Akbar Amat, Arienne Dwyer, Gülnar Eziz, Alexandre Papas, C.M. Sperberg-McQueen
Version 1.0 (2018-06-03) how to cite this work
This handbook aims to guide the reader and analyst of Central Asian Chaghatay manuscripts. Both practical and theoretical, our approach is drawn from a collaborative work on the manuscript material of the Gunnar Jarring Collection at the Lund University Library. Moving among various formats and genres, it occurred to us that, instead of listing general guidelines, we could propose a gradual and multidisciplinary process of annotation, which leads the reader from the first look at the book or document to its scholarly analysis. Each section of this manual corresponds to a specific set of actions, illustrated by a scanned sample from the Jarring Collection, on the model of Gacek (2009), and provides resources to solve potential issues, including bibliographical references for readers who need to extend the set of actions in relation with their discipline. The second section provides an introduction to the codicology of Chaghatay manuscripts, and allows to use the main elements of description to situate the manuscript within the multilingualism and literary production of Central Asia. The thrid section presents the different items that usually structure the contents of the writing, in order to obtain a complete overview of the source as well as a general understanding of its contents. In the fourth section, we develop tools and methods to facilitate the edition as well as the translation of Chaghatay books or documents, so that readers can avoid a number of philological pitfalls and reduce misinterpretations. The sixth and last section goes a step further towards the manuscript annotation by offering examples of scholarly analysis - especially but not exclusively historical and anthropological - of writings according to the genre or sub-genre they belong to.
Let's look at some sample pages of Turki manuscripts from the Jarring Collection, Provs. 41 and 365. The terms introduced here are explained in detail in the rest of the manual.
First, take a look at [Prov. 41], the Risale of the Bakers (رسالۀ نانوایلق). It has are double incipit (opening-words) pages, a ruled border with frames inside of it. On top middle of the pages, there is an initial textual formula which has frame of its own. However, in the Jarring collection, apart from some risale manuscripts, most others don't have rule-border and frames. Under the formula there is the rubricated (red) title that says, رسالۀ نانوایلق بو ترور 'This is the risale of the bakers'. Many manuscripts in the Jarring Collection lack titles; where titles appear, they are often written in black ink. Under the title there is the standard superscription which starts with Surāh Al-Fatiha, the first verse of the Qur'ān. The Turkic text starts after the superscription, and there is a catchword on the first incipient page. (A catchword is found at the bottom of one page that repeats the first word on the following page.)
The Quranic text was written in a different script from the Turki text: The Turki text was written in a script calledNastaʿlīq, while the Quranic texts were usually copied in Naskh or Thuluth scripts.
Folio 4 of Prov. 365 is a a poetry manuscript in the Jarring Collection. The page's text was written in Nastaʿlīq script, and there are mid-line breaks between verses. The poem is read right to left, from the first verse to the next separated by mid-line break. The subtitles were rubricated, and one can identify catchwords in the lower left corner of the page, which correspond to the beginning words of the first line in the next page.
Some manuscripts such as Folio 56 of Prov. 365 have marginalia, that is, additional writing around the edges of the pages, serving as notes (and written by the current scribe or a later reader). On this page's religious poem, these notes are marginalia glossing, i.e. translations or explanations of unfamiliar Arabic or Persian terms for the Turki reader.
These are some of the common features of these manuscripts.
No matter where in the world, manuscript production, emendation, conservation, and variation share many common features. Some of these features were introduced above, and include using different scripts for religious vs. secular texts, and adding borders, marginalia, and or catchwords. Another unavoidable feature of manuscripts anywhere in the world is their instability, that is, changeability. This section "Textual instability in a Manuscript Culture" has been adapted from a 2016 version of the Harvard Chaucer project's METRO resources.
When we talk of "text", below, we are generally referring to the content (what the manuscript says). “Handwritten texts are subject to a number of errors and inconsistencies. We will first describe some common errors you should be aware of, and briefly outline the ways in which this has affected modern studies of medieval texts.”
Scribal Error and Emendation
“A scribe might omit a word or write the wrong word when copying a manuscript. Scribes would sometimes inadvertently skip a line, an error known as "eyeskip." Frequently, either the scribe or another reader of the manuscript corrects the error....If a scribe's error was too big to correct with an insertion, he might erase the text by scraping a thin layer of parchment away, and write over the erasure.” (METRO 2018)
Degradation and Loss of Text
“Many medieval texts do not survive in complete versions. Partial or complete loss of a text can be caused by a number of factors. Manuscripts are fragile, and can degrade over time if not stored and handled correctly.... Sometimes single leaves or folios of another text were used in the bindings of later books." (id.) Many Turki manuscripts are missing initial or final pages. It is also common for several different works (on different topics written by different scribes) to be bound together, possibly incompletely, in one composite volume. Finally, book owners may make use of any partially blank page as scrap paper (for example, Prov. 351, a medical handbook, has a list of someone's sheep); in so doing, the manuscript text may be partially obscured.
Mouvance and Variance
“Different versions of a text frequently circulated..., and not all scribes had access to the same exemplar when they copied a text." Nineteenth and twentieth century European textual scholarship focused on determining which version of the text was "closest to the original, and demoting any alternate readings to "variant" status. Editors were supposed to create, from the body of manuscript evidence, a singular text that bore the most resemblance to the original author's intention." (id.)
"In 1972, the French structuralist Paul Zumthor made the case for what he called "mouvance" as an approach to editing and thinking about medieval texts" (Zumthor 2000 [1972]). For many medieval texts, especially those in the vernacular, attempts to find the "original" text obfuscate the actual mode of circulation of those texts. Texts were adapted and modified at will, and the intention of the original author did not have the same weight with medieval audiences that it does with modern ones. Extending Zumthor's work, Bernard Cerquiglini published Eloge de la variante (In Praise of the Variant) in 1989, in which he makes the case for embracing the variant versions of medieval texts. For an excellent, in-depth introduction to mouvance and variance, see Bella Millett's website ” (id.)
Central Asian manuscripts have been less studied than European ones, and may have been produced later, yet the approach of mouvance ultimately facilitates the analysis of multiple variant editions.
Codicology is the study of books as physical objects. It concerns materials, such as papers, inks, and techniques used to make books, also including their binding.
Describing the tangible aspects of a manuscript includes identifying visible and measurable physical characteristics. The Jarring Collection metadata includes paper type, extent (page size and number of folios), format, binding, decoration, foliation, and ink color. Manuscripts may have different versions, and their appearance and characteristics depend on the copyist. Some details on each of these physical characteristics follow.
The size of manuscripts is measured as width by length, typically in millimeters. At least ten different sizes can be found in the Jarring Collections. Among the tiniest are fingertip-sized scrolls meant to be put inside an amulet worn around the neck; at the other end of the scale, some scrolls are 15 meters long. Likewise, bound manuscripts range from the size of the palm of one's hand to less-portable volumes A4 size or greater.
Some examples can be found here: 150x120mm [Prov. 4]; 180x110mm [Prov. 6]; 200x165mm [Prov. 7].
A Turki manuscripts may be bound as a book (usually rectangular), rolled up in a scroll, or consist of one or more loose sheets of paper. Some manuscripts are bound in a book form, with folios (sheets) bound or pasted together, usually with an outside binding. Folios are folded once and stitched together to make a book, such as [Prov. 351]. Sometimes several gatherings of folios are put together into a book, as can be seen in [Prov. 207]. Some are composed of one or two leaflets alone, such as [Prov. 222]. There are also long scrolls like Prov. 575 (about 15 meters).
Manuscripts may be bound (sewn or pasted together), or unbound (loose sheets). An example of a bound manuscript is [Prov. 7].
We can identify paper of indigenous, Russian, and Swedish origin in the Jarring Collection:
A small minority of the manuscripts have watermarks in the paper. These include Provs. 1, 135, 140 etc., as well as the ones with watermarks analyzed by Kim (2010: 60): Prov. 223, 226, 221, 225, 224, 220, 227.
The most common ink color is black. Some texts use only one specific ink color, e.g. black or brown. Some other texts use different colors for specify the elements of the text. For example, chapter and section headings and key words might be one color, and the body text another color. Red ink (rubrication) is commonly used for frames, tables, and sometimes emphasis.
