Talk:SignWriting

Wikipedia in American Sign Language proposed

Please see meta:Requests for new languages/Wikipedia American Sign Language 2. Thank you.--Pharos 21:11, 9 March 2007 (UTC)

Popular? An opinion

I would seriously disagree that SignWriting is "popular" in the ASL community. There are not many ASL users, apart from those who study ASL as a lingustic field, that are familiar with SignWriting. The average ASL user does not know about SignWriting and those that might be familiar with SignWriting's existence do not use it. It is grossly misleading to say that it is "popular." -Mela —Preceding unsigned comment added by 156.99.135.160 (talk) 07:56, 23 August 2009 (UTC)

Orthography

A standardized symbol set is not the same thing as a standardized orthography. The IPA is a standardized symbol set, but there are numerous ways to transcribe English in the IPA, both phonemic and phonetic. There are some rather nasty arguments here on Wikipedia about how English should or should not be put in the IPA. SignWriting isn't the same thing as SignWriting ASL, and it sounds as though there's a lot of variability in the SW community. That isn't necessarily a bad thing—English before Caxton was written however the writer saw fit—but it does cause difficulties. Actually, those difficulties may soon be a thing of the past, just as the problem with typography is now ancient history: with an electronic dictionary, the software could lead you to the entry no matter how you spell an entry. But I imagine that with a print dictionary, you often won't find an entry for a spelling you come across, which means you need to convert things in your head before you can look them up. — kwami (talk) 00:36, 12 March 2008 (UTC)

"Actually, those difficulties may soon be a thing of the past, just as the problem with typography is now ancient history: with an electronic dictionary, the software could lead you to the entry no matter how you spell an entry." Eye guess ewe haven't had two youse eh spell chequer. (I guess you haven't had to use a spell checker.) Errors caused by reliance on spell checkers are numerous. --Thnidu (talk) 23:17, 8 January 2009 (UTC)

Collation

I removed the following statement from the problems section because it is incorrect and unsourced.

The lack of overt phonemic structure or standardized orthography makes alphabetization difficult. For example, the hand used to sign a word makes no difference to the meaning, but it does change the spelling.

SignWriting has had computerized collation since 1998. Users have found this method of collation to be natural and intuitive. You can read about it in detail in the SignSpelling Guidelines 2008. You can view a few dictionaries. You can download and use the SignBank Database. Click on any of the symbol in SignPuddle symbol frequence page and view the signs collated properly.

Not understanding how SignWriting does collation does not mean that it is difficult or a problem. Slevinski (talk) 18:48, 2 July 2010 (UTC)

I see Kwami put this back. I'm deleting it again because it's just original research. -- Evertype·✆ 22:30, 2 July 2010 (UTC)

Okay, that difficulty has evidently been overcome. We should, however, retain a note that SW is not phonemic.

What of the lack of a standardized orthography? Has that been solved? Can't just write what you sign, since there is no inherent way to decide which elements are to be written. — kwami (talk) 22:37, 2 July 2010 (UTC)

I don't think you've "proved" that it isn't phonemic. Or shown what it is if it is not. If you are moved to try to do this, you should post your essays here on Talk to see if there is consensus. It sounds like OR to me, though. -- Evertype·✆ 22:49, 2 July 2010 (UTC)

SignWriting is more phonetic that Stokoe Notation, but SignWriting is phonemic. A phonemic script will use a restricted set of symbols to capture the meaningful contrasts of a language. In sign language, there are 4 types of phonemes: location, hand shape, motion, and facial expressions. Let's analyze the hand shapes.

Computerized SignWriting has a limited number of hand shapes (261), more than Stokoe Notation, but still limited. For each hand shape, SignWriting adds 6 palm facings, 8 primary rotations (45 degree increments) and 8 mirrored rotations. These 6 palm facings are a restriction, but are they meaningful? If we can find a minimal pair of signs that only differ in palm facings, then the palm facings are meaningful. The signs for "ask you" and "ask me" use the same hand shape (index finger up) and the same motion (squeeze finger closed). The only difference is palm facing (palm facing away from the signer and palm facing towards the signer); therefore palm facings are indeed phonemic. The same can be done with rotation. For SignWriting to be truly phonetic, hand shapes would need to be able to be created with any finger arrangement, with any degree of palm facing, and with any degree of rotation. This type of phonetic writing is possible with SignWriting by hand, but computerized SignWriting uses meaningful restrictions and is phonemic. Slevinski (talk) 02:22, 3 July 2010 (UTC)

There are several sources which note that SW is not phonemic. I didn't think that was being questioned.

In order for it to be phonemic, all 8 orientations would need to be shown to contrast, not just rotation in general. Every variation in sign space would need to be shown to be distinctive as well, and AFAIK it hasn't. Also, to be phonemic, it would need to be based on a phonemic analysis of the language it is used to write, and our article (and AFAIR the Sutton site) brags that no such analysis is needed. Using SW for ASL is like using aspiration, vowel devoicing, dark els, and variant stress levels for writing English, and arguing that it's phonemic because English uses voicing, velars, and stress phonemically. — kwami (talk) 02:42, 3 July 2010 (UTC)

