Talk:Canadian Aboriginal syllabics

From Wikipedia, the free encyclopedia

Chinese character "Book" This article falls within the scope of WikiProject Writing systems, a WikiProject interested in improving the encyclopaedic coverage and content of articles relating to writing systems on Wikipedia. If you would like to help out, you are welcome to drop by the project page and/or leave a query at the project’s talk page.
B This article has been rated as B-Class on the Project’s quality scale.
High This article has been rated as high-importance on the Project’s importance scale.
This article is part of WikiProject Indigenous peoples of North America, which collaborates on Native American, First Nations, Inuit, Métis and related subjects on Wikipedia. If you would like to participate, you can edit the article attached to this page, or visit the project page, where you can join the project and/or contribute to the discussion.
B This article has been rated as B-Class on the assessment scale.
Suggested article edit guidelines:

This article has no comments yet. If appropriate, please review the article and leave comments here to identify the strengths and weaknesses of the article and what work it needs.

Contents

[edit] Title

This page replaces Unified Canadian Aboriginal Syllabics, which is not what anyone except the Unicode committee calls syllabics writing. Diderot 10:12, 15 Nov 2004 (UTC)

I dislike this redirect. The Unicode Standard and ISO/IEC 10646 call it Unified Canadian Aboriginal Syllabics because that's what the Canadian government's CASEC committee called it when they engaged in the unification of the different varieties of "Canadian Syllabics". What you've written "syllabics writing" isn't what people say. People just call them Syllabics, generally. You've kept "aboriginal" which is governmentalese at its best. I think this should be reverted. Evertype 22:24, 2004 Nov 28 (UTC)
I disagree. UCAS is a code block in the Unicode standard. It is not a writing system. It would be as inappropriate as replacing Roman alphabet with ISO Latin 1. "Canadian Aboriginal Syllabics" would be acceptable, but the "unified" part is purely a matter of Unicode blocks. It is not unified in anyone's practice except standardisation bodies. Users think of it as various kinds of Cree syllabics, Ojibwe syllabics, Dene syllabics and Inuktitut syllabics. Diderot 06:19, 29 Nov 2004 (UTC)

[edit] The alphabets

They all become squares on my computer. Is that normal?

Yes. Download and install the fonts here and the problem will go away. Diderot 23:02, 22 Jan 2005 (UTC)

[edit] Inuktun and some Inuit

I reverted back to my earlier version from the reversion by Kwamikagami, and I want to explain why. Inuktun is a not the name of a language that is distinct from Inuktitut. I'm not sure quite what it's supposed to refer to - it sounds like a the kind of word you might use in a western dialect to describe Inuktitut (Inuktun = like an Inuk). And, not all Inuit use syllabics. The Alaskans, Greenlanders, and Inuvialuit, Inuinnaqtun and Labridorimiut in Canada have never used syllabics. A minority of Inuit globally and many Inuit in Canada do not and have never used syllabics to write their language. --Diderot 16:42, 24 August 2005 (UTC)