In Prov. 2 for example, the frames are drawn in red ink. And some lines were overlined in red ink for emphasis. In Prov. 5 the title is rubricated too. In Prov. 327, green ink was also used for writing the colophon.
Some manuscripts in the Jarring Collection themselves describe how this ink is made (Dwyer 2017).
Pages are often counted by collectors and archivists; a manuscript's original scribe does not necessarily include page numbers in the original manuscripts. When written by original scribe or by the book's owner(s), pagination is indicated with Arabic numerals (see Prov. 5). Most of Jarring Collection, page numbers However, Latin page numbers also occurs on the top of the page. It might be written by a reader or a cataloger (see also Prov. 5).
Perso-Arabic scripts | Letter Forms | Diacritics | Abbreviations | Punctuation | Writing styles | Differences from MSU
Paleography is the study of handwriting, for the purposes of deciphering and analyzing not only the content of the text, but possibly also the scribal hands, and the cultural milieux in which these manuscripts circulated.
Chaghatay is written with a Perso-Arabic based script that resembles but is not identical to modern Persian orthography. The Persian alphabet itself includes the 28 Arabic letter glyphs, with four additional letter glyphs for sounds not found Arabic (the letters peh پ (IPA /p/), tcheh چ /č/, jeh ژ /ž/, and gah گ /g/). Chaghatay has the same 32 letter glyphs as in Persian, although combination rules are slightly different in Chaghatay. Diacritics, which are marks placed over, under, or through a letter to show that the letter should be pronounced in a particular way, are used similarly in Chaghatay and Persian, with the exceptions described below.
Chaghatay is written from right to left, and the orthography does not distinguish upper and lower case.
The Chaghatay alphabet and the transcription alphabet adopted in this manual has the following letters, in alphabetic order: alef (ا), beh (ب), peh (پ), teh (ت), theh (ث), jeem (ج), tcheh (چ), hah (ح), khah (خ), dal (د), zal (ذ), reh (ر), zain (ز), jeh (ژ), seen (س), sheen (ش), sad (ص), dad (ض), tah (ط), zah (ظ), ain (ع), ghain (غ), hamza (ء), feh (ف), qaf (ق), kaf (ک), gaf (گ), keheh (ك), ng (ڭ), noon+kaf (نک), lam (ل), meem (م), noon (ن), heh (ه), teh marbuta (ة, ـة), waw (و), yeh (ی), yeh (ي). Some scholars count hamza (ء) as a letter (Eckmann 1966: 27 and Polat 2004: 72) and some do not (Bodrogligeti 2001: 12).
Of these alphabetic letters, most represent consonants; vowels are represented always by glyph alef (ا), and often by the glyphs heh (ه), waw (و), yeh (ی). , yeh (ي); the latter four may also represent semi-vowels.
In contrast with Arabic and Persian, Chaghatay uses noon+kaf (نک) and ng (ڭ) for velar nasal sound /ŋ/; Mostly uses kaf (ک) for voiceless velar stop /k/, however, sometimes keheh (ك) also occurs in the final position.
Due to a lack of standardization, there are occasional inconsistencies in the spelling of some of the words in the same manuscript and sometimes spelling differs from manuscript to manuscript. In line 5 Prov. 369, which belongs to the late 19th century, Kashgar was uniformly written as کاشقر not as کاشغر, as in Prov. 327, line 4. There are some inconsistencies which began to appear in manuscripts written in the early 20th century. For example in Prov. 56 which belongs to the early 20th century we find both طرپیدین (Prov. 56, line 13) and طرفیدین (Prov. 56, line 1), and ژیغلاب (Prov. 56, line 9) while the more common from being یغلاب (i.e, in Prov 327, line 9, a late 19th century document), it was possible that the discrepancy had to do with the spoken form of the words being different from how they were written. Also in Prov 56 (line 3) the scribe wrote زمان as زامان adding an 'alef'.And the writer/scribe of Prov. 365 had the habit of writing the all the catchwords unpointed, which might be a calligraphic style of that time.
These letter forms are associated with particular script styles (e.g. Nastaʿlīq), which is covered in Script types below.
Words are not abbreviated. Commonly in Arabic manuscripts, a word is abbreviated by the first letter; this abbreviation of words does not occur in the Jarring Collection that we have seen so far.
Punctuation is not consistent in Chaghatay. Common punctuation marks occurring the Jarring texts are:
The four-dot mark indicates a verse break. It is used at the beginning and end of a verse, especially to separate verse from prose. It may occur at the beginning or end of lines, or in the middle of a page. See [Jarring_Prov_351_040 line 10-11] and [Prov_9_fol_007_v line 11-13]
The eight teardrop-spoked propeller asterisk indicates a decoration for title. See example [Jarring_Prov_24_023_r line 1] the title was written in between the two eight teardrop-spoked propeller asterisk. [Jarring_Prov_5_019_r] This mark occurs end of the title. This mark also occurs end of a poem. See [Jarring_Prov_365_fol_013_v line 2]. This mark occurs end of a prayer in Jarring texts, see [Jarring_Prov_5_005_v], [Jarring_Prov_5_009_v], and [Jarring_Prov_5_032_v] However, this mark did not occur consistently.
The period is a punctuation mark placed at the end of a sentence. However, this mark did not occur consistently in Chaghatay manuscripts until the later period (e.g. manuscipts. on Russian paper). For example, see this images: [Jarring_Prov_2_004_r line 6], the period only occur end of the first paragraph. See another example here: [Jarring_Prov_9_fol_006_r], the period only occurs in line 3. Period also occurs end of a the title in Prov. 5 [Jarring_Prov_5_023_r], [Jarring_Prov_5_023_v]
Dash: The dash, mostly with red ink, occurs on the top of names, prayers, and also highlighted questions, answers and important outline numbers. For example, [Jarring_Prov_2_004_v]
Rare punctuation
Colons: appear extremely rarely (we've found to date just one), preceding a direct quote: [Jarring_Prov_9_fol_013_v line 5] and [Jarring_Prov_5_012_v] See rubrication. This colon occurs beginning of dialogue: [Jarring_Prov_9_fol_013_v line 8]
Central Asian scribes use different calligraphic styles for different purposes: some styles were more appropriate for formal documents (such as legal contracts), others were more decorative (for illustrated manuscripts), while others were more everyday (such as narrative accounts). Here is a brief introduction to the types of writing likely to be encountered and their purposes. For more information, see the Further Reading section below.
The following Perso-Arabic scripts are used:
Most of Turki manuscripts were written in Nastaʿlīq (نستعلیق), although we find occasionally Thuluth, Naskh and Muhaqqaq, and for some letters Shikasta.
For an indigenous analysis of writing and script styles, see the Jarring Collection texts "Things necessary for writing" and "A Description of reading" (and Dwyer 2017 (pdf).
For list and description of all of the glyphs used in late Chaghatay manuscripts, see the next section, Glyphs.
The following discussion about easily confused glyphs assumes that the users of this manual are familiar with printed Perso-Arabic fonts and have less familiarity with handwritten scripts such as Nastaʿlīq which is the main script of the regions with Persian influence. By easily confused glyphs we mean that some of the handwritten letters can be confusing for or can be misread by users who nevertheless familiar with printed Perso-Arabic fonts. The glyphs to be discussed here belong to Nastaʿlīq script.
Naturally, the Arabic orthography also have orthographic variants - also differing only in the presence or absence of diacritics - that may or may not represent different sounds, that is, that are distinctive. For example ع (without a diacritic) and غ, or ح and خ. While distinctive in Arabic, these sounds were likely not distinctive in Chaghatay, so it would be logical for a scribe to use ع for غ, and ح for خ.
How should the omission of diacritics be handled in the transcription? There are two options: The glyph can be transcribed as written (e.g. without the expected diacrtics), and the transcriptionist can note (via a regularized orthography or a note) that the corresponding glyph with diacritic that is expected. This option requires the transcriptionist to use a standard set of non-diacritic variant Arabic character glyphs (such as qaf ق without the two dots). But not all such variant glyphs appear in Unicode. (For example, non-diacritic forms of intial yeh يـ, fa ف, and qaf ق are not (yet) in Unicode. If so, one can't type what one sees. The second option is to transcribe the missing diacrtics (e.g. in Prov. 365's catchwords) as if there were present, and note their absence. For either option, the transcriptionist should be sure to record both the expected and the actual glyph.
