Wiktionary talk:About Alemannic German

Mentionings

WT:CFI: "For terms in extinct languages, one use in a contemporaneous source is the minimum, or one mention is adequate subject to the below requirements [...] the community of editors for that language should maintain a list of materials deemed appropriate as the only sources for entries based on a single mention" (bolding adding)
Mentioning sources which are already used although without being present in said list and without further explanation:

• Emil Abegg, Die Mundart von Urseren (Beiträge zur Schweizerdeutschen Grammatik. Im Auftrag des Leitenden Ausschusses für das Schweizerdeutsche Idiotikon herausgegeben von Albert Bachmann. IV) [The Dialect of Urseren], Verlag von Huber & Co. in Frauenfeld, [1911] (HathiTrust-US) -- by WP's date of death, it's not PD/gemeinfrei
  • But: The spelling is different in the main text and in the Wörterverzeichnis at the end. Wörterverzeichnis: "Durch Aufhebung von Entrundung, Verdumpfung und Diphthongierung sowie der sekundären Dehnung und Kürzung ist der Lautstand soweit als möglich dem gemeinalemannischen Status angenähert worden." That sounds like: For the Wörterverzeichnis, the proper Urseren form as present in the main text was altered, "de-Useren-ed".
    WT's entries seem to be based on the Wörterverzeichnis with replacement of ^ by a doubled vowel (like â -> aa, ä̂ -> ää) but give the page for the Urseren form and not the Wörterverzeichnis as source.
• Le nostre parole – Unsere Wörter, several works, maybe only the one for Lusern is used ATM in WT (isolelinguistiche.it, istitutocimbro.it, lusern.it)

-84.161.37.130 15:26, 7 June 2018 (UTC)

The sorry state of gsw

Working on this lect recently, I couldn't help but notice the disorderly state its documentation on Wiktionary currently unfortunately is in.

Orthography

About gsw: It's a dialect continuum spanning 6 countries with no official governing body. I'd wager that at least 90% of written gsw occurs in text messages and on the internet. There are no official documents in gsw to the best of my knowledge and gsw literature (apart from from children's books, audio books and poems) is rare. As such, the vast majority of gsw is written subject to no codified rules but merely the speakers' personal preferences.

Spelling generally falls on a spectrum with the two extremes being: staying as close to the German cognates' / French donor words' writing as possible vs. representing the actual phonemics as accurately as possible. Most people probably fall somewhere in the middle or a bit closer to the former camp.

The two above paragraphs explain why for every single word there are multiple, sometimes even dozens of writings. I've written word in italics because the usual "that which is separated by spaces" definition of the term "word" is completely meaningless in the context of this lect (it's also wrong in the context of other languages in my opinion but that's a different discussion). To give an idea of just how many spellings there might be, a simple term such as iigsee consisting of /iː/, /ks/ and /eː/ may be rendered by any combination of {"i", "ii", "ih", "yy"} × {"gs", "x" (rare)} × {"e", "ee", "eh"} for a grand total of 24 spellings (for this exact pronunciation, of which there are also multiple).

So which spelling should be followed? To list some possiblities:

1. Every spelling variant that can either be attested or analogously attested (what I mean by this: if a suffix construction 'X-Y' is attested, then 'Z-Y' is attested through analogy if Y represents the same morphological and phonetic unit in both variants; analogously for prefixes) may have its own article.
2. Create full articles only for distinct phonetic units (which can be named either according to IPA or a phonemic writing, more later) and let all the attested or analogously attested actual spellings merely be stub articles that point to the real article of the phonetic unit. This would be similar (but exactly in reverse) to how Japanese romaji entries merely point to the real writing of the words ('in reverse' because in the case of Japanese, the romaji is the useful nonsense that points to the actual word, whereas for gsw, the actual attestable forms would point to the useful nonsense). This also has the big advantage that we can document regional changes in phonology a lot better because often times the same spelling is used for different pronunciations (e.g. /ʏ/ and /y/ are usually written as just 'ü' despite there being clear geographical borders of their use in certain syllables). As for the main articles, they could either be named according to:
   1. A phonemic writing system (the most prominent of which probably is Dieth-Schreibung). Much to my surprise, I have recently learned that this one is actually used somewhat regularly on the gsw wiki (see e.g. [1]). It introduces numerous new letters (ì, è, ä̃, ǜ, ...), none of which are ever seriously used in the real world. Technically, it's a pretty good system, although I honestly don't know why /e/ and /ə/ are both spelled 'e' but /ɛ/ (which is AFAICT never phonemic but only a different realization of /æ/) gets its own symbol 'è'. There are currently some articles following this convention, e.g. èèrnscht (well, almost; 'sch' should be 'ʃ').
   2. IPA (broad). This may seem like a rather radical change but it eliminates the middle man that is the aforementioned phonemic writing system. Unlike these arcane orthographies, IPA is used and well-known. To see how such a page could look like under this proposal, refer to ˈxʏb̥l̩. The articles Chübel etc. would then be made to be simple redirects (compare e.g. genki). In the context of this option, we would have to specify a transcription system for gsw (hotly debated here) for the creation of which I would definitely need assistance of a more linguistically educated person.

