Category: Writing, fonts & typography

Globalizing the Korean alphabet

By Simon

A group of linguists in Korea are looking into giving people with no written form of their language ways to write using the Korean alphabet (hangŭl), according to this article.

A number of communities they visited in Indonesia were keen on using hangŭl to write their languages and plan to send representatives to Korean to learn the alphabet, who will then to teach it to their communities.

The Korean alphabet is currently used only to write Korean, so it will be interesting to see how well it will work for other languages.

Writing systems and manuscripts

By Simon

The other day I came across an interesting article about writing systems and manuscripts in which the author divides writing systems into a number of different types such as bilinear, trilinear and quadrilinear, and equates different characteristics to the cultures using each type.

Bilinear scripts occupy all the space between notional upper and lower limits, as do CAPITAL LETTERS in the Latin alphabet. Other examples include such ancient scripts as Etruscan, Roman, Egyptian and Babylonian, and modern scripts like Chinese, Japanese and Korean. The author suggests that the main aim of such scripts is “to confine and constrain the written word” in order to fix words and to “to preserve the magical power of the word and to control people and things”, and that societies orientated towards magic and mysticism tend to choose such scripts.

Trilinear scripts occupy the same space as bilinear ones, but the letters can be positioned with their tops snug either with the upper or lower limits. Letters such as p, b, d and q have these characteristics. The author suggests that such scripts are “practical and dynamic”, that they are indended “to record speech as-spoken” and were used by practical societies. The Sumerian and Akkadian cuneiform scripts are trilinear.

Quattrolinear scripts, which include lower case versions of the modern Latin and Greek alphabets, have letters than can occupy the whole of the space, extend only to the upper of lower limits, or sit in the middle of the space.

The article also explains how different sizes and formats of written documents were used for different purposes in the ancient world, and that the influence of such practices is still with us. In ancient Sumeria, for example, the largest clay tablets were used for writing legal documents issued by the government – the law. The size of such tablets was limited by the weight of the clay though and they tend to be 14-15 inches high by 8-9 inches wide – the largest size one person could lift. Over time this size became fixed.

Clay tablets didn’t need margins as they were baked hard to preserve them and their edges didn’t crumble and they were completely covered in writing. However when people started to write on parchment, which did need margins, they had to use larger pieces to preserve the standard dimensions of the writing area. Meanwhile the Egyptian were finding it difficult to make papyrus rolls big enough to conform to the standard measurements for legal documents. After Alexander the Great conquered Egypt, the Greeks came up with a simple solution to this problem – they rotated the direction of writing by ninety degrees, after which Greek legal documents were 9-10 inches wide by 14 inches high. The Romans adopted the same practice and made their legal documents wider.

Standard written form of Cornish

By Simon

At a meeting of the Cornish Language Partnership last month, a new standard written form (SWF) of Cornish was ratified after much discussion, according to a report on maga.

A specification of the SWF, which is also available on that site, states that:

“The SWF is not meant to replace other spelling systems, but rather to provide public bodies and the educational system with a universally acceptable, inclusive, and neutral orthography. As such, it incorporates features drawn from a number of different Cornish orthographies, including Unified Cornish, Kernewek Kemmyn, Modern Cornish, Unified Cornish Revised, Kernowak Standard, and Kernewek Dasunys.”

The SWF is designed so that speakers of all forms of Revived Cornish will be able to learn it quicly and easily, and so that even those without any formal instruction in it will find it fairly easy to read.

Here’s an example of Cornish in the SWF:

Lyther Wella Bodinar
Bloodh vy ew trei ugens ha pymp. Th ero’vy den bohojek an puskes. My rug dyski Kernowek y’n termyn my veu maw. My veu dhe mor gen sira vy ha pymp den moy y’n kok. My rug skant lowr klowes udn ger Sowsnek kowsys y’n kok rag seythen warbar’. Na rug evy byskath gweles lyver Kernowek. My [rug] dyski Kernowek o’ mos dhe mor gen tus koth. Nag eus moy ’vel pajar po pymp y’n drev nei ’ell klappya Kernowek lebmyn, pobel koth pajar ugens bloodh. Kernowek ew oll nakevys gen pobel younk.

The same text in Kernewek Kemmyn:

[Ow] bloedh vy [yw] tri ugens ha pymp. Yth ezov vy den boghozek an puskez. My a wrug dyski Kernewek [y’n] termyn [ha] my a veu maw. My a veu dhe’n mor gans [ow] sira vy ha pymp den moy y’n kog. My a wrug skantlowr klywez unn ger Sowsnek y’n kog rag seythun warbarth. Ny wruga vy bythkweyth gwelez lyver Kernewek. My a wrug dyski Kernewek ow moz dhe’n mor gans tuz koth. Nynz eus moy ez pezwar po pymp y’n trev ni a yll klappye Kernewek lemmyn, pobel goth pezwar ugens bloedh. Kernewek yw oll ankevyz gans pobel yowynk.

