Minhast

= Overview = Minhast, the national language of Minhai, is spoken by nearly 59 million people. Approximately 1 million speakers live in expatriate communities throughout the globe, with the largest concentrations residing in the U.S., Xayda, Mexico, the Middle East, Kallaxwān and Canada. Significant numbers also exist in Southeast Asia and Norhern Europe.

Minhast is divided into two major dialects. Upper Minhast, which is centered in the highlands of Kilmay Rī, Ešked (Ekšed), and Attum Attar; the northeastern coastal provinces of Iskamharat and Perim-Sin; and the Capital Region, consisting of Āš-min-Gāl, Ankussūr, Huruk, Nammadīn, and Kered. Lower Minhast is spoken mainly in the southeastern coastal provinces of Neskud, Yaxparim, Senzil, and Rēgum. The two dialects differ chiefly in phonetics and the lexicon, with Lower Minhast containing some Spanish loanwords absorbed during the Colonial Occupation ("Ayawaškum min Karkarakt"). Otherwise, the two dialects are mutually intelligible. Nevertheless, it is Upper Minhast that is the standardized form of the language, used in government, commerce, and the media. A now extinct dialect, known as Šarmakandast, had an even larger number of Spanish loandwords, and certain changes to certain elements of its grammar, namely its numeric system, were significantly influenced by the colonialists during this time. Šarmakandast died out during the early 1970's, supplanted mostly by Upper Minhast speakers.

Minhast possesses a complex grammar, demonstrated in particular by the elaborate polysynthetic morphology of its verbal system. The Minhast verb inflects not only for tense and aspect, but can inflect to indicate mood, modality, causation, potentiality, intensity, and other functions. The verb also possesses a well-developed set of pronominal affixes used to cross-reference the core arguments of a clause. These affixes indicate both gender and number of the nouns they cross-reference, an essential function as Minhast nouns themselves do not have any markings to indicate these two classifications. Additionally, the verb can carry out three other operations, that of noun incorporation, antipassivation, and applicative formation, used by speakers for discourse purposes such as backgrounding previously established information and for promoting oblique noun phrases to core status. This polysynthetic characteristic can lead to very long verbs that can express an entire sentence. To demonstrate, the English phrase, "You did not even try to get them to reconsider the matter with this evidence" requires only three words in Minhast: "Keman yattah tašnišpipsaryentinasummatittaharu", meaning literally "To them the evidence, not-try-cause-return-look.at-yet-matter-with-it.you-did." The verb "tašnišpipsaryentinasummatittaharu", when parsed to its individual morphemes, yields "ta-šn-šp-b-sar-yenti-nasum-mat-tittah-ar-u" (neg.-conative-causative-resumptive-look.at-yet-matter-instr.applicative-3rd.inanim.sg.patient/2nd.sg.agent-past-transitive).

Minhast is classified as an ergative language. There are two clause types, intransitive and transitive. Each clause type has a set of core argument types; for the transitive clause, there are two core arguments, the agent and patient. For the intransitive clause, there is only one core argument, the subject. In a nominative-accusative language, such as Latin, the agent and subject are grouped together and are inflected with one case (e.g. -us in mundus "world"), while the patient, aka object,is marked with a different case (e.g. -um in mundum). In an ergative language such as Minhast, it is the subject and patient that are grouped together, and both receive null marking (e.g. nassa--Ø "brother"). In contrast, the agent is grouped separately, marked with the ergative clitic -de (e.g. nassa-de) or one of its allomorphs.

Word order is SOV. In single clause sentences, this word order is free, although the verb rarely deviates from its clause-final position. Deviation from the unmarked SOV word order is used for discourse purposes; an argument that is to be focused is fronted to the head of the clause. During clause-combining operations, the verb obligatorily remains in the final position, but the other arguments of the clause, core and oblique, still display free word order.