Again this can be ascribed to the measures scribes took to make their pen more fluid. Some of the cursive simplifications we have observed in the Nastaʿlīq and Shikesteh script are:
Simplifications of diacritics
1. Diaresis (two dots) abbreviated as a bar (line):
2. Three dots abbreviated as a bar:
Simplification of letter forms
1. The two denticles of sheen (ش) are sometimes reduced to a shallow arc or a slash, which is characteristic of Nastaʿlīq and Shikasta. 2. gaf (گ) was abbreviated to kaf(ک) in most places in most manuscripts, but Prov. 158 has (گ) for where we would expect (ک) (see Prov. 158).
Other script styles, like Naskh, do not typically use these simplifications.
Many readers will begin their exploration of Chaghatay as native speakers or learners of modern Uyghur and Uzbek. The modern Uyghur Perso-Arabic orthography differs in three main ways from written Chaghatay: in the representation of vowel values, vowel length, and (mostly velar) consonants. Additionally, written Chaghatay shows more intrinsic variation since it was not standardized, and used over a large area for many centuries. The orthographic differences are as follows:
The representation of vowels
Not all vowels are represented overtly in Chaghatay, and what they represent varies sometimes depending on their position in the word:
One glyph or series of glyphs can therefore represent several sounds or sound sequences. Deciding how to transcribe them is based on two clues: (1) the surrounding consonants (and vowels), and (2) knowledge of the language (or access to a lexicon).
The surrounding consonants matter, too. For example, اي adjacent to back consonants yields [a], while adjacent to front consonants yields [é] or [e]: e.g. ïsh ایش, er ایر, iki ایکی.
By contrast, Modern Standard Uyghur (MSU) has one glyph representation for each vowel except the ï [ɨ]:
Vowel Length
Length is at least represented for a/aː/e. Specifically (at least of /a/), aː is represented with an alif ا ; that short a is not represented. Further, the backness of this unrepresented vowel is inferred from the surrounding consonants:
The representation of consonants
Chaghatay consonants are mostly used identically to Persian, and are used similarly to the modern Uyghur orthography. A few places to pay attention to are:
Now we're in a position to revisit the quotation that began this Handbook. We can see that letter forms and script styles have social connotations:
This orthographic commentary is found on f003 of Prov. 158. ا alef is written as a straight vertical line, and ﻻ lam-alef are written as the beginning of a line is curved and crossed the end. So uprightness and righteousness are likened to the straight shape of alef while unfairness and dishonesty are likened to the twisted shape of lam-alef.)
Paleography: Further reading:
[Deroches], Deroche 2006, Gacek 2009, Hanaway and Spooner 2007.
Online sources:
(Maktub is a letter from one person to another. There are few personal letters of this sort in the Jarring Collection. Most are to or from Swedish missionaries, e.g. many in [Prov. 524] and [Prov. 476], one in [Prov. 542], a partial draft of one in [Prov. 312],
Modern Uyghur salam xet are similar.
(content )
There are yarligh 'decree' documents in the Jarring Collection, seven of which have been translated and studied by Hodong Kim who found that they were similar in structure to the ones issued during the Mongol empire (Kim 2010:61). There is a superscription basim subhane on top of the decrees. After that in sequence are: the name of the ruler who issued the decree, the name of the person(s) to whom the decree was addressed and the benefits and the privileges accorded to the addressee, warnings about consequences of not abiding by the decree, date and place of the decree, and finally the certification. The most conspicuous feature of the decree documents is the royal seals affixed to them.
An example from the Jarring Collection is [Prov. 458] an order to capture fleeing soldiers. (See Kim 2010 for other earlier 17th c. examples from the Jarring Collection.)
waqf < A. وقف dedication of property to charitable or religious uses; cf. Raquette, Eine kashgarische Wakf-Urkunde aus der Khodscha-Zeit Ost-Turkestans (1930). waqf documents usually feature the title وقف in large script centered at the top of the document, which is the distinguishing feature of such documents (Gacek 2009: 17). In the two page "waqf" published by Raquette, the name of the person who made the decree was written on the first page. the names of officials were listed in the beginning of the document. In the beginning of the second page, the names of officials were listed and the location was mentioned. The inviolability of the decree was emphasized for the land bequested.
How does one identify a manuscript? Looking at its physical characteristics and then starting to read through in order to discover the genre, language used, and topics discussed.
The first step is to examine its physical characteristics: how is bound - book format, scroll, or a single leaf? What kind of paper is used? What color and kind of inks? What style of handwriting? Are there different sections? Are there page numbers in Arabic script? (The latter could indicate that the manuscipts. copy dates from early 20th c.)
Titles are usually not given by the author or copyist and do not appear. If they appear, then some of them are rubricated (e.g. Prov. 24).
Looking at the front and back of the ms. can often reveal clues about its origins. Even if manuscripts often do not have titles, they often begin with a prayer. The back matter, in particular, is noteworthy. Any juridical document ends with a date (at least a year). And many documents have a colophon.
Most manuscripts are multilingual, with different passages or sections in different languages (see section on multilingualism). The most common text languages are Chaghatay and Persian; for prayers, Arabic is used. In Chaghatay texts, verse is often in Persian.
Sudden shifts of scripts is a sign that there's been a switch in language or text genre.
In the colophon, even if the manuscript is completely in Chaghatay, the colophon is likely to be in Persian. Possibly, authors and scribes learned to make a colophon in Persian in school, and it has become a habitual behavior.
Often, Chaghatay and Persian synonyms are used in texts, e.g. Prov. 351 and Prov. 9. Also, more rarely we find adjacent synonyms, sometimes even helpfully set off in parentheses: in Prov. 24, 3v., we can find birinj (gurunj) 'rice' as an example of adjacent synonyms. Also in Prov. 24 8r., there is an example of Arabic glossed in Persian, as hezer (ihtiyat) qil-.
If the catchwords do not match the first word on the subsequent page, then it is likely that leaves have become disordered. Leaves that are out of order may be more common in unbound manuscripts, but may even be so in bound manuscripts.
Often given at the end of the manuscript in the colophon.
The colophon may provide a date. If not, the date of writing can be deducted from the explicit data (see 2.2) while the date of copy can be roughly estimated from the writing style (see 1.1.6), and the kind of paper in some cases.
Further reading: Aigle 1995; Aigle 2000, Chih and Gril 2000, Mayeur-Jaouen and Papas 2014, Papas 2005, Thum 2014
Almost all manuscripts contain some Quranic quotations in Arabic; many also cite verses from famous poets or the names of famous people or toponyms.
The quotation in Arabic of religious sources is a feature of most of these manuscripts; such quotations are particularly common at the beginning of manuscripts. All risale begin with superscript bismillah and contain Quranic quotations (quotations from the Qur'ān and Hadith can be tracked through the Qur'ān and Hadith indexes ([Wensinck 1936]).
An Quranic citation example can be seen in Prov. 365 While the citation in this example accurately reproduces the original Quranic source, the diacritics in a number of manuscripts are not accurate when compared to the modern canonical text of the Qur'ān. The scribe's level of education, their memory, and perhaps the availability of the Qur'ān for reference at the time of writing determined the accuracy of citations.
Such citations are usually written formally with full diacritics. The Quranic citations were usually written in a calligraphic (script) style different from the one used for the Turkic text. In this example the calligraphic script of the Quranic text is Naskh.
Hadith (commentary) citations are usually not written in a different calligraphic style from the main text body. Hadiths, also always in Arabic, can be exemplified by Prov. 9. Also in the hadith in Prov. 500, its calligraphic styles is not different from that of the Turkic text. In both of the above manuscripts, the Arabic text has full diacritics.
Citing verse from known poets (especially Persian ones) is also common, as in this excerpt (whose author we haven't yet identified) in Prov. 351. There, the poems were copied in two columns in this example. The language is Persian and about medicine, probably quoted from a Persian medical handbook. Unlike Quranic quotations, poems were copied in the same popular Nastaʿlīq or Shikasta styles that are also used for the Turkic text.
Referring to book sources is also fairly common. Book references can point to the received authority on a subject,for example, at the end of Prov. 29, the author states that a book called Seyr Sharif has for more information on the topic.