Translation
I’m sixty-five years old. I’m a poor fisherman. I learnt Cornish when I was a boy. I was at sea with my father and five more men in a fishing boat. I hardly heard a single word of English in the boat for a week at a time. I’ve never seen a Cornish book. I learnt Cornish going to sea with the old fellows. There’s no more than four or five in our village who can talk Cornish now, old folk eighty years old. Cornish had been entirely forgotten by the young.

Source: http://corpus.kernewek.cymru247.net/wb.txt

This is the text of a letter written by William Bodinar in 1776.

Origins of the Celts

By Simon

http://thetranslatorscafe.wordpress.com/2008/06/03/celts-are-from-spain-says-professor/”>An article I found recently questions the commonly-held belief that the original Celtic homeland was in central Europea around Hallstatt in what is now Switzerland.

In a theory based an extensive overview of the linguistic and archaeological evidence, Professor John Koch of the University of Wales Centre for Advanced Welsh and Celtic Studies proposes that a Celtic civilisation and culture first developed in the west of Europe in the Bronze Age rather than in central Europe.

Inscriptions found on the Iberian peninsula and dating from 800 BC to 400 BC are in a Celtic language, Professor Koch believes. The language is known as Tartessian and is believed to be the oldest written language in western Europe.

There is also an Irish legend in Do Suidigud Tellaich Temra (The Yellow Book of Lecan) about the origins of the Gaelic Celts – “We are born of the children of Mile, of Spain.”

The Nagari Alphabet

By Simon

Today we have a guest post by Marcis Gasuns.

The Nāgarī (lit. ‘of the city’) or Devanāgarī (‘divine Nagari’) alphabet descended from the Brahmi script sometime around the 11th century AD. It was originally developed to write Sanksrit but was later adapted to write many other languages. The origin and meaning of devanāgarī (also written as devnagari, devanagari, deonagri) remains dubious. It comes from the Sanskrit words deva (god, celestial; brahman), and nāgarī (city, possibly from tamizh, where it means “temple”). Together they probably mean, literally, (the most popular explanation) “script of the city”, “heavenly or sacred script of the city” or “[script of the] city of the Gods or priests” in Buddhist texts.

At the End – a farewell from Elizarenkova:

elizarenkova-avtograf.jpg

Creating fonts

By Simon

If you would like to turn your constructed alphabets into fonts, there are a number of ways to do so: you could buy one of the professional font creation tools available from Fontlab, you could use a free font editor such as FontForge or Softy, or use the font creation service Fontifier.

Today I found out about another font tool, FontStruct, a free online font editor which looks good and fairly easy to use. The site also has a gallery where you view fonts created by other people and add your own creations. When I can find a spare moment or two, I’ll have a go at converting some of my ideas for con-scripts into fonts.

Devanagari Ligatures in Sanskrit Fonts

By Simon

This is another guest post by Marcis Gasuns.

Namaskar,

It’s been a problem ever since the first font was created and never truly solved until this day. The typefaces used to print Monier-Williams Sanskrit dictionary or MacDonnell’s Grammars have not been beaten by the PC fonts – they are much weaker. The god of typography is in the small details. And without those nuances, fonts are worthless. Below is an example of Mangal Devanagari Unicode font, installed on every XP, Vista PC:

बहवो न विरोद्घव्या दुर्जयो हि महाजनः।

स्फुरन्तम् अपि नागेन्द्रं भक्षयन्ति पिपीलिकाः॥

No variants are available for vowels (in MSS there are variants for – two for “a”, three for “e” etc.); there are no udatta markers as in Boehtling’s dictionaries, which remain the best Sanskrit dictionaries to this day. Different fonts have different issues.

There is no perfect TrueType or OpenType font at the moment, no matters whether it’s free or commercial. So there are the Unicode Devanagari and non-Unicode Devanagari fonts, from which we have to choose – or one can scan old books and Xerox them. I choose to use imperfect fonts rather than to copy old books.

The Unicode Devanagari fonts don’t support several (Vedic) accent marks and they’re rather low on all kinds of typographic nuances (ligatures like “drsthva” are totally wrong). The best Unicode Devanagari font for Windows available today is Ulrich’s Sanskrit 2003. It does look like the more up-to-date Hindi fonts (which is no good, as we deal with Rigveda and other rarer ancient sacred texts), but it does have a few hundred ligatures. However, as with all the Unicode fonts, it can’t have variants of the same ligature, some of them have up to four known variants, as “la”, for example.