I've started to like the last proposal quite a lot, even though it is totally unlike anything else presented on Wiktionary. In all fairness, the situation around gsw is also probably unlike anything else; desperate times call for desperate measures.

Under any proposal, I'd additionally advocate for a small set of prescriptivist rules regarding the spellings:

1. Disallowing ß in all gsw lemmas (see e.g. Chaabiß)
2. Prescribing that all nouns and proper nouns be capitalized (see e.g. the mess in the alternative forms of Chatz).

These should be fairly obvious so I'm not going into great lengths as to why they are required in my opinion.

Verbs

I have found no literature on this topic but I think it is possible that gsw verbs are agglutinative (the stem is still fusional and there are predictable sound changes at the borders of the parts):

The unstressed /xøm:(ə)ɾsnə/ (chömmersne (“can we ... it to them”)) is to be constrasted with the stressed /xøm:(ə)ɾs ɪnə/ (pitch accent and focus on ihne (ɪnə, “to them”)) and /xøn miə̯ɾsnə/ (pitch accent and focus on mir (miə̯ɾ, “we”)). Any hypothetical variation of these phrases using es (əs, “it”) as opposed to the reduced affix /s/ sounds ungrammatical to me (probably because the stressed accusative form for 'it' doesn't exist and one is forced to use the demonstrative pronoun instead). For this reason and because the pitch accent and semantics (focus) change between the forms, I don't think this is merely connected speech, though I would love to hear the input of somebody more linguistically knowledgeable than me.

This is relevant for the creation of conjugation tables that we are currently lacking.

Practical considerations

Us having documented only 1632^[here] gsw lemmas at the time of this writing makes it still possible to implement such wide-ranging changes in policy. It is however to be noted that I am unable to change everything on my own because a considerable bulk of our gsw lemmas are from rare varieties of gsw from the northern Italian alps with missing pronunciation information.

Pinging the gsw editors: @Widsith, Linshee

--Fytcha (talk) 17:02, 1 November 2021 (UTC)

---

@Fytcha: Hello, I haven't been that active in Alemannic lately, but I would certainly like to discuss it! As for the Italian stuff, Upper German varieties in Italy are a special interest of mine so that's why there's so many. Certainly there should be more entries for the dialects of Zürich, Basel, Bern, Luzern etc.

Regarding standardization and spelling, I like your option 2, using Dieth-Schreibung for the main entry rather than IPA. There are several languages on here that use IPA, Americanist or something similar, but they generally aren't written at all. Since Swiss German is written, albeit not in any standard way, I think we should still have actual spellings. The variant entries could have something like

1. Issime form of Chatz “cat”.

on the definition line. Would require some new templates.

We could also make a few adjustments to Dieth-Schreibung, like no one uses <ʃ> in real life right? So just use <sch> instead.

Also I agree with getting rid of ß, however I would like to allow my beloved Italian dialects to keep their lowercase nouns. The few speakers are mostly only literate in standard Italian, so if they would try to write it down I imagine they'd use lowercase. This guy Pietro Axerio from the 19th century only capitalized proper nouns in his writings. And the linguists who write about these dialects also use lowercase.

On the verbs: My first instinct is that the /s/ bit on verbs is just a clitic. The sound change seems like assimilation. Nasal place assimilation is super common and it happens in English too, like <it's been mad> /ɪts bɪn mæd/ [ˌʔɪts pɨ̃:m‿ˈmæ:d̚]. However I don't speak Swiss German so take my words with a grain of salt. I do speak standard German and am about to get a BA in linguistics though.

Also once we figure out a good system I'll be able to help you with the Italian Walser entries! Linshee ☺ ☆ 18:43, 1 November 2021 (UTC)

Oppose, as it's prescriptive and not descriptive.

