700 years after the fall of Rome ~500ad, the Carthaginian languages descended from Qartədšatim have split into three main dialect continua, Qartim, the varieties of Africa (and Cadiz); Barqim the varieties of Iberia; and Itlim the varieties of Italy, Illyria, and Gaul
These varieties are perceived as having a primary split between Qartim and Barqim (the latter including the Itlim dialect group). Itlim is perceived as having secondary splits between Itlim proper (the dialect group of Italy, Corsica, and Illyria); Yeyim (the dialect group of the Balearic Islands, Sardinia, Malta, and Sicily); and Gillim (the dialect group of Gaul and Britain
There are a few main isoglosses relevant for distinguishing between these varieties
The voiceless labial p:
p: the Eastern varieties generally retain the plosive /p/
f: the Southern and Western Iberian varieties have a fricative pronunciation /f/
fh: Eastern Iberia have /hφ/ for earlier p ~ f
The voiced labial b:
b: Central North Africa, Italy, Gaul, and Illyria retain the plosive pronunciation /b/
v: Iberia, Western Mauretania, and Eastern Libya have a fricative pronunciation /v/
The labial glide w:
w: in Africa, Illyria, and Gaul, the original glide /w/ is retained
v: in Iberia, and Italy the fricative is pronounced /v/, merging with original b in Iberia
Geminates:
CC: in the East, geminates are preserved as geminates
NC: in the Southwest, geminates prenasalise and lose gemination
Emphatic spreading:
C_C̣_C: in the North, emphaticness doesn’t spread
C̣_C̣_C̣: in the South, emphaticness spreads
The guttural h:
h: in the South and East, h is generally preserved
x: in the North, h is pronounced /x/
Ø: in the West h is lost
The hushing fricative š:
š: in the Southwest, the pronunciation of š as /ʃ/ is preserved
s: in the Northeast, š merges into s
Voicing of emphatics:
ṭ: in the South, emphaticness is retained
t: in the East emphatic consonants merge into voiceless consonants
d: in the West emphatic consonants merge into voiced consonants
Elyŝiskää is a complex dialect continuum, with most varieties lying on one of five groups of dialects, together with two isolated dialects. This grouping is based on relatively significant concentrations in isoglosses
There are 10 main isoglosses:
The umlauted vowels:
The pronunciation of the vowels ä, ö, y varies
æ ø y: the umlauted vowels are always /æ/ /ø/ /y/
e e iu: ä and ö are /e/, y is /i/ except in a few anharmonic words where it harmonises to /u/
(e e iu): in unstressed syllables behaves as e e iu, otherwise they take their standard values
((e e iu)): in unstressed syllables, short umlauted vowels behave as e e iu, otherwise vowels take their standard values
(((e e iu))): in syllables without primary stress behaves as e e iu, otherwise vowels take their standard values
Consonantal j:
The glides are phonemically divided between vocalic (where it appears after a consonant in the same syllable) and consonantal (all other occurrences)
j: there is no difference in pronunciation between consonantal and vocalic j
š: consonantal j is pronounced /ʃ/
h: consonantal j merges into h
g: consonantal j is pronounced /g/
Ø; consonantal j is dropped completely
Consonantal w:
The glides are phonemically divided between vocalic j (where it appears after a consonant in the same syllable) and consonantal j (all other occurrences)
w: there is no difference between consonantal and vocalic w
f: consonantal w is pronounced /f/
h: consonantal w merges into h
b: consonantal w is pronounced /b/
g: consonantal w is pronounced /g/
Ø: consonantal w is dropped completely
Voicing:
Differing varieties condition voicing in different environments, this is usually only directly represented in the fricatives and the affricate č
1: voicing occurs between vowels (or between a vowel and a vocalic glide), or adjacent to a nasal or voiced stop
V: voicing occurs between vowels (or between a vowel and a vocalic glide)
N: voicing occurs between vowels (or between a vowel and a vocalic glide), or adjacent to a nasal
D: voicing occurs between vowels (or between a vowel and a vocalic glide), or adjacent to a voiced stop
Velar Nasal ň:
The velar nasal ň has many variant pronunciations
ŋ: ň is always pronounced /ŋ/
m: in clusters with a labial, ň merges into m
n: ň merges into n in all positions
(n): in clusters with a dental, alveolar, or postalveolar, ň merges into n, otherwise it is pronounced /ŋ/
nm: in clusters with a labial, ň merges into m, in clusters with a dental, alveolar, or postalveolar it merges into n, otherwise it is pronounced ŋ
j: ň is always pronounced /j/
(j): ň is pronounced /j/ in voiced positions, and /ŋ/ in all other positions
The Postalveolars š and č:
The postalveolars š and č have several different pronunciations
ʃ tʃ: the postalveolars are pronounced /ʃ/ & /tʃ/ when voiceless and /ʒ/ & /dʒ/ when voiced
s ts: the postalveolars are pronounced /s/ & /ts/ when voiceless and /z/ & /dz/ when voiced
