François Déroche, with contributions by Annie Berthier, Marie-Geneviève Guesdon, Bernard Guineau, Francis Richard, Annie Vernay-Nouri, Jean Vezin, Muhammad Isa Waley
A text might be copied for various reasons, and a student transcribing a treatise required for his studies would not proceed in the same way as a scribe commissioned by a prince to copy a fine volume of poetry. The final appearance of a manuscript very often reflects these different requirements, particularly in terms of the care given to the presentation of the text on the page. Indeed, differences between a working copy and a manuscript of high quality were already apparent in the preparatory stage, notably in the ruling of pages. Not all manuscripts were ruled; when close examination leads to the conclusion that no specific measures were taken to guide the writing, the codicologist should try to ascertain whether the copyist may have used a substitute - the laid-lines of the paper, for example - to organise the lines of text or whether he carried out his task without overly concerning himself with evenness. It sometimes happens that pre-ruled lines were disregarded. In one Paris fragment (BNF arabe 383a) the parchment was ruled with a hard point, but the lines were never used. Another copyist re-cut the parchment and then transcribed the Qurʼān onto it without regard for the ruled guides.[1]
Apart from its usefulness in producing straight lines, ruling could help calibrate the length of a text. In a chapter of the Fihrist devoted to poets, Ibn al-Nadīm provides readers with a way of gauging works to be presented: ‘If we say that the poetry of a certain man fills ten leaves [𝓌araqāt], we mean Sulaymāniyya ones holding twenty lines-that is, on each side of the leaf [fī ṣafḥa al-𝓌araqa]ʼ.[2] It is possible that the ruling methods described had already been standardised to some degree in the fourth/tenth century.
The decision to draw ruled lines is immediate proof of a scribeʼs concern for page layout, a concern which in fact surfaced quite early in Islamic manuscripts, given that some copies of the Qurʼān written in the so-called Ḥijāzī script - datable to the late first/seventh century or early second/eighth century - have retained the marks left by the hard-point tool used for that operation.[3] The layout was still relatively crude, for the number of lines varies noticeably from one page to the next. But more complex ruling techniques, well known to copyists from other Middle Eastern manuscript traditions, were obviously mastered by Muslims beginning in the early second/eighth century or so, for they were soon able to produce sophisticated layouts. In a Qurʼānic fragment from Ṣan‘āʼ (Dār al-Makhṭūṭāt Inv. Nr. ɪ7-ɪ5. 3), the copyist arranged the writing in such a way as to produce geometric patterns on the page.[4] In another example, Istanbul fragment TIEM ŞE 362, the scribe began by carefully tracing the ruling pattern (see below); then, alternating the colour of ink according to a carefully devised system, he composed checkerboard-or diamond-patterns that stand out from the background thanks to their colour scheme.[5]
A scribeʼs intention to control page layout can therefore be determined through clues provided by ruling procedures or by unusual writing patterns. On the other hand, it is often difficult to ascertain the actual principle governing the copyistʼs choice of one layout over another. Formulas devised by the more skilful practitioners for their fellow copyists might provide useful information on the proportions that were favoured at any give time; but, as we shall see below, only one text of this type has apparently survived - and even then its precise significance remains obscure. Another potential approach, therefore, entails examining manuscripts and attempting subsequently to re-establish, by deduction, the rules that underlay their production
Ruling
Basic concepts
For anyone examining manuscripts, the most visible sign of a decision to organise the area of the page is undoubtedly the ruling, by which is meant the line or lines traced on the writing material in such a way as to allow the copyist to write as straight and as evenly as possible. Ruling was not, however, a preparatory technique used only by copyists; manuscript illuminators also used it to guide the execution of decoration, and binders often traced lines on leather coverings prior to decoratively stamping the boards.
Ɉustification of text blocks and measurement of dimensions
Although ruling is related to the codicological definition of ‘justificationʼ as the area of the page surface allocated to writing,[6] the fact that many manuscripts have no ruling requires that a distinction be made between ruled lines and the actual block of text. Nor - in manuscripts written in Arabic script - is the text block necessarily identical with the area within the horizontal and vertical ‘bounding linesʼ marking the four sides of the rectangular block allocated to text, which are sometimes divided into columns.[7] Horizontal measurement poses few problems, since overruns into the margin are rare. Manuscripts were composed in ‘long linesʼ with certain exceptions - primarily poetry and certain Christian Arabic texts.[8] Vertical measurement, on the other hand, is less straightforward. It was normal practice for copyists to write on the topmost line of the ruling where ruling existed,[9] and so the written surface is effectively greater than the ruled area, especially since the bottom line contains Arabic letters with descenders that extend below that line (although this phenomenon is less significant than top-line overrun). Hence when it comes to measuring the written area, it is recommended that the following method be applied: the distance separating the base lines of the first and last lines of a page should be recorded, and this number should be stated first, before the width. The dimensions of a manuscript whose writing area is higher than it is wide would be recorded as, for example, 25 x ɪ3.5 cm.; to write ɪ3.5 x 25 cm. would indicate that the justified area is greater in width than in height.
