Difference between revisions of "Praseo phonology"

m (Bot updated 1420394148)
m (Bot updated 1421088177)
Line 1: Line 1:
[[Category:Prasa]][[Category:Praseo language]]
+
% vim: ft=markdown fo=
  
= Consonants =
+
[[Category:Prasa]]
 +
[[Category:Praseo language]]
  
== Phoneme set and standard orthography ==
+
# Consonants
  
The following chart gives all of the consonant phonemes in Praseo in the linguistic orthography, together with their core phonetic values.
+
## Phoneme set and standard orthography
  
{|
+
The following chart gives all of the consonant phonemes in Praseo in the linguistic
!
+
orthography, together with their core phonetic values.
!Labial
+
!Dental
+
!Retroflex
+
!Palatal
+
!Velar
+
|-
+
|''Voiceless stop''
+
|''p'' [p]
+
|''t'' [t]
+
|
+
|
+
|''k'' [k]
+
|-
+
|''Voiced stop''
+
|''b'' [b]
+
|''d'' [d]
+
|
+
|
+
|''g'' [g]
+
|-
+
|''Voiceless fricative''
+
|
+
|''s'' [s]
+
|''ś'' [ʂ]
+
|''š'' [ɕ]
+
|''h'' [x]
+
|-
+
|''Voiced fricative''
+
|
+
|''z'' [z]
+
|
+
|
+
|
+
|-
+
|''Voiceless affricate''
+
|
+
|''ts'' [ts]
+
|
+
|''č'' [tɕ]
+
|
+
|-
+
|''Voiced affricate''
+
|
+
|
+
|
+
|''j'' [dʑ]
+
|
+
|-
+
|''Nasal''
+
|''m'' [m]
+
|''n'' [n]
+
|
+
|
+
|
+
|-
+
|''Liquid''
+
|
+
|''l'' [l]
+
|
+
|
+
|
+
|-
+
|
+
|
+
|''r'' [r]
+
|
+
|
+
|
+
|-
+
|''Semivowel''
+
|''o'' [w]
+
|
+
|
+
|''y'' [j]
+
|
+
|}
+
  
The most peculiar aspect of the orthography is that the semivowel [w] is consistently rendered as ''o'', regardless of position in the word. Thus, the word ''oirde'' is [wirde], and ''taoma'' is [tawma]. The phoneme [j], on the other hand, is only spelled as ''y'' when it is a syllable onset; as the second part of a diphthong it's spelled ''i'', giving ''yalu'' [jalu], but ''aisa'' [ajsa].
+
                        Labial      Dental      Retroflex      Palatal    Velar
 +
----                    -------    -------    ----------      --------    ------
 +
*Voiceless stop*        *p* [p]     *t* [t]                                 *k* [k]
 +
*Voiced stop*          *b* [b]     *d* [d]                                *g* [g]
 +
*Voiceless fricative*              *s* [s]     *ś* [ʂ]        *š* [ɕ]    *h* [x]
 +
*Voiced fricative*                  *z* [z]
 +
*Voiceless affricate*              *ts* [ts]                  *č* [tɕ]
 +
*Voiced affricate*                                              *j* [dʑ]
 +
*Nasal*                *m* [m]    *n* [n]
 +
*Liquid*                            *l* [l]
 +
                                    *r* [r]
 +
*Semivowel*            *o* [w]                                *y* [j]
  
= Vowels =
+
The most peculiar aspect of the orthography is that the semivowel [w] is consistently
 +
rendered as *o*, regardless of position in the word. Thus, the word *oirde* is [wirde], and
 +
*taoma* is [tawma]. The phoneme [j], on the other hand, is only spelled as *y* when it is a
 +
syllable onset; as the second part of a diphthong it's spelled *i*, giving *yalu* [jalu],
 +
but *aisa* [ajsa].
  
Praseo has four oral vowels, three nasal vowels, and three diphthongs which can also be nasalized.
+
# Vowels
  
{|
+
Praseo has four oral vowels, three nasal vowels, and three diphthongs which can also be
!align="center"|Oral vowels
+
nasalized.
!align="center"|Nasal vowels
+
|-
+
|align="center"|i u
+
|align="center"|
+
|-
+
|align="center"|e
+
|align="center"|ẽ [ɛ̃] õ [ɔ̃]
+
|-
+
|align="center"|a
+
|align="center"|ã [ã]
+
|}
+
  
== Diphthongs ==
+
Oral vowels        Nasal vowels
 +
-------------      --------------
 +
    i      u
 +
    e              ẽ [ɛ̃]    õ [ɔ̃]
 +
        a              ã [ã]
  
