THE TRANSLITERATION OF THE HARIVA.M/SA (HV) (Version of 30.6.2007; previous version 5.9.2005) ACKNOWLEDGEMENT The project of preparing an electronic transliteration of the HV was carried out with the explicit permisson of the Executive Board of the Bhandarkar Oriental Research Institute (BORI) (Resolution 5/19.05.2000). In the name of the international community of Sanskritists we wish to acknowledge the cooperation of the BORI. The copyright of the constituted text of the HV lies with the BORI, including commercial use of the transliteration. The contributors to this project have pledged to establish the transliteration exclusively for research oriented purposes. HISTORY It was decided during the second Dubrovnik International Conference on the Sanskrit Epics and Pur-a.nas (DICSEP2, 1999) to make “transitional texts” the main topic of the next conference in 2002. The HV being an exemplary text in that respect marking the transition between Epic and Pur-a.na as literary genres, and discovering that among the international community of scholars specialising on the Epics and Pur-a.nas there was a sizeable group who are already working on the HV, the scholars assembled for DICSEP 2 accepted the proposal by Prof. Dr. Horst Brinkhaus (University of Kiel, Germany) to make an electronic transliteration of the HV the goal of a joint effort and project. For reasons different for each of the contributors the project was completed only for DICSEP 4 in September 2005, even though it should be recorded that some of those who had committed themselves to a section delivered their portions in good time. As each chapter carries a header specifying its origin and history it is sufficient to list here the names of our contributors (in alphabetical order): Atul Agarwala J. Wayne Bass Julie Bélanger Peter Bisschop Horst Brinkhaus John Brockington André Couture Eva De Clercq James Fitzgerald Arlo Griffiths N. Hanemann Petteri Koskikallio Kreso Krnic Anne Mossner Luther Obrock Fran‡ois Painchaud Utz Podzeit Peter Schreiner Sandra Smets Renate Söhnen--Thieme Christophe Vielle Andreas Viethsen INPUT CONVENTIONS We decided to give preference to the maximum amount of text to be transliterated in the given time period (by 2002), even at the expense of some loss of flexibility of analytic mark--up. Thus, we began by transliterating the text as printed and reserved further mark--up (e..g. sandhi, separation of compounds, etc.) to a later stage of the project. The transliterated sections received from different contributors and sources were cumulated and, where necessary, revised and adjusted by Peter Schreiner (University of Zuerich). However, contributors were invited, encouraged and urged to adopt the following conventions as far as possible -- the most likely exception being the typing of conventions for diacritics. The present input conventions were chosen largely because there already exist programs and procedures to generate indexes etc. for texts conforming to these conventions. To change the coding of diacritics to another system on the basis of input already cumulated and standardized is not difficult. We need not waste time and energy discussing alternatives as long as everybody observes the fundamental rule that mark--up has to be EXPLICIT and UNAMBIGUOUS. This document was created by utilizing the conventions described below. When modifying the transliteration scheme according to the different schemes offered by GRETIL, Reinhold Grünendahl discovered a considerable number of typing mistakes as well as deviations or variants of our conventions. He very kindly corrected them in the modified files and provided a list of these mistakes. At the same time he found it necessary and convenient to change the format of the textual references and the coding of da.n.das separating verse quarters in order to make the file conform to established GRETIL conventions and to be able to provide a p-ada--index. In order to avoid duplication of work (the correction of input mistakes) as well as to avoid the coexistence of different versions, the modifications and corrections implemented by Reinhold Grünendahl were accepted in the file on the Zürich server while, however, the original transliteration scheme was retained. The present text records the original conventions for the sake of recording the project history, but it adds the GRETIL modifications which describe the actual format of the file. Special thanks are due to Reinhold Grünendahl for the time, the attention and all the effort he has invested to improve this electronic text of the HV! 1. HEADER: Each file (in fact, each chapter) begins with information about the source, in this case the critical edition of the HV and the text contained in the file, e..g. "HV (CE) chapter 28." The header also contains the names of the person to whom the transliteration is to be credited and of other persons who worked on the text; and it contains a date or dates to identify the version. The header is delimited by [h: ... :h]. 2. TRANSLITERATION 2.1. DIACRITICS: For typing diacritics, the contributors to our HV--project actually used the conventions familiar to their fingers and keyboards, on condition that they produced an unambiguous ASCII--version of their file in which all codes are preserved (e..g. by saving the file in RTF--format, by converting the input codes to TeX--format, to ITRANS--format, to the so--called Harvard--Kyoto--format, or to whatever else may be in use). For the cumulated output, we use what has been called the TZ--format (for _T_uebingen--_Z_uerich) as standard. In the TZ--format (and thus in this document) diacritical signs are represented by punctuation marks typed before the corresponding letter: - (hyphen) = macron . (period) = subscript dot ; (semicolon) = superscript dot / (slash) = aigu ? (question mark) = tilde Where a punctuation mark before a letter is NOT a diacritical sign, it will be doubled. The same rule is applied by contributors if their files contain non--Sanskritic material (e..g. in the header or in comments). 