Black inks in medieval Jewish manuscripts
p. 13-30
Texte intégral
Introduction
1Although manuscript studies have attracted a great deal of scholarly attention, the ink composition is only partly considered within codicology.1 However, thanks to the multidisciplinary approach adopted by the Bundesanstalt für Materialforschung und -prüfung (BAM) in Berlin and the Centre for the Study of Manuscript Cultures (CSMC) of the University of Hamburg, it has been possible to conduct the research presented here, using archaeometric analysis of inks in addition to the traditional codicology and palaeography. Our aim was to obtain information about the use of different black inks in medieval Hebrew manuscripts, focusing on the question of possible preferences in the use of certain inks by different scribes.
2Among recent studies dedicated to ink characterization, only a small number deal with medieval documents2 with a few of these considering Hebrew manuscripts.3 Recently, we pointed out the importance of the writing materials used for the Hebrew documents from the Cairo Genizah.4 As a large number of the fragments have already been thoroughly studied by scholars, this corpus is highly suitable for scientists who seek a solid historical context for their studies. The fact that many fragments from the Genizah explicitly give information about their date and location of composition offers the necessary frame for building a database of materials and gaining insights into the possible evolution of technologies in medieval writing materials. In addition, analysing Hebrew manuscripts from the East allows the results to be compared with extant recipes and handbooks which were in circulation in the Fatimid Empire. The handbooks are particularly interesting because they connect specific recipes to the type of document to be written, the writing support or the ingredients available.5
3In her seminal study on the evolution of ink, Monique Zerdoun Bat-Yehuda dedicates one chapter to the countries under Muslim domination and another to the specificity of black inks for religious purposes in the Jewish world.6 It is unfortunate that this study of the extant records has never been paralleled by experimental research, and therefore we cannot assess systematically the differences between the theory derived from the written sources and the experimental assessment. As a contribution to this task, we present here an overview of the inks detected in a selection of Hebrew documents penned in the Middle East in the period from the 11th century to the 13th century.
Types of inks based on the extant sources
4The most common ink used to copy medieval documents is black ink – although the category of black ink includes “brown inks”, since the current appearance of an ink is the result of ingredients, ageing, degradation and restoration. Black inks can be divided into four chemically different types, carbon (soot or charcoal), plant (tannin), iron-gall and mixed:
- carbon ink consists of a fine dispersion of carbon pigments in a water-soluble binding agent;
- plant inks are made from tannins extracted from tree bark or gallnuts;
- iron-gall ink is based on the black pigment (iron gallate) produced from the reaction between iron (II) and gallic acid, which is contained in tree bark and gallnuts;
- mixed inks result from the mixing of carbon ink with plant or iron-gall ink.
5We find all four types in medieval recipes and treatises from about the 10th century, indicating that they were well known in the Islamicate world during the Middle Ages. Arabic sources preserved many recipes; some are personal notes, but the detailed ones are found within instructions for scribes. The most famous handbook was written by Al-Mu‘izz Ibn Bādīs7 (1008-1062 CE)8 in the first half of the 11th century in North Africa.
6The ornament of the scribes [Zīnat al-kataba], written by Muḥammad b. Zakariyyā’ al-Rāzī, is the oldest known Islamic treatise about ink-making, dating to the 10th century.9 By far more known and better disseminated was the work on bookmaking authored by Ibn Bādīs. Zerdoun believes that the wide circulation of The support of the scribes and the tool for the wise men [ʿUmdat al-kuttāb wa-ʿuddat ḏawī al-albāb]10 by Ibn Bādīs indicates that scribes of the Islamicate world knew how to prepare different types of ink as early as the 11th century.11 Ibn Bādīs presents a wide variety of ink recipes12 that differ in their main ingredients, in minor additives and in quality. Among these recipes, one can find all four types of ink described above: inks based on charcoal or soot, various plant inks, iron-gall inks and mixed inks. When it comes to the iron-gall ink recipes, the majority require vitriol (a mixture of hydrated metal sulfates) as the source of iron, sometimes specifying that the vitriol should be green (i.e., ferrous sulfate); this indicates that the authors of the recipes already knew that iron was the only metal which reacted with gallic acid to produce black ink. However, other recipes based on iron filings demonstrate that vitriol was not the only possible source of iron for medieval iron-gall ink.
Methods and devices
7Several methods of analysis have been used recently to investigate black inks.13 The choice of an analytical protocol is commonly dictated by the question to be addressed and by the study conditions. The latter are usually imposed by collection conservation and determine whether the study has to be conducted in situ or only using non-invasive equipment.
