This section will describe in some detail the scope of our research, its geographical coverage, the structure and arrangement of the database, as well as some of its key features. It also explains what to expect and what not to expect (at present) from the electronic archive presented here, and what we hope to do in the future. Finally, it gives credit to the work of those who have contributed to the compilation and development of this innovative resource.
The archive database is arranged by manuscript and by illustration. A manuscript is defined as an item with a separate accession number in a library, and can therefore be either a bound codex or a single sheet, or an album containing separate folios. Each illustration in the ‘manuscript’ is then recorded separately.
The manuscripts are described, so far as possible, by their provenance, date, and place of production, by the number of folios they contain, and the dimensions of the folio and of the text block (written surface). The latter measurement is taken within the ruled margins, i.e. not including the width of the margins. Notes are also made on their illuminations, binding, and any other particular features that might seem of interest. Where possible, an image has been acquired of any colophon, which might provide the name of the calligrapher (copyist), and the date and place of completion of the manuscript. Sometimes, also, a sample page has been photographed, to illustrate the layout, handwriting and decorative features of the folios. To provide even fuller data, for example concerning paper, pigments, calligraphy and other matters, would have made the database impossibly large, and greatly increased the time needed to examine each manuscript. A line had to be drawn somewhere, to identify priorities and make best use of the limited time and resources available for such an ambitious programme. In fact, as the Project has progressed, more effort has been made to acquire images of the illuminations and bindings, for example, which were not included in the early days. We hope that libraries and holders of the manuscripts will be willing in future to contribute images of these and other elements, to increase the research value of the website.
Similarly, each illustration in a manuscript is identified by its folio number (recto and verso), by the title of the painting (see below: Scene titles) and the lines of poetry that come before and after the picture (see below: Breaklines). The name of the painter is also recorded, where known or identified in secondary studies. The database also has a record of the dimensions and shape of the picture (rectangular, stepped, irregularly breaking out of the text frame, etc.), and miscellaneous notes. Again, the fields contained in the database were to some extent restricted by a sense of the feasible. In particular, elements of image recognition (i.e. the presence of specific features of clothing, weapons, animals, the natural environment, etc.) were excluded: we hope that by providing images, researchers can build up their own inventory of matters of personal interest.
Dates are given according to the Islamic hijri calendar, where this is recorded in a colophon or signed painting, in which case it is also converted automatically into the equivalent date A.D. The notation is a little idiosyncratic. It was necessary to have a way of recording dates A.D. that were estimated in secondary studies, either by century, with margins of uncertainty (such as early, middle or late, or first quarter, last quarter, etc.), or within a range of dates, as determined from the style, or the known periods in which certain artists were active.
Most of this essential data is carried through into this website. The data can be accessed by two main routes, either according to the Collection (that is, usually a Library or Museum), arranged geographically by country, or by the Chapterin the Shahnama. The first route follows the way in which data were collected (see below), which naturally concentrated on the major collections worldwide. Information is given (or will be provided in future), on the Collection, and then on the individual manuscripts. When this information has been contributed by people outside the Project, this is indicated. For each manuscript, there is a small thumbnail display of the illustrations it contains, ordered by folio number and identified by their scene title (see below).
The alternative route into the data is by the Chapter and scene of the Shahnama , which allows one to view and compare the different illustrations of the same scene, arranged chronologically. As with the Collection and Manuscript entries, notes are provided (or will be provided in future) on the subject of the scene, and on the individual picture. These notes are not yet systematic, but reflect individual observations by members of the Project, remarks on the text, and editorial comments. Since in most cases, an image of the painting is provided, a description of the picture is not particularly necessary; but such descriptions may be found and are more helpful, when no image is currently available.
For the moment, it is assumed that most researchers will wish to view individual manuscripts, or to compare the depiction of individual episodes and scenes in the Shahnama over time. The new Notebook feature allows individual images to be selected and set aside for study. The search feature, which is still being developed, allows more personal queries to be made.
It is well known that no definitive text exists of the Shahnama, that is to say, of the poem as first composed by Firdausi. The earliest manuscript, in the National Library (Biblioteca Nazionale) in Florence and dated 1217 (which is incomplete), was written 200 years after the poet’s death. The problem of establishing the original text is made difficult not only by the normal problems of literary manuscript transmission, but also by the work’s strong popular, oral, tradition. Oral performance and transmission, and the nature of the epic itself, readily lend themselves to the ‘corruption’ of the text. The result is that every manuscript copy is different.