{|
+
## Diphthongs
!align="center"|Oral diphthongs
+
!align="center"|Nasal diphthongs
+
|-
+
|align="center"|ei [ej]
+
|align="center"|ẽi [ɛ̃j]
+
|-
+
|align="center"|eo [ew]
+
|align="center"|ẽo [ɛ̃w]
+
|-
+
|align="center"|ao [aw]
+
|align="center"|ão [ãw]
+
|}
+
  
All of the diphthongs in Praseo are falling diphthongs, and the exact quality of the second part of the diphthong is not explicitly specified. I've shown them above as semivowels, but the second part of ''ei'' may be as low as [ɪ], and the second part of ''ao'' may be [ɔ]. In nasalized diphthongs, the nasalization generally holds throughout the nasalized syllable.
+
Oral diphthongs       Nasal diphthongs
 +
-----------------    -------------------
 +
ei [ej]               ẽi [ɛ̃j]
 +
eo  [ew]              ẽo [ɛ̃w]
 +
ao [aw]              ão [ãw]
  
== Stress placement ==
+
All of the diphthongs in Praseo are falling diphthongs, and the exact quality of the second
 +
part of the diphthong is not explicitly specified. I've shown them above as semivowels, but
 +
the second part of *ei* may be as low as [ɪ], and the second part of *ao* may be [ɔ]. In
 +
nasalized diphthongs, the nasalization generally holds throughout the nasalized syllable.
  
Praseo has a very strong word-level stress, which is predictable based on the morphology and phonology of the word. Stress in a Praseo word falls on the ''penultimate'' syllable, unless:
+
## Stress placement
 +
 
 +
Praseo has a very strong word-level stress, which is predictable based on the morphology and
 +
phonology of the word. Stress in a Praseo word falls on the *penultimate* syllable, unless:
  
 
* The final syllable is morphologically marked as stressed (as in a few noun cases and verb forms)
 
* The final syllable is morphologically marked as stressed (as in a few noun cases and verb forms)
Line 134: Line 65:
 
In these cases, the final syllable is stressed.
 
In these cases, the final syllable is stressed.
  
Morphologically stressed final syllables are indicated with an acute accent on the vowel. This is the only way to distinguish between, for example, ''razi'' [ˈra.zi] ("heads (nom.pl.)") and ''razí'' [raˈzi] ("head (construct sg.)"). Other cases in which the final syllable is stressed are not marked, since the final stress can be inferred from the nasalization of the final vowel or the presence of a diphthong.
+
Morphologically stressed final syllables are indicated with an acute accent on the vowel.
 +
This is the only way to distinguish between, for example, *razi* [ˈra.zi] \("heads
 +
(nom.pl.)") and *razí* [raˈzi] \("head (construct sg.)"). Other cases in which the final
 +
syllable is stressed are not marked, since the final stress can be inferred from the
 +
nasalization of the final vowel or the presence of a diphthong.
  
Because Praseo stress is so regular, it is not usually indicated in my phonetic transcriptions.
+
Because Praseo stress is so regular, it is not usually indicated in my phonetic
 +
transcriptions.
  
= Book orthography =
+
# Book orthography
  
The orthography given above is used for precise transliterations of Praseo, and is used throughout this site and in the Praseo lexicon. However, a slightly different orthography is used for names and bits of text which occur in ''Storm Bride'' and other books featuring the Praseo language. (See [[Why I Use a Book Orthography]].)
+
The orthography given above is used for precise transliterations of Praseo, and is used
 +
throughout this site and in the Praseo lexicon. However, a slightly different orthography is
 +
used for names and bits of text which occur in *Storm Bride* and other books featuring the
 +
Praseo language. (See [[Why I Use a Book Orthography]].)
  
 
For the Praseo book orthography, the following substitutions are made:
 
For the Praseo book orthography, the following substitutions are made:
  
{|
+
Standard orthography   Book orthography
!align="center"|Standard orthography
+
----------------------  ------------------
!align="center"|Book orthography
+
    ś                         s
|-
+
    š                         sh
|align="center"|ś
+
    č                         ch
|align="center"|s
+
                            e
|-
+
    ã                         a
|align="center"|š
+
    õ                         o
|align="center"|sh
+
|-
+
|align="center"|č
+
|align="center"|ch
+
|-
+
|align="center"|
+
|align="center"|e
+
|-
+
|align="center"|ã
+
|align="center"|a
+
|-
+
|align="center"|õ
+
|align="center"|o
+
|}
+
  
As a consequence of these substitutions, there are no diacritics found in the book orthography of Praseo: nasal vowels are not distinguished from oral vowels, and the phonemes ''s'' [s] and ''ś'' [ʂ] are not distinguished.
+
As a consequence of these substitutions, there are no diacritics found in the book
 +
orthography of Praseo: nasal vowels are not distinguished from oral vowels, and the phonemes
 +
*s* [s] and *ś* [ʂ] are not distinguished.
  