= Phonology =

Syllabic Structure and Phonemic Interactions
Minhast words are subject to complex morphophonemic changes resulting from interactions with other morphemes occuring in the word. The verb is particularly complex in the various sound changes that may occur as a result of noun incorporation as well as the aggluginative processes involved in conjugation or other inflectional processes. These phonemic changes can be broken down according to the following classifications:
 * 1) Assimilation
 * 2) Metathesis
 * 3) Syncope
 * 4) Epenthesis
 * 5) Voicing/Devoicing

 No syllable can have a consonant cluster of more than two consonants. Syncope can be applied only if a biconsonantal cluster is formed, and the vowel is not a part of a heavy syllable (i.e. the vowel is long, or it occurs in a VCC sequence).   No Minhast word can have an initial consonant cluster. After any initial consonant cluster results from one or more of the possible morphophonemic alternations described below, an epenthetic is automatically appended to the head of the word to form the permissible iCC- pattern.  An epenthetic vowel is always inserted between two syllables if combining the syllables results in a triconsonantal cluster. The default epenthetic vowel is, but the other 3 vowels may also be used, depending on multiple factors (e.g. vowel harmony, an underlying quiescent initial vowel as part of the attached morpheme, etc.) 

Minhast has a strong tendency to form intermedial clusters, either or, providing that Rules #1-#3 are observed. If necessary, an epenthetic vowel may be added before or after the syllable to create these syllabic patterns, e.g. e.g. kanut-maris-kar- &gt;&gt;  -kant-(u)-maris-kar   The tendency to form intermedial consonant clusters creates complex assimilation interactions that nevertheless are predictable and almost always regular. These interactions are illustrated in Table X below:  

These complex morphophonemic interactions operate according to the general phonological principals outlined below:
 * 1)                     No syllable can have a consonant cluster of more than two consonants. Syncope can be applied only if a biconsonantal cluster is formed, and the vowel is not a part of a heavy syllable (i.e. the vowel is long, or it occurs in a VCC sequence).
 * 2) No Minhast word can have an initial consonant cluster. After any initial consonant cluster results from one or more of the possible morphophonemic alternations described below, an epenthetic is automatically appended to the head of the word to form the permissible iCC- pattern.
 * 3)                      An epenthetic vowel is always inserted between two syllables if combining the syllables results in a triconsonantal cluster. The default epenthetic vowel is, but the other 3 vowels may also be used, depending on multiple factors (e.g. vowel harmony, an underlying quiescent initial vowel as part of the attached morpheme, etc.)
 * 4)                      Minhast has a strong tendency to form intermedial clusters, either or, providing that Rules #1-#3 are observed. If necessary, an epenthetic vowel may be added before or after the syllable to create these syllabic patterns, e.g.
 * 5)                       The tendency to form intermedial consonant clusters creates complex assimilation interactions that nevertheless are predictable and almost always regular. These interactions are illustrated in Table X below:


 * 1)                       Vowels are classified according to a "weak-strong" gradient, where the "strong" vowels are more resistant to syncope than neighboring "weak(er)" vowels. All long vowels are by definition "strong", so the weak-strong gradient really applies to short vowels.

Table X: Vowel Gradients In Order of Increasing Strength

 * 1) The shape of a -CVCVC- syllable may contract either to a -CCVC- or -CVCC- pattern, depending on the strength gradients of the vowels with respect to one another. The -CaCaC- syllable pattern is the only one that does not contract. Syllables consisting of the same vowels may appear in either -CCVC- or -CVCC- patterns; the pattern they resolve to is influenced by interactions from surrounding syllables. These contractions are summarized in Table X:


 * 1)                       A verb root or an incorporated noun tends to lose one or more vowels to form at least one biconsonant cluster. The vowel that is lost depends on its strength gradient in relation to the noun of the neighboring syllable.
 * 2)                       With the exception of pattern -CaCa-, when two adjoining syllables have vowels within the same gradient, vocalic syncope resolves to CVCC.
 * 3)                      The pattern (C)VVCC always resolves to (C)VCC
 * 4)                      Compared to nominal and verbal roots, inflectional morphemes (e.g. theme, aspect, tense, person, etc) are resistant to syncope because this may lead to the inflectional morpheme to be changed beyond recognition. For example, he informed (him) (lit. he caused him to know) does not resolve to, even though this would prevent the impermissible CCV pattern from occurring. Instead, an epenthetic vowel is added before the causative affix to prevent this impermissible consonant cluster from occurring.
 * 5)                     Although inflectional morphemes do not experience syncope, they still may experience phonological changes in the form of metathesis and devoicing.
 * 6)                      Vowel devoicing occurs in CVħC, CVxC, CVsC, or CVC syllables, where C is any of the unvoiced consonants listed in Table X.
 * 7)                      Two consecutive syllables with the pattern CVCVC resolves to CVCCVC, due to the difficulty of pronouncing the  allophone in two consecutive closed syllables. Additionally, the vowel in the previous syllable may be devoiced if its adjacent consonants are voiceless, as in Example A, where the verb root vowel which occurs the voiceless consonants and, devoices to . Note also the epenthetic vowel appearing before the verb root and the 1st person incl. pl. affix , e.g.:
 * 8) *                                                        "We were (being) annoying" (lit.: annoying-we.and.you-[past]-[ imperf.]-[intrans])
 * 9) *                                                        "I was avoiding..."
 * 10)                      Dissimilation occurs in CVC-patterns involving š-Vš, resolving to s-Vš. A prime example is the number "twenty", e.g.                                            
 * 11)                      Dissimilation occurs in CVC-patterns involving mVm, resolving to nVm.
 * 1)                      Dissimilation occurs in CVC-patterns involving mVm, resolving to nVm.

= Noun Morphology =

Overview Although Minhast nouns have characteristics that make them structurally simpler than the nouns in many Indo-European languages, while at the same time having other characteristics that make them more complex.

The Minhast noun contains an inherent gender, but this is not marked on the noun but by the cross-referencing affixes of the verb. Similarly, the noun does not inflect for number; this too is marked by the verbal cross-referencing system. The noun does take case marking in the form of postpositional clitics. These clitics attach to the end of a noun phrase (NP) and indicate whether the NP is an ergative argument, or whether it is an oblique relation. As for the absolutive NP, it receives null marking.

A feature of Minhast nouns, absent in Indo-European nouns, is it can take verb-like inflection, taking markers identical to the tense and aspect affixes of the verb. This form of the noun is called the Stative. The Stative often used to emphasize that the noun in question exists in a particular point in time, not present at the moment of the speech event. More importantly, these affixes function in place of the copula: Minhast has no verb "to be" and instead uses two nouns in apposition, e.g. "Dirruk Anyar" for "Anyar is a young man". To say "Anyar was a young man", the word "dirruk must receive past tense marking, i.e. "Dirruk-ar Anyar" . The Stative is also used for identificational purposes, e.g.  "Anyar-ek"  e.g. "I am Anyar".  The Stative evolved diachronically, originally consisting of the noun or NP followed by an independant pronoun, e.g. *"Anyar yak" .  Tense-aspect markings were added later, as a result of generalizing from intransitive verb forms, e.g. "Karan-ek-ar-an" "I was sad".

Case

There are three cases in the Minhast noun: the Absolutive, Ergative, and Oblique. The Absolutive case is used for the patient of a transitive sentence, or for the subject of an intransitive sentence. It is also used as a citational or vocative case. Nouns in the Absolutive case are null-marked Ø ; no affixes or clitics are added to the noun stem to mark the Absolutive case.

In contrast to the Absolutive, the Ergative case is used to mark the agent of a transitive sentence. Nouns in the Ergative case take the clitic =de or one of its allomorphs, based on sound change interactions with the preceding phoneme. These sound change rules here described earlier in Section Xx. Table X summarizes the possible forms of the Ergative enclitic:

The Oblique case is the form used with postpositional clitics. Whereas the Ergative and Absolutive cases indicate the core arguments of a clause, the Oblique case is used for non-core arguments, delineating spatial and directional relations between and among the primary and secondary participants in the clause. The most common postpositional clitics are listed in Table X:

Personal Pronouns

The personal pronouns make three-way distinctions in syntactic role (ergative, absolutive, oblique), gender (masculine,             feminine, neuter), and a two-way distinction in number (singular, plural), and animacy (animate, inanimate). The

personal pronouns also are represented in five different forms: as independant roots used primarily for discourse functions, such as emphasis or disambiguation; as bound roots representing oblique arguments; as stative clitics attached to noun phrases; as verbal affixes performing coreferencing functions of noun phrase arguments; and as nominal affixes bound to head nouns in possessive constructs. The Minhast independant personal pronouns are divided into ergative and absolutive forms, both of which are irregularly inflected. Unlike English, the independant pronouns are infrequently used. When they appear in a clause, they serve to emphasize their coreferents or clarify the syntactic roles or attributes of the core arguments, such as gender or number. Otherwise, their use is             completely optional.