8The standard procedure developed by the BAM for work in libraries collections, which was adopted in this study, is based on a systematic reflectographic examination in the near-infrared (NIR) region followed by X-ray fluorescence (XRF) analysis.
9The first of these methods, infrared reflectography, can determine the type of ink used: the colour of carbon ink is independent of the wavelength between 300-1700 nm; iron-gall ink changes opacity at long wavelengths (i.e., 750-1000 nm)14 and becomes transparent at around 1400 nm; while plant ink turns transparent at wavelengths close to 750 nm. To disseminate the use of this method among conservators and codicologists, the BAM has suggested a simplified approach, measuring opacity changes from visible to near-infrared light.15 In addition to the information obtained from changes in opacity, the appearance of a halo around a letter, better noticeable under ultraviolet light, can be attributed to the presence of tannins. In the present study, reflectography measurements have been taken using a USB microscope – DinoLite AD4113T-I2V – equipped with three different light sources: ultraviolet (395 nm), visible (external lamp) and near-infrared light (940 nm).
10The second technique, X-ray fluorescence analysis (XRF), allows us to identify the elementary composition of materials and is commonly used on various objects in the field of cultural heritage, including manuscripts. This technique is of particular interest when comparing iron-gall inks, since the majority of medieval Arabic recipes for iron-gall ink contains vitriol as the source of soluble iron, the main inorganic ingredient of the ink. Mineralogically, vitriol is a mixture of hydrated metallic sulfates, containing manganese, iron, copper, zinc and other metallic sulfates in different proportions depending on the geological source of the vitriol.16 Though metals other than iron do not participate in forming the black ink, their presence allows us to classify different iron-gall inks using a fingerprint model.17 This model has been developed specifically for the differentiation of iron-gall inks, by quantifying the XRF spectra to calculate the relative amounts of the various metals in the vitriol in relation to iron, the main component of the ink. Initially developed to allow a quantitative comparison of manuscripts written on paper, a semi-quantitative version of the fingerprint model has been successfully applied to medieval parchment and paper manuscripts.18 In short, X-ray incident radiation causes inorganic components of the ink such as iron, copper, zinc, etc. to emit characteristic X-rays that are recorded as spectra by the detector of an XRF spectrometer. Evaluation of the spectra results in a fingerprint of the ink under investigation. In the studies presented here, we used ArTAX (Bruker Nano GmbH), a spectrometer specially designed for the objects of cultural heritage, and well-known in museums, libraries and archives. All measurements were performed at 50 kV and 600 µA, with an acquisition time of 40 seconds for each point. We conducted multiple line scans, each consisting of ten points, with a minimum of three measurements on the ink and the writing surface to obtain valid average values.
Results
11In this article, we will focus on Jewish documents produced between the 11th and the 13th century. The first part of the corpus consists of two sets of documents found in the Cairo Genizah, all of which date from the 11th or 12th century – 96 documents written in al-Fusṭāṭ, Egypt, and 34 documents penned in the Land of Israel;19 in addition, a single document from the Cairo Genizah that was written in Aden is also examined. The second set of documents in the corpus is composed of 14 Jewish documents found in the so-called Afghan Genizah, written between the 11th century and the 13th century. All the documents presented – unless otherwise stated – were penned by Jews and address various subjects: some are legal documents, some are private and some are religious documents.
12Previous studies of documents from the 11th century found in the Cairo Genizah has revealed that all four types of ink (iron-gall, carbon, mixed and plant inks, in descending order of frequency) were in use concomitantly in al-Fusṭāṭ.20 Moreover, some iron-gall inks investigated in this study contained no transition metal component other than iron. Since the complete purification of vitriol was not known at the time under investigation, we believe that these inks were produced without using vitriol, and to differentiate this iron-gall ink variety from the one based on vitriol, we refer to it as non-vitriolic ink in what follows. All the inks discovered in this study correlate well with the extant recipes preserved in the treatises mentioned above.
13One of the questions asked in our study was whether it was possible to find a solid correlation between specific types of documents and the use of specific inks. This was based on the assumption that scribes prepared their own inks, which Zerdoun implies by suggesting that they were well informed about the composition of their inks. In fact, our results regarding the inks used in this period strongly suggest that the scribes of Jewish manuscripts did not prepare inks themselves but rather bought them. The detailed study of 96 Hebrew documents (covering different subjects, written on different supports) penned by five scribes in al-Fusṭāṭ has unequivocally shown that there is no preference for one ink type over another within any of the specific Jewish communities nor for any specific document type with the exception of the religious documents produced in a form of a roll which needed to fulfil precise requirements.21 These results indicate that the scribes did not care much about the specific composition of their ink but were only interested in its colour.
