Etrandish phonology

Like many other languages, Etrandish has wide variation in pronunciation, both historically (see Old Etrandish and Middle Etrandish) and from dialect to dialect. In general, however, the regional dialects of Etrandish share a largely similar (but not identical) phonological system.

Phonotactics
In Etrandish, the construction of a syllable is, O standing for Obstrudent (stop, fricative, affricate), L standing for sonorant (nasal, rhotic, lateral approximant, semivowel), and V standing for vowel. There are some limitations however.
 * Syllable-initially, two stop or affricate consonants cannot follow each other in the same syllable. Only stop-fricative and fricative-stop clusters are allowed (not counting the optional glide or rhotic that may follow).
 * Syllable-initially, fricative-stop clusters are only tolerated if they are voiceless. Clusters like are not tolerated, only.
 * In native Etrandish vocabulary, stop-fricative clusters also take place in their voiceless forms, but voiced clusters like are in fact tolerated in foreign loanwords, such as the personal name Gzaxia.
 * Only up to two sonorants may form a cluster within a single syllable. Such a cluster can only be followed by a fricative within the same syllable. It may not be followed by a stop or affricate consonant, unless it is in the following syllable.
 * Sonorant-clusters can only exist at syllable coda, not the syllable onset.
 * Some consonants - such as the velar nasal - can only take place in the syllable coda, never at the offset.
 * The placeless nasal - which only occurs in lower-class dialects - appears only at the end of words, and only when following a vowel.

Now that we know how Etrandish syllables are structured, we also need to realize that Etradish also has sandhi - in other words, if a syllable is followed by a syllable that begins with a vowel (has no initial consonants), the first syllable's last consonant gets transferred over to the second syllable. This also extends beyond word boundaries. For example, (Calder) is pronounced  in isolation, while  (may, might, possibly) is always. Come them together, and instead of, we get or. The explanation is that +  equals, not. Normally, the end result would produce a single flapped instead of the geminated trill, but since this is the combination of two words, the  gets treated as a word-initial  - a trill. The only exception to this is if the final vowel of the first word is long, in which case a flapped is produced instead.

Other than, also becomes geminated, if a word ending with it is followed by a word starting with a vowel - so long as the first word's final syllable has a short vowel, that is, as a long vowel would block the gemination. Stops, fricatives and nasals - other than - simply get transferred over without any gemination. inserts a onto the following word instead.