= Allophonic processes =
+
# Allophonic processes
  
== Nasal place assimilation ==
+
## Nasal place assimilation
  
Sound changes early in the history of Praseo resulted in most coda nasals being lost and converted to nasalization on the preceding vowel. However, when the coda nasal was immediately followed by an obstruent (stop or fricative), the nasal is retained in pronunciation and orthography, and is pronounced at the same place of articulation as the following stop. These homoorganic stops are always written with ''n'' except when followed by a labial, in which case they are written with ''m''.
+
Sound changes early in the history of Praseo resulted in most coda nasals being lost and
 +
converted to nasalization on the preceding vowel. However, when the coda nasal was
 +
immediately followed by an obstruent (stop or fricative), the nasal is retained in
 +
pronunciation and orthography, and is pronounced at the same place of articulation as the
 +
following stop. These homoorganic stops are always written with *n* except when followed by
 +
a labial, in which case they are written with *m*.
  
{|
+
Orthography   Pronunciation
!align="center"|Orthography
+
-------------  ---------------
!align="center"|Pronunciation
+
    mb             [mb]
|-
+
    nd             [nd]
|align="center"|mb
+
                [ɳʂ]
|align="center"|[mb]
+
    nj             [ɲʑ]
|-
+
    ng             [ŋg]
|align="center"|nd
+
|align="center"|[nd]
+
|-
+
|align="center"|
+
|align="center"|[ɳʂ]
+
|-
+
|align="center"|nj
+
|align="center"|[ɲʑ]
+
|-
+
|align="center"|ng
+
|align="center"|[ŋg]
+
|}
+
  
Vowels preceding such syllable-final nasals are still nasalized, but they are written without the tilde.
+
Vowels preceding such syllable-final nasals are still nasalized, but they are written
 +
without the tilde.
  
== Nasal vowel spreading ==
+
## Nasal vowel spreading
  
Vowel nasalization spreads to the left in Praseo until it meets a voiceless obstruent or a word boundary. Thus a word with a final nasal vowel which does not contain a voiceless obstruent, the ''entire'' word may be nasalized. As suggested by the vowel chart above, vowels in Praseo are lowered under nasalization: ''i'' and ''e'' become '''', and ''u'' becomes ''õ''. This applies to allophonically nasalized vowels as well as underlying nasalized vowels; however, allophonically nasalized ''i'' and ''u'' are generally spelled using the underlying vowel, if it can be recovered.
+
Vowel nasalization spreads to the left in Praseo until it meets a voiceless obstruent or a
 +
word boundary. Thus a word with a final nasal vowel which does not contain a voiceless
 +
obstruent, the *entire* word may be nasalized. As suggested by the vowel chart above, vowels
 +
in Praseo are lowered under nasalization: *i* and *e* become **, and *u* becomes *õ*. This
 +
applies to allophonically nasalized vowels as well as underlying nasalized vowels; however,
 +
allophonically nasalized *i* and *u* are generally spelled using the underlying vowel, if it
 +
can be recovered.
  
{|
+
Orthography   Pronunciation   Remarks
!align="center"|Orthography
+
-------------  ---------------  --------
!align="center"|Pronunciation
+
ezande         [ɛ̃zãnde]       No vowels are written with the tilde; nasalization of preceding vowels is implied by the *n*.
!Remarks
+
zulõ           [zɔ̃lɔ̃]         Nasalization spreads to first *u*, pronounced [ɔ̃], but the spelling from the root form *zulu* is maintained.
|-
+
urhõ           [urxɔ̃]         Initial *u* is not nasalized, as the voiceless obstruent *h* blocks further nasalization.
|align="center"|ezande
+
|align="center"|[ɛ̃zãnde]
+
|No vowels are written with the tilde; nasalization of preceding vowels is implied by the ''n''.
+
|-
+
|align="center"|zulõ
+
|align="center"|[zɔ̃lɔ̃]
+
|Nasalization spreads to first ''u'', pronounced [ɔ̃], but the spelling from the root form ''zulu'' is maintained.
+
|-
+
|align="center"|urhõ
+
|align="center"|[urxɔ̃]
+
|Initial ''u'' is not nasalized, as the voiceless obstruent ''h'' blocks further nasalization.
+
|}
+
  
== Coronal place assimilation ==
+
## Coronal place assimilation
  
Coronal obstruents (dental, retroflex, and palatal) assimilate for place across intervening vowels. This place assimilation spreads leftward across vocoids, but stops when interrupted by a non-assimilating segment such as a non-coronal or a coronal liquid (''l'' or ''r'').
+
Coronal obstruents (dental, retroflex, and palatal) assimilate for place across intervening
 +
vowels. This place assimilation spreads leftward across vocoids, but stops when interrupted
 +
by a non-assimilating segment such as a non-coronal or a coronal liquid (*l* or *r*).
  
