Savvinic (Modern Day Oscan)

[kanejam]

Oscan is an Italic language, related to Latin and Faliscan (and possibly more distantly to Venetic and Lepontic), but more closely related to Umbrian, Paelignian and South Picene. It was spoken by several tribes and cities, and at one point the Oscan tribes enjoyed a thriving culture which produced many books and plays. They were eventually absorbed by the Roman empire and the language ceased to be spoken by about 50BCE

Alas, all of the literature has been lost, including an Oscan dictionary written by a Roman emperor which would have been comparable to the Latin literature of the time. What's left are inscriptions which tend to be of a religious nature which hampers reconstruction. For a discussion on reconstruction Oscan, see the link in my sig. The idea of reconstructing Oscan and creating a modern day descendent was inspired by Znex and Lambuzhao.

Let's start with the phonology of Classical Oscan.
/m n/
/p t k/
/b d g/
/f s x/
/w l r j/

Monophthongs in stressed syllables: /iː u uː e eː ɛ ɛː a aː o/
Monophthongs in unstressed syllables: /i u e ɛ a o/
/ai ɛi oi au ou (ɛu)/

- Consonants can appear geminate.
- Stress is always word-initial
- The fricatives /f s/ were voiced intervocalically.

In terms of morphology and syntax it was very similar to Latin. It had 3 genders, 7 cases, 5 declensions, 4 conjugations, 3 moods, 2 aspects and 3 tenses.

I haven't quite got all the sound changes sorted yet, but for now I'll give a quick sketch of the Modern Oscan language:

/m n ɲ (ŋ)/ <m n gn (ng)>
/p t ts tʃ k/ <p t z c(i) c(h)>
/b d (dz) dʒ g/ <b d (ż) g(i) g(h)>
/f v s z ʃ ʒ/ <f v s ṡ sc(i) ṡg(i)>
/w l r j/ <u l r j>

Stressed syllables: /a aː ɛ ɛː e eː i iː o oː u uː ai ei au eu iɐ ui/ <a a è è é é i i o o u u ai ei au eu ia ui>
Unstressed syllables: /ɐ ɪ ʊ ɐɪ ɐʊ/ <a i u ai au>

Stress is usually initial and will be marked with a grave accent on non-initial stressed short vowels and diphthongs. The language is SVO and has two genders and three cases on its nouns, a direct case as well as a genitive and a dative. The language is fairly conservative in terms of final vowels and consonants, mainly because I wanted to keep a case system (although I think the reduction to three keeps it realistic) and to keep the cool -u feminine ending, although this will likely be eclipse by the massive amount of Romance loans.

An example from the accomplishment thread:

Numbers 1 - 10: ein/einu, dûj/dôj, trêj, pezzur, pompe, ses, seftun, ôtu, noun, deccen.

Spoiler:

The conjugations of the first four:
..... direct ... genitive ... dative
1.M ein ... eigni ... eini
1.F einu ... eigni ... eini
2.M dûj ... dûn ... dûn
2.F dôj ... dôn ... dôn
3 ... trêj ... trên ... três
4 ... pezzur ... pezzun ... pezzir

Sound changes up next. Feedback welcome!

[clawgrip]

How did you decide on the sound changes? Just based on personal preference, or are you picking things common to Romance languages?

[kanejam]

Well there are changes in both the phonology and the morphology that follow specifically the South Italian languages such as the palatalisations and the articles and word order etc.

For the most part and especially for the phonology I wanted it to be more original. After all, the Welsh and English languages or the Albanian and Greek languages aren't all that similar despite being in very close proximity to each other. I'll concede that those languages weren't so similar to begin with whereas Oscan was very close to Latin, but they would have been separate languages for maybe as long as 3000 years, twice as long as any Romance languages.

They are at the moment quite conservative and pretty basic (not too many interesting conditional changes and the vowels stay pretty boring) so I'll need to flesh them all out.

[kanejam]

From Oscan to Old Savvinic:
Roughly 0CE to 1000CE. The general framework here is unlikely to change, but more sound changes or revision to the history are likely. Comments on plausibility etc are welcome.

