West Germanic conlang

[Ekke]

Hi everyone,

I’ve been working on a West Germanic conlang on and off for many years now. It’s always been more of a side project, so I’ve set it aside for long periods before picking it up again when inspiration strikes. I joined this board a few months ago with the goal of finally sharing this and a few other projects.

So far I have different versions of this language, ranging from what could be described as Proto-West Germanic with the morphological complexity of an average modern West Germanic language to a phonologically conservative quasi-Low German dialect. However, while organizing my notes to present the project here clearly and concisely, I decided to take a different approach and try something new. This new version takes a different direction and shares more features with the Anglo-Frisian branch.

Below are the main phonological changes from PWGmc to Modern X that form the basis of this current version. Most of the changes are attested somewhere within WGmc. Do you think the overall result feels plausible and naturalistic, or does it come across as a bit like a "Frankenstein's monster"?

Changes up to Old X:
Ingvaeonic nasal spirant law.
Long ā fronting > /æː/ except when followed by a nasal. (Corresponds to “First a-fronting” in Anglo-Frisian phonological history).
Long ā + nasal > /oː/.
Monophthongization of *ai > /aː/.
Monophthongization of *au > /ɔː/
I-mutation. Most notably, this creates the front rounded vowels /y, yː, ø, øː, œː/ from /u, uː o, oː, ɔː/
*eu becomes a diphthong written ⟨ia⟩, the quality of which I have not yet defined exactly.

I think the system would be more symmetrical if /æː/ > /ɛː/, so:
Late Old X vowel system
Short: i y u e ø o a
Long: iː yː uː eː øː oː ɛː œː ɔː aː
Diphthongs: ia iu

Early Middle X:
⟨ia⟩ > /jaː/
Diphthongization of /eː øː oː/ > /ei øy ou/ > /ɛi œy ɔu/ spelled ⟨ei, eu, ou⟩
/ɛː œː ɔː/ > /eː øː oː/

Late Middle X:
/aː jaː/ > /æː jæː/
/iu/ [iy] > /yː/

Changes regarding the fricative /ɣ/:
- Word-initial /ɣ/ > /g/
- Word-final /ɣ/ > /x/ ([ç] after front vowel)
- Word-internal after /a/ and front vowels /ɣ/ > /j/
- Word-internal after back vowels /ɣ/ > /w/

Open syllable lengthening:
- Lengthened /ɪ ʏ ʊ/ > /ɪː ʏː ʊː/ > /eː øː oː/
- Lengthened /ɛ œ ɔ/ > /eː øː oː/
- Lengthened /a/ > /ɑː/

Late Middle X vowel system
Short: ɪ ʏ ʊ ɛ œ ɔ a ə
Long: iː yː uː eː øː oː æː ɑː
Diphthongs: ɛi ɔu œy

Early Modern X:
/ɛi ɔu/ > /ai ɑu/
I'm tempted to have /œy/ > /œi/ or /ɔi/
Degemination of geminated consonants: /hɛbbə, bɪddə, sɛggə, wɪnnə/ > /hɛbə, bɪdə, sɛgə, wɪnə/
/θ/ > /t/, /ð/ > /d/ (this could have happened in Middle X)

Modern X vowel system
Short: ɪ ʏ ʊ ɛ œ ɔ a ə
Long: iː yː uː eː øː oː æː ɑː
Diphthongs: ai ɑu œy~œi~ɔi

Edit: I'll post a few example words and sentences in the romanization challenge thread.

[Travis B.]

Ik see, dat du ook andskeden heb, en germaniske spraak tu maken, aber dain spraak is keen helpsspraak in de underskiid tu main spraak.

Bearbeting: Ik heb andskaiden tu andskeden wichselt wegen dain kommentaren.

[Ekke]

Hi, Travis! /tuːzənd tank foːr də roːmanizeːrɪŋ/! (or maybe better /roːmanizatsjoːn/?)

To be honest, I can't write much in this conlang yet because I'm still in the phase of deciding on basic vocabulary and syntax. My previous versions were more influenced by Low German and Standard German, but with this one, I feel like I should look for more inspiration in the Frisian languages and maybe even Old English. On the other hand, the modern Frisian languages are significantly influenced by Low German, Standard German, and/or Dutch.

By the way: I just realized that, despite what it may seem in StG (“entscheiden, entschied, entschieden”), this wasn’t a strong class 1 but a strong class 7 verb (PGmc *skaiþaną), so: /andʃæːdə/, Present /ɪk andʃæːd/ (and if I want to keep i-umlaut in the 2nd and 3rd Sg, /duː andʃeːds/, etc.), Past /ɪk andʃaid/, Perfect /ɪk hɛb andʃæːdə/.

[Travis B.]