2.2. WORD SEPARATING BLANK: Word separating blank is inserted wherever the transliteration allows it, e..g., apy eva (not apyeva). 2.3. DA.N.DA: A sign representing the da.n.da (vertical bar, which serves as punctuation mark in Sanskrit texts and for which we generally used exclamation mark) is typed at the end of each metrical line (i..e., blank da.n.da). The quarter of a verse (p-ada) should also be marked by something representing a da.n.da sign (e..g., an exclamation mark) in order to enable the generation of a P-ada--index. The da.n.da sign is NOT followed by a blank before verse quarters 2 and 4 (b and d) in anu.s.tubh metre (/sloka), in order to be able to distinguish even and uneven p-adas. E..g., vistare.naiva sarv-a.ni !karm-a.ni ripugh-atina.h ! /srotum icch-amy a/se.se.na !hare.h k.r.s.nasya dh-imata.h !!30.1! The da.n.da sign (e..g., exclamation mark) is also used to delimit the textual references (i..e. the numbers identifying textual units). A double da.n.da (two exclamation marks without intervening blank) is written at the beginning of the reference figures, a single da.n.da is written at the end of the reference figures (cf. below, on "line division" and "references"). GRETIL modifications: / = da.n.da in constituted text | = da.n.da in *--passages Pada markers: For a four--pada verse: ........ $ ........ & ........ % ........ // For a six--pada verse: ........ $ ........ & ........ % ........ \ ........ # ........ // 2.4. LINE DIVISION: In verses of anu.s.thubh--metre a new line begins after quarters 2 and 4. Longer metres (longer than the /sloka, that is) are typed with each p-ada on a separate line. Prose passages are divided according to where the source prints references and da.n.das (cf. below, section on references). 2.5. PROSE PASSAGES: The main part of the text being in verse, only the beginning and end of prose passages is marked, e..g., by [pa: "Prosa--Anfang" (beginning of prose) :pe] "Prosa--Ende" (end of prose) GRETIL modification: ||* ... *|| = prose 2.6. REFERENCES: The full reference (chapter and verse) is given at the END of the verse to which it refers (as in the Devan-agar-i text). (The full reference needed to be typed only for the first verse of each chapter, e..g. !!30.1!, while for the following verses only the verse numbers needed to be typed, e..g. !!2!, !!3!, etc. The chapter number was inserted automatically at a later stage. This avoided unnecessary typing of numbers, reducing the chance of typing mistakes! On the other hand, mistakes could be “automatically” multiplied.) The beginning of references is marked by a double bar and the end is marked by a single bar. A reference is always followed by a new line. GRETIL modifications for references: Text: STRUCTURE OF REFERENCES (added): Text: HV_nn.nn = HV(constituted text)_adhy-aya.verse *HV_nn.nn*nn:nn = HV(*--pass.)_adhy-aya.verse*pass.--number:line 2.7. STAR--PASSAGES: Star--passages are typed at the place were they are inserted. Each line of a star--passage begins with "x " (i..e., letter x followed by a blank). The information about the source of the passage is copied at the beginning of the passage concerned and ends with TWO colons, e.g. "x N (except /S1) T1.3.4 ins.:: ..." N..B.: NO NEW LINE after the double colon! This "source information" may include statements about different or repeated occurrences of the passage inserted at different places as given in the CE. The reference of star--passages comes at their very end. The reference must be complete, i..e. include chapter and verse to which the passage refers plus the number of the star--passage, e..g. "!!30.20*451!", or "!30.33ab*452!" if the passage is inserted after the first line of a verse. If the star passage is inserted before the first verse of a chapter, it may carry the verse number zero, e..g., "!!30.0*449!" Since the identification of a speaker (e..g. k.r.s.na uv-aca:) is not counted as a separate line, insertions "after the reference" (thus the terminology of the CE) are put _after_ the line of text which follows. Since the asterisk serves to separate and identify the following part of a reference, it is always written BEFORE the number of the star passage (this deviates from the convention of the Critical Edition which puts the asterisk after the number). Thus, references to other star--passage in the information about the mss. in which a *--passage is found (to be copied at the beginning of the passage, see above) are modified, e..g. "T1 after *479" in the introduction to *480. In star--passages the CE counts lines (not verses); these numbers are copied after the da.n.da(s) (to keep track of the transliterated lines for proof--reading etc.), e..g. ! [5], !! [10], ! [15], etc. (cf. sample input, *449). If the last line of a *--passage (which ends with the reference) carries also a line counter, the latter is omitted. *--passages within *--passages (second level star--passages) FOLLOW the star--passage to which they belong and carry the same verse number of the constituted text as the corresponding first level star--passage (e..g., "31.148*482A"). The exact place of insertion will be evident from the introductory information about mss. evidence. And the fact that it is a second level star--passage is evident from the capital letter in the reference. GRETIL modifications (formatting of star--passages): Text: ALL INSERTIONS begin with 5 blanks. Thereafter, one of the following options may apply: 5 additional blanks for the first line of insertions after UNEVEN pad-as 3 additional blanks for EVEN p-adas in longer metres with split lines ALL 1--PADA LINES end with asterisk (*). 