Sound changes

Small vowel shift:
e → i / _[+stress] / _[+stress]$ (unconditional changes are less fun!)
e, i → j / _[-stress]V
o, u → w / _[-stress]V
eu → ou (rare)
e.g. ísík 3SG.MASC.NOM ['ezek] → ['izek], íúnk 3SG.MASC.ACC [eoŋk] → [jok]

Consonant weakening:
g → j / V_[ɛei](ː)
b → w / V_[ou](ː)
x → ː / V[+stress]_
x → h / #_ (not sure about this, I should probably just get rid of it)
x → ∅
e.g. flagis 'fire?' ['flagis] → ['flaji], trííbuf 'chief' ['treːbuv] → ['treːwuv], úhtu 'eight' ['oxtu] → ['oːtu]

First palatalisations:
(k)kj → ttʃ / V_
(k)kj → (t)tʃ
(g)gj → ddʒ / V_
(g)gj → (d)dʒ
(k)k → (t)tʃ /_[ɛei](ː)
(g)g → (d)dʒ /_[ɛei](ː)
e.g. tangiuf 'opinion' ['taŋgjuv] → ['taɲdʒuv], kersnú 'meal' ['kɛrsno] → ['tʃɛrsno]

Nasal + stop:
ng → ŋŋ /_V
nd → nn /_V
mb → mm /_V
nt → tt /_V (not sure about these three)
nk → kk /_V
mp → pp /_V
e.g. kúmbenis 'meeting' ['kombɛnis] → ['kommɛni], kúntrud 'against' ['kontrud] → ['kottruz]

Metathesis with s (maybe), stop + stop:
gm → jm
gn → jn/nj → (ɲ)ɲ
ŋŋ → ɲɲ
mn → nn
kt → tt
ks → sk?
ks → ss (probably not word-finally)
e.g. egmú 'thing' ['ɛgmo] → ['ɛimo], gnaattiuf 'race, culture' ['gnaːttjuv] → ['ɲaːttsuv]

Second palatalisations:
(t)tj → (t)ts
(d)dj → (d)dz
(s)sj, stʃ → (ʃ)ʃ
zj → ʒʒ
(r)rj, (l)lj → (ʎ)ʎ
(n)nj → (ɲ)ɲ
*Consonants doubled intervocalically
e.g. pettiur 'four' ['pɛttjur] → ['pɛttsur], kasíu 'I should' ['kazeu] → ['kaʒʒu], kariú 'bread?' ['karjo] → [kaʎʎo]

Reduction of geminate consonants:
maybe? something cool here to mess with the compensatory lengthening later on
ll, rr, m, n → ?
I'll need to fix this later on

Final consonants shift:
s → j / V_#
m → n / V_#
d → z / V_#
ns → n / V_#
nt → ns / V_# (inconsistent with other changes?)
n → ∅ / _[kt]#
s → ∅ /[kpts]_#
C → ∅ / s_#
e.g. duus 'two' ['duːs] → [duːj], kúntrud 'against' ['kontrud] → ['kottruz], kúlúps 'thief' ['kolops] → ['kolop]


Morphosyntactic changes

As with the other Romance languages, Oscan underwent a reduction in morphology, losing its passive verb forms (and maybe some or all future forms) as well as a reduction in the number of cases from seven to five (the vocative being lost and a merger of the dative and locative). The neuter gender merged with the masculine, except for a few neuter plurals reanalysed as singulars and merging with the feminine.

The anaphoric demonstrative (ísík, cognate to Latin is) was reinforced in its pronoun functionality, as well as forming a new definite article (although I may change my mind and use essuf, cognate to Latin ipse). An indefinite article formed out of the number one.

Word order switched to strongly SVO except when using pronouns, which tended to come before the verb. In terms of syntax it is still basically identical to that of the surrounding Romance languages (and for the moment, will remain so to the present day, other than the small differences that case makes).

Examples will follow up later.


History

TL;DR ~100BC part of Roman Republic and Empire until its fall, 493 independent, ~580 part of Lombardia Minor as the Samnite Duchy, ~780 part of Duchy of Benevento, 899 independent, 962 southernmost part of Holy Roman Empire, 1053 conquered by Normans.

So here's a bit of backstory. By the start of the Roman empire, the various Osci tribes were already taken over. But instead of quickly assimilating into Roman culture and language as they did in real life, they were allowed a much greater level of freedom and self-rule.

This was the same for the Umbri and North Oscan tribes, creating two separate Sabellian states (unsure how realistic this is), although the belligerent Samnites to the south were put under harsher rule and eventually assimilated completely.

They underwent a relatively quiet existence until the end of the Roman empire in 493, absorbing hundreds of both Classical and Vulgar Latin loanwords (the first for technical vocabulary, the latter more common).

They started in the north of Campania and the east of Lazio, but after the fall of the empire, they stretched through Appenines and took over much of the Molise region (I'm not confident about the geography) and established a somewhat feudal state out of the existing provincial government.

It was taken over by the Lombards in the late 500's, forming the Samnite Duchy (Duukatúm Savínieís) in between the Duchies of Benevento, Spoleto and Capua (real world Samnite Duchy was another name for the Duchy of Benevento).

As with the other Southern Lombard states, it was essentially independent, and helped to establish the idea of an Oscan nation, with Oscan being reintroduced as the official language (though most records were still kept in Latin, there were however a handful of examples of written Oscan termed Old Oscan or Old Savvinic).