As for the additional prescriptive rules:

1. Alemannic is also spoken in places, where ß is (or was) common.

2. Italian Alemannic might have Italian-influenced features like lack of capitalisation (vruantag/frontog/fròntag/vröntàg).
Maybe Italian Alemannic (and also Italian Bavarian) should be a separate language how there's also Dutch Low German (nds-nl)?
Patuzzi calls it "Walser Gressoney", "Walser Issime", "Walser Carcoforo", "Walser Rimella / Campello Monti", "Walser Formazza" and there's a separate ISO-code for Walser (wae), but Walser and Italian Alemannic aren't the same.

--Sasha Gray Wolf (talk) 19:28, 1 November 2021 (UTC)

@Sasha Gray Wolf: The only four articles that contain 'ß' in the whole gsw category are Biisääß, Chaabiß, schieße, süeß, all of which are labeled "Uri". Note that Bavarian (bar) and Swabian (swg) have their own codes and categories, so it is not entirely clear to me what you're referring to right now. Alsatian? Vorarlberg? Fytcha (talk) 20:15, 1 November 2021 (UTC)

There's Low German, and Dutch Low German (nds-nl) which often lacks capitalisation. And there are for example summer#Alemannic German and summer#Bavarian which also lack capitalisation. Like with Low German, it could make sense to split Alemannic (gsw) into Alemannic and Italian Alemannic and Bavarian (bar) into Bavarian and Italian Bavarian. The Italian form then could lack capitalisation. --Sasha Gray Wolf (talk) 20:33, 1 November 2021 (UTC)

@Linshee: Hey, glad to see you chime in as well!

Re "Since Swiss German is written, albeit not in any standard way, I think we should still have actual spellings.": Note that under both the IPA proposal as well as the Dieth-Schreibung proposal, we would still retain the actually attested spellings as separate articles and also under both would the "main part" be in an article whose name is not used (ˈxʏb̥l̩ is just as unattested as Chǜbl), though I see that the name of the Dieth-Schreibung might sometimes coincide with an actually used form, which is not the case for the IPA.

Re "We could also make a few adjustments to Dieth-Schreibung, like no one uses <ʃ> in real life right?": Nobody in real life uses ǜ etc. either (I don't count als.wiki here, they are not citeable as per WT:ATTEST anyway). Note also that Swiss German has both /ʃ/ as well as /ʃ:/. Would you propose using 'ssch' for that (similar to 'cch')?

Re "however I would like to allow my beloved Italian dialects to keep their lowercase nouns.": Sure! Maybe we can just treat the different sub-lects of gsw differently in regard to policy? I think, concerning Swiss German at least, these two prescriptions are very sensible.

Btw, are your Italian Alemannic articles in Dieth-Schreibung?

(Addendum:) I've now created the pair ˈxʏb̥l̩ and Chübel according to how I had envisioned them. I think I might have not done a good job of putting my ideas into words so I just went ahead and created them for illustration.

(Addendum:) Another big advantage of IPA as compared to Dieth-Schreibung as the primary article is that it is simply less work. The IPA should ideally be written in any article anyway so it is no additional work. However, transcribing into Dieth takes additional time (because we wouldn't do it otherwise). What's more, if, at a future point, we were to decide that we want all Dieth-Schreibungen for whatever reason, it would be as simple as adding an IPA to Dieth conversion template and creating articles for all such forms by a bot. This is not possible in the reverse direction because Dieth-Schreibung hides some details (most importantly stress and schwas); one cannot automatically derive the IPA from Dieth-Schreibung.

Thanks for your help! Fytcha (talk) 19:59, 1 November 2021 (UTC)

Re "are your Italian Alemannic articles in Dieth-Schreibung?": Considering Dieth-Schreibung#Gross-/Kleinschreibung, the answer is no or only in 1st-edition Dieth-Schreibung. Considering Dieth-Schreibung#Konsonanten (with ʃ) and e.g. herbscht, the answer is no or non-1st edition. Put together, it's not Dieth. --Sasha Gray Wolf (talk) 20:19, 1 November 2021 (UTC)

@Sasha Gray Wolf: But it was supposed to be prescriptive so could even guess consistently a priori where a word is even found. I think the prescriptive part was not so much making up spellings like Wikipedia but choosing certain schemes rather than always picking the most common form of a word (which you won’t be sure about either if it is only some people on the internet writing out the words.)