Table X gives the forms of the independant personal pronouns:

In addition to the independant ergative and absolutive personal pronouns, there is a set of bound personal pronominal roots to represent oblique arguments. Because they are bound roots, they must appear with postpositional affixes or other suffixes. The bound personal pronominal forms are illustrated

below in Table X:

The pronouns also occur as stative clitics, which are attached to noun phrases. These clitics are similar in form to the pronominal affixes of           the intransitive verb. As Minhast does not have a verb equivalent of "to be", the stative pronominal clitics' primary function is           to substitute for equational sentence-type structures, e.g.            N ē y ū nek , "I am N ē y ū n", Minhastit ạ ħ , "You are

Minhast",           satt ē k, "Here I am",            etc.  The forms of the stative pronominal            clitics are given below in Table X:

The pronominal verbal affixes are divided into two groups. One group is a set of portmenteau pronominal affixes coreferencing the ergative and absolutive arguments of the transitive clause. The other group contains pronominal markers coreferencing the absolutive argument of the intransitive clause. Because of their role in verbal morphology, the pronominal verbal affixes will be             dealt in Chapter X, Section X "Verbal Pronominal Affixes".

The final group of morphemes representing personal pronouns are the possessive pronominal suffixes. They are identical in form to the

&lt;a href="portmenteau_affixes"&gt;portmenteau pronominal affixes in the transitive verb&lt;/a&gt; ,

but here the ergative and absolutive segments of the possessive suffixes are used for different purposes. The ergative segment of the possessive pronominal suffix coreferences the possesser, while the absolutive segment coreferences the possessum. Like the portmenteau affixes in the verb, the portmenteau possessive pronominal suffixes mark gender, number, and animacy, important functions that are exploited in Minhast to disambiguate the attributes of the possessor and possessum nouns as described in Chapter X, Section X

"Possession". The possessive pronominal suffixes, without any additional affixes, by themselves represent absolutive arguments. They can take postpositional suffixes to mark them for oblique status, or they can take an ergative clitic to             mark it as the agent of a transitive clause, e.g. absolutive Aħyan-min-gelīšet

vs. ergative Aħyan-min-gelīšet=e , "...Aħyan's horse". The full paradigm of the possessive pronominal suffixes, as they have the same form as the transitive portmenteau pronominal affixes, can be referenced to in Table X of Chapter X,             "Verbal Pronominal Affixes".

Numbers Cardinal and ordinal numbers are one of the [two/XX] groups of true adjectives in the Minhast language. Minhast employs a vegisimal, i.e. base-20, counting system. Numeric expressions involve binding             the number and modified noun in a specific construct involving the  ligature:              Both cardinal and ordinal numbers can take possessive pronominal suffixes (see Part III "Syntax -              Possession" for discussion of possessive constructs), which then convey "X number of..." in the case of             cardinal numbers, and "the Xth one of/among..." for ordinals, e.g.:
 * 1) "Four of us went out there into the forest.
 * 2) "The fourth one among them went into the forest."

The numbers 1-10 even have intransitive verbal forms, meaning "There were X number of us/you/them."

The cardinal, ordinal, and verbal forms are summarized below:

Notice that for the numbers (21)and higher that the linker is inserted between the vegisimal number and the unit number. The numbers saššentāz (40) and(60) are etymologically derived from("two twenties") and("three twenties") respectively, due to regular sound changes from consonantal assimilation patterns.

Quantifiers and Interrogatives Quantifiers and Interrogatives are the second group of adjectives in Minhast. Additionally, they can stand on their own as independant nouns in their own right. They convey the sense of "some","each", "all" etc. Like numbers, these adjectives form syntactically bound constructs and thus require the ligature min, using the same syntax as that used for both numeric and possessive phrases. In addition, when used as independant nouns, quantifiers exhibit Ergative-Absolutive inflection, whereas is defective and receives Absolutive and Oblique inflections only. The Distributive form of the verb is used to indicate the distributive nature of an action or state across patients of the VP.

