Peter Schreiner [ed.]: N-ar-aya.n-iya--Studien

The files contained in the package NARAYANIYA.ZIP accompany the following
book:

Reinhold Grÿuunendahl ; Angelika Malinar ; Thomas Oberlies ; Peter
Schreiner: N-ar-aya.n-iya--Studien. Herausgegeben von Peter Schreiner.
Wiesbaden: Harrassowitz, 1997.  (Pur-a.na Research Publications
Tÿuubingen ; 5) 15, 642 p., DM 128.- 

The table of contents of the book is made available as Postscript file
for printout (NAR_CONT.PS). A working report, including the description
of the input conventions used in the present file forms part 2 of the
introduction to that book. The relevant sections are reproduced below
under NARR. 

CORRECTIONS: The kind attention of users of the printed lemma--index is
drawn to the passages below, marked "CORRECTIONS"! They are also
collected in the file CORRECT.

The package NARAYANIYA.ZIP includes the following files which are made
accessible as additional material for the textual analysis of the
N-ar-aya.n-iya. The file names are:
NARR
NARV0T
NARMET
NARWIR
LMIDXR
NAR_CONT.PS
CORRECT

The following remarks concern the peculiarities of each file: 

=====

NARR: 

This file contains the transliterated text of the N-ar-aya.n-iya
according to the Critical edition, including all "star--passages" and
appendices, but not including variant readings.

Transliteration:
Diacritical marks are transliterated by punctuation marks typed in front
of the letter to which they belong; this imitates the traditional
"layout" of typewriters where accents etc. are placed on dead keys typed
before the character. 

 . = subscript dot (e..g., k.rta.h)
 ; = superscript dot (e..g., a;nga)
 ? = tilde (superscript) (e..g., praj?n-a)
 - = superscript hyphen (macron) (e..g., -atm-a)
 / = aigu (superscript) (e..g., /s-astra) 

Where, as in an introductory documentation or in comments to variant
readings within the text, the use of punctuation marks in their proper
function has to alternate with their function as diacritical marks,
their use as punctuation marks is distinguished by a following blank (or
other signs of punctuation including parentheses) or by doubling where a
blank is not possible (e..g. in abbreviations). 

The quarter of a verse (p-ada) is marked by | (vertical bar Ä written as
exclamation mark in the input of plain Sanskrit text files). This bar
(da.n.da) is not followed by a blank before verse quarters 2 and 4 (b
and d) in anu.s.tubh metre (/sloka). After quarters 2 and 4 a new line
begins.

Longer metres (longer than a /sloka, that is) are typed in such a way
that each p-ada begins a new line.

The full reference (part, chapter and verse) is given at the end of the
verse to which it refers. The first segment of the reference includes
the number for the book or part according to the formula: book times
1.000 plus chapter <full stop> verse; 12,321.5 becomes 12321.5.

The beginning of references is marked by a double bar and the end is
marked by a single bar. After each reference a new line begins.

In addition to the reference as given in the printed text, each sentence
(i..e. line) is numbered in accordance with the internal reference
system of TUSTEP is written between pointed parentheses before the text;
its first and second segments correspond to the textual reference as
explained above (parvan, chapter, verse); the segment after the slash
counts the lines within each verse. Lines of the base text are numbered
by hundreds (with the zeros on the left margin not being printed:
12321.1/1(00), 12321.1/2(00); lines which belong to star passages are
numbered by the units and decimal digits: 12321.1/001 if the addition is
placed before the first line of verse 1; or 12321.1/101 if the addition
is placed after the first line (i..e. before the second line) of the
first verse. An exception is made for ch. 325 (the prose stotra), where
the number of sub--units requires three digits, and for the prose
passages in ch. 329. 

Star--passages begin with [X: and end with :X]; each line of a
star--passage begins with X. The (two) appendix passages are placed at
the end of the text; they have been given the virtual chapter number
999. The reference of star--passages begins with single da.n.da (while
references in the constituted text begin with double da.n.da); to the
verse number is added the number of the star--passage (e.g.
!12321.26*800! means "star--passage no. 800 which belongs to verse 26 of
chapter 321 of the /S-antiparvan") or the number of the appendix (e..g.
!12999.31/5! means "line 5 of appendix 31 in the /S-antiparvan"). 

