Getic
Posted: 21 Jun 2018 20:42
0. Table of contents
1. Introduction and essential sound laws
2. Nominal morphology
2.1. Thematic nouns
2.2. Athematic nouns
3. Pronominal morphology
4. Adjectival morphology
5. Verbal morphology
5.1. The present active
5.2. The past active
5.3. The perfect active
1. Introduction and essential sound laws
Getic was the Indo-European language spoken by the tribes known as the Getae (Ancient Greek: Γέται), who lived in modern-day Romania and Bulgaria near the Black Sea, north and south of the Danube River. Most ancient sources identify the Getae with the Dacians and as members of the Thracians. For my purposes I’ve assumed that although these groups may have been related, they did not necessarily speak the same language.
During Augustus’ reign, the Roman Empire conquered the Getae living south of the Danube River and eventually incorporated their land into the province of Moesia. According to a poem framed as a letter to a friend, the Roman poet Ovid learned Getic and composed a “little book” of nationalistic poetry in the language while sentenced to relegatio in Tomis (modern-day Constanța) during the early first century CE. This thread will describe the sub-Danubian Getic language spoken that would have been familiar to Ovid.
As I've imagined it here, the Getic language resembles some of its neighbors around the Black Sea: it is a satem language, it merged voiced aspirated and plain voiced stops, and it resolved dental-dental clusters to *-st-. I also applied some sound changes reported to have operated in Dacian, such as the diphthongization of *é. Getic has a complicated accentual system, partially because it preserves aspects of the PIE system and partially because of the effects of laryngeals.
What follows is an account of the major sound laws which operate to produce Getic from Proto-Indo-European.
Decoupling of syllabic resonants: Before the syllabic resonants *m̥ n̥ r̥ l̥ an epenthetic vowel was inserted. After labialized consonants, the vowel was *u; otherwise, it was *i. Original sequences of syllabic resonant plus laryngeal resulted in *(i/u)HR through metathesis.
Sievers’s Law: After a consonant cluster, *y w became *iy uw.
Boukólos Rule: Labialized velar stops lost their labialization before rounded vowels and *w.
Assibilation in coronal-plosive clusters: Before another stop, coronal stops became *s or *z depending on the voicing of the following stop.
Ruki Rule: *s was backed to the fricative *š after *r w ḱ ǵ ǵʰ k g gʰ kʷ gʷ gʷʰ y, except word-finally. Clusters of *rs and *sr merged into *š.
Laryngeal coloring and merger: The laryngeals caused the typical coloring. All three laryngeals then collapsed into a single phoneme *h, which was lost word-initially. When syllabic between two consonants, this phoneme *h became *a.
Hirt’s Law and corollary: The accent was retracted to the leftmost syllable containing *h. If *h preceded the vowel, the newly-accented syllable received either high pitch (on a short vowel) or falling pitch (on a long vowel); if *h followed the vowel, the vowel was lengthened and received rising pitch. In syllables in which Hirt's Law did not operate, original accented long vowels and diphthongs became characterized by falling pitch.
Loss of laryngeals: *h was lost, lengthening all preceding vowels.
Wódr Rule: *w was lost before rounded vowels except intervocalically.
Vowel merger: Before *w and sporadically before *m, *o oː became *u uː. The diphthongs *oy oːy became *e eː. All other instances of *o oː merged with *a aː.
Merger of voiced stop series: The breathy voiced stops merged with the plain voiced stops.
Satemization and Matasović's Law: The palatal stops affricated to *ts dz, except sporadically before *m n r l, where merger with the plain velar stops was possible; *ǵ was more likely to become *g than *ḱ to become *k. Before *m n l, *ts dz became *s z; before *r, *ts dz became *str dr. All labialized velar stops lost their labialization. Before front vowels, the plain velar stops affricated to *tʃ dʒ.
Winter’s Law: Before a voiced stop or affricate, short vowels became long. In addition, when followed by a voiced stop or affricate, *z ž disappeared and lengthened a preceding vowel.