I think that your insistence on comprehensiveness here is unwarranted. 'Every' handshape need not be shown to contrast "all" 8 orientations for orientation to be considered phonemic. Plenty of languages use voicing and voicelessness as phonemically distinctive features and yet do not distinguish /s/ from /z/. Irish is one of these. Velarization and palatalization are two other features which are not realized comprehensively throughout the phonology of Irish. I don't find your argument here convincing, Kwami. And indeed I find your tone to be very POV ("brags"). SW is a practical writing system for Signed Languages, just as Latin is a practical writing system for many spoken languages. In the main, most languages using Latin use it phonemically. (Latin distinguishes b and p, d and t, g and k, etc.) The same can be said for SW, and exceptions to the rule do not make either non-phonemic. -- Evertype·✆ 10:18, 3 July 2010 (UTC)

Well sure, it can be used phonemically. It could be used as an abugida for English, for that matter. But it's presented as non-phonemic, that no phonemic analysis is required. The hand locations, for example, could encode a lot of phonetic detail. Irish may not have palatalization on all consonants, but it only encodes a 2-way distinction, which is what the phonology is based on. If it encoded a 4-way distinction, that would be non-phonemic.

If SW is used phonemically, we should be able to attest to that. — kwami (talk) 17:06, 3 July 2010 (UTC)

Phonetic and Phonemic SignWrting

While we all like hard boundaries, there is usually more of a spectrum. Instead of phonetic and phonemic we can talk about emic and etic accounts (an observer versus a person within a culture). SignWriting started as a etic account (by someone outside the culture) and produced a phonetic script. The turning point for SignWriting happened when the people inside the culture started to use the script. The change is cemented in SignWriting history as the change from the receptive viewpoint to the expressive viewpoint. The receptive viewpoint is used in transcription. The expressive viewpoint is used for authorship. With this change, requested by those inside the culture, the SignWriting script moved from the purely etic account to include an emic component. This emic component has continued to grow over 30 years. The SignWriting script is neither phonetic nor phonemic, it depends on how it is used.

The latest symbolset of SignWriting (the ISWA 2010) could be said to be phonetic, but it is not equivalent to the IPA. The IPA contains every sound it is possible to use for a voiced language. If there is any argument within the IPA community, it is very small and very insignificant. The ISWA 2010 is extensive covering a reasonable 652 symbol bases and organizing 36,600 static and unique graphemes. However, the ISWA 2010 is not exhaustive like the IPA. The ISWA 2010 includes 261 hand shapes. There are hundreds of possible hand shapes that could have been added, but were not. The choice of hand shapes was based on the experience of an international user base. The choice of hand shapes is neither phonetic nor etic. The ISWA 2010 is a phonemic account of a phonetic ideal. Some international sign languages have multiple generations of writers. The most experienced are writing phonemically right now. Beginning writers most often write phonetically. The phonetic writing is useful for analysis to discover the phonemic component. The hand shape frequency reports are especially relevant. Low frequency hand shapes can be identified and possibly combined with a similar hand shape to reduce the size of the ISWA 2010 subset for a specific language. Slevinski (talk) 17:09, 5 January 2011 (UTC)

SignWriting is not featural

SignWriting is not featural. SignWriting does not identify features, but segments of gestures. A feature is an analysis of the physical mechanics of sound production. A feature is defined as meaningless because it is a physical account of a sound. For a voiced language, the sound is meaningful and the physical account is meaningless. Of course, it is not entirely useless because there is the skill and craft of lipreading. Lumping SignWriting as featural is a travesty because it has absolutely nothing to do with the original development nor the existing classification structure. I understand that Wikipedia can only reference existing information, but using a flawed classification is worse than no classification. The SignWriting script is better thought of as segments of articulatory gestures. Slevinski (talk) 17:09, 5 January 2011 (UTC)

Of course it's featural. It illustrates the features of the elements of sign. The symbols for different phones are similar depending on how similar the phones are. If it weren't featural, the symbol for a palm-up orientation would look nothing like the symbol for a palm-down orientation. And since when is a feature "defined" as meaningless? — kwami (talk) 18:57, 4 May 2011 (UTC)

SignWriting is definitely featural, it specifies both phonetic features and segments. A SW character typically (eg. the sign MEET) indicates the beginning, middle and end segments of a one-syllable sign, and it also specifies many many features, notably [+/- bend] for each joint of each of the fingers.

Sound is irrelevant to the discussion. A linguistic feature is an analysis of the physical mechanics of language production. For a voiced language the sound is meaningful and for a signed/visual language the visual image is meaningful. The physical account describes via distinctive features how that aural or visual image is created. Sometimes the description is in terms of acoustic properties, more often it is in terms of the physical movements of articulators e.g., [+/- nasal] describes whether or not the velum is raised; [+/- bent] describes whether or not a joint is straight; [+/- round] describes lip positions used with both signed and spoken languages. Dzho (talk) 03:39, 19 February 2012 (UTC)

How popular is SignWriting?

As a user above noted, the article contains no hard facts on how widely used SignWriting is. Is it even well known among the deaf? It mentions some "student magazines" that use SignWriting, but how many magazines? Are these magazines widely read? Are they notable? Why isn't there a criticism section? I would be quite surprised if nobody had any problems with this. Does the article even mention how SignWriting is written? All I see is a listing of symbols. cntrational (talk) 16:04, 19 March 2011 (UTC)

Supposedly there is a manual script, not just computer input, but I haven't seen it. You're right, arguments of it being popular could be propaganda, but it would appear to be more popular than Stokoe notation, which is almost entirely unknown in the Deaf community. — kwami (talk) 18:59, 4 May 2011 (UTC)

A wide variety of texts are available, including bible translations. SignWriting is used in many countries. There are no censuses though. -- Evertype·✆ 00:45, 5 May 2011 (UTC)