Consonants
''This is about Standard Etrandish phonology. If you want to read about dialects, see Etrandish dialects.''
 * appears only in loanwords, and most Etrandish-speakers have difficulty pronouncing it. Even educated speakers tend to replace it with or.
 * may be pronounced as either or .  may become
 * has a relatively large amount of allophones depending on their position:
 * Voiceless glottal approximant word-initially, and after nasals, voiceless stops and voiceless fricatives
 * Voiced glottal approximant between two vowels
 * doesn't occour after voiced stops and fricatives because causes said voiced stops and fricatives to devoice, due to regressive assimilation
 * Voiceless velar fricative between a back/central vowel and a voiceless stop, fricative or word boundary. Some speakers may use a voiceless uvular fricative  instead.
 * Voiceless palatal fricative between a front vowel and a voiceless stop, fricative or word boundary
 * Voiced velar fricative between a back / central vowel and a voiced stop or fricative. Some speakers may use a voiced uvular fricative  instead.
 * Voiced palatal fricative between a front vowel and a voiced stop or fricative.
 * The voiceless is present only in conservative speech. In every other variant, it is merged with the voiced.
 * The voiceless stops and affricates are aspirated  when not in the syllable coda or proceeding a sibilant.
 * There is no aspiration in conservative speech.
 * The voiceless stops are typically unreleased  in the syllable coda
 * In emphatic speech, they may be aspirated instead
 * The voiced stops and affricates are slack-voiced  word-initially and word-finally.
 * In the dialects of Northern Etrand, they may be completely devoiced in these environments. This is a marked feature of the stereotypical Steelhelm accent.
 * Modally voiced stops and affricates are always fully voiced in conservative speech.
 * The dental fricatives are often shifted to other positions.
 * is shifted to or even merged with  in most dialects
 * is merged with in a lot of lower-class speech. Generally,  is preserved in its original form (as ) much more widely than.
 * are preserved as not only in conservative speech, but also in most educated speech
 * The placeless nasal only exists in middle-class and lower-class speech as a merger of coda-position,  and  when no consonant follows
 * Conservative speech preserves, and  fully intact, as ,  and
 * Word-final is vocalized to  or, forming nasal diphthongs. Otherwise, it assimilates to the following stop, becoming  before labial consonants,  before velar consonants,  before all the other consonants.
 * is realized in multiple ways, depending on the envorniment
 * In Standard Etrandish - as well as in the dialect of Western Etrand -, is realized as an...
 * apico-alveolar trill word-initially and when geminated
 * apico-alveolar / retroflex flap medially when not geminated
 * pharyngealized alveolar / retroflex approximant in the syllable coda
 * In several dialects, things are different:
 * in Northern Etrand, the Guttural R reign supreme.
 * In Southern Etrand, is consistently realized as a post-alveolar / retroflex approximant  and may or may not be elided when no vowel follows (non-rhotic), based on speaker preference (speakers who elide the final  pronunce  as ).
 * In the Eastern Mountains, is realized as an alveolar / uvular trill  consistenly. The flap and approximant are never used.
 * Non-humans will usually pronounce as in their native, non-human languages:
 * Dwarves realizing as a uvular trill
 * Halflings realizing as either an alveolar tflap  or a uvular trill, depending on the context.
 * Lizardmen realizing as a retroflex approximant.

Vowels
''This is about Standard Etrandish phonology. If you want to read about dialects, see Etrandish dialects.''

Monophthongs

 * Unlike, isn't entirely back, and tends to be centralized to a certain degree . Some speakers may go as far as fronting  to , especially after palatal(ized) consonants.
 * The status of the schwa as a phoneme is questionable, because it only appears before non-prevocalic . Rather than a separate phoneme, it can be also interpreted as the allophone of  and  before non-prevocalic , creating a completementary distribution.
 * is also pronounced as when it's preceeded by another vowel (forming a vowel cluster) and proceded by a non-prevocalic.
 * In Standard Etrandish, and all dialects other than Northern Etrandish, the vowels are also merged into the / realized as a schwa  before non-prevocalic . The long vowels  become centralizing diphthongs  before non-prevocalic, while in the same context, the second element of the diphthongs  gets replaced by a schwa.
 * The cluster usually coalesced into an R-coloured vowel  when it's not followed by any vowels.
 * The vowels, and  only exist before  and
 * is primarily pronounced as or, but also has the following allophones:
 * before coda-position and
 * before coda-position clusters and
 * in diphthongs, word-finally, and also in vowel clusters following, , , or . in other words,  are pronounced as.
 * The clusters are pronounced as  or  when they aren't being followed by a vowel.
 * In Standard Etrandish - and other dialects that realize non-prevocalic as some kind of approximant, rather than a flap, trill or fricative - non-prevocalic  tends to coalesce with vowels that precede it, creating R-coloured vowels, allowing for the realization of  as.

Diphthongs

 * Standard Etrandish - and all other dialects that realize non-prevocalic as some kind of approximant, rather than a flap, trill or fricative - also has some extra dipththongs, though they are allophones of a vowel (or dipththong) followed by non-prevocalic :
 * The are respectively realized as either  or . The clusters  are respectively pronounced as  or.
 * In lower-class and middle-class speech, the nasal diphthongs exist too, as allophones of
 * does not exist in conservative speech. Conservative speech preserves, and  fully intact, as ,  and
 * Even in most non-conservative speech, only exists word-finally