Appearance
Phonology and phonotactics
How to sound out Agazan. Each letter has one pronunciation. Syllables end with a consonant only at the end of the word (one rare number-word exception is under Advanced), so you know when words stop.
On this page
Beginner
Pronounce each letter the same way every time. A syllable is one beat with one vowel. Spell the beats you actually say: a consonant starts a beat, and the last consonant of a content word (a word built on a root, such as a noun or verb) is where that word ends.
Write native Agazan in lowercase. Two vowel letters in a row are two syllables. Say each vowel as its own beat.
Vowels
Agazan has four vowel letters. In English, a vowel letter’s sound depends on the letters around it; each Agazan vowel always has the same sound.
The cue words below are pronounced as in Standard American English.
| Agazan | IPA | Cue |
|---|---|---|
u | /ʉ/ | you (without the y) |
a | /ä/ | spa (like Spanish casa) |
o | /o̞/ | Cambodia (no glide, like Spanish todo) |
e | /e̞/ | bet |
Audio is from Wikimedia Commons under CC BY-SA 3.0. /ʉ/ and /ä/ are by Denelson83, and /e̞/ and /o̞/ by TFighterPilot.
Opposite vowels
The four vowels form two opposite pairs. Within a pair, the tongue moves along one line and the lips change shape:
| Pair | Tongue | Lips |
|---|---|---|
u ↔ a | high ↔ low, centered | rounded ↔ relaxed |
o ↔ e | back ↔ front, same height | rounded ↔ spread |
To find u, start from a: keep the tongue centered, close the jaw, and round the lips. To find o, start from e: keep the jaw where it is, pull the tongue back, and round the lips. Moving between the two vowels of a pair is a small step.
Consonants
A consonant starts a syllable. Many consonants have a voiced sound (vocal cords buzzing, as in zoo) and an unvoiced one (as in sea), and Agazan treats the two as the same letter. Prefer the voiced sound so you can hold a sung note; the unvoiced one is fine for style.
| Agazan | IPA | Cue | Unvoiced variant |
|---|---|---|---|
w | /w/ | we | |
d | /d/ | do | /t/ , toe |
r | /ɹ/ | red | |
m | /m/ | me | |
v | /v/ | vie | /f/ , fee |
g | /ɡ/ | go | /k/ , kite |
l | /l/ | lie | |
h | /ɦ/ | ahead | /h/ , hat |
n | /n/ | no | |
z | /z/ | zoo | /s/ , sea |
b | /b/ | be | /p/ , pay |
y | /j/ | yes | |
th | /ð/ | this | /θ/ , thin |
x | /ʒ/ | vision (the si) | /ʃ/ , shy |
Remember
x sounds different from English x. The rest are familiar. y is always a consonant, never a vowel.
th is one letter written with two characters. Agazan has no t, so th never means t followed by h.
Audio is from Wikimedia Commons under CC BY-SA 3.0 by Peter Isotalo, except /ɹ/ by Erutuon.
Word edges
A content word ends in one of four consonants: -l, -m, -n, or -r. That last consonant is the audible end of the word. Inside the word, consonants start syllables. They do not close a syllable in the middle.
azawan = a-za-wan. odogal = o-do-gal.
Intermediate
Reminder:
Intermediate sections assume you have read the beginner sections of every page.
Phonotactics (word shape)
Beginner already used word edges: a content word ends in -l / -m / -n / -r. Here is the full shape of that word, in order, written as one lowercase token (zazawan).
| Agazan | Use | English |
|---|---|---|
| Role letter | first letter | role in the clause (subject, verb, …) |
Optional l after /ɡ/ | adjective before the noun | gl- looks ahead to the next noun |
| Root(s) V(CV)+ | content | start with a vowel; each later consonant starts a new syllable |
Mid-word x | productive compound seam | joins two roots inside one word |
Mid-word th | stance seam | joins a sake to its stance vowel (sakes), any other root to a scope vowel (label scope), or a direction to whose facing counts (viewpoint laterals) |
Dictionary stem (no x) | lexical compound | one long simple-looking root (ebedaluruhe bedroom) |
-l / -m / -n / -r | word ending | audible end of the content word |
Name instance -ln | one of a name | word-final coda ln |
Optional -x | plural after the ending | word-final -lx / -mx / -nx / -rx / -lnx (letter x) |
Stand-in -rl / -rm | dependent clauses | word-final coda rl / rm |
Backward stand-in -rth | pointing back | word-final coda rth; the only word-final th |
A syllable ends with a consonant only at the end of the word. Inside a root, l and r always have a vowel after them, so they start a syllable rather than sounding like a suffix (zubuhel: prefix z, root ubuhe, ending -l). Each spelling has only one pronunciation. Writing does not mark stress. Musical rhythm may still place emphasis.
