Experiment and Rule Updates

So just as an experiment, I decided to translate, verbatim, the pater noster from the original gothic. Just the words, mind you, not actually fiddling around with the grammar, to see if any new ideas or questions might reveal themselves in the manipulation of the phonology.

The original, with my gloss (and we can argue that on another blog):

Gothic:

swa nū bidjáiþ jūs:
atta unsar þu in himinam,
weihnái namō þein.
qimái þiudinassus þeins.
waírþái wilja þeins,
swē in himina jah ana aírþái.
hlaif unsarana þana sinteinan gif uns himma daga.
jah aflēt uns þatei skulans sijáima,
swaswē jah weis aflētam þaim skulam unsaráim.
jah ni briggáis uns in fraistubnjái,
ak lausei uns af þamma ubilin;
untē þeina ist þiudangardi jah mahts
jah wulþus in aiwins. amēn.
untē jabái aflētiþ mannam missadēdins izē, aflētiþ jah izwis atta izwar sa ufar himinam.
iþ jabái ni aflētiþ mannam missadēdins izē, ni þau atta izwar aflētiþ missadēdins izwarōs.

Gothic Gloss:

swa nū bidjɛ̄þ jūs:
atta unsar þu in himinam,
wīhnɛ̄ namō þīn.
kwimɛ̄ þiuđinassus þīns.
wɛrþɛ̄ wilja þīns,
swē in himina jah ana ɛrþɛ̄.
hlɛ̄f unsarana þana sintīnan gif uns himma daǥa.
jah aflēt uns þatī skulans sijɛ̄ma,
swaswē jah wīs aflētam þɛm skulam unsarɛm.
jah ni briŋgɛs uns in frɛstuƀnjɛ̄,
ak lɔ̄sī uns af þamma uƀilin;
untē þīna ist þiuđangardi jah mahts
jah wulþus in ɛ̄wins. amēn.
untē jaƀɛ̄ aflētiþ mannam missađēđins izē, aflētiþ jah izwis atta izwar sa ufar himinam.
iþ jaƀɛ ni aflētiþ mannam missađēđins izē, ni þɔ̄ atta izwar aflētiþ missađēđins izwarōs.

Gytc Gloss:

swā nau bidʒeþ jaus:
atta unsra þau in himinma,
waihne namo þain.
kwime þȳðnassas þains.
werþe wili þains,
swī in himin jā an erðe.
hlēf unsran þan sintainan gif uns himdag.
jā aflīt uns þat ai skulans saijem,
swaswī jā wais aflītma þem skulma unsarem.
jā nai bringes uns in frestyvni,
ak lœ̄si uns af þam yvlan;
unte þaina ist þȳðnagarþ jā māts
jā wulþas in ēwins. amīn.
unte jave aflītiþ manma misðīðnas iʒe, aflītiþ jā iʒus atta iʒur sā uvra himinma.
iþ jave nai aflītiþ manma misðīðnas iʒe, nai þō atta iʒur aflītiþ misðīðnas iʒuros.

Gytc:

swā nū bidʒeþ jūs:
atta unsra þū in himinma,
wīhne namo þīn.
kwime þȳðnassas þīns.
werþe wili þīns,
swē in himin jā an erðe.
hlaif unsran þan sintīnan gif uns himdag.
jā aflēt uns þat ī skulans sījem,
swaswē jā wīs aflētma þem skulma unsarem.
jā nī bringes uns in frestyvni,
ak lœ̄si uns af þam yvlan;
unte þīna ist þȳðnagarþ jā māts
jā wulþas in aiwins. amēn.
unte jave aflētiþ manma misðēðnas iʒe, aflētiþ jā iʒus atta iʒur sā uvra himinma.
iþ jave nī aflētiþ manma misðēðnas iʒe, nī þau atta iʒur aflētiþ misðēðnas iʒuros.

Preliminary Thoughts:

Final –wa, –wi → u
Final –u → a
Final –ja → i… implications here for class 1 weak verbs, end in /–in/, not /–na/. Class 2, /–on/ > /–an/, Class 3 /–na/, Class 4 /–nan/.

Some specifics about vowel reduction: Short vowels are deleted in unstressed interior syllables, but that has to happen after vowel reduction in final syllables, and if a final syllable is reduced to a syllabic (which will later be expanded), the unstressed internal vowel cannot be deleted (e.g. himinam > himinm̩ > himinma, not himinam > himnam > himn̩m̩ > himnama) …or do I actually like that better?