z g: the postalveolars are pronounced /z/ and /g/ in all positions
h th: the postalveolars merge into h and ŝ
j j: the postalveolars merge into j in all positions
(j j): the postalveolars merge into j when voiced and /ʃ/ & /tʃ/ when voiceless
The clusters sj and tj:
In different dialects the clusters sj and tj are pronounced differently
sj tj: sj, tj, & dj are pronounced as the clusters /sj/, /tj/, & /dj/
š č: sj & tj merge into š & č, dj is pronounced /dʒ/
ž dž: sj, tj, & dj are pronounced as /ʒ/, /dʒ/, & /dʒ/
j š: sj merges into j; tj & dj are pronounced /ʃ/ & /ʒ/
j j: sj, tj, & dj merge into j
Voiced h:
The pronunciation of h in voiced positions varies between dialects
h: h is pronounced as h in voiced positions
ɣ: h is pronounced as /ɣ/ in voiced positions
Ø: h is dropped in voiced positions
Voiceless h:
The pronunciation of h in voiceless positions varies between dialects
h: h is pronounced h in voiceless positions in voiceless positions
k: in clusters voiceless h merges into k, otherwise it is /h/
2: in clusters voiceless h leads to gemination of the other consonant, otherwise it is /h/
(2): in clusters voiceless h leads to gemination of a preceding consonant, in all other positions it is /h/
L: voiceless h is lost with lengthening of any preceding vowel, even with intervening consonants
(L): voiceless h is lost with compensatory lengthening if immediately following a vowel, otherwise it is /h/
L2: in clusters, voiceless h is lost with compensatory lengthening of the preceding segment, leaving gemination of a preceding consonant, or lengthening of a preceding vowel; otherwise it is /h/
The dental ŝ:
The dental fricative has varying pronunciations in different dialects
θ: ŝ is pronounced /θ/ when voiceless and /ð/ in voiced positions
s: ŝ merges into s
f: ŝ merges into f
t: ŝ merges into t
h: ŝ merges into h
r: ŝ merges into r
(r): ŝ merges into r in voiced positions, is /θ/ otherwise
l: ŝ merges into l
(l): ŝ merges into l in voiced positions, is /θ/ otherwise
The consonant inventory of Aud̄d̄aun̄yo is fairly complex, whilst the vowel inventory is fairly small. Meanwhile the phonotactics are fairly simple, albeit with some substantial sandhi processes
Aud̄d̄aun̄yo Consonants
consonants are listed here in their standard romanised orthography
all consonants except w & y can be geminated in which case they are written twice. If the consonant is written as a digraph, only the first character is repeated
horizontal bars mark either fricatives or dentals
dots mark emphatic (ejective) consonants or uvulars
acute accents mark lateral fricatives or palatal nasals
carons mark palatal fricatives
Aud̄d̄aun̄yo Vowels
vowels here are shown in their standard romanised orthography
macrons mark long vowels
grave accents mark short low-mid vowels
carons mark long low-mid vowels
ai and au are pronounced [aɪ̯] and [aʊ̯] respectively
Aud̄d̄aun̄yo Phonotactics
The possible syllables are:
CV̆
CV̄
CV̆+
CV̄+
CV̆K
where C is any consonant, V̆ is a short vowel (including diphthongs), V̄ is a long vowel, + is used for a coda consonant matching the onset of the following vowel, K is any consonant other than w or y
w & y are not allowed in the coda, but are sometimes morphologically expected to occur in them. In these cases, the semivowel contracts with the preceding vowel resulting in raising (and fronting or backing depending on the semivowel)
In colloquial speech, and in pausa, final short monophthongs are usually dropped, with final long monophthongs shortening, and final rising vowels (diphthongs) contracted to the corresponding low-mid short vowel. This can lead to final consonant clusters although if these are difficult for the speaker, a short vowel /a/ may be inserted between the two consonants
Aud̄d̄aun̄yo Sandhi
affricates lose their fricative when followed by a dental, alveolar, or palatal fricative
glottals are lost following another consonant, causing gemination of that consonant instead
most consonants (all except approximants or trills) assimilate in place of articulation to a following “close” consonants (provided that consonant is also not an approximant or trill)
dentals are close to both alveolars (e.g. t̄s > ts, d̄ś > dś)
alveolars are close to each other, dentals, and palatals (e.g. ṣt̄ > ṯ̣t̄, sc > šc)
palatals are close to both alveolars and velars (e.g. št > st, cḵ > kḵ)
velars are close to palatals and uvulars (e.g. kš > cš, ḵq > q̄q)
uvulars are close to velars and glottals (e.g. q̄k > ḵk, q̄’ > h’)
glottals are close to uvulars (e.g. hq > q̄q)
if two consonants in a cluster are close but the first cannot assimilate due to a gap in the phonology, the second consonant assimilates instead (e.g. tst̄ > tst, ġh > ġq̄)
if two consonants (again not approximants or trills) are “close” and share manner of articulation, they also assimilate voicing reducing the cluster to gemination (voicing is dominant, emphatic/ejective intermediate, and voicelessness is recessive) i.e.
