Thursday, November 03, 2005

To Point or Not to Point

In a recent comment Mark brought up the issue of tone marks in Pinyin,

"I tend to think that the best system for native speakers would be the least complicated, thus omitting most diacritics. ... I'd like to learn more about the use of pointed/unpointed text with the Hebrew and Arabic scripts. I wonder if you or any of your readers know of studies on the efficiency (for reading, but also for writing, if available) of using or omitting diacritics with these scripts. Or perhaps the partial omission of diacritics."

I can't say much about Arabic, since it is a bit too complicated to compare to tone marks. For Hebrew, most of the text that I have seen on the internet is unpointed except for some religious sites.

However, the topic of pointing or not pointing in Cree Syllabics is one that I have thought about for many years. Too many.

The original Cree Syllabary was designed by a missionary to mark vowel length and preaspiration of consonants with marks. If all of these features are marked all the time then the text is called 'fully pointed.' If they are omitted, then 'unpointed'. Sometimes the text is pointed occasionally to disambiguate, for certain words, for stylistic issues, etc. Then it has been called strategic pointing.

While I have a some old copies of the Wawatay News in a filebox somewhere, I have also found a copy of an article from this Oji-Cree newspaper on the internet. Although the online version of the paper is in English only, the newspaper that I saw was always in English and Syllabics. This is an aboriginal publication and as such represents one very common way to write in Oji-Cree.

The image posted above represents the same text in unpointed and pointed syllabics.

As a retrospective on this for Cree, originally it was intended by the missionary to be pointed. The Cree used it as unpointed text, except for one or two families who had a tradition of pointing. This corresponded to their reputation as priests and elders in the Anglican tradition.

Most missionary publications have only been partially pointed. Now on the internet we can see variations by community. The Cree School Board has a policy of using pointed text. The Naskapi unpointed. Most Cree and Oji-Cree text written by the Cree elsewere is unpointed or occasionally partially pointed.

Is it necessary to point text? Evidently some communities thought not. So why are others using pointed text? Hmm.

Thanks to Chris Harvey for great resources.

Mark invites readers to respond to this issue of whether to use fully marked text or not, whatever the language.


Blogger WERBEH said...

The text of the Hebrew Scriptures was not pointed until midway through the first millennium (~600 C.E.) by the Masoretes. The purpose of the vowel and accentual markings was to render accurately the ancient synagogue reading. This tradition was continued by the Tiberian school under the Ben Asher family and is reflected in the Aleppo and Leningradensis B19a Codices.

Identical forms are particularly problematic in Hebrew because of the tri-radical (i.e. three letter) root system of most words. Without vowel points many forms, particularly verbs, would be indistinguishable. Context would be the only guide to the specifics of certain forms.

The pointed Hebrew text is available at these sites:
Mechon Mamre
Daily Hebrew includes vowels

10:57 PM  
Anonymous Anonymous said...

And speaking of Hebrew, it is worth 'pointing' out that the convention that has developed in Modern Hebrew is to point only classical texts and poetry, and leave everything else unpointed.

The story with Yiddish is interesting as well. Over the centuries vowel points have been used in Yiddish to varying degrees, and by the twentieth century the convention became to use them only to distinguish between 'a' and 'ɔ/u', and sometimes to distinguish stops and continuants, but these are used only in printed texts; in handwritten Yiddish points are hardly ever used. This is possible because, in the grand tradition of Indo-European languages adapting semitic scripts, obsolete consonants were put to use as vowels, so the Tiberian system of diacritics became largely unnecessary for Yiddish. The so-called standard orthography makes greater use of points, but this system has only ever been used in a small minority of texts which have had very little impact on the masses of Yiddish readers and writers.

9:59 AM  
Blogger michael farris said...

From what I remember of Arabic (long time ago and I never got that far) I disliked the choice of full vowel marking or none. The full system was too repetitive while no marking was just hell. For my own purposes I worked out a compromise with far fewer (but some) marks that was very readable to me at least (y or w between consonants was enough to indicate long i or w for example)

With pinyin, I think the Taiwanese system of using no mark for the first tone (and not marking zero tone somehow or just not marking it at all) would be a good innovation.

10:29 AM  
Anonymous Anonymous said...

Hi Werbeh,

Thanks for the details and resources. I read some of your blog and it was neat to read that a commenter on your blog attends Regent College here in Vancouver.

When I read your site in Internet Explorer I saw those empty boxes again but no missing characters. I pasted a paragraph into Notepad and BabelPad with no better success. However, when I opened your site in Firefox everything came together. Three cheers for Firefox.

I was also interested in the input method you recommend. Lots of practical information on your blog - thanks.


8:33 PM  
Anonymous Anonymous said...


What you say about Yiddish sounds a little like Cree. A standard orthography may exist with all the points but is infrequently used. Interesting how there is a split between the 'standard' and the most common practice.