Maybe, but for that matter Inuktitut isn't a language either. Inuit is a dialect chain, going by various local conventional names: Iñupiaq in Alaska, Inuktun in the NW Territories and western Nunavut, Inuktitut in eastern Nunavut, Kalaallisut in Greenland. The point was that it is used both in "Inuktitut" and "Inuktun", though the latter only in Nunavut. However, if you would prefer the name Inuvialuktun over Inuktun, that's fine. John D Nichols, writing in Daniels & Bright, had this to say: "the current form [of syllabics], used for most dialects of Eastern Canadian Inuktitut and for the Natsilingmiut dialect of Western Canadian Inuktun, derives from the 1865 reforms ..." He gives as his reference Kenn Harper, 1985, "The Early Development of Inuktitut Syllabic Orthography", Études/Inuit/Studies 9:141-62.
As for the second point, this is true for all nations that use syllabics. For instance, although it is used to write all Cree dialects within a certain geographic range, it is not used for all communities speaking those dialects. The Inuit should not be singled out for the "some"/"many" comment. kwami 19:00, 2005 August 24 (UTC)
Yes, Inuktitut can be viewed as a number of related languages, but there is no simple criteria for making the distinction between a group of languages and a single one. Politics is the usual way - but our article on Inuktitut covers the entire dialect continuum - which is the usual way of doing it, both among the Inuit and in the technical literature. The Inuit do not always use consistent names for their language, so the notion of an Inuktun separate from Inuktitut is a bit strange. Inuvialuktun is a purely political construct, a point made in the corresponding article, which even the Inuvialuit are hesitant to use.
I made the syntax of where syllabics are used in Inuktitut a bit more precise and more in line with the claim for Cree, since it is an eastern vs. western arctic issue.
In as much as one wishes to consider the western dialects separate from Inuktitut, I don't think there was ever a real tradition of syllabics in the west. The Inuinnait (Copper Inuit) were the last Inuit reached by missionaries - they were not in full contact with European society until the 1920s. By then, a few syllabics readers had learned to write that way from Inuit further east, but missionaries already believed quite strongly in the exclusive use of Roman letters and promptly began using exclusively Latin alphabet schemes. Still further west, the Mackenzie delta Inuit were likely never exposed to syllabics, and the Alaskans would likely have never heard of syllabics. Certainly syllabics aren't used in the Inuinnait areas today except where the new Nunavut government has imposed them, and they are totally absent from the NWT which uses only Roman alphabet writing officially and in schools, to the total exclusion of other writing schemes.
The "eastern Inuktitut" vs. "western Inuktun" dichotomy is easily overestimated. I suppose you could define the line as the s/h isogloss, which puts Natsillingmiutut on the other side, but that's more than a bit arbitrary and I don't think anyone does so anymore.
--Diderot 17:44, 25 August 2005 (UTC)

[edit] Sekani

I've eliminated the references to Sekani as there is no evidence that the Sekani ever wrote their own language in syllabics. Some Sekani people did know syllabics, but they used them to read and write Carrier. Bill 08:42, 28 December 2005 (UTC)

Just eliminated some more references to Sekani which I must have overlooked. There is only one syllabic text known in a Sekani context, a gravestone at McLeod Lake. I have read it, and it is in Carrier, not Sekani. Bill 19:23, 17 March 2006 (UTC)

[edit] Carrier Transcription

I've changed the transcription of the vowel-only series for Carrier. The previous transcription was not IPA - it was evidently based on a misunderstanding of the Carrier Linguistic Committee writing system, which is the Roman system introduced by the SIL that is now the most widely used writing system for Carrier. In that writing system <u> represents /ʌ/ and <oo> represents /u/. Someone evidently thought that <oo> stood for long /o/. I have substituted the correct IPA symbols. Bill 19:23, 17 March 2006 (UTC)

[edit] Syllabics : an Abugida?

Am I the only one who would dispute that the various Canadian Syllabics are not abugidas? In Cree syllabics for example, there's no definite order for which the characters are placed. In addition to this, the said rotation/reflection done to the characters is not consistent throughout the entire writing system(s). - In a regular abugida system, such as the Khmer script, every consonant has an inherent vowel (usually /a/) which is modified by the use of diacritics (or by 'modifying a base character'). In Syllabics however, the use of diacritics is rare, and since there is no one script order, it's impossible to define an inherent vowel or a base character from which all others are derived.

A lot of this is better explained on the Language Geek website, where it also cites that "Peter T. Daniels, [the one] who invented the term abugida, calls Cree a ‘sophisticated grammatogeny’, certainly not an abugida." - Io Katai (talk) 05:13, 7 March 2008 (UTC)