A word such as ''niśa'' is pronounced [ɳiʂa], with the initial ''n'' assimilated to a retroflex in accordance with the following ''ś''. This can apply to multiple segments across multiple syllables: the word ''dusiśa'' is [ɖuʂiʂa], with all three coronal obstruents pronounced as retroflex. Note, however, that this is a form of the word ''dusu'', which in its root form reveals that the underlying segments are non-retroflex, and the orthography reflects this, marking only the last ''ś'' in the word as retroflex. More generally, since ''ś'' is the only phonemic retroflex, there is no orthographic indication of retroflexion on the stops ''t'' or ''d'', and ''s'' is only written as the retroflex ''ś'' if it is phonemically retroflex.
+
A word such as *niśa* is pronounced [ɳiʂa], with the initial *n* assimilated to a retroflex
 +
in accordance with the following *ś*. This can apply to multiple segments across multiple
 +
syllables: the word *dusiśa* is [ɖuʂiʂa], with all three coronal obstruents pronounced as
 +
retroflex. Note, however, that this is a form of the word *dusu*, which in its root form
 +
reveals that the underlying segments are non-retroflex, and the orthography reflects this,
 +
marking only the last *ś* in the word as retroflex. More generally, since *ś* is the only
 +
phonemic retroflex, there is no orthographic indication of retroflexion on the stops *t* or
 +
*d*, and *s* is only written as the retroflex *ś* if it is phonemically retroflex.
  
The palatals ''č'', ''j'', and ''š'', are also assimilated to other coronals. However, the affrication of the stops ''č'' and ''j'' is entirely automatic: if an underlying ''č'' is assimilated to a dental, then it becomes ''t'' (the affrication is lost), and if an underlying ''t'' is assimilated to a palatal it becomes ''č'' (a palatal affricate).
+
The palatals *č*, *j*, and *š*, are also assimilated to other coronals. However, the
 +
affrication of the stops *č* and *j* is entirely automatic: if an underlying *č* is
 +
assimilated to a dental, then it becomes *t* (the affrication is lost), and if an underlying
 +
*t* is assimilated to a palatal it becomes *č* (a palatal affricate).
  
 
Examples:
 
Examples:
  
{|
+
Orthography   Pronunciation   Remarks
!align="center"|Orthography
+
-------------  ---------------  --------
!align="center"|Pronunciation
+
dusiśa         [ɖuʂiʂa]       Retroflex articulation spreads to the left from the underlying *ś*. Orthography indicates underlying form.
!Remarks
+
ličitsu       [lititsu]       Underlying palatal *č* is assimilated to [t] and from following *ts*; affrication is lost. Orthography indicates underlying form.
|-
+
nuči           [ɲutɕi]         Underlying *n* is assimilated to [ɲ] from following *č*
|align="center"|dusiśa
+
|align="center"|[ɖuʂiʂa]
+
|Retroflex articulation spreads to the left from the underlying ''ś''. Orthography indicates underlying form.
+
|-
+
|align="center"|ličitsu
+
|align="center"|[lititsu]
+
|Underlying palatal ''č'' is assimilated to [t] and from following ''ts''; affrication is lost. Orthography indicates underlying form.
+
|-
+
|align="center"|nuči
+
|align="center"|[ɲutɕi]
+
|Underlying ''n'' is assimilated to [ɲ] from following ''č''
+
|}
+
  
== Voice assimilation ==
+
## Voice assimilation
  
Obstruent clusters in Praseo always agree in voice, and the first element in the cluster is changed to agree in voice with the second. (This does not occur with clusters of obstruents and sonorants, however.) This occasionally generates phones which do not otherwise occur, such as the voiced retroflex spirant [ʐ], an allophone of ''ś'', or the voiced palatal fricative [ʑ], an allophone of ''š''.
+
Obstruent clusters in Praseo always agree in voice, and the first element in the cluster is
 +
changed to agree in voice with the second. (This does not occur with clusters of obstruents
 +
and sonorants, however.) This occasionally generates phones which do not otherwise occur,
 +
such as the voiced retroflex spirant [ʐ], an allophone of *ś*, or the voiced palatal
 +
fricative [ʑ], an allophone of *š*.
  