One thing to consider doing is introducing palatalization of onset /ɣ/ like in Anglic or some Low German varieties (even though from a quick glance it seems this did not happen in Frisian).

[Moose-tache]

This is very fun!

I do think your changes are very naturalistic. I have also made a WG conlang by cobbling together attested or near-attested changes, and found the result to be somewhat boring. Have you encountered that problem, or have you found ways to make your language interesting?

Also, do you know where it's spoken?

[Ekke]

One thing to consider doing is introducing palatalization of onset /ɣ/ like in Anglic or some Low German varieties (even though from a quick glance it seems this did not happen in Frisian).

It seems the palatalization of velar stops *k and *g did not occur in Frisian under the same conditions as in Anglic, but it did occur initially before front vowels:
PWGmc *geban > Old Frisian jeva, West Frisian jaan, North Frisian jiw / jiiw / jeewe...
PWGmc *geld > Old Frisian ield, jeld, West Frisian jild, Saterland Frisian Jield, Jäild, North Frisian jild.
PWGmc *kirikā > Old Frisian tsirke / tzirke, West Frisian tsjerke, North Frisian Sērk / schörk etc.
PWGmc *kāsī > Old Frisian tzīse, *tzēse, West Frisian tsiis, tsjiis (modern Frisian languages have ts-, tsj-, s-, sch-)

You can find non-palatalized *k, for example, in West Frisian "tsjerke, bjirk, ik" that were palatalized in English (church, birch, Old English iċ).

I don't really know much about the Frisian languages —certainly not as much as I should to attempt a naturalistic quasi-Frisian? Saxo-Frisian? conlang (I'm doing it anyway )— but my impression, based on the online texts and dictionaries I've checked, is that West Frisian has borrowed many words from Dutch, and Saterland Frisian from surrounding Low German and Standard German. I guess that if you’ve come across Frisian words with an unpalatalized “g” where you’d expect it to be palatalized, they were likely loanwords.

I have seriously considered the possibility of palatalizing initial velars before a front vowel in this conlang, specifically /k/ > /t͡ʃ/ and /ɣ/ > /j/, but that presents other dilemmas I don't feel like dealing with right now, such as what to do with *ggj (currently I simply have /g/, as in /sɛgə, lɪgə, brʏg/), or how to spell /t͡ʃ/... In any case, the idea is definitely in the back of my mind.

[Ekke]

This is very fun!

I do think your changes are very naturalistic. I have also made a WG conlang by cobbling together attested or near-attested changes, and found the result to be somewhat boring. Have you encountered that problem, or have you found ways to make your language interesting?

Also, do you know where it's spoken?

Thanks!

I don't know... It's true that I don't work on a particular conlang until it's “complete”; instead, I set it aside and come back to it later, but I wouldn't say it's because I get “bored”... I think it's more that I have to attend to real-life things, and after a while, new conlang ideas come to mind.

I find working on a-posteriori conlangs has its own special appeal, especially if one is more or less familiar with the related real-world languages and/or there are enough of them to draw inspiration from (as is the case with the Germanic, Romance, or Slavic languages).

Unfortunately, I haven’t decided where this conlang is spoken, beyond “somewhere near the North Sea”. I’ve even considered a fictional North Sea island (I think there are quite a few alt-history conlangs set in unsunk parts of Doggerland).

By the way, what made you find the result of your WGmc conlang boring? Can you describe it or provide a link to a description so I can take a look?

[Travis B.]

Maybe set your language in Jutland? (E.g. a Jutland that never became Danish-speaking?)

[Ekke]

Maybe set your language in Jutland? (E.g. a Jutland that never became Danish-speaking?)

Yes! A not-completely-Danified Jutland is a very good setting for several reasons. It would make various features of my conlang more convincing, such as the lack of final-obstruent devoicing (which I believe began in Old Dutch), the preservation of the dental fricative(s) well into the Middle or Modern period, etc. Its proximity to the homeland of the Angles and the Jutes might even justify the early vowel changes it shares with the Anglo-Frisian languages (*ā fronting, *ai > ā), assuming that these changes were already underway before the migration. Btw, is anything known about the language spoken by the Jutes? (Some people think that the Frisians were a branch of the Jutes.)

In this setting, my conlang would still be subject to a significant influence from Middle Low German during the Hanseatic period, which would explain the lexical, grammatical and syntactic similarities with Low German.

[Moose-tache]

This is very fun!

I do think your changes are very naturalistic. I have also made a WG conlang by cobbling together attested or near-attested changes, and found the result to be somewhat boring. Have you encountered that problem, or have you found ways to make your language interesting?

Also, do you know where it's spoken?

Thanks!

I don't know... It's true that I don't work on a particular conlang until it's “complete”; instead, I set it aside and come back to it later, but I wouldn't say it's because I get “bored”... I think it's more that I have to attend to real-life things, and after a while, new conlang ideas come to mind.