What would you classify it as, then? I assume, since you're reading Greek Wikipedia, that you know that the term "grammatogeny" has nothing to do with the form of the script, whether an abugida or anything else.
I have no idea what you mean by "there's no definite order for which the characters are placed."
An abugida is a script where the letters stand for consonants, and you modify them for vowels. People take the Brahmi family to be representative, because it's so widespread (as in Khmer). However, the term abugida itself comes from Ethiopic, and that is also irregular: You can't predict how to represent the vowels. In Cree, you have letters for consonants, and modify them to indicate vowels, so it fits the basic definition of 'abugida'. Yes, it's a unique modification, and it's a little irregular, mostly because of somewhat haphazard evolution, but it's not a true alphabet, not an abjad, and not a syllabary. I don't know that anyone has proposed a new term just for Syllabics. — kwami (talk) 07:23, 7 March 2008 (UTC)
If you read the website I linked, you would know that I'm citing Peter T. Daniels, the first person who actually used the term "abugida" to refer to this type of script. (It's even noted in the Abugida article). I'm not saying that his terminology is any better, but he certainly distinguished Cree syllabics from all other abugidas. And although I used Khmer as an example, any other abugida still follows the general pattern of having a base character + modifications (Ge'ez, Kharoṣṭhī, among others).
I also never said it was an abjad, syllabary, nor alphabet; but it's neither an abugida. - The main reason for this, as I said, there has never been a standard on how to order the characters, which means that it's impossible to say that there is a base character which is modified to change its vowel (even though they bare ressemblance). In Cree for example, nearly every community has a differently established order of characters[1]; whereas in most other abugidas, the base character which is modified is more than clear. The only actual diacritic used in Syllabics is the 'dot', which generally indicates a long vowel, similar to how the Hepburn romanization uses a macron. As for the syllable-final consonants (ᒃ,ᑦ,ᔥ,ᒻ...), most variants don't agree on the same characters, and some don't bear a ressemblance to their former (ᑕ,ᐊ...).
So how do you argue that the Canadian Aboriginal Syllabics are an abugida system, when they're far closer to a syllabary with alphabetical elements? From the abugida article : "An abugida is a segmental writing system in which each letter (basic character) represents a consonant accompanied by a specific vowel; other vowels are indicated by modification of the consonant sign, either by means of diacritics or through a change in the form of the consonant. - Io Katai (talk) 21:11, 7 March 2008 (UTC)
I'm not following your wording on the final consonants, but in any case they're not relevant here.
I'd have to read Daniels again to see how closely that site represents him, but it does say,
Abugida is a term coined by Peter T. Daniels for a script whose basic signs denote consonants with an inherent vowel and where consistent modifications of the basic sign indicate other following vowels than the inherent one.
The only way Cree does not fit this description (arguably) is that it's arbitrary which inflected form you pick to be the base; that is, which is the inherent vowel. That's certainly a significant difference from Brahmi, but Cree is nothing like a syllabary in his scheme, which is what you're claiming. It sounds to me like Daniels was just trying to word things in a way that was intelligible.
'Consistent' doesn't mean 100%. If it did, many of the Brahmic scripts could not be considered abugidas.
Daniels divided phonographic scripts into four groups: syllabaries, which have no letters than can be identified as consonants; abjads, which only have letters for consonants; abugidas, which have letters for consonants that are modified to indicate vowels; and alphabets, which have distinct letters for consonants and vowels. No script is going to fit perfectly, but Cree could only be an abugida in this classification.
If you take things too literally, hardly any scripts are going to match Daniels' definitions. For instance, an alphabet is supposed to be a phoneme-based system. Therefore, the English script I am using now is not an alphabet, because it requires digraphs. Shall we create a new category, 'digraphary', and say that English is 'a digraphary with alphabetic elements'? I don't see how English diverges from the alphabet prototype any more than Cree diverges from the abugida prototype.