ss > ss
sṣ > ṣṣ
sz > zz
ṣs > ṣṣ
ṣṣ > ṣṣ
ṣz > zz
zs > zz
zṣ > zz
zz > zz
Aud̄d̄aun̄yo Orthography
So far we have used the standard romanised orthography. Additionally, in dictionaries, class vowels are frequently marked with an underdot
There is additionally a native orthography but this is so far a complete WIP
The Phonology of Qartədšatim has diverged a bit from its earlier Canaanite roots, having lost or greatly reduced its laryngeal consonants. It has however retained a three-way distinction between its voiceless, emphatic, and voiced stops and coronal fricatives. The voiceless stops are aspirated, and, whilst the emphatic stops are still pharyngealised, many speakers treat them simply as tenuis
Qartədšatim Consonants
the two allophones of p are in free variation, although when geminated, the stop is predominant
the voiced and emphatic coronal spirants have two allophones, the first listed is the usual form, but in consonant clusters the stop moves to the centre of the cluster (i.e. whilst zr is /zdɾ/, rz is /ɾdz/)
all consonants except h, w, & y can be geminated
h can only occur word-initially, word-finally, and between vowels. It is lost adjacent to another consonant. In coda-position (word-finally) h is preserved as a furtive vowel /a̯/
Qartədšatim Vowels
due to loss of laryngeals, vowels can sometimes occur in hiatus, often resolving into diphthongs ending in one of the two glides
Active Sound Processes
ə is a reduced vowel and is lost adjacent to any other vowel
ə assimilates to an adjacent glide wə, yə, əw, əy > u, i, u, i
word-initial consonant clusters are disallowed
in native vocabulary, this would usually occur with a consonant followed by a glide in which case an epenthetic vowel is inserted corresponding to the glide in question Cj, Cw > Cej, Cow
initial CC clusters are broken with an epenthetic ə in one of two ways, either CC > CəC, or əCC
stress is on the last syllable of the underlying form
if the final syllable is doubly-closed (i.e. ends in two consonants), an epenthetic ə is inserted between the two consonants unless the first is more sonorant than the second i.e. melk “king” does not receive an epenthetic vowel, but úmər “word” does
unlike Hebrew segholates, this epenthetic vowel is not lost when suffixes are added
in words where the stress is not on the final syllable, the stressed syllable is marked with an acute accent in the romanisation
glides are disallowed in the same syllable as their corresponding vowel
this could only occur through the application of prefixes or suffixes
with suffixes, the stressed vowel retains its quality with whatever vowels need to be lowered (to e or o) or alternate to their corresponding glide (y or w) adjusting to compensate
with prefixes, the stem is generally unchanged with the vowel or glide of the prefix being lowered or reduced to the corresponding glide as necessary
the vowels i and u alternate with the glides y and w when adjacent to another vowel
In a stressed final open or singly closed syllable (i.e. one ending only in a single consonant), a is raised and rounded to o
propretonic vowels tend to reduce
voiceless consonants usually voice allophonically when adjacent to a voiced consonant. There are a few exceptions though, word-final -t causes a preceding voiced consonant to devoice
Orthography
There are three distinct orthographies used for Qartədšatim, one in the Punic alphabet, a direct continuation of the Phoenician alphabet, there is a Latin orthography, and the last is a Greek orthography
The Punic Orthography:
The Punic alphabet has no diacritics or vowel pointing, but makes strong use of matres lectionis
𐤀 alp: represents an unreduced vowel. Particularly adjacent to an original ʔ. Also used for word-initial and word-final vowels
𐤁 bet: represents b
𐤂 gáməl: represents g
𐤃 dalt: represents d
𐤄 he: represents h and adjacent unreduced vowels. Particularly adjacent to an original h
𐤅 waw: represents w, u or o
𐤆 zen: represents z
𐤇 het: represents h and adjacent unreduced vowels. Particularly adjacent to an original ḥ
𐤈 ṭet: represents ṭ
𐤉 yod: represents y, i, or e
𐤊 kapp: represents k
𐤋 lamd: represents l
𐤌 mem: represents m
𐤍 nun: represents n
𐤎 samk: represents s
𐤏 ayn: represents an unreduced vowel, usually a. Particularly adjacent to an original ʕ
𐤐 pe: represents p
𐤑 ṣade: represents ṣ
𐤒 qop: represents q
𐤓 reš: represents r
𐤔 šin: represents š
𐤕 taw: represents t
