I'm making up my own language!
This article doesn't deal with the language's alphabet and punctuation marks. Here the letters are identified by their romanizations only.
Manner | Labial | Dental | (Denti-)Alveolar | Post-Alveolar | Palatal | Velar | Uvular | Pharyngeal | Glottal | ||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Plain | Emphatic | ||||||||||||||||||||
Nasal | m | [m] | n | [n̪] | |||||||||||||||||
Stop | Voiceless | p | [p] | t | [t̪] | t | [t̪ˤ] | k | [k] | q | [q] | ' | [ʔ] | ||||||||
Voiced | b | [b] | d | [d̪] | d | [d̪ˤ] | |||||||||||||||
Fricative | Voiceless | th | [θ] | s | [s] | ts | [t͡sˤ] | sh | [ʃ] | x | [x] | ch | [ħ] | h | [h] | ||||||
Voiced | v | [v] | dh | [ð] | z | [z] | dh | [ðˤ] | j | [ʒ] | gh | [ʁ] | ` | [ʕ] | |||||||
Approximant | l | [l] | l | [ɫ] | y | [j] | |||||||||||||||
Trill | r | [r~ɾ] | |||||||||||||||||||
Lateral Fricative | ś | [ɬ] |
In a few cases the headings on the chart are approximate, perhaps reflecting sound changes that occurred during a transition period prior to those addressed by this document. The glottal stop is omitted in romanizations when it occurs at the beginning of a word. Every consonant can be geminated, which occurs when the same consonant functions as both the coda to one syllable and as the onset to the next. This is generally shown in romanization by doubling the sound's romanized letter although digraphs are not doubled. In romanizations a full stop appears in the middle of a word immediately before a digraph when the previous syllable is open and the romanizations of the individual letters of the diagraph correspond to romanizations of other sounds. When written in informal environments that lack a wide variety of characters and sufficient formatting options then the emphatic consonants can alternatively be written using capital letters.
The ideal pronunciation of /r/ is as a trill but I'm bad at pronouncing it so sometimes it's a tap. As for the two allophones of /l/, Dark L [ɫ] occurs:
Clear L [l] occurs after front vowels and at the beginning of words when followed by a front vowel.
Quality | Bright | Dark | Proto-Language | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|
Short | Long | Short | Long | Short | Long | Quality | |||||
Closed | i | [i] | ie | [iː] | u | [u] | uu | [uː] | [i] | [iː] | Closed Back |
Mid | e | [e] | ee | [eː] | o | [ɔ] | oh | [oː] | [u] | [uː] | Closed Front |
oh | [oː] | [aw] | [aːw] | Open | |||||||
Open | a | [a] | aa | [aː] | a | [a] | aa | [aː] | [a] | [aː] |
The vowels are divided into "bright" vowels and "dark" vowels. These correspond to the language's two noun classes. Vowel harmony is triggered by the first vowel that comes after the first root consonant and it is this vowel that additionally serves to indicate the noun's class. Harmony extends progressively until it reaches the end of the word. [a] and [aː] are neutral transparent vowels. /j/ is romanized as i when it appears in a coda, forming ai, aai, ei, oi and ui. That last one is pronounced as [ui].
The permissible syllable structures are almost always Cv and CvC with few exceptions. A choice of vowels when combined with possibilities for consonant gemination forms a template. Templates may only include /iː/ and /oː/ in syllables that are either open or stressed and this implies a knock on restriction for their pairwise counterparts /uː/ and /eː/. However, when a template containing /u/ is inserted into a root with where the result has u from the template followed by v from root (i.e. the consonant was formerly pronounced as [w]) and if these occur together inside the same closed syllable then the final result has /u/ lengthening to /uː/ with a corresponding re-romanization from uv into uu. Because of vowel harmony, a prototypical uw will sometimes be expressed as ie [iː] instead of uu. Similarly, a root with a v when combined with a pattern containing a preceding a or aa yields oh [oː] if they fall within the same syllable. Inserting a triconsonantal root into a template produces one of following forms.
Prototypical Syllable Pattern | Phonotactically Valid Templates | Notes | Regular Examples | Irregular Examples |
---|---|---|---|---|
CvCvC | 6×4 = 24 | bajaad, bavaaj, buvaaj | bajoh, bajuu | |
CvCCvC | 4×4 = 16 | Second consonant doubled | bajjaad | bohvaaj, buuvaaj, bajjoh, bajjuu |
CvCCv | 4×6 = 24 | bajvaa | bohjaa, buujaa | |
CvCvCv | 6×6×6 = 216 | bavajaa, buvajaa, bajavaa, bajuvaa | ||
CvCCvCv | 4×6×6 = 144 | Second consonant doubled | bajjavaa bajjuvaa | bohvajaa, buuvajaa |
CvCvCCv | 6×4×6 = 144 | Third consonant doubled | bavajjaa, buvajjaa | bajohvaa, bajuuvaa |
Total | 568 |
Word stress indicated by vowel length. The stressed syllable undergoes vowel lengthening for /e/, /i/, /o/ and /u/ (becoming /eː/, /iː/, /oː/ and /uː/ respectively) but /a/ is not lengthened. When the coda is the glide sound /j/ then lengthening only occurs when the vowel was /i/, which lengthens to become /iː/. Combinations such as /eːj/ and /oːj/ do not occur.
The spoken word is always chanted with a rhythm and intonation similar to Qur'an recitation. However, in general the rules of tajwid do not apply unless otherwise stated. Of particular note is that the consonants are always pronounced clearly (idhar) and never with nasalization (ghunnah). Qalqalah (echo) does occur when a word ends in an emphatic consonant though.