Table X contains a list of the most common quantifiers in use, plus rek, "each":

Two quantifiers, bāk and ādan

Possession

To express possessive phrases, Minhast uses the ligature particle -min- to link heads with their dependant arguments in structures represented in the following diagram:

[possessor] - min - [possessum]

The ligature -min- is used to bind the posessor with the possessum, e.g. ipnines-min-itar "sword" + [ligature] + "edge" i.e. "(the) sword's edge". Note that number, let alone other syntactic characteristics (e.g. gender, animacy, etc) cannot be discerned for either the head or dependant nouns. Although the gender and animacy of the head and dependant nouns are unmarked, these attributes are nevertheless known as they are inherent, albeit memorized lexical characteristics. The same cannot be said for number, which is not a fixed attribute in nouns. The default number of both head and dependant noun in this type of possessive construct is singular, but by no means is this absolute, and sometimes a plural translation of one or both nouns in the possessive phrase must be used, as is illustrated in the phrase, Birīħ-min-Hūr, "The Lions' Mountain", not "The Lion's Mountain". The correct, former reading is known only through historical and literary context.

Because of the potential for ambiguity in this type of possessive construction, an alternate structure predominates, which employs portmenteau pronominal affixes similar in form to the verbal pronominal affixes to bind to the posessum, as illustrated by the following formula:

[possessor] - min  - [possessum + ergative portmenteau pronominal affix] - t

Modern Minhast prefers this construction, so today constructions like tazer-min-erak.massešt ,

(literally "a/the bird - its feathers") predominate. The agent segment of the portmenteau pronoun, -ssešt-, refers to the possessor head noun tazer "bird", and marks it as singular and animate. The patient segment -ma- refers to the dependant noun erak and marks it as plural and inanimate. Any case clitics used to specify the word's grammatical role are appended at the end of the NP, e.g. tazer-min-erak.massešt(i)=kī "on the bird's feathers".

The portmenteau affixes are also used in expressing direct pronominal possession, e.g. iššū-tirekt "my head", or ezab-lent "his sister". Case clitics are attached after the portmenteau affixes.

In the case of nouns derived from nominalized VPs, the situation becomes more complex. In particular, nominalized transitive verbs, which are able to encode agent-patient relationships, can secondarily express possessive relationships. An example would be astekkenareft, literally "they that begat me", a formal term for "my father". Here, the portmenteau affix -ekkek- denoting the ergative 3rd person common plural and the absolutive 1st person singular, paraphrases the possessive relationship using verbal syntax to describe agent-patient relations.

= Verb Morphology =

Overview
The verb is the most complex part of Minhast grammar. Because the noun can only indicate case and deixis, the verb performs additional functions, such as specifying the definiteness, gender and number of the core arguments, whether the core arguments are collective or not, whether an action was carried out or experienced by individual members of the core arguments, etc. In addition, the verb can indicate the speaker's attitude towards the sentence, including evidentiality, assertion, etc.

The Minhast verb is organized into a template containing eight affix slots (called by traditional Minhast grammarians the šabāye). Verbal affixes, depending on their classification, fall within one of these slots. Multiple affixes may occupy a given slot, but all affixes follow a rigid order, and an affix that is assigned to one slot may not appear in another slot; to do so would result in an unintelligible verb. Many slots are themselves divided into smaller sections. Not all slots are filled at a given time - in fact, some affixes are mutually exclusive with others, whose combination would otherwise create nonsensical meanings. The šabāye are displayed in the table below, in the order of their respective positions within the verbal template:

Verb Structure
The Preverb contains two slots. The first slot is reserved for theme markers, which are used to indicate the quality or nature of the action or state; these may include such things as degree or intensity, potentiality, etc.

The second slot contains affixes used to denote the direction of an action in relation to the ergative argument in transitive sentences, or the absolutive argument in intransitive ones.

Mode Affixes
The Inverse Volitional affix is used to mark whether a verb is deliberate, or accidental. This affix interacts with the semantic characteristics of the verb root, expressing the accidental or involuntary occurance of an act in verbs which under normal circumstances indicate an action or state that is semantically deliberate. Thus, a verb like hitting carries as part of its semantic nature a deliberate act. The presence of the Inverse Volitional affix converts the deliberative nature of the act of hitting to that of an accidental or involuntary nature. Conversely, verb roots which are normally acts or states that semantically involuntary become deliberate when the Volitional affix is applied.