“Dutch Low German” was believed to be removed, only that there are no Low German editors to care. If you split you only get silly names like ***Bavarian Swiss German and ***Swiss Swiss German—country borders are really no argument and spelling differences do not justify a split either since they are only projections of language(s) as a separately understood entity. You should come loose from spellings, they aren’t the language. Fay Freak (talk) 20:35, 1 November 2021 (UTC)

Any source? WT:About Low German and WT:About Dutch Low Saxon don't seem to support it. --Sasha Gray Wolf (talk) 20:51, 1 November 2021 (UTC)

Wiktionary:Beer parlour/2016/April#Let's kill nds-de/nds-nl. And that’s even without me partaking. There are a lot holding this opinion. It is telling that you tried to make any conclusion from the presence of langcodes on Wiktionary. It is basically the is–ought fallacy. Langcodes are just copied over from Wikipedia, (of course) with an idea of only few, they don’t mean to serve as an argument “it’s a language”. We also have to consider practical purposes of documenting the languages that don’t matter for encyclopedias like Wikipedia and Ethnologue, aside from some of the language names not or barely existing outside of Ethnologue or such reference works. Not every distinction must be reflected in a new language header or a separate dictionary. Fay Freak (talk) 21:29, 1 November 2021 (UTC)

@Fytcha I don't speak this at all, but I've been meaning to find out more about it for quite a while. It probably has something to do with a certain Johann Friedrich Enz from Herisau, Appenzell Ausserrhoden, who somehow ended up as a US immigrant named John Frederick Entz a couple of centuries ago (my great great grandfather). At any rate, I only have one reference bookmarked: the Schweizerisches Idiotikon, which would seem to fall in the 10 percent that's not in text messages or the internet (the online resource is OCRed from a dead-tree original). I've only started to look through it, but it seems like it might provide some guidance re: orthography. Chuck Entz (talk) 08:39, 2 November 2021 (UTC)

@Chuck Entz: Really cool to see that there's some interest in gsw from outside the German speaking world as well!

I've worked with schweizerisches Idiotikon before and I think they use their own system that I haven't seen outside of that dictionary:

1. ë for either [æ] or [ɛ] (hard to tell right now from the examples)
2. ā, ē, ī, ō, ū, ǟ, ȫ, ǖ, for long vowels
3. Use of apostrophe and parentheses ("G(e)fäll", "G'fäll"), especially in the definite article ("z'", "'s")
4. Hyphens for (some?) contractions ("dass-d^u")
5. Use of superscript for some sounds (to me it seems like they use it for optional sounds that are missing in some dialects): "I^ch könnt deⁿ Bademerēⁿ nümmeme^r Rëcht gëⁿ." ^[2]
6. Use of small letters ("Viȫl_e̥li")^[3]
7. â, î which I haven't found out the meaning of
8. í

There are some inconsistencies, e.g. they reflect the difference between /k/ and /kː/ but not between /x/ and /xː/ (probably because the latter is represented by a digraph).

--Fytcha (talk) 12:32, 2 November 2021 (UTC)

@Chuck Entz: Just for curiosity's sake: Have you ever been to Switzerland? --Fytcha (talk) 17:23, 3 November 2021 (UTC)

Aside from a couple of trips across the Mexican border and one to Samoa, I've never been outside the US. I don't get out much... Chuck Entz (talk) 15:00, 4 November 2021 (UTC)

@Chuck Entz: Couldn't you say the checkuser role allows you to travel the world, though? PUC – 15:15, 4 November 2021 (UTC)

• Some comments…I think lemmas should have some kind of system, but I think they should be based around real usage, not an invented system much less an IPA-based one. The way the Idiotikon does it is kind of a disaster, so I think we should avoid going in that direction. I actually think there's more printed Swiss German than you suggest, I always pick up something from the Mundart shelf whenever I'm in Orell Füssli and apart from the obvious big hitters like Pedro Lenz there are loads of poets working in ‘dialect’ as well as columns in newspapers (including printed text messages like that weekly feature in 20 Minuten). This includes many historical writers like Meinrad Lienert that I've studied with interest. All of these are pretty much untapped resources at this point (albeit books favour some dialects like Bernese over others, for mysterious reasons). In terms of principles for lemmas, probably we can say that nouns should be capitalised and we should avoid ß, beyond that for now I don't think we can say much for certain. In my opinion the way to proceed is to gather real-life examples first and draw conclusions about standardising lemmas afterwards, but obviously this requires some work on our part. Ƿidsiþ 13:30, 2 November 2021 (UTC)

I would agree that the organization of the work borders on... how shall I say it... Idiotik. I'm sure it makes perfect sense to an expert on Swiss dialectology, but the rest of us wouldn't be able find anything without their... rather idiosyncratic... search function. It would never have occured to me to look for "chabiss" under "kab, keb, kib, kob, kub." It's got lots of data in a unified format, though. Whatever orthography we choose, we can tie it to theirs.