The language was originally written using an impure abjad with one letter dedicated exclusively to the /aː/ vowel and the letters representing /w/ and /y/ additionally serving as indicators for /uː/ and /iː/ respectively, and all other letters representing consonants only. The written form changed to an alphabet later on.
Syllable Final Letter | Letters | Romanization | |||||||
---|---|---|---|---|---|---|---|---|---|
aː | a | i(ː) | u(ː) | aː | a | i(ː) | u(ː) | ||
Any other letter | a? | ? | aa | a | i | u | |||
y | ay | y | aai | ai | ie | ui | |||
w | aw | w | aaw | aw | iw | uu |
The table on the left uses neither Romanizations nor the letters of the actual language but rather Latin letters that have been chosen to have a one to one correspondence with letters of the actual abjad. While the script remains an abjad this table on its own doesn't show much of interest, merely that writing some letter causes that letter to appear in the document! However, the companion table on the right shows the possible Romanizations that can result from that letter, and from our previous discussion of the Romanization system you can infer the sounds. Thus, the two tables taken together show, for instance, that the letter shown as y in the left hand table can represent any of the sounds /j/, /aj/, /iː/ or /ui/. For clarity of presentation the unknown consonant that comes after the vowel in the "Any other letter" row has been omitted from the Romanization columns rather than being shown as a wildcard.
Syllable Final Letter | Letters | Romanization | |||||||
---|---|---|---|---|---|---|---|---|---|
aː | a | i(ː) | ɔ, oː | aː | a | i(ː) | ɔ, oː | ||
Any other letter | a? | ? | aa | a | i | o | |||
y | ay | y | aai | ai | ie | oi | |||
w | aw | w | aaw | aw | iw | oh |
Syllable Final Letter | Letters | Romanization | |||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
aː | a | i(ː) | u(ː) | ɔ, oː | e | aː | a | i(ː) | u(ː) | ɔ, oː | e | ||
Any other letter | a? | ? | aa | a | i | u | o | e | |||||
y | ay | y | aai | ai | ie | ui | oi | ei | |||||
w | aw | w | aaw | aw | iw | uu | oh | ew |
Syllable Final Letter | Letters | Romanization | |||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
aː | a | i(ː) | u(ː) | ɔ, oː | e | aː | a | i(ː) | u(ː) | ɔ, oː | e | ||
Any other letter | a? | ? | aa | a | i | u | o | e | |||||
y | ay | y | aai | ai | ie | ui | oi | ei | |||||
w | - | w | aw / w | w | - | iw | uu | oh | ew |
Syllable Final Letter | Letters | Romanization | |||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
a | i | u | ɔ | e | aː | iː | uː | oː | eː | a | i | u | ɔ | e | aː | iː | uː | oː | eː | ||
Any other letter | ? | a? | ? | a | i | u | o | e | aa | ie | uu | oh | ee | ||||||||
y | y | - | y | ay | y | - | ai | - | ui | oi | ei | aai | ie | - | |||||||
w | - | w | - | w | - | w | aw / w | w | - | iw | - | ew | - | iew | uu | oh | eew |
Syllable Final Letter | Letters | Romanization | |||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
a | i | u | ɔ | e | aː | iː | uː | oː | eː | a | i | u | ɔ | e | aː | iː | uː | oː | eː | ||
Any other letter | ? | a? | ? | a | i | u | o | e | aa | ie | uu | oh | ee | ||||||||
y | y | - | y | ay | y | - | ai | - | ui | oi | ei | aai | ie | - | |||||||
v | - | v | - | v | - | v | av / v | v | - | iv | - | ev | - | iev | uu | oh | eev |
Syllable Final Letter | Letters | Romanization | |||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
a | i | u | ɔ | e | aː | iː | uː | oː | eː | a | i | u | ɔ | e | aː | iː | uː | oː | eː | ||
Any other letter | a? | i? | u? | o? | e? | aa? | iy? | uv? | av? | ee? | a | i | u | o | e | aa | ie | uu | oh | ee | |
y | ay | - | uy | oy | ey | aay | iy | - | ai | - | ui | oi | ei | aai | ie | - | |||||
v | - | iv | - | ev | - | iyv | uv | av | eev | - | iv | - | ev | - | iev | uu | oh | eev |
Sounds | Letters | Can Have Coda V? | ||||
---|---|---|---|---|---|---|
Short | Long | With Y | Short | Long | With Y | |
a | aː | aj, aːj | a | aa | ay, aay | N |
e | eː | ej | e | ee | ey | Y |
i | iː | i | iy | |||
ɔ | oː | ɔj | o | av | oy | N |
u | uː | ui | u | uv | uy |
There's no copula verb in the present tense.
Nouns are not marked for masculine or feminine gender and verbs are not conjugated according to human gender. Objects don't have gender roles and people shouldn't be judged by their sex or gender! However, nouns are assigned to one of two grammatical genders (also known as noun classes) and these operate in the same way as divisions into masculine and feminine words operate in European languages. The two noun classes are named "bright" and "dark" and the words that belong in each group exhibit some of the properties present in the following lists.
Things that have some qualities from both lists and things that have none of the qualities from either list are assigned to one of the two noun classes based upon my general feeling about them. Ironically, there are two irregular words such that the word meaning night is a bright word and the word meaning day is a dark word.
Number | Person | Suffix |
---|---|---|
Singular | 1st | |
2nd | (ee)s | |
3rd | (ee)th | |
Plural | All | (e)n |
Adverbs are marked with an affix.
The language uses a consonantal root system. Roots come from two sources.
Where the Hebrew and the Arabic have conflicting meanings then the Hebrew meaning takes precedence.