Osthoff’s Law: Before any two consonants, long vowels became short. Affricates counted as single consonants for this rule. Osthoff’s Law remained productive in Getic until after the loss of its geminate consonants (see below).
Affricate simplification: The clusters *sts stʃ šts štʃ became *sː ʃː sʲː sʲː. This rule remained in effect to operate on secondary palatal affricates from clusters of velar stops plus *y (see below). Finally, *ts dz became *s z.
Vowel-breaking and compensatory lengthening: Accented short *é became *í after *m n r l y w when word-initial or after a consonant, *yá in a closed syllable, and *yé elsewhere.
Nasalization: Before a nasal, *i iː e eː lowered to *e eː a aː, except sporadically after *y. Nasals in word-final position or before another consonant were lost, nasalizing and lengthening the previous vowel; if the vowel was stressed, it either received falling pitch (if originally short) or retained its original pitch (if originally long).
e-raising: Before a syllable containing *i iː, *e eː became *i iː unless immediately followed by *r.
Palatalization: Before *y, all consonants were geminated, coronals were palatalized, and velars affricated to *tʃ dʒ. Before *i iː e eː y, *š became *ʃ; elsewhere, it was further backed to *x. *y was then lost after consonants.
Glide developments: The sequences *yy ww became *dʒː bː. The treatment of sequences containing glides and vowels varied slightly between the dialects spoken north and south of the Danube. (More information and examples to come.)
Lenition: Between two vowels, the second being unstressed, plosives and *s became voiced. All geminate consonants became short.
Dactyl Rule: If a syllable before the antepenult was accented, the accent moved to the antepenult; newly accented short vowels received high pitch, and long vowels received falling pitch.
Spondee Rule: Before a long vowel, falling pitch became rising pitch.
These changes result in the following phonemic inventory:
/m n nʲ/ m n ň
/p b t d tʲ dʲ tʃ dʒ k g/ p b t d ť ď č ǧ k g
/v s z sʲ zʲ ʃ x/ v s z š ž ś h
/l r j/ l r y
/i iː e eː ẽː a aː ãː u uː ũː/ i ī e ē ę a ā ą u ū ų
/ej eːj ew eːw aj aːj aw aːw/ ei ēi eu ēu ai āi au āu
In most words, one syllable received the accent, which involved an increase in the pitch of that syllable’s vowel. Short accented vowels are said to have high pitch. Long vowels and diphthongs could have either falling or rising pitch, depending on whether the first or second half of the syllabic nucleus was accented. High pitch is written as á, falling as â ái âi, and rising as ǎ aí ǎi.
Some sample vocabulary highlighting the sound changes given above:
brǎdēr ‘brother’ < *bʰréh₂tēr
ďěgą ‘earth’ < *dʰéǵōm
dūgatêr ‘daughter’ < *dʰugh₂tḗr
gūkálas ‘oxherd’ < *gʷoukʷolos
íhsas ‘bear’ < *h₂ŕ̥tḱos
šárdīas ‘shepherd’ < *ḱérdhyos
ťármę ‘border’ < *térmn̥
vą̂das ‘wind’ < *h₂wéh₁ntos
yūgą̂ ‘yoke’ < *yugóm
śistás ‘broken’ < *skidtós
vēráderas ‘truthful’ < *weh₁róteros
bérid ‘he carries’ < *bʰéreti
ēdá ‘I know’ < *wóydh₂e
If the sort of thing interests you, it might be a fun exercise to work out which sound laws operated on which words, and why for example vą̂das has falling pitch on the root.
Postscriptum: Thanks to anyone who's read this through; I hope you've enjoyed it. Soon I'll post about nominal morphology, and later a bit about allophony and dialects. I'd like to conclude by recognizing Dewrad's conlang Wenetic as a major source of inspiration for this project, not only in content but also in presentation. I expect someone familiar with his work will recognize the debts I owe, but hopefully not find Getic too derivative.
