Brahmic Scripts

Discussion of natural languages, or language in general.
User avatar
Drydic
Smeric
Smeric
Posts: 1652
Joined: Tue Oct 08, 2002 12:23 pm
Location: I am a prisoner in my own mind.
Contact:

Re: Turkestan Brahmi

Post by Drydic »

...only?
Image Image
Common Zein Scratchpad & other Stuffs! OMG AN ACTUAL CONPOST WTFBBQ

Formerly known as Drydic.

User avatar
clawgrip
Smeric
Smeric
Posts: 1723
Joined: Wed Feb 29, 2012 8:21 am
Location: Tokyo

Re: Turkestan Brahmi

Post by clawgrip »

When you compare it to my original idea of making each conjunct as an independent glyph, which Richard W rightly pointed out was a waste of time, it seems like less trouble. Plus, most of those variations are not big variations, just minor adjustments to allow them to join up.

User avatar
Morrígan
Avisaru
Avisaru
Posts: 396
Joined: Thu Sep 09, 2004 9:33 am
Location: Wizard Tower

Re: Turkestan Brahmi

Post by Morrígan »

clawgrip wrote:When you compare it to my original idea of making each conjunct as an independent glyph, which Richard W rightly pointed out was a waste of time, it seems like less trouble. Plus, most of those variations are not big variations, just minor adjustments to allow them to join up.
It's not much worse than Ge'ez, except that you have to set up the substitution rules.

User avatar
clawgrip
Smeric
Smeric
Posts: 1723
Joined: Wed Feb 29, 2012 8:21 am
Location: Tokyo

Re: Turkestan Brahmi

Post by clawgrip »

Oh right, Ge'ez. I guess that must be pretty time-consuming to make a font of.

If my calculations are right, I should need 95 glyphs to be able to make all the conjuncts. The majority of them are only slight alterations, like removing serifs or slightly adjusting or deleting lines on the base characters so that a subscript can attach smoothly, so it shouldn't take a really long time to get all of them. I've already got the 44 base characters done, and 21 of the remaining 51 done, so only 30 left. I also have a couple of the independent vowels left to do. Then I will have to look more seriously at how to code it.

User avatar
clawgrip
Smeric
Smeric
Posts: 1723
Joined: Wed Feb 29, 2012 8:21 am
Location: Tokyo

Re: Turkestan Brahmi

Post by clawgrip »

I'm making pretty good progress with completing all the necessary Tocharian glyphs.

Here's an example of the kind of variation that's necessary for a single letter:
Image

Thankfully, the majority of them only require one or two forms.

User avatar
clawgrip
Smeric
Smeric
Posts: 1723
Joined: Wed Feb 29, 2012 8:21 am
Location: Tokyo

Re: Turkestan Brahmi

Post by clawgrip »

Finished the Tocharian numerals:

Image

I have to wonder about a culture that feels that is an acceptable digit for 4.

User avatar
Ser
Smeric
Smeric
Posts: 1542
Joined: Sat Jul 19, 2008 1:55 am
Location: Vancouver, British Columbia / Colombie Britannique, Canada

Re: Turkestan Brahmi

Post by Ser »

Latin used .VIIII. for nine, which probably took just as much or even more time to write than the Tocharian four, if that helps.

User avatar
KathTheDragon
Smeric
Smeric
Posts: 2139
Joined: Thu Apr 25, 2013 4:48 am
Location: Brittania

Re: Turkestan Brahmi

Post by KathTheDragon »

Serafín wrote:Latin used .VIIII. for nine
I thought it was IX

User avatar
Ser
Smeric
Smeric
Posts: 1542
Joined: Sat Jul 19, 2008 1:55 am
Location: Vancouver, British Columbia / Colombie Britannique, Canada

Re: Turkestan Brahmi

Post by Ser »

I've even seen manuscripts where the same scribe varied between .iv. and .ix. and .iiii. and .viiii., so that's that.

User avatar
Jipí
Smeric
Smeric
Posts: 1128
Joined: Sat Apr 12, 2003 1:48 pm
Location: Litareng, Keynami
Contact:

Re: Turkestan Brahmi

Post by Jipí »

IIII, VIIII and such were definitely used, I've seen it in (transcripts of) manuscripts myself before, too. Cases in point: http://urts55.uni-trier.de/cgi-bin/iCor ... &such=iiii (searches ~4000 mostly High German, but also some Low German, Dutch and Latin deeds from the 13th century more or less well).

