Old Xšali

To Be Continued... Xshalad is still working on this article. The contents are incomplete and likely to undergo changes.

Xṣâli is a language spoken in Xshalad.

Xṣâli [ǁʃa˦˨li˧]
Period	c. -1200 YP
Spoken in	Xṣâli Tṣíkâ, aka Xšalad
Total speakers	c. 8 million
Writing system	Xṣâli Logographic Script
Classification	!Ho   Xṣâli
Typology
Basic word order	VSO, OVS
Morphology	analytic
Alignment	NOM/ACC
Credits
Created by	Xshalad

Phonology

Xṣâli’s phonology is not unusual by Akana standards, except for its non-pulmonic consonants, including clicks, which give it a very distinctive sound. Typologically the language favors complex initials and open syllables. Vowels carry a complex of tone, phonation, and nasalization.

Onset

There are forty two single consonants and thirty eight initial clusters, giving eighty possible initials (null onsets are not allowed). The initial clusters take two forms: any non-click consonant with a preceding nasal (which matches the place of articulation of the second consonant), or a plain click followed by the fricatives s, ṣ, or lh. Below are the thirty eight initial consonants. The form on the left is the default IPA pronunciation, and the form on the right is the orthographic representation used in this document. The native orthography of Xṣâli is a logo-phonemic system, which will be dealt with in a later document.

	labial	dental	lateral	postalv.	palatal	velar	glottal
plain stops	p p				c ḳ	k k	ʔ ‘
aspirated stops	pʰ ph				cʰ ḳh	kʰ kh
affricates		ts ts	tɬ tl	tʃ tṣ
fricatives		s s	ɬ lh	ʃ ṣ			h h
implosives	ɓ b	ɗ d			ʄ j	ɠ g
nasals	m m	n n			ɲ ṇ	ŋ ng
approximants	w w	r r	l l		j y
plain clicks		ǀ c	ǁ x	! q	ǂ z
aspirated clicks		ǀʰ ch	ǁʰ xh	!ʰ qh	ǂʰ zh
glottalized clicks		ǀˀ c’	ǁˀ x’	!ˀ q’	ǂˀ z’
nasalized clicks		ǀᶰ cn	ǁᶰ xn	!ᶰ qn	ǂᶰ zn

Aspiration shows some allophonic variation. Aspirated clicks are often pronounced as if they have a [x] or [χ] offglide. The pulmonic aspirates are often pronounced as clusters with [ɦ] before a tone 3 vowel (see below).

Nucleus

Xṣâli has five vowels with no length distinction. However, vowels carry one of six “tones.” These are actually six states within a suprasegmental complex. For the purposes of this grammar, each tone will be given a number.

• Tone 1 – mid tone: pronounced with a level medium tone.
• Tone 2 – high tone: pronounced with a level high tone, and creaky voice.
• Tone 3 – low tone: pronounced with a level low tone, and breathy voice.

Tones 1 to 3 can be seen as distinguished by tone with secondary phonation, or as distinguished primarily by glottal restriction with a secondary tonal realization.

• Tone 4 – high nasal: pronounced with a level high tone, and nasalization.
• Tone 5 – low nasal: pronounced with a level low tone, and nasalization.
• Tone 6 – falling: pronounced with a high-low falling tone, also with no phonemic phonation.

Tones 4 to 6 have no phonemic phonation. However, glottal restriction tends to match nearby vowels of tones 1 to 3, if present.

Clicks can be glottalized or nasalized, and vowels can have glottal and nasal features as well. These features can combine in complicated ways. For many speakers, in rapid speech, clicks match the following vowel in terms of nasal or glottal features, effectively neutralizing two entire rows of phonemic distinction among click consonants. However, this is uncommon in careful speech.

Orthographic representation:

Tone 1	a e i o u
Tone 2	á é í ó ú
Tone 3	ạ ẹ ị ọ ụ
Tone 4	ã ẽ ĩ õ ũ
Tone 5	ą ę į ǫ ų
Tone 6	â ê î ô û

Syllable Structure

Phonetically, coda consonants may appear due to contraction. However, phonemically codas are not allowed. An onset is required, meaning the only allowable syllable structure would be CV or CCV if one of the thirty eight initial clusters is present. In compound words, medial clusters are not separated by a syllable boundary. A word with structure CVCCV would be phonemically CV.CCV, and phonetically realized as either CV.CCV, or possibly CV.C.CV, but never CVC.CV.