Nevertheless, for the purposes of the database and the systematic identification of the title of the paintings (see below), it is necessary to refer to a standard text. On the one hand, it is tempting to refer to the only complete modern edition, by Y.E.Bertel's and others, 9 vols. (Moscow, 1960-71), which is the basis for many modern reprints in Iran, and most recent translations. There are some difficulties with this (absence of headings in the text, all the alternative readings in a separate annex), the chief of which applies equally to the other authoritative modern edition, by Djalal Khaleghi-Motlagh, 5 vols. (New York, 1988-97), continued by Mahmoud Omidsalar, vol. 6 (New York, 2005), which is still in progress (down to the death of Bahram Gur). Both these critical editions, by definition, seek to purge from the text any lines considered to be ‘spurious’. However, many of these lines, and other variations, are often encountered in the illustrated manuscripts with which we are working. What is needed is a less restrictive approach, which preserves more of the verses to be encountered (and incidentally, better allows one to contemplate alterations to the text, and the timing and possible reasons for the growth of the text at specific points).
For this purpose, two other printed texts are particularly useful. The edition by Seyed Mohammad Dabirseyaqie [Dabir-Siyaqi], 6 vols. (Tehran, 1935 shamsi/1956), apart from being conveniently laid out, has the great advantage that the editor also produced an index of the lines (Kashf al-abyat-i Shahnama-yi Firdausi, Tehran: Mudabbir, 1368 shamsi/1990). This makes it relatively easy to look up the verses surrounding the miniature paintings, although success depends of course on the verse in the manuscript starting in the same way, and there are often discrepancies of a few lines between the index and the printed text. Dabir-Siyaqi follows the chapter numbering found in the classic work of Fritz Wolff, Glossar zu Firdosis Schahname(Berlin, 1935, pp. vi-vii), and is close to the edition of Jules Mohl, making correspondence between the two quite easy.
Mohl’s edition, in 7 vols. (Paris, 1838-78), although now a rarity, has been chosen as the reference text, since it contains the largest number of verses, was the first standard edition (based on 35 manuscripts), and has thus influenced later publications. It was accompanied by a complete French translation, which was published separately in 1876 and has recently been reprinted. This also formed the basis for the only complete English verse translation, by A.G. and E. Warner, 10 vols. (London, 1905-25). In addition, much of the preliminary statistical work done by Farhad Mehran has been on the basis of the Mohl edition, and it seemed important to retain this. Amin Mahdavi, as a by-product of his own work for the Project, has worked on a concordance between seven standard editions, with support from the British Institute of Persian Studies (BIPS); see his recently published article in Persica 15 (2003), pp. 1-31.
While all four editions have been routinely used, for the present all chapter numbers and verse lines are given here according to the Wolff numbering scheme and the Mohl edition, which has been reprinted several times in Iran, though sometimes unfortunately without the line numbers, and sometimes not preserving the original page numbering.
Each image is fixed in the text, so to speak, by the coordinates of the lines appearing before and after it. In the database, the first and last verses on each illustrated page are recorded, together with the verses immediately before and after the painting: these are called the ‘break-lines’. These break-lines are provided on the website, numbered according to the line in the relevant chapter of the Mohl edition of the text.
The lines are not always found in Mohl, in which case occasionally the closest line is given. Sometimes the line has only been located in one of the other editions (see above). Comments on the textual context of the painting may be found in the notes accompanying each image (i.e. important departures from the printed editions, the inversion of lines, etc.). In future, it is intended to provide lines in all four standard editions, which will also permit a comparison between them. The aim of all this is to make it reasonably simple for the user to find the text surrounding the painting, in at least one of the main printed editions.
The breakline is a powerful tool in the statistical analysis of the text of the Shahnama and in discussing such questions as the placement of images on the page, and the frequency of illustration in a given manuscript. The concept of the break-line belongs to Farhad Mehran, who has developed it in three publications: Farhad Mehran, “Frequency distribution of illustrated scenes in Persian manuscripts”, Student 2, no. 4 (Neuchâtel, 1998), pp. 351-79; “Missing paintings in dismantled Persian manuscripts”, Student 4, no. 1 (2001), pp. 61-78; and “The break-line verse: the link between text and image in the ‘First Small’ Shahnama”, in Shahnama Studies I, ed. Ch. Melville, Cambridge (2006), pp. 151-70.
The break-line also provides a useful method for identifying the correct title of the painting.