1. Introduction and essential sound laws
2. Nominal morphology
2.1. Thematic nouns
2.2. Athematic nouns
3. Pronominal morphology
4. Adjectival morphology
5. Verbal morphology
5.1. The present active
5.2. The past active
5.3. The perfect active
1. Introduction and essential sound laws
Getic was the Indo-European language spoken by the tribes known as the Getae (Ancient Greek: Γέται), who lived in modern-day Romania and Bulgaria near the Black Sea, north and south of the Danube River. Most ancient sources identify the Getae with the Dacians and as members of the Thracians. For my purposes I’ve assumed that although these groups may have been related, they did not necessarily speak the same language.
During Augustus’ reign, the Roman Empire conquered the Getae living south of the Danube River and eventually incorporated their land into the province of Moesia. According to a poem framed as a letter to a friend, the Roman poet Ovid learned Getic and composed a “little book” of nationalistic poetry in the language while sentenced to relegatio in Tomis (modern-day Constanța) during the early first century CE. This thread will describe the sub-Danubian Getic language spoken that would have been familiar to Ovid.
As I've imagined it here, the Getic language resembles some of its neighbors around the Black Sea: it is a satem language, it merged voiced aspirated and plain voiced stops, and it resolved dental-dental clusters to *-st-. I also applied some sound changes reported to have operated in Dacian, such as the diphthongization of *é. Getic has a complicated accentual system, partially because it preserves aspects of the PIE system and partially because of the effects of laryngeals.
What follows is an account of the major sound laws which operate to produce Getic from Proto-Indo-European.
Decoupling of syllabic resonants: Before the syllabic resonants *m̥ n̥ r̥ l̥ an epenthetic vowel was inserted. After labialized consonants, the vowel was *u; otherwise, it was *i. Original sequences of syllabic resonant plus laryngeal resulted in *(i/u)HR through metathesis.
Sievers’s Law: After a consonant cluster, *y w became *iy uw.
Boukólos Rule: Labialized velar stops lost their labialization before rounded vowels and *w.
Assibilation in coronal-plosive clusters: Before another stop, coronal stops became *s or *z depending on the voicing of the following stop.
Ruki Rule: *s was backed to the fricative *š after *r w ḱ ǵ ǵʰ k g gʰ kʷ gʷ gʷʰ y, except word-finally. Clusters of *rs and *sr merged into *š.
Laryngeal coloring and merger: The laryngeals caused the typical coloring. All three laryngeals then collapsed into a single phoneme *h, which was lost word-initially. When syllabic between two consonants, this phoneme *h became *a.
Hirt’s Law and corollary: The accent was retracted to the leftmost syllable containing *h. If *h preceded the vowel, the newly-accented syllable received either high pitch (on a short vowel) or falling pitch (on a long vowel); if *h followed the vowel, the vowel was lengthened and received rising pitch. In syllables in which Hirt's Law did not operate, original accented long vowels and diphthongs became characterized by falling pitch.
Loss of laryngeals: *h was lost, lengthening all preceding vowels.
Wódr Rule: *w was lost before rounded vowels except intervocalically.
Vowel merger: Before *w and sporadically before *m, *o oː became *u uː. The diphthongs *oy oːy became *e eː. All other instances of *o oː merged with *a aː.
Merger of voiced stop series: The breathy voiced stops merged with the plain voiced stops.
Satemization and Matasović's Law: The palatal stops affricated to *ts dz, except sporadically before *m n r l, where merger with the plain velar stops was possible; *ǵ was more likely to become *g than *ḱ to become *k. Before *m n l, *ts dz became *s z; before *r, *ts dz became *str dr. All labialized velar stops lost their labialization. Before front vowels, the plain velar stops affricated to *tʃ dʒ.
Winter’s Law: Before a voiced stop or affricate, short vowels became long. In addition, when followed by a voiced stop or affricate, *z ž disappeared and lengthened a preceding vowel.