14To illustrate our conclusion, we are presenting here a 12th century letter, preserved today in the papyrus collection of the Österreichische Nationalbibliothek in Vienna as manuscript H 00161 Pap.22 This document, shown in figure 1a, was found in the Cairo Genizah, and is dated to 1467 of the Seleucid era, that is, 1156 CE. The letter, written in Judeo-Arabic is a legal deposition regarding a ship that sank off Aden (Yemen) on its way to Quilon (India).23 The author of the letter reports the discussion on the on Halakha-based status of the wife of one of the victims. The letter is signed in the margin of the verso by five notables of Aden, who consent permitting the widow to remarry.24

a)
b) |
b) |
c) |
c) |
d) |
d) |
Fig. 1 — Manuscript H 00161 Pap.;25 a) detail from the image with the rectangles indicating the locations of the micrographs presented on the right, b) visible (VIS) and near-infrared (NIR) micrographs of the first line of the document, c) VIS/NIR micrographs of the second line of the document, and d) VIS/NIR micrographs of the margin.
15The main text and the notes in the margin are penned in a black ink that shows no signs of deterioration. The NIR reflectographic observation of the document shows clearly that the first line and a half of the document and the marginalia are both written with iron-gall ink (figure 1b and figure 1d): the ink changes opacity but does not disappear under NIR light. Starting from the second line, the NIR observations reveal, as no change of opacity can be noticed, that carbon ink was used (figure 1c). On the verso, the ink varies as well, changing from iron-gall ink to carbon ink and vice versa, several times. Only reflectography has been performed on this document, so no composition study of the iron-gall inks or of the carbon-based inks could be made.26 Therefore, one cannot be sure that a single iron-gall ink was used or that the carbon ink remains the same consistently throughout the document. Furthermore, reflectography alone does not allow us to differentiate between pure carbon and a carbon-based mixed ink. Despite all these limitations, it is clear that the scribe who penned this letter used more than one type of the ink. A change of ink occurring within the space of two lines is especially remarkable. We interpret it as a clear sign of the fact that the scribe did not care much about the composition of the ink but was interested in the colour, which is black independently of the ink type. Furthermore, we can see that iron-gall and carbon ink were equally available at that time in Aden (Yemen), confirming – at least, partially – the results obtained already in al-Fusṭāṭ (Egypt).
16Every known type of ink has been identified as occurring in documents written during the period under investigation found in the Cairo Genizah, although iron-gall ink dominated the palette. In addition, some scribes seem to have had preferences for a specific ink type, which must have been due to their personal inclination rather than being imposed by the community to which the scribe belonged.27 We speculate here that the price of ink might have been a relevant factor because we find inks greatly varying in quality, which in turn might be connected to the place of procurement or the price of ingredients.
17Contrary to the findings concerning the variety of inks used in al-Fusṭāṭ, documents originating from the area corresponding to nowadays Israel reveal a change in the pattern. All 34 documents found in the Cairo Genizah that were penned in the land of Israel28 – either in Ramla or in Jerusalem – that we were able to analyse were written with an iron-gall ink. Interestingly, although we have analysed only a very limited number of documents (eight)29 written by Avraham ben Šelomoh ben Yehudah, the son of Šelomoh ben Yehudah Ga’on, the only one written with a carbon ink originates from Munyat al-Qā‘id (Egypt).30
18Let us go further east and look at similar analyses conducted on the inks used on 14 documents31 from the so-called Afghan Genizah,32 preserved today in the National Library of Israel in Jerusalem. All the documents were written between the very beginning of the 11th century and the first half of the 13th century.33 All the inks analysed show a predominant use of carbon-based inks, except for ms. Heb. 8333.86=4,34 which is completely written in iron-gall ink. The results of our analysis strongly suggest that in Afghanistan, al-Fusṭāṭ and Aden, it was the colour of the ink that was decisive rather than its composition. For example, ms. Heb. 8333.206=4, shown in figure 2, has been written with more than one type of ink, just like manuscript H 00161 Pap. presented above (figure 1). Reflectographic observation shows carbon ink (figure 2b), iron-gall ink (figure 2c) and what is most probably mixed ink (figure 2d).