When a nasal precedes the phoneme /ɠ/, the cluster is written ngg, to distinguish it from /ŋ/, written ng.

All native roots are historically monosyllabic. Due to compounding and borrowing, many polysyllabic words are not easily parsed into monosyllabic elements, but generally there is a strong overlap between the phonology of syllable structure and the phonology of root structure.

Here are some examples of legal syllables, in standard orthography and modified IPA.

• chį – ǀʰiⁿ˨
• ndá – nɗaˀ˦
• qsû – !su˦˨
• mpha – mpʰa
• znọ – ǂⁿoʱ˨

Word Classes

In any language, parts of speech are not slots to be filled but roughly defined groupings based on usage, with exceptions and subcategories. By the standard analysis, there are eight parts of speech in Xṣâli: verbs, nouns, preverbal adverbs, free adverbs, quantifiers (including numbers), particles (conjunctive, aspectual, and interrogative), postpositions, and pronouns (including demonstratives). All of these are closed classes except verbs, nouns, and to some extent free adverbs.

Compounding

For most parts of speech, roots and words are functionally the same thing. However, this is not true for nouns and verbs, due to compounding. Lexical nouns (as opposed to nominal roots) and lexical verbs (as opposed to verbal roots) are words treated as noun and verbs, respectively, within a clause. Lexical nouns can consist of a bare nominal root or a compound. Nominal compounds may use a verbal or nominal root as the first element, but nearly always end in a nominal root. Lexical verbs can consist of a bare verbal root or a compound. Verbal compound verbs may use a nominal or verbal root as the first element, but nearly always end in a verbal root. Their meaning is not always predictable, so they are treated as separate lexical verbs. Other types of roots do not form compounds. Uncompounded nominal roots are almost never used as verbs, and vice versa; compounding is the main derivational strategy for crossing over from one part of speech to the other.

Here are just a few common patterns that emerge from productive compounding. The root xa is rarely used on its own, but probably once meant “exist” or “existence.” It sometimes forms inchoatives (though these are syntactically nouns, since they cannot take agents).

• dâxa – get bigger
• pâxa – be like this

Many derived verbs use final ḳho to form causatives, or pą to form passives.

• îḳho – operate, use
• capą – get eaten; this derivation is frequently used to create stative verbs from verbs that normally allow only active intransitive forms.

Reduplication is a nonproductive derivational process in which a root is repeated in a single word. Not every root has a reduplicated form, and the meaning of each form is not always predictable. Nominal roots often make augmentative or collective nouns. Verbal roots usually make participles, but it is not consistent whether they are agentive or patientive (other methods can be used to make predictable, productive agentive or patientive forms).

mǫ – mother

mǫmǫ – ancestors, family

da – father (more commonly dasó)

dada – grandfather

qạ – speak

qạqạ – testament

cî – cut (more commonly tṣícî)

cîcî – knife

It is not clear if reduplication is a form of compounding. Only roots can be reduplicated, not compound words, and these reduplicated forms rarely go on to form more complex compounds. However, they differ from normal compounds in that many qualifiers and free adverbs may be reduplicated. Also, unlike compounds, the part of speech may not be identical to the part of speech of the final element (e.g. some reduplicated qualifiers become free adjectives).

While there are no dedicated derivational affixes, there is historical evidence of derivational suffixes. In the modern language this manifests itself as a high tone on a root that would otherwise have a middle or low tone. The starting root is usually a verbal root, and the resulting word is almost always a noun. Some common examples include distributive pronouns, results or patients, and augmentative or resultative nouns.

• xá – everyone, derived from xa
• ngá – cloth, derived from nga, “weave”
• q’á – battle, derived from q’a, “fight”

Adverbs

Adverbs come in two varieties. Preverbal adverbs are a closed class that always appears immediately before the first word in a verb phrase. Free adverbs are a semi-closed class that can appear anywhere in a clause so long as they do not split a verb phrase or noun phrase.