It looks like I’m going to want to start separating the clitics before vowel raising and umlaut, or ‘þatei’ is going to become ‘þetē’ instead of ‘þat ī’.

I’m starting to like the idea of intervocalic /f/ and /þ/ becoming voiced; I think I’m going to make that an official rule, since too many years of studying Old Norse are making me constantly do it accidentally anyway. I don’t think I’ll extend it to /s/, though.

Interestingly enough, despite all of the sound changes I’ve implemented, the orthography I’ve proposed makes it all very similar to the original; I’d imagine that a Gytc speaker would be able to read Gothic, though brutally mispronounce it, much in the same way Icelandic speakers can read Old Norse. Or, well, for that matter, the way we spell English in, ostensibly, Late Middle English.

New rules:

• [ij]V→īV (this can be stuck in just before long vowel raising, maybe as part of Final Short Vowel Lengthening… only no longer just final), and 
• Ø → j / V[+stressed]____+v (this is a persistent rule from early on in Gothic). 
• Hence, sija > sīja, ijōs > ījos, etc. Looks like this is going to tromp on my dreams of ija becoming iʒa, but that’s okay; that needed to be sorted out anyway. 
• [f,þ] → [+voiced] / V_____S. (Any reason not to add this in as an extension of stop-to-fricative expansion?)
• Clitic separation
• –ei#, -u-, -uh# → #ei#, #u#, #uh# (ev. ī, ū, ō)
• /ū/ precedes the primary verb.
• Deus Ex Machina
• A nice little persistent rule I made up that can contain non-locatable changes or things I just think would sound better (like mf# → m), but don’t feel like explaining or putting into historical context.

I've updated the rules list at http://ling.everywitchway.net/germanic/east/gothic/gytc/rules, along with still-outstanding questions. Bulleted and numbered lists are a bear in any system, and forget about moving from MS Word to Google Sites, so please forgive any inconsistencies for the time being. Once they're a little more solidified I'll try to go through the actual HTML code and make sure they're polished up a little.

Revision of Rules, and Unanswered Questions

I’ve been trying to simplify and clarify the rules I’d created at http://ling.everywitchway.net/germanic/east/gothic/gytc/rules, which were getting a little unwieldy. So far I’ve broken them down into 12 distinct stages, which may be able to be further reduced, but I want to be careful not to break anything (that I don’t want to break) as I manipulate them.

The rules are as follows:

