A Cross-Cultural Writing System
Checkout the Discord server with any thoughts/comments/questions you have, or start a GitHub discussion or add something to the Reddit group if you'd like as well.
Tone Text (esoteric name Hanákana) is a Universal Writing System (UWS), capable of representing all human voice sounds in a concise and readable script. We like to think of them as modern runes. This means you can write any human language on Earth in tone text (constructed language or natural language) and read it without having to learn that language's specific quirks. Here is an example of the modern runes using the new ToneEtch font. You can try it in the interactive online Editor.
Tone text is a sound-based script like the Ancient Runes, which means when you read see a symbol/letter, you know exactly how it is pronounced. There are no weird edge cases or specific words for which you have to memorize the sound. You simply learn the script and can pronounce the words and texts.
You may note that Linguistics also has the International Phonetic Alphabet (IPA), which seems to accomplish a similar thing. But the IPA is not for writing anything but the most detailed pronunciation guides, not for writing sentences or long texts. Tone Text is meant for the later, for writing like you would in a normal language.
The ASCII encoding is mapped to IPA for your knowledge in the Chat Text. Ideally, Tone Text would be included in the Unicode standard, but that is unlikely. We have suggested it be used as a writing system on Mars 😂.
It doesn't necessarily play well with dyslexia, which might find the highly symmetric patterns difficult. We think there are far better methods of learning outside of writing systems than trying to invent the perfect script for all cases. To most natural languages, dyslexia is an unsolved problem. We instead strive to make a minimal seed from which knowledge can be modelled from primitive parts, memorized, and then used to derive further knowledge.
Tone Text is also a monospaced font, to be used eventually for
programming. All sounds are derived from 3 base shapes, that for i,
a, and u. Rotating and extending tails on them gives you the
remaining base letters. Then adding dots to them extends the sound of
the base letter slightly. From this it should be straightforward to
memorize how to pronounce words in a day or so, and from there you can
read and write any language pretty well.
import tone, {
symbols,
consonants,
vowels,
numerals,
} from '@termsurf/tone'
// Makes TalkText ASCII format into Tone-compatible font format.
tone('aa+xa+a-ita-') // => aa3xa3a4ita4These are the main sounds. Everything else are variations of these sounds. These "variations" are accomplished in a structured way, see the guide. Indic retroflex consonants are added with a dot on the tailless side. Arabic sounds are added with a dot on the tailed side usually. Clicks are mapped to the closest mouth-position-oriented sound. Ejectives and implosives are simply added with accent marks. Etc.
Alphabetical order is this (1-36):
i
e
a
o
u
A
E
I
U
O
u$
o$
m
n
q
g
d
b
p
t
k
h
s
f
v
z
j
x
c
C
y
w
Q
'
l
r
Everything else is a modification on this base. Here are many more sounds. They are given the numerical value of the corresponding base symbol, and are ordered correspondingly, with retroflex, then pharyngealized, then ejectives, implosives, stops, tenses, clicks. The stops and pharyngealizations don't get any extra points.
N n
L l
X x
D d
J j
T t
R r
H h
K k
V v
F f
d~ d (pharyngealization, 1 dot on tailed side)
t~ t
s~ s
C~ C
l~ l
p! p (ejective)
t! t
k! k
s! s
x! x
T! t
K! k
g? g (implosive)
b? b
p. p (stop, 2 dots on tailless side)
t. t
k. k
p@ p (tense, 3 dots on tailless side)
t@ t
k@ k
s@ s
d* d (click, arc on tailless side)
p* p
t* t
k* k
l* l
Here is the consonant IPA phonology chart mapping.
There are at least 10 classes of numerological encodings:
- 36 sound encoding. This is where letters are in order based on their sound, numbered 1-36.
- 29 sound encoding. This counts the 14 voiced and unvoiced consonants as 7 instead of 14, as described below.
- 22 shape encoding. These are the unique shapes ignoring mirror symmetry. qjfvQxschwkbaoeiE2uAI1.
- 16 shape encoding. These are the unique shapes ignoring rotational symmetry. qnfrQyszhmkpaeIO.
- 9 shape encoding. These are the unique shapes ignoring rotational and mirror symmetry. They are also numbered similarly. qfQshkaeI.
- 12 shape encoding. These are the tailless shapes, including all symmetries. They are numbered i, a, o, u.
- 10 shape encoding. These are the tailless shapes, ignoring mirror symmetry. They are numbered i, a, u.
- 4 shape encoding. These are the tailless shapes, ignoring rotational symmetry. They are numbered i, a, o, u.
- 3 shape encoding. These are the tailless shapes, ignoring rotational and mirror symmetry. They are numbered i, a, u.
- 2 shape encoding. These are comparing tailed vs. tailless characters, ignoring rotational and mirror symmetry.
The most common are the 36 sound encoding, the 16 shape encoding, and the 9 shape encoding. The 16 shape encoding is how you would play a game with one-sided tiles. The 9 shape encoding is how you would play a game with two-sided tiles. The 16 shape encoding can be encoded in hexadecimal. The 9 in digits 1-9.
Here is the 29 sound encoding.
i
e
a
o
u
A
E
I
U
O
u$
o$
m
n
q
g k
d t
b p
h
s z
f v
j x
c C
y
w
Q
'
l
r
To get access to the font for your own system, just click and download the ToneEtch.otf directly.
- Keyman Development for the Keyman keyboard.
pnpm make:keyboardMIT
This is being developed by the folks at TermSurf, a California-based project for helping humanity master information and computation. Find us on Twitter, LinkedIn, and Facebook. Check out our other GitHub projects as well!