Modern Gothic

[Emily]

introduction

Well, everyone knows Wulfila's Gothic died in the first millennium. What this conlang presupposes is: maybe it didn't?

as mentioned in other threads, i've been working for a while on a project to develop a modern-day descendant of wulfila's gothic. gothic, of course, is the only member of the east germanic subfamily that we have any significant corpus for, and seems to have completely died out by 1000 AD. (another east germanic language was discovered several centuries later to have been spoken in crimea, but we only have one problematic source of data for that and it too is long dead.) gothic tribes were pretty highly mobile, and at various points occupied territories as far east as modern-day ukraine and as far west as spain. their christianization is attributed to the bishop wulfila, who is also credited with developing the gothic alphabet (closely modeled on greek but with some latin letters and possibly runic characters pressed into service as well) and translating the bible. the form of christianity they practiced, however, was declared a heresy at the council of nicaea and was stamped out in the ensuing years, a major factor in why so few gothic texts have survived to the present day

the premise behind this project is that gothic survived in eastern europe, where the speakers practiced their religion in secret and mostly kept to themselves, living almost like the amish. (in the story, in fact, the "discoverer" of modern gothic is initially led to believe the community she embeds herself in to be a low german-speaking mennonite village.) crucially, the speakers have had essentially no contact with other germanic speakers for 1500 years, and instead have been surrounded mostly by slavic speakers

broadly speaking, my design goals were for the resulting language be strikingly different from any existing germanic languages and yet still be true to its roots but also to the realities of developing as a minority language in a majority slavic region. methods for achieving this (and different things i've tried to strike the balance between) include

• it needs to show slavic influence (this comes out especially in the phonology)
• it should not undergo post-proto-germanic changes (umlaut, etc.) that are common to the existing germanic languages
• it should preserve archaic features that gothic had but the other languages lost
• if gothic has some weird difference from the other languages (e.g. the word for "house" is razn while every other germanic language has hūs), do my best to keep it around

methodologically the way i've been working on this is to develop it at three separate stages. treating wulfila's gothic as "old gothic", i apply a first set of sound changes and then play with the effects to arrive at "middle gothic" (or "phase 1"), then do the same for "early modern gothic" (phase 2) and finally "modern gothic" (phase 3). this is the best way for me to strike the balance between trying to create 1500 years of realistic linguistic change while still being able to actually do it in a reasonable amount of time. obviously this involves a lot of working back and forth, but i'll try to keep my notes clearly labeled

a note on transcription
generally speaking, phonology will be in the IPA, but with the following exceptions:

• ipa "θ" is written /þ/, and affricates are written with ligatures: /ʦ ʣ ʧ ʤ/
• stress is indicated with acute accent rather than the typical stress symbol
• some morphemes will include the symbols /-ʲ/ or /-J/, which indicate certain types of palatalization which i will explain when we get there

lastly, i'll try to remember to write the romanized gothic text in bold

[hwhatting]

I hope I said that already when you mentioned that in those other threads, but I really like your project.
Some comments:

another east germanic language was discovered several centuries later to have been spoken in crimea, but we only have one problematic source of data for that and it too is long dead.

You're very cautious here; AFAIK most people who accept the source call it Crimean Gothic and assume that it's a descendant of Wulfila's Gothic.

it needs to show slavic influence (this comes out especially in the phonology)

Influence on phonology is actually stronger in language change situations than in language contact situations; language contact predominantly leads to similar semantic developments (like similar division of semantic fields, calqued expressions, etc.) and similarities wrt to syntax and gramatical categories (e.g., Yiddish having question-forming perticles like Slavic, or the similarities in the Balkan Sprachbund - subordinated clauses instead of infinitives, development of a definite article, etc.).

[Emily]

sound changes and phonology
most of the posts in this thread will focus on one phase of the language, but since the sound changes are the backbone to the entire project i figured it made sense to post them all together

there's debate over the pronunciation of certain gothic letters and digraphs, most notably ai and au. i assume here that the forms marked in academic notation as aí and aú are short /ɛ ɔ/, while the forms ái áu and ai au are both /ɛː ɔː/. ddj could theoretically be any palatal consonant, but almost certainly a geminated one; i assume /ɟː/. i also assume that gothic w representing a borrowed greek /y/ was pronounced /i/. lastly, for the voiced series, i assume a voiced fricative pronunciation V_V, V_j, V_w; a voiceless fricative pronunciation word-finally or before a voiceless consonant; and voiced plosive pronunciations elsewhere

phase 1 (from old gothic to middle gothic)