As for the orthography, we need to learn from their mistakes: it should be accessible to someone like me who has no experience with the language but wants to learn it. While IPA has the benefit of being an internationally-recognized, uniform standard, there again, we don't want the organization of our lemmas to be only accessible to someone who's taken linguistics classes. I had great teachers- they pretty much all have Wikipedia articles- but it's been 35 years and I'm a bit rusty. It should be something that someone who is neither a linguist nor a native speaker can use, but that a linguist or a native speaker would recognize. Since standard German is an important part of the linguistic environment for native speakers, perhaps something based on German might work- though you always have to be careful not to reinforce the stereotype that "dialect" is just an inferior form of Schriftdeutsch. Chuck Entz (talk) 15:07, 2 November 2021 (UTC)

@Chuck Entz: To be fair, searching "Chabis" (the most common spelling, corresponding to the most common pronunciation) yields the correct result right away. --Fytcha (talk) 17:23, 3 November 2021 (UTC)

@Widsith: Huh, that's interesting. Yesterday, I've been to both the book stores in my small town and neither had a single work in Swiss German (apart from the funny dictionary below; I've explicitly asked). Which Orell Füssli in Zürich did you refer to? The one in the underground part of the main station (probably not) or the one at the Europaallee? I want to have a look too next time I'm in Zürich. --Fytcha (talk) 17:23, 3 November 2021 (UTC)

Yeah the main one off Bahnhofstrasse has lots, but even the small one at Oerlikon station has a few. The "Spoken Script" editions are really good too because they preserve Swiss-German spoken-word stuff, which is surprisingly quite an active scene here! Ƿidsiþ 08:32, 4 November 2021 (UTC)

• I just got myself this cute little one. They only document the Zürich dialect and they use four additional symbols: è for /ɛ/ (they use ä for /æ/~/a/) and ẽ, ĩ, õ for nasalized counterparts. The issue is that, despite the artificial symbols that don't occur in most natural writing, they still don't provide a phonemic spelling system. I can give more information if somebody is curious (they actually describe their spelling in the first few pages) but I don't think this writing system is an option for us. --Fytcha (talk) 17:23, 3 November 2021 (UTC)

  I have a few: https://www.goodreads.com/review/list/53165-warwick?shelf=swiss-german. Two dictionaries of Zurichdeutsch and one of the Einsiedeln dialect, plus some other general resources which vaguely try to cover all dialects. I think there are a few points we should reach consensus on, and we can probably do it without too much stress – all recorded variants can be included in "Alternative forms" sections anyway:

  1. Capitalisation of nouns (I think we should do this, since most books do – though not all)

  2. Infinitive verb ending (I prefer <-e> to <-ä> but I guess it doesn't matter too much)

  3. Representation of long vowels (I prefer doubled characters, <aa>, <üü> etc rather than extra diacritics)

  4. Whether/how to distinguish between /ɛ/ and /æ/ (in my experience this is not usually distinguished in orthography, so my preferred solution would be to lemmatise <ä> for both and distinguish between them in the Pronunciation section. This also goes for some other vowels, e.g. in ZDeutsch there is a split between the vowels of böös /bø:s/ and nööch /nœ:x/ but usually this isn't written and when it is it requires a complicated Unicode diacritical solution)

  5. Representation of /ʃp/ and /ʃt/ (there seems to be an even split between writers who follow Standard German by using <sp, st> and writers who prefer <schp, scht>. I don't really care but we should probably pick one)

  I'm sure you could suggest other points. Perhaps we could compile a list of these to serve as the basis for an eventual Policy page. Ƿidsiþ 08:33, 6 November 2021 (UTC)

  @Widsith: Those are some good points and I'd be on board with most. I think our general objective should be: Trying to find the best orthographic system without resorting to non-German diacritics/symbols. The key word in here is of course best which I'd define as a mixture of reflecting the pronunciation and being actually used in the real world (or at least being readily understood by natives).

  We agree on most points. Some things that we have to consider still:

  1. <schp,scht> look really awkward to my eyes at syllable onsets but I agree that they are valid (maybe even more common) in syllable codas (probably because of German orthography where <sp,st> stands for /ʃp/, /ʃt/ at syllable onsets only). I would not like to have <schp,scht> at onsets so for me the choice is between:
     1. Writing <sp,st> everywhere (hasple, Chiste)
     2. Writing <sp,st> in onsets and <schp,scht> in codas (haschple, Chischte)

     Note that there are words such as Spast/Spasst /ʃpɐst/ (not sure how many more there are though), which for me are an argument for the second option. When it comes to use, I'd claim that the second option is more common on the internet and in chats (I'll take your word regarding the writers, you're more familiar with Alemannic literature than I am).

