis of this

passage, Dozy (in a note loosely

inserted in the MS) assumed that

the author of this note is of the

Jewish faith, apparently because

of the abhorrence of idols. Over

the text the numeral '39'has

been written in the older ghubart

foliation. Leiden, University

Library, Cod. Or. r5, fol.49a

B5

*

t**=*.. ; rAl rlbii-r '-*r f tri r L /

{VJ--,IS:+{Pi+'r ,tl t Lr .* t t,yt ,.

i

. + . -

: ' * - J

-? - 1 " :

''a ,i: ;1- : . f

. " f

I*J;' ;; }il, ;W;;.U :* t7fiS."r4lidr -J,.r;, I : i -"*.,"i '*' : g1*', .,.1- i*,,, i *$i t, 5X;*t* i, ,.;rrn---;"a"j g:i.a't +;.;.b,.-;ir #lr3$;1.* p t

F:T/ i

e.-f i".j*i!g.i:Ylrr k F n * - - *

+rr.' ir;{r,{r r.. i I inr"1;|p;; li;J/i+et ry:+ror r/S il**'; vfr#l*j#,,.d I *'--i5+ W b f ^b *, *y * i;'t* *,t' ,"it&n, ^va$.u|t, r, ;^ii;4'U;.tt+1/,1raa !U.?^;,.1'.,i.14.b.u-;*1.=y.:;s1c..r-*.,r'.[.--*-*tri-r*r

. tt. j {t r* il*l+u'; b.rb-i{rlfi.- *i i,p t*,*"* i*''.i+r lr4i,.r4trf".tl

t /1,-*.,r,

J f*3q: rgb i ib_r^trr i ri-. i 1 2t lr*t 7I,r;, i-t,ltL- ^.-r".rt.udr.r\i. aa1gd,! +. l.

t

v-*!ln*ir".,rt\isa3--.tl{Fj LA ff

"*' -'+ gA-4Lj t'-'v';$tI'j"Ffuldi il|!|JUqgFrft' Jl,r!s-.

,e o*--*'*'!Er=* *+**ur? ?:*,*I4ge1rdpt;

ii-jii;ti *,rirg" L*I*S.1 Ct{ }.*: 'r' L:'''U'L.'r'#

,t' '"'^'':;i' l .5,J l;:'!s

t'i,Wt s;+i4 {''

4 ll ',

I

I *.ttt i*nr-" g*91-r {s6r #l: .}rrj f a e. -Ll-l liJ I t'',t

', ,.-d b*,, qalq-re'- jJ-i'i.rr,l,r'. itullt l *cg4 c L . "LE y" r i r ' l r . * j l r rc . f t l lF - lF t l

: t*, F;n r",i e, Y *.-*' r, l--I i' t "'f- b tu L4+t? f,+*': : ;J'11.J' +' i tt

, f i4gr^.ia1o r'rl rl-.ili .cv+'. *,f,?:]rirlpllt.i.,HlF"J

,*- ']; ++it''ii- li; .;.o^ I r ": # l+t b11

J A N J U S T \ v I T K A M

can be observed on the old paper, from fols. jga to tz7a. On fol. 59a the thin ghubartnumbering has

'48', whereas the thickg/z ubarT nurnbering has '49' . The thinly written

large ghuban nurnerals continue until the last leaf of old paper ('tt7'; Fig. 33).This series too has lacunae. Its '5o' is on fol. 6ra, whereas its'60'is on fol. 7oa. Thismeans that somewhere one old leaf is missing (or otherwise that this is an instanceof ancient miscounting). Later on, the leaf that should have carried'9o'is missing.It should have been between fols. 99a and rooa, but these are numbered '89' and'9r'respectively. At the end of the section written on old paper, which is far moreheavily damaged than the part in the middle of the book block, several leaves wouldbe missing if the defective numbering were a decisive argument.