In chapter 325 the numbers in parentheses count the epithets listed in
this hymn.

Explicit statement of the speaker (xxx uv-aca) are transliterated as
printed; they begin a new line and are marked off from the following
text by colon (e..g. bh-i.sma* uv-aca:). 

Sandhi:
Markers are added to the transliteration of the printed (Devan-agar-i)
version in order to mark sandhi changes. A sandhi change is defined with
regard to the "pausaform" of a word, i..e., the form a word would take
at the end of a line or out of context (vigraha). Note that this
pausaform need not be identical with the stem in a dictionary (e..g.
pausaform puna.h vs. lemma punar, pausaform tama.h vs. stem tamas).

Consonants which have undergone a sandhi change in the text are marked
by asterisk (*). Similarly, final vowels which have changed due to
sandhi are marked by * (e..g., -as-id* r-aj-a nalo* n-ama).

Vowel sandhi is dissolved and marked (e..g., na*asti, ca*eva).
Similarly, avagraha is reconstituted, the originally omitted initial "a"
being marked as a sandhi vowel (e..g., devo* *api).

In some special cases the marking of sandhi has to be extended to
include some disambiguating information:
 Ä to half--vowels which substitute for a long vowel the diacritic for
 "long vowel" is added (e..g., devy-* api);
 Ä if final --a in sandhi does not stand for --a.h (with visarga), then
 the original vowel which has been substituted by the --a is added
 (e..g., "lokae* eva", where "loka eva" is printed, which is the sandhi
 form for "loke eva").

In case of "double sandhi" the sandhi is marked by double **, e..g.
"sa**eva" in case of printed "saiva" (instead of the correct "sa eva").

Blank is inserted between words wherever this is possible in
transliteration (but not necessarily in Devan-agar-i), e..g., hy* api,
devo* *api.

Separation of compounds:
Separation of compounds is marked by inserting + between the members of
a compound (e..g., nara+n--ar--aya..na). In case of sandhi, the +
functions also as a sandhi--marker, i..e., no additional sandhi--marker
is used (e..g., tapo+vane, mah--a+--atmana..h).

Separation of compounds is restricted to nominal compounds (including
upapada--compounds like ura+ga, go+p--i) and does not include
grammatical analysis. 

=====

NARV0T: 

This file contains the text (WITHOUT star--passages and appendices) in
"text format", i..e. the running text without markup of sandhis,
compounds, p-adas, disambiguating figures (but WITH prefixed numerical
reference for each line).

=====

NARMET: 

This file contains the text (as in NARV0T) of all metrical passages in
"text format" with a schematic representation of the metrical pattern of
each line added in front of the text. 

The following is a list of passages in meters other than Anu.s.thubh:
322.1-12
327.89
327.107
329.12
330.25
334.13-17
335.88-89
336.77
337.3-5
337.47-48
337.68-69
339.10
339.16-19

The following is a list of the prose passages (omitted in NARMET):
325.4/001-325.4/171
328.53
329.3-329.8/5
329.13-50

In the presentation of the metrical patterns, the figure zero stands for
"long syllable", the figure one for "short syllable". This code was
suggested by B. van Nooten ("The /sloka in the Sabh-aparvan." In:
Journal of the Oriental Institute 17 (1967-68), p. 353-364); it allows
to read the pattern of each verse quarter as a binary number which can
be transformed into a decimal notation. (The use of the decimal code of
each line makes an index of metrical pattern considerably more
transparent.)

[m: marks the beginning of the metrical pattern; [t: marks the beginning
of the text. The comma after the 8th syllable of anu.s.thubh lines has
been inserted automatically to demarcate verse quarters. In longer
meters each quarter forms a separate line.