a)
b) |
b) |
c) |
c) |
d) |
d) |
Fig. 2 — Jerusalem, The National Library of Israel, ms. Heb. 8333.206=4,35 fol. 23; a) detail of the manuscript with rectangles indicating the locations of the micrographs presented on the right, b) visible (VIS) and near-infrared (NIR) micrographs from rectangle 1, c) VIS/NIR micrographs from rectangle 2, and d) VIS/NIR micrographs from rectangle 3.
19In this document, the XRF analyses have shown that the carbon ink contains elevated amount of copper. Similar inks have been found on paper manuscripts from the Silk Road.36 At the moment, it is not possible to unequivocally determine whether this is a carbon ink to which a copper salt has intentionally been added, or if it is a mixed ink consisting of a carbon ink combined with a primitive iron-gall ink, in which copper vitriol was used rather than iron vitriol. In any case, it is noteworthy that the Afghan documents are primarily written with carbon ink and the ones penned in the Land of Israel with iron-gall inks, while the documents in Egypt present a broad distribution of inks with iron-gall ink dominating the distribution.
20This correlation between the predominant use of certain ink types and the geographic origin of documents needs to be verified on the basis of further studies. Currently, we are constructing the map of the use of the black writing ink in the medieval Middle East, and we would like to show the contribution obtained in the present study, in figure 3. In this figure, the size of the circles is proportional to the number of observations.

Fig. 3 — Map showing the different ink types used to write Hebrew documents, depending on where the document was copied.
21According to the version described and translated by Levey in 1962,37 Ibn Bādīs’s book contained 39 black inks recipes:38 22 iron-gall inks, 11 carbon inks, five plant inks and one mixed ink. Let us roughly compare the results obtained from analysing documents written in al-Fusṭāṭ to the distribution of black inks recipes by Ibn Bādīs in Levey’s translation. Among the 96 documents penned by the scribes presented above, the distribution of inks used is as follows: 66% iron-gall inks, 32% carbon inks, 1% mixed inks and 1% plant inks. A comparison of the two distributions is presented in figure 4, with different magnitudes on right and left y-axes. The change of magnitudes allows comparison of the overall distribution of the ink types identified analytically from the manuscripts written in al-Fusṭāṭ and of the one found in the treatise by Ibn Bādīs. Except for plant inks, the proportion for each ink type measured during our analyses reflect the proportion for the same ink type observed in the recipes: iron-gall inks were used more frequently in everyday life and non-sacred documents, followed by carbon inks, and finally mixed inks. There are more plant ink recipes in Ibn Bādīs’s treatise than in the results obtained from the corpus of Genizah documents that were analysed, and we can only speculate on why plant inks were only rarely used, but it is worth noting that plant inks have a brown colour, unlike iron-gall and carbon inks that are originally black. It is possible that the scribes under study preferred the other types of ink (iron-gall and carbon inks) to plant ink because of their colour.39

Fig. 4 — Distribution of the ink types in Ibn Bādīs according to the 1962 translation of Levey and as measured in manuscripts from the Cairo Genizah that were written in al-Fusṭāṭ by one of the five known scribes.
22Let us look now at the composition of the iron-gall inks. As mentioned earlier, the majority of the recipes for iron-gall ink that circulated at this time in Egypt include vitriol. Al-Rāzī, the author of Zīnat al-kataba speaks of different types of vitriol. Their colours and names correlate well with their ancient Greek counterparts, but, just as with the Greeks in Antiquity, there is a confusion between green vitriol (ferrous sulfate) and blue vitriol (copper sulfate), since they are both commonly given the same name, qalqand (ou qalqant), the Arabic transliteration of the Greek chalcanton χαλκάνθου.40 This fact and the lack of a purifying technique that would isolate the iron salt required for production of iron-gall ink may be responsible for the great variety of metallic elements discovered in the iron-gall inks in this study. Elementary ink compositions attributable to the same vitriol were only ever found within single manuscripts.41
23In addition to vitriolic iron-gall inks, we also found a small number of non-vitriolic ones. Iron nails, slag and iron filings have been traditionally used in Africa to obtain solvated iron that reacts with gallic acid to produce black ink.42 On the other hand, we found a number of recipes in Islamic sources that reflect this tradition.43 Ibn Bādīs,44 al-Marrākušī45 and al-Qalalūsī46 have recipes that include iron filings as the source of iron. It is noteworthy that use of metallic iron to produce black pigment continues to be known today through oral transmission;47 iron filings and similar products are probably used to avoid the difficulty of obtaining vitriol.