While the classical Osci did have a literary culture, the books were rare and not often preserved by scribes or monks as Latin literature was. Thus by this stage very few people were acquainted with classical Oscan, and without the stigma of speaking 'degraded Latin', were happy to write down their language as was.

They were briefly conquered by the Franks, but in 776 they rebelled along with the other southern Lombard Duchies. They eventually became part of the Duchy of Benevento rather than join the Franks (although with a fair amount of coercion). The Oscan language lost its official status once more, but had taken enough of a hold among the people to ensure its survival.

Throughout its history as part of Lombardia minor, it continued to gain loans from the central dialect of Vulgar Latin, as well as a handful of Germanic loans from the Lombards. These loans tended to be common vocabulary.

The Oscans now suffered the brunt of the Frankish hostilities, but thankfully missed the Saracen sackings of Southern Italy. In 899, after the Duke of Capua conquered a weakened Benevento, the Samnite Duchy briefly re-established itself, this time siding against the Lombard nations and joining the Holy Roman Empire in 962 under King Otto. A handful of West Romance words as well as one or two Frankish words entered the language, as more technical and judicial vocabulary.

This lasted until 1053 when it was conquered by the Normans, and this marks the end of the Old Oscan period and the start of the Middle Oscan Period.

[atman]

As a long-time resident of Southern Italy, and as a fan of Indo-European languages, I think I should say something about this.

So, a few friendly suggestions:

Nasal + stop:
ng → ŋŋ /_V
nd → nn /_V
mb → mm /_V

Wouldn't these create a phonemic intervocalic velar nasal? Those are very rare in Europe, never mind Italy. I'd rather go with ng → gg, nd → dd, mb → bb.

nt → tt /_V (not sure about these three)
nk → kk /_V
mp → pp /_V

Basically all southern Italian languages go nt > nd, nk > ng and mp > mb. And you could perhaps find a way to have some words start with nasal+stop clusters, a typical southern italian feature.

gm → jm

I turned Ancient Greek /gm/ clusters into Old Atlántika /ŋm/ and from there /ɲɲ/ > /ɲ/. See the development of the word for "thing" is pregma > preŋma > preɲɲa > preɲa, written prènha (compare English pragmatic)

kt → tt
ks → sk? maybe something initially? a few other changes will probably appear here as I resolve various clusters...

I'd say ks > ss, but ks > s word-initially would do. In Atlántika, ks > xs > ʃʃ > ʃ unconditionally, always

s → j / V_#

One of the signature Italian changes! Classical Armenian (being classical Armenian) did it even stranger: s > kʰ word-finally!

Please keep up the good work! Bravo!

[kanejam]

As a long-time resident of Southern Italy, and as a fan of Indo-European languages, I think I should say something about this.

Don't just say something, say as much as you can! This is exactly the feedback I need, especially as realism is a strong goal of mine (although I somewhat ignore the question of how Oscan would survive 600 years being dominated by Latin) and resources on South Italian are scarce if you don't speak Italian. So thanks!

So, a few friendly suggestions:

Nasal + stop:
ng → ŋŋ /_V
nd → nn /_V
mb → mm /_V

Wouldn't these create a phonemic intervocalic velar nasal? Those are very rare in Europe, never mind Italy. I'd rather go with ng → gg, nd → dd, mb → bb.

nt → tt /_V (not sure about these three)
nk → kk /_V
mp → pp /_V

Basically all southern Italian languages go nt > nd, nk > ng and mp > mb. And you could perhaps find a way to have some words start with nasal+stop clusters, a typical southern italian feature.

Well, the change nd → nn is a fully established sound change in Oscan, so I thought of the others as 'completing' the change. As much as I want the velar nasal it might be a bit unrealistic, so maybe it will turn into a palatal nasal as soon as it appears.

The nk → kk was meant to be a unique change to try to inject some original character into the language rather than being a South Italian bogolang. However I think the South Italian change you mentioned happens a fair bit later, and Savvinic may well share the change later on.

I didn't know about the initial nasal + stop clusters, very interesting! I'm not sure I can get it into native words as Oscan has strong initial stress (which I think will be the most striking difference) but it will certainly come through in loans of the same shape as Napoleon > Mpoljan (assuming second syllable stress, this is just an example).

gm → jm

I turned Ancient Greek /gm/ clusters into Old Atlántika /ŋm/ and from there /ɲɲ/ > /ɲ/. See the development of the word for "thing" is pregma > preŋma > preɲɲa > preɲa, written prènha (compare English pragmatic)

kt → tt
ks → sk? maybe something initially? a few other changes will probably appear here as I resolve various clusters...