8:36 PM  
Anonymous Anonymous said...

Hi Michael,

I like what you are saying about a compromise. Altogether this discussion makes diacritics much more interesting. I do know that some Cree used points in a decorative sense just for stylistic effect without bothering with whether they had a phonemic function. Somehow they do appeal to people in the right quantity.


8:41 PM  
Blogger michael farris said...

I studied Aymara for several years. Aymara orthography still isn't completely standardized though it's much closer than it used to be.
Anyway, no written form of Aymara uses more than two diacritics, the n-tilde (though IME neither Aymara or Spanish speakers think of the tilde as a diacritic) and umlaut for long vowels.

There are three kinds of long vowels in Aymara:
lexical (part of the root or suffix),
morphological (an underlying short vowel is lengthened as part of some morphological process)
expressive (some vowels in some suffixes can be lengthened for expressive purposes, though it's hard to say just what is being expressed)

For a paper on competing orthographies I was writing I examined a lot of written materials (not that much, but what I had available at any rate) and found that native usage was inconsistent, but there were clearly discernible trends.
Basically they were least likely to write lexical vowel length, a little more likely to write (some kinds of) morphological vowel length and most consistent in writing expressive vowel length.

12:58 AM  
Anonymous Anonymous said...


Thanks for the example of expressive marking. A similar effect was also found in Cree orthography. The preaspiration was used for expressive emphasis but not in the context where it was phonemic. Nevertheless the linguists in some areas have persuaded the Cree to adopt a rigidly phonemic way of marking preaspiration.

I found a link for Aymara here.

1:21 PM  
Anonymous Anonymous said...

Please go to:

and try insert the following RSO transcript at "Plains Cree Y dialect"
Syllabics Transcriptop page!

Weshkac miina kihci-ihkwe eshinihkaasokopanen Noohkomihs. Mii tahsh okii-ayaawaakopan

naapenhsan e-ishinihkaasonikopanen Cahkaapehsh. Mii tahsh aha Cahkaapehsh kii-minoki

kaye kii-mashkawisii.

Mii tahsh miina kaye kii-nihtaa-papaami-ayaa aha Cahkaapehsh kaye tahsh

kii-nihtaa-mamiihkintisi awiiyan iko kaye awiiyaashiihshan kaye piko nipiihkaank kaa-ayaanic

ekwa kaye anihshinaape. Mii piko kaye kaa-ishi-wanihikaaniwatinik e-kii-ishaac,

e-kii-mamiihkincihtooc wanihikanini kaye piko nakwaakanini. Ahpan wententic kape-kiishik.

Paanimaa piko e-aakwaa-tipihkaanik e-pi-kiiwec. Mii tahsh aha Noohkomihs e-kanoonaac,

"Aaniin kaa-ishihcikeyan minikohk kaa-kii-inentiyan kape-kiishik?"

Mii tahsh aahpici e-naawinaakwatinik waahsa noohpimink entaawaac Noohkomihs kaye

Cahkaapehsh. Mii tahsh kaa-inaac Noohkomihs Cahkaapehshan, "Maacaan naasipiin

ci-naasihkaman nipi eka kekoon nipi e-ayaayank, aamaan kikashkihtoon maminsi

e-tootaman, aamaan ikaa kikohtaacihsiin."

Mii tepwe kaa-ishi-naasipiinishahwaakaniwic ci-naatahipiic. Mii tahsh Noohkomihs kaa-inaac,

"Manaa-kanawaapam tipihki-kiisihs, kaawin kaye kanoonaahken," okii-inaan. Mii tepwe

kaa-ishi-naasipiic e-kii-naatahipiic otahkihkwan e-tahkonaac. Ahpii tepwe shemaak

okii-kakanoonaan tipihki-kiisihsoon kaye okii-mamiihkinsomaan kaye piko

okii-paahpinontawaan. Mekwaac naasipiihkanaank ontahipaanink mii kaa-ishi-papaakimaac

ini tipihki-kiisihsoon. Piinihsh tahsh kii-pi-pehshonaakosi tipihki-kiisihs e-kii-nishkimikoc

Cahkaapehshan. Mii hsa ahpan e-kii-manipinikoc tipihki-kiisihsoon, anihsh kaawin

okii-nihtaa-noontansiin aana-kii-wiintamawaakaniwic.

Mii kekahpii Noohkomihs e-kwiinawi-piihaac ini Cahkaapehshan. Mii ehkitoc, "Aan inaap

ehpiihci-kinowensh-ontentic, ntashkaa e-mamiihkincihaakwen tipihki-kiisihsoon,

aana-kii-oncihakipan ekaa ci-tootank".

Let me know your opinion! At this time we are in discuss to use this UST progranam for Cree language (all dialects) in Alberta.

Best regards,

Edward Brabec, author

12:11 PM  

Post a Comment

<< Home