24During the analyses conducted on documents from the Cairo Genizah, we found that a total of six of the 96 manuscripts were written with an iron-gall ink that displayed the main feature of a non-vitriolic ink, namely the lack of metallic elements besides iron that would commonly be present as contaminants. An example of the analysis of such an ink can be seen in figure 5. Figure 5a shows a detail of ms. T‑S 24.73 of the Cambridge University Library. The manuscript is a power of attorney, dated 25 Adar I, 4807 (i.e., 26 March 1047). The reflectographic images in figure 5b show that the ink of the signature of one of the witnesses, Avraham “the Humble” ben Ayyub [אברהם הענו ביריבי איוב]48], loses opacity under NIR light, which is characteristic for iron-gall ink. Figure 5c displays the portions of interest of the XRF spectra of the support (red), the ink of the main text (solid black) and that of the signature in focus (dotted black). Comparing these three spectra, we can see that the signal from the only element that can be associated with iron-gall ink, namely iron, is significant in the spectra of the inks, whereas no peaks from other metallic vitriolic components, such as copper (Cu), manganese (Mn) or zinc (Zn), emerge in the spectra of the inks.

a)
b) |
b) |

c)
Fig. 5.49 T‑S 24.73 a) Detail of the photograph, b) visible (VIS) and near-infrared (NIR) micrographs of the ink used by Avraham ben Ayyub to sign (highlighted by a square on the manuscript), c) portion of the XRF spectra collected from the ink of the main text (solid black), the ink of the signature of Avraham ben Ayyub (dotted black) and the paper (red).
25The small number of extant recipes for the production of non-vitriolic inks (in average, each treatise contains one recipe) contrasts with the number of occurrences of this ink type, seeming to indicate that in Egypt – or at least in al-Fusṭāṭ – during the time in focus, it was rather common to prepare iron-gall ink without vitriol. Furthermore, its popularity in Egypt outside of the Jewish context is reflected in its presence (representing about a third of a corpus of 100 documents) in a corpus of Coptic documents written between the 4th and the 11th centuries in northern Egypt.50 Such inks were also found in the context of early medieval European manuscripts.51 These findings are especially important because they negate the possibility that non-vitriolic inks might be a “Jewish speciality”, though this type of the ink seems to be well represented in the Jewish manuscripts outside of Egypt.52
26In general, we believe that the correspondence between the types of ink found analytically on the analysed Jewish manuscripts and in the Arabic ink recipes indicates that Jews were well integrated in the countries of the Fatimid Empire. To check the degree of popularity of non-vitriolic inks in documents written by the Islamic majority would require further analytical studies, and indeed such a study is currently being conducted at the University of Hamburg, which focuses on the transition from papyrus to paper in the Islamic world and includes an experimental characterization of the inks.53
Conclusion
27In this article, we have looked at the four types of black inks known in the Arabic sources: carbon ink, iron-gall ink, plant ink and mixed ink. All four types of inks have been identified in Jewish documents written between the 11th century and the 13th century in al-Fusṭāṭ, Egypt. In a single document written in Aden, two types of inks have been detected: both iron-gall ink and carbon ink. In contrast, a single ink type, namely iron-gall ink, has been identified in documents written in the Land of Israel, while another single, carbon ink, type was prevailingly found in Jewish documents in Afghanistan. A further study is now underway to expand the analysis to manuscripts written by non-Jewish scribes in the same geographic areas.
Acknowledgements
28The archaeometric analysis was supported by the Cluster of Excellence “Understanding Written Artefacts”, funded by the German Research Foundation (Deutsche Forschungsgemeinschaft, DFG) and under the Centre for the Study of Manuscript Cultures (CSMC) at the University of Hamburg. Our thanks go to Olivier Bonnerot, who accompanied us on each of our trips to Cambridge University Library (CUL). We would like to acknowledge the help of the staff of the various manuscript collections in allowing us access to the manuscripts and for supporting us during the analytical studies. Our special thanks go to the staff of the CUL, especially to Ben Outhwaite, James Bloxam, Rebecca Goldie, Anna Johnson, Emma Nichols and Ngaio Vince-Dewerse for their precious and constant help, as well as to all those in the Digital Content Unit for hosting us on our trips: Maciej Pawlikowski, Blazej Mikula, Scott Maloney, Marc Box, Amelie Deblauwe, Domniki Papadimitriou. We would like as well to thank the Syndics of the CUL for permission to reproduce the image of the manuscript T-S 24.73. Many thanks to Simone Suppan and Bernhard Palme from the papyrus collection of the Austrian National Library in Vienna for allowing us access to the Cairo Genizah manuscripts in that collection and for permission to reproduce an image of manuscript H 00161 Pap. Our deepest gratitude goes to Haim Ofir and the staff of the National Library of Israel who hosted us and helped us to analyse the manuscripts of the Afghan Genizah: Alexander Gordin, Marcela Szekely, Hagar Milman, Haim Shushan and Denis Shorr.