More: show

1. intial /ɸl ɸr/ > /þl þr/ (this process had taken place in certain gothic words that begin with fl- in proto-germanic; this change expands it to all fl-/fr- words)
2. vowels are rounded after the labial consonants/:w kʷ gʷ hʷ/: /a(ː) ɛ(ː)/ > /ɔ(ː)/, /eː/ > /oː/; i(ː) > u(ː); /iu/ > /uː/
3. /hʷ/ becomes /ɸ/
4. /h/ becomes /x/ if _#, V_C, C_C (assuming this allophonic variation didn't already exist)
5. word-initial /h/ becomes /ɸ/ when followed by a consonant
6. /ngʷ/ > /mb/, /nkʷ/ > /mp/ (inspired by Romanian)
7. /kʷ/ > /k/
8. geminates collapse, causing a preceding short vowel to lengthen
9. /ɟ/ (which collapsed in the previous rule) > /ʤ/
10. /iu/ > /yː/
11. /w/ > /v/
12. [ɸ β] > [f v]
13. palatalization: non-labial consonants palatalize before /yː iː eː i j/
    • /t k þ/ > /ʧ ʦ ʃ/
    • [x h] > /ç/
    • [d ð g ɣ] > /ʤ ʒ ʣ ʝ/
    • /s z n/ > /ʃ ʒ ɲ/
    • /lj rj/ > /ʝ/; otherwise /l r/ > /j/
    • these changes happen to whole clusters (e.g. /styː/ > /ʃʧyː/), except that the new affricates /ʦ ʣ/ block any further regressive palatalization: /skyː/ > /sʦyː/ not /*ʃʧyː/
    • after the above changes take place, /j/ is deleted after a palatal
    • j is deleted after a palatal
14. unstressed short /i u/ become /ɛ ɔ/
15. stressed long vowels become diphthongs:
    • /yː iː eː/ > /ja/
    • /uː oː/ > /wa/
    • /ɛː ɔː aː/ > /ɛʊ ɔʊ aʊ/
    • if the resulting /ja/ or /wa is preceded by a liquid (/l r j/), that liquid disappears
16. unstressed long vowels shorten
17. final unstressed vowels delete if they are preceded by at least one unstressed syllable and are immediately preceded by a consonant
18. consonant clusters with mixed voicing level the voicing:
    • an unvoiced consonant adjacent (on either side) to a nasal or liquid, including /j w/ resulting from earlier long vowels, becomes voiced
    • otherwise, the voicing assimilates to the voicing of the final consonant in the cluster
19. /u/ is inserted before /l r m n ɲ/ if said consonant is found between two consonants, or is word-final following a consonant

for those keeping track, this leads to the consonant inventory /m n p b t d k g f v þ ð s z x ɣ h l j r w ʧ ʦ ʃ ç ʤ ʣ ʒ ʝ ɲ/, with /w/ only appearing before /a/. we also have an unusual situation with the vowels: the stressed vowels are /a ɛ i ɔ u aʊ ɛʊ ɔʊ/, but the unstressed vowels are /a ɛ e i ɔ o u y/, with three different vowels only appearing in unstressed positions. this will be remedied in the next stage

phase 2 (from middle gothic to early modern gothic)

More: show

1. [h] is lost
2. /aʊ ɛʊ ɔʊ/ become /aw ɛw ɔw/
3. after fricatives, affricates become stops: /ʧ ʦ/ > /t/, /ʤ ʣ/ > /d/ (e.g. /ʃʧ/ > /ʃt/)
4. otherwise, affricates become fricatives: /ʧ ʤ ʦ ʣ/ > /ʃ ʒ s z/
5. /ç/ merges into /ʃ/ and /ʝ/ merges into /ʒ/
6. /j/ is deleted after a palatal consonant (this deletes many instances of /j/ created in rule 15 above)
7. in word-initial clusters consisting of any consonant followed by a nasal consonant, the nasal consonant loses its nasalization:
   • /m/ > /w/
   • /n/ > /l/ after an alveolar (but not a dental), /r/ otherwise
   • /ɲ/ deletes after a palatal, becomes /j/ otherwise
   • for clusters found in inherited words, this works out to: zm gn vn zn ʒɲ zɲ vɲ > zw gr vr zl ʒ zj vj
8. nasalization: vowels and some consonants become nasal
   • vowels are nasalized if followed by a nasal consonant that is not immediately followed by a vowel
   • most non-nasal consonants that follow a vowel and precede a nasal consonant become nasals (i.e. VCN > VNN):
     • /p b f v w/ > /m/
     • /t d þ ð s z l r/ > /n/
     • /k g x ɣ/ > /ŋ/
     • /ʃ ʒ j/ do not change
     • with this stage in the sound changes, /ŋ/ is effectively phonemicized; all instances of /n/ before a velar become /ŋ/ (its already existing allophone in this position)
   • vowels are nasalized (again) if followed by a geminated nasal consonant
   • /ɲ/ > /j/
   • nasal consonants are deleted after nasal vowels
   • /m n ŋ/ > /b d g/ after a nasal consonant
9. low or back vowels become high front vowels after /j/: /ja jɔ jo ju jã jɔ̃ jõ jũ/ > /jɛ jɛ je ji jɛ̃ jɛ̃ jẽ jĩ/
10. front vowels become rounded back vowels after /w/: /wa wɛ we wi wy wã wɛ̃ wẽ wĩ wỹ/ > /wɔ wɔ wo wu wu wɔ̃ wɔ̃ wõ wũ wũ/
11. nasal vowels lower: /ã ɛ̃ ẽ ĩ ɔ̃ õ ũ ỹ/ > /ɒ̃ æ̃ ɛ̃ ẽ ɒ̃ ɔ̃ õ ø̃/
12. stops are deleted before fricatives, and fricatives are deleted before other fricatives
13. intervocalic consonant lenition: in the sequence V₁CV₂,
    • if the word stress is on V₁, voiceless consonants become voiced
    • if the word stress is on V₂, no change takes place
    • if both V₁ and V₂ are unstressed:
      • voiced fricatives become approximants: /v ɣ ð z ʒ/ > /w w l r j/
      • voiced stops and voiceless fricatives both become voiced fricatives
      • voiceless stops become voiced stops
      • note that this one has a tendency to be reversed through paradigm leveling
14. unstressed vowels reduce:
    • /e i y/ > /ɨ/
    • /a ɛ/ > /ə/
    • /ɔ o u/ > /ɵ/
    • /ẽ ø̃/ > /ɘ̃/
    • /æ̃ ɛ̃/ > /ɜ̃/
    • /ɒ̃ ɔ̃ õ/ > /ɞ̃/
15. in the position VC_CV, an unstressed oral vowel is deleted, and an unstressed nasal vowel becomes an oral vowel: /ɘ̃ ɜ̃ ɞ̃/ > /e ɛ ɔ/ (in a long enough word, this can happen multiple times, essentially going from left to right: */ásekuzino/ > /áskuzno/)
16. /ɣ/ > /h/
17. /w/ > /f/ before voiceless consonant, /v/ otherwise
18. any geminated consonants degeminate (without compensatory vowel lengthening this time)
19. remaining reduced vowels harden:
    • /ə ɨ ɵ/ > /ɛ ɪ ʊ/
    • /ɘ̃ ɜ̃ ɞ̃/ > /ɪ̃ ã õ/
20. stress shifts to the penultimate syllable
21. nasal vowels lose nasalization (but retain vowel quality)
22. vowels merge:
    • each pair of /æ a/, /ɛ e/, /ɪ i/, /ɔ o/, /ʊ u/ merges, with the former of each pair becoming the unstressed allophone and the latter becoming stressed
    • /ɒ/ merges into /o/ ([ɔ]) if unstressed, /a/ if stressed