Preverbal adverbs are a small closed class that appear immediately before the first verb. Only one preverbal adverb may appear before a given verb phrase, except for bî, which may appear before one other preverbal adverb.

• tṣạ - negative
• ce - not yet
• tsá - not anymore
• xã - already
• mpą - still
• zî - almost
• nẹ - unintentionally
• bî - reciprocally

ka bî ce tsanké-k la

MEDI RECP NEG displease-MEDI ASP

“They haven’t ever displeased each other yet.”

Free adverbs are a larger group, and occasionally new words may emerge in this class. They may give information about manner, time, or place, and in some cases information that would be more overtly grammatical in other languages.

Qọlhã lí xṣâqa ha.

1st know now ASP

“Now I get it.”

Demonstratives

Demonstratives have a four way distinction. Pâ is a proximate marker, used for things that are close to the speaker, and when introducing new topics. Ka is used for things that are close to the speaker and listener, or between them. It is by far the most common way to refer to previously established topics, and serves as a default third person pronoun. Sũ is a distal marker, used for things that are far from both the speaker and the listener, or previous topics. Bi is an indefinite marker. It is only used for non-specific referents, so a sentence like “Somebody called for you. You’ll never guess who!” would use one of the first three demonstratives instead of bi.

Third person pronouns do not exist as a separate class; instead the demonstratives listed above fill this role. First and second person pronouns exist, although they are all historically recent nouns, and behave like nouns except for that fact that they rarely appear with quantifiers. The first and second person pronouns, unlike demonstratives, have different forms depending on social context.

• ??? - first person, formal
• qọlhã - first person, informal
• ??? - second person, formal
• hilhó - second person, informal
• daxṣâ - second person, honorific

Because they can serve as predicates, the heads of noun phrases, and modifiers of nouns, demonstratives and pronouns are syntactically similar to nouns. However, they cannot be quantified, and have other subtle differences in usage.

Postpositions

Postpositions indicate a noun phrase’s role within a clause. There are only five postpositions in Xṣâli, although each one has a range of uses and meanings.

• tsǫ – comitative, instrumental, range of comparison, also phrase-level conjunction (“and” between nouns). The contracted form is never used as a functional postposition, but does appear in some number phrases.
• nga – location, direction, goal, motion toward, beneficiary, recipient
• mạ/-m – source, motion away, location, non-standard topic marker. The second form is a contraction, which is optionally used anytime the word does not appear at the end of a clause. There is some ambiguity between the use of mạ as an ablative and as a topic marker, but this can usually be cleared up with periphrasis.
• ké – possessive marker, genitive
• ṣi – without, also phrase-level conjunction (“but not”). The contracted form is never used except in some number phrases.

Some nouns frequently appear before postpositions to further specify their meaning. Technically these take over as the new head noun of the phrase, but they are thought of more as extensions of the postpositions themselves, and so the combination of noun and postposition is memorized as one unit. Since the postpositions have such a wide range of meanings, some of these noun+postposition combinations are very common, and some of the nouns in question have taken on slightly different meanings than their default meaning as stand alone nouns.

• qhapi nga – across, through
• qsų nga – outside
• qsų mạ – away from
• tsá nga – inside
• gâqha nga – in the vicinity of
• dú nga – to the side of, behind

Particles

Particles come in three varieties: conjunctive, aspectual, and interrogative.

The conjunctive particles connect complete clauses to form complex sentences. One of the most common is nį, the relative marker. This particle indicates that the previous clause is a modifier of the next head noun. The relative clause is intact, i.e. no arguments are omitted even if they appear in the following clause, though they may be replaced by demonstratives. The other conjunctive particles are ki (if), wạ (even though), and mú (when, because). This last is used in a wide variety of situations, almost as a generic connector of two clauses.

Sũkâ tṣạ xṣâxa qọlhã mú?

there NEG exist 1st CONJ

“I’m not there, so [what then]?”

There are also aspectual particles that affect how the information in a sentence should be placed in a larger temporal context. La is used for habitual actions, resultant states, or reliable predictions. Curiously, la is rarely used to mark gnomic truths. Ha is used for events with a dynamic relationship with the rest of the relevant timeline, often similar to English phrases “just then” or “as it happens.” Usually ha implies some sort of change, either in the past or future, or a sudden or truncated event.