Mentioning a personage (a religious, historical, or quasi-historical figure) also occurs. In Prov. 351, for example, has a citation, "Luqmān Hakim said that..." (written in the same calligraphic style as the main text). Unlike religious or verse citations, which are set off with indentations and vertically aligned verse lines, the form of book and personage references are not usually visually differentiated from the main text in any way.
Further readings: Brill Encyclopedia of Islam, Encyclopedia Iranica, Islam Arastirmalari Merkezi, Iranian Encyclopedia of Islam, Historical Atlas of Islam, Bregel 2003, Jarring 1997, Shinjang Uyghur Aptonom Rayoni kheritiler toplimi 2005, Encyclopedia of the Quran, Quran concordance, Calendar converter
Once you are able to identify the manuscript, it is necessary to situate the work within the literary production of Central Asia. Aside from the Jarring Collection, the most important collections of manuscipts from Eastern and Western Turkestans are preserved in the Institute of Oriental Studies in Tashkent, the Institute of Oriental Manuscripts in St-Petersburg, and the currently inaccessible Office of the Xinjiang Uyghur Autonomous Region Minority Nationalities Leading Unit for the Collection, Organization, and Planning for Publication of Ancient Texts. Some manuscripts and documents are to be found in Staatsbibliothek in Berlin, Bodledian Library at Oxford, Bibliothèque de l'Institut de France in Paris, the Staatsbibliothek in Berlin, and Houghton Library at Harvard University.
Further readings: Semenov et al. 1952-1987, Abdurrahman and Jin 2000, Central Asian manuscipts. 1952-1987, Hofman 1969, Hartmann 1904, Muginov 1962, Raquette 1940, Veli 1988, Persian manuscipts. 1972, Iranian sources database/encyclopedia: http://dl.nlai.ir
The previous sections focused on the evaluation of the characteristics of the manuscript; in the current section, we shift our focus to scholarly editing proper. The creation of a digital or print scholarly edition entails decisions on the types of analytic transcription (how the manuscript contents are represented); how writing systems might be transliterated; how the text is translated; and how various characteristics of the text are best represented.
Creating a literal ("diplomatic", that is, character-for-character) transcription of the manuscript page is fundamental to preserving information and transforming it into a digital, machine-readable form. Whatever is legible on the page should be typed as written, even if considered variant or erroneous. (Scribal variations and errors can be noted elsewhere, for example in footnotes or comments, and a separate regularized transcription can be made.)
Structure vs. content: Depending on the manuscript and the aims of the project, the transcriptionist will decide how to represent the structure of the page. If the page has only prose or verse, the decision is simply if or how to represent sentence, line, page and/or paragraph breaks, as well as catch words. Other structures on the page, such as tables, charts, and marginalia, may be more challenging to represent. This section focuses instead on the written content.
Transcribing the content of a Turki manuscript simply entails typing Perso-Arabic characters using a Persian keyboard setting. The software (e.g. text/XML editor, word processor) should render the characters into their proper forms for the context (initial, medial, final).
Sometimes, software will join Perso-Arabic characters in the middle of a word, even when they are not represented as joined in the manuscript. For example the word turnjabin ‘manna’ is written as ترنجهبن, with a final-form ه and an initial form ب (since in Persian this was originally two words). Typing it as one word, without whitespace in the middle, results in erroneous medial (joining) forms of ه and ب , like this: ترنجهبن .
To avoid this error, we recommend adding a non-joining character from Unicode. Although not visible, there is a "Zero-width space", U+200B in the transcription ترنجهبن. (It is also possible to simply add whitespace within the word, like this ترنجه بن, but this method may not accurately represent what the scribe wrote, and also fails to distinguish whitespace within a single word from whitespace between two words.)
Note: There are a few zero-width characters in Unicode; here are how the others look:
Transliteration entails rendering one orthographic version of the text into another writing system. Often, non-Latin writing systems are transliterated into Latin systems; in the ATMO project, we transliterate Perso-Arabic to a Latin-based script, whether the original manuscript was printed or handwritten.
But deciding what kind of transliteration system one needs depends on the the language(s) represented in the text. “Transliterating Chaghatay is somewhat complicated because one is often transliterating Turkic, Persian, Arabic, and Mongol words all together in the same sentence. The standard is to transliterate Arabic as Arabic, Persian as Persian, and Turkic as Turkic (as best can be approximated). This means: Arabic and Persian words will indicate vowel length with macrons over long vowels (ā ī ū), while Turkic words will not. قیلماق, for example, is written qilmaq, not *qīlmāq. It's also good to indicate vowel quality when you can, inserting ö, ü, ï, ä, etc., where it is correct to do so. كورماك, for example, is probably körmäk.”(Schuessel 2011).
Also, transliteration systems best follow international scholarly conventions, if they exist. For Chaghatay manuscripts, this principle means incorporating conventions for transcribing Arabic, Persian, and Ottoman Turkic, but also adapting them (to reflect how scribes adapted scripts to represent sounds in their language). One such reference for Arabic, Persian, and Ottoman are the American Library Association and U.S. Library of Congress' tables (ALA-LC 1997-2018).
Whichever the transliteration system, it must represent all the orthographic distinctions of the original writing system. For those of us who may be more familiar with Uyghur or Uzbek, it's particularly important to not ignore long vowels (surāh, not surah), and to include for example consonant distinctions found in Arabic or Persian but not modern Turkic (mithqāl).
A separate issue is that typing transliteration can be cumbersome: characters with diacritics are not found on a normal keyboard. For this problem, either keyboard shortcuts can be made, or characters can be cut and pasted, or a place-holding character can be substituted and then replaced at regular intervals.
The transliteration system of the ATMO project has been published for public comment: ATMO 2017, and will be discussed in the next sections.
The transliteration system of the ATMO project is available here: ATMO 2017b.
Since most manuscripts lack punctuation, one step before linguistic annotation is to mark sentence boundaries. Further, other text boundaries such as paragraph and section boundaries are of interest. We have marked boundaries using the following conventions in our ATMO:Lat (Latin script transliteration):
/ marks sentence boundary. // marks paragraph boundary. /// marks major text boundary (e.g. section boundary) //// marks boundary between texts, in collective manuscripts.
We find it most efficient to add these at the stage of tandem proofreading.
Regarding the marking of word, affix, and morpheme boundaries, please see the Morphological Annotation Handbook of this project.
People are referred to by title, epithet or personal name.
bu faqir, bu kamina (this poor man, this humble man) almost systematically refers to the author; the plural suffix -lar is a honorific mark at the end of a proper name (ex: hazrat-larim = my/the saint)
Author or scribe's name and title usually appear in manuscript colophons in Jarring Collection (see also Gacek 2009: 66). For example, in Prov. 369, the scribe, a certain Hekimjan, called himself as faqir bichare فقر بی چاره 'hopeless poor man'. However, it was not always the case, since not all the manuscripts in the Jarring Collection have colophons. Even a well executed manuscript like Prov. 365 doesn't have a colophon at all. And aside from that, not all the colophons contain author's or scribes name and sometimes if there is any, it was written without title. For example, in the first manuscript in Prov. 327, the scribe simply called himself without any titles as Molla Abdulqadir Yarkendi. While the second manuscript does have a colophon, it doesn't contain any information about the author or scribe, neither his name nor his title.
Honorific titles like hazrat (majesty, excellency) is more common in biographies of saints and other important religious figures. Prophets were also called with this title sometimes, however prophets' titles were usually followed by the phrace علیه السلام "peace be upon him". And prophet Muhemmed's name is generally followed by the phrase صلى الله عليه وسلم 'God bless him and grant him salvation'. Khoja title was exclusive to the people who claimed themselves to have been from the prophet Muhammed and Satuq Bughrakhan lineage. For example, the Prov. 369 contains the family tree/pedigree of the Khojas along with the biography of the most famous among them, the Afaq.
Book owners sometimes called themselves with the title Akhon (mullah). In Prov. 41, the owner name was written as Damolla Abubakri Akhon. In the second manuscript in Prov. 327 the owner called himself as محمت اخوند.