  2. I think we shouldn't use the other nontransparent German digraphs for anything other than their transparent pronunciation
     1. <ah> etc. shouldn't be used for long vowels
     2. <ie> shouldn't be used for /iː/ but for /iə̯/
     3. <ei> shouldn't be used for /ɑi̯/ but for /ei̯/
     4. <eu> possibly shouldn't be used for /ɔɪ̯/ but admittedly <oi> looks awkward, most people write <eu> (this one is a hard call), similarly <äu> for /ɔɪ̯/.
  3. <y> is a tough call too. I personally never use it and neither do any of my friends (not even the Bernese), but it has a strong tradition in Switzerland; even a canton's official name is written with <y> for /iː/. I'm not a fan of it because its only purpose and distinguishing feature from <ii> is etymology (Zyt but not yne; has to do with whether there was an /ei̯/ there in Middle High German AFAIK).
  4. I'm personally not a fan of the apostrophe either even though it is used (d'Frau). I'd be fine with banning it entirely (except for in alternative forms and the likes, as usual). FWIW the Alemannic Wiki isn't particularly fond of it either: [4] (see also how they write Schrybig and Schriibwiise right under lol, this is probably what's going to happen if we use <y> as well)

  Note that all those rules for spelling should in my opinion only apply to the main article and translations in translation boxes. It's perfectly fine to have articles for all sorts of spellings (articles consisting only of {{alternative spelling of}}). It's just not sensible to have all alternative forms of e.g. chnuuschte in translation boxes so there needs to be some set of rules to determine the principal spelling and once we have that it's nothing but sensible to write the main article under that principle spelling too (otherwise people would always have to click twice to get from a translation box to the meat of the article).

  I will write a draft for Wiktionary:About Alemannic German soon. Grammar is also something that needs to be discussed, see the (in my opinion incorrect) inflection table here: Zit Fytcha (talk) 18:52, 8 November 2021 (UTC)

  I agree with all of that! I wouldn't use <y>, it was very common in older writing but I had the impression it was a bit old-fashioned now. And yes, all recorded forms would be under "Alternative forms" anyway. Ƿidsiþ 14:17, 9 November 2021 (UTC)

  @Widsith: What do you think about <eu>/<oi>? Compare for instance [5] with [6]. Maybe we should be inconsistent there just because it's so widespread. Fytcha (talk) 15:08, 9 November 2021 (UTC)

  @Fytcha: I forgot about that one. Both my ZTüütsch dictionaries use <öi> and in fact they distinguish between <öi> ([øɪ]) and <ö̀i> ([œɪ]), as in the difference between Höi and Hö̀i, which I have to confess I never realised sounded different. But yeah this is not something most writers would represent in print (not least because it took me a few minutes to work out how to actually type it), and I would guess (out of pure speculation) that in some dialects these sounds merge anyway. Incidentally, referring to a point you made upthread somewhere, these dictionaries do not distinguish between short and long forms of /ʃ/, /x/ or /ŋ/, and I think that is normal. Another question that occurs to me is how to represent the /kx/ sound. Some people don't write it but other sources do, e.g. I have a copy of the script for Die Schweizermacher (!) which uses forms like Dankche throughout. Ƿidsiþ 09:02, 13 November 2021 (UTC)

  @Widsith: Interesting. I've had a conversation with a friend today about <öi> (caused by your comment) and he slightly annoyedly brought forth the claim that there's many people who write <öi> in places where they pronounce neither [øɪ̯] nor [œɪ̯]. He (born and raised in Stadt Zürich) pronounces both the greeting as well as the word for hay as [hoɪ̯] (I use this for the greeting but for hay I use [hɔɪ̯]). As for 'height', we both don't use this word but instead [høːχɪ]. Not sure if there's a minimal pair in our dialects but (unlike in Standard German) Swiss German doesn't have a tense/lax division, compare e.g. [høːχ]/[nœːχ] (pronouncing them as though merged (both with [œ]) sounds very Baslerisch to me).

  We speculated that that 'misspelling' was born out of people's aversion to writing <oi> (I bet you can imagine how often primary school kids are corrected upon writing things like "hoite" etc.) so people who want to neither write <eu> (because it looks German and isn't phonemic) nor <oi> (for the above reason) might use the widespread <öi> even if it doesn't fit their personal dialect perfectly. This theory is corroborated by the fact that words without a High German equivalent (like the greeting Hoi) are actually written with <oi> instead of <eu>. Another theory is related to the fact that the words that are pronounced with [oɪ̯] in e.g. Zürich are pronounced with [øɪ̯] in Bern (though I'm not sure if it's all the words or what kind of correspondence there is exactly). Don't take this as conclusive though, it's just armchair speculations.