Some short grammar words stack two vowels (a join such as ua). A stack always puts its vowels in alphabetical order (u, a, o, e): u comes first and e comes last. That gives six stacks: ua, uo, ue, ao, ae, and oe. The two vowels stay two separate syllables.
Related form: word-initial x is the continue prefix (discourse), not a compound seam.
Number-word exception
When English says how many or which place, Agazan uses a number word. A number word puts r right after the role letter. Ordinary words never have that cluster, because their roots start with a vowel, so a role letter followed directly by r tells you a number is coming. The marker is ra, ru, re, or ro; counting from the end is written #-, spelled and spoken rue before the digits. Two vowels in a row stay two separate syllables.
zagadulx grarel.z-cat-x | g-three
"Three cats."
grarel is g + ra + digit re + -l. In a long number, each written comma is spoken as a group separator, th plus the marker’s vowel.
Compare with: ordinary endings on content words use word ending senses. Number words reuse those same four letters with number-specific endings.
Letter names
When you spell a word aloud or name a letter, say the Agazan name for it. Pause between names so two names do not run into one syllable.
The table is in alphabetical order. The ten letters that begin a digit syllable come first, in digit order (1 to 9, then 0), so w is the first letter and z the tenth. The other consonants follow, and the vowels come last.
| Agazan | Name | Cue |
|---|---|---|
w | we | wet |
d | da | Dada |
r | ro | row |
m | me | met |
v | vu | voodoo |
g | ga | gaga |
l | lo | low |
h | hu | who |
n | nu | noon |
z | ze | zen |
b | be | beg |
y | ya | yacht |
th | tho | though (no glide) |
x | xu | shoe |
u | u | you (without the y) |
a | a | spa |
o | o | Cambodia (no glide) |
e | e | bet |
text
`agadu` → `a` `ga` `a` `da` `a`In a clause, a letter you talk about (such as z) is a mention; read it aloud by its letter name (ze).
Ten letters also begin a digit syllable. The letter name uses the opposite vowel (a↔u, o↔e), so naming the letter is not the same as counting.
| Agazan | Digit syllable | Name | Cue |
|---|---|---|---|
w | wo (1) | we | o ↔ e |
d | du (2) | da | u ↔ a |
r | re (3) | ro | e ↔ o |
m | mo (4) | me | o ↔ e |
v | va (5) | vu | a ↔ u |
g | gu (6) | ga | u ↔ a |
l | le (7) | lo | e ↔ o |
h | ha (8) | hu | a ↔ u |
n | na (9) | nu | a ↔ u |
z | zo (0) | ze | o ↔ e |
Advanced
Singability constraints
These choices about word shape make Agazan easier to sing:
| Constraint | Why it helps | How Agazan keeps it |
|---|---|---|
| One vowel per letter | Sliding from one vowel to another mid-note forces the mouth to change | Stacked vowels = separate syllables |
| Few consonant clusters | Clusters are harder to say quickly | Ordinary shape is consonant-then-vowel at the start of a syllable; limited clusters below |
| Voiced preferred | Voiceless stops cut the note; singers often voice them anyway | Voiced preferred; unvoiced allowed as style only |
| No mid-word syllable-final consonant | Ending a syllable on a consonant breaks a held note | Roots are V(CV)+; a final consonant only at the word edge or a lexical join (-l / -m) before a number marker r |
| No lexical stress | Music already places emphasis | Stress always falls on the same beat (rhythm and stress); spelling does not mark it, and music may move it |
| Spelling = pronunciation | You do not memorize special readings | One path from letters to sound |
| Audible word edges | Song often removes speech pauses | Content words end in -l / -m / -n / -ln / -r / -rl / -rm (optional -x); stand-ins end in -rl / -rm / -rth |
Legal clusters: left-hanging gl-; number-word role letter + r; tag pronoun role letter + w (zw, dw, bw); lexical join -l / -m plus number marker r on a kind morph; word-final -lx / -mx / -nx / -rx; name instance -ln / -lnx; stand-in and sense-pinned resume -rl / -rm, stand-in -rth. The lexical join before r is the only syllable-final consonant inside a word.