Notes de bas de page
1M. Geissbühler et al., “Advanced Codicological Studies”.
2R. J. Díaz Hidalgo et al., “New Insights into Iron-Gall Inks”; M. Aceto, E. Calà, “Analytical evidences of the use of iron-gall ink as a pigment on miniature paintings”; G. M. Contreras Zamorano, La tinta de escritura; I. Rabin et al., “Why Do We Need to Study Inks?”; G. Nehring et al., “Writing and Correcting”; O. Hahn et al., “A Study of Early European Inks”.
3O. Hahn et al., “The Erfurt Hebrew Giant Bible”; S. A. Centeno, N. Stavisky, “The Prato Haggadah”; I. Rabin et al., “Tintenarten”; S. Wijsman et al., “Uncovering the Oppenheimer Siddur”.
4Z. Cohen et al., “Composition Analysis of Writing Materials in Geniza Fragments”.
5C. Colini, From Recipes to Material Analysis, p. 49 discusses in figure III.16 and figure III.17 the distribution of the different ink typologies recommended for use on papyrus, parchment and paper according to al-Rāzī. M. Levey, “Mediaeval Arabic Bookmaking”, pp. 60–61 translates three examples of black inks for religious books, allegedly given by Ibn Bādīs. However, it is worth noticing that Levey translates both indiscriminately qarāṭīs (papyrus) and waraq (paper) by “paper”. In some cases, the Arabic term is present in the note, in some other, the Arabic term is not there.
6M. Zerdoun Bat-Yehuda, Les encres noires au Moyen Âge.
7M. Levey, “Mediaeval Arabic Bookmaking”; M. Zerdoun Bat-Yehuda, Les encres noires au Moyen Âge, pp. 123–43; S. Fani, Le arti del libro, pp. 50–55.
8Ibn Bādīs was the fourth ruler of the Zirid dynasty in Ifriqiya, reigning from 1016 to 1062 (M. Talbi, “Al-Muʿizz b. Bādīs”), although certain sources attribute this handbook to this son, Tamīm Ibn Bādīs. See C. Colini, From Recipes to Material Analysis, p. 36.
9M. Zaki, “Early Arabic bookmaking techniques”.
10The translation of the title here is the one given by C. Colini, From Recipes to Material Analysis, p. 35, who is quoting Fani’s translation (S. Fani, Le arti del libro, p. 50). As Colini noted, Levey translates the title of Ibn Bādīs’s treatise very differently, as “staff of the scribes and implement of the discerning”. See M. Levey, “Mediaeval Arabic Bookmaking”, p. 13.
11M. Zerdoun Bat-Yehuda, Les encres noires au Moyen Âge, p. 128.
12The precise number of recipes differs in different copies of the manuscript.
13O. Hahn et al., “Characterization of Iron-Gall Inks”; A. S. Lee et al., “Raman Analysis”; M. Aceto et al., “The Vercelli Gospels Laid Open”; A. Gambaro et al., “Chemical and statistical characterization”; R. L. Easton, W. Noel, “Infinite Possibilities”; I. Nastova et al., “Spectroscopic Analysis of Pigments and Inks”; V. Tanevska et al., “Spectroscopic Analysis of Pigments and Inks”; V. Mocella et al., “Revealing Letters”; I. Rabin, “Digital and Scientific Approaches to Oriental Manuscript Studies”.
14In R. Mrusek et al., “Spektrale Fenster zur Vergangenheit”, the authors indicate that iron-gall ink becomes transparent at 1200 nm. However, our recent tests have shown that iron-gall ink becomes completely transparent at longer wavelengths than this.
15I. Rabin et al., “Identification and Classification”.
16V. Karpenko, J. A. Norris, “Vitriol in the History of Chemistry”.
17O. Hahn et al., “Characterization of Iron-Gall Inks”.
18M. Geissbühler et al., “Advanced Codicological Studies”; G. Reeg, “Codex hebraicus 18”.
19This period has been very well studied from historical and philological points of view: considerable in-depth research has been carried out both on the scribes and on the documents studied here. In addition, the differences between the Jewish communities are well marked at this period; and the documents reflect a rich panoply of subjects (legal, private and religious). The large number and diversity of documents allow us to have a more representative sample of the written production and of its evolution. For the detailed explanation for the selection of the corpus of documents see Z. Cohen et al., “Composition Analysis of Writing Materials in Geniza Fragments”, or Z. Cohen, Composition Analysis of Writing Materials in Cairo Genizah Documents.