== Affricate simplification ==
+
## Affricate simplification
  
If an affricate occurs in a consonant cluster, it is reduced to its fricative part. The spelling of the affricate is generally unchanged in this scenario. This includes clusters with sonorants, so that ''ačru'' is pronounced [aɕru].
+
If an affricate occurs in a consonant cluster, it is reduced to its fricative part. The
 +
spelling of the affricate is generally unchanged in this scenario. This includes clusters
 +
with sonorants, so that *ačru* is pronounced [aɕru].
  
== Palatal absorption ==
+
## Palatal absorption
  
When the finite verbal ending ''-ya'' is attached to a stem ending in a velar, the velar becomes palatalized (see below). The ''y'' of the verbal ending is often absorbed into this palatal, though in careful speech they may still be distinguished. It is always present in the orthography. Thus a work like ''ejya'' is usually pronounced simply [edʑa], but when enunciated carefully is [edʑja].
+
When the finite verbal ending *-ya* is attached to a stem ending in a velar, the velar
 +
becomes palatalized (see below). The *y* of the verbal ending is often absorbed into this
 +
palatal, though in careful speech they may still be distinguished. It is always present in
 +
the orthography. Thus a work like *ejya* is usually pronounced simply [edʑa], but when
 +
enunciated carefully is [edʑja].
  
= Morphophonemic alternations =
+
# Morphophonemic alternations
  
The following alternations began as allophonic alternations in an earlier stage of Praseo, but have become morphologically conditioned following the loss of some aspects of their conditioning environment. Unlike many of the allophonic alternations above, these alternations are always reflected in the orthography.
+
The following alternations began as allophonic alternations in an earlier stage of Praseo,
 +
but have become morphologically conditioned following the loss of some aspects of their
 +
conditioning environment. Unlike many of the allophonic alternations above, these
 +
alternations are always reflected in the orthography.
  
== Palatalization ==
+
## Palatalization
  
The velars ''k'', ''g'', and ''h'' are palatalized to ''č'', ''j'', and ''š'' when they occur at the end of a stem, and one of the following affixes is attached to the stem:
+
The velars *k*, *g*, and *h* are palatalized to *č*, *j*, and *š* when they occur at the
 +
end of a stem, and one of the following affixes is attached to the stem:
  
* The plural ending ''-i''
+
* The plural ending *-i*
* The feminine singular ending ''-e''
+
* The feminine singular ending *-e*
* The accusative endings ''-ẽo'' and ''-ẽi''
+
* The accusative endings *-ẽo* and *-ẽi*
* The finite verbal ending ''-ya''
+
* The finite verbal ending *-ya*
* The participles ending ''-eo'' and ''-eśa''
+
* The participles ending *-eo* and *-eśa*
  
Thus, we have the following forms derived from the word ''aku'' "knot":
+
Thus, we have the following forms derived from the word *aku* "knot":
  
* aku ''knot''
+
* aku *knot*
* ači ''knots''
+
* ači *knots*
* ačẽo ''knot (acc.sg.)''
+
* ačẽo *knot (acc.sg.)*
* ačya ''to tie''
+
* ačya *to tie*
* ačeo ''tying''
+
* ačeo *tying*
  
However, as a trap for the unwary, it is ''not'' simply the case that these velars are palatalized before any front vowel. The following front vowel affixes (among others) do ''not'' trigger palatalization:
+
However, as a trap for the unwary, it is *not* simply the case that these velars are
 +
palatalized before any front vowel. The following front vowel affixes (among others) do
 +
*not* trigger palatalization:
  
* The instrumental plural ''-ẽi''
+
* The instrumental plural *-ẽi*
* The construct case endings ''-ei'' and ''''
+
* The construct case endings *-ei* and **
  
So for ''aku'' the following non-palatalized forms occur:
+
So for *aku* the following non-palatalized forms occur:
  
* akẽi ''with knots (instr.pl)''
+
* akẽi *with knots (instr.pl)*
* akí ''knot (cons.sg)''
+
* akí *knot (cons.sg)*
  
Throughout the rest of the grammar, affixes indicated whether or not they trigger palatalization when they are introduced.
+
Throughout the rest of the grammar, affixes indicated whether or not they trigger
 +
palatalization when they are introduced.
  