I'd say ks > ss, but ks > s word-initially would do. In Atlántika, ks > xs > ʃʃ > ʃ unconditionally, always

Funny how similar the ends of those words are (egmú is from earlier *egma). I'm okay with this change for now. I think ks → ss makes a lot of sense seeing as I already have kt → tt. I'm not sure quite where I was going with the epenthesis thing. As far as I know though ks doesn't occur initially.

s → j / V_#

One of the signature Italian changes! Classical Armenian (being classical Armenian) did it even stranger: s > kʰ word-finally!

Please keep up the good work! Bravo!

Well I thought about getting a series of pharyngealised ejective affricates but decided against it

Once again, thanks for the great advice! I should bring out the next set of changes in the next day or so.

[kanejam]

From Old Savinnic to Middle Savinnic:
Roughly 1000CE to 1500CE.

Sound changes

Intervocalic approximants disappear:
w → ∅ /[uo](ː)_V
w → ∅ /V_[uo]
j → ∅ /[ieɛ](ː)_V
j → ∅ /V_[ieɛ]
e.g. treeuuf 'chief' ['tre:wuv] → ['trɛ:uv]

Vowels contract, diphthongs merge:
ɛa → a(:)
ua, oa → o(:)
uo, ou → u(:) *these are long in stressed syllables
oe, oɛ → oi
ue, uɛ → ui (→ oi if unstressed)
ea, ia → iɐ
io, eo, eu, ɛo → ɛu
oi, ei → ɛi *these can all be long if the first vowel was originally long
e.g. oin 'one' ['oin] → [ɛin], eoc 3SG.FEM.NOM ['eok] → ['ɛuk]

Vowel syncope:
Vowels after stressed short vowels are deleted as long as it won't produce an illegal cluster. This is also blocked by an /l/ or an /r/ either before or after the unstressed vowel, and so far any deletions in grammatical endings will be restored analogically pretty quickly. Might reformulate this a bit to get it to fit in a bit better.
e.g. colop 'thief' ['kolop] → ['kolp]

Compensatory lengthening:
C → CC / V[+short, +stress]_[Vlrwj]
- Any single stop turns to a geminate when following a stressed short vowel and follwed by either a vowel or a liquid/approximant
ww → vv (or should it be ggw or mm? hmm)
ww → mm
jj → ɲɲ
e.g. mano 'hand' ['mano] → ['manno] (unfortunately I can't find any example for the doubled semivowels)

Vowel breaking and small shift:
a, ɛ → ɐ / _[-stress]
iː, eː → jɛː / _[lr]$
uː, oː → woː / _[lr]$
ɛ, e → jɛ / _C[+velar]$
ɛ → iɐ / _C$
a → ɛ / _C*[ji]
a → o / _C*[wu]
a → ɛ / _[ɲʎ]²
i → e / _[ɲʎ]²
e.g. casgiu 'I should' ['kaʒʒu] → ['koʒʒu], cersno 'meal' ['tʃɛrsno] → [''tʃiɐrsno
*some of these changes might be a bit rare, although once I work on loans they might become more common. They also feel a tad disparate but I want to change the vowels up a bit because otherwise it's a bit too conservative.

Palatal reduction:
(ʎ)ʎ → (ʒ)ʒ
(d)dz → (z)z
e.g. alli 'whole, entire' ['aʎʎi] → ['ɛʒʒi]


Morphosyntactic changes

I suppose if major changes to verbs happen, they'll happen here. For now all I know is that the noun system collapses further, with a gradual loss of the accusative case as well as a merger of the genitive and ablative case (taking the forms of the ablative in the singular and the genitive in the plural).


History

TL;DR - 1053 Conquered by Normans, split into County of Molise and Loritello but rejoined in 1184, 1130 part of Kingdom of Sicily, 1254 crown passed to French rulers, 1435 Aragonese Rule, 1516 part of Hapsburg Empire and Kingdom of Spain until 18th century.

The region was invaded by the Normans in 1053 and split into two Counties, Molise and Loritello, both a part of the greater County of Apulia and Calabria. The County of Molise led a quiet existence, but Loritello was confiscated in 1147 after the count's son tried to overthrow his father and become independent. It was restored in 1154 until 1184 when it was rejoined into the County of Molise.

Meanwhile, in 1130 the count of Apulia and Calabria also became the ruler of Sicily, thus starting the Kingdom of Sicily. In 1254 the crown passed to the French, and in 1282 Sicily was invaded and the kingdom (although they still called themselves the Kingdom of Sicily) became the Kingdom of Naples.

During this time the language picked up a lot of Norman loans, but few French loans as French was really only the language of highest government.

It stayed under French rule until 1435 when it passed to the Aragons (although France contested the region until 1559 when they eventually gave up). For the last few centuries, the region had led a very quiet existence; they were allied to the neighbouring Papal states at this point and so had no major conflicts. This continued after the transfer to Aragon rule, although the region lost its position of count.