1. Expansion of Intervocalic Voiced Stop to Fricative
1. [b],[d] → [v],[ð] / V____[+sonorant]
• Intervocalic /b/ and /d/ become continuant (if this hadn’t already happened in Gothic by this point).
• This does not affect /g/ at this time. OR, Gothic realization of intervocalic /g/ as [γ] reverts to [ɡ].
• Gothic realization of intervocalic /b/ as [β] becomes [v].
• Expands to include /b/ and /d/ before any sonorant (any vowel or l, r, n, m)
2. Devoicing of Obstruent Clusters
1. CC[+voice] → [-voice]
• Voiced obstruent clusters become unvoiced.
• [Must precede Rhotacism]
3. Rhotacism
1. z → ʒ
• Rhotacism begins with all instances of /z/.
• [Must follow Devoicing of Obstruent Clusters] 
4. Final Short Vowel Lengthening
1. V́# → V̄
• A stressed final short vowel becomes long.
• [Must precede vowel raising]
5. Stressed Long Vowel Raising & Diphthong Contraxion
1. V̄́[-low] → [+high] 
• A stressed long vowel is raised.
• /ī/ and /ū/ are raised to diphthongs /aj/ and /aw/, respectively.
• /ē/ and /ō/ are raised to /ī/ and /ū/, respectively.
• /ɛ̄/ and /ɔ̄/ are raised to /ē/ and /ō/, respectively.
• /ā/ is not affected.
• [Must follow Short Vowel Raising]
• [Must precede h-Assimilation]
2. VV → V̄
• Dipthongs (i.e. ai, au, iu) become condensed into long vowels (ē, ō, ȳ, respectively).
• Do you see how I cleverly averted all the controversy about the pronunciation of /ai/ and /au/ by making both /ɛ̄/,/ɔ̄/ and /ai/,/au/ end up as /ē/,/ō/, respectively? Please note, though, that the short forms remain short.
6. [h]-Assimilation
1. Vh → V̄Ø
• /h/ is deleted after a short vowel, and the vowel becomes long.
• [Must follow Stressed Long Vowel Raising]
7. Umlaut
1. V́[-front] → [+front] / ____(σ)/ī/,/j/
• A stressed non-front vowel (i.e. a, ā, o, ō, u, ū) becomes fronted (i.e. e, ē, œ, œ̄, y, ȳ, respectively) when /ī/ or /j/ occurs in the following syllable. 
• (Not affected by [i] at this time.)
8. Assimilation of Final [s] After a Defricate*
1. s# → Ø / [sp],[st],[sk]_____
• /s/ is deleted word-finally after /st/ or /ʃ/.
• [Concurrent with P&A?]
• *Defricate is a completely made-up word. Is there a better (non-lengthy) term for what I’d consider the opposite of an affricate? At least in terms of [s]+stop?
9. Palatalization & Affrication
1. [sk] → [ʃ] / V[+front][+high] _____
• /sk/ becomes palatalized (“/c/”) when it follows a high front vowel (i.e. e, ē, i, ī)
2. [tj],[kj] → [ʧ] and [dj],[gj] → [ʤ]
• /tj/,/kj/ and /dj/,/gj/ become affricates (/tc/ and /dʒ/, respectively).
10. Vowel Reduction
1. V̄[-stress] → V
• Unstressed long vowels become short.
• [ī,ē,ā,ō,ū,(ȳ) → i,e,a,o,u,(y)]
2. V[-stress] → [+reduced]
• Unstressed short vowels are deleted or reduced.
• a → Ø
• i,e,o,u → ə
• [Must follow Umlaut, Palatalization, and Affrication]
• [Must it? Maybe this needs to happen before Umlaut to make sense?]
11. Final Obstruent Devoicing (persistent)
1. C[+obstruent,+voice,+continuant)]# → [-voice]
1. Word-final [v,ð,z] → [f,þ,s]
2. [Is this even necessary, since it’s a persistent rule?]
12. Syllabic Expansion
1. S[+syllabic] → [-syllabic]ə

There are still a few unresolved issues I need to work out or work into this system.  Among them (complete with some of my scrawled unanswered questions):

• Voicing of intervocalic fricatives: f,þ,s → +voice / V_____V/Son.? 
• Is this necessary? 
• Why do I want to do this? 
• This would give us [ēði] (< aiþei) instead of [ēþi], but what about aiþþau? [ēþo]? [ēðo]? 
• Would have to happen after rhotacism, or that could get ugly. 
• g → Ø / ŋ____[+nasal] 
• I just think it would sound better when you end up with words like gangna or gangma. 
• What else is it going to impact? 
• Where to put it? Can this be concurrent with any other rules? 
• jj → ʒ 
• Why? 
• I kind of want /ija/ to become [iʒə], but at what cost? Maybe it should just end up as [ī]? 
• How? 
• Geminates? 
• mf# → m, re: fimf > fim 
• Expansion of ŋ-deletion to include Vmf → V̄f? 
• No, that would result in fīf instead of fim. 
• I hate the /f/ there! I want it gone! 
• Could I live with [fīf] instead? 
• No, way too Ingvaeonic. It’s got to come out [fim].
• Some sort of [f/b] interaction after [m]? 
• Does this violate Verner or the Prime Directive? 
• Why not? English does it plenty (comb, climb, lamb...) 
• Can I live with fimb? Maybe...
• Hey, what if I did expand ŋ-deletion not only to f but also s and/or þ. It could do some cool things to plural endings, turning the acc.pl. into some nifty shapes.
• Yeah, and also give you uns > ūs, fimf > fīf, and tanþus > tāþa.
• Hello, north-germanic sea coast. No.
• According to the rules above, “badja” would decline thus: 
• sing: baða, beʤis, beʤ, baða 
• pl: beʤ, beʤe, beʤma, beʤ 
• Weird contrast between [ð] and [ʤ]. But cool weird? Dunno yet. 
• Re: the h-assimilation rule, what about faíhu? 
• [fēu > fēa] is weird and awkward, and I don’t want the vowels to get too uppity. 
• Clarify the rule to only apply to [h] when it’s a coda to the short vowels nucleus? 
• That would give us [feha], which is even more awkward-sounding. 
• [fē]? 
• No, that’s Old Norse. 
• Still a conundrum.
• For that matter, there are a lot of problems with vowels crashing into other vowels they shouldn't be associating with. kniu/kniwa? Not to mention faíhiwē... yuck.
• Unstressed Short Vowel Reduction: 
• The rule I wrote above I think just applies to word-final short vowels. Should it be complete deletion for non-final short vowels, e.g. gytc from gutisk, instead of gytac, as the rules would spit out? Need more examples. 
• Clitics become separated from roots. 
• –ei (relative) > ī [aj]. 
• –u– (interrogative) > u > ū > ū [aw]. 
• Has to occur before final short vowel lengthening. 
• Where to put “ū”? Before the verb? 
• –uh > ō [ū]. 
• Frequent cases where it can still be clitic, but immune to unstressed vowel reduction? 
• weizuh > wīʒū, or wīs ū? 
• Remains clitic for pronouns and determiners only?