20Z. Cohen, Composition Analysis of Writing Materials in Cairo Genizah Documents.
21Z. Cohen, Composition Analysis of Writing Materials in Cairo Genizah Documents.
23E. Ashtor, “A journey to India”; S. Shaked, A tentative bibliography of Geniza documents, p. 235; M. A. Friedman, S. D. Goitein, India Traders of the Middle Ages, pp. 530–40; R. E. Margariti, “Wrecks and Texts”.
24For more details on the document and a complete translation, see M. Friedman, S. D. Goitein, India Traders of the Middle Ages, pp. 530–40.
25Copyright Papyrussammlung der Österreichischen Nationalbibliothek. We would like to thank them for permission to reproduce this image.
26However, it would be very beneficial to see how the composition of inks variates in the document, and specifically among the signatures and the region of the margins. The authors therefore hope to be able to perform these analyses in the future on this document, as well as on different contemporary documents of this collection.
27Z. Cohen, Composition Analysis of Writing Materials in Cairo Genizah Documents.
28The documents which have been analysed are all written by the Ga’on of the Palestinian yeshiva, Šelomoh ben Yehudah; by his son Avraham ben Šelomoh ben Yehudah; or by the leader of the Palestinian community in Ramla, Šelomoh ben Samaḥ. The documents are stored in Cambridge University Library (CUL) under the following classmarks: T‑S 10J10.22, T‑S 10J10.9, T‑S 10J11.29, T‑S 10J11.30, T‑S 10J12.17, T‑S 13J14.23, T‑S 13J14.8, T‑S 13J17.17, T‑S 13J23.11, T‑S 13J34.10, T‑S 18J3.9, T‑S 24.43, Mosseri Ia.4, Mosseri II.141.1, Mosseri II.141.2, T‑S 13J11.5, T‑S 13J13.17, T‑S 13J13.28, T‑S 13J15.1, T‑S 13J16.14, T‑S 10J29.13, T‑S 13J31.7, T‑S 13J31.8, T‑S 13J33.6, T‑S 13J26.1, T‑S 13J34.11, T‑S 13J34.2, T‑S 13J17.4, T‑S 13J21.19, T‑S 13J23.1, T‑S 13J23.19, T‑S 13J9.2, T‑S 8.220, T‑S 18J2.16.
29The documents are stored in CUL under the following classmarks: T‑S 10J10.9, T‑S 13J14.23, T‑S 13J17.17, Mosseri II.141.1, T‑S 13J26.1, T‑S 13J17.4, T‑S 13J21.19, T‑S 18J2.16.
30The document is a letter stored in CUL under the classmark T‑S 13J14.23. The localization of the town is based on the mention made by Yāqūt (al-Ḥamawī Yāqūt, Le dictionnaire des pays [Mu’jam al-buldān], vol. V, p. 219). We would like to thank Wissem Gueddich for bringing this information to our intention and translating the text for our needs.
31The documents analysed can be found under the following classmarks: ms. Heb. 8333.215=4, ms. Heb. 8333.216=4, ms. Heb. 8333.217=4, ms. Heb. 8333.21=4, ms. Heb. 8333.12=4, ms. Heb. 8333.23=4, ms. Heb. 8333.206=4, ms. Heb. 8333.161=4, ms. Heb. 8333.49=4, ms. Heb. 8333.86=4, ms. Heb. 8333.73=4, ms. Heb. 8333.176=4, ms. Heb. 8333.17=4, ms. Heb. 8333.25=4.
32The “Afghan Genizah” is the name given to a corpus of documents found in central Afghanistan. It is not established if all the documents come from the same cache, where it was found or even if the cache was indeed a Jewish genizah. However, due to the fact that many of the documents are written in Hebrew script, the corpus was named “Afghan Genizah”.
33O. Haim, “What Is the ‘Afghan Genizah’?”.
35We would like to thank the National Library of Israel for permission to reproduce this image.
36S. van Schaik et al., “Writing, Painting and Sketching”; R. Nöller, O. Hahn, “Illuminated Manuscripts”.
37M. Levey, “Mediaeval Arabic Bookmaking”. The author is using a copy from the 20th century (Chicago, University of Chicago, Oriental Institute, ms. A12060) based on a copy from the 19th century (Berlin, Staatsbibliothek, ms. Landberg 637).