I find working on a-posteriori conlangs has its own special appeal, especially if one is more or less familiar with the related real-world languages and/or there are enough of them to draw inspiration from (as is the case with the Germanic, Romance, or Slavic languages).

Unfortunately, I haven’t decided where this conlang is spoken, beyond “somewhere near the North Sea”. I’ve even considered a fictional North Sea island (I think there are quite a few alt-history conlangs set in unsunk parts of Doggerland).

By the way, what made you find the result of your WGmc conlang boring? Can you describe it or provide a link to a description so I can take a look?

I think the issue was that the WG languages are so similar to one another, that mixing and matching changes that are characteristic of the group creates a language that just sounds like all the others. It was hard to make it really distinct. But I guess it's all in the details. Or you could have one or two really unique changes that give it its own character.

[hwhatting]

PWGmc *kirikā > Old Frisian tsirke / tzirke, West Frisian tsjerke, North Frisian Sērk / schörk etc.

Just in case you're interested, it's Särke in Saterfrisian.
I grew up near the area where it is spoken, and our local newspaper occasionally published materials in Seelters (Saterfrisian). Most of it was collected by a linguist from Texas, Marron C. Fort, who studied the language and became a bit of a local celebrity ("The American guy who is interested in our weird dialect!")
I used to collect these materials, but they were lost at some point after I moved out of my parents' house.

[WeepingElf]

Maybe set your language in Jutland? (E.g. a Jutland that never became Danish-speaking?)

Yes! A not-completely-Danified Jutland is a very good setting for several reasons. It would make various features of my conlang more convincing, such as the lack of final-obstruent devoicing (which I believe began in Old Dutch), the preservation of the dental fricative(s) well into the Middle or Modern period, etc. Its proximity to the homeland of the Angles and the Jutes might even justify the early vowel changes it shares with the Anglo-Frisian languages (*ā fronting, *ai > ā), assuming that these changes were already underway before the migration. Btw, is anything known about the language spoken by the Jutes? (Some people think that the Frisians were a branch of the Jutes.)

In this setting, my conlang would still be subject to a significant influence from Middle Low German during the Hanseatic period, which would explain the lexical, grammatical and syntactic similarities with Low German.

Such a language may have survived on an island off Jutland's coast.

[Ketsuban]

You can find non-palatalized *k, for example, in West Frisian "tsjerke, bjirk, ik" that were palatalized in English (church, birch, Old English iċ).

You also find birk and ik in Scots. (And kirk, which is probably a Norse loan.)

[Ekke]

@ hwhatting: It’s a pity that you lost those Saterfrisian materials. I haven't found any long texts in the language, but I have found some online resources that include short samples, like a 162-page Sprachlehre des Saterfriesischen (which mentions Marron C. Fort) here: https://www.oldenburgische-landschaft.d ... friesisch/, a couple of online dictionaries here: https://www.xn--saterfriesisches-wrterbuch-tvc.de/ (which lets you browse Fort’s Saterfriesisches Wörterbuch) and here: https://www.seeltersk.de/deutsch-saterfriesisch/.

@ WeepingElf: Yes, maybe Rømø or Fanø, where, apparently, only Danish is spoken. Or a fictional island in that area. It might seem silly, but when I imagine an alternate history to set a conlang in, I don’t want it to “replace” a real language; for example, I wouldn’t want to imagine an alt-history where Söl’ring doesn’t exist...

@ Ketsuban: I didn’t know about ik in Scots. In fact, don’t have a clear idea of when it diverged from English. I’m curious to know if (wild speculation) it happened when the ⟨c⟩ after a front vowel was still a palatal /c/ and in some cases like ⟨ic⟩ the palatalization was reverted, or if that word is also influenced by Norse. Slightly off-topic: When I was considering if I should front my ā (< PGmc *ai), I took into account that ā fronts in Scots (āc, hām, stān > aik, stane, hame).

I found a mistake in the romanization challenge thread: /wiː høːr/ should be /wiː høːrə/.
Also, the m. sg. demonstrative /deː/ might be better as /dai/: /dai apəl/ "that apple".

Edit: Replaced the direct pdf link to the Saterfrisian grammar with the page that contains the pdf link to avoid download prompts (plus, it has extra materials!)

[WeepingElf]

@ WeepingElf: Yes, maybe Rømø or Fanø, where, apparently, only Danish is spoken. Or a fictional island in that area. It might seem silly, but when I imagine an alternate history to set a conlang in, I don’t want it to “replace” a real language; for example, I wouldn’t want to imagine an alt-history where Söl’ring doesn’t exist...

Yes, that was my idea.

[hwhatting]

@Ekke: Thanks for the links!