the new consonant inventory is /m n ŋ p b t d k g f v þ ð s z x h l j r ʃ ʒ/. the vowels have greatly simplified to /a e i o u/, with the unstressed allophones [æ ɛ ɪ ɔ ʊ]

phase 3 (from early modern gothic to modern gothic)

More: show

1. /l/ becomes /u/ after a vowel unless followed by another vowel (/ul/ > /u/, not /*uː/)
2. velar (and glottal) consonants palatalize following /i/ (including [ɪ]) or /j/:
   • /ŋ/ > /ɲ/
   • /x h/ > /ç ʝ/
   • /k g/ > /ʧ ʤ/
   • this only affects the immediately following consonant, not entire clusters
3. diphthongs ending in /j/ or /u/ shift:
   • /aj/ > /e/
   • /ej ij/ > /i/
   • /oj uj/ > /wi/
   • /au eu/ > /o/
   • /iu/ > /ju/
   • /ou/ > /wu/
4. vowel hiatus simplifies:
   • an unstressed vowel is deleted after a stressed vowel
   • before a stressed vowel, or between two vowels, an unstressed /a o u/ > /w/ and unstressed /e i/ > /j/
   • unstressed /aa ee ii oo uu/ collapse into /a e i o u/
   • certain combinations of VCV treated as hiatus and collapse (this takes place after the above hiatus changes however):
     • awa, aha > a
     • owo, oho > o
     • uwu, uhu > u
     • eje, ehe > e
     • iji, ihi > i
5. word-initial stressed vowels develop glides in the onset: /a o u/ take /w/ (e.g. /a/ > /wa/), while /e i/ take /j/
6. a consonant followed by an unstressed vowel followed by the same consonant is reduced to just the consonant (e.g. /kik/ > /k/, /mum/ > /m/, etc.)
7. an unstressed vowel is deleted in the position #K_CV where K is a stop or fricative
8. a stop following a nasal consonant assimilates to the nasal's POA: (e.g. amda > amba)
   • after /m/: voiceless /p/, voiced /b/
   • after /n/: voiceless /t/, voiced /d/
   • after /ɲ/: voiceless /ʧ/, voiced /ʤ/
   • after /ŋ/: voiceless /k/, voiced /g/
9. consonant cluster simplification:
   • homorganic voiced stop inserted after a nasal consonant when followed by /l r w/: /mr/ > /mbr/, /nw/ > /ndw/, etc.
   • fricative + fricative becomes stop + fricative
   • a nasal consonant is deleted before another nasal
   • a stop is deleted when following another consonant and preceding another stop
   • a stop is deleted following a stop preceded by a vowel and preceding another consonant or word-final (VK_C or VK_# where K = stop)
   • word-initial stop + stop becomes fricative + stop
   • if a nasal or liquid consonant is between two other consonants, a /u/ ([ʊ]) is inserted before it
   • in a sequence of three obstruents (stops/fricatives), a /u/ ([ʊ]) is inserted before the third
10. /h/ becomes /x/ before a consonant or word-finally
11. an unstressed vowel in the final syllable is reduced:
    • /a o u/ > [ə]
    • /e i/ > [ɨ]
12. stressed vowels in open syllables shift:
    • /a/ > /au/
    • /e/ > /a/
    • /i/ > /e/
    • /o/ > /ɑ/
    • /u/ > /o/
    • /a/ does not shift to /au/ if preceded by /j/ or /w/
    • this change is originally allophonic but becomes phonemic through paradigm leveling
13. voicing is leveled in consonant clusters to the final obstruent in the cluster (unlike the leveling in phase one, liquids and nasals do not affect this)
14. word-final obstruents and clusters are devoiced

this leaves us with our final phonology. the consonants are /m n ŋ p b t d k g f v þ ð s z x ɣ h l j r w ʃ ʒ ç ʝ c ɟ ʧ ʤ/. the stressed vowels are /a ɑ e i o u au/, the unstressed vowels in non-final syllables are [æ ɛ ɪ ɔ ʊ], and the unstressed vowels in final syllables are [ə ɨ]

[Emily]

I hope I said that already when you mentioned that in those other threads, but I really like your project.

thank you!