38The count of ink recipes is taken from chapter 2 (soot inks) and chapter 3 (black gallnut inks).
39Some of the plant inks given by Ibn Bādīs (at least in the version translated by Levey) are described as shiny black inks; but as they are plant inks (M. Levey, “Mediaeval Arabic Bookmaking”, p. 18), they should be brown. Perhaps Ibn Bādīs, one of the scribes copying the handbook during its transmission, or even Levey himself is leaving out one of the ingredients in these recipes, which would of course change the number of actual plant inks in the corpus.
40The Hebrew transliteration of chalcanton, קנקנתום qanqantum, appears in the Talmud with the same meaning (bGiṭṭin 19a; b‘Eruvin 13a; bSoṭah 20a; bMegillah 17a). For more details on the vitriol, see V. Karpenko, J. A. Norris, “Vitriol in the History of Chemistry”.
41Z. Cohen et al., “Composition Analysis of Writing Materials in Geniza Fragments”.
42M. Biddle, “Inks in the Islamic Manuscripts”, pp. 14–16.
43C. Colini, “‘I tried it and it is really good’”; C. Colini, From Recipes to Material Analysis, p. 19.
44M. Levey, “Mediaeval Arabic Bookmaking”, p. 20.
45A. Schopen, Tinten und Tuschen, p. 98; S. Fani, Le arti del libro, p. 111.
46A. Schopen, Tinten und Tuschen, p. 93.
47C. Colini, From Recipes to Material Analysis, p. 94.
48This signature is difficult to read and could be as well transcribed as [אברהם הענו ביריבי איוב], i.e., Abraham “the Wretched” ben Job, which is the option defended in Goitein’s index cards (nos. 100297 and 100155). The authors would like to thank Vered Raziel-Kretzmer who directed us to the relevant bibliography. However, the name איוב is more likely to be interpreted as Ayyub rather than as Job.
49We would like to thank the Syndics of Cambridge University Library for permission to reproduce the images.
50T. Ghigo, I. Rabin, “Archaeometric Study”; T. Ghigo et al., “Black Egyptian Inks in Late Antiquity”.
51O. Hahn et al., “A Study of Early European Inks”.
52N. Gordon et al., “Inks Used to Write”.
53This project, “The Scribe’s Choice: Writing Supports in Arabic Documents of the Early Islamic Centuries”, is hosted by the University of Hamburg, in “Field A: Artefact Profiling” under the cluster “Understanding Written Artefacts (UWA)”. See https://www.csmc.uni-hamburg.de/research/cluster-projects/field-k/rfk01.html.
Auteurs
-
Zina Cohen
Zina Cohen est chercheuse post-doctorale en archéométrie au Bundesanstalt für Materialforschung und -prüfung (BAM), l’Institut fédéral de recherche sur les matériaux dans la division Analyse des artefacts et des biens culturels. Sa thèse a été réalisée sous la direction conjointe de Judith Olszowy-Schlanger et Oliver Hahn entre l’École pratique des hautes études (EPHE-PSL) et l’université de Hambourg, en étroite collaboration avec Ira Rabin. Dans sa thèse, elle a étudié la composition du matériel d’écriture des communautés juives du Caire dans les documents médiévaux. Elle s’intéresse particulièrement à la production et à l’utilisation des matériaux d’écriture à l’époque médiévale.
-
Ira Rabin
Ira Rabin est scientifique senior au Bundesanstalt für Materialforschung und -prüfung (BAM) et professeur à l’université de Hambourg. Elle a étudié la chimie à l’Université hébraïque de Jérusalem. En 1987, elle a obtenu son doctorat en chimie physique à la Freie Universität de Berlin. De 2007 à 2010, elle a coordonné le projet international « Qumran » au BAM. Actuellement, elle concentre ses recherches sur l’histoire des encres d’écriture noires et des supports d’écriture (papyrus, parchemin).

Le texte seul est utilisable sous licence Creative Commons - Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International - CC BY-NC-ND 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.
Fragments, manuscrits, livres dans le monde juif
Journées d’études en paléographie et diplomatique hébraïques : les dix premières années
Giacomo Corazzol et Sarah Fargeon (dir.)
2025
Corps et âme
Leurs interactions selon les médecins et les philosophes de l’Antiquité à l’époque moderne
Antoine Pietrobelli et Laetitia Loviconi (dir.)
2025
La mission de Charles d’Ochoa
Un jalon dans l’histoire des collections indiennes de la Bibliothèque nationale de France
Jérôme Petit
2025