Ahmad b. Qasim' s restoration of the Leiden manuscript

The tabular structure of the Kitab al-Musta'TnT makes it possible for the copyist tocopy the work double-page by double-page. Both the Leiden and the Arcadianmanuscripts have a layout of seven rows per page, the upper row being used in bothmanuscripts for the titles of the columns, the six remaining rows for the explanatorytext. By maintaining the tabular layout of their copy according to the one in

the exemplar copyists made their task easier. In this case, this may be somewhatmisleading, however, since this open structure of the Kitab al-Musta'rni also allowsfor all sorts of additions by owners and readers, which may affect the integrity

of the text. That this has happened is clear from the numerous variants which canbe observed between the Leiden and the Arcadian manuscript, but I leave remarks

about variant readings in the manuscripts to the editors of the critical edition of theKitab al-Musta'tnT. Another aspect of the vulnerable integrity of the text in tabularbooks such as the present one is that, when one leaf is lost, the right and left sides ofthe opening are no longer a conceptual unit.

Notes and additions by a great number'4 of readers can be observed on many ofthe pages on the old paper, some very faded. They are often introduced with theword quf, the equivalent of nota in Latin manuscripts. It is tempting to identify thethickly wrirren nore on dawa al-khanazir (fol. 57a) with the hand which also wrotethe thlck gL ub arT msrner als.

On fol. g7a of the Leiden manuscript (a leaf of modern paper) we can observethat an extra or explanatory text was struck out at the bottom of the page. The textcontains an old medical judgement against the use of copper eating and drinkingvessels (ad *c;). Ahmad b. Qasim, the copyist, may have struck it out because hebelieved it was not part ofthe text. The Arcadian manuscript (pp.tg6-Z) has a similaraddition at the bottom of the p^ge at the same lemma, but of two lines only, with themention of Aristotle (as in the penultimate line of the Leiden manuscript). It shows,again, the enormous variety of readings. The philological significance, however, ofthese variants, notes and additions falls outside the scope of the present study.

A few more isoiated details in the Leiden manuscriot of Kitab al-Musta'Tn7 mav beobserved here:

24. I have desisted from speciS'ing and counting them.

8 8

T H E L E I D E N M A N U S C R I P T

- The title in the Leiden manuscript (over the beginning of the text, on fol. rb

[Fig. :21) is given in a somewhat confused way, certainly as comPared to the

Arcadian manuscript, which has a regular dtle page. Title and author are mentioned

in a marginal addition written over the beginning of the introduction:

.,$IJ*.)l g;t\ dl Jl- 'l i i,r-f +dU;4&-:r .,jc' L.";r;'iJl L.;':)l .,! .,]5i"*ll -rlis

4llL J^ijJl ; :^-:l1s'+.rrl ,# ulll+,Jil'L*il

Kitab al-Musta'tnt on simple medicines from what was collected and composed by Yunus

b. Ishaqb, B.k.lar.sh al-Isra'iltfor al-Musta'Tn bi-llah (the Exalted) Abu Ja'far Ahmad

b. al-Mu'tamin bi-llah

Over the beginning is also written in bold script:

4!tll crluill i .i.si,.*ll gUS

Kitab al-Musta'Im on medical simples

This may be due to a damaged exemplar. Textually speaking, the addition in the

Leiden manuscript does not greatly dlffer from the title in the Arcadian manuscript.- There are a few conjectures of Ahmad b. Qasim in the margin, which are not

confirmed by the Arcadian manuscript. For example, on fol. rb: .l-=., aitl the

Arcadian manuscript has haqTq, the reading in the body of the Leiden manuscript.