Prepronominal Affixes
The morphemes in this verb slot, named because they come immediately before the pronominal affixes, are also known as the "social affixes". These affixes indicate the the nature of an event or state within, between or across the core arguments of a clause. The first three affixes of this slot are mutually exclusive, but may occur in combination with one of the final two affixes. The Distributive and Partitive can combine to create certain idiomatic expressions, e.g. "I haphazardly spread it across the table".

Pronominal Affixes
The pronominal affixes present one of the greatest challenges to the students of the Minhast language due to their inherent complexity in structure and morphosyntax. These affixes serve important functions to the core arguments they coreference, such as indicating syntactic roles, gender, animacy, and number. These affixes, along with the role affixes, also serve to identify the verb as transitive or intransitive. For the transitive verb, the pronominal affixes present greater complexities than those of the intransitive verb - the transitive affixes, representing both the ergative and absolutive arguments of the clause, are portmenteau affixes; although some patterns can be discerned from this fusion of the segments representing the ergative and absolutive components, the transitive pronominal affixes are mostly irregular and have to be memorized individually. As expected, the affixes may change shape due to the sound changes created by adjacent morphemes. However, many of these sound changes deviate from the normal assimilation patterns described earlier in Chapter X "Phonology".

The pronominal affixes distinguish three genders, masculine, feminine, and neuter. The neuters are further differentiated into animate and inanimate; the masculine and feminine genders are inherently animate and thus require no special marking. These affixes also indicate singular and plural numbers. Both the masculine and the feminine 3rd person plurals have merged into one common gender, while the gender for animate and inanimate neuter nouns are still distinguished.

Due to the complexity of the transitive pronominal affixes, their full forms are summarized in Table X below:

Portmenteau Pronominal Affixes for the Transitive Verb: Ergative by Absolutive
In comparison to the transitive pronominal affixes, the affixes for the intransitive verb are much simpler. There forms are listed below in Table X:

Absolutive Pronominal Affixes for the Intransitive Verb
Tense Affixes

Aspect Affixes

Section 7: Role Affixes The Role Affix is the last obligatory slot of the verb. The affixes in this slot serve to mark the verb's transitivity, and to establish its role as either a true verb or a nominalized VP. The nominalizer suffixes -eft and -naft cannot be followed by any postverbal clitics.

Section 8: Conjunctive Affixes The Conjunctive Affixes slot contains optional affixes that serve to identify the relation of a preceding clause with that of the following clause. These clitics occupy the final position of the verb complex, and are mutually incompatible with each other. The Conjunctive clitics are all recognizable by the final -m ā  suffix, hence the Conjunctive clitics are called by Minhast grammarians "M ā -min-tallannaft" , i.e."The M ā of Binding", "The M ā  which Binds".

In addition to serving as coordinators, the Conjunctive Affixes often serve to mark following clauses as dependant on the states or events of the preceding clause. The subset of Conjunctive clitics that perform this function are aptly called the Mood Affixes.

Noun Incorporation Minhast, like many polysynthetic languages, employs extensive noun incorporation (NI), where a noun is absorbed into the verb complex. Only nominal stems may be incorporated; enclictics and inflection morphemes are excluded from the incorporataion process.

NI is used for a variety of purposes. A common function of NI is to background previously established information. Also, when the core arguments of a clause refer to human beings, native speakers prefer to retain them as core arguments throughout an entire narrative; NI, often in combination with applicative formation, will often be exploited on non-human nouns to ensure that the human nouns remain as core participants.

Nouns in an instrumental oblique role, like nouns that are patient core arguments, may be incorporated into the verb. When instrumental oblique nouns are incorporated, additional issues arise. When there is a semantic patient and an instrumental oblique noun occurring in the same sentence, it is possible that the semantic patient remains an independant word while the instrumental oblique noun is incorporated. Applicative formation cannot take place in this instance because the patient noun retains its status as the direct object of the clause - applicatives affixes are used only to indicate that a noun in an oblique role was promoted to Absolutive status. Thus, the only sign that the incorporated noun in this situation is a semantically instrumental noun is the absence of applicative markers on the verb.

