Möxali

Möxale is the primary language of the Lavondu people of T'ugü, and a lingua franca throughout surrounding coastal regions of Fountainhead, particularly those connected to T'ugü by maritime trade. Möxale emerged as a mixed language in bilingual communities of Xiyeru in T'ugü following the Xiyeru Exodus.

Phonology
Möxale is most commonly spoken as a second language, by many peoples over a wide area throughout its long history, so there is much variation in the phonology. There are, however, some patterns and general tendencies common to all speakers or particular subgroups. In order to view Möxale as a whole, its phonology is discussed here using the romanisation of its core phonemic inventory, which is common to almost all varieties of the language. Each phoneme can be understood as representing a diaphoneme, an abstracted phonological unit with an analogous phoneme in each variety of Möxale which may have different phonetic realisations depending on the dialect or historical stage. Particular phonetic varieties are also given in square brackets where relevant.

Front-back allophones
Some consonants have a pair of major allophones determined by the presence of a front (ä ö e i) or back (a o u) vowel in the host syllable. These allophonic pairs or some variants of them are present in the vast majority of speakers, particularly those who speak Möxale as a first language or are in regular contact with those who do. The typical phonetic values of these allophonic pairs are shown below: Öb speakers commonly realise j as [dʒ] at the start of syllables and use bilabial and bidental fricatives [ɸ β θ ð] in all phonological environments where Lavondu and most other speakers have labiodental and dentalised sibilant fricatives [f v s̪~θ z̪~ð] for f v s z. For all speakers, v is typically a bilabial approximant when acting as the medial approximant in an onset cluster, but with no velar coarticulation (i.e. [β̞] not [w]).

Vowels
The vowels (a e i o u) are most typically realised as [ɑ e i ɔ o], though Öb tend to realise e u as [ɪ ʊ]. The vowels ä ö can vary in height quite freely between [e̞ ø] and [æ œ] depending on speaker and phonological environment, but the height hierarchy ä < ö < e is usually maintained. Adjacent vowels at the phonemic level are always separated by a non-phonemic glottal stop [ʔ], which is not represented in the native script but romanised as ⟨’⟩ within words.

Phonotactics
Möxale's syllable structure is (OR/C)V(F), where O is an obstruent (p t c k b d f s x v z j), R is an approximant (v l y) with a different place of articulation to O, C is any consonant, V is any vowel (ä ö e i a o u), and F may be any of (f s x h v z j n l). Coda n nasalises the preceding vowel and, if there is a following consonant, manifests as a nasal stop which assimilates to the place of articulation of that consonant. since v patterns both as a fricative and an approximant in the context of syllable structure, it is the only consonant that can occur as a single onset, be either consonant in an onset cluster or be a syllable coda.

Morphophonology
It is useful to analyse the structure of Möxale as having an underspecified morphophonemic level, with phonotactic constraints and other regular processes yielding the phonemic level. Morphophonemic forms are shown in |vertical bars|. There are four important phenomena; soft morphemes, inversion, vowel elision, and coda neutralisation.

Soft morphemes
Soft morphemes carry a special property which triggers regular internal sandhi processes on its initial phonological segment. This can be any vowel or one of a limited set of consonants which are underspecified and subject to a voicing rule which then yields the surface level phoneme. Soft morphemes with an initial vowel (V₂) avoid placing that vowel adjacent to a previous vowel (V₁) by a process determined by the qualities of the two vowels. The pairs of vowels (ä ö), (o u) and (e i) are labelled Ä, O and E respectively and follow a height hierarchy a < Ä < O < E. If V₁ is higher on this hierarchy than V₂, either y or v is inserted between the vowels, else the two vowels merge. Note that the height hierarchy of the vowel pairs does not necessarily coincide with the relative heights of individual vowels, thus u-o and i-e do not yield *uvo and *iye, but u and i respectively. A table of all possible combinations of vowels V₁-V₂ is shown below: The vowels in brackets are subject to regular elision as described in the next section. Note that these processes apply only to the initial segment, thus soft morphemes are free to express the full range of Möxale’s phonology elsewhere within the morpheme, e.g. |Fei| manifests as fe’i, not *fi.