This way of measuring the height of a writing surface offers the advantage of providing a relatively reliable value for the height of individual lines, which is obtained by dividing the height of the total area (in this example, 25 cm.) by the number of lines minus one. Take, for example, a manuscript with eleven lines per page having a total text height of 25 cm.: 25 divided by (ɪɪ-ɪ) = 2.5 cm. per line. This figure constitutes the ‘ruling unitʼ. In fact, computing this value for two groups of palaeographically coherent series of early Qurʼānic manuscripts, B II and D I, has revealed the specific characteristics of each one in terms of dimensions.[10] Some manuscripts display the particularity of having been copied (though normally by a single copyist) in two or more different types of hand which may present perceptible differences in size; among the best known examples are Qurʼān manuscripts - especially those including a commentary and/or translation - and copies of the Burda, a religious poem by al-Būṣīrī. In such cases, the ruling unit of each hand should be noted.
Patterns of ruling
Patterns of ruling varied significantly across the Islamic world, especially during the period when parchment was used. Subsequently, the introduction of an instrument that made it easy to rule sheets of paper - the misṭara, discussed below - led to a relative standardisation of patterns. Where manuscripts had horizontal guidelines to regulate the distance between each individual line of text, scribes might write so that letters either straddled the line or rested on it. Such guidelines were not always drawn, however, in the Maghrib, the two vertical lines marking the text area were often the only reference points employed by copyists.
Any visual trace of the ruling operation had to be as discreet as possible, while remaining distinct enough for the scribe to see. Methods that left visible marks on the writing surface were known (lead point or ink, for example), but those that only left indentations were less noticeable and therefore greatly preferred (illus. 63). With the latter method, a single operation produced the same pattern on both sides of the writing material, the implement leaving an indented furrow on one side and a raised ridge on the other. In contrast, ruling done in ink or pencil had to be repeated on each side, this made the task more laborious, but had the advantage that one could modify the layout from recto to verso.
Study of a manuscript therefore calls for meticulous examination of any lines on the leaves, since the rubbing and pressure produced by frequent use of a manuscript can reduce the trace of marks ruled in relief, especially on paper.
Special lighting is sometimes required to detect furrows and ridges. As pointed out above, it may reasonably be assumed that the ruling followed principles derived from overall page layout, since it is unlikely that the number of lines, the height of each line (or ‘ruling unitʼ), and the relationship between the height and width of the writing area were matters left to improvisation. It is therefore important that the codicologist note all these features accurately, for they may provide a better understanding of the copyistʼs aesthetic concerns. It is nevertheless true that any trimming of manuscripts (during the process of rebinding or re-margining, for example) will have modified the dimensions of the leaves, thereby reducing the possibility of establishing the accurate ratio between overall format and text block.
Ruling in Arabic and Islamic manuscripts
As already mentioned, ruling was used by Muslim copyists as early as the first/seventh century. This makes it all the more curious to note the near-total absence of ruling marks on most Qurʼāns written in early ‘Abbāsid script on parchment.[11] The regularity of the lines nevertheless implies that copyists employed some method for guiding their writing even though no vestige of it remains,[12] which perhaps means that the marks were erased. One European treatise contains a recipe for a special ruling ink that could be erased with bread after use.[13] For manuscripts written in Arabic script, paper became the most widely used writing material at an early date. This may explain the popularity of methods that left an impressed or scored mark on the surface, notably the misṭara or ruling frame.
Scored ruling
A hard point[14] or fingernail[15] could be used - with or without a straightedge - to score writing surfaces as different as parchment and paper. To the best of current knowledge, there is no indication that Muslim copyists took advantage of a hard pointʼs potential for scoring several stacked sheets of paper at once, the top leaf receiving a rather deep furrow whereas the bottom one would be less visible; this was common practice in Europe. On the other hand, the tools used for scored ruling permitted a wide variety of lines, including circular shapes for decoration. This meant that copyists enjoyed great latitude. To establish a typology of such patterns would certainly be of interest, but research in this area is still at a very early stage so far as manuscripts in Arabic script are concerned. Another, rather crude method of scoring might be noted in passing: it involved folding the leaf along the vertical line of the outer margin. Ruling done with a hard point on parchment was frequently accompanied by ‘prickingsʼ, holes made by perforating the parchment or paper with the ruling tool.[16] When made close to the outer edge of the leaf, these prickings were often removed during subsequent trimming operation.