Richard W
Avisaru
Avisaru
Posts: 363
Joined: Sat Oct 16, 2010 8:28 pm

Re: Turkestan Brahmi

Post by Richard W »

clawgrip wrote:I have to wonder about a culture that feels that is an acceptable digit for 4.
The digits 4 to 7 look like compounds of the forms the Unicode chart gives for positional and non-positional notation! How reliable is your source?

User avatar
clawgrip
Smeric
Smeric
Posts: 1723
Joined: Wed Feb 29, 2012 8:21 am
Location: Tokyo

Re: Turkestan Brahmi

Post by clawgrip »

I discovered the sign for 1000, so I've updated the original image.
Richard W wrote:
clawgrip wrote:I have to wonder about a culture that feels that is an acceptable digit for 4.
The digits 4 to 7 look like compounds of the forms the Unicode chart gives for positional and non-positional notation! How reliable is your source?
My sources are TITUS and images of the original manuscripts. The shapes of the Tocharian numerals are also clearly related to Khotanese numerals. My guess is, if anything, it seems at least somewhat likely that the positional digits you're talking about were based on these non-positional ones, perhaps as Katakana was derived by taking specific setions out of kanji (e.g. イ is just the left side of 伊), though to be honest I don't know where they came from. But we can easily track the evolution from Brahmi to Tocharian and Khotanese if we imagine an intermediate form:

Image

EDIT: a few sources here:
4:
http://titus.fkidg1.uni-frankfurt.de/te ... 1135va.jpg second line, leftmost letter, thinner strokes
http://titus.fkidg1.uni-frankfurt.de/te ... 0401ra.jpg first line, first letter

5
http://titus.fkidg1.uni-frankfurt.de/te ... 0741ra.jpg last line, in the middle. Has a distinctive swash ascender

6
http://titus.fkidg1.uni-frankfurt.de/te ... 0300ra.jpg the ink is flaking and it's on the crease of a fold, but it's on the top line, directly under the 3 of 300

7
http://titus.fkidg1.uni-frankfurt.de/te ... 0187va.jpg second line, third letter (after the one with the dot on it)

This one has 5, 6, and 7:
http://titus.fkidg1.uni-frankfurt.de/te ... 0725ra.jpg
5: 2nd line, last letter
6: 4th line, last letter
7: last line, last letter

User avatar
clawgrip
Smeric
Smeric
Posts: 1723
Joined: Wed Feb 29, 2012 8:21 am
Location: Tokyo

Re: Turkestan Brahmi

Post by clawgrip »

Not much to show right now because I'm working on trying to nail down all the variant forms of each letter I'll need so that I can make any potential conjunct. It's difficult because Tocharian has some pretty unwieldy conjuncts that necessitate a lot of slight glyph variations. Here are just a few complicated conjuncts that can be found in actual Tocharian manuscripts:

Image

lska, kṣtsā, śtwa, dd, mbu, wñu, mt, mwtā, lymāṃ

These were not made with opentype; I just made them as individual glyphs to test how well the pieces fit together.

User avatar
Morrígan
Avisaru
Avisaru
Posts: 396
Joined: Thu Sep 09, 2004 9:33 am
Location: Wizard Tower

Re: Turkestan Brahmi

Post by Morrígan »

clawgrip wrote:Not much to show right now because I'm working on trying to nail down all the variant forms of each letter I'll need so that I can make any potential conjunct. It's difficult because Tocharian has some pretty unwieldy conjuncts that necessitate a lot of slight glyph variations. Here are just a few complicated conjuncts that can be found in actual Tocharian manuscripts:

Image

lska, kṣtsā, śtwa, dd, mbu, wñu, mt, mwtā, lymāṃ

These were not made with opentype; I just made them as individual glyphs to test how well the pieces fit together.
For things like these, I would probably just use use these as ligature substitutions rather than attempting any kind of positioning. Looks legit though, this is why I really loved Tocharian in my early days as an IE-ist.

User avatar
2+3 clusivity
Avisaru
Avisaru
Posts: 454
Joined: Fri Mar 16, 2012 5:34 pm

Re: Turkestan Brahmi

Post by 2+3 clusivity »

I am liking your working on both scripts. Keep it up.
linguoboy wrote:So that's what it looks like when the master satirist is moistened by his own moutarde.

