- 1 Introduction
- 2 History
- 3 Phonology
- 4 Verbal Morphology
- 5 Noun Morphology
- 6 Syntax
- 7 Common Expressions
Ajrato [ɑʑˈɾato], a creation of Dunomapuka, aka boy #12, is a language of the Akumin family spoken in Ajrat, a quiet and rather drizzly nation on the west coast of Maiakumi, on the planet Randa. In times past Ajrat colonized areas in far-off Rumial and Arvang, and was once widely spoken there, though has given way to either to indigenous languages or various intriguing Ajrato-based creoles. The language has had continued prestige throughout Randa as language of international communications, though somewhat outdone in this regard by Twe and Skri-Manu. It is, however, the dominant language for arty movies, certain academic circles, and diplomacy, and to some may carry an air of stuffy elitism. It has often been voted "Randa's prettiest language" in opinion polls.
Extra-diagetically, the language strives towards some ideal of aesthetic perfection in the author's mind. Phonologically it is variously influenced by Sanskrit, the Romance Languages, and Japanese (though only the yamatokotoba vocabulary!), while the grammar tends toward fairly familiar European models, though with a few quirks thrown in -- note the 1p inclusive/exclusive distinction, the null copula, the "collective" number. I will move on to greater levels of weirdness in other projects (two being conceived are Twe and Mojido); meanwhile, enjoy Ajrato for being pretty, graceful, and fairly unintimidating.
The Intebaram (i.e. Akumin) Empire, based in sunnier lands to the south, slowly expanded up the coast of the Western Ocean in Classical times, spreading its language and culture, though this territory did not extend into Ajrat itself yet. The successor state to Intebaram in this area, occupying a narrow coastal strip, was the Maritime Empire of Akumi, a commercially-minded, pious state that looked toward the sea, and defined its identity through struggle with the Koyaric peoples and Tosa barbarians to the north. The area carved out existed for a time as several dukedoms at the northern frontier of Akumin civilization, but the frontier kept moving northward, and more people settled down. Co-operative defense against the Tosa led to national unity, with the enthronement of a certain local dialect (from Ruya, the capital) as the national language.
Ajrato is influenced by the Tosa and Koyaric languages of the conquered territories. The Koyarien, in particular, probably contributed to the development of a more open syllable structure and an aversion to voiceless fricatives.
There are sixteen consonant phonemes:
|Plosive/Affricate||p /p/ b /b/||t /t/ d /d/||dj /dʑ/||k /k/|
|Fricative||v /v/||dh /ð/||s/ss /s/ s /z/||j /ʑ/||h /h/|
|Nasal||m /m/||n /n/|
|Liquid||r /ɾ/||y /j/|
The fricative j and affricate dj are alveolo-palatal /ʑ/ and /dʑ/ in standard speech, though many speakers pronounce these [ʒ] and [dʒ].
Orthographic s represents /s/ initially and /z/ medially; medial /s/ is indicated with a single letter I romanize as ss (historically this represented the cluster /st/).
/p/ is somewhat infrequent in the language, having shifted to /h/ or /f/ and thence to /v/ in the Late Akumin period, and appearing only in later borrowings. /g/ is missing from the stop system, though may occur in recent unassimilated borrowings; many speakers will substitute /k/.
There are five vowel phonemes: a e i o u /a e i o u/.
In closed syllables, /e i o u/ lax and lower slightly to [ɛ ɪ ɔ ʊ]; some dialects omit this rule for the high vowels. There is also a tendency to turn /a/ into [ɑ], with different environments depending on the dialect; see Dialects for specifics. The standard language applies this rule in only in closed syllables, but not before a nasal or before /j/, thus
- nakra [nɑkɾa] river, but
- hanta [hanta] potato, and
- anai [anaj] holy.
