Kezhwa Grammar
Kezhwa is a language with both analytic and agglutinative features. The language makes use of clitics and grammatical particles, as well as affixes. Intervocalic voicing of voiceless obstruents is a prominent feature of the language's morphology. The language follows subject-verb-object (SVO) word order and is largely head-initial. It has prepositions, as opposed to postpositions. Additionally, relative clauses follow the nouns they modify, however, adjectives precede the nouns they modify.
Background
In the comic books that serve as a basis for the TV series Paper Girls, the time travelers are shown speaking in characters which are a simple one-to-one substitution cipher of English, which was intended for readers to be able to decode it. The studio, as well as the Petersons, wanted to maintain the spirit of the original cipher, i.e. a type of code that could be broken, without it being an actual cipher. Their vision was to create type of creole that borrowed words from languages all over the world at any time in history, making the vocabulary mostly a posteriori. Kezhwa's phonology and grammatical system, however, are largely a priori.
Borrowings and semantics
When the Petersons borrowed words into Kezhwa, the first question they considered was how non-native speakers would hear the word being borrowed. Secondly, they considered how that perceived phonological unit would shift to fit the phonology of Kezhwa. Thirdly, they considered what the resulting meaning would be in Kezhwa. In Kezhwa, there are almost no straightforward semantic borrowings, the meaning, as a general rule, being shifted in some way. This is illustrated in the examples below:
- Saga is from "It's all good," which can be written phontetically as one unit, [ɪtsalɡʊd]. However, said quickly, it can be realized as [saɡʊ]. In Kezhwa, unstressed vowels turn into schwas, and so [sagʊ] becomes saga [sagə]. In Kezhwa, this is used as a response roughly equivalent to "It's fine" or "It's all right".
- Datep comes from "duct tape" [dəktep], which can be reduced to [dətep]. Any schwa in a stressed syllable becomes an [a] in Kezhwa, and in a closed syllable, /e/ becomes [ɛ], resulting in datep [datɛp], which means "to repair, to fix, to patch up, to heal, to ameliorate," which comes from the versatility of duct tape as a tool for repair.
- Duma is taken from the Swahili duma, meaning "cheetah." The unstressed [a] became a schwa, and so [duma] becomes [dumə]. The Kezhwa word came out to mean "to hurry, to hustle." The basis for this shift in meaning was that cheetahs appear to hurry when moving at speed.
- Tika is borrowed from Finnish tikka, meaning "dart." The geminate [kk] is reduced to a single [k], and because the a is in an unstressed syllable, it becomes [ə], giving us tika [tikə], meaning "to try," because trying something can be compared to throwing a dart at a dartboard.
- Borrowed compounds, with their parts not necessarily coming from the same language, are also found in Kezhwa. In the word genzhín, the first half is from Mandarin gēn, meaning "to follow," and the second is from Czech stín, meaning "shadow." Kezhwa has SVO word order, meaning that the verb in a compound will always come before the object. In stín, the t is dropped, because consonant clusters with stops in the second position are not allowed, and the s [ʃ] is voiced to to zh [ʒ], because of the preceding n. The Kezhwa word means "to track," coming from the fact that tracking can be likened to following a shadow.
Nouns
Nouns are marked for definiteness and number.
Definiteness
Nouns may be either definite or indefinite. The distinction of definiteness is only present for singular nouns; plural nouns make no such distinction. For singular nouns, the definite is the unmarked form, and the indefinite is marked with a proclitic (with the proto-form *i-), the form of which depends on the sound the noun begins with. If the noun begins with a consonant, it is i-, if it begins with the vowel i, zh-, and if it begins with vowel other than i, y-. The i- proclitic causes voicing of voiceless obstruents in consonant-initial nouns.
Number
Nouns can take singular or plural number. The singular is the unmarked form, and plurals are marked with the particle an, which precedes the noun. The particle an causes voicing of voiceless obstruents in consonant-initial nouns.