The Aragons encouraged Neapolitan as the language of the Kingdom in contrast to every other overlord of the region who all imposed their own language. They also didn't mind the use of Oscan for the regional affairs of the County of Molise and for record keeping. This is the period of the language known as Middle Savvinic. Despite this a number of Catalan and later Spanish loans entered the language.

In 1516 the Oscans became part of the Hapsburg empire (officially part of the Kingdom of Spain) when the Aragonese and Spanish crowns merged. The province led another long period of quietness until the 18th century. Yes, this is a little bit boring but there's isn't much I can think of for a small region to keep exciting.

As always, comments are welcome.

[kanejam]

From Middle Savinnic to Modern Savinnic:
Roughly 1700CE to 2000CE.

Sound changes

Voicing around a liquid:
C[-voice] → C[+voice] / _[lr] except word initially
C[-voice] → C[+voice] / [lr]_
e.g. castre 'head' ['kastrɛ] → ['kazdrɐ]

Nasal + stop:
ng → ŋŋ /_V
nd → nn /_V
mb → mm /_V (yes, again )
nk → ng / _V
nt → nd / _V
mp → mb / _V
e.g. these should be common but need loan words to work

Coda loss after long vowels:
C² → C / V:_
l → w / V_$
rC → Cr / V:_V
V: → V / _C$ **or** C → ∅ / V:_C (unsure, maybe have a separate outcome for long diphthongs? [du:j] to [duj] or [du:]?)
e.g. colp 'thief' ['kolp] → ['koup] (oblique stem koub-), no immediate examples of the other changes

Reduction of unstressed vowels:
a ɛ → ɐ / _[-stress]
e i oi ui → ɪ / _[-stress]
o u ou → ʊ / _[-stress]
ai ɛi → ɐɪ / _[-stress]
au ɛu → ɐʊ / _[-stress]

Word final devoicing:
C[+voice] → C[-voice] / _#
e.g. cottruṡ 'against' ['kottruz] → ['koddrʊs]

S/Z + stop:
s → ʃ / _D (dental)
z → ʒ / _D
w → ∅ / [ʃʒ]_
e.g. sueṡṡur 'sister' ['swɛzzur] → ['ʃwɛzzur] → ['ʃɛzzʊr]

Rhotacisation of d:
d → r / _
rr → ddr / _
e.g. Ital. Madonna → ['mbronnɐ], arraito 'silver' ['arraito] → ['addrɐɪtʊ] (this word is cognate with Ital. argento, how cool is that!?)

Nasalisation?:
Should I have nasalisation? I'm thinking it's possibly allophonic, with the nasal lost word-finally except in stressed monosyllables.


Morphosyntactic changes

Syntactic doubling: Many unstressed particles ending in long vowels or consonants lose them, but to preserve the length the first consonant of the following word is usually doubled. As an example, ioc [jok] 3SG.atonic loses the final [k] but lengthens the first consonant of the following verb: ju pprovvau 'I like him'.

The doubling only happens where doubling can happen, so a word such as spagnul 'Spanish' doesn't undergo syntactic doubling. It also only happens to full verbs, nouns and substantive adjectives. The semivowels which can't be geminate become nasals (as in the compensatory lengthening).


History

In 1712, the Kingdom of Naples was given to the Holy Roman Empire, but Austrian rule only lasted until 1734 when it was reconquered by Spain. In 1799 it was part of the revolt supported by the French that led to the formation of the Parthenopean republic.

This lasted only briefly as the lower classes and the clergy were still loyal to the king, and ended less than five months later. It was a rather rough time for southern Italy, but it was certainly not over.

In 1806, Napoleon took Naples under his control as part of his empire. When Napoleon was defeated, the previous king was reinstalled but this time as the ruler of the new Kingdom of the Two Sicilies, which once again joined southern Italy with the island of Sicily.

The merger in the already war torn region led to abject poverty, and many Oscans along with other southern Italians emigrated in search of a better life. As the region was brought out of feudalism, banditry and brigandage rose dramatically.

In 1860 it was annexed by the Kingdom of Sardinia, which became the Kingdom of Italy a year later. From here the history should look very familiar: it joined the Triple Alliance with Germany and Austria-Hungary in 1882, before switching sides during WWI in 1915.

In 1922 the Fascist party was elected and Benito Mussolini became the totalitarian dictator, allying with Nazi Germany. Under the harsh rule, the Oscan language, which was seen as inferior along with the majority of Italian languages, suffered greatly and was largely replaced by Standard Italian.

In 1943, the region was heavily occupied by the Allies, and this led to a flood of English loanwords. In the following years, the region had an influx of southern workers which further endangered the Oscan language.

In 1946, the Republic of Italy was established, which still exists today. The language continues to see a large number of Standard Italian loans as the language is completely unrecognised at the state level. It is however promoted by the region of Molise.