When two vowels are in hiatus, usually only one is written. The reduced vowel ə is not usually written except word-initially
The Latin Orthography:
a: represents a
b: represents b
c: represents q
ch: represents k
d: represents d
ds: represents z following another consonant
e: represents e, also a particularly in closed unstressed syllables
f: represents p when pronounced /f/
g: represents g
h: represents h although this phoneme is often not marked at all
i: represents i
j: represents y
l: represents l
m: represents m
n: represents n
o: represents o
p: represents p when pronounced as /p/
ph: represents p
qu: represents qw or kw, also an alternative to c and ch
r: represents r
s: represent s or š
sd: represents z except following another consonant
st: represents ṣ before another consonant
t: represents ṭ
th: represents t
ts: represents ṣ except before another consonant
u: represents u
v: represents w
y: represents ə
The Greek Orthography:
α: represents a
β: represents b
γ: represents g
δ: represents d
ε: represents e
ζ: represents z
η: represents e or i particularly in stressed and/or open syllables
θ: represents t
ι: represents i
κ: represents q
λ: represents l and occasionally r
μ: represents m
ν: represents n
ξ: represents any cluster of k, q, or g followed by s, š, ṣ, or z
ο: represents o
ου: represents u
π: represents p when pronounced as /p/
ρ: represents r
σ: represents s or š
στ: represents ṣ when followed by another consonant
τ: represents ṭ
τσ: represents ṣ when not followed by another consonant
υ: represents ə
φ: represents p
χ: represents k
ψ: represents any cluster of p, or b followed by s, š, ṣ, or z
ω: represents o particularly in stressed and/or open syllables
ωυ: represents u particularly in stressed and/or open syllables
initial h is usually written with the rough-breathing diacritic but otherwise it’s left unwritten
Correspondences
(Ṣīdūnīm represents the relict of Phoenician in the Eastern Mediterranean)
the colouring in the Arabic column here shows that these have merged despite not being adjacent rows
Hebrew śin and šin are merged
there is no begedkefet spirantisation, as in Arabic
waw retains its glide pronunciation, as in Arabic
the gutterals are lost in many positions
as in Aramaic, initial ʔa is is sporadically lost, especially in names with repair of the initial consonant cluster
original long vowels typically agree with Hebrew
original short vowels often disagree with Hebrew
original diphthongs typically agree with Hebrew except in closed stressed syllables
The Phonology Qartadšoti has diverged a fair bit from its earlier Canaanite routes, having lost all its laryngeal consonants. It has however retained a three-way distinction between its voiceless, emphatic, and voiced stops and coronal fricatives although the pharyngealisation of the emphatics is frequently lost, with the emphatic stops becoming tenuis.
Consonants
voiceless consonants voice allophonically when adjacent to a voiced consonant
in clusters, ṣ loses its pharyngealisation and an epenthetic t is inserted between the ṣ and the other consonant e.g. ṣr is [stɾ], rṣ is [ɾts]
in clusters with z, an epenthetic d is inserted between the z and the other consonant e.g. zr is [zdɾ], rz is [ɾdz]
all consonants (except the glides w & j) can be geminated
Vowels
due to the loss of laryngeals, vowels can be placed in hiatus, which can resolve into diphthongs ending in one of the glides w or j
Active Sound Processes
/ff/ is fortified to /p/
/f/ is fortified to /p/ word-initially
y is lost adjacent to any other vowel
y assimilates to an adjacent glide i.e. wy, jy, yw, yj > u, i, u, i
word-initial consonant clusters are disallowed
in native vocabulary these clusters can only occur with the second element as a glide with the introduction of an epenthetic vowel i.e. Cj, Cw > Cej, Cow
in borrowings, clusters with the second element as a glide are resolved the same way as in native vocabulary i.e. Cj, Cw > Cej, Cow
in borrowings, other clusters are resolved through the introduction of an epenthetic vowel i.e. CC > CyC
the glides are disallowed adjacent to their corresponding vowel
in native vocabulary, this can only occur through the addition of suffixes or prefixes. If the affix contains the glide, the full vowel is lowered i.e. [word]i-j, [word]u-w > [word]e-j, [word]o-w
