09-325

ISO/IEC JTC1/SC2/WG2 N3692 L2/09-325 2009-09-28 Universal Multiple-Octet Coded Character Set ... names, and some writers equate name with script ident...

0 downloads 146 Views 2MB Size
ISO/IEC JTC1/SC2/WG2 N3692 L2/09-325 2009-09-28 Universal Multiple-Octet Coded Character Set International Organization for Standardization Organisation internationale de normalisation Международная организация по стандартизации Doc Type: Working Group Document Title: Roadmapping the scripts of Nepal Source: UC Berkeley Script Encoding Initiative (Universal Scripts Project) Author: Michael Everson Status: Liaison Contribution Date: 2009-09-28 Analysis of the Nepalese scripts is slightly complicated because different font styles have different names, and some writers equate name with script identity. Analysis of the chief features suggests that a number of these can be unified on structural grounds. 1

RAÑJANA encompasses: Rañjana (Figure 1, 2, 3) Wartu (Figure 4)

This is the most complex of the scripts, in particular because it is used for Tibetan, which has introduced some innovations from Tibetan script itself which do not occur in other Brahmic scripts. (An example is the use of an explicit “head-ra” alongside regular ra.) One unique feature is the long or short mark used for -ā; compare  ka,  kā and  ga,  gā. Named scripts which should be unified here are Rañjana (Tibetan  lañdza), and Wartu (Tibetan ཝརྟུ wartu). Stylistically, Wartu is much the same as Rañjana but with a wiggly headline on many of the letters. See Figures 3 and 4. 2 PRACALIT There are many scripts in this group, mostly distinguished by their headlines. The major difference between the PRACALIT scripts and the RAÑJANA scripts is the way in which -e and -ai are made by changing the top bar (see Figure 20). Two major varieties are distinguished, and there is not yet enough evidence available to determine whether or not it is appropriate to encode them separately from one another. At this stage, at least we can distinguish: 2.1 PRACALIT flat-headed script, which encompasses: Pracalit (Figures 5, 6, 7) Pācūmol (Figure 8) Hiṁmol (Figure 9) Kuṁmol (Figure 10) 2.2 BHUJIṀMOL curve-headed script, which encompasses: Bhujiṁmol (Figures 11, 12, 13) Golmol (Figure 14) Kveṁmol (Figure 15) Litumol (Figure 16) Page 1