During the Chaghatay period (ca. 14th to early 20th c.), Chaghatay (a Middle Turkic language) and Persian were the two dominant languages in use. Chaghatay was spoken east and west of the Pamir mountains between the 14th and 20th centuries, especially in cultural centers such as Samarkand, Bukhara, Herat, Khiva, Kokand and Kashgar (Eckmann 1996: 2), and also including a broad swath of Central Asia from Transoxiana (between the Syr Darya and Amu Darya), Khorasan, Ferghana to the Tarim basin and Ghulja valley. Chaghatay speakers were mostly sedentary farmers and traders; transhumant and sendentary Turkic speakers have lived east and west of the Pamir since at least the 9th century CE, when the Old Turkic Khanate dispersed southwards and westwards. The Chaghatay language was widely used in private and public domains, especially in the home and in the marketplace.
Before and during the Chaghatay period, Persian cultural and linguistic influence was very strong in these regions. The 11-13th century Khwarezm empire included Transoxiana and was founded by Turks, who nonetheless promoted Persian; for example, Firdausi finished the Shahname under Khwarezm sponsorship. During the Chaghatay period, Persian was the prestige language of intergovernmental communication , literature, and education. Leaders of Transoxiana used Persian in correspondence with the Manchu Qing empire. Persian speakers were élites: madrassa teachers, government officials, literati and high-value commodity traders. Chaghatay Turkic text authors and copyists sometimes code-switch into Persian in order to cite poetry and/or the sayings of notable persons. Persian quotations in the colophon are also fairly common.
The early 20th c. Jarring manuscripts include many quotations of Persian classical poetry. Some of these can be identified with Iranian editions of classics, but others may well be from Persian medical texts. These mostly occur in the middle of text, sometime in the colophon. These poetry quotations are often marked off either by a glyph (four dots, flower, etc.) or a rubricated title (nazm 'verse'). For example, see the word nazm and four-dot marker with red ink in the second and third line from button to top on the right side of this folio [Jarring_Prov_351_040], [Jarring_Prov_351_032] and see just nazm with read ink [Prov_351_031].
Arabic by contrast, was present in the area exclusively in written form. Since the Arabs withdrew from the Central Asian region after their defeat during the 7th c. CE, the influence of Arabic is limited to religious texts, and otherwise parts of formal written language. In non-religious texts, Arabic is primarily found as an embellishment. There was a convention of writing Arabic prayers (ayat) at the beginning of texts, at least from the time of the early Middle Turkic masterworks Qutadgu Bilig [the Wisdom of Royal Glory] and the Turki Tillar diwani [Dictionary of Turkic Languages] (11th c.). These initial prayers ('superscriptions') are often from surāh 1 and yet 1. Additionally, ayet prayers in middle of texts are very common , as well as prayers at the end as 'subscriptions'.
On the use of Arabic: Most the manuscripts contain Quranic quotes. The Qur'ān is divided into chapters (surāh), and these are divided into verses (ayat). The Quranic quotes occur at multiple points in the manuscripts. Almost every single text begins with an Arabic prayer, specifically the first verses of the Surāh Al-Fatiha: Bismillāh ir-rahman ir-rahim 'In the name of Allāh, the most Gracious, the most Merciful.' For a text example, see [Jarring_Prov_2]. Prayers sometimes also occur at the end of texts, see example [Prov_2_045_r]. Turki manuscripts feature many laudatory formula and canonic citations in Arabic (see 2.2 and dictionaries below).
The Turkic Chaghatay language is named after the eponymous Mongol ruler, Chaghatay (1183-1242; reigned 1226-1242). The Mongolian conquerors were all schooled in Persian (source needed) and interacted with local Turkic and Persian speakers in (Turkic? Persian? Both?). There was no need and no point of using Mongolian, so there is little trace of that language in the region. The Chaghatay language developed as a regional koiné. It represented the eastern variety of Middle Turkic and integrated significant amounts of Persian. Besides lexical copies from Persian, the use of ki and kim as relativizers spread under Persian influence, and vowel harmony was limited (to backness harmony). Arabic influence was limited to the lexicon. And of course, Chaghatay and its descendants Uzbek and Uyghur were written with Perso-Arabic scripts.
Further, some proportion of the population attended local religious schools. Arabic were used in these madrassas for religious educational (Teklimakani 2004: 8). In addition to the Qur'an (in Arabic), pupils were also taught a great deal of Persian literature and poetry. Thus, elites were fully bilingual in Persian and Turkic Chaghatay, and also had reading knowledge of Arabic. Therefore more and more bilingual and multilingual highly educated people came out. These pupils later produced literature and scholarly work in Arabic and Persian (especially). Women and uneducated people presumably only spoke Chaghatay.
Some highly prestigious poets like Lutfi, Sekkaki and Nava'i actively opposed use of Persian and Arabic in their writing, but even their own peoples didn't entirely avoid using Perso-Arabic vocabulary.
The Chaghatay period is therefore characterized by diglossia, where Persian was written and to some extent spoken language of the elites, and Turkic was the spoken language of ordinary people. Written documents appear to have been increasingly produced in Turkic, but Persian was still the prestige language. Literacy events such as oral and written storytelling and poetry, contracts, and the circulation of religious texts must have been common, because of the abundance of primarily Turkic-language manuscripts of diverse genres from at least the 16th century.
Further, many Perso-Arabic words and some morphemes have been nativized and incorporated into the Chaghatay language itself, such as Arabic dunyā 'world' and Persian ki, often 'that'. Thus, the Chaghatay language itself incorporates Perso-Arabic strata onto a Middle Turkic matrix language, as well as some Mongolian. But in addition, the documents of the Jarring Collection also are embellished with quotations and verses in Arabic and Persian, often of a Quranic and/or religious nature. Thus, the manuscripts reflect interlingual code-switching (Chaghatay to Arabic or Persian and back) as well as the multilingual basis of Chaghatay itself.
East of the Pamirs, whence most of the Jarring manuscripts stem, the Turkic Chaghatay language interacted with the Turkic spoken by the Karakhanids and their subjects, 10-13th c. (They were arch-enemies of Persians, but the language was Persian-influenced nonetheless.) There, in the Tarim Basin, we also see a similar diglossia, with elites being educated in Persephone and Arabophone madrassas, and later writing primarily in Persian and Chaghatay, while speaking Chaghatay Turkic at home.
Some texts in the Jarring Collection are atypically in colloquial Turkic, and not literary or religious in nature. These include Prov. 207 and Prov. 208; several of the manuscripts of this type were commissioned by Ambassador Jarring or his predecessor Gustav Raquette.
What does this tells us about what the scribes knew? Some scribes handwriting was very beautiful and also applied the Perso-Arabic diacritics correctly. However, some scribes overuse the diacritics. The scribe might not know the Arabic or Persian well. Frequent code switching between Turkic and Persian (and Arabic) sometimes result in a concept or term expressed as forms from two languages, e.g. Prov. 9 alternates between Per. doz and Turki ogri 'thief'.
Social aspects of multilingualism:
The Arabic prayers are instrumentalized to ensure a positive outcome in a range of text genres. Risale manuals include Arabic language prayers at every step in the process is accompanied by prayer so that the artisan will go to heaven. If he does not read the manual, it will be bad for his business and he may well go to hell. Medical texts include Arabic prayer in order to ensure that the treatment is proper. Tezgire include Arabic prayers if an Imam dies (and ... in what other circumstances.)
In contrast, passages in Persian are typically verse. Use of Persian verse is associated with erudition and education, since it requires familiarit with classical poetry and Persian heritage. Persian verse may be used in a text for metaphorical, poetic purposes, or to give scholarly or historical weight to an argument.
Certain genres have a particular abundance of Persian terminology. Medical manuscripts (the focus of this project) are one such example. Healers habitually use Persian terms for herbs (esp. in Prov. 351, in contrast to Prov. 28); perhaps the healers use the more 'foreign' term to show their superior knowledge and intentionally obfuscate. In any case, healers have a stock vocabulary that is different from what other people use.
Multilingual manuscripts offer the opportunity to study semantic change. For example, one of the Persian terms for medicine majduq now in MSU has semantically narrowed to refer specifically to a relaxant drug.
Further readings: Steingass Persian, Dekhoda, Lane 1863, Wehr 1976
Further reading: Ehet 2011, Teklimakani 2004, Bodrogligeti, 2001, Clauson 1972, Eckmann 1966, Jarring 1964, Kunos 1902, Bahawudunet al. 2002, Zener 1866.
