the documented Franco writing system.
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

22 KiB

title githubrepo description date tags
compliant franco suckless, standardized, documented and eloquent Arabic writing system 2019-03-01 [writing systems Arabic linguistics]

state of the project

personally i stopped using this system, now i use english to solve the problems discussed here. but it's usable for any one agrees more with my old position. if you want to change something or you want to mantain the project i can help you, pull requests are welcome on the github repository.


this project is supposed to be an implementation of ideas introduced by this research article. to understand the rationale behind all of this, or to have a better insight about the issue of the Arabic writing system, please refer to the article.


  • to have a standardized, documented, eloquent, and community developed Franco flavor.
  • to have a consistent, searchable and bidirectional friendly Arabic writing system.
  • a reference for people with unreadable Franco

Boundary conditions

  1. Franco is a writing system for Arabic speakers, not a way to write Arabic to be pronounced by English speakers.
  2. Franco constrains itself to ASCII characters not to English alphabet, however, it should not conflict with it (English alphabet).
  3. as much as we can, there should be only one "right" way to write most word.


those abbreviations and acronyms are used throughout the document :

  • TAWS
    the Traditional Arabic Writing System
  • CvC
    consonant + short vowel + consonant


this is not a transliteration system, but for your convenience i've compared the symbols of Franco to those of TAWS which we are used to - instead of IPA for example.

TAWS Franco
ا a
ء 2
ب b
ت t
ث $
ج g,j
ح 7
خ 5
د d
ذ 6
ر r
ز z
س s
ش 4
ع 3
غ 8
ف f
ق 9
ك k
ل l
م m
ن n
ه h
و w
ى y


Digraphs are two letters that behaves as a unit and is not the sum of its parts, e.g. "ch", "sh", "kh" and "th" in English.
Digraphs are basically an ugly hack to the English writing system, when used with other languages which uses them a lot, as in our case, it makes words longer and harder to read, specially for repeated letters, i.e. shadda.

English Franco
kh 5
sh 4
gh 8
dh D
th in throw $
th in the 6

don't start with "2"

since in most Latin languages e,o,a in the beginning of words contain the "2" in their pronounciation, we can borrow this property.

example should be
2ana ana
2egry egry
2oddaam oddaam
2e2laam e2laam
2ul ul


consonants in many languages can be velarized, velarization is a property that gives a duller sound to the consonant which also affects the neighboring vowels and consonants. in the egyptian language/dialect we often prefer to have an entire veralized syllable, so for convenience and readiability (but not necessarily phonetically accurate) we assume that any vowel between velarized consonants is also veralized (the nucleus of the syllable).

from wikipedia

Velarization is a secondary articulation of consonants by which the back of the tongue is raised toward the velum during the articulation of the consonant.

in TAWS "ض ط ظ ص" are examples of veralized consonants that have a character representation, but in reality other consonants such as "l,r,b,m" can also be velarized. those consonants are assigned capital letters in compliant Franco.

since Capital letters doesn't have any phonetic differences than lower case letters in all alphabetic writing systems, it's perfect to represent velarized consonants in compliant Franco (and long vowels also).

one great thing about this representation is its backward compatibility, by backward compatibility i mean if someone wrote ط as T instead of t, it won't effect the interpretation of the word if the receiver doesn't know this mapping. we basically lose nothing - kinda - but we gain more clarification and disambiguation.

importance of case sensitivity

without those we have no distinction between many words, for example:

  • fos7a and foS7a
  • bass and baSS
  • 7aZZ and 7azz
  • GAARy(my neighbor) and gaary (current)
  • garraa7 and gaRRAA7
  • etrabbeet and eTRABBeet
  • SAAB and saab
  • mafRUD and mafrud
  • enSef and ensef
  • DALL and dall
  • BA3D and ba3d
  • TAAR and taar
  • BAAT and baat
  • DAHR and dahr

what we lose

we obviously lose capitalization of names in Franco text, which is not used already.

diacritics and vowels

short vowels

the table below shows the basic diacritics transliterations.

traditional Franco Unicode name

long vowels

traditional Franco Unicode name
ا + َ aa FATHA + ALEF
و + ُ u DAMMA + WAW
ى + ِ i KASRA + YEH

there is also vowels that can't be expressed by TAWS, as traditional Arabic has no such vowels in its inventory. for those we use "ee" and "aa"

diacritic examples
ee leeh (why), la2eet
oo moot (death), noom

importance of long vowels

again, without those we have no distinction between lots of words, for example:

  • zaman and zamaan
  • men and min
  • 7ARAM and 7ARAAM
  • malak, malaak, maalak

why we need new vowels not in TAWS

the traditional system has no way to resolve the difference between for example:

  • moot and mut
  • kun and koon
  • ru7 and roo7
  • TIR and TEER
  • betu3 and betoo3
  • din and deen
  • fu2 and foo2

more examples

for didactic purposes someone may want more examples.