Anyway, since you can choose any vowel to be the inherent one, you could argue that the choice is irrelevant. Your argument is like saying the Roman script isn't an alphabet because different communities make different choices as to which letters are consonants and which are vowels: i, r, u, w, y, and z are all consonants in some Roman scripts, and vowels in others. Similarly, b, s, m, g may be consonants or tones, while j and v may be consonants, vowels, or tones. This doesn't mean the Roman script is not an alphabet. Also, having charts with different orders, as in your link, doesn't mean that different Cree communities have different alphabetic orders in their dictionaries. However, even if there were no dictionary standard for which vowel was basic within a single community, that's no more relevant than the days when there was no dictionary standard for whether i and v were consonants or vowels in English. — kwami (talk) 04:16, 8 March 2008 (UTC)
I'm not arguing alphabets though, I'm arguing that the Canadian Aboriginal Syllabics are not an abugida script. Like I said, if you can't prove what makes them an abugida, then why should they be classified as so? If you don't actually read the website that I keep citing (which explains in more details what I've been trying to convey), then your arguments are kind of flawed. Arguing that it doesn't have to be 100% could also mean that you could classify CAS as a syllabary. In fact, Omniglot classifies CAS under syllabary. And if you look down, there are also scripts such as the Kpelle syllabary which also have alphabetical elements. The Iberian scripts are also classified as a syllabary, although the Wikipedia article classifies them as semi-syllabaries because they have alphabetic elements.
I'll even go so far as to cite Unicode, which says "In other featural syllabaries, such as the Canadian Aboriginal Syllabics, the relationship between the sound and graphics parts is less systematic."[2] - There are also books that back the term "syllabary"[3] rather than "abugida"[4]. I also found it hard to prove that CAS were an abugida because most search results return quoted wikipedia information. - Io Katai (talk) 05:04, 8 March 2008 (UTC)
Maybe I'm just not familiar enough with it, but I fail to see at a quick glance how Kpelle is alphabetic. pi and li are identical, as are po and su, but glyphs with common consonants or vowels don't seem particularly similar. How is that alphabetic?
I have read that site, as should be obvious from my last comments. In the opinion of the author, Cree deviates enough from Daniels' definition of abugida to argue that it should not be considered one. I'm giving you the counter argument. My analogies with alphabets are just that: analogies to show you where your logic leads us. The difference hinges on one thing: Must the inherent vowel be objectively obvious to an outsider, as in Hindi, or can it be a matter of convention, as in Cree? You're arguing that it cannot be the latter, but you have little to back you up. Granted, it's hard to find outside support either way, because abugida is not a term in widespread use.
The Unicode site is ... bizarre. It claims that hangul is a syllabary. Hangul! Can there be a more perfect example of an alphabet than hangul? It's more alphabetic than Greek! I think that shows where they're coming from: A "syllabary" is a script where each syllable has a separate Unicode designation. It has nothing to do with the script itself; it depends on how the Unicode Consortium decided to encode it! Because of their differing structures, different encoding strategies work better for Hindi and Cree. That's not what Daniels had in mind! Their definition of 'abugida' is specifically a "Brahmic family script"; they even say that Ethiopic is more-or-less a syllabary; perhaps they're worried they'd look foolish if they definitively claimed it was one.
You gave me this definition of an abugida: a script whose basic signs denote consonants with an inherent vowel and where consistent modifications of the basic sign indicate other following vowels than the inherent one. Cree fits your definition. — kwami (talk) 07:54, 8 March 2008 (UTC)
As for the Hangul example, Unicode analyzes the whole segment as a syllable 한 [han] (and they go so far as to explain their definition of 'featural syllabary'[5]), and each individual jamos (elements that compose the whole syllable/character block[6] - 한) as alphabetical ㅎㅏㄴ. And the definition of abugida is mentioned in the second paragraph, not the first (It says that most SouthEastern scripts historically derived from the Brahmic script are abugidas, not all Brahmic family scripts are the only existing abugidas). As for the Ethiopic example, it's explaining how in Unicode it's usually treated as a syllabary, because the encoding required differs from from standard abugidas in the SouthEast. Yet, they still classify it as an abugida, or else it would be under the (featural) syllabary section along with Cree et al. -