Genre is a general term for text (or discourse) type. A text genre will include the text's form, conventions, formality, stylistic features and content type. Manuscripts may be prototypically one genre, or they may be primarily of one genre, while including other genres (for example, a risale professional manual in prose, with interspersed Quranic verse). Sometimes a text will incorporate materials from multiple sources, for example, a hagiography may be copied from a number of other source hagiographies. Identifying sources and text genres aid both historical and philological analyses.
Now we'll talk about several of the most important genres in this collection. Safarnāma is a particular case in the Jarring Collection. Note that the risale (treatise) is a large category: Prov. 351 is a treatise on medicine, not a professional manual.
Recent research done on Islamic hagiographies revealed the historical value of these popular genre in Central Asia and elsewhere in the Muslim world. Again, an accurate review of the manuscripts (especially in composite volumes including several hagiographies or several types of texts), a close attention to what the author says or alludes to (particularly in doctrinal aspects), and comparison with hagiographies produced in other Muslim contexts may open new perspectives.
Further reading: Aigle 1995; Aigle 2000, Chih and Gril 2000, Mayeur-Jaouen and Papas 2014, Papas 2005, Thum 2014
The genealogical documents known as shajara are important in Central Asia and deserve more attention than they are typically accorded. Once discussed in terms of political history and religious anthropology, the scrolls are not simple diagrams of descendants but artistic piece which projects the power of the dynasty and are linked to religious rituals within Sufi milieux.
See Prov. 219 and Prov. 561
Further reading: Shinmen and Sugawara 2002, Papas 2012
Professional manuals are well represented in Turki literary production but remains understudied, despite recent high-quality publications. Here again, we recommend to study these manuscripts from a multidisciplinary perspective: precise reading of contents, reference to Islamic doctrine, and recourse to ethnographic material.
see Prov. 2, Prov. 41, Prov. 68.
Further readings: Centlivres-Demont 1997, Dağyeli 2011a, Dağyeli 2011b, Gavrilov 1928, Olesen 1994, Gavrilov 1928.
Written in mathnawi style (rhyme aa bb cc), very common in Central Asia, the safarnāma of the Sufi poet Muhammad Siddiq Zalili (17-18th c.) is the typical example of a neglected source. Between the poetical figures and the relatively difficult rhetorics, historians and anthropologists find numerous data on toponyms, land descriptions, and religious practices. Ultimately, the manuscript appears as both a real and spiritual journey.
This second mathnawi is the only known work authored by Muhammad Kharabati (17-18th c.). Often described as a Muslim englightener, if not an early Reformist (jadid), a complete reading of the manuscript shows almost the reverse, that is to say, a desperate mystic haunted by death and absurdity. Moreover, the text looks very close to oral language of the time, suggesting that here we are dealing with popular preaching rather than poetry.See Prov. 90, Prov. 365.
This genre has not (as yet) been encountered in the Jarring Collection, but may be relevant for other collections.
These are biographical dictionaries (of jurists, scholars, Sufi shaykhs, saints, etc.) should be approached not only as prosopographic sources but also as conceptualized works and practical objects. Read carefully (especially the introductions), studied step by step according to our methology, tabaqat books (such as the Turki translations of Jami's Nafahat al-uns) appear as complex descriptions of individuals, rather than groups, which are the subject of readings within educational circles.
Further readings: Papas 2010, Papas 2015
While most of us use manuscripts for one or more academic disciplines, incorporating other scholarly approaches can offer corroborating evidence for particular arguments. The following exemplify a few of the many possible scholarly approaches to these manuscripts, using Prov. 28 as an example text.
Prov. 28 is entitled "Prescriptions for curing of different diseases, Partly with traces of ancient shamanistic rites." Most sentences include a diagnosis and a remedy; the manuscript is a list of these diagnoses and remedies. Diagnoses such as being afflicted by uchqun 'sparks' and pir spirits suggest that the healers are mediators between the spirit world and the human world and are thus spirit mediums (what some would loosely term "shamans"). The text describes the implements used by the healer, auspicious dates and substances for healing, and measures should the patient not respond to treatment. Textual actors include several kinds of healers, community and family members, and spirits, as well as the patient.
The manuscript itself dates from around 1930 and Jarring's metadata notes that it was "probably copied from some indigenous medicine book." As is common with these manuscripts, neither authorship, nor a precise place nor a date can be associated with this text. Still, as this copy was in circulation in the 1930s, it must reflect one stream of healing practice common in the southern Tarim Basin during the late 19th and early 20th century.
Let us now turn to several possible disciplinary takes on this material. A caveat: These are of course not the only possible analyses within these disciplines. Much of the following discussion on Prov. 28 is excerpted from a forthcoming article (Dwyer and Amat in preparation 2018.)
The development of manuscript production had to do with the rise in the literacy among the population of Tarim Basin. Between seventeenth and twentieth centuries there was a rise in vernacular and manuscript production (Thum 2014: 59). According to Prov. 207 Scribe’s Necessities a scribe acquired the writing as an organic skill which included finding raw materials for a pen, ink, and paper, making not only pen and ink but also paper. The scribe would rule the papers, write on them and finally bind them in book format. The Jarring manuscript collection includes European and Russian origin papers which were used at the same time with locally produced papers. The availability of industrially produced paper must have gradually led to the narrowing down of skills necessary for becoming a scribe. The separation of writing from papermaking must have made writing a social act increasingly available to a large mass. As in the case of Prov. 28, the availability of industrially produced paper enabled ordinary people easier access to the medium of writing. And with Jarring manuscripts, it became possible for the first time to see writings left by ordinary people rather than professional scribes, poets, and belles-lettres.
For historians, Prov. 28 offers interesting perspectives at least in two domains. The writing production of Central Asia in the early modern and modern periods included a medical component, which consisted in talismans, recipes, healing vademecums, magical and medical treatises. These documents mixed various techniques and disciplines (Greco-Islamic theories, divination, traditional knowledge of nature, etc.), which evolved over time. Compared with other sources, either contemporary or earlier ones, such as the Tabiblik kitabi by Subhan Quli written in the seventeenth century (Karoly 2015), Prov. 28 can be analyzed as a piece to reconstitute the long history of medical writing in Central Asia. A second domain of interest regards the history of health and health-care. Beyond the somewhat repetitve and undecidable debate on Shamanism in Central Asian Islam, where healing is summoned as a witness of the survivance of Shamanic traces or, on the contrary, of evidences of strong Islamic traditions, the techniques described in Prov. 28 can be discussed from the perspective of patients and illness. Associated with alternative sources, such as chronicles and biographies which mention causes of death, current diseases, epidemics and so on, Prov. 28 helps to better understand the health and even social problems East Turkestan people faced at the time of the manuscript production.
More concretely, Prov. 28 can be read along with what can be called "performative manuscripts", like the talismans (tumar) preserved in the Gunnar Jarring Collection. Talismanic scrolls such as Prov. 14 and 452, for example, are historical documents, which illustrate the culture of magical healing in late nineteenth- and early twentieth-century Tarim Basin. Prov. 14 measures 110 x 12 cm. The text is in Arabic and late Chaghatay/early Uyghur mixed with some Persian. The scroll is made for both men and women and should be kept with oneself to be efficient. The aim is to awaken love and to unite couples, either by gathering lovers with beloved or by maintaining long-term relationships (Papas 2014). As for Prov. 452, the manuscript is a very long scroll of approximately 5m length. This is more precisely a counter-charm (radd nāma) against all kinds of magic and harmful events, written in basic Persian and ended by a series of invocations in Arabic. These scrolls were prepared by low-class mullahs who belonged to the same social milieu as healers.
Prov. 28 can be used to analyze socio-cultural phenomena historically. The healing method known as pir oynash was practiced until very recently in the region. The practices seamlessly incorporated spirit mediumship and Islamic components, as well as material and spiritual dimensions. These dimensions were not viewed oppositionally, as they might be today.
The healing practice presupposed a communal social organization (Beller-Hann 2008). [As the manuscript indicates, the healing practice required the participation of the community members, hence it had strong social character.] Until very recently, the family was an independent economic unit and each family was connected to others through reciprocal gift-giving relations, forming mehelle 'neighborhood.' Families could satisfy their demands outside the market through either direct barter or gift-exchange. Such was the case until very recently in places like the city Atush. Surplus products entered the market, hence the market was an accessory to the society. Both the urban and rural people carried out agricultural production (Beller-Hann 2008: 103-104).