You're very cautious here; AFAIK most people who accept the source call it Crimean Gothic and assume that it's a descendant of Wulfila's Gothic.

the sources i've seen have all classified it as east germanic, but conclude that it was unlikely to be a direct descendant of wulfila's gothic

Influence on phonology is actually stronger in language change situations than in language contact situations; language contact predominantly leads to similar semantic developments (like similar division of semantic fields, calqued expressions, etc.) and similarities wrt to syntax and gramatical categories (e.g., Yiddish having question-forming perticles like Slavic, or the similarities in the Balkan Sprachbund - subordinated clauses instead of infinitives, development of a definite article, etc.).

interesting point! really the biggest slavic influence on the phonology is the palatalization in phase 1. it seems like such a distinctive change that i can see it having an influence on a neighboring minority language. (and at any rate, it's certainly feasible for it to have independently developed as well)

[WeepingElf]

This is a very nice and promising project. However, I'd switch the palatalization results such that the dentals become alveolars and the velars become postalveolars. The way you have them now, their trajectories in the phonetic space cross, which seems unlikely to me.

[alice]

Nice!!! This brings back some memories of when I used Gothic as the ancestor of one of my conlangs; I notice that you've done many of the things I did.

[Starbeam]

I absolutely love these prospects! Makes me wanna work on my Gothlang(s).

I wonder, how much of this is based of your personal analyses of the language's phonology, how much is your own artistic invention? How much influence are you willing to use from neighboring/ closely related languages? Likewise with other parts of the language, but i don't see much beyond the phonology.

Would you use word-final devoicing? I always wonder why, of the Slavic languages, only Ukrainian and all but one of the Southwest Slavic languages lacked it. Woudl you mind showing some example passage to demonstrate the sound changes? No worry if no, i get i'm blabbing haha.

Also, i appreciate you using þorn instead of theta in the IPA. Not the official symbol, sure, but i like it.

[Emily]

This is a very nice and promising project. However, I'd switch the palatalization results such that the dentals become alveolars and the velars become postalveolars. The way you have them now, their trajectories in the phonetic space cross, which seems unlikely to me.

interesting point, i hadn't thought of that. i'll keep it in mind if i do a revision of the phonology at some point

I wonder, how much of this is based of your personal analyses of the language's phonology, how much is your own artistic invention? How much influence are you willing to use from neighboring/ closely related languages? Likewise with other parts of the language, but i don't see much beyond the phonology.

well, my analysis of the language's phonology really mostly comes into play in the first phase; after that it's gotten far enough away from the original phonology that it's basically pure invention from that point

Would you use word-final devoicing?

it goes away by the end of phase 1, but comes in at the tail end of phase 3

Also, i appreciate you using þorn instead of theta in the IPA. Not the official symbol, sure, but i like it.

that's mostly because there's no alt code for lowercase theta lol

[Moose-tache]

2.12: stops deleted before fricatives
Is this only within morphemes, or across? Do we end up with words like "das" for "day?"

[Emily]

nouns (phase 1—middle gothic)
i don't want to spend too much time going over irl gothic (sometimes called "bible gothic", referred to in this project as "old gothic"); wikipedia has a decent overview of noun declension, and the university of texas has a great introductory text on gothic as a whole. in a nutshell, gothic nouns had several "strong" declensions and fewer "weak" ones, the main difference being the weak declensions usually had /n/ in the oblique endings. there were also a couple of minor declensions that didn't necessarily fall into either category. all declined for nominative, accusative, dative, and genitive cases (as well as vocative, but those endings almost always matched those of other cases), had singular and plural number, and like many other IE languages had some declensions that were gender-specific and some that were not

my process, predictably, began with running declensions through the SCA and seeing what the results were. primarily i was looking for which declensions seemed similar enough that speakers would start lumping them together. in phase 1, however, something that immediately jumped out was palatalization. the genitive singular and plural endings for a-stem masculine and neuter nouns (by far the largest group of nouns in gothic) both palatalized the final consonant(s) of the stem due to the presence of front vowels in the genitive endings; the same was true of one or both of the genitive endings in several other declensions, as well as in much of the adjective declension. as an example:

• gothic skōhs (nom sg), skōhis (gen sg) "shoe" -> ph1 /zgwáxs/, /zgwáçɛs/
• land, landis "land" -> /lánd/, /láɲʤɛs/

given the distinctiveness of this phenomenon and the fact that it took place in the most common noun and adjective declensions, i decided that palatalizing the final consonant of the stem would be a distinguishing feature of the genitive case, and i spread it to other declensions where it had not organically arisen. in my documentation, i used the mark /-ʲ/ to indicate this process, a practice i will carry over into this thread. there was a second source for palatalization in some declensions, however: a following /j/. this appeared in different endings (most prominently the nominative plural of the so-called u-stems), and it also affected stems that didn't end in consonants affected by the previous palatalization by inserting a /j/ after them. because /j/ disappears in the sound changes after palatalizing consonants, this goes from simply being "an ending that starts with /j/" to a second form of palatalization (recall also the different effect on /r l/). so this form is written with /-J/. (also any time i use the term "final consonant", it's shorthand for the last consonant or consonant cluster of a word's stem before any endings)

lastly, some stems' final consonants were labialized in old gothic (ending in w q gw ƕ). in the sound changes, the vowels in the declension endings were often colored; this didn't result in enough words to form whole new declensions, but they ended up merging with other declensions (mostly the u-stems) that they bore similarities to

the general evolution from old gothic to middle gothic declensions:

More: show

old gothic declension	old gothic final consonant	middle gothic declension
a-stem masc	any	strong 1
a-stem neut	/w hʷ kʷ gʷ/	strong 5
a-stem neut	other	strong 6
ja-stem masc (long)	/w hʷ kʷ gʷ/	strong 3
ja-stem masc (long)	other	strong 1
ja-stem masc (short)	any	strong 1
ja-stem neut	any	strong 6
ō-stem fem	/w hʷ kʷ gʷ/	strong 5
ō-stem fem	other	strong 4
jō-stem fem	any	strong 4
i-stem masc/fem	/w hʷ kʷ gʷ/	strong 1
i-stem masc/fem	other	strong 2
i-stem fem (-eins)	any	strong 1
u-stem masc/fem	any	strong 3
u-stem neut	any	strong 5
an-stem masc (male human/animal referent)	/w hʷ kʷ gʷ/	weak 1
an-stem masc (other)/neut	/w hʷ kʷ gʷ/	weak 3
an-stem masc	other	weak 1
an-stem neut	other	weak 2
ōn-stem fem	any	weak 3
in-stem fem	/w hʷ kʷ gʷ/	weak 3
in-stem fem	other	weak 4
-nd stem	any	minor nd
r-stem	any	minor r
other consonant stem	any	consonant

the middle gothic noun declensions:

More: show

Strong 1

Case	Sg	Pl
Nom	s/z*	os
Acc	–**	anz
Dat	a	am
Gen	ʲɛs	ʲe

Strong 2

Case	Sg	Pl
Nom	s/z*	ʲis
Acc	–**	ʲɛnz
Dat	a	ʲɛm
Gen	ʲɛs	ʲe

Strong 3

Case	Sg	Pl
Nom	ɔs	Jɔs
Acc	ɔ	ɔnz
Dat	ɔ	ɔm
Gen	ʲɔs	ʲɛv

Strong 4

Case	Sg	Pl
Nom	a	os
Acc	a	os
Dat	ɛ	om
Gen	ʲos	ʲo

Strong 5

Case	Sg	Pl
Nom	ɔ	os
Acc	ɔ	os
Dat	ɔ	om
Gen	ʲos	ʲo

Strong 6

Case	Sg	Pl
Nom	–**	a
Acc	–**	a
Dat	a	am
Gen	ʲɛs	ʲe

Weak 1

Case	Sg	Pl
Nom	a	anz
Acc	an	anz
Dat	ʲɛ	am
Gen	ʲɛnz	ʲaɲ

Weak 2

Case	Sg	Pl
Nom	o	on
Acc	o	on
Dat	ʲɛ	am
Gen	ʲɛnz	ʲaɲ

Weak 3

Case	Sg	Pl
Nom	o	onz
Acc	on	onz
Dat	on	om
Gen	ʲonz	ʲon

Weak 4

Case	Sg	Pl
Nom	i	inz
Acc	in	inz
Dat	in	im
Gen	inz	in

Consonant

Case	Sg	Pl
Nom	s/z	s/z
Acc	–	s/z
Dat	–	ɔm
Gen	ʲɛs	ʲ–

Minor Nd

Case	Sg	Pl
Nom	z	z
Acc	–	z
Dat	–	am
Gen	ʲɛs	ʲ–

Minor R

Case	Sg	Pl
Nom	ar	ʲɔs
Acc	ar	rɔnz
Dat	ur	rɔm
Gen	ʲurz	ʲe

* /ɛs/ after a palatal, a cluster ending in /j/ or /w/, or (inconsistently) after /s/ or /z/
* /ɛ/ after a palatal or a cluster ending in /j/ or /w/

(sorry, i don't know if it's possible to do colspans or different colored cells [other than header cells] -- the board's bbcode documentation doesn't mention tables at all! )

the vocative (almost) always matches the accusative in the singular, and the nominative in the plural

examples with real words:

More: show

Strong 1 — bards "beard"

Case	Sg	Pl
Nom	bárdz	bárdos
Acc	bárd	bárdanz
Dat	bárda	bárdam
Gen	báʤɛs	báʤe

Strong 2 — laiks "dance, ball"

Case	Sg	Pl
Nom	lɛ́ʊks	lɛ́ʊʦis
Acc	lɛ́ʊk	lɛ́ʊʦɛnz
Dat	lɛ́ʊka	lɛ́ʊʦɛm
Gen	lɛ́ʊʦɛs	lɛ́ʊʦe

Strong 3 — haimaus "village"

Case	Sg	Pl
Nom	hɛ́ʊmɔs	hɛ́ʊmjɔs
Acc	hɛ́ʊmɔ	hɛ́ʊmɔnz
Dat	hɛ́ʊmɔ	hɛ́ʊmɔm
Gen	hɛ́ʊmɔs	hɛ́ʊmɛv

Strong 4 — gilþa "sickle"

Case	Sg	Pl
Nom	ʣílða	ʣílðos
Acc	ʣílða	ʣílðos
Dat	ʣílðɛ	ʣílðom
Gen	ʣíʒos	ʣíʒo

Strong 5 — eiþau "fermented drink"

Case	Sg	Pl
Nom	jáþɔ	jáþos
Acc	jáþɔ	jáþos
Dat	jáþɔ	jáþom
Gen	jáʃos	jáʃo

Strong 6 — boþ "blood"

Case	Sg	Pl
Nom	bwáþ	bwáþa
Acc	bwáþ	bwáþa
Dat	bwáþa	bwáþam
Gen	bwáʃɛs	bwáʃe

Weak 1 — garazna "neighbor" (masc)

Case	Sg	Pl
Nom	garázna	garáznanz
Acc	garáznan	garáznanz
Dat	garáʒɲɛ	garáznam
Gen	garáʒɲɛnz	garáʒɲaɲ

Weak 2 — augo "eye"

Case	Sg	Pl
Nom	ɔ́ʊɣo	ɔ́ʊɣon
Acc	ɔ́ʊɣo	ɔ́ʊɣon
Dat	ɔ́ʊʝɛ	ɔ́ʊɣam
Gen	ɔ́ʊʝɛnz	ɔ́ʊʝaɲ

Weak 3 — garazno "neighbor" (fem)

Case	Sg	Pl
Nom	garázno	garáznonz
Acc	garáznon	garáznonz
Dat	garáznon	garáznom
Gen	garáʒɲonz	garáʒɲon