- ai /aj/
- ei /ej/ = [ɛj]
- oi /oj/ = [ɔj]
There is disagreement over whether these are phonemes in and of themselves or syllables that close with /j/. I prefer the second analysis because it accounts for a number of issues more neatly; see Syllabification below.
Rising diphthongs. Rising diphthongs are permitted in stressed closed syllables only; otherwise these sequences split into two syllables. Historically, when unstressed, these diphthongs lost the inital glide.
- ie /ie/ > [jɛ] in closed syllables; otherwise [i.e]
- ue /ue/ > [wɛ] in closed syllables; otherwise [u.e]
The pronunciation of these diphthongs, thus, is affected by the addition of suffixes:
- hiet [hjɛt] egg > hieten [hi.e.tɛn] eggs
- yued [jwɛd] leaf > yueden [ju.e.dɛn] years
Other than these recognized sequences, Ajrato shows a general aversion to having two vowels in hiatus, except for the /i.a/ that occurs in certain verb forms, such as suriakte [su.ɾiˈɑx.te] "we drink," and certain noun forms, such as kajiari [ka.ʑiˈa.ɾi] "of a thought." Other sequences may occur in recent borrowings, such as /u.a/ in satsetua "banana," though some speakers may substitute /uja/.
Neither initial nor final consonants clusters are permitted; medial clusters are limited and all have a coronal or palatal as the second element. Those permitted are:
- sr /zɾ/
- jr /ʑɾ/
- kr /kɾ/
- kt /kt/ = [xt]. This is spelled kd in a single verbal inflection -kdi.
- ny /nj/
- ry /ɾj/
- nt /nt/
- nd /nd/
- nj /nʑ/ = [ndʑ]. This is spelled ndj in certain passive forms such as hanandja "I am seen."
- jn /ʑn/ This last is a development out of Koyaric borrowings and occurs only in a few toponyms: Kijni (the capital city), Sajna, Yenejno.
Other clusters may occur in recent borrowings, such as /ts/ in satsetua "banana," though some speakers will substitute, say, [s] (thus a possible, though stigmatized, pronunciation of this word is /saˈse.tu.ja/, as though sassetuya).
A limited number of consonants are allowed word-finally: dental stops /t/ and /d/, nasals /n/ and /m/ (the second only after /a/), and the flap /ɾ/.
Accepting a non-phonemic analysis of the diphthongs, the permitted syllable structure can be reduced to (C)(i, u)V(z, ʑ, n, ɾ, k, j) in medial syllables; in final syllables the permitted coda consonants are instead (t, d, n, m, ɾ, j).
Consonant clusters are treated as being split between the two syllables. Coda consonants trigger laxing of the nuclear vowel. If the syllable is closed than the rising diphthongs collapse into a single syllable as shown above.
Since clusters of three consonants never occur, /j/ never appears adjacent to a consonant cluster (i.e., the falling diphthongs don't occur before a cluster and the rising diphthongs don't occur after).
Ajrato verbs are conjugated along the following paradigms:
- Four tenses, including an irrealis,
- Affirmative, negative or imperative mood, and
- Active versus passive voice.
Each combination of these (besides the imperative) has seven finite forms and an infinitive (also a participle; this will be called the Participle I in some usages). The finite forms are conjugated for Ajrato's seven personal pronouns:
- da I
- ne you (sing.)
- ku he/she/it
- an exclusive we (not including "you," the listener)
- nan inclusive we (including "you;" operates separately from an)
- nie you (pl.)
- men they
The citation form of a verb is the infinitive of the present active affirmative; this ends in -ki. This form is given in dictionaries, though linguistic resources may just list the verb stem (suri- "to drink," etc.). The verbs are not grouped into any conjugation classes except by stem vowel; that is, the vowel ending the root, coming right before any suffixes. All five vowels are represented, and thus verbs may end in any of -aki -oki -uki -eki -iki.
These can be further subdivided into those with a front stem vowel and those with a back stem vowel, a distinction relevant in how the plural conjugations are formed, but to begin, we will conjugate one verb from each of the five groups in its present active affirmative conjugation.