if the affix contains the full vowel and borders an alternating string of glides and the corresponding lowered vowels, the affix retains its quality & becomes a glide, and the rest of the chain adjusts i.,e. [word]owow-u, [word]ejej-i > [word]wowo-w, [word]jeje-j (this can lead to the insertion an epenthetic vowel to resolve an initial consonant cluster)
in all other situations where the affix contains a full vowel, it is absorbed into the preceding glide i.e. [word]j-i, [word]w-u > [word]j, [word]w
in native vocabulary, prefixes follow the same pattern with the quality of the affix’s vowel being preserved, and the affix vowel only being absorbed as a last result to prevent illegal clusters i.e. j-i[word], i-jeje[word], i-j[word] > j-e[word], j-ejej[word], j[word]
in borrowings, the vowel is usually dropped in word-internal syllables, otherwise it resolves by lowering the full vowel
vowel/glide alternations occur word-internally, between a word and any affixes or attached particles, and between the components of a genitive construction
n assimilates to a following full consonant (so not glide). Unlike vowel/semivowel alternations, this does not occur between components of a genitive construction, only word-internally, or between a word and any affixes or attached particles
Orthography
The Qartadšoti alphabet is a cursive form of the Phoenician alphabet. For convenience of display it is represented here with the corresponding Hebrew letter. There are no diacritics and vowels, when represented, are represented through matres lectionis
א olf: represents o. Can also represent a or e, particularly when adjacent to an original (now lost) ʔ
ב bet: represents b
ג gymol: represents g
ד delt: represents d
ה e: represents a or e, particularly when adjacent to an original (now lost) h
ו u: represents w if adjacent to a vowel, or u if between consonants
ז zen: represents z
ח et: represents a or e, particularly when adjacent to an original (now lost) ḥ
ט ṭet: represents ṭ
י id: represents j if adjacent to a vowel, or i if between consonants
כ kof: represents k
ל lamd: represents l
מ mem: represents m
נ nun: represents n
ס samk: represents s
ע an: represents a
פ fi: represents f & p
צ ṣad: represents ṣ
ק quf: represents q
ר ruš: represents r
ש šin: represents š
ת to: represents t
Any epenthetic t or d resulting from a cluster containing ṣ or z is not written. The vowel y is also not written except word-initially where it is written variously with א ה & ,ח
Where two vowels occur in hiatus, usually only a single is written. If either is “a”, an “an” is used, otherwise it is usually the first vowel that is written. If the definite article causes hiatus, both vowels are written
Correspondences
The main correspondences with Hebrew are as follows
Hebrew śin and šin have merged into Qartadšoti šin
the begedkefet letters do not spirantise (other than p which is always a fricative in Qartadšoti)
waw retains its glide pronunciation
Hebrew laryngeals are all lost (although an colours neighbouring, particularly preceding vowels to a)
Hebrew ō is usually Qartadšoti u
a is raised and rounded to o in final open or singly-closed syllables
many vowels are reduced to y in non-final syllables
as in Aramaic, initial ʔV and ʕV (particularly where V=a) are sporadically lost, especially in names, with repair of particularly egregious consonant clusters by insertion of a vowel within the cluster, this often leads to doublets e.g. ʕabd “servant/slave” > *bd > *bad > bod whence both names ʕAbdaštart and Bodaštart.
Most elements of Iliš’s phonology straightforwardly continue that of British English in the South East of England; there have however been a few major changes; the development of the palato-velars into a full retroflex series, and universal monophthongisation
Consonants
The consonants are all fairly self-explanatory, there is widespread assimilation of alveolar consonants to retroflex consonants when in the same cluster. /n/ also assimilates to /ŋ/ when followed by a velar (except /j/). All consonants can be geminated
Vowels
The vowels are grouped, somewhat loosely, according to their origin in British English. This grouping is important for the orthography as it determines what diacritic (if any) is used in spelling the vowel. Vowel length is not phonemic. The diphthong vowels (which are phonetically monophthongs) only occur before consonants, otherwise they reduce to the corresponding short vowel followed by the approximant /j/. There is widespread alternation between /ə/ and /ɻ/ depending on the environment and many vowels (including all the rhotics, /ɜ/, & /ə/) produce an epenthetic /ɻ/ before another vowel, or cause a following consonant (cluster) to become retroflex