Weak 4 — aiþi "mother"

Case	Sg	Pl
Nom	ɛ́ʊʃi	ɛ́ʊʃinz
Acc	ɛ́ʊʃin	ɛ́ʊʃinz
Dat	ɛ́ʊʃin	ɛ́ʊʃim
Gen	ɛ́ʊʃinz	ɛ́ʊʃin

Consonant — gaits "goat"

Case	Sg	Pl
Nom	gɛ́ʊts	gɛ́ʊts
Acc	gɛ́ʊt	gɛ́ʊts
Dat	gɛ́ʊt	gɛ́ʊtɔm
Gen	gɛ́ʊʧɛs	gɛ́ʊʧ

Minor Nd — þijonds "friend"

Case	Sg	Pl
Nom	ʃíjondz	ʃíjondz
Acc	ʃíjond	ʃíjondz
Dat	ʃíjond	ʃíjondam
Gen	ʃíjoɲʤɛs	ʃíjoɲʤ

Minor R — swustar "sister"

Case	Sg	Pl
Nom	zvústar	zvúʃʧɔs
Acc	zvústar	zvústrɔnz
Dat	zvústur	zvústrɔm
Gen	zvúʃʧurz	zvúʃʧe

[Emily]

2.12: stops deleted before fricatives
Is this only within morphemes, or across? Do we end up with words like "das" for "day?"

generally within words, although when the fricative in question is the NomSg -s/-z, the deleted sound is usually put back in by analogy with the rest of the inflection. (also, in "dag-", the "g" is a fricative, though of course those are subject to the same rule)

[alice]

Did you forget an /r/ in the gen pl of "beard"?

[Emily]

Did you forget an /r/ in the gen pl of "beard"?

no-ish. /r/ and /l/ palatalize to /j/, which i've been deleting adjacent to another palatalized consonant. but looking at the rules (and the SCA rules as well) i see that i only actually have that deletion if it comes after the consonant but not before. so i think that's something i need to rethink what i want to do

[foxcatdog]

Where's the dual number?

[alice]

Where's the dual number?

It didn't survive into the Gothic noninal declension, only in the first and second person in verbs and some pronouns.

[foxcatdog]

Where's the dual number?

It didn't survive into the Gothic noninal declension, only in the first and second person in verbs and some pronouns.

I know but i wanna see it.

[Emily]

it'll turn back up! honestly i expanded the use of the dual so that it exists throughout the entire morphological system by the modern gothic period. is it historically justifiable that this would be the case? not really. but if it were to happen, i think the way i did it makes the most sense. will touch on it in future posts about nouns, adjectives, pronouns, verbs

[Emily]

speaking of,

pronouns (phase 1—middle gothic)
the middle gothic pronouns descend of course from the old gothic pronouns. the biggest changes other than the basic sound changes outlined above are:

• the personal pronouns develop two sets of descendants: those based on stressed versions, and those based on unstressed. the stressed versions serve as the basis for the spellings, but are generally only used for emphasis or in isolation; the unstressed forms are by far the most commonly used
• unstressed accusative and dative pronouns can also attach to the verb phrase as a clitic prefix, often with an accompanying change of form (pronouns listed below with a vowel in parentheses add said vowel when attaching to a word beginning with a consonant; other pronouns will change the voicing of their final consonant to match that of the first consonant in the following word)
• 3rd-person dual pronouns are introduced. they began as simply third person plural pronouns followed by the appropriate inflection of twái "two" ("I gave it to them two"), but in fairly short order, presumably by analogy with the 1st/2nd person dual, became mandatory, and were pronounced as one word and worn down as such. scribes, however, clearly felt they didn't "count" as distinct pronouns, as they are meticulously written in materials of the period as two separate words
• the 1du acc and dat pronouns, which in old gothic were uns and unsis but frequently interchanged with each other, settle down into the short form (/únz/ stressed, /ɔnz/ unstressed)
• the demonstrative sa, þata, sō, which in old gothic seemed to be in the early stages of turning into a definite article, has by middle gothic actually split into 1) demonstrative pronouns descending from the stressed form (sá, þát(a), zwá "this"), and 2) definite articles descending from the unstressed form (sa, þat, so "the"), though the difference isn't very stark in this stage (and they're both basically spelled the same). it should be noted that the former acts strictly as a pronoun and is not found in attributive/adjectival position
• the middle gothic relative pronouns descend from unstressed versions
• the feminine interrogative pronoun ("who/what? (f)") is lost
• the old gothic demonstrative jáins "that there, yon" has split descendants similar to sa: the stressed forms become the demonstrative pronoun "that", while the unstressed forms become a demonstrative adjective that combines the meanings "this" and "that"
• there are two competing indefinite pronouns: one descending from unstressed áins "one", and the other from unstressed sums "some, a certain". the meanings don't completely overlap but they also grow fuzzier over time