[kanejam]

Personal Pronouns

I'm doing these because they are pretty easy. There are four forms for each pronoun. The first is the tonic form e.g. meun 1SG.TONIC, which is essentially the direct case but is only really used after prepositions and for emphasis. This is because Savvinic is pro-drop in terms of its subjects due to person marking on verbs, and because when a pronoun is the direct object of a verb it takes its atone form e.g. mi 1SG.ATONE. This form precedes the verb (as in the Romance languages and in contrast to normal objects).

After that there is an atone dative form e.g. mai 1SG.DAT. It is only used before the verb to mark a benefactor or indirect object. It is also used for many reflexive verbs. The last form is the possessive, e.g. mei, meu 1SG.POSS.M.NOM, 1SG.POSS.F.NOM. As the gloss might suggest, the possessive forms agree in all of case, number and gender with the possessed noun, and precede the verb as opposed to genitive nouns. The have an irregular declension.

There are three persons and two numbers, with gender also being distinguished in the third person for both numbers. There is also a reflexive pronoun (with only the two atone forms) which can also be used reciprocally.

1SG ... meun, mi, mai, mei, meu
2SG ... teun, ti, tai, tei, teu
3SG.M ... iṡṡic, ju+, ṡi+, sei, seu
3SG.F ... euc, ja+, ṡai+, sei, seu
1PL ... nù(i), nu, novi, nosdra, nosdru
2PL ... ù(i), u, uevi, uesdra, uesdru
3PL.M ... eus, ju(s), (i)ṡi, eiṡi, eiṡu
3PL.F ... ias, ja(s), (i)ṡai, eiṡi, eiṡu
REFL ... -, si, sciai, -, -

The pluses in the 3SG forms indicate that they cause syntactic doubling in a verb that directly follows them. The letters in brackets in the 3PL forms are the official forms but in informal speech are often dropped, meaning the forms are distinguished solely by the doubling of a following verb. The letters in brackets in the 1PL and 2PL are because I haven't worked out what happens to long diphthongs yet.

Here are the possessives in full for those interested. The genitive and dative singulars come straight from Classical Oscan unlike for the nominals, so they look a bit different. However they are very regular other than the nominitives

... NOM.SG ... GEN.SG ... DAT.SG ... NOM.PL ... GEN.PL ... DAT.PL
SG.M ... -ei ... -ei ... -oi ... -oi ... -un ... -oi
SG.F ... -eu ... -ai ... -ai ... -ai ... -un ... -ai
PL.M ... -a/-i ... -ai ... -i ... -ai ... -i ... -un ... -i
PL.F ... -u ... -ai ... -ai ... -ai ... -un ... -ai

E.g. mei casdra, meu mannu, soi casdrin, nosdri manni, uesdrun madrun
1SG.POSS.M.NOM head-NOM 1SG.POSS.F.NOM hand.NOM 3SG.POSS.M.DAT head-DAT 1PL.POSS.F.PL.NOM hand-PL.NOM 2PL.POSS.F.PL.GEN mother-PL.GEN
'my head, my hand, to his/her head, our hands, of your mothers'.

[Man in Space]

ww → vv (or should it be ggw or mm? hmm)
jj → ɲɲ

Given the second sound change, I'd advise the change to [mm].

[atman]

What a well-done language! Congratulations! When I'll be less busy (and less lazy, read: never) I'll make a thread like this for Atlántika .

In the meantime, just a couple of notes / ideas / alternatives:

V: → V / _C$ **or** C → ∅ / V:_C (unsure, maybe have a separate outcome for long diphthongs? [du:j] to [duj] or [du:]?)

There was a similar problem for me to solve, and this is what I did:

Consider the Atlántika pronouns watos, wata "he, she" "himself, herself" (compare English automatic)

Their nominative plurals "they.MASC, they.FEM" in Old Atl. were /'watoj/ and /'wataj/ respectively, and these became /'wati/ and /'wate/ in the modern language.

The dative singulars of the same pronouns had long diphthongs in them /'wato:j/, /'wata:j/, here the diphthongs merely shortened in the later language, to /'watoj/, /'wataj/, to later lateralize and depalatalize the approximant /'watoʎ/, /'wataʎ/ and finally /'watol/, /'watal/ in modern Atlántika.

S/Z + stop:
s → ʃ / _D (dental)
z → ʒ / _D
w → ∅ / [ʃʒ]_

Similarly in Atlántika, s → ʃ and z → ʒ before any stop. In my language, /w/ undergoes fortition in various places, though I haven't quite decided the environments, turning into /p/ (after /s/ ?) or into /g/ (after a resonant /m n l r/ ?)

[DesEsseintes]

What a well-done language! Congratulations!

This is very impressive and praiseworthy! This thread and clawgrip's Yabushionese have really ignited a little a posteriori spark within my a-priori-beating heart...