Sometimes a non-instrumental oblique noun is promoted to Absolutive status, which requires the verb to take the appropriate applicative affix. An instrumental oblique noun can still be incorporated in this case, but now an ambiguous situation arises: outside of contextual cues, there are no formal grammatical structures that indicate whether the incorporated noun was originally a Patient, or whether it was originally an instrumental noun now incorporated into the verb complex. The listener must therefore rely on contextual cues to determine the incorporated noun's true role.

Antipassivation Antipassivation is a valency-decreasing process whereby the absolutive argument of a transitive verb is demoted to oblique status or is deleted from the clause, thereby demoting in turn the ergative argument to absolutive status. Antipassivation is used for discourse purposes, with the following functions: To delete the Object when it is unknown or not topic-worthy To mark an Object as indefinite by demoting it to Oblique status, usually the Instrumental case.</li> To mark the Subject as indefinite by demoting it to Absolutive status</li> To create a semantic pivot in clause combining operations.</li> </ol>

Antipassivation is accomplished by transforming the Transitive verb into an Intransitive verb (by replacing the Transitive -u Role suffix with the -an Role suffix and replacing the Erg.-Abs. portmenteau pronominal affixes eith the Abs. pronominal affixes). The clictic =de is removed from the Agent NP. The Patient may either be deleted, or the Instrumental affix -pār- or other oblique marker is added to the demoted Patient if retained.

Passive Voice Periphrasis, aka "The Pseudo-Passive"

Minhast, like many ergative languages, does not have a formal passive voice (see Dixon). However, it is capable of expressing the passive voice with a curious construction involving, paradoxically, a semantically transitive verb. This "Pseudo-Passive" construction employs the verb "to take" -itta š - and NI of a verb root. Because the S of a passive voice construction in prototypical nominative-accusative languages is usually an involuntary patient, the Inverse Volitional affix is used to change the semantically deliberate nature of -itta š - to an accidental or an involuntary event. The Pseudo-Passive construction can be illustrated by the following formula:

-ittaš - [incorporated semantically transitive verb root] - kah-

Thus, to express the equivalent of the English construction, "Jeff was assaulted last night", the Minhast equivalent becomes, literally, "Jeff took the hitting against his will last night", i.e. Dief kayyarkī ittaš nikaharan.

It must be noted, however, that the Pseudo-Passive construction occurs quite rarely in Minhast and is considered stilted by native speakers. When it does occur, it is usually found in translations of a passive phrases in texts from nominative-accusative languages. The prefered construction used to convey the same D-structure is to employ an Antipassive construction, with the indefinite noun keytān "someone", and demoting the recipient of the action to an oblique role, e.g. Keytān Diefpar kayyarkī ušnaran, "Someone hit Jeff last night."

There is a similar structure, found mostly in literature quoting oracular speech, but this structure lacks the Inverse Volitional affix:

ittaš - [incorporated semantically transitive verb root]

This structure occurs in highly marked discourse, such as in emphatic indirect commands, e.g. ittašullumtahan "You shall take the hearing" (i.e. Take heed and listen carefully!). This particular style is noticeably prevalent in quoted oracular speech found in the Minhast epics "Anyaddaddaram", "Renkun u Minsun", and "Karabtanarek"

The Split Ergative System

Although classified as an ergative language, Minhast displays some nominative-accusative morphology in its verb structure. This phenomenon, known as split ergativity, is virtually universal among ergative languages, although different languages manifest this split in different parts of their respective grammars.

Minhast marks the ergative role on NP's and the independant pronouns, using the ergative enclitic =de or one of its allophones. However, the verbal pronominal affix system clearly demonstrates a nominative-accusative pattern. The portmenteau affixes, upon closer examination, cross-reference core roles by the positions of the pronominal segments underlying the portmenteau affixes. The absolutive NP is cross-referenced by the first segment of the pronominal affix, and the ergative is cross-referenced by the segment of the second segment of the portmenteau affix. The majority of these segments, although phonologically eroded through time by sound shifts and morphophonemic interactions, otherwise have no special morpheme to differentiate their functions. Only two exceptions exist: the first variant of 3rd person singular masculine ergative has its Agent segment marked by -(n)n- ; the third person neuter inanimate singular has an Agent/Patient opposition of -maħ- and -tir- respectively.

= Dictionary = ...

= Example text = ...