| THE STUDY OF TAFS¯IR IN THE 21st CENTURY: |
| E-TEXTS AND THEIR SCHOLARLY USE |
| Andrew Rippin |
| University of Victoria |
There is a CD-ROM which may be purchased online for $30 from the IslamiCity
Baz@r which includes the Qur’¯an in Arabic, plus translations into
English, Malay, Turkish, French and German (fully searchable), a version of
the Qur’¯an recited in Arabic, and the Arabic text of the tafs¯irs
of the Jal¯alayn, Ibn Kath¯ir and Qur.tub¯i. Another such
CD-ROM, available more easily in the Middle East and which costs the equivalent
of only $10, provides the recited Arabic Qur’¯an plus five classical
tafs¯irs (Jal¯alayn, Ibn Kath¯ir, Shawk¯an¯i,
Baghaw¯i, and Bay.d¯aw¯i) and three related works from
other genres. If you bought all these texts in their traditional hardbound book
form, not only would they take up a lot of space on your bookshelves but also
they would cost at least 10 times as much. But the question is, what are we
getting for our money? That is the issue which I wish to address in this paper.
Before launching into that, however, it is best to review the inventory of texts that are available in electronic form from one or more sources. When I first proposed this paper in October for the Annual Meeting of the American Oriental Society, I listed some ten e-texts of tafs¯ir that I knew of. By March that list had swollen to well over 100 texts (see Appendix 1), mainly due to the willingness of Avraham Hakim at Tel Aviv University to share with me the mine of information which he has about some remarkable CD-ROMs from Jordan and Iran. The large number of Sh¯i‘¯i tafs¯irs in the list reflects two CD-ROMs from the Nashr-e .Had¯ith-e Ahl al-Bayt Institute containing some 900 volumes by Sh¯i‘¯ authors, the first containing the full texts of works up to the time of .T¯us¯i, the second providing a full index to the corpus of later printed Sh¯i‘¯i material, such that one can find the page and line in a book by searching for a word or a string of words (thus those texts marked “Index Only” in Appendix 1 are really not of much use unless one has access to the printed text as well). Some of the Sunn¯i texts are found on a CD-ROM from Markaz al-Tur¯ath in Amman called Maktabat al-Fiqh wa-U.s¯ulihi , once again a disk which contains 900 volumes. We have also discovered that that institution has published a Maktabat al-Tafs¯ir wa-‘Ul¯um al-Qur’¯an which I have not actually seen yet but the approximate list of the contents is provided in Appendix 2 (see http://www.turath.com/ [updated 01/28/05]). Those two sources from Iran and Jordan thus account for a majority of the available texts, but there are a few more to be found in various odd sources, and many of the texts are more easily and sometimes more cheaply available from other vendors. Other works mainly come in two ways-commercial Qur’¯an CD-ROMs with tafs¯ir s included, and free downloadable texts from the project known as al-Mu.haddith based in Paris (Markaz al-Tur¯ath also provides some texts for free, but no tafs¯ir s as yet apparently). More Sh¯i‘¯i works are available through the work of the Computer Research Center of Islamic Sciences in Qom, published by Noor Software, who have produced a CD called Anw¯ar al-N¯ur with various tafs¯irs included.
Of course, the production of the inventory of these works has been a task in itself and realizing what this list represents needs a moment's attention. All of these books are immediately accessible on one's computer. All come with some sort of “front end” that allows either for full text searching (in the case of al-Mu.haddith texts and those from Markaz al-Tur¯ath and Iran) or for quick reference indexed according to Qur’¯an verse. It is worthy of note that the tafs¯irs of Ibn Kath¯ir, .Tabar¯i, Jal¯alayn, and Qur.tub¯i, and the A.hk¯am al-Qur’¯an works of Jass¯as, Ibn ‘Arab¯i and that ascribed to Sh¯afi‘¯i are available for use free online at the Sakhr Web site-that is, you access the web site with your Arabic-enabled browser, type in the s¯ura and ¯aya number, and there you have the tafs¯irs. No more searching through obscure indexes! In most instances, it is possible to “cut and paste” between these texts and one's word processor, although that brings up a technical point. The basic commercial CD-ROMs tend to incorporate their own Arabic support. The CDs of the Sh¯i‘¯i texts from the Ahl al-Bayt Institute run primitively under DOS (not a DOS window) and allow printing only if you happen to have an old Epson dot matrix printer still lying around. Other CD-ROMs, however, do require Arabic Windows. Regardless of the CD, any cutting and pasting between programs will require that Arabic system support. Virtually all the programs run on Windows: Mac support appears quite limited.