Before one can compare the way in which Persian artists depicted different scenes in the Shahnama, it is of course necessary to be sure that one is comparing like with like. At present, it would not be unfair to state that, even if not actually inaccurate, scene titles such as ‘Battle between the Iranians and the Turanians’, or ‘Ruler enthroned’, are not helpful in identifying the actual moment in the story that is being represented. In addition, it does matter, from this point of view, whether Farud is killing Rivniz or Zarasp, particularly if one wishes to compare the artist’s depiction of the scene with Firdausi’s text. In the past, many books on Persian miniature painting considered only the picture itself, and not the context, which was often even cropped from the illustration.
Taking Mohl’s printed text as the reference edition, the Shahnamahas been broken down into chapters (or episodes) using the scheme established by Wolff, and these subdivided into sections (or scenes), following Mohl’s own subheadings. These are delimited by the lines at the start and end of each scene. As a first approximation, any painting located - by its break-lines - within those lines, is given the associated scene title. These titles are then numbered, and used consistently for all other pictures located within the same passage of the text.
Naturally, this simple procedure has to be modified in practice. In the first place, it is frequently the case that a painting is positioned more or less outside the ‘correct’ place in the text, and then of course it is necessary to name the picture according to its contents, rather than according to its position.
Similarly, but more problematically, the image may not be very closely related to the text in which it occurs. This may be either because it has been painted later, by an artist ignorant of the context, or taken from a different source and pasted into the manuscript, or simply because the painter’s interpretation of this scene may show considerable departures from what might be considered canonical. Such idiosyncratic paintings are most common in later manuscripts from Qajar Iran, or in Indian or Kashmiri manuscripts, though the problem is not exclusive to them. In such cases, the picture would be normally be given the title that might be expected from the context and the subject of the poem. This allows the rogue or aberrant image to be seen grouped together with the other more normal representations of the scene, rather than hidden behind a private title assigned to it on an exclusive basis.
Thirdly, the list of titles needs constant revision and updating. While there are good arguments for retaining generic titles, such as ‘The fire ordeal of Siyavush’, or ‘Rustam kills the White Div’, to describe an unfolding scene, it is often necessary to make further subdivisions and to identify particular ‘moments’ in the action. The need for such further subdivisions is driven primarily by the discovery of paintings that refer to different moments within a scene, and especially when any one manuscript contains more than one picture from a given scene: they cannot both be given the same title. This arises most naturally when scenes are short and full of action, as for instance in the single combats of the Eleven Champions. It is also the case, however, that many of the scenes in Mohl’s edition (as determined by his section headings) are extremely long, and some further refinement is inevitable.
The need for subdivisions thus arises out of a closer scrutiny of the relationships between the iconography of the image and the language of the text, which is one of the main rationales for the Shahnama Project.
It remains a matter for discussion how far this subdivision needs to be taken, and at what point it is more useful to preserve a generic title and make broad comparisons across the range of associated images, than to break the representations down into smaller groups.
At present, it must be understood that the list of scene titles is not fixed, and indeed, that logically it cannot be finalised until every existing Shahnamapainting has been recorded and identified. The subdivision of scenes may inevitably require the revision of titles already given according to the broader generic definitions. A complete standard list of titles can thus be regarded as a future by-product of the Project. These issues have been addressed in the PhD dissertation of Amin Mahdavi, “An event-driven distribution model for automatic insertion of illustrations in narrative discourse: A study based on the Shahnama narrative”, vol. 1 (Edinburgh, 2004), and will be discussed in his forthcoming paper in the proceedings of the second Edinburgh Shahnamaconference.
Every effort has been made to employ a consistent spelling for the names of the characters appearing in the Shahnama, and for other words and terms transliterated from Persian. There are many rival systems and possible alternatives for the correct spelling of names. The choices adopted here are not intended to be regarded as definitive, but I trust at least they will not prove too divisive.
The text of the Shahnama as written by Firdausi excluded, so far as we know, various other related epic cycles, generally pursuing in more detail the exploits of some of the leading characters in the ‘Sistan cycle’ of the National Epic. The most substantial of these are the Garshaspnama, composed by Asadi Tusi (a Khurasanian compatriot of Firdausi’s) about half a century later, and the Barzunama. These, and shorter epics, are often incorporated into later copies of the Shahnama, either in whole or in part, and scenes from them are often illustrated. They have thus been recorded along with the other miniatures investigated by the Project, but they still await more detailed classification.