Orthography
Iliš uses the basic Latin alphabet (without <c>, <q>, <w>, <x>, or <y>) along with five diacritics (four of which occur on vowels, and one on consonants). The caron is used to mark retroflex consonants of the corresponding alveolar consonant (e.g. <š> represents /ʂ/, <ť> represents /ʈ/, <ň> represents /ɳ/); the acute accent marks tense vowels of the corresponding type (e.g. <á> represents /æ/); the grave accent represents rhotic vowels of the corresponding type (e.g. <à> represents /ɑ/); the circumflex accent represents long vowels of the corresponding type (e.g. <â> represents /ɐ/ ); and the dot diacritic represents diphthong vowels of the corresponding type (e.g. <ȧ> represents /a/)
<a> represents a-type vowels
<b> represents /b/
<d> represents /d/ or /ɖ/
<e> represents e-type vowels
<f> represents /f/
<g> represents /g/
<h> represents /x/
<i> represents i-type vowels
<j> represents /j/
<k> represents /k/
<l> represents /l/ or /ɭ/
<m> represents /m/
<n> represents /n/ or /ɳ/
<ng> represents /ŋ/
<nng> represents the geminate /ŋŋ/
<o> represents o-type vowels
<p> represents /p/
<r> represents /ɻ/
<s> represents /s/ or /ʂ/
<t> represents /t/ or /ʈ/
<u> represents u-type vowels
<v> represents /v/
<z> represents /z/ or /ʐ/
Sound Changes
The main sound changes between modern English of South East England and Iliš are as follows:
th-fronting is complete (although in some rare environments, the dental fricatives can become approximants instead)
l-vocalisation is complete
neighbouring /r/ causes alveolar consonants to become retroflex, a feature that spreads throughout the cluster (and across most word boundaries)
homorganic stops assimilate to a neighbouring /n/ or /m/
Both standards of Elyŝiskaa have phonologies reminiscent of Karelian, with a series of postalveolar fricatives, and an eight-vowel system with three harmonic pairs & two neutral vowels. The most obvious differences are in the Latin standard based on the variety of the Kalevalsky district (the Cyrillic standard based on the Arkhangelsk variety resembles Karelian more closely)
Consonants
The Latin standard, representing the Kalevalsky variety has the following consonants
All consonants except /w/ & /j/ can be geminated. /h ~ x/ generally appears as /x/, with /h/ occurring word-initially and between vowels when ungeminated. Voicing is only phonemic in the stops, but the fricatives and affricate allophonically voice between vowels (even if followed by a semivowel), or if adjacent to a voiced stop, or nasal.
The Cyrillic standard, representing the Arkhangelsk variety has the following consonants
As in the Kalevalsky variety, all consonants (except /j/) can be geminated, and /h ~ x/ follows the same rules of allophony. /n ~ ŋ/ takes the form /ŋ/ in a cluster with a velar, but is otherwise alveolar. The Arkhangelsk variety (and the Cyrillic standard) shows considerable influence from Russian.
Vowels
Both varieties exhibit the same vowels
Each of these vowels can occur long or short. A word (even many compounds or recent loanwords) will generally only feature vowels of consistent harmony, i.e. it won’t include both front and back vowels. There are a wide array of diphthongs, each with the second element non-syllabic. As most pairs of compatible vowels can occur as diphthongs, it is more helpful to consider syllables with diphthongs as actually containing two vowels in the nucleus. The full set of available diphthongs are shown below
Syllables are of the form (C)(J)V(V)(C) where J is a semivowel. The phonotactics of the semivowels are not entirely simple though, they can occur after a consonant before a long vowel or diphthong, or most intervocalic positions. Stress is always initial, and polysyllabic words are required to have an even number of syllables to maintain trochaic rhythm. This is accomplished through the addition of an additional /e/ at the end of the word, between the stem and ending, or within certain clusters as needed; in some instances this is not possible and those words are allowed an odd number of syllables
Orthography
The Latin orthography uses the basic Latin alphabet with five additional letters (ä, č, ö, ŝ, & š), and five letters (c, q, v, and z) which are only used in loanwords.