User avatar
clawgrip
Smeric
Smeric
Posts: 1723
Joined: Wed Feb 29, 2012 8:21 am
Location: Tokyo

Re: Turkestan Brahmi

Post by clawgrip »

Thanks. Here are a few more conjuncts that make me wonder whether I will ever successfully be able to make this font:

Image
kntsu, ṅkli, tswsaṃ, ñwmāṃ, küle
The last one is noteworthy because it is stacking two complete aksaras. Again, these are all actual conjuncts that can be found in extant Tocharian manuscripts. And I didn't have to search hard to find them, either. There may be worse ones lurking somewhere, though so far I have not seen one with more than four consonants.

I'd have to make thousands of GSUBs to account for all these ridiculous things. I think the positioning is probably easier somehow, though I don't even know how to do it yet.

User avatar
2+3 clusivity
Avisaru
Avisaru
Posts: 454
Joined: Fri Mar 16, 2012 5:34 pm

Re: Turkestan Brahmi

Post by 2+3 clusivity »

What lines of text are you pulling these conjuncts from? I am guessing Sanskrit loan words. Without seeing the words, I wonder if the scribes are just writing shorthand by dropping out the inherent vowel and mashing words--or parts thereof--into conjuncts.

Could you show a transliteration of some of the lines with these clusters?

If they are too numerous, you may wish to limit your goal to the most common conjuncts. Consider that not every workable Devanagari script needs the following <ddhrya> conjunct:

Image
linguoboy wrote:So that's what it looks like when the master satirist is moistened by his own moutarde.

User avatar
clawgrip
Smeric
Smeric
Posts: 1723
Joined: Wed Feb 29, 2012 8:21 am
Location: Tokyo

Re: Turkestan Brahmi

Post by clawgrip »

Sure, I will make one up for you. But I was about to post about Tocharian's weird virama, so I will post it first.

Instead of placing a mark above or below or after to indicate a loss of the inherent vowel, as most Indic scripts with a virama do, Tocharian usually attaches the consonant to the previous one by means of a horizontal line. This just makes it that much more confusing for encoding and it kind of makes me want to give up. Here are some examples:

Image

twṢaL, ntwä, wñunT, praṅK

Pretty sure this only happens with "foreign letters" though (the ones I underline/capitalize). I could probably make ligature substitutions for those.

User avatar
clawgrip
Smeric
Smeric
Posts: 1723
Joined: Wed Feb 29, 2012 8:21 am
Location: Tokyo

Re: Turkestan Brahmi

Post by clawgrip »

Here are a few examples:

tsraṣiśśi mā praski naṢ : TAmyo kāsu tsraṣṣune pÜkaṃ pruccamo ñi PAlskaṃ :

-Ṣ PAñ kurSArwā ārplāsyo rarkusāṃ tkanā KALK : tMAṢ rākṣtsāśśi dvipaṃ yeṢ tMAṢ yakṣāśśi

baladvipaṃ yeṢ TMAṢ śtwar-wäknā ārṣlāsyo rarkuñcäS iṣaNAS kcäK śtwar-wäknā spe_

rñe wawiK ślaK śkaṃ || ṣāmnernaṃ || māski KAtkāLAṃ KTAṅkeñcÄ tsraṣiñ sāmuddRA : traidhātuK saṃ[s](ār) (tsra)-

-s wramnāśśi puK kāryaPAṃtwāśśi SAtkāluneyaṃ tsmār nāṃtsu knānmune nu ākntsuneyiS

(preṃ) ākntsuneyiS empelune kratswsaṃ kar wrasoM TApreṃ prāKAr tuṅk yāmtRA ṣokyo nu krañcäśśi kārme (we)-

Ṣ śtärT kuLMAṃtsyo wär camā eṢAK paṃPArS TMAK SAM poTAk pañwmāṃ śla śewiñlune lakeyäṢ kā(KA)-
2+3 clusivity wrote:If they are too numerous, you may wish to limit your goal to the most common conjuncts.
Since I have no particular goal or time limit and am only doing this out of personal interest, I guess I don't have to limit myself. I can just take forever to do it.

User avatar
clawgrip
Smeric
Smeric
Posts: 1723
Joined: Wed Feb 29, 2012 8:21 am
Location: Tokyo

Re: Turkestan Brahmi

Post by clawgrip »

Also:
2+3 clusivity wrote:Have you also done: http://upload.wikimedia.org/wikipedia/c ... t_4-1.jpeg as well?
can be found here

There are significant stylistic differences between the original manuscript and the style I chose for my font. Also, there was a scribal error in the original manuscript near the end of line 4 that I have fixed in mine.

