A guide to writing systems

[Richard W]

Of course, these definitions are not the only possible ones. I know that many people consider the essential component of abugida-hood to be the writing of C+V in a single ‘syllable block’ (as I’ve been calling it). By this standard, scripts like Lao would be abugidas, while Meroitic and 'Phags-pa would be alphabetic. And, indeed, by this standard, ‘abugidas approaching syllabaries’ would be a somewhat redundant phrase.

This is where I would insist that the Lao script with the modern Lao spelling is an alphasyllabary, rather than an abugida. And, as the combinations of consonant and vowel become opaque, an alphasyllabary moves towards being a syllabary.

[Richard W]

Now these two points are interesting… they remind me strongly of how consonant-doubling gets used with the Latin script (especially in Germanic). What vowel quality difference does consonant-doubling achieve in Thai?

The consonant doubling made the vowel of the closed syllable into /a/. Without the doubling, it would be /o/. Nowadays doubling is restricted to <r>, which it never used to be applied to, to memorialise a now-silent <r> from Sanskrit. So, Sanskrit dharma and Pali dhamma, which in abugidic Thai script are ธรฺม and ธมฺม, became in Thai ธมม and then ธรรม (both /tʰam/). The Thai phonetic spelling of the word is ธัม.

[Richard W]

Another innovation from that area is the use of matres lectionis in an abugida, not an abjad. The subscript letters for Indic /a/, /v/ and /y/ are used for representing vowels (Khmer and Tai languages) … The Bengali writing system has also developed a mater lectionis.

I don’t find this particularly obvious when looking at the scripts. Do you have a source for this history?

The use of subscript <v> and <y> for vowels like [ua] and [ia] goes back to Angkorian Khmer (source: Jenner, 2010, p7, in Jenner & Sidwell). This original situation is best preserved in closed syllables in the Lanna script, which borrowed from the Fakkham member of the Thai script family when it was adapted to write the Tai vernacular in addition to Pali. This can be seen in the right hand column under the heading 'diphthongs' in the table at https://en.wikipedia.org/wiki/Tai_Tham_ ... ent_vowels. (I for one am having trouble getting a privately installed font picked on my PC on Windows 10, and had to copy and paste to Notepad to check its contents.) In orthographically open syllables, a vowel is added in the Tai languages, and this has become mandatory for /ia/ in both the Thai (which has added a second vowel) and Khmer languages. Lao keeps the subscript form for /ia/ in closed syllables, as is not explained by the Wikipedia article. The use of original <ind. a> in the vowel /ɯa/ in the various Tai writing systems seems to be analogous.

The use of subscript <ind. a> for the vowel [ɔ:] can be found in both Lanna and in Southern Thai in the Khmer script (source: Antelme, note 18, on p29). Martin Hosken wouldn't believe it was the subscript letter in the Lanna script, and so it was encoded as yet another dependent vowel. Independently encoded texts have revealed the subscript letter being used for the vowel symbol. Other modern Thai scripts (other than New Tai Lue), which lack subscripting as an active process, just write the vowel using <ind. a>. (New Tai Lue has raised the subscript vowel to the baseline, so it contrasts with <ind. a>).

I've done some more research on Bengali ya-phalaa, and it's not quite as much of a mater lectionis as I thought. It's the combination of subscript <y> plus <dep. aa> that has its own interpretation as a vowel symbol. Source: https://www.unicode.org/versions/Unicod ... df#G664195.

[bradrn]

Of course, these definitions are not the only possible ones. I know that many people consider the essential component of abugida-hood to be the writing of C+V in a single ‘syllable block’ (as I’ve been calling it). By this standard, scripts like Lao would be abugidas, while Meroitic and 'Phags-pa would be alphabetic. And, indeed, by this standard, ‘abugidas approaching syllabaries’ would be a somewhat redundant phrase.

This is where I would insist that the Lao script with the modern Lao spelling is an alphasyllabary, rather than an abugida. And, as the combinations of consonant and vowel become opaque, an alphasyllabary moves towards being a syllabary.

Hmm, so you consider ‘alphasyllabary’ and ‘abugida’ to mean different things? How so? (I tend to agree with masako that most people use them as synonyms.)

Another innovation from that area is the use of matres lectionis in an abugida, not an abjad. The subscript letters for Indic /a/, /v/ and /y/ are used for representing vowels (Khmer and Tai languages) … The Bengali writing system has also developed a mater lectionis.

I don’t find this particularly obvious when looking at the scripts. Do you have a source for this history?