The same goes for the conjecture in the last line of fol. ib of the Leiden manuscript,

where the text in the body of the page is not so clear, whereas the copyist in the

margin conjectures: ki-6- a:!1. The copyist has also written under the last line: \-4g-,-2 1s2d1ng supported by the Arcadian manuscript, p. 5, line 6. There is a marginal

correction for eL.-it in the body of fol. roa of the Leiden manuscript :+:.; i JE

e ljl LgJil which refers ro a manuscript other than the exemplar. This remark may be

understood in connection with the other manuscript mentioned by Ahmad b. Qasimin his letter to Golius of z February 1624. The marginal conjecture al-zaj seems to

be correct. The originally written al-rajlaj does not seem to make sense, nor does

al-zajaj in the Arcadian manuscript , p. 24,line 4 from the bottom.- The first double page of the tabular text in the Leiden manuscript Kitab al-

Musta'TnT (fols. rzb-r3a) is infact the beginning of the text, with the six lemmata

amrr barTs, aqaqiya, aqhawan, abhal, anjara, anjaran. The Arcadian manuscript opens its

tabular rext (pp. 3213) with the same six lemmas in the same order. This continues

for a while, but then the two manuscripts {irst diverge and later reconverge again:

- Arcadian , pp. 34-�s : Leiden Or. r5, fols. zob-zra- Arcadian,pp.36-7 : Leiden Or. r5, fo\s. ztb-zza- Arcadian, pp. 38-9 = Leiden Or. r5, fols. r4b-r5a- Arcadian,pp. 4o-i = Leiden Or. r5, fols. i5b-r6a- Arcadian , pp. 4z-3 : Leiden Or. r5, fols. r6b-i7a- Arcadian,pp.44_�5 : Le iden Or. r5 , fo ls . r7b-r8a

89

Frg . _ t : .Openrng page of the Leid. ln_MS oirhe Kttab al-Musta'mt,

shovn'ing the beginning of theauthor's introduction, copied byAhmad b. Qasim al-Andalusi.Leiden, University Library,

Cod. Or . r5 , fo ls . rb

N E X T O P E N I N G

F i g . 3 3 .

The last available leaf of o1d

paper of the Leiden MS of theKi t ab al- Mu s ta' tnt. This page

was damaged with loss of

text. Ahmad b. Qasim hastily

restored it with modern paperand supplied the missing text.

See a lso the numera l ' r r7 ' in

ghubarl numerals written in

the upper outer margin. The

number was written when this

leaf was already considerably

damaged. Leiden, University

Library, Cod. Or. r5,fol. rzTa

F ig ' :+ .Marginal addition in the

modern part of the Leiden MS

of rhe Kimb al-Musra'rnr.The

marginal addit ion supplies

text which initially had been

overlooked by Ahmad b. Qasimal -Anda lus i . Le iden. Un ivers i t l 'Library, Cod. Or. r5, fol. rob

J A N J U S T W r r K A \ 1

- Arcadian, pp. 46-7 = Leiden Or. r5, fols. r8b-r9a- Arcadian, pp.48-9 = Leiden Or. r5, fols. rgb-zoa- Arcadian,pp. Jo-r : Leiden Or. r5, fols.4b-z4a- Arcadian , pp. S2-3 = Leiden Or. r 5, fols. z4b-z5a- Arcadian, pp. 54-S : Leiden Or. r5, fols. z5b-z 6a, andso on.

It is evident that the copyist of the modern part of Leiden or. r5, Ahmad b. easim,had become confused, probably because the beginning of the old part of themanuscript, the text of which he intended to complete, was still quite defective. Hetherefore skipped foI. r4a, whlch contains the old version of the text (now struckout), and replaced it by a new text on fol. zza (see the corresponding rext in theArcadian manuscript, pp. 36-7). The newly made text on fol. 4a was skipped aswell, since it was akeady available on fol. r5a on old paper.- All lemmata have been provided with Spanish translations in the margins. Thesemay be identified with the translations mentioned by Ahmad b. Qasim in his letterto Golius ofz February fi24- Not only did Ahmad b. Qasim number the openings, which was essenrial formaintaining order in what was then just a pile of loose leaves, and not only did hetranslate the lemmata in order to help his young friend Golius about whose pro-ficiency in Arabic he apparently had his doubts, but he also improved the legibilityof some of the illegible and faded text by writing over it, and he added text inside therows and in the margins of many pages [Flg. :+].- on fols. ro8, rog, r27, Ahmad b. Qasim has partially repaired the old paper bysticking some modern paper on the upper side of the original, and substituted thelost text on these strips in his own handwriting [Fig. 33]. This restorarion lacks rheneatness with which he usually worked, and we may, perhaps, assume that he was inhaste, and lacked time now that he had come near the end ofhis restorarion work.