personal pronouns
listed in order nom, acc, dat, gen

More: show

• stressed
  • 1sg: ík, mík, mís, mjána — ik, mik, mis, meina
  • 2sg: þú, þúk, þús, þána — þu, þuk, þus, þana
  • 3sg m: ís, ína, jáma, ís — is, ina, eima, is
  • 3sg n: íta, íta, jáma, ís — ita, ita, eima, is
  • 3sg f: ʃí, íja, ízɛ, ízos — si, ija, izai, izos
  • 1du: vút, únʣɛs, únʣɛs, úngar — wut, ungis/uggis, uggar
  • 2du: jút, ímbɔs, ímbɔs, ímbɔr — jut, imbaus, imbaur
  • 3du m: jázdvɔ, ínzdvɔnz, ímdvɔm, íʒedvɔʤ — eis dwau, ins dwauns, im dwaum, ize dwauddj
  • 3du n: íjadvɔ, íjadvɔ, ímdvɔm, íʒedvɔʤ — ija dwau, im dwaum, ize dwauddj
  • 3du f: íjozdvos, íjozdvos, ímdvɔm, ízodvɔʤ — ijos dwos, im dwaum, izo dwauddj
  • 1pl: vwás, únz, únz, únzar — wos, uns, unsar
  • 2pl: jús, ízvɔs, ízvɔs, ízvɔr — jus, izwaus, izwaur
  • 3pl m: jás, ínz, ím, íʒe — eis, ins, im, ize
  • 3pl n: íja, íja, ím, íʒe — ija, im, ize
  • 3pl f: íjos, íjos, ím, ízo — ijos, im, izo
  • refl: —, ʃík, ʃís, ʒjána — sik, sis, zeina
• unstressed/clitic
  • 1sg: ɛk, mɛk, mɛs, min
  • 2sg: þɔ, þɔk, þɔs, þin
  • 3sg m: ɛs, ɛn(a), im(a), ɛs
  • 3sg n: ɛt(a), ɛt(a), im(a), ɛs
  • 3sg f: ʃɛ, ɛj(a), ɛz(ɛ), ɛzos
  • 1du: vɔt, ɔnʣɛs, ɔnʣɛs, ɔngar
  • 2du: jɔt, ɛmbɔs, ɛmbɔs, ɛmbɔr
  • 3du m: izv, ɛnzvɔnz, ɛmvɔm, ɛʒevɔʤ
  • 3du n: ɛjav(ɔ), ɛjav(ɔ), ɛmvɔm, ɛʒevɔʤ
  • 3du f: ɛjozvos, ɛjozvos, ɛmvɔm, ɛzovɔʤ
  • 1pl: vus, ɔnz, ɔnz, ɔnzar
  • 2pl: jɔs, ɛzvɔs, ɛzvɔs, ɛzvɔr
  • 3pl m: is, ɛnz, ɛm, ɛʒ
  • 3pl n: ɛj(a), ɛj(a), ɛm, ɛʒ
  • 3pl f: ɛjos, ɛjos, ɛm, ɛz
  • refl: ʃɛk, ʃɛs, ʃin

in this period the 3sg masc/neut genitive is sometimes replaced with the reflexive genitive to avoid the ambiguity with 3sg nominative; by early modern gothic this replacement is complete. conversely, the 2sg genitive has reintroduced the nonpalatalized /þ-/ to avoid confusion with the reflexive. the genitives here are pronouns only; possessive adjectives are related but decline as regular strong adjectives: 1sg /minz/ mins, 1du /ɔngarz/ auggar, 1pl /ɔnzarz/ aunsar, 2sg /þinz/ þins, 2du /ɛmbɔrz/ aimbaur, 2pl /ɛzvɔrz/ aizwaur, reflex/3rd (all numbers/genders) /ʃinz/ sins

interrogative pronouns

• "who": fɔ́s fɔ́na fɔ́ʊma fús — ƕaus, ƕauna, ƕauma, ƕus
• "what": fɔ́ fɔ́ fɔ́ʊma fús — ƕau, ƕauma, ƕus
• "where": fɔ́r — ƕaur
• "to where": fɔ́ʤe — ƕauddje
• "from where": fɔ́ðro — ƕaudro
• "when": fɔ́n — ƕaun
• "why": dɔvwá — dauwo
• "how, by what means": vwá — wo
• "how, in what manner": fɔ́ʊvɔ — ƕauwau

these are the only words in middle gothic that retain the letter ƕ, and even here it is lost when old gothic /hʷ/ developed into middle gothic /v/ instead of /f/

demonstratives
the demonstrative pronouns are split into proximal ("this, these") and distal ("that, those"). the demonstrative adjectives make no such distinction. note that only the masculine gender makes any distinction between nominative and accusative forms