== Affrication ==
+
## Affrication
  
The dental stops ''t'' and ''d'' at the end of a stem are affricated into ''ts'' and ''z'' (historically ''dz'', but simplified in Praseo) respectively when the following affix is one of the following:
+
The dental stops *t* and *d* at the end of a stem are affricated into *ts* and *z*
 +
(historically *dz*, but simplified in Praseo) respectively when the following affix is one
 +
of the following:
  
* The plural ending ''-i''
+
* The plural ending *-i*
* The accusative endings ''-ẽo'' and ''-ẽi''
+
* The accusative endings *-ẽo* and *-ẽi*
* The finite verbal ending ''-ya''
+
* The finite verbal ending *-ya*
* The participle ending ''-eo''
+
* The participle ending *-eo*
  
Thus, the word ''rada'' "leader, chief" has the following forms:
+
Thus, the word *rada* "leader, chief" has the following forms:
  
* rada ''leader''
+
* rada *leader*
* razi ''leaders''
+
* razi *leaders*
* razẽoa ''leader (acc.sg.)''
+
* razẽoa *leader (acc.sg.)*
* razya ''to lead''
+
* razya *to lead*
* razeo ''leading''
+
* razeo *leading*
  
When a given stem forms both a U-class noun and another noun in some other word class, the final consonant of the stem is affricated in like manner throughout the U-class paradigm. Thus, the stem ''rad-'' gives us both ''rada'' "leader" and ''razu'' "head (of the body)", and the word ''razu'' uses the stem ''raz-'' in all cases and numbers. Thus you have the following forms, for contrast:
+
When a given stem forms both a U-class noun and another noun in some other word class, the
 +
final consonant of the stem is affricated in like manner throughout the U-class paradigm.
 +
Thus, the stem *rad-* gives us both *rada* "leader" and *razu* "head (of the body)", and the
 +
word *razu* uses the stem *raz-* in all cases and numbers. Thus you have the following
 +
forms, for contrast:
  
{|
+
Case       *rada* "leader"    *razu* "head"
!Case
+
-----      ----------------    --------------
!''rada'' "leader"
+
Acc. sg.   razẽoa             razẽo
!''razu'' "head"
+
Acc. pl.   razẽoi             razẽi
|-
+
Dat. sg.   radoa               razú
|Acc. sg.
+
Dat. pl.   raduśi             razuśi
|razẽoa
+
|razẽo
+
|-
+
|Acc. pl.
+
|razẽoi
+
|razẽi
+
|-
+
|Dat. sg.
+
|radoa
+
|razú
+
|-
+
|Dat. pl.
+
|raduśi
+
|razuśi
+
|}
+
  
As this illustrates, the final ''z'' of the stem is preserved throughout the paradigm of ''razu'', even in those cases which do not trigger affrication of the ''d'' in the paradigm of ''rada''. This serves to distinguish the dative plural forms, which would otherwise be identical. See the full declension of ''razu'' [[Praseo%20nouns#U-Class%20Nouns|on the noun declension page]].
+
As this illustrates, the final *z* of the stem is preserved throughout the paradigm of
 +
*razu*, even in those cases which do not trigger affrication of the *d* in the paradigm of
 +
*rada*. This serves to distinguish the dative plural forms, which would otherwise be
 +
identical. See the full declension of *razu* [on the noun declension page](Praseo
 +
nouns#U-Class Nouns).

Revision as of 10:43, 12 January 2015

% vim: ft=markdown fo=

  1. Consonants
    1. Phoneme set and standard orthography

The following chart gives all of the consonant phonemes in Praseo in the linguistic orthography, together with their core phonetic values.

                       Labial      Dental      Retroflex       Palatal     Velar

------- ------- ---------- -------- ------
  • Voiceless stop* *p* [p] *t* [t] *k* [k]
  • Voiced stop* *b* [b] *d* [d] *g* [g]
  • Voiceless fricative* *s* [s] *ś* [ʂ] *š* [ɕ] *h* [x]
  • Voiced fricative* *z* [z]
  • Voiceless affricate* *ts* [ts] *č* [tɕ]
  • Voiced affricate* *j* [dʑ]
  • Nasal* *m* [m] *n* [n]
  • Liquid* *l* [l]
                                   *r* [r]
  • Semivowel* *o* [w] *y* [j]

The most peculiar aspect of the orthography is that the semivowel [w] is consistently rendered as *o*, regardless of position in the word. Thus, the word *oirde* is [wirde], and

  • taoma* is [tawma]. The phoneme [j], on the other hand, is only spelled as *y* when it is a

syllable onset; as the second part of a diphthong it's spelled *i*, giving *yalu* [jalu], but *aisa* [ajsa].

  1. Vowels

Praseo has four oral vowels, three nasal vowels, and three diphthongs which can also be nasalized.