Those sdr clusters are so yummy! And many word forms are coming out very convincingly as something not quite Italian, but definitely from that part of the world. Again, very impressive!

[kanejam]

What a well-done language! Congratulations!

Aww shucks guys! It's great to know I'm doing okay!

@atman, what is the setting for Atlàntika? I'm guessing on the name that it has something to do with Atlantis I would love to see a thread on it.

@Des, you should try a posteriori sometime, I can see you creating a previously undiscovered sister lang to Blackfoot. The thing I like is that the framework for the grammar is there already (more so in my case as I'm basically copying romance developments) and you don't need to worry about word formation.

[kanejam]

Yikes, relegated to the third page! I have been busy as uni started three weeks ago but I'm using a little bit of spare time to get some work on Savvinic done. It doesn't help that I've been toying with the idea of developing a tricon IElang (what, I skipped the tricon phase of conlanging maturation, okay?) and so I've been reading up quite a bit on Proto-Semitic.

Anyway, I'm in the middle of working out verbs so I'll post those.


Verbs

So if we derive verbs straight from Oscan we get a table that looks like the following:



We have dropped a few poorly attested forms: the passive which likely wouldn't make it anyway, the subjunctive imperfect, the pluperfect and the (possibly non-existent) imperfect. The present and past participles (It. -ando, -ato) were horridly similar for all verbs other than the e-conjugation, so I thought the gerund would take over that function. This leaves us with the table above, which in terms of morphology looks like any other Romance language apart from the lack of imperfect.

The lack of an imperfective is a bit odd, so I think that function should be filled in. The two main options I see are a periphrastic construction (probably the preterite of 'be' + present pp) or, less likely, the preterite would become the new imperfect and a periphrastic perfect would emerge (probably the present of 'have' + past pp). I'll probably go with the first one for now.

The future could survive where it was lost in the Romance languages as its form is phonologically separate from all the others. There could well develop a periphrastic form, possibly of the form 'have' + infinitive or 'go' + infinitive as in the Romance languages. I'll keep the synthetic future for now though. The conditional is actually just the future perfect, but I thought that it would make sense based on the development of the conditional in the neighbouring Romance languages. That might be a bit unoriginal though, and it might just keep it's future perfect function or be lost entirely, and the conditional function taken over by the subjunctive.

The analytic passive ('be' + past pp) which existed in Oscan and was identical to the Latin periphrastic perfect which survived until today. So after that, I just need to go through and work out the kinks in the conjugations. I'm thinking that the í- and i-conjugations would merge, as they come out nearly identical. That leaves us with an a-conjugation, an i-conjugation and a consonant one that I'll call the e-conjugation for simplicity and historical correctness. The preterite forms will probably go through a fair amount of regularisation as well, leaving only some common verbs with weird alternations such as facci ~ fèfcis ~ fat, fattu 'does, did, done', taggni ~ tèttais ~ tat, tattu 'has, had, had'. Of course 'be' is going to do some really crazy. It should be ès and èṡṡun in the table though.



So anyway, here's an example sentence I've thrown together. I'm still working out articles and I'm lacking in vocab a bit, so this example will probably be unnatural and get replaced with a 'I call myself' type of construction later on.

Uau joccau, nunu mei ès Jacum. Pérpi ès nuna tei?
[wəʊ̯ˈjɔkkəʊ̯, ˈnuːnʊ ˈmɛɪ ɛs ˈjaːkʊ̃m. ˈperpj͜ ɛs ˈnuːnə ˈtɛɪ]
good day, name POSS.1SG be.3SG Jacob. Q.NOM be.3SG name POSS.2SG.M
"Hello, my name is James. What is your name?"

Next up I'll work out nouns and do some work on how loanwords will be incorporated.

Edit: changed 'name' from nuna to nunu which, other than sounding funny, shows something cool. In Oscan, the word is nuumen, where the e is epenthetic and in the nominative and vocative singulars only. The old plural is nuumnú and this gets reinvented as a feminine singular (which is a neat trick that I stole from Albanian). The nasals assimilate and then degeminate after the long vowel and then the unstressed vowel merges into /u/.

Second edit: it turns out that famaun should mean something like 'live' or 'stretch' according to Buck, rather than 'call'.

[kanejam]

Feedback is always appreciated!

Articles

These aren't too complicated and should be familiar to any student of Romance. There are two articles, the definite and indefinite. These agree in case and gender and, in the case of the definite, number. Each article has two forms depending on whether the following word starts in a vowel or not.