2.8 APPENDIXES Appendix passages are strictly speaking star--passages which are too long to be printed at the bottom of the page. Since, however, this transliteration aims at reproducing the printed edition, the appendixes (unlike star--passages) are not inserted in the constituted text. However, a note mentioning that an appendix passage is inserted should be inserted at the appropriate place. This note, i..e., the cross reference to the appendix, is treated either like the introductory part of a *--passage (concluding with a double colon but with no text following), e..g., "X After 7.45, /S1 K1.3 Dn Ds D1.4 ins. a passage given in App. I (No. 1)::" Or this information is given as a comment delimited by [k: ... :k]. The appendix passages are treated like chapters. They are numbered as if they were chapters above 9000 and are counted by tens, e..g. 9010, 9020, etc., for Appendix 1, 2, etc. Thus, 9421 represents Appendix 42A, while 9422 can represent Appendix 42B. (Since Appendix II is being postponed for the time being, the Roman numeral identifying Appendix I can be dropped.) The lines of appendix passages do not begin with "x ". The line numbers, however, are copied as for the star--passages. The reference ("chapter", i..e. appendix number, plus line number) is given at the beginning of the passage as a separate reference delimited by [r: :r], e..g. [r: 9400 :r] at the beginning of appendix 40. It is sufficient to give this reference only once. Star passages in appendixes are treated like star--passages in the main text, i..e. they begin with "x " and carry their reference at the end, e..g. "!!9422.71*10!". A single asterisk suffices! Chapters (adhy-ayas) are marked by "Colophon" in the printed edition, but they are not numbered. The word "colophon" may be considered as an "editorial addition" (see below, 2.15) and is thus typed in single square brackets, i..e. "[Colophon]", on a separate line. To facilitate orientation in the often lengthy appendix passages, an automatic line count was executed and the automatically generated line number was inserted along with the number of the appendix at the beginning of each line in pointed parentheses, e..g. <9400.24> (i..e. line 24 of appendix 40). Where applicable, the number of the star passage is added after asterisk, the line numbers of the star passage being added after slash, e..g., <9400.24*5/7> (i..e. line 7 of star--passage 5 added after line 24 in appendix 40 -- an invented example!). The application of this program implied a welcome checking of the typed line numbers as any discrepancies could be discovered by comparing the two line counts. GRETIL modifications in Appendixes: STRUCTURE OF REFERENCES (added): HV_App.I,nn.nn = HV_App.I,section.line **HV_App.I,nn.nn**nn:nn = HV_App.I,section.line**pass.--number:line ALL INSERTED LINES (= **--passages) begin with 5 blanks. ALL EVEN p-adas in longer metres (with split lines) begin with 3 blanks. (Consequently, EVEN p-adas in **--passages begin with 8 blanks.) / = danda in regular verse line (i.e., a line containing two p-adas) $ = p-ada separator in regular verse line | = danda in split--line verses @ = marks end of uneven pada in split--line verses, e.g.: " @ HV_App.I,20.664 @" 2.9 SANDHI: Sandhi is not marked and vowel sandhi was not dissolved at this stage. Thus, we typed "caiva" (for "ca eva"), "mah-atman" (for "mah-a+-atman"), etc. 2.10 SEPARATION OF COMPOUNDS: Compounds are not separated but are typed as single words. 2.11 VARIANT READINGS: Variant readings are disregarded at this stage of our project; we transliterate the text as constituted by the CE. 2.12 PROSE IDENTIFICATIONS OF SPEAKER: The inserted identifications of speaker(s) are copied as printed and are concluded by a SINGLE colon, e..g., vai/sa.mp-aya.na uv-aca: (As these lines are not counted in the CE, the speaker identifications are not retained in a separate line in our output.) GRETIL modification: {...} = interlocutor (these lines are not counted in *--passages!) 2.13 COMMENTS BY THE TRANSLITERATOR: Passages which are not part of the transliterated, printed text but which contain comments, annotations (e..g. about printing mistakes), additional information, etc., are clearly delimited, e..g. by [k: ... :k] enclosing "K"ommentar (i..e., comments) by the transliterator. 2.14 EDITORIAL ADDITIONS: Any annotations by the editor(s) of the edition used for the transliteration (e..g. conjectures, markers for lacunae etc.) which are part of the printed edition are enclosed in single parentheses. 2.15 COLOPHONS: Colophons are not transliterated. But note that the first verse of each chapter must include the chapter number in the reference! "[Colophon]" is typed in appendixes to mark the beginning of a new chapter; cf. above, 2.8. Kindly address questions, suggestions, criticisms, etc., to: Peter Schreiner Horst Brinkhaus Abteilung fuer Indologie Abteilung fuer Indologie Universitaet Zuerich Christian--Albrechts--Universitaet Raemistr. 68 Leibnizstr. 10 CH--8001 Zuerich D--24118 Kiel Switzerland Germany email: pesch@indoger.unizh.ch email: hbrinkhaus@indo.uni-kiel.de tel.: +41-1-634.2036 (or 634.2057) tel.: +49-431-880.3317 fax +41-1-634.4958 fax: +49-431-880.1598