Atlantic/Older version: Difference between revisions

Jump to navigation Jump to search
Line 38: Line 38:
/l/ was later reintroduced into the language through Arabic loans and later Latin and Greek learned reborrowings – cf. ''luua'' "dialect" from Arabic لغة ''luḡa''.
/l/ was later reintroduced into the language through Arabic loans and later Latin and Greek learned reborrowings – cf. ''luua'' "dialect" from Arabic لغة ''luḡa''.


Atlantic palatalized /k ɡ/ as most Western Romance languages did, to /ts dz/; /ts/ also resulted from <small>-TI-</small> and <small>-TR-</small> (but not <small>-STR-</small>) clusters. These were later deaffricated to /s z/ in most dialects, but a few Numidian ones still retain the affricates. See <small>RATIŌNEM</small> > ''raçon'' "reason, cause" /raˈtson/ > /raˈson/, <small>TRIA</small> > ''çìa'' "three (disj.)" /ˈtsi.a/ > /ˈsi.a/, <small>GELĀRE</small> > ''giuò'' "to freeze" /dziˈwo/ > /ziˈwo/. <small>-DI-</small> developed differently depending on whether it was followed by a front or by a non-front vowel. Before non-front vowels, the palatal element was lost so that it developed as regular intervocalic /d/ (see below); before front vowels, it palatalized to /dz/ much like <small>-TI-</small> did; cf. <small>NVMIDIAM</small> > *Numida > ''Numìua'' "Numidia" /nuˈmiwa/ and <small>NVMIDIĒNSEM</small> > ''numigens'' /numiˈdzens/ > /numiˈzens/ "Numidian".<br/>As mentioned before, <small>-S-</small> blocked <small>-TR-</small> from shifting to /ts/; this /s/ was, in this environment, backed to /ʃ/, orthographically denoted by a preceding '''i''', as in <small>CASTRVM</small> > *castr > */ˈkastur/ > ''caistur'' "city" /ˈkaʃtur/<ref>Phonemic /ʃ/, however, only arose because of successive palatalizations, and before /tr/ only because of learned Latin or Greek borrowings, as ''stratòs'' /straˈtos/ "army" - cf. <small>EXTRĀNEVM</small> > *istroniu > ''istroin'' /iʃˈtroɲ/ "foreign".</ref>.
Atlantic palatalized /k ɡ/ as most Western Romance languages did, to /ts dz/; /ts/ also resulted from <small>-TI-</small> and <small>-TR-</small> (but not <small>-STR-</small>) clusters. These were later deaffricated to /s z/ in most dialects, but a few Numidian ones still retain the affricates. See <small>RATIŌNEM</small> > ''raçon'' "reason, cause" /raˈtson/ > /raˈson/, <small>TRIA</small> > ''cia'' "three (disj.)" /ˈtsi.a/ > /ˈsi.a/, <small>GELĀRE</small> > ''giuò'' "to freeze" /dziˈwo/ > /ziˈwo/. <small>-DI-</small> developed differently depending on whether it was followed by a front or by a non-front vowel. Before non-front vowels, the palatal element was lost so that it developed as regular intervocalic /d/ (see below); before front vowels, it palatalized to /dz/ much like <small>-TI-</small> did; cf. <small>NVMIDIAM</small> > *Numida > ''Numìua'' "Numidia" /nuˈmiwa/ and <small>NVMIDIĒNSEM</small> > ''numigens'' /numiˈdzens/ > /numiˈzens/ "Numidian".<br/>As mentioned before, <small>-S-</small> blocked <small>-TR-</small> from shifting to /ts/; this /s/ was, in this environment, backed to /ʃ/, orthographically denoted by a preceding '''i''', as in <small>CASTRVM</small> > *castr > */ˈkastur/ > ''càistur'' "city" /ˈkaʃtur/<ref>Phonemic /ʃ/, however, only arose because of successive palatalizations, and before /tr/ only because of learned Latin or Greek borrowings, as ''stratòs'' /straˈtos/ "army" - cf. <small>EXTRĀNEVM</small> > *istroniu > ''istroin'' /iʃˈtroɲ/ "foreign".</ref>.


Intervocalic lenition of stops also follows the Western Romance pattern: unvoiced stops become voiced and voiced stops become fricatives, with original /d/ becoming */ð/ > /w/ and original /ɡ/ becoming */ɣ/ > /w/ or */ɣ/ > /j/ according to nearby vowels:
Intervocalic lenition of stops also follows the Western Romance pattern: unvoiced stops become voiced and voiced stops become fricatives, with original /d/ becoming */ð/ > /w/ and original /ɡ/ becoming */ɣ/ > /w/ or */ɣ/ > /j/ according to nearby vowels:
: <small>CAPERE</small> > *cabi > ''cab'' "to make an effort to understand"
: <small>CAPERE</small> > *cabi > ''cab'' "to make an effort to understand"
: <small>VĪTAM</small> > ''uaida'' "life"
: <small>VĪTAM</small> > ''uaida'' "life"
: <small>CRVCEM</small> > *crugi > ''ecrug'' "cross"
: <small>CRVCEM</small> > *crugi > ''carug'' "cross"
: <small>QVOQVE</small> > *pubi > ''pub'' "also"
: <small>QVOQVE</small> > *pubi > ''pub'' "also"
: <small>CABALLVM</small> > *caual > ''cauài'' "horse"
: <small>CABALLVM</small> > *caual > ''cauài'' "horse"
: <small>PEDEM</small> > *piði > ''piu'' "foot" /piu̯/
: <small>PEDEM</small> > *piði > ''piu'' "foot" /piu̯/
: <small>FRIGIDVM</small> > */wriɣiðu/ > */urijið/ > ''uriu'' "cold"
: <small>FRIGIDVM</small> > */wriɣiðu/ > */urijið/ > ''urìu'' "cold"
: <small>MAGVM</small> > */maɣu/ > ''mau'' "nomad"<ref>Originally "magician, fortune-teller", this term came to identify wandering fortune-tellers.</ref>
: <small>MAGVM</small> > */maɣu/ > ''mau'' "nomad"<ref>Originally "magician, fortune-teller", this term came to identify wandering fortune-tellers.</ref>
A preceding nasal prevented the consonant from being lenited, while a preceding /r/ or /l/ didn't, e.g. <small>DIEM MARTIS</small> > ''gimàrdit'' "Tuesday".
A preceding nasal prevented the consonant from being lenited, while a preceding /r/ or /l/ didn't, e.g. <small>DIEM MARTIS</small> > ''gimàrdit'' "Tuesday".
/kt/ and /pt/ clusters were resolved by turning the backmost element to /s/, i.e. to /st/ and /ps/ respectively, see <small>LACTEM</small> > ''nast'' "milk"; <small>APTVM</small> > ''aps'' "ready".
/kt/ and /pt/ clusters were resolved by turning the backmost element to /s/, i.e. to /st/ and /ps/ respectively, see <small>LACTEM</small> > ''nast'' "milk"; <small>APTVM</small> > ''aps'' "ready".