Documentary evidence for the popularity or use of these so-called “nine scripts” (the eight PRACALIT scripts plus Rañjana) is rare except for Rañjana, Golmol and current Pracalit script in which the majority of Nepalese manuscripts and inscriptions were written from the 15th to 18th centuries CE. Golmol is said to have been introduced in Nepal by Bhaumagupta (ca. late 6th century CE.). The heyday of its use was, however, the early medieval period, i.e., the 13th-15th century CE. Though it was widely used in the 13th century, even then its use was not universal. Traditionally, Rañjana was restricted to copying sacred texts—Hindu as well as Buddhist. It was rarely, if ever, used for secular purposes. More recently, however, Newars have used Ranjana as an ornamental script for invitations, very formal manuscripts and so on, and until about 1900 used it for their Sanskrit and Newari manuscripts. In the 1990s it was common to see posters and banners advertising Buddhist lectures, new year's parties and so forth with the heading in Rañjana and body in Newa Lipi (Pracalit) which only Newars could read. Encoding considerations. It should first be said that some members of the user community have criticized the idea of unifying these “scripts”. It may be that this is a misunderstanding of the UCS; the analogy of the Latin script with its Gaelic and Fraktur variants, however, is probably applicable, which is why the recommendations here have been made. A complete inventory of the shaping behaviour of any of the PRACALIT scripts (apart from Pracalit itself) has been unavailable, but it seems unlikely that any of them will exhibit behaviour that should warrant anything but the standard VIRAMA encoding model ( KA + VIRAMA +  SHA =  KSHA) to be used for them. It has been suggested, however, that the Tibetan stacking model (ཀ KA + ྵ SUBJOINED-SHA = གྴ KSHA) could be used for the RAÑJANA scripts (Rañjana and Wartu). It does not seem that there is any benefit on either structural or practical grounds to doing so. Structurally, Rañjana’s behaviour is classically Brahmic; གྴ is a stack:  is a conjunct. Using the Tibetan model would seem to be artificial. Further discussion of the options and elaboration on this point is expected during the proposal process itself. The Rañjana scripts differ in many ways from Tibetan. Both of them (and all the PRACALIT scripts) have unique Brahmic voiced aspirates  gha,  jha,  ḍha,  dha, rather than the Tibetan stacked གྷ, ཇྷ, ཌྷ, དྷ, བྷ, which are decomposed for Tibetan in Normalization Form C, so a one-to-one transliteration is in any case not feasible. Figures 17, 18, and 19 show examples of Rañjana consonant clusters. These are not simple stacks or even complex consonant conjuncts: they are monograms representing seed syllables, and should probably be encoded as such. Some of these may also be “example” clusters, such as kkhgghṅa (ka + kha + ga + gha + ṅa), and three different examples of cchjjhña (ca + cha + ja + jha + ña). Although it is in principle possible to predict what a string of a Tibetan base + n subjoins will look like, it wouldn’t be possible to do the same for Rañjana; in Figure 18, for instance, several of the conjunct monograms have more than one rendering. It is unlikely that any rendering engine should be expected to render these dynamically; rather, they ought to be font-specific special ligatures, or, as mentioned above, encoded as monograms. (It should be pointed out that these are typically not set in type, but hand-drawn.)

Page 2

Bibliography Gu Wenyi & Shi Xueli. 1995. Gaṅs-can mkhas paʼi phyag bris sna tshogs phyogs bsdus rin chen phreṅ ba (= Zang wen shu fa hui cui). Lanzhou : Kan-suʼu mi rigs dpe skrun khaṅ : Kan-suʼu Źiṅ-chen Źin-hwa dpe khaṅ gis bkram pa. ISBN 7-5421-0048-3 Hemraj Shakyavansha. 1985. Nepalese Alphabets = Nepāl lipi saṁgraha. Seventh Edition. Shakya, Rabison. 2002. Nepāl lipi Varṇamālā = Alphabet of the Nepalese script. Patan: Motirāj Shakya. ISBN 99933-34-36-7 Tshe-tan Zabs-drun. 1992. Landzai thig rtsai dka gnad gsal byed dba brgya ldan pai glog sgron zes bya ba dpe ris dan bcas pa. ISBN 7-5420-0248-1 Acknowledgements. This project was made possible in part by a grant from the U.S. National Endowment for the Humanities, which funded the Universal Scripts Project (part of the Script Encoding Initiative at UC Berkeley) in respect of Rañjana and Pracalit encoding. Any views, findings, conclusions or recommendations expressed in this publication do not necessarily reflect those of the National Endowment of the Humanities.

Page 3

Figures

Figure 1. Consonants with vowel matras in Rañjana script from Shakya 2002. Note the particular shapes of -ā, -ī, -e, -ai, -o, and -au.

Figure 2. Consonants in Rañjana script from Shakya 2002.

Page 4

Figure 3. Example text in Rañjana script from Gu & Shi 1995. Note that the shapes of a, e, and o are unrelated to one another, that the voiced aspirates are Brahmic, and that ḷ and l are distinct.

Figure 4. Example text in Wartu script from Gu & Shi 1995. Note that the shapes of a, e, and o are unrelated to one another, that the voiced aspirates are Brahmic, and that ḷ and l are distinct. Page 5

Figure 5. Consonants with vowel matras in Pracalit script from Shakya 2002. Note the horizontal head-line, and the particular shapes of -u, -e, -ai, -o, and -au.