The use of subscript <v> and <y> for vowels like [ua] and [ia] goes back to Angkorian Khmer (source: Jenner, 2010, p7, in Jenner & Sidwell).

p4, surely? But thanks for all the info!

[Richard W]

Hmm, so you consider ‘alphasyllabary’ and ‘abugida’ to mean different things? How so? (I tend to agree with masako that most people use them as synonyms.)

Bright's definition of an alphasyllabary and Daniels' definition of an abugida are quite different. That's why Meroitic is not an alphasyllabary and the Lao script for modern Lao is not an abugida.

The use of subscript <v> and <y> for vowels like [ua] and [ia] goes back to Angkorian Khmer (source: Jenner, 2010, p7, in Jenner & Sidwell).

p4, surely? But thanks for all the info!

Both pages give the information, but it's easier to extract from p4.

[bradrn]

Hmm, so you consider ‘alphasyllabary’ and ‘abugida’ to mean different things? How so? (I tend to agree with masako that most people use them as synonyms.)

Bright's definition of an alphasyllabary and Daniels' definition of an abugida are quite different. That's why Meroitic is not an alphasyllabary and the Lao script for modern Lao is not an abugida.

Ah!

…you know, consulting WWS (bottom of p4), I find it immensely ironic that the definition of an ‘alphasyllabary’ gives precisely the same two criteria of mine which were so controversial earlier:

refers to the formal property of denoting vowels by marks that are not of the same status as consonants, and do not occur in a linear order corresponding to the temporal order of utterance

Compared to:

1. Graphemes are arranged in a single line in one direction
2. Consonants and vowels are graphically equal

So, clearly, my earlier idea of ‘linearity’ was on good authority! (Even if I’ve largely rejected it now.)

[Richard W]

• Khmer had 2 orthographic vowel notions, <a> (null glyph) and <ā>, to cover 2 degrees of aperture and 2 lengths. They prioritised aperture.

I don’t quite understand this… consulting Wikipedia, it looks like all Khmer vowel sounds are written separately?

If you look at the notes in the vowel table, you will be referenced to the 'Modification by diacritics'. There you will read that bantak is used to shorten the vowels represented by អ and អា. I'm not sure of the age of bantak - it seems not to have been used by Old Khmer. The association of Khmer vowel length with vowel symbols seems to have been conserved through the ages. Also, if you read Jenner, you will see that the vowel length for these two vowels was apparently not marked in Old Khmer.

[bradrn]

• Khmer had 2 orthographic vowel notions, <a> (null glyph) and <ā>, to cover 2 degrees of aperture and 2 lengths. They prioritised aperture.

I don’t quite understand this… consulting Wikipedia, it looks like all Khmer vowel sounds are written separately?

If you look at the notes in the vowel table, you will be referenced to the 'Modification by diacritics'. There you will read that bantak is used to shorten the vowels represented by អ and អា. I'm not sure of the age of bantak - it seems not to have been used by Old Khmer. The association of Khmer vowel length with vowel symbols seems to have been conserved through the ages. Also, if you read Jenner, you will see that the vowel length for these two vowels was apparently not marked in Old Khmer.

Hmm, I see. But since I asked that question, you linked Jenner, which helped:

The correspondence of the vowel symbols to their phonological actuality is poor. Ambiguity is noted in the case of written a, which may represent /ɔː, ɔ, aː, a/, written ā /aː, a/, written o /oː, ɔː, ɔ/, and above all written e and ē /eː, e, ɛː, ɛ, ɤː, ɤ/.

[Richard W]

Ferlus offers answers to most of Bradn's questions about Thai.

[bradrn]

Ferlus offers answers to most of Bradn's questions about Thai.

…in French, a language I don’t speak. (I do have plans to learn it soon, but that only helps so much.)

[Richard W]

Ferlus offers answers to most of Bradn's questions about Thai.

…in French, a language I don’t speak. (I do have plans to learn it soon, but that only helps so much.)

I must say the French looks very English; it's not like a newspaper. Also, there are a lot of pictures. Still, it may be a lot of work to push it through Google Translate bit by bit (the text is machine readable, but there seem to be a lot of errors in it after cut and paste, which Google Translate takes in its stride. Weird!). The URL is more accessible than I thought - it's https://shs.hal.science/halshs-00922719 ... Coedes.pdf . The only subtlety is that thaï with a diaeresis means 'Siamese', while without it it means 'Tai', and he doesn't inflect the latter word, which could confuse Google. I can't get the translation of the PDF as such to work. May be the English at the start defeats Google Translate.

The biggest problem may be comprehending the transliterated Thai and Lao!

[bradrn]

Ferlus offers answers to most of Bradn's questions about Thai.