Asides, the issue about classifying CAS as a featural syllabary was raised on the Unicode Mail List, and discusses reasons for classifying scripts as they were (as well as why CAS are both 'featural', and unlike 'abugidas').[7][8][9][10]etc.
As for your comparaisons with an alphabet, it's been long known that they don't represent sound-to-sound accurately, even Unicode acknowledges that. Those which are though, are typically called 'true alphabets'. Abugidas can be the same, the most notorious one being the Khmer script because it has no spelling standards. And although I did use Daniels as an example, he isn't the only person to argue that CAS should be reclassified; as I did in linking with my sources. I would still appreciate on the other hand, reliable sources that point to all CAS being abugidas.
And the definition (the one which you quoted is not the one found in the abugida article), if you read clearly, defines that all consonants must have an inherent vowel. That's ok, it's the same as a syllabary. Then it goes on to say that the basic sign must be modified in order to change the vowel. - If you want to argue that CAS fit this definition, then what's the basic sign? Just because some symbols (not all) express a relationship (which differs from character to character), doesn't mean they can necessarily be bunched together under a term that only vaguely represents, when other terms would be more accurate. CAS were in essence based off of the Pitman Shorthand phonetic notation, so it shouldn't come as a surprise that they have featural elements ("the symbols do not represent letters, but rather sounds(...)for the most part, written as they are spoken"). Unlike this article though, the Cree syllabics, Carrier syllabics, and the Ojibwe writing systems articles also refer to their branches of syllabics as syllabaries. - Io Katai (talk) 16:47, 8 March 2008 (UTC)
Okay. ᐸ is the consonant /p/. It has the inherent vowel /a/. To change the vowel to /i/, you rotate it clockwise. To change the vowel to /e/ etc. etc. To cancel the inherent vowel, you superscript it.
That fits the definition of an abugida. Your objection is that it's arbitrary which form you take to be basic. It isn't arbitrary in those languages where the superscript form is derived from the basic form. However, even where it is arbitrary, I don't see how that's relevant. It still fits the definition, which never defines which form must be basic. You also object that which form is basic (so you admit there is a basic form!) varies from language/dialect to language. That again is irrelevant: Which Roman letters are consonants or vowels varies by language, century, and dictionary, but they're still alphabetical. That kind of variation is true for all kinds of scripts.
If you take Daniels' other definition, where abugidas use diacritics for vowels, then Cree would not be an abugida. However, Daniels is clearly inconsistent here. There are three types of consonantal modification in the world's scripts: diacritics (Brahmic family), modifying shapes (Ethiopic family), and rotation/reflection (Cree family). You've directed me to talk pages where people discuss the relative merits of calling each an abugida, but all fit Daniels' broader definition. If you want to say there are proper abugidas with diacritics for vowels, and others without diacritics, okay, but you could also argue that there are proper alphabets with a one-to-one glyph-to-phoneme structure, and others like English which are not proper alphabets.
I guess you could also argue that the difference depends on how children learn the script in school. The Ethiopic and probably the Cree families are learned as syllabaries. That seems to be the point of much of the Unicode mail discussion. However, I have never seen a formal definition where scripts are classified based on how they are taught. Rather, all definitions I've seen classify scripts based on their form. — kwami (talk) 19:59, 8 March 2008 (UTC)
Yes, you could say that ᐸ is /p/ with the inherent vowel /a/, but you could also say that it represents the whole syllable /pa/. The argument is the fact that no one script order defines the base character from which all others are derived. Most people who define the script as a syllabary will acknowledge that there is a pattern (why can't there be? It's based on a constructed featural script after all), but it is one unique to syllabics (rotation, reflexion: symetric, asymetric). If you completely define all CAS as abugidas, then it should be impossible for lone consonants to exist, as they too should have the inherent vowel /a/. Finals in Syllabics can function as regular consonants, whether they're placed in syllable coda, used in consonant clusters[11], or placed in initial position (like ᐦ in Ojibwe) [12].