Definite Article

Code: Select all

SG .... MASC .......... FEM
NOM ... i+/ṡṡ' ........ ju/j'*
GEN ... ṡu+/ṡgg(i)' ... ṡa+/ṡṡ'
DAT ... ṡai ........... ṡai
PL
NOM ... ju/u'* ........ ja/j'*
GEN ... ṡu+/ṡgg(i)' ... ṡu+/ṡgg(i)'
DAT ... ṡi/ṡg(i)' .... ṡai

The form on the right of the slash is used before consonants. Those with a '+' after them will cause syntactic doubling of the following consonant if possible. The other form is the one used before vowels, with the i not written before an i or e. The asterisked forms change slightly, with j' changing to u' before i and u' changing to j' before u. The M.DAT.PL is usually confused with the other forms ṡgg(i)' and ends up being either ṡi/ṡgg(i)' or ṡi+/ṡgg(i)' in the spoken dialects that use it.

Indefinite Article

Code: Select all

SG ... MASC ...... FEM
NOM ... ai+/nn' .. na/n'
GEN ... ni ....... ni
DAT ... ni ....... ni

These are pretty basic and work the same way as the definite articles above.

I love the little bit of ambiguity in this system, although it is almost always clear from the noun which case is being used.

Here are a set of example nouns:
'hand'
- ju mannu, ṡa mmannus, ṡai mannai, ja manni, ṡu mmannun, ṡai manni, na mannu, ni mannus, ni mannai
'bonfire' or maybe 'perfume'
- j'accendu, ṡṡ'accendas, ṡai accendai, j'accendai, ṡggi'accenżun, ṡai accendai, n'accendu, ni'accendas, ni'accendai
'thief'
- i ccoup, ṡu ccoupus, ṡai coupai, ju coupi, ṡu ccoupun, ṡi coupi, ai ccoup, ni coupus, ni coupai
'man'
- ṡṡ'umuf, ṡggi'umunus, ṡai umunai, j'umun, ṡggi'umunun, ṡgi'umunis, nn'umuf, ni'umunus, ni'umunai

Tell me what you think. The syntactic doubling hasn't worked quite like I want it to but it's certainly not bad. Aesthetics question: do you prefer ṡai accendai, ṡai umunai or ṡaj'accendai, ṡaj'umunai?

I'll do noun declension next and then hopefully I'll get to loan words before too long.

[DesEsseintes]

ṡggi'umunun

Beautiful!

[atman]

Edit: changed 'name' from nuna to nunu which, other than sounding funny, shows something cool. In Oscan, the word is nuumen, where the e is epenthetic and in the nominative and vocative singulars only. The old plural is nuumnú and this gets reinvented as a feminine singular (which is a neat trick that I stole from Albanian). The nasals assimilate and then degeminate after the long vowel and then the unstressed vowel merges into /u/.

I like that, and it's also OK historically. After all, Indo-European languages did the most incredible things with h3nómn̩, which should be the old word for 'name': analogical changes, contaminations, all kinds of linguistic phenomena. It's perhaps the most difficult word to reconstruct in the whole PIE lexicon, even if reflexes are found throughout the family.

The lack of an imperfective is a bit odd, so I think that function should be filled in. The two main options I see are a periphrastic construction (probably the preterite of 'be' + present pp)

I think this option is best, my lang Atlántika also lost the Graeco-Aryan imperfect (Greek épheron, Sanskrit ábharam) and now needs to use a paresent participle + the conjugated form of the verb xtème 'I stand'.

Aesthetics question: do you prefer ṡai accendai, ṡai umunai or ṡaj'accendai, ṡaj'umunai?

I'd say ṡaj'accendai, ṡaj'umunai.

[zyma]

I feel bad about never having much to say other than that this continues to be very impressive and interesting, and that looking at it makes me want to drop everything and make an a posteriori language (I know how you feel, @DesEsseintes).

I agree with @atman about preferring ṡaj'accendai and ṡaj'umunai.

[kanejam]

ṡggi'umunun

Beautiful!

I'm glad you like it!

I like that [...]

Yeah, it'll happen to a handful of other words but I'll keep it as an irregular change. My thinking behind nunu (I wish it didn't look so silly) was that you would discuss or ask for someone's names pretty often, rather than just their praenomen, cognomen or gentilis.

I feel bad about never having much to say [...]

That's fine, I always feel the same way reason other people's posts, which is why I tend to lurk. Sometimes it's nice though to see that people are in fact reading and enjoying this.

And it's settled: ṡaj'accendai and ṡaj'umunai it is!

---

Also, in the introduction I had something about a Roman emperor writing an Oscan dictionary, because I had read that somewhere. But when I tried to go back and find it, I couldn't. It turns out I was thinking of Claudius and his Etruscan dictionary. Whoops! Anyway, I've fixed up the phonology in the first post so just about everything is up to date.

[DesEsseintes]

nunu (I wish it didn't look so silly)

It doesn't look silly; it looks awesome and in line with the aesthetic, which - as I've mentioned before - does a brilliant job of looking South Italian and yet not.