Ca mmuḳạ qọlhã dibą la.

eat finish 1st idli PRCL

“I eat all my idli.”

Ca mmuḳạ qọlhã dibą ha.

eat finish 1st idli PRCL

“I’ve just eaten the last of my idli.”

The syntax of questions is dealt with in a later section, but polar questions are marked by two particles, either or both of which may be used. Nọ appears at the beginning of a polar question, and sí appears at the end.

Nọ pâ-m xṣâ sí?

PRTC PROX-POST correct PRTC

“Is this right?”

Quantifiers

Quantifiers and numbers are a closed class that indicate the number of the head noun in a noun phrase. They cannot serve as the head of a noun phrase on their own without a place holder. Numbers are dealt with in more detail in the numbers section below.

Ṣáká ṣínngu ca wạ ka z’ãtsą qsųqha.

cat many eat CONJ MEDI few hunt

“Many cats eat, but few of them hunt.”

Phrase Syntax

All phrases not headed by an adverb or particle can be classified as either a verb phrase or noun phrase. The internal syntax of verb phrases is complicated by the use of serial verbs and auxiliaries. Noun phrases are even more internally complex, with various types of modifiers and quantifiers.

Serial Verb Constructions

A verb phrase may consist of a single lexical verb (either a verbal root or a compound verb), but multiple verbs may appear together. There are two ways this can happen. The first is serial verb constructions. A serial verb construction strings together two, or rarely three, verbs into a single phrase in which each verb retains its original meaning and function (in other words, there is no grammatical or semantic change to each verb). Theoretically any combination of lexical verbs could form a serial verb construction, but in practice some verbs combine in this way more often than others.

q’anã qha pą hilhó

city go see 2nd

“Go and see the city.”

Note that verbs denoting emotions are often causative by nature, so something like “get scared and run” would not be handled with a serial verb construction, as in some other languages with SVCs, but with separate clauses. Chaining clauses together will be dealt with in more detail in later sections.

Ka c’ạnké sũ mú xhųwa-k ha

MEDI scare DIST CONJ run-MEDI PART

“It scared her, so she ran.”

Auxiliary Constructions

The other way that verbs can combine is through auxiliary constructions. As with SVCs, auxiliaries share one set of agents and patients, and serve as a single verb phrase. In this case the second verb modifies the first, or rarely the last two verbs modify the first. The verbs that can be used as modifiers are a closed class, and while they have a separate life as ordinary verbs, their use as auxiliaries can be unpredictable.

Mwi wáqṣẹ hẽxị hilhó

seed throw want 2nd

You want to throw seeds.

Structure of Noun Phrases

A noun phrase may consist of a single lexical noun or pronoun, or a head noun or pronoun along with a number of modifiers. Lexical nouns, including compounds, may appear immediately before a pronoun or another noun to modify it somewhat like an adjective. A head noun may have a number of modifiers, in which case relative clauses appear first, followed by pronouns (which in this context are possessive) or demonstratives, followed by nominal modifiers, followed by the head noun. Quantifiers and numbers usually appear immediately after the head noun. Postpositions, if present, always appear at the very end of a noun phrase.

The simple juxtaposition of nouns is a major aspect of Xṣâli grammar. Title and appositives share this same formula. However, the chaining of nouns is sometimes ambiguous. In theory, ṣáká ho (cat food, literally "cat rice") can be food made from a cat, or food for a cat. Although the main tool against this ambiguity is context, the problem is ameliorated somewhat by the fact that the order of modifiers is strictly determined.

[TBD: order of modifiers within a noun phrase]

Clause Syntax

Alignment

The constituent parts of a clause follow strict rules of word order. The agent immediately follows the verb phrase. The patient can appear anywhere except immediately after the verb phrase, i.e. before the verb phrase or after the agent, although the former is much more common than the latter. This means that OVS and VSO clauses are possible. With topical fronting SOVS is also possible. Intransitive verbs follow an active-stative alignment. If the subject has low agency, it appears before the verb, and if it has high agency it appears after. A single lexical verb can be transitive, active intransitive, or stative intransitive. Since Xṣâli does not allow core arguments to be dropped, it will always be clear whether a verb is transitive or intransitive, active or stative. This is important, since the particular meaning and usage of a verb of one state might be subtly different than the same verb in another state.