Volume 6 of the edition by Dabir-Siyaqi conveniently contains a text of three of these related epics, first the ‘Dastan-i Jamshid’ (pp. 1-43), second, the further feats of Rustam, ‘Sar-guzasht-i Rustam’ (pp. 44-76), and finally, more importantly, the Barzunama, or story of Barzu son of Suhrab, ‘Sar-guzasht-i Barzu pisar-i Suhrab’ (pp. 77- 246). The classic account of the Persian epic narrative tradition is Zabihallah Safa’s seminal work, Hamasa-sara’i dar Iran (Tehran, 1954); see also M. Mole, “L’Epopee iranienne apres Firdosi”, La Nouvelle Clio 3(1953), pp. 377-93.
For the moment, all illustrations that depict these various epics interpolated into the text of the Shahnama have simply been grouped together, according to the story, but no effort has yet been made to sort them according to the scenes they depict. In the absence of standard critical editions for most of these epics, and in view of the considerable variations found between different manuscripts, this represents a substantial task for the future. The Barzunama is the focus of two papers, by Gabrielle van den Berg and Eleanor Sims, published in Shahnama Studies I (2006). Work on these ‘secondary epics’ is now being carried out at Leiden University by the Project directed by Gabrielle van den Berg, ‘The Persian Epic Cycle and the Shahnama of Ferdowsi’.
At present, only minimal bibliographic references are contained in the website. Future updates from the archive will seek to rectify this. It remains the intention to supply details of the main publications that refer to collections, manuscripts and individual paintings, including their provenance, exhibition history and previous publication. This work is still in progress.
It will seem a bizarre paradox that while the Shahnama corpus contains thousands of records not to be found in the “Preliminary Index” of Jill Norgren and Edward Davis, the archive does not include a large number of the paintings they catalogued a quarter of a century ago. Why is this?
Our aim has been to provide a global coverage of the illustrated manuscripts of the Shahnama, and this remains the final objective. In practice, however, and especially given the guiding principle that we would only include manuscripts that we have seen for ourselves, a truly global coverage has not yet been achieved. As in other aspects of the database, much more information has been collected than is presented here, largely because the manuscripts have not been examined first-hand, and important information is either unavailable or still not verified.
As a result, the current archive concentrates on the main European collections, starting with those in Britain, and also in Tehran. It was clearly most efficient to work in places with a concentration of manuscripts, such as London, Dublin, Paris, Berlin and St Petersburg, before moving on to smaller or less accessible collections.
The main areas under-represented or entirely absent from the current archive are (1) North America; (2) Turkey and (3) India. The reasons are different in each case, although they have in common the fact that they are all relatively distant and require a substantial period of travel and residence to yield results.
A certain amount of information has been obtained from the rich collections in the USA, but more work is needed, and further progress is being made in the second phase of the project.
Istanbul houses perhaps the richest and most important collections of illustrated Shahnamas in the world, and the absence of much of this material from the database is regrettable. However, given the long-standing and almost completed work of Filiz Cagman and Zeren Tanindi on the Topkapi manuscripts, the decision was made to wait for their catalogue to appear, as it seemed pointless to duplicate research already being done by two such fine specialists on the spot. We can only hope that their results will be published before too long. The University library’s important collection was still not accessible in 2003, following the damaging earthquake. Some useful progress was however made in the Turk ve Islam Eserleri Muzesi.
As for India, it was clear from an exploratory visit (by Amin Mahdavi) that it was going to be very difficult to gain the access to the manuscripts necessary to work on them properly, and particularly to obtain photographs. As issues of conservation are involved in the case of many of these manuscripts, it is rather a matter of urgency that they are catalogued and documented before they deteriorate further. The recent publication of the catalogue of the collection in the Raza Library, Rampur by Barbara Schmitz and Ziyaud-Din A. Desai, goes some way to meeting the need for accurate information, but there is much more to be done.
A start on the collections in Pakistan has been made by Dr Zahra Hassan-Agha, particularly for the library of the Punjab University in Lahore, and we are hoping to make further progress during the coming six months.
In addition to these main geographical gaps and concentrations in our data, we have not systematically pursued examples in private collections, nor surveyed the records of the main auction houses, whose sales catalogues provide a valuable but intimidating volume of information on the existence and movement of Shahnama illustrations.