Vowel Arrangement

I’ve been trying to expand on my idea from the last post about using the lowered equivalents of raised long vowels to avoid the awkward diphthongs /ai/ and /au/, but I ran into a bit of a roadblock when I realized that that was going to “break” ē and ō, which were already raised from ɛ̄ and ɔ̄. I think I’ve got it solved, though, by giving a little nod to Wulfilas and rearranging my prejudices about diphthongs. So here’s my revised vowel system to Gytc:

Short vowels

i [i]
y [y]
e [e]*
œ [ø]*
a [a]
a [ə]
o [o]*
u [u]

* The short middle vowels o, e, and œ, fall somewhere between [o,e,ø] and [ɔ,ɛ,œ], respectively. Their long equivalents are more closed.

Long vowels

ē [i:]
ȳ [y:]
ai [e:]
œ̄ [ø:]
ā [a:]
au [o:]
ō [u:]

Diphthongs

ī [ai]
ū [au]
ei [ei]*
eu [eu]*

* /ei/ and /eu/ are the i-umlaut forms of /ai/ and /au/, respectively.

Okay, back to work.

So after a bit of a hiatus, I’ve been prompted by one of my conlang groups to keep working on gytc, so I’m getting back to work.

Yeah, um... this:

So that prompted this:

I’ve still got some work to do on it -- finding better unicode letters to represent various letters without mucking around with the whole private use subsets and creating a font and whatnot; finding better transliterations for a couple of characters; reevaluating my phonemic inventory -- but it’s a start.

Otherwise, not much has changed on the Gytc front, except for a sort of wicked idea: I think, to avoid “ugly” orthography like “mains” for [maɪns], I want to pull a page out of the English handbook and use the lowered version of long vowels (except ā) to represent their raised or diphthong equivalents, e.g. “mīns.” Also “hūs” ([haʊs] = house), fōts ([fu:ts] = foot), mēna ([mi:nə] = moon).

I’m still a little non-committal about certain sounds, though. II still have no idea whether /ʒ/, which is intended to be some sort of rhotic, should be [r], [ɾ], [ʁ], or even [ʒ]. I’m not sure I even want to make that determination – I just want to be sure that it’s differentiated from /z/ [z], which has to come from a voices /s/, and /r/ [ɾ] which comes from a “real” Protogermanic /r/.

04.20.13.04.19.09.35

A few more updates. There have been a few new words added, and I’ve rearranged a bit to simplify things. 

Specifically, I’ve replaced /c/ with /k/ and then /š/ with /c/, as i threatened to do in the fall, though i haven’t yet committed to changing /ǧ/ to /q/. Really, these are all just placeholders until i find the “right” alphabet, which i’m coming closer to. I’ve actually made a good deal of progress on this, but it doesn’t quite look “right” yet, so nothing formal so far.

In other news, i added a slightly unorthodox system of deixis, on top of the already questionable system of “ul,” “úli,” and “úlót;” adding the modifiers “la,” for “last” or “yester-” (e.g. yesterday, last night, last year) “ja” for “to-“ or “this” (e.g. today, tonight, this afternoon), and “þa” for “next,” (e.g. tomorrow, next week, next month). I’ll elaborate on these a little more when i’m not busy at work.

04.20.12.11.11.12.53

So, a few updates here.  Some minor vocabulary and some less minor phonological changes (or, rather, clarifications).

Let's tackle the big one first:

Long vowels

For a while now i've thought about adding in some sort of long/short vowel clarification to the orthography, but there is already so much going on with the markers for primary and secondary stress and those annoying letters that don't have a clear equivalent (like š and ǧ), that i wanted to be careful about not introducing macrons or something that would further complicate the orthography, at least until i create a writing system that I'm happy with.