The Present Tense
sudaki to wash
ayoki to work
danuki to look
moreki to say; tell
yutiki to sew
This is the basis for the system, modified slightly for the other conjugations. After the stem comes a consonant or cluster marking the person (1st: kt; 2nd: j/s; 3rd: r. The ss of nan is a later development based on combining the forms for an and nie), and then -a for the singular and -e for the plural.
Formation of the plural requires a change to the stem vowel. Verbs with front stem vowels, we see, all take the plural stem -ia- and back stem vowels take the plural stem -a-. Now that the reader is familiar with this principle we will illustrate the rest of the conjugations with one verb from each group (sudaki, yutiki).
The Recent Past
The recent past is formed by suffixing -ja- after the stem. It is the only verbal tense that puts a suffix before rather than after the personal ending, the relic of an earlier system wherein affixes for aspectual distinctions were always placed at this position -- this form once denoted the perfective aspect. In the modern language the aspectual system has collapsed and this meaning of this conjugation has switched to one of tense; it can no longer be combined with the other tenses.
|sudajaki to have (just) washed
|yutijaki to have (just) sewn|
The stem vowel never changes in the recent past conjugation. Note that -jaja- in the 2s form has dissimilated to -jadja-.
The Simple Past
The simple past is formed by suffixing -n after the present form. The infinitive ends in -ken.
|sudaken to have washed
|yutiken to have sewn|
The irrealis, translating a range of conditional and subjunctive meanings, but treated as a tense (in older usage it's a future form, in fact), is formed by suffixing -t after the present form. The infinitive ends in -ket.
|sudaket to have washed (irr.)
|yutiket to have sewn (irr.)|
The Passive Voice
All of the above conjugations are active, but Ajrato possesses a simple inflection to make verbs passive. The passive was originally formed by adding -n- before the personal endings but sound change has caused some idiosyncrasies as will be illustrated below. We will provide an i-group verb as well as an e-group verb because the i-group has the added irregularity that the stem vowel breaks to -ie- in the singular and infinitive.
|sudanti to be washed
|morenti to be said
|yutienti to be sewn|
The mutations to the stem vowel in the plural are preserved. Another quirk is that the forms for nan merge with nie, but the subject pronoun can always be included to clarify.
The passive can be combined with any of the tenses. We will not write all possible forms, but consider:
- sudanti to be washed (present)
- sudajanti to have been washed (recent past)
- sudanten to have been washed (simple past)
- sudantet to be washed (irrealis)
The negative of any verb (except imperatives, see below) is formed with a prefix: u- before consonants and m'- before vowels. The second, following Ajrato convention, is separated from the verb with an apostrophe indicating the morpheme boundary. To demonstrate:
- usudaki to not wash
- uyutikta I'm not sewing
- m'ayoja you're not working
If the verb has initial /s/ this voices to /z/ when the prefix is added (indicated by not changing the spelling).
The result of all this is a verbal system with a rather flexible ability to combine forms, neither extremely agglutinating nor extremely synthetic. Examples of verb forms incorporating multiple inflections:
- umorejasse we didn't (just) say
- sudandjat I (may; would) be washed
- m'ayanden they weren't dealt with (passive of ayoki)
The Participle II
This is one of two forms that exist outside the combining framework described above; the Participle II form is always the verb stem suffixed with -kdi (representing /kti/), and there are no varying forms for tense or person. The basic meaning is "who does," forming a relative construction; the Participle I forms can be used for this also, but the Participle II is used to indicate a general truth about what the modified noun does. This will be discussed in more detail under Syntax.
- yutikdi dhien the man who sews
There are no forms for different tenses, but there is a negative:
- uyutikdi dhien the man who does not sew
The other form which exists outside the framework of conjugations is the imperative, which likewise does not conjugate for tense or person. For verbs with non-monosyllabic roots it is formed with the bare stem:
- suda wash up!