Oral vowels         Nasal vowels

--------------
   i       u
    e               ẽ [ɛ̃]    õ [ɔ̃]
       a               ã [ã]
    1. Diphthongs
Oral diphthongs        Nasal diphthongs

-------------------
ei  [ej]               ẽi [ɛ̃j]
eo  [ew]               ẽo [ɛ̃w]
ao  [aw]               ão [ãw]

All of the diphthongs in Praseo are falling diphthongs, and the exact quality of the second part of the diphthong is not explicitly specified. I've shown them above as semivowels, but the second part of *ei* may be as low as [ɪ], and the second part of *ao* may be [ɔ]. In nasalized diphthongs, the nasalization generally holds throughout the nasalized syllable.

    1. Stress placement

Praseo has a very strong word-level stress, which is predictable based on the morphology and phonology of the word. Stress in a Praseo word falls on the *penultimate* syllable, unless:

  • The final syllable is morphologically marked as stressed (as in a few noun cases and verb forms)
  • The final syllable is nasalized
  • The final syllable contains a diphthong

In these cases, the final syllable is stressed.

Morphologically stressed final syllables are indicated with an acute accent on the vowel. This is the only way to distinguish between, for example, *razi* [ˈra.zi] \("heads (nom.pl.)") and *razí* [raˈzi] \("head (construct sg.)"). Other cases in which the final syllable is stressed are not marked, since the final stress can be inferred from the nasalization of the final vowel or the presence of a diphthong.

Because Praseo stress is so regular, it is not usually indicated in my phonetic transcriptions.

  1. Book orthography

The orthography given above is used for precise transliterations of Praseo, and is used throughout this site and in the Praseo lexicon. However, a slightly different orthography is used for names and bits of text which occur in *Storm Bride* and other books featuring the Praseo language. (See Why I Use a Book Orthography.)

For the Praseo book orthography, the following substitutions are made:

Standard orthography    Book orthography

------------------
   ś                          s
   š                          sh
   č                          ch
   ẽ                          e
   ã                          a
   õ                          o

As a consequence of these substitutions, there are no diacritics found in the book orthography of Praseo: nasal vowels are not distinguished from oral vowels, and the phonemes

  • s* [s] and *ś* [ʂ] are not distinguished.
  1. Allophonic processes
    1. Nasal place assimilation

Sound changes early in the history of Praseo resulted in most coda nasals being lost and converted to nasalization on the preceding vowel. However, when the coda nasal was immediately followed by an obstruent (stop or fricative), the nasal is retained in pronunciation and orthography, and is pronounced at the same place of articulation as the following stop. These homoorganic stops are always written with *n* except when followed by a labial, in which case they are written with *m*.

Orthography    Pronunciation

---------------
   mb              [mb]
   nd              [nd]
   nś             [ɳʂ]
   nj              [ɲʑ]
   ng              [ŋg]

Vowels preceding such syllable-final nasals are still nasalized, but they are written without the tilde.

    1. Nasal vowel spreading

Vowel nasalization spreads to the left in Praseo until it meets a voiceless obstruent or a word boundary. Thus a word with a final nasal vowel which does not contain a voiceless obstruent, the *entire* word may be nasalized. As suggested by the vowel chart above, vowels in Praseo are lowered under nasalization: *i* and *e* become *ẽ*, and *u* becomes *õ*. This applies to allophonically nasalized vowels as well as underlying nasalized vowels; however, allophonically nasalized *i* and *u* are generally spelled using the underlying vowel, if it can be recovered.

Orthography    Pronunciation   Remarks

--------------- --------
ezande         [ɛ̃zãnde]        No vowels are written with the tilde; nasalization of preceding vowels is implied by the *n*.
zulõ           [zɔ̃lɔ̃]          Nasalization spreads to first *u*, pronounced [ɔ̃], but the spelling from the root form *zulu* is maintained.
urhõ           [urxɔ̃]          Initial *u* is not nasalized, as the voiceless obstruent *h* blocks further nasalization.
    1. Coronal place assimilation

Coronal obstruents (dental, retroflex, and palatal) assimilate for place across intervening vowels. This place assimilation spreads leftward across vocoids, but stops when interrupted by a non-assimilating segment such as a non-coronal or a coronal liquid (*l* or *r*).

A word such as *niśa* is pronounced [ɳiʂa], with the initial *n* assimilated to a retroflex in accordance with the following *ś*. This can apply to multiple segments across multiple syllables: the word *dusiśa* is [ɖuʂiʂa], with all three coronal obstruents pronounced as retroflex. Note, however, that this is a form of the word *dusu*, which in its root form reveals that the underlying segments are non-retroflex, and the orthography reflects this, marking only the last *ś* in the word as retroflex. More generally, since *ś* is the only phonemic retroflex, there is no orthographic indication of retroflexion on the stops *t* or

  • d*, and *s* is only written as the retroflex *ś* if it is phonemically retroflex.