The healing practice was a part of the general gift-giving reciprocity, which also included various ceremonies such as weddings and funerals. The social character of performance arts, the dance, and music performance as described in the manuscript, shows that entertainment was a part of communal life, in the service of communal life before its separation from the organic totality of life, its compartmentalization, and professionalization in modern times.
The use of the dutar lute and the dap drum as instruments of healing also demonstrates the fluidity between communal rituals. The healing practice consisted of the use of material and spiritual substances and their apparent dualism was overcome in socially organized healing. The absence of mind-body, matter and form dualisms points to the organic totality of life which could integrate disparate systems of medicines and practices for the service of the community.
The type of paper and the style of writing (more on that in History of book production) indicates that the manuscript was written by a person who was not a professional scribe. The carelessness of the writing indicates that there were increasing literacy among the ordinary people and that the medium of writing became increasingly available for them.
Historical anthropological analysis thus can provide ethnographic details on the relationship between social organization, life-cycle rituals, economy, religion, and in the case of this corpus, also healing.
With the exception of discourse analysts, linguists tend to focus at the level of sentence or below, rather than on the page or on the entire manuscript. Further, whatever the aims of the individual researcher, a defacto standard that allows for many kinds of linguistic research is so-called Interlinear Glossing (ILG). ILG is in at least a three-tier format: (1) a version of the text, in the original or a transliterated script, segmented into morphemes and words; (2) an ILG, with literal translations of substantives like nouns and verbs, and all-caps abbreviations of grammatical morphemes; and (3) a free translation.
Again using Prov. 28 as an example. Each sentence is segmented and the parts of speech for each sentence is analyzed as the following (lines 1-3):
This approach allows a linguist (even one unfamiliar with the relevant languages) to analyze areas such as morphology, syntax, and the lexicon (frequencies, collocations, etymologies, dependencies, and so on). To improve the text's accessibility to the non-specialist, a regularization of the first text line (the segmented transliteration) would help:
The word-for-word transliteration is:
Regularized (i.e. how people probably pronounced it):
In any case, the scheme for Interlinear Glossing should be documented: how morphemes are segmented and which glosses correspond to which linguistic forms. (The ATMO project's ILG documentation is available as a Technical Report (ATMO 2017a).
The above four analytic examples are by no means exhaustive, but they provide a basis for scholarly exploration.
Abdurrahman, Amina and Jin Yu-Ping. 2000. Une vue d'ensemble des manuscrits tchagatay du Xinjiang. Cahiers d'Asie centrale, 8, pp. 35-62. https://asiecentrale.revues.org/589
Aigle, Denise (ed.) 1995. Saints orientaux. Hagiographies médiévales comparées. Paris: De Boccard.
Aigle, Denise (ed.) 2000. Miracles et karama. Hagiographies médiévales comparées. Turnhout: Brepols.
ALA-LC 1997-2018. (American Library Association and U.S. Library of Congress) ALA-LC Romanization Tables: Transliteration Schemes for Non-Roman Scripts.
ATMO 2017a. ATMO Technical Report 2017-01: Part-of-speech tag set. (Annotating Turki Manuscripts from the Jarring Collection Online, Arienne M. Dwyer and C.M. Sperberg-McQueen, PIs).
ATMO 2017b. ATMO Technical Report 2017-02: ATMO transliteration table (Annotating Turki Manuscripts from the Jarring Collection Online, Arienne M. Dwyer and C.M. Sperberg-McQueen, PIs).
Bahawudun, Muhemmettursun, Ghenizat Gheyurani, Ismail Qadir, and Ablimit Ehet. 2002. Chaghatay Tilining Izahliq Lughiti [Dictionary of Chaghatay Language]. Ürümchi: Xinjiang Xelq Neshriyati [Xinjiang People's Publishing House].
Bodrogligeti, András J.E. 2001. A grammar of Chagatay. (Languages of the World: Materials ; 155). München: LINCOM Europa, 2001. (Repr. 2007).
Boeschoten, Hendrik, and Marc Vandamme. 1998. Chagatay. In Johanson, Lars, and Éva Á Csató eds., The Turkic Languages, pp. 166-178. London and New York: Taylor and Francis.
Bögü, Ablimit Ehet. 2011. Chaghatay Uyghur Tili [The Chaghatay Uyghur language]. Ürümchi: Xinjiang University Press.
Bregel, Yuri. 2003. An Historical Atlas of Central Asia. Leiden: Brill.
Brill Encyclopedia of Islam. http://referenceworks.brillonline.com/
Calendar converter. www.aly-abbara.com/utilitaires/calendrier/calendrier_hijir.html
Centlivres-Demont, Micheline. 1997. Un corpus de risâla du Turkestan afghan. in Grandin, Nicole, and Marc Gaborieau (eds.), Madrasa: la transmission du savoir dans le monde musulman. Arguments.
Central Asian documents: manuscript, 1558-1804. Networked Resource Houghton Library, Harvard University. http://ocp.hul.harvard.edu/dl/ihp/011812637
Cerquiglini, Bernard. 1989. Éloge de la variante: Histoire critique de la philologie. Éditions du Seuil.
Chih, Rachida & Denis Gril (eds.) 2000. Le saint et son milieu, ou comment lire les sources hagiographiques. Cairo: IFAO.
Clauson, Gerard. 1972. An etymological dictionary of pre-thirteenth-century Turkish. Oxford University Press.
Dağyeli, Jeanine Elif. 2011a. La construction des identités collectives d’après les chartes des corps de métier (risāla) en Asie centrale. Cahiers d’Asie centrale, 19-20, pp. 73-94. http://asiecentrale.revues.org/1381
Dağyeli, Jeanine Elif. 2011b. Gott Liebt das Handwerk: Moral, Identitat Und Religiose Legitimierung In der Mittelasiatischen Handwerks-Risala (Iran-Turan). Munich: Reichert.
Déroche, François (ed). Le manuscrit médieval en écriture arabe. Online: http://www.irht.cnrs.fr/sites/default/files/images/images_contenu/images_contenu_site/pieces_jointes/livret_arabe.pdf
Déroche, François. 2006. Islamic codicology: an introduction to the study of manuscripts in Arabic script. London: Al-Furqan.
Dehkhoda, Ali Akbar. 1998. Dehkhoda dictionary. Sixteen volumes. second edition. Tehran: Tehran University 19. Dehkhoda Lexicography (Dekhoda, Lughatname) (Iranian lexicographer). http://parsi.wiki (includes definitions, citations from classical literature. Extremely useful for phrases)
Dwyer, Arienne M. 2017. Manuscript Technologies, Writing and Reading in Early 20th Century Kashgar. In Ildikó Bellér-Hann, Birgit N. Schlyter & Jun Sugawara (eds.), Kashgar Revisited: Uyghur Studies in Memory of Ambassador Gunnar Jarring, 34–57. Benjamins. proof (pdf)
Dwyer, Arienne M. and Akbar Amat. In preparation 2018. Trance-medium healing remedies in an Islamicized context. Ms.
Eckmann, János. 1966. Chagatay manual. Vol. 60. Bloomington: Indiana University.
Encyclopedia Iranica. http://iranicaonline.org
Encyclopedia of the Quran (Leiden: Brill, 6 vols.). http://referenceworks.brillonline.com/
Gacek, Adam. 2009. Arabic Manuscripts. A Vademecum for Readers. Brill.
Gavrilov, Michel, Joseph-Antoine Castagné, and Joseph-Antoine Castagné. 1928. Les corps de métiers en Asie centrale et leur statuts (Rissala). Revue des études islamiques No. 2, pp. 209-230.
Hadith concordance. N.D. sunnah.com (searchable in English, Arabic, Urdu, Indonesian)
Hanaway, William L., and Brian Spooner (eds.), 2007. Reading Nasta'liq: Persian and Urdu Hands from 1500 to the Present. Series: Bibliotheca Iranica: Literature Series 3. Mazda Pub.
Hartmann, Martin. 1904. Die osttürkischen Handschriften der Sammlung Hartmann. Mitteilungen des Seminars für Orientalische Sprachen zu Berlin, 7: pp.1-21.