The Misṭara or masṭar
Another tool which scored the surface was used very widely for paper manuscripts, namely the misṭara (or masṭar)[17], a frame made of cardboard or occasionally of wood on which cords of various thickness could be stretched, corresponding to the text frame lines and guidelines.[18] A copyist would place the misṭara beneath a sheet of paper and then rub his thumb along the cords, perhaps wrapping his thumb in a cloth to avoid soiling the paper; in this way, a light indentation was produced (illus. 63). Some scribes invariably placed the misṭara on the verso of the leaves of a quire, whereas others probably ruled across each bifolium successively; the difference can be observed by noting the respective positions of the furrows and ridges in the various quires. A misṭara offered extreme flexibility of use, since highly complex patterns of ruling could be prepared and then conveniently and quickly imposed on dozens of quires. If necessary, the copyist could add further lines - with a hard point, for example - thereby employing what might be termed ‘mixed ruling methodsʼ. The main ruling of the Takhmīs on al-Būṣīrīʼs Burda in Paris BNF Arabe 6072 was laid out with a misṭara, but the copyist then drew two vertical lines with a hard point to rule the areas reserved for the commentary[19]; a similar method was used in MS. Brussels BR ɪ999ɪ, in which hard-point lines indicate the position of decorative elements.[20] Some of the manuscripts of poetry studied by Paola Orsatti contain four or even six columns of poems, flanked by two smaller columns or diagonal margin lines.[21] Many copies of works such as Niẓāmīʼs Khamsa or Five Poems (one example being MS. Geneva Bibliotheca Bodmeriana 523, dated 889/ɪ484), feature three double columns of text.[22] For works in mixed prose and verse, such as a copy of Nigāristān-i Muʼīnī (MS. Rome Accademia dei Lincei, Caetani 62), the ruling pattern done with a misṭara includes two columns and a double margin, but the copyist only kept to it fully when transcribing passages in verse.[23]
Other ruling techniques...
*The remainder of this article is exclusively available in the printed version of the related book. The book is available in both electronic and printed formats within Our Publications in the following link:
http://doi.org/10.56656/100099
[1] Ruled guides might include horizontal ‘guide-linesʼ for individual lines of text and/or marginal ‘bounding linesʼ for the justification of the block of text.
[2] Ibn al-Nadīm, Kitāb al-First, ed. G. Flügel (Leipzig, 1871), p. 159 and ed. R. Tajaddud (Tehran, 1350/1971), p. 181; translated into English by B. Dodge as The First of al-Nadîm: a tenth-century surυey of Islamic culture (New York/ondon, 1970), vol. 1, p. 351.
[3] See, for example, the fragments Paris BNF Arabe 328a and 328 (E. Tisserant, Specimina codicum orientalium [Bonn, 1914), PL 41b; G. Bergstrasser and O. Pretzl, Die Geschichte des Korantexts, GdQ, vol. III [Leipzig, 1938], fig. 8; N. Abbott, The Rise of the North Arabic script and its Kurʼanic deυelopment [Chicago, 1939] p. 24; Déroche, Cat. I/1, p. 61, no. 7, F. Déroche and S. Noja Noseda, eds., Le manuscrit Arabe 328 (a) de la Bibliothèque nationale de France [Lesa, 1998]).
[4] H.C. von Bothmer, ‘Ein seltenes Beispiel für die omnementale Verwendung der Schrift in frühen Koranhandschriften: die Fragmentgruppe Inv. Nr. 17-15.3 im ‘Haus der Handschriftenʼ in Sanaaʼ, Ars et Ecclesia, Festschrift für Franz Ɉ. Ronig zum 60. Geburtstag, H.-W. Stork, C. Gerhardt and A. Thomas, eds., [Veröffentlichungen des Bistumsarchivs Trier 26] (Trier, 1989), pp. 45-67.
[5] F. Déroche, ‘Coran, couleur et calligraphieʼ, I Primi sessanta anni di scuola: Studi dedicati dagli amici a S. Noja Noseda nello 65° compleanno, 7 leglio 1996 (Lesa, [2004]), p. 131-154.
[6] The term ‘justificationʼ has several meanings, of course, including - in the context of printed type-‘to adjust spacing along a line of text to a prescribed measure so that adjacent lines are of equal length.ʼ (The Shorter Oxford Dictionary.)
[7] Michelle Brown defines bounding lines as ‘the marginal lines supplied during ruling to guide the justification of the text and its ancillaries (such as initials). ʼSee her Understanding illuminated manuscripts: a guide to technical terms (Malibu, CA/ London, 1994).