  1. ee

    • yareet
    • been el-beneen
    • ee (what)
    • leela
    • 8eer
    • 4ee2
    • feen
    • etneen, meteen, toffa7teen, etc.
    • la2eet, 7abbeet, geet, etc.
    • 3eneek, regleek, edeek, etc.
    • 3eneeha, regleeha, edeeha, etc.
  2. oo

    • 4ook
    • 4oo2
    • dool
    • loola
    • Soom
    • loon


shadda in TAWS is a diacritic that is phonetically equivalent to a repeated letter. in Franco it is represented by - yeah, you guessed it - a repeated letter, e.g. YALLA, ba7ebbek, kollo.

people with poor Franco often omit the second letter causing one of the biggest sources of ambiguity in mainstream Franco text, as omitting the second letter normally yields a different word. try to convince your wife that the message "hagElak bel-MaRa" she found on your phone is supposed to be "hagElak bel-MaRRa" to know how important this is. i can't emphasize more that the second letter is not optional and text compliant to this spec should always write the shadda appropriately wherever it appears.


  • 4adda
  • howwa
  • heyya
  • lessa
  • 5omsomeyya
  • 9essa
  • mosta3edda

The hyphen

the hyphen, i.e. "-", has three use cases:

  • after the definite article
  • when we want to emphasize that two characters should not be interpreted as a digraph. this occurs a lot in the "sh" and "th" sequence. for example: sme3t sot-ha, byet-hayya2lak, as-hal, btet-haz
  • to fit English words in Arabic grammer like: aa3ed ba-chill, 3reft a-configure el-btA3.

the definite article

the definite article in compliant Franco is not as simple as "the" in English or "-ال" in TAWS. generally it can appear as "el-…", "l-…", "eX-X…" or "X-X…"(where X represents on of the sun letters discussed here).

for programmers, this Emacs-style regular expression will match all possible forms a definite article can take in compliant Franco:


sun letters

some words start with letters called sun letters (shown below) that has the effect of turning the "l" to a copy of itself, i.e. shadda. for example el naharda should be written as en-naharda like how it's actually pronounced

some examples:

letter example
t et-tegaah waa7ed
$ e$-$a9AFA el-3ARABEYYA
6 wa9t e6-6orwa
d omm ed-donya
D el-wa2t eD-DAAYE3
r ya de7ket eR-ROMMAAn
z ez-zaman TAWIl
Z eZ-ZAAher keda
n ER-Rakk 3an-neyya
s zayy es-soKKAR
S eS-SO7AB fe agaaza
4 3ala add e4-4oo2
k fek-konaafa
g eg-gee4 we4-4ORTA

moon letters

moon letters are straight forward, they are not sun letters and they don't touch the "el".


  • el-lo8a
  • el-3ammeyya
  • el-9AhERA
  • el-ward
  • el-balady
  • w keda..

to "e" or not to "e"

that is the question. the rule is similar to what's described later in shortening. it depends on the preceding word, if it ends with "a" or "o", the e is ommited. for example:

  • ana l-awwal wana l-aa5er
  • 3ala l-maa4y
  • lessa d-donya b 5eer
  • yabo r-reggaala


TAWS Franco example
ب b/be b we44 reglak
و w/we a7la w a7la
l/le a4ky l min
في f/fe kalaam fe kalaam
على 3ala 3ala maagy
من men 5o44 men hena
ف fa konn fa yakun
اللى elly elly hnaak da
ما ma 3omrena ma hanerga3

for description of the cases where the e is removed from be/we/le/fe, see shortening.

prepositions compound with definite article.

traditional Franco example
من ال mel mek-ko4k
ما ال mal tamAm TOl mal-balad tamAm
عال 3al 3al-mA4y
وال wel bel-yoom wel-yoomeen
لل lel Ray7a lel-banAt
في ال fel fek-konAfa
ب ال bel bel-habal

those prepositions inherit the same rules of the definite article, by that i mean for example "fel-konAfa" should be written "fek-konAfa".

demonstrative pronouns

traditional Franco example
دي de el-betA3a de
ده da el-betA3 da
دول dool el-betO3 dool


in the Egyptian dialect/language, we tend to not pronounce the first occurrence of short vowel in some words, for example sometimes we say "kfaaya" and sometimes we say "kefaaya", in this section i will discuss this phenomena. all the conditions in this section also applies to prepositions be/we/le/fe.

conditions for shortening

whether shortening will take place or not is governed by the ending of the preceding word, that is, if the ending of the preceding word is one of the following three cases, the vowel is deleted,

ends with a vowel(e,o,a), w or y.