More: show

• proximal demonstrative pronouns
  • m sg: sá, þána, þáʊma, þís — sa, þana, þama, þis
  • n sg: þát(a)*, þáʊma, þís — þat(a), þama, þis
  • f sg: zwá, þízɛ, þízos — so, þizai, þizos
  • m du: þɛ́ʊdvɔ, þánzdvɔnz, þɛ́ʊmdvɔm, þíʒedvɔʤ — þai dwau, þans dwauns, þaim dwaum, þize dwauddj
  • n du: ðwádvɔ, þɛ́ʊmdvɔm, þíʒedvɔʤ — þo dwau, þaim dwaum, þize dwauddj
  • f du: ðwázdvos, þɛ́ʊmdvɔm, þízodvɔʤ — þos dwos, þaim dwaum, þizo dwauddj
  • m pl: þɛ́ʊ, þánz, þɛ́ʊm, þíʒe — þai, þans, þaim, þize
  • n pl: ðwá, þɛ́ʊm, þíʒe — þo, þaim, þize
  • f pl: ðwás, þɛ́ʊm, þízo — þos, þaim, þizo
• distal demonstrative pronouns
  • m sg: jɛ́ʊnz, jɛ́ʊnan, jɛ́ʊnam, jɛ́ʊɲɛs — jains, jainan, jainam, jainis
  • n sg: jɛ́ʊnat, jɛ́ʊnam, jɛ́ʊɲɛs — jainat, jainam, jainis
  • f sg: jɛ́ʊna, jɛ́ʊnɛ, jɛ́ʊnɛzos — jaina, jainai, jainaizos
  • m du: jɛ́ʊnɛdv, jɛ́ʊnanzdvɔnz, jɛ́ʊnɛmdvɔm, jɛ́ʊnɛʒedvɔʤ — jainai dwau, jainans dwauns, jainaim dwaum, jainaizj dwauddj
  • n du: jɛ́ʊnadv, jɛ́ʊnɛmdvɔm, jɛ́ʊnɛʒedvɔʤ — jaina dwau, jainaim dwaum, jainaizj dwauddj
  • f du: jɛ́ʊnozdvos, jɛ́ʊnɛmdvɔm, jɛ́ʊnɛzodvɔʤ — jainos dwos, jainaim dwaum, jainaiz dwauddj
  • m pl: jɛ́ʊnɛ, jɛ́ʊnanz, jɛ́ʊnɛm, jɛ́ʊnɛʒ — jainai, jainans, jainaim, jainaizj
  • n pl: jɛ́ʊna, jɛ́ʊnɛm, jɛ́ʊnɛʒ — jaina, jainaim, jainaizj
  • f pl: jɛ́ʊnos, jɛ́ʊnɛm, jɛ́ʊnɛz — jainos, jainaim, jainaiz
• demonstrative adjectives
  • m sg: jɛnz, jɛnan, jɛnam, jɛɲɛs
  • n sg: jɛn, jɛnam, jɛɲɛs
  • f sg: jɛna, jɛnɛ, jɛɲɛz
  • m du: jɛnɛv, jɛnzvɔnz, jɛmvɔm, jɛɲʒvɔʤ
  • n du: jɛnav, jɛmvɔm, jɛɲʒvɔʤ
  • f du: jɛnzvos, jɛmvɔm, jɛnzvɔʤ
  • m pl: jɛnɛ, jɛnanz, jɛnɛm, jɛɲɛʒ
  • n pl: jɛna, jɛnɛm, jɛɲɛʒ
  • f pl: jɛnos, jɛnɛm, jɛɲɛz

* /þáta/ before a consonant, /þát/ before a vowel
the demonstrative adjectives are spelled the same as their respective distal pronouns

relative pronouns
the relative pronouns at this stage are essentially just the definite articles with either /i/ appended to the end (after a vowel) or, if it ends in a consonant, the final consonant being palatalized. in addition to the familiar 3rd-person relative pronouns, a set of 1st- and 2nd-person pronouns remains from earlier stages. these were rare even in old gothic, however, and in middle gothic they basically exist only as an archaism found in bibles and no other text type, and even in bibles their use is haphazard

definite article
the definite articles likewise do not distinguish nominative and accusative in the neuter or feminine genders

More: show

• m sg: sa, þan, þam, þɛs — sa, þan, þama, þis
• n sg: þat, þam, þɛs — þat, þama, þis
• f sg: so, þɛz, þɛs — so, þizai, þizos
• m du: þɛv, þanzvɔnz, þɛmvɔm, þɛʒvɔʤ — þai dwau, þans dwauns, þaim dwaum, þize dwauddj
• n du: þov, þɛmvɔm, þɛʒvɔʤ — þo dwau, þaim dwaum, þize dwauddj
• f du: þozvos, þɛmvɔm, þɛzvɔʤ — þos dwos, þaim dwaum, þizo dwauddj
• m pl: þɛ, þanz, þɛm, þɛʒ — þai, þans, þaim, þize
• n pl: þo, þɛm, þɛʒ — þo, þaim, þize
• f pl: þos, þɛm, þɛz — þos, þaim, þizo

indefinite articles
lastly, below is the indefinite article ains; the competing indefinite article saums declines exactly the same except with the stem /sɔm-/ instead of /ɛn-/

More: show

• m sg: ɛnz, ɛnan, ɛnam, ɛɲɛs — ains, ainan, ainam, ainis
• n sg: ɛn, ɛnam, ɛɲɛs — ain, ainam, ainis
• f sg: ɛn, ɛn, ɛnzos — aina, ainai, ainaizos
• m du: ɛnɛv, ɛnzvɔnz, ɛmvɔm, ɛɲʒvɔʤ — ainai dwau, ainans dwauns, ainaim dwaum, ainaizj dwauddj
• n du: ɛnav, ɛmvɔm, ɛɲʒvɔʤ — aina dwau, ainaim dwaum, ainaizj dwauddj
• f du: ɛnzvos, ɛmvɔm, ɛnzvɔʤ — ainos dwos, ainaim dwaum, ainaiz dwauddj
• m pl: ɛn, ɛnanz, ɛnɛm, ɛnɛʒ — ainai, ainans, ainaim, ainaizj
• n pl: ɛn, ɛnɛm, ɛnɛʒ — aina, ainaim, ainaizj
• f pl: ɛnos, ɛnɛm, ɛnɛz — ainos, ainaim, ainaiz

for obvious reasons, the dual forms of ains were much slower to be adopted than those of saums

anyway, sorry for those who are not interested in inflection tables, these first several posts are pretty morphology-heavy

[Moose-tache]

Marvelous. More please.

I notice the proximal demonstrative, which does not have an attributive form, does have a genitive. How and when do Middle Gothic speakers use þis and jainis?

Using sin for is replaces one ambiguity with another. Do Middle Gothic speakers ever get confused by sentences like “Jack showed John what was left of his book?”

Making all the pronouns reduced when not isolated or emphasized is pretty bold. English, for example, reduces pronouns after verbs and prepositions much more often than pronouns before those words. Is it possible to have three unstressed pronouns in a row?

[alice]

Using sin for is replaces one ambiguity with another. Do Middle Gothic speakers ever get confused by sentences like “Jack showed John what was left of his book?”

Since this is precisely what happened in Dutch and German, I don't think so.