[8] For example, MS. Paris BNF Arabe 181 (FiMMOD 31).
[9] Exceptions exist, for example when the top line served as a ‘lintel ʼrather than being written on. In one London Qurʼān (N.D. Khalili Collection Qur 4), the ruling for each line of writing consists throughout of three lines, the upper one marking the top limit of ascenders, thereby playing the role of lintel (D. James, After Timur [London, 1992], p. 42, no. 9, the manuscript is tentatively dated to the period 1480-90). In an Iranian manuscript (sixth/twelfth or seventh/thirteenth century) of the tafsīr of Ṭāhir ibn Muḥammad al-Isfarāyinī, four lines were marked to guide the writing of the large hand reserved for the text of the Qurʼān: two of them, above the ruling line, served to mark the height of the main body of the letters and the tallest ascenders, while a third one, below the ruling line, indicated the maximum limit of descenders; see The Qurʼan, scholarship and the Islamic arts of the book: a further selection of fine manuscript material [Bernard Quaritch Ltd., unnumbered catalogue] (London, 1999) p. 20.
[10] F. Déroche, ‘A propos dʼune série de manuscrits coraniques anciensʼ, MSS du MO, PP. 102-103; ‘The Qurʼān of Amājūrʼ, MME5 (1990-91), pp. 61-62, Chart I. The designations of the various ‘Abbāsid scripts (B II, DI, etc.) are given in Déroche, Cat. 1/1 and Abbasid Tradition.
[11] Two exceptions to this trend should be noted here: firstly, manuscripts on tinted parchment (notably the case with the ‘Blue Qurʼanʼ) display traces of an elaborate ruling pattern; secondly, illuminators often left preparatory ruling marks for the decoration of manuscripts that did not employ ruling elsewhere.
[12] According to E. Whelan (‘Writing the Word of God: some early Qurʼān manuscripts and their milieux, Part Iʼ, Ars Orientalis 20 [1990], p. 115), the presence of ruling on decoration and, even more so, the tiny irregularities in the base lines of lettering are proof that text was never ruled, but that copyists relied on their eye. This argument does not seem to be decisive, however, ruling does not mean that a ruler was used - a line traced on parchment (or paper) was only a guide and did not prevent hands from writing above or below it. Nor, for that matter, does a comparison between techniques used by illuminators and copyists seem relevant to resolving the question.
[13] M. Dukan, La Réglure des manuscrits hébreux au Moyen-Âge (Paris, 1988), pp. 15-16; the source is A. Piemontois, De secretis libri septem, l. V (Lyon, 1558), p. 316; see also The Secrets of the Reυerende Maister Alexis of Piemont [..[, trans. William Warde (London, 1558).
[14] Hard points - also known as drypoints - were made of various metals. Some of the more malleable metals (lead, silver and so on) might leave traces on the writing surface, which would be oxidised on contact with air.
[15] Although the present writer has not positively identified any ruling done with the fingernail, al-‘Almawī mentions it, advising copyists to be careful not to tear the paper. See F. Rosenthal, The Technique and approach of Muslim scholarship (Rome, 1947), p. 11.
[16] Prickings also appear in manuscripts written on paper. These marks seem to be related to preparations for illumination in two Qurʼāns, namely MSS. Berlin Museum für islamische Kunst Inv. Nr. I. 42/68 (Museun für islamische Kunst Berlin, Katalog 1979, 2nd edition [Berlin-Dahlem, 1979], pp. 9-11, no. 1 and pl. 1) and Paris BNF arabe 418 (Déroche, Cat. I/2, pp. 128-129, no. 535 and pl. I).
[17] Gacek, AMT, p. 68.
[18] An eleventh seventeenth-or twelfth eighteenth-century misṭara is reproduced in NEW YORK 1994, p. 127, fig. 88 (Metropolitan Museum of Art, Islamic art, accession no. 1973.1). An Ottoman example is illustrated in U. Derman, ‘Hatʼ, Sabancı koleksiyonu (Istanbul, 1995), p. 23, A.
[19] FiMMOD 3.
[20] FiMMOD 195.
[21] Paola Orsatti,‘Epigraphes poétiques de manuscrits persansʼ, Mss du MO, pp. 73-75.
[22] FiMMOD 178.
[23] Orsatti, op. cit., p. 73.
| Source note: This article was published in the following book: Islamic Codicology: An Introduction to the Study of Manuscripts in Arabic Script _ English version, 2005, Al-Furqan Islamic Heritage Foundation, London, UK, pp. 159-184. |