Figure 6. Consonants in Pracalit script from Shakya 2002.

Page 6

Figure 7. Alphabet in Pracalit script from Shakya 2002. Note that the shapes of a, e, and o are unrelated to one another, that the voiced aspirates are Brahmic, and that ḷ and l are distinct.

Figure 8. Alphabet in Pācūmol script from Shakya 2002. Note that the shapes of a, e, and o are unrelated to one another, that the voiced aspirates are Brahmic, and that ḷ and l are distinct. Page 7

Figure 9. Alphabet in Hiṁmol script from Shakya 2002. Note that the shapes of a and e are unrelated to one another, though o is based on a, and that the voiced aspirates are Brahmic.

Figure 10. Alphabet in Kuṁmol script from Shakya 2002. Note that the shapes of a, e, and o are unrelated to one another, and that the voiced aspirates are Brahmic. Page 8

Figure 11. Consonants with vowel matras in Bhujiṁmol script from Shakya 2002. Note the curved a head-line, and the particular shapes of -ā, -u, -e, -ai, -o, and -au.

Figure 12. Consonants in Bhujiṁmol script from Shakya 2002. Page 9

Figure 13. Alphabet in Bhujiṁmol script from Shakya 2002. Note that the shapes of a and e are unrelated to one another, though o is based on a, and that the voiced aspirates are Brahmic.

Figure 14. Alphabet in Golmol script from Shakya 2002. Note that the shapes of a, e, and o are unrelated to one another, and that the voiced aspirates are Brahmic. Page 10

Figure 15. Alphabet in Kveṁmol script from Shakya 2002. Note that the shapes of a, e, and o are unrelated to one another, and that the voiced aspirates are Brahmic, and that ḷ and l are distinct.

Figure 16. Alphabet in Litumol script from Shakya 2002. Note that the shapes of a and e are unrelated to one another, though o is based on a, and that the voiced aspirates are Brahmic, and that ḷ and l are distinct. Page 11

Figure 17. Examples of monograms text in Rañjana script from Gu & Shi 1995. These read in transliteration from left to right as: kkhgghṅa, cchjjhña, ṭṭḍḍhṇa, tthddhna, pphbbhma, yrlwa, śṣsha.

Figure 18. Example text in Rañjana script from Gu & Shi 1995. These read in transliteration from left to right in rows as: ¶, kkhgghṅa, cchjjhña, cchjjhña, cchjjhña ṭṭḍḍhṇa, ṭṭḍḍhṇa, tthddhna, pphbbhma, yrlva, śṣsha hkṣmlwryaṁ (ha kṣa ma la va ra yaṁ), madagalaṁ ~ maṁdagala

Page 12

Figure 19a. Monogram in Rañjana script from Tshe-tan Zabs-drun 1992.

Figure 19b. Example text in Rañjana script from Tshe-tan Zabs-drun 1992. The text of the syllable is from the Kalacakra Vajra mantra. The green elements in this rendering should read kṣa ma but it is difficult to parse as it is the lowest element that ought to be the ma. The la wa ra ya (yellow, white, red, black) are quite clear. Page 13

Ra

Pra

Bhu

Ra

Pra

Bhu

Figure 20. Summary chart comparison of Rañjana, Pracalit, and Bhujiṁmol from Gu & Shi 1995; These are taken from Figures 1, 5, and 11.

Page 14

Ra

Wa

Pra



Hiṁ

Kuṁ

Bhu

Gol

Kveṁ

Li

a

e

o

gha

jha

ḍha

dha

bha



la Figure 21. Summary chart comparison of Rañjana, Wartu, Pracalit, Pācūmol, Hiṁmol, Kuṁmol, Bhujiṁmol, Golmol, Kveṁmol, and Litumol, from Gu & Shi 1995; These are taken from Figures 3, 4, 7, 8, 9, 10, 13, 14, 15, and 16.

Page 15