You also state that finals are related to their counterparts, so that ᐸ is /pa/, thus ᐸ should be /p/. This is inconsistent throughout all CAS, the so-called 'base character' isn't always used, and in some cases the consonants bear zero relation to the consonant-vowel series, as in Maskwacis Plains Cree Syllabics. And unlike any other abugida, including Ge'ez, they don't normally deal with completely independant consonants (diacritic modifications at most). Whereas a few syllabaries deal with some (or one) independant vowels, such as Cherokee.
Anyhow, saying that CAS as a whole doesn't have to be a perfect abugida, could mean that it's also an imperfect syllabary, or alphabet. I would also like to see sources. Making claims is one thing, but if you can't follow them up with sources, then what good are they? Even in this article, the external links and references define Syllabics as a syllabary [13][14], perhaps a unique system or a syllabary with alphabetical elements [15], characters with alphabetical elements[16], and an alphabet[17]. - The reason for calling it an alphabet is because "since more complex syllables are written with more than one symbol, and since even simple syllables such as these, consisting of only a consonant and a vowel, are actually decomposed into their component consonant and vowel."[18] - Io Katai (talk) 03:35, 12 March 2008 (UTC)
I've already covered most of these points, so I won't bother to repeat myself.
"If you completely define all CAS as abugidas, then it should be impossible for lone consonants to exist": By that logic, if you completely define all CAS as syllabaries, then it should be impossible for lone consonants to exist.
Both syllabaries and abugidas do idiosyncratic things with final consonants, which in any case aren't used to classify scripts.
Yes, it's common to use various terms for syllabics, just as it's common to speak of the Hindi syllabary or the Arabic alphabet. However, all of the scripts on Wikipedia are classified by Daniels' terms. You may take issue with that, but that's a larger issue than just this one article, and needs to be taken up with the people working on articles such as writing system and abugida. — kwami (talk) 06:20, 12 March 2008 (UTC)
P.S. Several Brahmic scripts also mark final consonants with diacritics, like Cree, but they're not considered any less abugidas because of that.
Anyway, you're clearly persuaded by one website which takes issue with the classification of syllabics as an abugida. That's hardly reason to change, but can certainly be brought up as evidence on one of the more general pages. — kwami (talk) 06:42, 12 March 2008 (UTC)
Actually, it was the fact that the wiki article stated it as an abugida that made me use that original website. The LanguageGeek is after all, referenced throughout most native Canadian language articles. That aside, I used various different references to continue arguing the point. If there were other sources than Wikipedia, I might acknowledge CAS as a family of abugidas, but it's absurd to consider all of them abugidas when there are no sources to back that claim. And since Wikipedia prefers references over original research, I further provided resources to support my points.
And I do agree with your comment about syllabaries, but some do have alphabetical elements. Brahmic scripts can indicate final consonants or clusters through diacritics; but diacritics alone don't have any value, thus they are dependant. In syllabics, you can have things such as ᐯᐦ <peh>, ᐦᒑ <chaa>, ᐦᐊ <ha> (Lac Seul Ojibwe). This is comparable to how the letter <h> (among others) functions in the latin alphabet, where consonants are independant. In addition, loan words and loan names in Syllabics may use 'finals' as initials.
You also seem persuaded to put down Daniels' terms when he considers Cree a 'sophisticated grammatogeny' (a term he coined, and also used for the development of bopomofo); yet you say Wikipedia follows by his definitions. He himself wrote If one of the orientations were basic, then it could be considered an abugida., which is as I pointed out, not contradicting his definition of an abugida. Anyhow, I asked for one simple thing thrice, it's useless to continue arguing you at this point. - Io Katai (talk) 08:55, 12 March 2008 (UTC)
I'm not following what thing you're thinking of that you asked for thrice. AFAIK, I've responded to all the points you've brought up.
Again, 'sophisticated grammatogeny' isn't a kind of script, to be contrasted with abugida or anything else. It simply means that someone got creative in inventing it.
I wasn't familiar with that quote from Daniels. It does appear then that he considers it at best a marginal abugida. However, if memory serves, he certainly does not consider it a syllabary, featural or otherwise. Also, there are several other cases where he comments that a script is marginally this or that, since few scripts are consistently one thing or another. (The half alphabetic, half syllabic Iberian scripts are an extreme example of this.) It's certainly worth mentioning this in the article. — kwami (talk) 09:35, 12 March 2008 (UTC)