  As to what we should do on Wiktionary: I'm not so fond of merging [øɪ̯], [œɪ̯], [oɪ̯] and [ɔɪ̯] all into <öi> (but I think this is not what you've suggested anyway). What do you think about taking <eu> as the main spelling, as inconsistent as it is? The reason is simply how widespread it is and also the fact that theres no variation of /eu̯/ anyway that this takes the symbol away from. Fytcha (talk) 18:32, 13 November 2021 (UTC)

  @Fytcha: You say "Swiss German doesn't have a tense/lax division" [sic] but then you say that you DO distinguish between those vowels, so I'm a bit confused what you mean! I personally would use <oi> for /oɪ ~ ɔɪ/ and <öi> for /øɪ ~ œɪ/, especially if as you say there are no minimal pairs. Basically, I would argue that this is something generally distinguished in Pronunciation sections rather than in orthography. (I see that in some cases, the two alternate, e.g. my dictionary gives the word for ‘Heu’ as Hö̀i, Höi suggesting that there is variety in which vowel speakers use.) I am not so keen on <eu>…older writer used it but my impression is that modern writers avoid it for being too German. Work has been so busy that I haven't had time to really research this properly, but in the books I have within reaching distance I can do a quick survey. I see Pedro Lenz (Bern) uses <öi> (zwöi, öichi); he also has <äu>, but that is for a phoneme that I think would be <uu> in Zurich, e.g. säuber. ‘I schlofe töif und schnäu.’ Ernst Burren (Solothurn) uses <öi> (fröid, zwöi). Achim Parterre (Biel) uses <öi> in fröie (these last two are from the same anthology, so possibly there is a house editorial style?). Andreas Neeser (Aarau, I think) uses <öi> as well (öisereim, ghöiet). So actually in the books I have here it's quite unanimous, which I wasn't expecting! Ƿidsiþ 09:04, 14 November 2021 (UTC)

  @Widsith: Sorry, I've expressed myself quite poorly in my last message! To clear up what I intended to express: In High German, the vowels occur in pairs; their tenseness is linked to their length, i.e. there's /œ/ and /øː/ but not /œː/ nor /ø/. This is not the case for Swiss German, for which I provided an example.

  There's another mistake in my last message: I've said that using <eu> for /oɪ̯/ (and similar) would be fine since /eu̯/ doesn't occur. This is wrong. I didn't think of the Bernese dialect where /lː/ is famously replaced with /u̯/ in (some) syllable codas, which gives rise to /eu̯/ in words like zeue (zelle, "to count").

  "he also has <äu>, but that is for a phoneme that I think would be <uu> in Zurich, e.g. säuber. ‘I schlofe töif und schnäu.’" Just a side note, this is due to the same /lː/->/u̯/ process I've described above, so this would be ‘schnäll’ (or ‘schnell’) in more central/eastern dialects.

  I've changed my mind and I think I would be somewhat on board with your suggestion "<oi> for /oɪ ~ ɔɪ/ and <öi> for /øɪ ~ œɪ/". I still maintain that (at least on the internet) <eu> is vastly more common than <oi> (1240 vs. 1, 67 vs. 4, 404 vs. 6) but at least it's neat and more consistent this way.

  I forgot to reply to your question with /kx/: I haven't really thought about this before but I'm a bit skeptical as to how often <kch> is actually reflected in writing (<kch> 0 vs. <kh> 54 vs. <k> 410, <kch> 0 vs. <kh> 22 vs. <k> 149, <kch> 0 vs. <kh> 0 vs. <k> 69 (I hope my Google comparison are at least somewhat of value to you)). Be advised that this phoneme is pronounced more like [kʰ] in Graubünden, Basel and Liechtenstein (possibly more, not sure). Additionally, I have to admit that I'm confused about even just the IPA transcription of the various forms of /k/ (and /p/). My own dialect features at least /ɡ̊/ (in luege), (possibly) /ɡ̊ʰ/ (in luegt; hard to discern, I've said it out loud at least 20 times now), /k/ (in goo), /kː/ (in Egge or läcke), /kʰ/~/k͡x/ (in Lack) and /k͡x/~/q͡χ/ (in umkheie). Really confusing and the common written renderings seem to be equally all over the place. I'm really not linguistically versed enough to make sense out of this mess. Fytcha (talk) 13:23, 14 November 2021 (UTC)