The difrerence is striking. In the beginning of his reconstruction work on thetext he replaced entire pages when this was not necessary (compare fols. r4a, 2za),but now he may have had a deadline to keep. From fol. try onwards the Leidenmanuscript shows, mostly at the top of the pages, considerable damage with theconsequent loss of entire lines of text. Apparently this lost text could not be sup-plemented. Ahmad b. Qasim's two other manuscripts may not have given himthe text he needed to complete rhe copy for Golius. Bur, again, lack of time mayalso have been the actual reason why his ambitious restoration programme of themanuscript was not entirely completed. Another consequence of lack of time maybe seen in the last quire, which is entirely in the hand of Ahmad b. Qasim. His hasteto finish the copying was apparently such that he forgot to use one rop row (fol.r3za) for the title of the page, a mistake he could only correct by leaving the borromrow empty. On fol. r3 ra he even forgot to make this correction and filled n sevenrows of text, but then struck out the text in the bottom row. These are not essentialmistakes, but Ahmad b. Qasim, who by now had almost completed what must havebeen an extensive editing job, made them, one assumes, because he was in a hurryand had no dme to replace the faulty leaves nearly.

9o

,, ],u Lr 1r-ffi- 1$ br --+r .^-'?, rJ, o. 2,y,r1 -,. -.; - -

r r r l l ' i r g * L l r i - l, ",; .. ri -Lqi{*H€'jbff""i*r': r:m'#gr,.1VJf 't.

ra , ̂ - / ,.1 (,

,= j!1r;,

1 ! . . ! +

o r ,e * * - * ; - i l l t * ' * \ - a ; t * ' JE* - i - ' . ; * : r - *+ "1 : : : i i g ' ; l * L ; ' - ' ' - r : r - s > " i ' a ! - -

' r ' t ' l ' I t

* ! : * : . . , ' : s ! . 1 , . o 5 ' - : o - - 1 - - l , L - .

o - ' : * j x o 1 * .

J - ' * - - { * ' - - * t 1 ' " - J ; : t i

. 1 ,1 * 1 { * r * ! i l a - . ; . , . . ; ; ; " b * i * , : " t ' ; t * i * - i * ! ' : + i ' ; ! t ' r } ' ' o

{ r i . , . . , , x : , 1 . : : + . ; ; . * $ l - i ' " X * . " 1 , e ' i ' o .

..";;

*;& .o

l . , r i - *U '1 , - . - * - - . * : I , o . - i ' * i * * } r - . - r - i . -Le ' , i , - - r . o>-

;u *.--,r *- ,:*'- : i .., ,,

,':, -t: .* 'r,:

:1"--,.. ' - '',-

*** yag*; * ** ;p''.tr- *,'x-iiltu**'*t J g* '1 *}1*: t*r I o j '-' : *

, . : a * Y l * 3 ' : b . ' ' ' - * , * - ' r ^ \ *

\J ' * '

' , "o **nt ' - t- +-r *-a='-L . . t" - L * '* | . r * ' \*- ' \ ' :

{."\$*rJui* "i*

iruly, . ',: .- o. - -=>^.sro->- > q *** ' ! *J i 'c )* - , ' * r . r - a * ' i J^^ '

T:* - 'J * - \ - r

- . 1 \ , . - * : : t - \ r 1 * - . i l - * * t " , L i r * " 1 : . * * ' \ - - L u " "

..4

-, ri.A

'T: :t4 ajrb- ;"""J,u ;1F

\lF'-'d -**.-ei,u:. i , f**jl L*:- ,r;*tj_&gtb

, n:::l

i :i

!'|

,.Ft

|�-,cb A .i. .* .r lr y*J 1.,;.i , ;, i *' {, rU Lr- t -, * ;d*"} l,'inn"J t riSS j t"t:.. O * ; ;,t ;