These have posed a particular problem of classification and cataloguing. Detached folios from unknown or unidentified original manuscripts will be found within the collections in which they are currently located. The same principle applies to dispersed folios from known manuscripts - many of which, necessarily, are among the finest and most celebrated manuscripts created, hence their commercial appeal. At the moment, folios from famous royal manuscripts such as the Great Mongol (Demotte), the Shah Tasmasp (Houghton) and the Shah Isma‘il II Shahnamas, and others such as the delightful “Big Head” Shahnama, are only included incidentally as they are found. It is envisaged that in future, these will be presented as distinct ‘reassembled’ manuscripts once more, as indeed some of them have been in recent scholarly publications. We also hope that it will prove possible to match dispersed sheets not presently identified with folios scattered in other collections.
The images presented on the website are of varying quality and resolution. This partly reflects their source, as they have been supplied to the Project directly on CDs, on high quality ektachromes, 35 mm slides, or photographed by members of the Project when kindly given permission to do so. On the database, images are stored at three sizes: thumbnails (long size 300 pixels), medium (800 pixels) and large (1600 pixels). On the website, images appear also at three sizes: small thumbnails (70 pixels), large thumbnails (300 pixels) and medium size (800 pixels). The latter can be magnified or enlarged for viewing purposes, but with no increase in resolution.
The colour values of the pictures are generally very inaccurate, compared with the original paintings. This is an inevitable consequence of every stage of their reproduction, from first photography and developing, to any scanning process that has taken place, and finally from viewing the image on screen. For detailed study, there remains no substitute for consulting the original manuscripts, and indeed it is an aim of the Project to encourage manuscript research, with all its rich possibilities. Nevertheless, the availability of so many images on the website should reduce unnecessary resort to manuscripts, and so help to preserve them.
It was been a fundamental principle of the Project that any picture is better than none, and no apology is made for the large number of poor images, whether black and white, out of focus or at very low resolution.
However, it is also axiomatic that the aim of the Project is to combine study of the paintings together with the surrounding Persian text. For this, a minimum requirement is that the text should be legible. I therefore make a plea, that in the future development of the website, a general consensus will emerge, that the images should be as good as possible. So far, only a few Collections have been willing for their images to be of high quality, but at present this is not reflected in the default presentation of the pictures. A document has been prepared concerning the reproduction of images for the Project, and is available on request.
The chief limiting factor on the quality of the images, apart from the practical question of their size and the speed of their appearance on screen, has been understandable concern over the misuse of the images, i.e. downloading and reproduction without permission. At present, the relatively poor quality of the images is the best safeguard against this, and is considered preferable to spoiling the images or watermarking them, or some other more elaborate intervention.
All images remain the copyright of the institution or individual in whose keeping the originals are preserved. The copyright of all photographs taken by members of the Project, or alterations to images supplied to the Project, are assigned absolutely to the institution where the originals are kept. Every image viewed on screen is accompanied by the appropriate copyright statements and credits. The details are provided of how to contact the copyright holders for permission to reproduce images, or how to order photographs.
In no case should the Project be contacted for permissions or for orders of images, although of course we will be happy to supply further information on the appropriate contact details if necessary.
In addition to the general interest and considerable assistance extended to the Project by all the librarians and curators of manuscripts represented on the database, with whom it has been a real pleasure to correspond and to meet, I would like to thank a number of colleagues for their particular personal contributions and assistance at different times over the last eight years. Such thanks go to Seyyed Mohammad Ahmadi Abhari, Raf Alvarado, Michele Bernardini, Sheila Canby, Massumeh Farhad, Selina Fellows, Dr Jurgen Wasim Frembgen, Linda Komaroff, Mary McWilliams, Francis Richard, Karin Ruhrdanz, Sharareh Salehi, Shreve Simpson, Robert Skelton, Abolala Soudavar, Lale Uluc, Olga Vasiliyeva, Muhammad Isa Waley and Elaine Wright.
This has been a group project in which many people have contributed in different ways over a long period. Several decisions about the design of the database, and later the website, have been reached after long discussions, and much of the data has been assembled and worked on by more than one person. Nevertheless, within the group, different specific tasks have been the responsibility of different people, as follows:
The original concept for the database came from Farhad Mehran. The database itself was created in Access 97 by Iain Murray, at that time an undergraduate at Pembroke College, Cambridge. Invaluable technical expertise was been available throughout the first phase of the Project from Amin Mahdavi.