<small>VI-</small> /wi/ was reduced to /j/ if a vowel followed, as e.g. in <small>VIĀTICVM</small> > ''iodig'' "travel".<br/>Sporadic posttonic syncope led to a range of newer clusters, which were often resolved through assimilation. For example, /mn/ assimilated to /nn/ and /nm/ to /mm/, as in <small>FĒMINAM</small> > *fēmna > ''fenna'' "woman"; <small>ANIMAM</small> > *anma > ''amma'' "person". Most modern dialects have further degeminated them (Senegal Riparian dialects are the main exception, as are a few ones in the Atlas valleys of Numidia), but they are still represented in the orthography.
<small>VI-</small> /wi/ was reduced to /j/ if a vowel followed, as e.g. in <small>VIĀTICVM</small> > ''iòdig'' "travel".<br/>Sporadic posttonic syncope led to a range of newer clusters, which were often resolved through assimilation. For example, /mn/ assimilated to /nn/ and /nm/ to /mm/, as in <small>FĒMINAM</small> > *fēmna > ''fenna'' "woman"; <small>ANIMAM</small> > *anma > ''amma'' "person". Most modern dialects have further degeminated them (Senegal Riparian dialects are the main exception, as are a few ones in the Atlas valleys of Numidia), but they are still represented in the orthography.


===Mauritanian palatalization===
===Mauritanian palatalization===
The so-called Mauritanian palatalization (''palataligeoçon muridoinensa'') is considered in Atlantic linguistics the main isogloss between Mauritanian and Numidian dialects; this change likely started in the 17th century and was successfully completed in the span of a few generations in the territories of Mauritania. Mauritanian palatalization, both progressive and regressive, was triggered by all instances of /j/, affecting nearby consonants (a very similar change before /i/ happened before and is reflected in all modern Atlantic dialects); the orthography still unambiguously reflects the situation as for a given word, Mauritanian dialects will have a palatalized consonant (not in the phonetic sense of "palatalized") while Numidian dialects will have /j/ and a "regular" consonant.<br/>Mauritanian palatalization resulted in the following changes:
The so-called Mauritanian palatalization (''palataligeoçon muridoinensa'') is considered in Atlantic linguistics the main isogloss between Mauritanian and Numidian dialects; this change likely started in the 17th century and was successfully completed in the span of a few generations in the territories of Mauritania. Mauritanian palatalization, both progressive and regressive, was triggered by all instances of /j/, affecting nearby consonants (a very similar change before /i/ happened before and is reflected in all modern Atlantic dialects); the orthography still unambiguously reflects the situation as for a given word, Mauritanian dialects will have a palatalized consonant (not in the phonetic sense of "palatalized") while Numidian dialects will have /j/ and a "regular" consonant.<br/>Mauritanian palatalization resulted in the following changes:
: /t d/ > /tʃ dʒ/, cf. ''fuit'' "he was" (Mauritanian /futʃ/, Numidian /fui̯t/), ''naidur'' "shore" (Maur. /ˈnadʒur/, Num. /ˈnai̯dur/).
: /t d/ > /tʃ dʒ/, cf. ''fuit'' "he was" (Mauritanian /futʃ/, Numidian /fui̯t/), ''nàidur'' "shore" (Maur. /ˈnadʒur/, Num. /ˈnai̯dur/).
: /n/ > /ɲ/, cf. ''uiçain'' "neighbour" (Maur. /wiˈsaɲ/, Num. /wiˈsai̯n/).
: /n/ > /ɲ/, cf. ''uiçain'' "neighbour" (Maur. /wiˈsaɲ/, Num. /wiˈsai̯n/).
: /r/ > /ʒ/, cf. ''Uiolarea'', capital city of the Atlantic Provinces (Maur. /ujoˈlaʒa/, Num. /ujoˈlarja/). This change also happened, as part of Mauritanian palatalization, before /i/; Numidian still has /r/ before /i/, cf. ''riduòi'' "ritual" Maur. /ʒiˈdwoi̯/, Num. /riˈdwoi̯/.
: /r/ > /ʒ/, cf. ''Uiolarea'', capital city of the Atlantic Provinces (Maur. /ujoˈlaʒa/, Num. /ujoˈlarja/). This change also happened, as part of Mauritanian palatalization, before /i/; Numidian still has /r/ before /i/, cf. ''riduòi'' "ritual" Maur. /ʒiˈdwoi̯/, Num. /riˈdwoi̯/.