<a> represents /ɑ/
<ä> represents /æ/
<b> represents /b/
<c> used in some loanwords to represent /k/ or /s/ before /i/ or /e/
<č> represents /tʃ/
<d> represents /d/
<e> represents /e/
<f> represents /f/
<g> represents /g/ or /tʃ/ in loanwords before /i/ or /e/
<h> represents /h ~ x/
<i> represents /i/
<j> represents /j/
<k> represents /k/
<l> represents /l/
<m> represents /m/
<n> represents /n/
<o> represents /ɔ/
<ö> represents /œ/
<p> represents /p/
<q> used in some loanwords to represent /k/ or the cluster /kw/ when followed by a <u>
<r> represents r
<s> represents /s/
<ŝ> represents /θ/
<š> represents /ʃ/
<t> represents /r/
<u> represents /u/
<v> used in some loanwords to represent /ʋ/, equivalent to native <w>
<w> represents /ʋ/
<x> used in some loanwords to represents the cluster /ks/
<y> represents /y/
<z> used in some loanwords to represents /s/
Any letter except the semivowels can be doubled in order to mark gemination of a consonant, or a long vowel. A sequence of two vowels marks the appropriate diphthong
The Cyrillic orthography
<а> represents /ɑ/
<а̄> represents /ɑ:/
<ӓ> represents /æ/
<ӓ̄> represents /æ:/
<б> represents /b/
<в> represents /v/
<г> represents /g/
<д> represents /d/
<дж> represents /dʒ/
<е> represents /e/
<е̄> represents /e:/
<ж> represents /ʒ/
<з> represents /z/
<и> represents /i/
<ӣ> represents /i:/
<й> represents /j/
<к> represents /k/
<л> represents /l/
<м> represents /m/
<н> represents /n/
<о> represents /ɔ/
<о̄> represents /ɔ:/
<ӧ> represents /œ/
<ӧ̄> represents /œ:/
<п> represents /p/
<р> represents /r/
<с> represents /s/
<т> represents /t/
<у> represents /u/
<ӯ> represents /u:/
<ӱ> represents /y/
<ӱ̄> represents /y:/
<ф> represents /f/
<х> represents /h ~ x/
<ц> represents the cluster /ts/ in some loanwords
<ч> represents /tʃ/
<ш> represents /ʃ/
<щ> represents the geminate /ʃʃ/ in some loanwords
The Cyrillic orthography mark vowel length with a macron rather than doubling. Geminates are marked by doubling the consonant. The fact vowels can occur with both umlaut and macron diacritics can make it difficult to render correctly in some fonts (including those that tumblr seems to be using here)
Correspondences between the standards
Kalevalsky /h ~ x/ is usually absent in the Arkhangelsk variety surviving only in geminates (which Arkhangelsk reduces to single consonants), and word-initially; otherwise it is dropped with compensatory lengthening of the preceding vowel or gemination of the preceding consonant
Kalevalsky /ʃ/ & /tʃ/ are usually preserved in the Arkhangelsk variety, but their voiced allophones are rendered /j/
Kalevalsky /sj/ & /tj/ are palatalised to /ʒ/ & /dʒ/ in the Arkhangelsk variety
otherwise, Kalevalsky /j/ & /w/ are strengthened to the voiced fricatives /ʒ/ & /v/ in the Arkhangelsk dialect
Kalevalsky /θ/ is fronted to /f/ (or /v/ depending on voicing) in the Arkhangelsk variety
Kalevalsky /ŋ/ assimilates to the consonant it is in a cluster with in the Arkhangelsk variety, it is /m/ in clusters with labials, /ŋ/ in clusters with velars, and /n/ otherwise
Kalevalsky geminate fricatives are degeminated in the Arkhangelsk variety introducing phonemic voicing in the fricatives
Sound Changes
The main sound changes in the history of Elyŝiskää from Proto-Indo-European are as follows:
the palatovelars and labiovelars merge into the plain velars with compensatory semivowels /j/ & /w/ added both before and after the stop
laryngeals colour /e/ and lengthen /i/ & /u/
stops become aspirated before a laryngeal
laryngeals between consonants, or between a consonant and the start or end of a word, vocalise to /e/, /a/, or /o/ as in Greek; otherwise they merge into /h/
voiced stops nasalise before a nasal, voiceless stops voice before a nasal, aspirates & fricatives become voiceless stops before a nasal. Before a homorganic nasal, voiceless stops, aspirates, and fricatives nasalise instead of their usual behaviour
aspirates become voiceless fricatives
stressed syllable determines vowel harmony; /i/, /i:/, /e/, & /e:/ set front harmony, /a/, /a:/, /o/, /o:/, /u/, & /u:/ set back harmony
/k/ and /g/ palatalise to /tʃ/ under the influence of neighbouring /i/, /i:/ or /j
The phonology of Þȳðsk is fairly typical of the old and conservative Germanic languages, featuring both dental fricatives, front rounded vowels, and phonemic vowel length. It's consonant inventory is relatively large, as it retains the dental fricatives, but the vowel inventory is much smaller than that of many modern Germanic languages, with many vowels straightforwardly continuing their Proto-Germanic values
Consonants
All consonants except /ŋ/, /ʒ/, /ɣ/, /ʃ/, /w/, and /j/ can occur as geminates. /r/ is realised as /ɾ/ singly, and /r/ as a geminate, geminate /r/ can also be syllabic if word-finally after a consonant. /x/ appears as /h/ in syllable onset and /x/ in the coda. /s/ voices allophonically between vowels. /ʃ/ & /ʒ/ are almost exclusively realised as affricates /tʃ/ & /dʒ/, with the simple fricative only occurring word-finally
Vowels
Each vowel quality in Þȳðsk appears as a long and a short phone, accompanied by differences in vowel quality as well. The positions of the vowels (labelled with their usual spelling, rather than phonetic transcription) on the vowel chart are shown below
There are eleven diphthongs, which lack a length distinction many of which occur as a result of the three different mutations on the Proto-Germanic diphthongs. The positions of the diphthongs (labelled with their usual spelling, rather than phonetic transcription) on the vowel chart are shown below
As is usual for the Germanic languages, stress is only the first syllable of the root (so excluding any prepositions, reduplicated syllables, or other prefixes)
Latin Orthography
The Þȳðsk alphabet contains the basic latin alphabet as well as four additional letters (æ, ø, þ, and ð). There is a single diacritic, this is usually a macron, but acute accents, dots, and circumflexes are also attested in some texts.