There is an alternative font, called Chandas, containing 4347 glyphs: 325 half-forms, 960 half-forms context-variations, 2743 ligature-signs (which should be enough, to print even Panini’s grammar), but the author, Mihail Bayaryn from Minsk, has abandoned the project and there are still some additional Devanagari marks from the MSS missing there. I have to admit that Ulrich’s font, on which he has worked more than year, looks better when printed, but contains fewer ligatures (so you can’t print Ashthadyayi with any of the Ulrich’s fonts).

So there remain the non-Unicode fonts (some of them even with Mac support). Ulrich is also the author of Sanskrit 99, which has a very interesting counterpart, entitled Ancient Sanskrit 98 with Bombay-Mumbay characters like Ancchar instead of the more common Delhi-Varanasi styled fonts. From a typographical point of view (not web), the best looking fonts are the “French” (called Gudakesha, as found in Bopp, 1816) and “German” (Shantipur, as found in Harvard Oriental Series vol. 14, 1914) fonts that we have created together with Tikhomirov, which have been made public at Ulrich’s website without my allowance, breaking our copyrights (the legal copy can be downloaded here). For 5 years I’ve been looking for each of more than 500 more common ligatures in the Sanskrit books published in Paris, Bonn, Oxford, to find every sign.

In late 2005 the fonts were still not finished, in 2006 they were hinted by Mihail Bayaryn (a private OpenType version was made) and the fonts have never been finished. They are replicas of old book fonts. You can see the history of making of both the fonts at Nagari Sanskrit Group and the second Devanagari font as well, also a specimen.

To put it in a nutshell. If you need an easy-to-type font – Sanskrit 2003 is the one to choose (forget about Mangal installed on Windows XP by default, it’s for Hindi, not for Sanskrit), but if you want to have the Indian flavour – use the unfinished fonts by Marcis Gasuns & Tikhomirov hinted by Mihail Bayaryn, having some of the missing characters from Ancient Sanskrit 98. If you have any questions, feel free to ask.

Devanagari Fonts

By Simon

Namaste,

Simon agreed that I’ll occasionally post reviews about Asia, India and devanagari, typography and some font & linguistic software related matters as well.

I’m a young man, born in Riga and spreading the Sanskrit message around the world. Working as a Sanskrit Reader in Russia, teaching at the Russian State University and writing a PhD about Sanskrit verbal roots. I’m planning to publish several Sanskrit manuals and reprints of old books in the near future. I’m an editor as well of the Sanskrit section of the Open Directory and http://nagari.southindia.ru. So you can download some dictionaries there, etc.

It is strange that 200 years after the first Nagari typefaces where cast in iron in India, we’ve got no fine Devanagari fonts at our disposal. Ok, we have quite a few Hindi fonts. But, hey, there are many differences. No Devanagari fonts at the moment supports the four variants of “la” or the northern (Varanasi) and southern (Mumbai) variant of the letter “a”. Ok, some may say that who cares about Nagari font, but you do know and notice the difference, if you’re a teacher – students get stuck seeing a letter they’ve never seen before and have no reference chart to look upon.

None of the True Type Unicode fonts have a precoded ligature for “sthva” (which means the “va” should be under the “sth” and not beside it) as it was in the good old times when Harvard Oriental Series was printed (even they have lost the type in the latest editions, e.g. the 50th edition was printed in transliteration only). All the letters in Windows fonts are written without even the slightest break, though it is well known that in the manuscripts and books printed before 1914 in Europe, Devanagari letters are separated buy 1-1.5 mm. Maybe it is not very good from the point of view of grammarians, but, sure, it looks much better. Devanagari font differences like these can be continued.
That’s it for today, next post will be in a while. Is there anybody who’s interested to hear about Unicode Devanagari font matters?

Word of the day – transliteration

By Simon

Transliteration, from the Latin trans, across, and littera, letter, is the practice of transferring a text from one writing system to another. Transliteration can be used to give people who can’t read other alphabets an idea how to pronounce words and names, though without some knowledge of the idiosyncrasies of pronunciation of the original languages, their pronunciation will probably sound only vaguely like the original.

In phrase books you often find transliterations designed to show how to pronounce things based on the mother tongues of the readers. Such guides tend to be confusing as the ‘phonetics’ might not make much sense to you. A few phrase books and language course provide IPA pronunciation guides, which are great, if you’re familiar with the IPA.

In the phrases section on Omniglot, I try to provide transliterations for all languages written with non-Latin writing systems. Many languages have several different transliteration systems and I usually choose the most widely-used. The transliterations may not make sense to everyone, but there are recordings of quite a few of the phrases, so you can at least hear how to pronounce them. Ideally there would be sound files for all the phrases, plus maybe IPA transcriptions, and with your help, maybe that will be the case one day.

I chose this word today because someone suggested that some of the transliterations on my phrases pages are confusing, especially to Americans. The trouble is, if I made the transliterations American-friendly, people from other countries might not find them useful.