Hard morphemes are those morphemes that do not undergo these processes when preceded by a vowel, either retaining their intrinsic onset consonant unchanged, or by inserting an epenthetic glottal stop before an initial vowel.

Inversion
Inversion is an ablaut process which acts on both vowels in a disyllabic stem. It developed out of an archaic stress rule in Xiri which is no longer active in Möxale, and through analogy has become a regular process triggered by certain suffixes. Vowels e i u in the first syllable are lowered to ä e o, and vowels ä e o ö in the second syllable are raised to e i u i. For example, consider the noun stem |Fedo| (clothing), which surfaces as fedo in the unmarked absolutive case:


 * |Fedo\ɪɴᴠ-yo| → fäduyo (clothing-ᴀᴄᴄ)

Inversion is only triggered by a small handful of suffixes, even within a grammatical paradigm only some suffixes may trigger inversion while others do not:


 * |Fedo-en| → fedin (clothing-ᴇʀɢ)

An inverted stem cannot be uninverted by some other morphological process. In compounds of two stems, the first stem is inverted.

Vowel elision
Stressed vowels may never be elided, but there are several circumstances which trigger the regular elision of unstressed vowels. One such circumstance is in the consonant epenthesis of vowel-initial soft morphemes described above:


 * |su-aze| → *s(o)ˈvaze → ˈsvaze

This elision is blocked by an approximant in the syllable onset:


 * |kyo-aze| ( → *ky(o)ˈvaze → kyoˈvaze

If the approximant has the same place of articulation as that preceding onset, it too is elided:


 * |eji-aze| → *ij(e)ˈyaze → *eˈj(y)aze → eˈjaze (Note also the inversion of the initial stem ije- → eji-).

Coda neutralisation
The underlying structure of a morpheme allows a final stop which is altered if it arises as a syllable coda at the phonemic level, i.e. if there is no following syllable in which the consonant can act as the onset. This applies to both hard and soft morphemes and occurs across word boundaries. Nasals neutralise into assimilatory coda n:


 * |cipäm-∅| → cipän (needle-ᴀʙꜱ)
 * |cipäm-en| → cipämen (needle-ᴇʀɢ)
 * |cipäm\ɪɴᴠ-yo| → cepeño (needle-ᴀᴄᴄ)

Plosives undergo spirantisation, in which they mutate to the corresponding fricative with the same voicedness and place of articulation: p t c k b d → f s x h v z:


 * |limok-∅| → limok (bee-ᴀʙꜱ)


 * |limok-en| → limoken (bee-ᴇʀɢ)


 * |limok\ɪɴᴠ-ä| → lemukä (bee-ᴀᴄᴄ)

Grammar
Möxale features three lexical classes distinguished by their valency, that is the number of arguments they take:


 * Nouns cannot be predicate (no arguments)
 * Stative verbs can be predicate with exactly one argument
 * Transitive verbs can be predicate with one or more arguments

Case
Nouns take an obligatory inflection for case, of which there are three which mark different morphosyntactic roles. The ergative suffix is a soft morpheme, and so its initial segment |e| is subject to vowel sandhi as described above:


 * |udok-en| (lizard-ᴇʀɢ) → udoken
 * |ana-en| (mouth-ᴇʀɢ) → anän
 * |teno-en| (salt-ᴇʀɢ) → tenin

The accusative case has two allomorphs depending on the final segment of the underlying form of the noun stem. The suffix |-yo| appears on stems ending in a vowel (ä ö e i a o u), a nasal (m n ñ g), a palatal (c x j) or h. In the case of a final nasal, |Ny| regularly neutralises to ñ, so the accusative suffix appears as -ño, while a final h similarly assimilates to the palatal yielding -xo:


 * |ibo\ɪɴᴠ-yo| (gourd-ᴀᴄᴄ) → ebuyo
 * |kixon\ɪɴᴠ-yo| (spear-ᴀᴄᴄ) → kexuño
 * |kabah\ɪɴᴠ-yo| (mountain-ᴀᴄᴄ) → kabaxo