Noun Declension
The different forms are shown for the examples tamagó "child", ala "walkie talkie", and iga "target" in the table that follows:
| Consonant-Initial | Vowel-Initial | I-Initial | ||||
|---|---|---|---|---|---|---|
| Singular | Plural | Singular | Plural | Singular | Plural | |
| Definite | tamagó | an damagó | ala | an ala | iga | an iga |
| Indefinite | i-damagó | y-ala | zh-iga | |||
Noun Derivation
Verbal nouns can be derived from verb roots using the suffix -(e)ga. The form with an initial e is used after consonant-final verb roots, and the other for vowel-final ones. For example akwat "to know" gives us akwádega "knowledge", and kewi "to help" gives us kéwiga "help".
Determiners
There are proximal ("this") and distal ("that") demonstratives, which appear as suffixes on nouns. The suffix used differs based on if the noun root ends in a consonant (-agí and -ayá) or vowel (-gí and-yá). This can be seen in the following table, using yambwas "foot" and ala "walkie-talkie" as examples:
| Consonant-Final | Vowel-Final | |
|---|---|---|
| Proximal ("this") | yambwazagí | alagí |
| Distal ("that") | yambwazayá | alayá |
Verbs
Verbs can be marked for polarity, tense and mood. Verbs also have a "plain" form, which in the positive is the unmarked verb root. This is generally translated as an infinitive.
Polarity
There is positive and negative polarity, where the positive is the unmarked form, while the negative is marked with the prefix '(a)- (from a proto-form *ha). Polarity is distinguished for all tenses and moods.
Tense
The tenses are present and non-present, with the present being marked with the prefix i(m)- (from a proto-form *ima "now"), and the non-present being marked with the prefix l(a)- (from a proto-form *la "then". The non-present can be translated as both past and future. These tenses are the result of the Kezhwa speakers being time travelers, for which the distinction between the time they are at right now, and all other times, would be more salient than other distinctions between time periods.
Mood
The moods are indicative and imperative. The indicative forms are those marked for tense above. Imperative forms agree with their subjects in number and are marked by means of enclitics (-(n)a for singular subjects and -(z)u for plural ones).
Verb Conjugation
Verbs are conjugated differently depending on if the verb root begins with a consonant or a vowel, and also depending on if it ends in a consonant or vowel for the imperative forms. Consider, as an example, the conjugation table for the vowel-initial verb akwat "to know":
| Plain | Present | Non-Present | Imperative | ||
|---|---|---|---|---|---|
| Singular | Plural | ||||
| Positive | akwat | imakwat | lakwat | akwad-a | akwat-u |
| Negative | 'akwat | i'akwat | la'akwat | 'akwad-a | 'akwat-u |
For verb roots beginning in i, the present tense prefix (i-) dissimilates in the negative form, and becomes e-. For example, the present negative form of inená "to breathe" is e'inená "I am not breathing," not *i'inená.
Now, compare the table above with the conjugation table for the consonant-initial verb kewi "to help" below:
| Plain | Present | Non-Present | Imperative | ||
|---|---|---|---|---|---|
| Singular | Plural | ||||
| Positive | kewi | igewi | lagewi | kewi-na | kewi-zu |
| Negative | 'agewi | i'agewi | la'agewi | 'agewi-na | 'agewi-zu |
Names of years
In Kezhwa, years have specific names derived systematically from how they are written in Roman numerals. First, the year is written in Roman numerals, and then the Roman numerals are converted to Kezhwa sounds according to the following table:
| Arabic | Roman | Kezhwa |
|---|---|---|
| 1 | I | i |
| 5 | V | v |
| 10 | X | s |
| 50 | L | l |
| 100 | C | k |
| 500 | D | d |
| 1000 | M | m |
After this, some rules are applied to the resulting string:
- e is inserted between consonants in clusters, to make the word abide by the CAVC (where A is "approximant") syllable structure of Kezhwa and other phonotactic rules, such as the prohibition against geminates.
- An i between two other i's becomes zh [ʒ], that is, iii becomes izhi [iʒi].
- A glottal stop ' [ʔ] is inserted in the middle of ii sequences, that is, ii becomes i'i [iʔi].
- s before a v becomes z, that is, sv becomes zv.
For example, 2019 in Roman numerals is MMXIX, which gives us the base string mmsis, and the final result memesis. A more complex year is 1988, or MCMLXXXVIII, which comes out as mekmelsesezvizhi.