Try singing a short Agazan line quickly at a high but comfortable pitch:
zazawan gamadam.
/ zä.zä.wän ɡä.mä.däm /
A line that piles up close vowels, clusters, and mid-word stops is harder to sustain even when it is only a little harder to speak:
/ seiɹ ˈʈʂuɹt tis ˈheb.ɡiɹn fuofts /
Rhythm and stress
Give every syllable about the same length, and keep its vowel full however fast you talk. English weakens unstressed vowels to uh (the second a in banana). Agazan never does, because many words differ by one vowel (val, vol, vul), and weak vowels would make them sound alike.
Each word takes a light stress on its first syllable. Since the rule never changes, spelling does not mark it. The ending consonant already tells a listener where a word stops; the stressed first beat tells them where the next one starts, so fast speech still splits into words. A number word stresses the last digit of each group instead.
zazawan gamadam.
/ ˈzä.zä.wän ˈɡä.mä.däm /
Hold the ending. Let the final -l, -m, -n or -r sound for a moment instead of clipping it. It is often the only difference between two words (vul, vum, vur), and all four are sounds you can hold. In singing, put the cutoff on it.
Words never said quietly
English swallows small words that change the meaning: can and can't often differ by one weak sound, and I think is usually mumbled. In Agazan, give these words their full stress even when the words around them are light:
| Words | Why |
|---|---|
| Negation: the u joins (negation) | Missing one flips the claim |
| MAY and its hold endings | Missing one makes a guess sound certain |
| How you know and how strong the evidence is | Missing one hides where the claim came from |
| The kind of can't (ability) | Missing one turns not right now into not ever |
Voice and certainty
Only the words and the written tone marks carry meaning. How firmly you stand behind a claim is the ending on the act word; how sure you are is MAY. Your pitch may agree with them, but an unwritten rise or fall adds nothing. A statement said with a rising voice is still a statement, and a firm -l claim said hesitantly is still firm.
So listen to the words, not the voice. A speaker whose voice naturally rises at the end is not hedging, and a confident voice is not evidence.
Speech edges
Each edge in a sentence has its own sound. From smallest to largest:
| Edge | Voice | Taught in |
|---|---|---|
| Word | Stress on the first syllable, ending held | rhythm and stress |
| Item in a list, before the join word | Level pitch, no fall: more is coming | right-close fence |
| Join word that ends a list | Fall | right-close fence |
| Scope island | Brief pitch reset, one tight phrase, pause after its last word | scope islands |
/x/ continue, linker, stand-in | Dip, no pitch reset | periods |
| Period | Fall on the last word, short pause | periods |
| Topic word | Pitch reset; same speech act | topic |
New /y/ turn | Full pitch reset | periods |
The level pitch on list items matters because the join word comes last: it lets a listener hear that a list is under way before the join arrives.
Singing on one note
Every distinction Agazan makes is spelled in its letters, so a line sung on a single pitch still says everything it says in speech. When a melody takes over your pitch, voice each tone mark with how you shape the note:
| Mark | In speech | In song |
|---|---|---|
! | Louder, stressed | Accent: a stronger attack, louder |
? | Rising, tentative | A lighter attack, slightly late |
% | Light, smiling | Short and detached |
& | Stressed and slowed | Held for its full length |
; | Soft, gentle | Smooth and connected, softer |
Thinking sounds
When you need a moment in the middle of speaking, English fills the gap with um or uh. Agazan has two sounds for that gap, and each tells the listener what would help. Neither is a word, and neither is written.
| Sound | Means | The listener | Cue |
|---|---|---|---|
| mmm, lips closed | I am thinking it over | waits and leaves the silence | Lips closed: not ready to speak |
A held a, e or o (any of them) | I know what I mean but cannot find the word | may offer a word | Mouth open: the word is on its way |
Do not hold u: it is the undo vowel, so a held u can sound like a denial. A thinking vowel is held well past one beat, so it does not sound like a letter name, which is short and followed by a pause.
Compare with: a thinking sound only holds the gap. To fix a word you already said, use a hook: el for a better wording of the same thing, ol for the thing you meant instead.
See also
- Citation (root + ending, no sentence): word-endings.md
- Role letters in a clause: clause.md
- Numbers (role letter +
rexception and digit syllables): numbers.md - Mention spans (letter as form in a clause): spans.md
- Mid-word
xandth: x-compounds.md