- ayo work!
- yuti sew!
Many verbs, however, have monosyllabic roots--we will illustrate with haki "to die" and toki "to put down"--and don't use a bare stem, which would be awkwardly short. Instead the irrealis is used, conjugated appropriately for the 2nd person singular or plural.
- hajat die! (said to one)
- haset die! (said to audience)
- moya tojat put down the cat! (said to one)
- moya taset put down the cat! (said to audience)
The negative of the imperative is formed with mei, the same word used to negate nouns.
- mei ayo don't work!
There are no third-person imperatives. An exhortative ("let's...") can be formed by adding the pronoun an before the imperative (instead of the expected nan, including the listener; this usage predates the formation of nan). Monosyllabic verbs conjugate this in the irrealis appropriately, but still require the pronoun.
- an danu let's look!
- an haktet let's die!
There are a number of particles that can be placed before a verb to clarify certain distinctions of tense and aspect (and mood, in the case of the irrealis). We might consider some of these particles to be conjunctions; others are more like adverbs, but Ajrato grammarians treat these as a single class. The usage of all of these is somewhat nuanced and will be further discussed under the relevant section in Syntax.
The particles can be divided up by what tense they are used with.
Used with the present. na future yune planned future (contraction of yuki ne, "to go and...") vi "have been doing;" "still do"
Used with the simple past. den remote past; marks especially great time depth
Used with the irrealis. ko "may;" possibility (ranging to probability) sun possibility (but doubtful) do "if" i "would" or "then" -- marks conditional or second half of an if/then clause
Irregularities in the verbal system are not too frequent, enough to be covered adequately in this brief section.
All verbs ending -raki and -riki change the 3s active (in the present, simple past, and irrealis) to -rera.
- nuraki to eat > nurera he eats (not *nurara)
- suriki to drink > surera she drinks (not *surira)
There are three verbs with u-stems with monosyllabic roots: ruki to do, accomplish; kuki to sleep; and yuki to go. In the plural forms these do not have the stem -u- mutate to -a-; rather it becomes -uya- (in both cases this is from original -ua-, with different results in initial and medial syllables).
- ruyakte we do (not *rakte)
- kuyasse we sleep (not *kasse)
- yuyare they go (not *yare)
A few auxiliary verbs have infinitves in -ri rather than -ki.
- sori to do to oneself
- yori "either"
Some of these have migrated into their own set called the "modal adverbs," no longer treated as verbs.
- teri together with
- juri to be able to
- kori to be obliged to
- duri "ought"
- dhari to need to
A small number of verbs use suppletive forms in some paradigms.
- The passive of mieki "to give" is santi (because mienti means "to be killed").
Ajrato nouns are declined for three numbers and three cases.
The plural of a noun always ends in -en.
Nouns ending in any of -a -o -e change the last vowel to -en.
- koya mound > koyen mounds
- toko bee > token bees
- sete pear > seten pears
Final -i becomes -ien, or -yen after another vowel.
- kaji thought > kajien thoughts
- mokai squid > mokayen squids
- soi fire > soyen fires
Final -u adds -yen.
- mitu llama > mituyen llamas
Final consonants add -en.
- dar bread > daren loaves of bread
- menet party > meneten parties
- nen point > nenen points
Monosyllables ending in -a -o -e add -yen.
- sa root > sayen roots
- dho water > dhoyen waters
- de hair > deyen multiple groups of hair
The collective is a separate number, used in theory to mark unmeasurable quantities. The usage is rather idiosyncratic in practice, and it's on the borderline between an actual inflection and a derivation. It always ends in -am.
Final -a -o -e are changed to -am.
- issa wolf > issam "the wolf;" "wolves" (generalized)
- kudo stone > kudam stone (material)
- ite bone > itam bone (as a substance)
Final -i becomes -iam, or -yam after another vowel.