Osthoff’s Law: Before any two consonants, long vowels became short. Affricates counted as single consonants for this rule. Osthoff’s Law remained productive in Getic until after the loss of its geminate consonants (see below).
Affricate simplification: The clusters *sts stʃ šts štʃ became *sː ʃː sʲː sʲː. This rule remained in effect to operate on secondary palatal affricates from clusters of velar stops plus *y (see below). Finally, *ts dz became *s z.
Vowel-breaking and compensatory lengthening: Accented short *é became *í after *m n r l y w when word-initial or after a consonant, *yá in a closed syllable, and *yé elsewhere.
Nasalization: Before a nasal, *i iː e eː lowered to *e eː a aː, except sporadically after *y. Nasals in word-final position or before another consonant were lost, nasalizing and lengthening the previous vowel; if the vowel was stressed, it either received falling pitch (if originally short) or retained its original pitch (if originally long).
e-raising: Before a syllable containing *i iː, *e eː became *i iː unless immediately followed by *r.
Palatalization: Before *y, all consonants were geminated, coronals were palatalized, and velars affricated to *tʃ dʒ. Before *i iː e eː y, *š became *ʃ; elsewhere, it was further backed to *x. *y was then lost after consonants.
Glide developments: The sequences *yy ww became *dʒː bː. The treatment of sequences containing glides and vowels varied slightly between the dialects spoken north and south of the Danube. (More information and examples to come.)
Lenition: Between two vowels, the second being unstressed, plosives and *s became voiced. All geminate consonants became short.
Dactyl Rule: If a syllable before the antepenult was accented, the accent moved to the antepenult; newly accented short vowels received high pitch, and long vowels received falling pitch.
Spondee Rule: Before a long vowel, falling pitch became rising pitch.
These changes result in the following phonemic inventory:
/m n nʲ/ m n ň
/p b t d tʲ dʲ tʃ dʒ k g/ p b t d ť ď č ǧ k g
/v s z sʲ zʲ ʃ x/ v s z š ž ś h
/l r j/ l r y
/i iː e eː ẽː a aː ãː u uː ũː/ i ī e ē ę a ā ą u ū ų
/ej eːj ew eːw aj aːj aw aːw/ ei ēi eu ēu ai āi au āu
In most words, one syllable received the accent, which involved an increase in the pitch of that syllable’s vowel. Short accented vowels are said to have high pitch. Long vowels and diphthongs could have either falling or rising pitch, depending on whether the first or second half of the syllabic nucleus was accented. High pitch is written as á, falling as â ái âi, and rising as ǎ aí ǎi.
Some sample vocabulary highlighting the sound changes given above:
brǎdēr ‘brother’ < *bʰréh₂tēr
ďěgą ‘earth’ < *dʰéǵōm
dūgatêr ‘daughter’ < *dʰugh₂tḗr
gūkálas ‘oxherd’ < *gʷoukʷolos
íhsas ‘bear’ < *h₂ŕ̥tḱos
šárdīas ‘shepherd’ < *ḱérdhyos
ťármę ‘border’ < *térmn̥
vą̂das ‘wind’ < *h₂wéh₁ntos
yūgą̂ ‘yoke’ < *yugóm
śistás ‘broken’ < *skidtós
vēráderas ‘truthful’ < *weh₁róteros
bérid ‘he carries’ < *bʰéreti
ēdá ‘I know’ < *wóydh₂e
If the sort of thing interests you, it might be a fun exercise to work out which sound laws operated on which words, and why for example vą̂das has falling pitch on the root.
Postscriptum: Thanks to anyone who's read this through; I hope you've enjoyed it. Soon I'll post about nominal morphology, and later a bit about allophony and dialects. I'd like to conclude by recognizing Dewrad's conlang Wenetic as a major source of inspiration for this project, not only in content but also in presentation. I expect someone familiar with his work will recognize the debts I owe, but hopefully not find Getic too derivative.