Tṣalhụ dibą ca qọlhã la

well idli eat 1st

I eat a lot of idli.

Tṣalhụ dibą ca

well idli eat

Idli is tasty (literally: Idli eats well).

The second verb must be patientive intransitive, because of the missing agent. Also note the slight difference in meaning.

One complicating factor is topicalization. The topic of a clause dictates what antecedent pronouns are likely to refer to. A topic may be introduced deliberately at the beginning of a clause, even if it occurs again later as a core argument. The topic may indicate what a clause is about, but it is not generally the most important information within a clause. The comment is usually some new information, and often replaces the existing topic as the new topic of subsequent clauses.

[example: topic introduced and then Patient-Verb-AgentMedial, next sentence Patient is Medial]

Pronouns are obligatory. If the same pronoun is used for two core arguments, the verb is reflexive. The pronoun MEDI most often stands in for the topic, or serves as a dummy pronoun. For some verbs, there is little distinction between a patientive verb with a single argument and a reflexive, but for others, this mark of transitivity can have an impact on meaning.

[example: patient-verb vs patient-verb-ka]

Some verbs of speech or handling can be ditransitive. This is an exception to the rule that all arguments must be present, as ditransitive verbs can have one or two stated objects. One will be a beneficiary or recipient, and the other will be a patient. Both objects are unmarked, and their order in the clause generally follows topicality. So context is necessary to determine the exact roles of each argument. Usually the more animate one will be the beneficiary/recipient. One important ambiguity about these verbs is that, since either object may be omitted, and neither object is marked, a verb might appear with only one object, and the listener must determine if it is a beneficiary/recipient or more traditional patient.

[example]

Special Semantic Topics

Numbers

Numbers in Xṣâli use a decimal system, with clear evidence of a vigesimal system underneath. The numbers one through five, and ten, have their own unique roots. The numbers six through nine are derived by a no longer productive process of addition.

1 – dã

2 – dú

3 – sê

4 – mị

5 – ḳi

6 – dãḳi

7 – dúḳi

8 – sêḳi

9 – mịḳi

10 – há’a

Each decade has its own unique term, but thirty, fifty, seventy, and ninety are historically derived.

10 – há’a

20 – ngạ’a

30 – hángú

40 – ngạ’ú

50 – hángê

60 – ngạ’ê

70 – hángị

80 – ngạ’ị

90 – háchó

In between the decimals, the first four numbers are derived by addition to the previous decimal, and the last five by subtraction from the next decimal. The postpositions can contract to -ts and -ṣ, something these postpositions normally do not do. One irregular case is the number twenty five, which usually uses addition rather than subtraction.

31 – dã tsǫ hángú (literally “thirty with one”)

32 – dú tsǫ hángú

33 – sê tsǫ hángú

34 – mị tsǫ hángú

35 – ḳi ṣi ngạ’ú (literally “forty without five”)

36 – mị ṣi ngạ’ú

37 – sê ṣi ngạ’ú

38 – dú ṣi ngạ’ú

39 – dã ṣi ngạ’ú

There are unique terms for one hundred, four hundred, one thousand, and ten thousand. These are multiplied by a following number. Speakers will usually avoid using the number six through nine when multiplying hundreds.

100 – chó’a

200 – chó’a dú

300 – chó’a sê

400 – ngạngạ

500 – chó’a ngạngạ ~OR~ chó’a ḳi

1000 – bọsé

10000 - [TBA]

Here are some examples of complex numbers. Some numbers have multiple strategies of formation. For example, larger powers of one hundred may be formed by adding powers of four hundred to powers of one hundred.

229 – dã ṣi hángú chó’a dú

497 – sê ṣi chó’a ḳi ~OR~ sê ṣi chó’a ngạngạ

900 – chó’a dã ṣi há’a ~OR~ chó’a ngạngạ dú

Cardinal numbers end a noun phrase. Ordinal numbers may begin it, or at the end immediately before the word khạznĩ, which becomes the new phrase head. When cardinal numbers are used without a head noun, the word ntsowį is inserted just before the number.

[example]