In terms of additional texts, I might point out that there are few Web sites around which have graphic reproductions of pages of certain printed tafs¯irs: these really are of minimal use (except to the extent that one may verify the accuracy of true e-texts) and they are not included in Appendix 1, but an example may be seen at the Ahlul Bayt Digital Islamic Library Project (http://www.al-islam.org/sources).
But the actual issue which I wish to address in this paper is the value of these texts to scholarly work. Certainly the convenience factor is immense. No more searching through the seemingly endless volumes of Fakhr al-D¯in al-R¯az¯i's Tafs¯ir in order to find the appropriate page for the treatment of a certain Qur’¯an passage! The ability to be able to search the full text of a tafs¯ir will revolutionize many studies: Terminology becomes much easier to extract, indexes are a snap to compile, and so forth. The texts from al-Muhaddith can be linked into an Arabic-English dictionary such that you can click on most words in a text and get a meaning, although the parsing of the words is less than perfect. But...
The basic inaccuracy of the available texts is certainly problematic. This manifests itself in a number of ways: simple textual errors, unexplained textual changes, and lack of clarification in text-comprehension matters and in text-critical matters. I will address each of these factors in turn.
First, that these texts should have textual mistakes should come as no particular surprise. Many of the works are extremely long and there is no reason to assume that the printed editions upon which they are all based are error-free either. I suspect that some of these texts have been transformed into their electronic versions through Optical Character Recognition processes (rather than being inputted through simple keying). This, of course, speaks highly of the abilities of the technology and how much it has progressed over the last 10 years-the very fact that this can be done with Arabic strikes me as astounding. However, OCR is never 100 percent accurate, not even in English, a much simpler orthographic form than Arabic. So, mistakes are likely to creep in. And it is true that mistakes are quite common in these texts, although efforts are certainly being made in various places to proofread them. Nor should the problem be exaggerated. Some of the texts I have checked are remarkably accurate. The worst I noticed was the Sakhr version of .Tabar¯i (in the online version, but that is presumably the same as the CD version). Of course, the text of .Tabar¯i's tafs¯ir is a tough one to conquer. The printed edition by the Sh¯akir brothers (Cairo, 1954-68) is clearly the best-but it is incomplete. There is some evidence of eclectic editing on the part of the people responsible for the three electronic versions of the tafs¯ir which I examined (Sakhr, Ariss, Mu.haddith), putting the various editions together with the aim of creating the “best” text. But the Sakhr online edition has a substantial number of typographical errors on just the first few screens; I counted eight major examples in the first twenty full-screen lines of text of the tafs¯ir on S¯urat al-F¯ati.ha , whereas both the Mu.haddith and the Ariss texts had two different mistakes in the same segment. That the errors differ between the various electronic versions that are available does at least suggest that the work of creating the e-texts was done independently and they have not been created simply through electronic copying. But it also is a fact that there are a substantial number of textual mistakes to grapple with in some instances.
In the category of unexplained textual changes, I include both additions and minor deletions. These can affect texts in relatively insignificant ways-whether .salla ll¯ahu ‘alayhi wa-sallam comes after the name of Mu.hammad (it is omitted in the Sakhr and Muhaddith versions of .Tabar¯'s tafs¯r at the very beginning of the comments on S¯urat al-F¯ati.ha , but it is present in the Ariss version) does not affect our reading of the text-but the omission and addition of subheadings, for example, may well change things more substantially. This is to be noted in the tafs¯ir of Ibn Kath¯ir, in which headings that help the reader in the printed text are left out (I must admit that I do not know whether they are included in the original manuscript tradition of this tafs¯ir ). And there are headings in the texts of Jass¯as and Ibn ‘Arab¯i which have been added-probably to make their application to questions of legal issues more apparent within the context of the Sakhr publication in which they appear (that is, a collection of fiqh texts). Most serious of all such changes, however, is the actual loss of text. The major instance of this which I have noted is in the Sakhr and the Ariss versions of .Tabar¯i's tafs¯ir . There, the introduction to the text is not present. Of course, it is apparent why this would happen. The tafs¯ir has been indexed to the specific verses of the Qur’¯an upon which it comments. .Tabar¯i's introduction does not apply directly to any verse, so it is omitted. The introduction is present in the version from al-Mu.haddith. It should also be noted that in both the Sakhr and Ariss versions the absence of the introduction has resulted in a renumbering of the .had¯ith reports, a scheme which was so helpfully added in the Sh¯akir printed edition. To add to the confusion, the two electronic versions have then provided different numbering schemes: the Sakhr numbering starts at 114, which seems to provide an implicit admission that the introduction is missing although the numbering is different from that in the Sh¯akir edition (in which the report in question is number 138); the Ariss version starts at number 1, so there is no trace of the missing hundred or so pages of text in that version at all.