The palatals *č*, *j*, and *š*, are also assimilated to other coronals. However, the affrication of the stops *č* and *j* is entirely automatic: if an underlying *č* is assimilated to a dental, then it becomes *t* (the affrication is lost), and if an underlying

  • t* is assimilated to a palatal it becomes *č* (a palatal affricate).

Examples:

Orthography    Pronunciation   Remarks

--------------- --------
dusiśa         [ɖuʂiʂa]        Retroflex articulation spreads to the left from the underlying *ś*. Orthography indicates underlying form.
ličitsu        [lititsu]       Underlying palatal *č* is assimilated to [t] and from following *ts*; affrication is lost. Orthography indicates underlying form.
nuči           [ɲutɕi]         Underlying *n* is assimilated to [ɲ] from following *č*
    1. Voice assimilation

Obstruent clusters in Praseo always agree in voice, and the first element in the cluster is changed to agree in voice with the second. (This does not occur with clusters of obstruents and sonorants, however.) This occasionally generates phones which do not otherwise occur, such as the voiced retroflex spirant [ʐ], an allophone of *ś*, or the voiced palatal fricative [ʑ], an allophone of *š*.

    1. Affricate simplification

If an affricate occurs in a consonant cluster, it is reduced to its fricative part. The spelling of the affricate is generally unchanged in this scenario. This includes clusters with sonorants, so that *ačru* is pronounced [aɕru].

    1. Palatal absorption

When the finite verbal ending *-ya* is attached to a stem ending in a velar, the velar becomes palatalized (see below). The *y* of the verbal ending is often absorbed into this palatal, though in careful speech they may still be distinguished. It is always present in the orthography. Thus a work like *ejya* is usually pronounced simply [edʑa], but when enunciated carefully is [edʑja].

  1. Morphophonemic alternations

The following alternations began as allophonic alternations in an earlier stage of Praseo, but have become morphologically conditioned following the loss of some aspects of their conditioning environment. Unlike many of the allophonic alternations above, these alternations are always reflected in the orthography.

    1. Palatalization

The velars *k*, *g*, and *h* are palatalized to *č*, *j*, and *š* when they occur at the end of a stem, and one of the following affixes is attached to the stem:

  • The plural ending *-i*
  • The feminine singular ending *-e*
  • The accusative endings *-ẽo* and *-ẽi*
  • The finite verbal ending *-ya*
  • The participles ending *-eo* and *-eśa*

Thus, we have the following forms derived from the word *aku* "knot":

  • aku *knot*
  • ači *knots*
  • ačẽo *knot (acc.sg.)*
  • ačya *to tie*
  • ačeo *tying*

However, as a trap for the unwary, it is *not* simply the case that these velars are palatalized before any front vowel. The following front vowel affixes (among others) do

  • not* trigger palatalization:
  • The instrumental plural *-ẽi*
  • The construct case endings *-ei* and *-í*

So for *aku* the following non-palatalized forms occur:

  • akẽi *with knots (instr.pl)*
  • akí *knot (cons.sg)*

Throughout the rest of the grammar, affixes indicated whether or not they trigger palatalization when they are introduced.

    1. Affrication

The dental stops *t* and *d* at the end of a stem are affricated into *ts* and *z* (historically *dz*, but simplified in Praseo) respectively when the following affix is one of the following:

  • The plural ending *-i*
  • The accusative endings *-ẽo* and *-ẽi*
  • The finite verbal ending *-ya*
  • The participle ending *-eo*

Thus, the word *rada* "leader, chief" has the following forms:

  • rada *leader*
  • razi *leaders*
  • razẽoa *leader (acc.sg.)*
  • razya *to lead*
  • razeo *leading*

When a given stem forms both a U-class noun and another noun in some other word class, the final consonant of the stem is affricated in like manner throughout the U-class paradigm. Thus, the stem *rad-* gives us both *rada* "leader" and *razu* "head (of the body)", and the word *razu* uses the stem *raz-* in all cases and numbers. Thus you have the following forms, for contrast:

Case *rada* "leader" *razu* "head"


---------------- --------------

Acc. sg. razẽoa razẽo Acc. pl. razẽoi razẽi Dat. sg. radoa razú Dat. pl. raduśi razuśi

As this illustrates, the final *z* of the stem is preserved throughout the paradigm of

  • razu*, even in those cases which do not trigger affrication of the *d* in the paradigm of
  • rada*. This serves to distinguish the dative plural forms, which would otherwise be

identical. See the full declension of *razu* [on the noun declension page](Praseo nouns#U-Class Nouns).