Hofman, Henry Franciscus. 1969. Turkish literature: A bio-bibliographical survey. Section III. Muslim Central Asian Turkish literature being in the main a list of Chaghatayan authors and works in Chaghatay as registered in Professor MF Köprülü's article: Çagatay edebiyati İA. vol. III (270-)(with some additions Navāʾīāna, however excepted). Vol. 1. published by the University of Utrecht under the auspices of the Royal Asiatic Society of Great Britain and Ireland.
Iranian sources database/encyclopedia. http://dl.nlai.ir
Iranian Encyclopedia of Islam (in Persian). Onine: http://rch.ac.ir/article/paging
Islam Arastirmalari Merkezi (in Turkish). http://www.islamansiklopedisi.info
Islamic manuscript. http://www.islamicmanuscripts.info
Jarring, Gunnar. 1964. An Eastern Turki-English Dialect Dictionary. Vol. 56. No. 4. Lund: Gleerup.
Jarring, Gunnar. 1997. Central Asian Turkic place-names: Lop Nor and Tarim area: an attempt at classification and explanation based on Sven Hedin's diaries and published works. Sven Hedin Foundation [Sven Hedins stift.], National Museum of Ethnography [Folkens museum].
Karoly, László. 2015. A Turkic Medical Treatise from Islamic Central Asia: A Critical Edition of a Seventeenth-Century Chagatay Work by Subḥān Qulï Khan. Brill.
Kennedy, Hugh. 2002. An Historical Atlas of Islam. [Atlas historique de l’islam]. http://referenceworks.brillonline.com/browse/historical-atlas-of-islam
Kim Hodong. 2010. Eastern Turki Royal Decrees of the 17th Century in the Jarring Collection. In Millward, James, Yasushi Shinmen & Jun Sugawara (eds). Studies on Xinjiang Historical Sources in 17-20th Centuries, pp. 59-119. Tokyo: Toyo Bunko.
Kunos, Ignaz. 1902. Shejx Sulejman Efendi's Chagataj-Osmanisches Wörterbuch [Lughat-i Chaghatayi we Turki-i Osmani].
Lane, Edward William. 1863. Arabic-English lexicon. 8 parts. Williams & Norgate. http://www.tyndalearchive.com/tabs/lane
Mayeur-Jaouen, Catherine, and Alexandre Papas (eds.) 2014. Family Portraits with Saints. Hagiography, Sanctity, and Family in the Muslim World. Berlin: Klaus Schwarz.
METRO (Middle English Teaching Resources Online) resources. 2018. Textual Instability in a Manuscript Culture website
Metropolitan Museum. N.D. Arabic script and the art of calligraphy. Curriculum Resources. http://www.metmuseum.org/learn/for-educators/publications-for-educators/art-of-the-islamic-world/unit-two
Millett, Bella. 2014. What is Mouvance? website
Muginov, A.m. 1962. Opisanie uigurskikh rukopisei Instituta narodov Azii. Moscow: Vostochnoi literatury.
Olesen, Asta, and Ida Nicolaisen. 1994. Afghan Craftsmen: The cultures of three itinerant communities. Thames and Hudson. (Esp. the appendix pp. 291-309.)
Papas, Alexandre. 2005. Soufisme et politique entre Chine, Tibet et Turkestan. Paris: J. Maisonneuve.
Papas, Alexandre. 2010. Mystiques et vagabonds en islam. [Portraits de trois soufis qalandar]. Paris: Cerf.
Papas, Alexandre. 2012. Joining the Dots between the Khwājas of East Turkestan: A Shajara Scroll Preserved at the Louvre Museum. Der Islam, 88, pp. 352-365.
Papas, Alexandre. 2014. Un rouleau talismanique du Xinjiang musulman. Eurasian Studies 12, pp. 555-660, xxiv-xxvi.
Papas, Alexandre. 2015. Kharābātī (1638-1730), un poète populaire du Turkestan oriental. Cahiers d’Asie centrale, 24, pp. 127-144.
Quran concordance. http://referenceworks.brillonline.com/ or: http://corpus.quran.com or: al-Mu'jam al-mufahras li-alfaz al-*Qur'an. https://ia600500.us.archive.org/7/items/AlMujamAlMufahrasLiAlfazhAlQuran/mmaqk.pdf
Raquette, Gustaf. 1940. Collection of Manuscripts from Eastern Turkestan. An Account of the Contents. särtryck ur C. G. Mannerheim, Across Asia from West to East 1906-08, Helsingfors.
Schluessel, Eric. 2011. Transliteration. https://wiki.harvard.edu/confluence/display/turkic/Transliteration
Semenov, A. A. et al. 1952-1987. Sobranie vostochnykh rukopisei Akademii nauk Uzbekskoi SSR. 1-12 vols. Tashkent.
Shinmen Yasushi and Onuma Takahiro. 2012. First contact between Ya'qub Beg and the Qing: The Diplomatic Correspondance of 1871. Journal of Asian and African Studies, No.84, pp. 5-37. http://repository.tufs.ac.jp/bitstream/10108/70942/1/jaas084001_ful.pdf
Shinmen Yasushi and Jun Sugawara. 2002. Kashugaru Hojake Afakuto no katsudo no ittan: Yaringu korekushon Prov.219 ni tsuite [Evidence of the activities of Khwāja Āfāq in Kashgar: on manuscript Prov. 219 of the Jarring Collection]. Toyoshi Kenkyu [Journal of Oriental Researches], 61/3, pp. 33-63.
Shinjang Uyghur Aptonom Rayoni kheritiler toplimi. 2005. Beijing-Ürümchi: Junggo Kherite Neshriyati-Shinjang Pen-Tekhnika Neshriyati.
Steingass, Francis Joseph. 1892. A Comprehensive Persian-English dictionary: Including the Arabic Words and Phrases to be Met with in Persian Literature. London: Routledge & K. Paul. http://dsal.uchicago.edu/dictionaries/steingass
Storey, Charles Ambrose. 1927. Persian Literature: A Bio-bibliographical Study, Vol. 1-3. Luzac. Translated into Russian and revised, with additions and corrections, by Y. E. Bregel. Edited by Y. E. Borshchevsky. 3 vols.: pp. 693; [ii], 695–1314; 1315–1886. Moscow: Central Department of Oriental Literature. 1972. Rbls. 3, 3.11, 2.90.
Sugawara Jun and Yayoi Kawahara. 2006. Mazar Documents from Xinjiang and Ferghana (Facsimile),1–3. Tokyo: Tokyo University of Foreign Studies.
Teklimakani, Abdureop Polat. 2004. Chaghatai Uyghur Tili Heqqide Mupessel Bayan [Complete narrative on Chaghatay Uyghur Language]. Beijing: milletler Neshriyati [The Publishing House of Minority Nationalities].
Thum, Rian. 2014. The Sacred Routes of Uyghur History. Cambridge: Harvard University Press.
Veli, Qurban. 1988. Uyghur, Özbek, Tatar Qedimki Eserler Tizimliki. Kashgar: Qeshqer Uyghur Neshriyati.
Wehr, Hans, and J. Milton Cowan. 1976 [1960]. A Dictionary of Modern Written Arabic: Edited by J. Milton Cowan. Spoken Language Services. online: https://archive.org/details/Dict_Wehr.pdf. (Considered the most comprehensive Arabic-English dictionary.)
Wensinck, Arent Jan. 1936. Concordance et indices de la tradition musulmane. Vol. 1. Brill Archive. (cross-indexed all hadiths.) Arabic pdf edition available online: http://waqfeya.com/book.php?bid=1236
Zenker, Julius Theodor. 1866. Türkisch-Arabisch-Persisches Handwörterbuch. http://reader.digitale-sammlungen.de/ (a dictionary of Ottoman, which includes Chaghatay.)
Zumthor, Paul. 2000 [1972]. Essai de poétique médiévale. Éditions du Seuil.
Amat, Akbar, Arienne M. Dwyer, Gülnar Eziz, Alexandre Papas, and C.M. Sperberg-McQueen. 2018. A Handbook of Chaghatay manuscript transcription. Lawrence, Kansas: Annotating Turki Manuscripts from the Jarring Collection Online (ATMO). (Principal investigators: Arienne M. Dwyer and C. M. Sperberg-McQueen; sponsored by the Henry Luce Foundation). On the Web at XHTML. Last examined dd Mmmmmm YYYY.
Credit: Painting of Mahmud al-Kashgari © 1981 by Ghazi Emet.