the most occuring case is if the preceding word ends with a vowel, for example we say "kefAya keda" and we say also "keda kfAya".

examples :

  • ana ktir
  • ta3aala hnaak
  • ya 5SAARA
  • ana ndemt
  • mahma kbert
  • alby m3aak
  • ta3aalu nnaam

ends with two consecutive consonants

another case where the vowel is deleted, is when the preceding word ends with two consecutive consonants like "4oft", "", "esta7melt".. etc, for example:

  • 4oft ktir
  • ro7t l3ebt
  • 4a3r TWIL
  • 4e3r kwayyes
  • la7m r5IS
  • 3e4t snin
  • 7elm kbir

you may have noticed that this case is reducible to the first case, since we can't pronounce more than two consecutive consonants (the two from the end of the word and the first of the following) we add a very short vowel - called epenthetic vowel - at the end of the first word to make it pronounceable, so it becomes "4ofte ktEr" which is exactly the first case.

this topic is actually more complicated than that, but most of the time it's clear for the Egyptian speaker which way to go, don't think about it too much.

please note that in searching franco text we should take into consideration both writings, for example this regex "ke?tir" will match every instance of the word regardless of whether shortening happened or not.

ending shadda

in compliant Franco some words must end with shadda for reasons discussed here, this shadda appears in three cases (or two actually).


for words that has the shape of (consonant + short vowel + consonant) like "7OR" for instance, the last consonant is written as if there a shadda at the end, "7ORR".

examples :

  • 7ORR
  • 7add
  • hezz
  • bass
  • zayy
  • koll
  • morr
  • sett
  • foll
  • TOLL
  • 3akk
  • fann
  • lakk
  • 3amm

the middle vowel must be a short one, for example "kaan" is not pronounced "kaann".

something + stop + CvC

some words doesn't have the shape of CvC, but ends with it preceded by brief stop, for example "haylef", those words are pronounced like "hay.lef" where the dot represent the brief stop. this is practically equivalent to pronouncing it as two separate words. here also the shadda is added by the same means we added it to "leff".


  • a7ebb
  • ye7ebb
  • ma3le44
  • ka2enn, akenn
  • momell
  • betmorr

note that the stop before the cvc is mandatory for the shadda to appear, for example "la2et" ends with CvC but the stop isn't there and so the shadda.

(a,o,e) + consonant

if you remember from the don't start with "2" section, words that start with "a","o" or "e", when pronounced they begin with a hidden "2", so "omm" in reality has the shape of CvC, that is, the first consonant is the hidden "2", the second vowel is "o" and the third consonant is "m", which is exactly the first case.


  • omm
  • ABB
  • A55
  • add (add ed-donya)
  • OTT
  • ayy
  • all (el-waga3 all 4wayya)

rationale behind the shadda at the end

you may ask the following: what's the problem with writing it without the shadda like "7or" or "bas" since there is no ambiguity the shadda resolves. for two reasons:

  1. the reality is that most people now write unreadable Franco, and most people are used to read unreadable Franco, this caused short/long vowel blindness. for example a word like "sab" will be interpreted by most people as "saab", but "sabb" will be easier for them to read
  2. the shadda actually exists if there is a following word, for example "sabb din". when you pronounce it you will notice a very short "e" after "sabb", it will sound something like "sabbe din".

3)consider this sentence - without the shadda at the end - "haylef ylef w yerga3ly", if you pronounced it in Egyptian Arabic, you will observe two things:

  1. you shorten "yelef" so that it becomes "ylef" and you shorten "we" becoming "w"
  2. you pronounce a shadda in both "haylef" and "ylef"

but by the rules of shortening the shortening in "yelef" and "we" shouldn't be there, since the preceding words "haylef" and "ylef" respectively doesn't end by a vowel nor a two consonants, but if we wrote explicitly the shadda observed by pronunciation, the second rule of shortening apply (the preceding word ends with two consonants), which explains why the shortening happens. so writing the sentence like this "hayleff yleff w yerga3ly" just solves every thing.


Assimilation here is a term for sound changes that occur within a word to ease pronunciation, for example "ganby" becomes "gamby".
not all writing system reflect assimilation in its orthography, but since this writing system relies heavily on phonetics, and wants to be predictable and consistent, i choose to reflect it, so the "right" way to write, e.g. "zanby" is "zamby", "anbuba" is "ambuba"

general notes

Arabic names

Arabic names should not be written in Franco, simply because no one has the right to decide how a name is pronounced other than the name holder himself, but it's ok to write a name in franco if you don't know how it's pronounced.

for example: mohammed mounir not ma7ammad monir.

e or y in the end

by specifications defined above, y and i in the end have no difference and can be used interchangeably, for consistency we will use y. e.g. "siby" not "sibi".