Richard W
Avisaru
Avisaru
Posts: 363
Joined: Sat Oct 16, 2010 8:28 pm

Re: Turkestan Brahmi

Post by Richard W »

clawgrip wrote:Instead of placing a mark above or below or after to indicate a loss of the inherent vowel, as most Indic scripts with a virama do, Tocharian usually attaches the consonant to the previous one by means of a horizontal line. This just makes it that much more confusing for encoding and it kind of makes me want to give up.
Why is it confusing for encoding? You don't expect characters to proceed in an orderly fashion, do you?

I think the font should handle it matching up a trailing virama with the preceding base consonant, and ligating them. It's what the skipping flags in the GSUB lookups are for. I recommend doing this ligation before any ligation of subscript consonants.

User avatar
2+3 clusivity
Avisaru
Avisaru
Posts: 454
Joined: Fri Mar 16, 2012 5:34 pm

Re: Turkestan Brahmi

Post by 2+3 clusivity »

Any progress?
linguoboy wrote:So that's what it looks like when the master satirist is moistened by his own moutarde.

User avatar
clawgrip
Smeric
Smeric
Posts: 1723
Joined: Wed Feb 29, 2012 8:21 am
Location: Tokyo

Re: Turkestan Brahmi

Post by clawgrip »

I had a really tight schedule in October and was able to put less and less time into this. I stopped for a little bit, but I would like to pick it up again. It's nice to see that someone is still interested.

cibeckwith
Niš
Niš
Posts: 1
Joined: Wed Jan 15, 2014 3:50 pm

Re: Turkestan Brahmi

Post by cibeckwith »

Hi. I am very much interested in your work on Tokharian Brahmi script. I taught Tokharian last semester here (Indiana University, Bloomington) and discovered that there was still no font available for it. In reading through the postings on this topic it appears that you might be unaware of a Tokharianist work that contains pages and pages of all possible individual forms, syllables, ligatures, etc. (This stuff is in one chapter of Melanie MALZAHN, ed., Instrumenta Tocharica. Heidelberg: Winter, 2007. (ISBN: 9783825352998 ~ 3825352994).) I discovered this book only partway through the semester, when the students had already acquired the basic writing system, and got it via inter-library loan. I think the script is actually simpler than it first seems. (We only encountered a very few examples of Sanskritic letters or opaque combinations; in general they could be acquired slowly after the letters actually used for ordinary Tokharian. Having discovered this, I will modify my approach the next time I teach it.) Some of the oddities mentioned in the postings seem to be rare, or possibly limited to Tokharian B (we mostly did Tokharian A). It seemed to me when I read through the postings (back in November; I've been busy or traveling) that you had gotten it right, at least as far as the very nice samples you posted are concerned. So, I want to encourage you to continue and produce a font I can use the next time I teach the language, and of course one that others interested in it can use too. I also would like to ask if anyone knows of a complete "Asokan" Brahmi font. I bought the nice-looking one produced by XenoType, and have used it, but it does not include ligatures (with one exception), which do occur in some of the inscriptions. Finally, I mentioned you to one of my students in the Tokharian class; he is a linguist who does computational linguistics and started working on a Tokharian font too. He might join the list soon and contact you.

With best wishes,
Chris Beckwith

User avatar
clawgrip
Smeric
Smeric
Posts: 1723
Joined: Wed Feb 29, 2012 8:21 am
Location: Tokyo

Re: Turkestan Brahmi

Post by clawgrip »

Thanks for your comments. I do not know the work you've mentioned, so I have just had to decipher it using images of the manuscripts and transcriptions of the language and observing how the glyphs are put together. I did this for Khotanese as well, but this was much more of a headache because Khotanese is less well-known (there is next to nothing on the internet explaining how it works), and the documents and transliterations I have are hosted on different sites (Dunhuang for images, TITUS for transliterations), are categorized in entirely different ways, and many of the Dunhuang images have no transliterations on TITUS.

I agree that the script is regular and intuitive for handwriting, but it requires a lot of little adjustments to make a clean-looking font, which is my intention. I want the font to be usable for any sort of Tocharian people may plan to write, though I am not aware of the difference between A and B.

I have put this on hold for now because I think I need to upgrade my operating system, font software, and knowledge of OpenType before I can really make anything really good.

Post Reply