After some thought, i've decided that a simple rule can supplant the need for long and short vowel markers.  Well, i thought it was a simple rule that i could write out with combinations of stressed and tense vowels, but there are, of course, exceptions, so until i work that out more succinctly, here's the bulk of it:

• The vowels e, i, o, ǫ, and u are always long, whether stressed or unstressed, unless part of a diphthong.
• The vowels ɛ and y are always short, whether stressed or unstressed.
• The neutral vowel a is short when unstressed, long when stressed.

There.  No need to add additional diacritics now.

Possible Orthography Change

On the same track, i've been toying with the idea of moving around some letters to get rid of the š and ǧ problem; possibly c > k, š > c, and ǧ > q, but i haven't decided yet.  I don't particularly care for q as a replacement for [γ]. I've also thought about replacing ǫ with w, but i don't really like that either.

New Vocabulary

Finally, i have a lot of new words that i've been adding to the lexicon, many through various translations in random "conlang" groups on Facebook, etc., and most recently by filling out an extended swadesh list.  One day i aspire to tackle the Universal Language Dictionary, but i don't think that's going to happen anytime soon, and if it does, it will be Northeadish before Maltšέgj. In the meantime, i'm going to work on getting the new words added to the lexicon today, but i'm a little distracted with the age-old problem of how best to keep the lexicon updated in a format that I can actually use effectively.  We'll see how it turns out... (For now, I'm creating entries in Word, stripping out the formatting, and then pasting them into Google Sites, which seems redundant and annoying.)

One exciting facet of new vocabulary building that i explored this morning was with familial relationships. Years ago i added the words "adína" (sister) and "ǫ́bri" (niece), after my sister Adina and her daughter Aubrey. I figured i'd continue with that theme, and added "tym" (father - Tim), crýstʌ (mother - Christl), crɛg (maternal uncle - Craig), kyp (paternal uncle - Kip), lýnda (materal aunt - Linda), hɛ́lɛn (maternal grandmother - Helen)... and so on. I don't have any brothers or nephews, so I might have to make something up on those fronts, and I had to choose among family members for who had the most "maltšέgj-able" name, but I thought it was a nice tribute.

04.20.11.11.24.10.16

I’ve recently had a bit of a chance to play around with Maltšɛ́gj in a “conlang” group on Facebook, which hasn’t led to much more than my reviewing some of the finer points of grammar and the creation of the word “djɛ́lyš” (coffee), but I have been considering a slight grammatical shift that could be quite interesting regarding the “free stress” of the language.

It seems that I, perhaps unconsciously, created most verbs with final stress. It’s not exactly a new concept, especially for English, but I think I would like to implement a “verbs have final stress, nouns/adjectives have primary stress” rule, along the same lines as English verbs borrowed from French or Latin, like présent/presént, cóntract/contráct, &c.

So in other words, leaving us with pairs like ðrɛ́pnid (permission) and ðrɛpníd (to be allowed, may); glácsi (memory) and glacsí (to remember); or bjóxɛf (full, complete) and bjoxɛ́f (to fill, to fulfill, to complete).

This would, of course, entail revising the lexicon quite extensively, and maybe modifying some verbs, but it’s just something I’m thinking about at the moment.

Silly things

I haven't had the opportunity to work on gytc for a while, but I was just playing around with some silly phrases and decided to include them here, gathering up a couple of fun new words.

• Flȳgwagnas mains ist ēla fuls! (My hovercraft is full of eels!)
• Ik kann glas itna.  It nī harmiþ mik. (I can eat glass.  It does not hurt me.)
• Ik im jausts. (I am cheese.)

The Jamin & Karen Supper Club: Midori's Floating World Café

Actually, the title is a little misleading: I went to Midori the other night with The Sweeties, not with Karen.  They treated me to what is possibly the only sushi place in the Twin Cities that i haven't been to yet in exchange for some laptop fixin'.

I wasn't really sure what to expect from Midori's Floating World Café, but what i definitely did not expect was how overwhelmingly cute it was!  It's just adorable.

We got some tasty beef pot stickers as appetizers with our drinks, and i was very pleasantly surprised by the delicious peach sake cooler - pretty much a sake spritzer with peach syrup in it.  Very refreshing. Then came little octopus dumplings and some sort of tofu salad.  As cute as they were, i need to accept the fact that i just don't like octopus. 