<a> represents /ɐ/
<ā> represents /ɑ:/
<āi> represents /ɑɪ̯/
<āu> represents /aʊ̯/
<æ> represents /æ/
<ǣ> represents /æ:/
<ǣi> represents /æɪ̯/
<b> represents /b/
<c> appears in recent or learned loanwords. Represents /k/ or /s/ if written before <e>, <i>, or <j>
<d> represents /d/
<e> represents /ɛ/
<ē> represents /e:/
<ēa> represents /eɐ̯/
<ēi> represents /ɛɪ̯/
<ēu> represents /ɛʊ̯/
<f> represents /f/
<g> represents /g/
<gh> represents /ɣ/
<gj> represents /dʒ/
<h> represents /x ~h/
<i> represents /ɪ/
<ī> represents /i:/
<j> represents /j/
<k> represents /k/
<kj> represents /tʃ/
<l> represents /l/
<m> represents /m/
<n> represents /n/
<ng> represents /ŋ/
<o> represents /ɔ/
<ō> represents /ɔ:/
<ōa> represents /oɐ̯/
<ōi> represents /ɔɪ̯/
<ōu> represents /ɔʊ̯/
<ø> represents /œ/
<ø̄> represents /œ:/
<ø̄i> represents /œʏ̯/
<ø̄u> represents /œʊ̯/
<p> represents /p/
<q> represents /k/ in recent or learned loanwords
<qu> represents /kf/ in recent or learned loanwords
<r> represents /ɾ/
<s> represents /s/
<sj> represents /ʃ/ which only occurs word-finally in native words, but in recent or learned loanwords can occur in any position (although it is sometimes confused with /tʃ/ in non-final positions)
<t> represents /t/
<u> represents /ʊ/
<ū> represents /u:/
<v> represents /v/
<w> represents /w/
<x> represents /ks/ in recent or learned loanwords
<y> represents /ʏ/
<ȳ> represents /y:/
<z> represents /s/ in recent or learned loanwords
<zj> represents /ʒ/ which only occurs word-finally in native words, but in recent or learned loanwords can occur in any position (although it is sometimes confused with /dʒ/ in non-final positions)
<þ> represents /θ/
<ð> represents /ð/
doubled consonants are geminated
Fuþark Orthography
There is also a runic fuþark orthography which survived well into the modern day and, whilst most Sāuvīlanders usually use the Latin alphabet, the fuþark is still used in some epigraphic contexts, similar to those where Latin might be used in the English-speaking world, and it is widely understood.
It is divided into four ætts (each attributed to a specific Pagan God whose name begins with one of the runes of that ætt) of eight runes, as well as a fifth ætt (attributed to Loki) of eight bindrunes used for numerals in a calque of the Roman numeral system
Neither gemination or vowel length are marked in the fuþark
In Lykjīnrǣihtr, the usual rune equivalent of Roman numberal “I” is identical to īsr, the bindrune ǣin is used whenever a single I would appear in the equivalent Roman numeral
Diphthongs are represented either as a sequence of two runes, or as bindrunes (which are now seen as archaic, similar to spelling Latin juvenis “young” “ivvenis” or using æ or œ ligatures)
You may recognise the banner at the top of this blog as being written in this runic orthography!
Sound Changes
The major sound changes in the history of Þȳðsk are:
i, a, & u umlaut
Holtzmann’s law (in this case Proto-Germanic jj ww > gj gw with compensatory lengthening of the preceding vowel)
resolution of diphthongs to a reduced set or to long vowels
nasals assimilate to a following consonant (unless it’s a glide) and lengthen the preceding vowel
Proto-Germanic w hw > v w
medial vowels reduce
reduction of -Vr endings to -r, loss of some medial syllables, and simplification of consonant clusters
loss of j before i, e, or y, loss of v & w before u, o, ø, or y with compensatory lengthening
palatalisation of k, g, & ɣ > tʃ, dʒ, & j before i, e, or y and palatalisation of k & g > ʃ & ʒ word-finally after i or ī, and loss of ɣ after i or ī with compensatory lengthening after short i