- kaji a thought > kajiam thought
- bedhai a muscle > bedhayam muscle
Final -u adds -yam.
- mitu llama > mituyam "the llama" (generalized)
Final consonants add -am.
- menet party > menetam the party scene
Monosyllables ending in -a -o -e add -yam.
- sa root > sayam root structure
- ho a light; a lamp > hoyam light (non-quantifiable, or as a concept)
- de hair > deyam hair, generally (rarely used)
The Oblique Case
The oblique of a noun has multiple uses, but its most basic function is as a genitive. It always ends in -ri.
Final -a -o -e simply add -ri.
- koya mound > koyari of a mound
- toko bee > tokori of a bee
- sete pear > seteri of a pear
Final -i adds -ari, the sequence becoming -yari after another vowel.
- kaji thought > kajiari of a thought
- mokai squid > mokayari of a squid
Final -u adds -yari.
- mitu llama > mituyari of a llama
Final consonants add -ari.
- dar bread > darari of bread
- menet party > menetari of a party
Monosyllables ending in -a -o -e add -yari.
- dho water > dhoyari of water
For nouns ending in -e (except monosyllables) this is identical to the singular oblique.
- seteri of a pear; of pears
The collective oblique is formed by changing the final -am of the collective to -ori. This is, in fact, a more important and consistent usage than the normal collective.
- koyam mound-dom > koyori of mound material
- mokayam squid-hood > mokayori of squid
- sayam root structure > sayori of the root structure; of root material
These are used in expressions specifying a measured quantity, such as mokayori sata doren "two kilos of squid" or dhayori kei boken "three cups of water."
For nouns ending in -o (except monosyllables) this is identical to the singular oblique.
- tokori of a bee; of bee-ness
The Equative Case
The equative case has the meaning of "like X" or "as X."
Nouns ending in vowels suffix -nun.
- koya mound > koyanun like a mound
- mitu llama > mitunun like a llama
Monosyllables ending in -a -e -o add -yanun.
- dho water > dhoyanun like water
Those ending in -n suffix -dun.
- an sun > andun like the sun
- nen point > nendun like a point
Those ending in other consonants suffix -enun.
- dar bread > darenun like bread
The plural equative is formed by changing the final -n of the regular plural to -nun.
- koyen mounds > koyenun like mounds
- mokayen squids > mokayenun of squids
The collective equative is formed by changing the final -am to -onun. There isn't much use for this form except for nouns that usually appear in the collective.
- kudam stone > kudonun like stone
The most common sort of irregularity is the irregular stem. In these cases the root differs in some way from the connecting stem form. Either the root contains a different vowel from the stem, or the stem possesses an "orphaned" extra consonant that drops off at the end of the root.
Monosyllablic nouns are very frequently irregular. In a sense, they are all irregular, in that they don't lose their final vowels to add suffixes; rather, they must have an intervening consonant -- as it happens, the default for this is -y-, but other consonants are seen. For example:
- i bud; stem idh-; thus idhen buds, idhari of a bud, etc.
- te honey; stem tek-; thus teken, tekari...
- na berry; stem nak-; thus naken, nakari...
- na hour; stem nav-; thus naven, navari...
Non-monosyllables may have "orphaned" consonants in the stem also.
- yuna mouth; stem yunav-; thus yunaven, yunavari...
- ani star; stem anik-; thus aniken, anikari...
There is a fairly frequent, but unpredictable alternation between final -n in the base form and -nt- in the stem.
- taben swamp; stem tabent-; thus tabenten, tabentari...
- nasan rust; stem nasant-; thus nasanten, nasantari...
There is also occasional alternation between a base ending in -aC and a stem with -oC-.
- dosat month; stem dosot-; thus dosoten, dosotari...
- yorad ball; stem yorod-; thus yoroden, yorodari...
There is also a tendency for some nouns ending in -i (and especially -ji and -dji) to act as though they ended in -e.