For an entrée, i got the sashimi and tempura bento dinner, and it was nice enough, pretty; nothing special.  The tempura was particularly good - especially the sweet potato; the sashimi was fine but nothing to blog about... and, having just relearned my lesson for the nth time, i gave the tako to Ms. Fledermaus.

I just killed syllabics :Þ

So I take some of that back.  I've just killed off syllabics altogether, so you can ignore the bit about them being their own letter à la icelandic and all that.

C[+sonorant,+syllabic] → C[-syllabic]a

Syllabic sonorant consonants become non-syllabic and are followed by -a.

swistr → swistra (sister)
rign → rigna (rain)

It also has some interesting implications for verbs:

dūn, stēn, gīn - no syllabics here.  But:

wisan > wisn > wisna
waírþan > werðn > werðna

Here's also a nifty twist for Class 1 weak verbs:

wiljan > wilin

Some new words and rules

I've been playing around with some new words, and by extension, new rules.  I added a rule to "Mora Loss" (which is really just a sort of catch-all bucket for about 30 other rules) to include long vowels:

V [+long]→[-long]/_______## (i.e. a long vowel becomes short at the end of a word or word segment.)

I also added an early rule which may be somewhat superfluous, as I think this already happened in Gothic, but not in the orthography:

b,d→v,ð/V_______V (i.e. intervocalic stops, not including g, become continutant.) 

I think g did this in Gothic as well, but I'm imagining something a little more interesting for g as time goes on...

I also added a bunch of new words, which are in the lexicon, mostly for the purpose of examples in the rules:

dōr n.st.n. door.
dūn v.t. to do.
fūts n.st.m. foot.
gangan v.i. to go. Also gīn.
gēts n.st.f. goat.
gīn v.i. to go. Short form of gangan.
gum n.w.m. man.  
haus n.st.n. house.
hōhs adj. high.
ja itj. yes, yea(h). 
kwīns n.st.f. woman.
lōmyna n.st.n. lightning.
man n.w.m. man, person (not gender-specific).
nī itj.  no, not.
standan v.i. to stand.  Also stēn.
stēn v.i. to stand.  Short form of standan.
þȳþs n.st.m. person.
werðan v.i. to become, to turn into.
wilen v.t. to want.
wisan v.i. to be.

I've also decided that syllabic sonorants can just sit there and be syllabic sonorants, icelandic style (e.g. sivn - "seven", unsr - "our").

Maltšέgj Project Reloaded... (only better than the Matrix)

I suppose this could be considered the first official revival of the Maltšέgj Project.  While codifying these entries into blog format, i've also undertaken to transcribe the lexicon on my linguistics site, and in doing so i've already made some changes, which in the sake of fairness, ought to be recorded here.

In part four of the project, i changed [ʃ] from § to ж, which is just hard to type and weird to look at.  For now, i've transcribed it as š (though i flirted briefly with the idea of ƨ, now that Times New Roman is a little more robust).  I've also replaced λ with ʌ, because it just looks a little neater, and all the various iterations of y and γ into plain old y.  I'd like to change ђ to something else, too, but that can wait... probably ǧ.

Transcribing these entries has reawakened Maltšέgj for me, and this might just be the beginning of a resurgence of the project.  Part Five, anyone?

Some More Thoughts on Rhotacism

I think i've been getting tripped up by other germanic languages when thinking about rhotacism.  In particular, I've been wondering how it would apply to the masculine singular nominative ending, and the fact is that by the time of gothic this ending has already become -s, and as such, rhotacism just can't apply to it.

I would still like to see some sort of rhotacism take place in gytc, but i don't want to see z and r turn into a single letter.  I think what i would like is a sort of runic-style "ʀ" with an indefinite pronunciation until i can come up with something more concrete, so for now i think it's safe to just use the ʒ character, and just wait to decide whether it should be pronounced [r], [ʒ], or, heck, maybe even [ɮ].  (I like ɮ.  It's fun to say.)

So we can put down as a solid rule that z > ʒ, but that this has no effect on s from a former z, and we'll decide what it sounds like later on.  I would also put this in a chronology where this change occurs after the rule in which voiced consonant clusters become unvoiced (e.g. razda > rasta > rast, not razda > raʒda > ract or rart).

I'm going to start a lexicon and rule page for gytc at http://ling.everywitchway.net/germanic/east/gothic/gutish to keep track of these sorts of changes as they occur.

Stay tuned!