A final palatal (c x j) or l also regularly elides the approximant, effectively manifesting the accusative as -o on these stems:


 * |dihac\ɪɴᴠ-yo| (leg-ᴀᴄᴄ) → dehaco
 * |lipel\ɪɴᴠ-yo| (tendon-ᴀᴄᴄ) → lepilo

On all other final segments (p t k b d f s v z), the accusative case takes the form -ä:


 * |anök\ɪɴᴠ-ä| (axe-ᴀᴄᴄ) → anikä
 * |iyos\ɪɴᴠ-ä| (fire-ᴀᴄᴄ) → eyusä

The allomorphy of the accusative suffix originates from a sound change which occurred between Late Xiri and Möxale in which unstressed ya yo yu wa we wi became e e i o ö ü where there was a preceding consonant, then later after an unstressed vowel shift these result in ä ä e o ö ö. This yields the change yo → ä in certain environments, of which the accusative suffix on obstruent-final stems is an example. The relative chronology of these changes indicates that plosive-final stems must have been incorporated into a very early stage of Möxale, when it was essentially just a divergent dialect of Late Xiri with a large number of Alöbi loans. Möxale's spirantisation of final plosives may have already been present as an active sound rule in this early stage, or developed as a direct result of the incorporation of plosive-final stems, as Xiri allowed coda fricatives but not plosives. The elision of -y- leaving just -o on stems ending in (c x j l) comes from Alöbi and Xiri, which both historically elided medial -y- in the same environments.

The underlying form of the stem is particularly important to consider for case inflection, compare the inflectional set taneh/taneken/tanekä (knife) to ineh/inehen/inexo (finger), from the underlying forms |tanek| and |ineh| respectively.

Number
Grammatical number can be optionally specified with a prefix with allomorphs for vowel-initial and consonant-initial stems. These correspond to roughly the same meanings as the equivalent prefixes used in Alöbi, from which the Möxale forms are derived. The dual number has slightly extended its usage from pairs of things to small complete sets of two, three and very occasionally four. For example, the zivon (64 day month) of the Avizñox calendar is sometimes referred to as kovyanä in reference to its four ovyanä "great segments" which are each incomplete and exist only as parts of the whole, hence the use of the dual number. Such situations would use the paucal number in Alöbi, which is no longer present in Möxale. The plural number represents any greater number, but unlike in Alöbi can also be used for as little as two referents, so long as those two things aren't considered to form a pair or complete set.

Number need not be overtly marked on a noun with a suffix, but the grammatical property is always carried by every instance of a noun. A number prefix is typically used when introducing a referent to discourse, to clarify which referent is being referred to, or to emphasise the number of a referent. The number of a noun is otherwise inferred from context or simply not relevant. A newly introduced referent which lacks a number prefix is usually understood to mean reference to something in general, rather than to a particular instance of it, i.e. "cats (in general)" rather than "a cat" or "the cat".

Transitive verbs
The transitive verb takes a core argument, the subject, which has its semantic role determined by a voice prefix. Depending on the verb voice, the verb may also take additional arguments. The morphological structure is an unusual mixture of Alöbi-style agglutination and Xiri morphemes and syntax.

Subject
The subject prefix is present only when an explicit subject noun is not. These come directly from Xiri and are the only remnant of the pronominal system, which was completely supplanted by the Alöbi system.

Mood
Möxale inherited most of Xiri’s evidential particles, but also adopted Alöbi’s strict verb-final structure in main clauses. This forced the evidentials in front of the verb, occupying the same position as the Xiri modal particle which was rarely used in conjunction with an evidential. The result is the merging of the evidentials and five of the Xiri irrealis moods (subjunctive, permissive, imperative, conditional, interrogative) into a single paradigm. The seven realis moods are mostly derived from evidentials and can be further categorised by type of evidence: first-hand assertion (visual sensory, non-visual sensory, inferential), second-hand report (quotative, hearsay) and third-hand statement of fact (narrative, common knowledge). Unlike Xiri, there is no unmarked indicative mood, evidentiality is necessarily implied through the mood.