Other problems of this nature may not be so severe, but they do raise a fundamental question. It may be thought bad enough that the transformation from book form to electronic form changes our sense of what an author is doing and what the reading experience is about. But it also clearly makes matters of form and sequence within a work much harder to perceive and much easier to gloss over. A minor example would be from Tafs¯ir al-Jal¯alayn . In many printed editions (but certainly not all) the tafs¯ir on S¯urat al-F¯ati.ha is found at the end of the book, because Ma.hall¯i apparently wrote it (plus the commentary on a few verses of S¯urat al-Baqara ) after having composed the commentary on s¯ura s 18 through 114. Suy¯u.t¯i then completed the entire text by providing the commentary on s¯ura s 2 through 17, and he left Ma.hall¯i's commentary on S¯urat al-F¯ati.ha at the end of the work (see M. H. al-Dhahab¯i, al-Tafs¯ir wa'l-Mufassir¯un , Cairo 1976, i, 334). In the CD-ROM versions, the intimate tying of the tafs¯ir to the text of the Qur’¯an means that the tafs¯ir on S¯urat al-F¯ati.ha is, of course, where it belongs, but the compositional history of the text has been obliterated.
Clarification of text-comprehension matters is more delicate. Paragraphing and punctuation (if even present) are notoriously arbitrary in many printed editions. Not a great deal of effort seems to have been spent on helping these matters in the e-texts; in some instances matters are definitely made worse by misplaced commas and quote marks. The issue of vowelling is likewise double-edged. We have all had the experience of reading what appears to be a well-edited text in which obvious or irrelevant i‘r¯ab is included, while tougher pieces, where somebody else's interpretation of the text might help one's reading, are noticeably undecorated. Vowelling varies among the e-texts-sometimes it is occasionally present, other times there is none. Some of the Sakhr legal texts can be fully vowelled by the click of a button-but I believe this is done mainly by automation, and I certainly would not want to trust it! The printed .Tabar¯i edition by Sh¯akir, once again, is not a perfect example of the use of punctuation or vowelling, but on many occasions those formalities do seem helpful and their general omission in the e-texts is lamentable.
The matter of the lack of textual critical apparatus is the last crucial matter, and it is, in fact, the common aspect underlying all of the foregoing complaints. In no case are these e-texts new scholarly editions. They are all based upon existing printed texts. Sometimes the e-text does not even make clear which printed edition is being used. But, more crucially, any scholarly apparatus-reference to manuscript readings and the like-have, in general, been dropped out. There is absolutely no real reason for this to be so-incorporating the printed footnotes would clearly be possible-but it has not been done.
This is not a problem unique to Arabic texts by any means. A recent article in Religious Studies News (Offline 67) (online at http://purl.org/CERTR/Offline/off67.html[dead link]) puts it very well:
The answer to this problem, according to the authors of this article, is for scholars to take back charge over the production of such texts. I am not sure this is a viable option in a field such as our own in which the scholarly community is small and commercial endeavors already command the field. I would note with pleasure, however, Michael Carter's S¯ibawayhi project (something certainly not irrelevant to tafs¯ir studies) found at http://www.hf.uio.no/east/sibawayhi/HomePage. That project will provide an electronically stored text that includes variants, glosses, translations, manuscripts copies, and references to secondary literature: a complete critical library attached to the text. As this project indicates, the possibility of accomplishing a scholarly version is there. As far as tafs¯ir texts go, the best place to start would certainly be with the Mu.haddith texts, since those are directly available in a pure text form, rather than in a proprietary compressed form as with most commercial products. Of course, as I am sure Michael Carter would tell you, such efforts at creating a truly useful electronic text requires a great deal of work. Determining if it would be worth the time and effort would be tough.
In sum then, there are a large number of texts of tafs¯ir available in electronic form. The study of tafs¯ir is definitely ready of the 21st century. But, clearly, these texts will not replace the printed word in the near future, at least as far as scholarly work goes.