*k &Cf.;i'lrJ,"r:1; ; : "' ;f. ;5,:,: T.'fi,i'd' IXI lf.';,,ff ,::T' fi

1:.

f.; -+ t r * i rr,fti:ci5 ury t# j.rlF l+r. r# r- l1:fi Li ri i *_r', * 4

*.rL rja* l: ;ryf ij*,:;i i "., ., , sil;*+*o;q @ *ru"? i i ;.: ;_-

*' :m:i*t} Li: f ;. :"? * ii"'r"sF$* u sL -, ir'* *1;l., : g r i ila

.,' :. i, ::y* Xj:yy r,-';"s r*' *---ii ; = Fi'; r * ;, ; *Itr:r v kr*<' ,Yu,tl t

1* ul|*# .r,nsL*i:gl:**lg*r

{;;.;r{r-sf *gu4J;, fu_-'J' j nrrd*r**r*r*s*!*t"-* x;xi-r ;*,'&-r-r*,-, q *;-g{;e

' i

J A N J U S T W I T K A M

Concluding reffiarks

In addition to the numerous new details that can be gathered from a close inspectionof the Leiden manuscript, the following conclusions may be drawn. The'earlierview that Leiden cod. or. r j was a copy of the Kinb ar-Musta'TnTmade by Ahmadb. Qasim at the request ofJacobus Golius, on the basis of a single -".rrr.ripr, is nolonger tenable. A newly discovered letter by Ahmad b. easim in the Mairchesterlibrary shows that at least two exemplariawere involved, one to be copied from andthe other for advanced collation work. Traces of both can be dir..rrr.i in the Leidenmanuscript.

Notwithstanding the new information, Ahmad b. Qasim's involvement in pro-curing a coPy for Golius has not become any clearer. It has, if anything, becomeharder to understand. His was an exhausting editorial task, for *hich time almostfailed him. An analysis of the traces of several pagination systems in the Leidenmanuscript shows that the old part, apart from being incomplete at the beginningand at the end, seems to have had lacunae in the text.

Finally, the description of the Leiden manuscript of the Kitab al-Musta,TnT,hereundertaken for the first time since Dozy's remarks in r 85 r and the description of r g65by DeJong and De Goeje, together with newly discovered material in Manchesrer,shows that the Leiden manuscript is one of several codices that clearly need to playa key role in a critical edition of the Kitab al-Musta'rnr.It is ro be hoped that sucha project wiil receive an extra impulse now that the importr.r, -r.rircript of theArcadian Library has come to light.

lI

94

I B N B A K L A R I S H , S B O O K

O F S I M P L E S

M E D I C A L R E M E D I E S B E T W E E N T H R E E F A I T H S

I N T ' W E L F T H _ C E N T U R Y S P A I N

edited by

Charles Burnett

T H E A R C A D I A N L I B R A R Y

in association with

O X F O R D U N I V E R S I T Y P R E S S

, ,. , - ' . ' , ,.,,, ' ,,n..lrt rhe Arcadian Librarv ln association rvith oxford l,niversitv Press

" . ' ' : : : ' : . : ' : - The Arcadian Ltbrar l ' , London, is an af i l ia te of

The Arcadian Group S.A., Case Postale 2ee, r2rr Geneva rz, Switzerland

The .4rcadian Library is a priuate,-family library specializing in the historic influences of the Leuant upon

Europe. Its aim is to promote this cultural transfer through exhibit ion and through publication

Oxford ljni.versitv Press, Great Clarendon Street, Oxford oxz 6op

Oxford University Press is a department of the University of Oxford.