  I get what you mean about tenseness and length; yes, good point. And don't worry, I don't think anyone is linguistically versed enough to handle the dialectal complexities of Swiss German! I take your point about spelling of [kx] – I think you're right, this is something again that probably is best not reflected in orthography for the lemmas. I think we have a good basis to start listing some principles on the ABOUT GSW page! I am pretty busy with work recently but I hope to put some time aside to help with this and find some useful examples to make further decisions. Ƿidsiþ 17:46, 14 November 2021 (UTC)

  Okay sure, I will also be more busy in the near future but I'm going to work on the about page in small steps. I'm sure more topics and points of discussion will come up in the course of that. Fytcha (talk) 18:14, 14 November 2021 (UTC)

• Another point that I've wanted to bring up: We maybe need a system to distinguish between same-dialect synonyms and other-dialect synonyms. I personally use both IPA^(key): /iː/ and IPA^(key): /iːx/ for the first person singular nominative pronoun but I don't use IPA^(key): /ɪːɡ̊/ (most commonly encountered in Bern), so IPA^(key): /iːx/ and IPA^(key): /ɪːɡ̊/ are different types of synonyms of IPA^(key): /iː/. --Fytcha (talk) 17:30, 3 November 2021 (UTC)

  But this is already distinguished in the orthography, since "ig" is often written in Berndeutsch and both "i" and "ich" are common in ZDeutsch. You're right though that we should probably explain in such entries which dialects use which and why. Ƿidsiþ 08:35, 4 November 2021 (UTC)

  @Widsith: Yes, they are differentiated in spelling, that's right, but this is not quite the point that I've wanted to make. What I've originally meant was: When writing the synonyms section, there should ideally be a consistent (better: machine-parsable) way to denote the fact whether the main word and the synonym have a mutual dialect or not. I hope I've explained myself better this time. Fytcha (talk) 14:46, 4 November 2021 (UTC)

  If every form (the lemma and each listed alternative form, synonym, etc) were {{{qualifier}}ed as to what dialects it is used in, readers would be able to see when a dialect uses more than one spelling. The lists of alternative forms and their qualifiers may be very long, but ... length may be the price of precision and intelligibility. - -sche (discuss) 11:08, 7 December 2021 (UTC)

• Another question: What to do in translation boxes? There's many words for which there are numerous dialectal variants and/or common spellings. I don't think we want all of our translation boxes to look like the one here: apple core (in this box, only variants that differ by pronunciation are included; if we were to include all attested spellings for every variant that might just be twice the entries). Also, this should be solved to allow editors to be more productive when adding gsw translations. --Fytcha (talk) 14:41, 4 November 2021 (UTC)

• I would propose not listing gsw as a sub-lect of German within translation boxes. For one, it looks very awkward when selecting gsw as a preferred language (it previews as "German: Alemannic German: ..." in the translation box header), and secondly, swg and bar aren't listed under German either despite being considerably closer to Standard German. Furthermore, renaming the whole category from "Alemannic German" to "Alemannic" should also be considered; it seems to me like that is the more common name in English literature [7]. --Fytcha (talk) 21:54, 16 December 2021 (UTC)

• I'm actively following this discussion because I'd like to make more contributions for Alemannic but I've been hampered by the lack of clear guidelines. My special area of interest is Südbadisch (particularly Breisgauer varieties) but I'm not a fluent speaker so I'm often unsure of certain details, such as vowel length. Thus, while I like the idea of using a standardised orthography such as Dieth-Schreibung for main entries, in practice I might have difficulty getting the correct forms.

How useful do folks find broader subdivisions like "Low Alemannic/Niederalemannisch", "Lake Constance Alemannic/Bodenseealemannisch", "Höchstalemannisch/Highest Alemannic", etc.? Would there be any usefulness to employing these in addition to labels identifying the specific local varieties?Linguoboy (talk) 17:18, 17 December 2021 (UTC)

@Linguoboy: I think this is already done pretty well on a category level, refer e.g. to Category:Regional Alemannic German. I'm not sure whether making explicit mention of these within the article is of any benefit though, given that a more precise classification is available. For me personally, my perception of the different dialects is tied much more strongly to the geographical regions than the "proper" linguistic classification. To give an example, the Urner dialect (Highest Alemannic) feels considerably closer to my native High Alemannic dialect than certain other High Alemannic dialects like the one spoken in the city of Bern. If a more precise classification is not possible or if a term's use is accurately confined by this linguistic categorization (as is the case for words like gschniit if I'm not mistaken), then I think using such a label (and only such a label with no mention of geographical regions) would be the way to go. I could add the necessary label data to allow for correct categorization and linking if needed. --Fytcha (talk) 05:59, 30 December 2021 (UTC)