=====

NARWIR: 

This is a reverse index of all occurring wordforms (14530 tokens),
sorted according to the Devan-agar-i alphabet but beginning from the end
of each entry. Before generating the index each word is automatically
transformed to its pausaform (resulting in 5642 types). Compounds are
treated as individual words; since the occurrence of each element word
is listed under its lemma in the printed lemma--index, the element words
constituting a compound are not included in this index. The absolute
frequency of each wordform is added in parentheses.

For the sake of briefness, the reverse index does not repeat the textual
references (to where a wordform occurs); this information is available
easily in the printed lemma--index or by a search in the input file
(NARR).

=====

LMIDXR: 

(NOTE: In this file diacritics are coded according to the
TUSTEP--conventions: %- = macron; %.. = subscript dot; %? = tilda; %. =
superscript dot. &.wu stands for root--sign, #'1 and #'2 stands for the
superscript disambiguating figures.)

This file contains a reverse index of the 2769 lemmas which make up the
vocabulary of the N-ar-aya.n-iya. This index is automatically generated
from the source file on which the printed lemma--index is based (for
corrections see below and in the note concerning the file NARR). It
repeats the main entries of the printed lemma--index, sorted in reverse
order, i..e. beginning from the end of each lemma, according to the
Devan-agar-i alphabet (the disambiguating figures are ignored for
sorting). In the printed index, prefixes to roots are placed behind the
root (marked by the root sign (here coded as "&.wu"), e..g. &.wu gam +
upa; &.wu gam + sam, upa) ; consequently, in sorting the reverse index
prefixes (and their combinations) come first. Nominal stems end with
hyphen; indeclinables (including non--paradigmatic forms like ablatives
in --tas) are unmarked. All additional information (frequency,
references, cross--references) is available from the printed
lemma--index.

=====

CORRECTIONS: (version of 10 July 1997)
(To be added as correction sheet to the printed version.)

Too late in the production process of the printed book it was discovered
that the identifications of speakers have been included in the lemma
index (cf. the entries uv-aca:, -ucatu.h:, -ucu.h:). This effects
especially the frequencies of entries of proper names (e..g. n-arada.h).

Only after the release of the printed book I discovered the mistakes in
lines 12329.27/5 and 12336.76/2 of the input file; the file NARR is thus
no longer identical with the text on which the lemma--index in the
printed book is based.

When generating the reverse lemma--index (unfortunately after the
publication of the book) the following errors were discovered in the
printed lemma--index:

&.wu -ap + sam, pra (instead of "sam + pra")

/sl-ila- (instead of "sl-ila")

&.wu han + sam, -a (instead of "sam + -a")

dhi- with a cross reference to &.wu dh-a including the tag "(vl:" should
be eliminated.

&.wu p-a_2 + adhi (root sign is missing)

&.wu j?n-a + abhi, sam (delete anusv-ara)

final -s (instead of visarga) in p-adatas, n-amatas, /s-astratas,
        p-ar/svatas, m-urtitas

t-avat (adverbial use of t-avat-) should have been lemmatized under the
        adjective t-avat-

pa/syat- (instead of pa/syant-, occurrences to be listed under pa/syat-)
bharat- (instead of bharant-)
bhagavat- (instead of bhagavant-, occurrences to be listed under bhagavat-)
vegavat- (instead of vegavant-)
gandhavat- (instead of gandhavant-)
himavat- (instead of himavant-, occurrences to be listed under himavat-)
sah-ayavat- (instead of sah-ayavant-)
praj?n-avat- (instead of praj?n-avant-)
sat- (instead of sant-, occurrences to be listed under sat-)

vanas- (occurring in "vanas+pati-") should have been lemmatized as vana-.

mukta- (one occurrence to be deleted)

These corrections have been implemented in the file from which the
reverse lemma--index is generated; to that extent the basis of the
reverse lemma--index is not identical with the printed lemma--index.

=====

The readers of this file and the users of the attached materials for a
textual analysis of the N-ar-aya.n-iya are cordially invited and
encouraged to send corrections, comments, criticisms (and perhaps
acknowledgments) to

Peter Schreiner
Abteilung fuer Indologie
Raemistr. 68
CH-8001 Zuerich
Switzerland

email pesch@indoger.unizh.ch

(4 September 1997)