- tadji valley; plural tadjen, oblique tadjeri, etc. (not *tadjien, *tadjiari)
- yunji wave; plural yunjen, oblique yunjeri, etc. (not *yunjien, *yunjiari)
Of final note, the word dho "water" has an irregular collective stem dhay-.
- dhayam the waters; dhayori of water
Full Declension of Selected Nouns
karu letter; symbol
ani star (stem anik-)
|Nominative||Object clitic||Object clitic - after /n/||Genitive||Dative|
|3 sing. animate||ku||-ku||-tu||kuya||kuri|
|3 sing. inanimate||--||-yo||-yo||--||--|
|1 pl. exclusive||an||-yan||-an||anya||ari|
|1 pl. inclusive||nan||-nan||-dan||nanya||nari|
The nominative, genitive and dative columns represent stand-alone pronouns. The object forms are clitics attached to the relevant verb in the phrase. The two alternatives represent the plain form, and a mutated form that appears after the -n of the past tense. This causes the velar /k/ to front to coronal /t/, while all /n/'s dissimilate to /d/ after another /n/ (this phenomenon also appears in the formation of the equative).
There is a third form for the objects that appears after the -t of the irrealis, formed by inserting -e- between this and the normal clitic, except that the exclusive we appears simply as -an (not *-eyan).
- miktane I'm killing you
- miktande I killed you
- ko miktatene I may kill you
The inanimate forms are not a full paradigm, only distinguished from animates when they're direct objects. Otherwise, inanimate things use ku etc.
"Animate" is limited to humans, or animals and abstract concepts if you want to anthropomorphize them, with no funny business. Babies and housepets are optional, but using the inanimate for them sounds a bit more refined -- or snobbish.
Of note is the dialect of the capital city Kijni, which generalizes the inanimate forms into a full set, always distinguished from animates; thus: yo it; yoya its; yori to it. This usage is rather stigmatized, though.
Let the reader not be confused by the dative forms, which look like the genitive (i.e. oblique) of nouns.
Infinitives with Object Clitics.
The object clitics alter the form of certain active infinitives. The simple past alters the clitic itself, and the irrealis inserts an epenthetic -e- as outlined above, but the present and recent past, which normally end in -ki, drop the -k- when the clitic is attached. Infinitive endings change as follows:
- -aki --> -ai-
- -eki --> -ei-
- -iki --> -ei-
- -oki --> -oi-
- -uki --> -oi-
-an changes the -i- to -y-, and -yo absorbs the -i-.
- nuraida to eat me
- yuteime to sew them
- sureyo to drink it
- danoini to look at you (pl.)
|Adjective||--||mo what?||kei this||yai that||nami said||ru some||mei no||sen other|
|Person||ku who||mu who?||keta this person||yata that person||nanta said person||ru ku someone||mei ku no one||sen ku someone else|
|Thing||yo which||mo what?||keta this one||yata that one||nanta said thing||ru yo something||mei yo nothing||sen yo something else|
|Place||ven where||mo ven where?||ura here||yara there||nami ven said place||ru ven somewhere||mei ven nowhere||sen ven somewhere else|
|Time||dere when||mo dere when?||si now||vana then||nami dere said time||ru dere sometime||mei dere never||sen dere some other time|
|Way||tien how||mo tien how?||--||--||--||ru tien somehow||mei tien not in any way||sen tien some other way|
|Reason||ja because||ja mo why?||--||--||--||--||--||--|
For syntax and usage notes, please see Ajrato Syntax.
- Mai. Yes./That's so.
- Med. No.
- Ya. Yeah.
- Neri miekta imen. Thank you (to one person).
- Nieri miekta imen. Thank you (to group).
- Imen. Thanks.
- Nemin moyen miakte. We have eleven cats.
- Keta djimuyari. This is bullshit.
- Yai duha sunte suriki hanaranan. That old crone is eyeing us, drinking her brandy.
- An de-nura. Let's do lunch.