It furthers the University's objective of excellence in research, scholarship, and education by

pubiishing worldwide in Oxford New York Auckland Cape Town Dar es Salaam Hong Kong

Karachi KualaLumpur Madrid Melbourne MexicoCity Nairobi Newpslhi sh,.-h.i

Taipei Toronto'With

ofiices in Argentina Austria Brazil Chile Czech Republic France Greece

Guatemala Hungary Italy Japan Poland Portugal Singapore South Korea Switzerland

Thailand Turkev Ukraine Vietnam

Oxford is a registered trade mark of Oxford University Press

in the UK and in certain other countrles

Published in the United States by Oxford lJniversity Press Inc., New York

Copyright e The Arcadian Library zoo8. The moral rights of the authors have been asserted

Database right The Arcadian Library (maker)

First published zoo8

Al1 rights reserved. No part of this publication may be reproduced, stored in a retrieval system,

or rransmitted, in any form or bv any means, without the prior permission in writing of The

Arcadian Library, or as expressly permitted by law, or under terms agreed with the appropriate

reprographics rights organization. Enquiries concerning reproduction outside the scope ofthe above

should be sent to The Arcadian Librarv. at the above address

You must not circulate this book in any other binding or cover and you must impose the same

condition on any acquirer

Illustration credits: the illustrations in this book have been provided by institutions each of which

is identified in the caption to their illustration. Copyright restrictions in favour of each institution

apply. O zoo8

British Library Cataloguing in Publication Data (data available)

Library of Congress Cataloguing in Publication Data (data available)

rsnN 978-o-i 9-9543c�6-9

Project management by RobertJones, Series editor Alastair Hamibon, Academic editor Charles Burnett

Copy-editor Bernard Dod, Designed and typeset by Phil Cleaver and Emily Berry o-f etal-design.com ltd

Photography at the Arcadian Library: John Stone and Karem Ibrahim

Printed in Creat Britain by Cambridge (Jniuersity Press, on Stow Book Wove, free from chlorine ( TCF )and OBA, and resistant to ageing in accordance with ISO 97oo6Typeset in Monotype Arcadian Bembo

T A B L E O F C O N T E N T S

The Contr ibutors

Pre tace;'i ' Charles Burnett

Ibrr Baklarish's Kitab al-Musta'Tnl: the historical contexr ro rhe discoverv of ai r e \ \ ' n l a n u s c r l P t . . . : . t 5: '1 -1na Labarm

The manuscript transmission of the Kitab al-Musta'TnT andthe contributions ofrhe Arcadian Library manuscripr 27;.1'Jodlle Ricordel

Ttrr'ards the study of the Romance languages in the Kitab al-Musta'Tnr . . . . . 43;, 1' _l ttrn Carlos Villauerde Amieu a

The Leiden manuscript of the Kitab al-Musta'TnT 75;1'Jat Just Witkam

The Syriac words in the Kitab al-Musta'TnT inthe Arcadian Libraryn r a n u s c r i p t . . . . . 9 Jt ^ f T t z l'1)'�Ge)JJrey Knan

Ibn Biklarish-Isra'ili ro5i,y Dauid J. Wasserstein

Ibn Baklarish in the Arabic tradition of svnonvmaric rexrs and tabularpresentationsr, y Emilie S au age- Smith

The zoological-medicinal material in the Arcadian Library manuscripr . . . . . r331,1, Anna Contadini

Brbl iography r j3

Index 16r

Plates: the Arcadian Library manuscript ,pp. r-52 . . . . 16 j

T H E C O N T R I B U T O R S

Charles BurnettProfessor of the History of Arabicf Islamic Influences in Europe, Warburg Institute,

Uniuersity of London

Ana LabartaProfessor of Arabic, (Jniuersity of Valencia

Joelle RicordelPriuate scholar, Paris

Juan Carlos Villaverde AmievaProfessor of Arabic, (Jniuersity of Ouiedo

JanJust WitkamProfessor of Palaeography and Codicology of the Islamic World, Leiden (Jniuersity

Geoffrey KhanProfessor of Semitic Philology, (Jniuersity of Cambridge

DavidJ.WassersteinProfessor of History and of Jewish Studies, Vanderbilt Uniuersity

Emilie SavaEe-Smith

Professor of the History of Islamic Science, (Jniuersity of Oxford

Anna ContadiniSenior Lecturer in the Arts and Archaeology of Islam, School of Oriental andAfrican Studies, London