…in French, a language I don’t speak. (I do have plans to learn it soon, but that only helps so much.)

I must say the French looks very English; it's not like a newspaper.

Usually, I do find that I can comprehend French papers in linguistics. (Although weirdly, Spanish is easier.) But for some reason this one seems to be a bit more difficult to read — fewer technical terms, I suspect!

[Richard W]

Usually, I do find that I can comprehend French papers in linguistics. (Although weirdly, Spanish is easier.) But for some reason this one seems to be a bit more difficult to read — fewer technical terms, I suspect!

Fewer technical linguistic terms, perhaps, but the paper seems unusually rich in words with semantically-matching English loanwords. What might be more difficult is the frequent use of ne ... que 'only', and the frequent use of the pronoun en. I'm currently extracting the French text with a view to throwing Google translate at it a page or so at a time. The text really is riddled with scannos. I can't help wondering if some of them are deliberate jokes, such as 'r' frequently being misalphabeticised as 'l'. One interesting feature of the content is that Ferlus sees the primary implicit vowel of Tai scripts of Khmer origin as being [ɔ], which to me is very dubious as Proto-SW Tai. He may mean what I think of as [ɔː] - I've always had trouble with Ferlus's 'Proto-Tai', which may exclude Northern Tai.

Another trick may be to skip the introduction (on pages 1 to 2).

[bradrn]

The text really is riddled with scannos. I can't help wondering if some of them are deliberate jokes, such as 'r' frequently being misalphabeticised as 'l'.

Hah, funny… except that I checked a bunch of random words, and ⟨r⟩ always seems to be transcribed as ⟨r⟩ (except for one word where it became ⟨i⟩).

One interesting feature of the content is that Ferlus sees the primary implicit vowel of Tai scripts of Khmer origin as being [ɔ], which to me is very dubious as Proto-SW Tai.

Surely Proto-SW Tai is sufficiently old that it predates the scripts?

Another trick may be to skip the introduction (on pages 1 to 2).

Yep, good point — skipping a few pages, there are far more loanwords.

[Richard W]

Surely Proto-SW Tai is sufficiently old that it predates the scripts?

No. The Khmer were chiselling away in Khmer by 611, while Pittayporn dates the start of the expansion of SW Tai to the 8th to 10th centuries AD.

[Richard W]

Surely Proto-SW Tai is sufficiently old that it predates the scripts?

No. The Khmer were chiselling away in Khmer by 611, while Pittayaporn dates the start of the expansion of SW Tai to the 8th to 10th centuries AD.

[bradrn]

Surely Proto-SW Tai is sufficiently old that it predates the scripts?

No. The Khmer were chiselling away in Khmer by 611, while Pittayaporn dates the start of the expansion of SW Tai to the 8th to 10th centuries AD.

Wow… interesting, thanks!

(Also, I think you quoted your post when you meant to edit.)

[Richard W]

The text really is riddled with scannos. I can't help wondering if some of them are deliberate jokes, such as 'r' frequently being misalphabeticised as 'l'.

Hah, funny… except that I checked a bunch of random words, and ⟨r⟩ always seems to be transcribed as ⟨r⟩ (except for one word where it became ⟨i⟩).

I was using a spell checker to detect scannos, so the statements are not incompatible. Anyway, I (with a lot of help from Google Translate) have knocked you up a translation to use with the original PDF. It was interesting that "tai noir" sometimes got translated as "black Thai" and sometimes as "tai noir". I did find a "tai noil" or some such. Anyway, I've corrected them all in the translation to "Tai Dam", which seems to be the commoner name in English nowadays. The translation has a horrible mix of fonts; I may tidy it up some day, and perhaps translate to HTML now that I've largely finished with the spell-checker.

[bradrn]

The text really is riddled with scannos. I can't help wondering if some of them are deliberate jokes, such as 'r' frequently being misalphabeticised as 'l'.

Hah, funny… except that I checked a bunch of random words, and ⟨r⟩ always seems to be transcribed as ⟨r⟩ (except for one word where it became ⟨i⟩).

I was using a spell checker to detect scannos, so the statements are not incompatible. Anyway, I (with a lot of help from Google Translate) have knocked you up a translation to use with the original PDF. It was interesting that "tai noir" sometimes got translated as "black Thai" and sometimes as "tai noir". I did find a "tai noil" or some such. Anyway, I've corrected them all in the translation to "Tai Dam", which seems to be the commoner name in English nowadays. The translation has a horrible mix of fonts; I may tidy it up some day, and perhaps translate to HTML now that I've largely finished with the spell-checker.

Thank you very much for this work!