no place for those

compliant Franco does not contain any of the following english characters: [q,c,x,u,p,v], except in English words or human names.
similarly the following sequences of characters doesn't appear in compliant Franco: [sh,kh,th,gh], instead, those characters appear hyphenated to prevent confusion, e.g. s-h,k-h,t-h. see The hyphen.

additional shortening cases

in "we" the e is deleted when the following word is starts with "a,o,e"

  • in addition to the normal conditions of shortening. for example in "a4ky l'mEn w'a7ky l'mEn" the "we" is shortened despite the normal conditions of shortening does not hold here, but it's because a7ky starts with "a"

    • haru7 w aagy b SOR3A

sample text

it's known that the beauty of a language comes from the beauty of text it has been written in. below are different pieces of text written in compliant Franco, that i think are beautiful.

"ek-kelma deen" by ahmed foad negm.

morr ek-kalaam.. zayy el-7osaam
YE2TA3 makaan.. maymorr
amma el-madi7.. sahl w mori7
ye5da3 SA7I7.. w y8orr
wek-kelma dein.. men 8eir edeen
bass el-wafa.. 3al-7orr

poems by mostafa ibrahem:

9awaa3ed ed-dayra es-saba3a (5):

ezzaay te3raf..
min saabe2 min gowwa ed-dayra..
w min fiha ATA3 ATWAL ME4WAAR..
me4 momken BARDO elly f DAHRAK..
YETLA3 sab2ak..
w enta elly 7MAAR ?!

9awaa3ed ed-dayra es-saba3a (3):

far2 ES-SOR3AAT gowwa ed-dayra
beenk we mabeen koll el-ma4yin
howwa elly haye2dar YEDMANLAk
en enta t2aabel naas tanyin

corner cases and limitations

searchability limiting case

due to the structure of the Egyptian dialect/language (and human languages in general) some sentences can be pronounced in multiple ways and the writing system can't force one way or the other.


  • "gondy f gonudek" or "gondy fe gnudek"
  • "3aayez anaam" or "3ayz anaam"

irregularity in the ending shadda

  • "lak" in "ana lak 3ala TOl", "lakk" is another word.
  • "men" in "men hena", "menn" is a another word.


what is the problem with the traditional system?

this article may answer this question, and more.

why this guideline written in English and not in Franco?

i agree this would be a cool idea, but this is an evolving guideline, if it is written in Franco, every update in the specification means rewriting every thing to be compliant with itself.

are you working on Franco to replace TAWS?

Franco is just an alternative writing system that can coexist with any other writing system. i think Franco is very capable and has more potential than TAWS in many situations, like writing modern Arabic dialects and bidirectional text(mixed Arabic and English), but compliant Franco is not yet near ready for replacing TAWS in writings such as the Quran for example. for now we just focus on making a superior system, and maybe we can ask this question again later.


independent map of consonants

| manner          | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
| --------------- | - | - | - | - | - | - | - | - |
| Plosives:       |   |   |   |   |   |   |   |   |
| \-\> unvoiced   |   |   | t |   | k | 9 |   | 2 |
| \-\>\> emphatic |   |   | T |   |   |   |   |   |
| \-\> voiced     | b |   | d |   | g |   |   |   |
| \-\>\> emphatic | B |   | D |   |   |   |   |   |
|                 |   |   |   |   |   |   |   |   |
| Fricatives:     |   |   |   |   |   |   |   |   |
| \-\> unvoiced   |   | f | s | 4 | 5 |   | 7 | h |
| \-\>\> emphatic |   |   | S |   |   |   |   |   |
| \-\> voiced     |   | v | z | j | 8 |   | 3 |   |
| \-\>\> emphatic |   |   | Z |   |   |   |   |   |
|                 |   |   |   |   |   |   |   |   |
| the R-sound     |   |   | r |   |   |   |   |   |
| \-\>\> emphatic |   |   | R |   |   |   |   |   |
|                 |   |   |   |   |   |   |   |   |
| the L-sound     |   |   | l |   |   |   |   |   |
| \-\>\> emphatic |   |   | L |   |   |   |   |   |
|                 |   |   |   |   |   |   |   |   |
| Nasals          | m |   | n |   |   |   |   |   |
| \-\>\> emphatic | M |   |   |   |   |   |   |   |
|                 |   |   |   |   |   |   |   |   |
| Semi-vowels     | w |   |   | y |   |   |   |   |

where the numbered columns represent the place of articulation:

  • 1= with both lips
  • 2= with the lower lip against the upper teeth
  • 3= with the tip of the tongue against the back of the teeth
  • 4= with the back of the tongue against the hard palate
  • 5= with the back of the tongue against the soft palate
  • 6= with the back of the tongue against the uvula
  • 7= in the pharynx
  • 8= in the glottis (glottal)