Initial work on the website was carried forward by Peter Batke, a former colleague of Jerome Clinton in Princeton, who produced Alpha and Beta test versions for distribution, and first pointed the way to the possibilities ahead. The website as it now stands is the work of Fetherstonhaugh Associates.
In the second phase of the Project, the features of the website have been considerably enhanced and developed by Dan Sheppard, with a view to reproducing many of the functions of the original Access database (especially for data entry) for the internet. Technical support for the database has also been provided by Raad Al-Rawi at CARET, Cambridge.
To avoid duplication of effort and confusion when merging the different copies of the database, responsibility for collecting and processing data was assigned on a geographical basis.
Amin Mahdavi, working half-time throughout the five years of the Project, collected data from the British Library in London; the John Rylands Library, Manchester; the Chester Beatty and Trinity College Libraries, Dublin and the main libraries in Iran. He also collected data in India, which has yet to be completed.
Christine van Ruymbeke, working half-time in the first two years of the Project (1999-2001), started the work on the collections in Cambridge, Oxford and Paris.
Gabrielle van den Berg, working part-time in the second part of the project (2001-2004) was responsible for data collection in the German-speaking countries, especially Berlin and later also Munich and elsewhere, and including Copenhagen.
Firuza Abdullaeva, working full-time throughout the second half of Project (2002-2004) has been responsible initially for data collection in St Petersburg and the Russian speaking countries, but later more widely for completing the work in Cambridge, Oxford and Paris, and many other locations, from Yerevan to Venice. It is regrettable that it has been possible to show so few of the paintings from the two important collections in St Petersburg; this is still the subject of protracted negotiations.
More recently, Afsaneh Firouz has researched data from the collections in Harvard and Boston, and in Geneva, and Laura Weinstein has worked on collections in New York and Washington D.C.
Finally, Charles Melville has collected data more randomly as the opportunity arose, in the UK, Europe, USA and Asia, particularly in Istanbul, Tashkent, Vienna, Philadelphia, Los Angeles and the main Italian libraries.
Darya Vasiliyeva (at that time a student in St Petersburg State University), Majied Robinson (at that time a student at Peterhouse, Cambridge), and Will Ward (St Antony’s College, Oxford) gave valuable help checking break-lines and recording manuscripts during their summer vacations.
In each case, preliminary responsibility for processing the data (checking break-lines, assigning scene titles, descriptions) rested with those who had collected it. Such comments are generally identified by the initials of the person concerned. All final editing of the data, and preparing the information for uploading to the website, has been the work of Charles Melville, who remains responsible for any errors and defects that remain.
Much remains to be done, both in the cataloguing of manuscripts and the acquisition of images, and in creating a more interactive structure for the website, permitting users to frame their own queries, and to contribute data online.
There are many important collections still to catalogue fully, especially in the U.S.A., Turkey, and India, and the value of the database as a research tool will only be fully realised when it is as complete as possible.
A second phase in the development of the website will extend the possibilities for entering data online and facilitate the exchange of ideas and information between individual users, and between users of the site and the Project itself. We are hoping also to provide printed Persian text and an English translation to accompany the display of individual images.
In addition, we intend to support the publication of a series of illustrated catalogues of the Shahnamas in different collections, starting with those in Cambridge, Oxford and Berlin. The publication of an important manuscript of the Shahnamain Oxford, Ms. Ouseley Add. 176, by Firuza Abdullavea and Charles Melville, is currently in press with the Bodleian Library.
Finally, it is hoped that we can use the materials assembled on the website to develop an educational as well as research resource, for instance in illustrating developments in the Persian arts of the book, and especially, calligraphy, illumination, binding and other elements of codicology. A small start has already been made in this direction, and we hope to attract additional funding to support this work, which lies somewhat outside the immediate purposes of the Project.
We invite all librarians, curators and colleagues interested in using and developing this resource, to send any corrections to the data presented here, or fresh information about collections, manuscripts or paintings that they have worked on, to allow us to supplement and extend the database. It is for the use of everyone, and it will therefore be most use, if it properly reflects the needs and interests of those using it.
Once the facility for entering information online is fully operational, Museums and Libraries and individual scholars or collectors will be able to update and correct the information about their own manuscript collections, and will perhaps be tempted to provide better quality images to be displayed on the website!
Please let us know how you would like to see the website develop.
For the present, the project can be contacted by email at: firstname.lastname@example.org or comments can be sent directly to
Dr Charles Melville
Faculty of Oriental Studies
University of Cambridge
Cambridge CB3 9DA