--- title: "Finite State Automata and Arabic Writing — Fanton 1998" description: "Fanton argues that both can be handled with finite state machines. 10 references. Arabic script poses two recurring problems for software." authors: ["Michel Fanton"] year: 1998 venue: "Computational Approaches to Semitic Languages" source: "https://aclanthology.org/W98-1004/" licence: "CC BY-NC-SA 3.0 (ACL Anthology, before 2016)" licence_url: "https://creativecommons.org/licenses/by-nc-sa/3.0/" summary_licence: "Elamur Cosmic License" canonical: "https://research.elamur.ai/free/en/papers/fanton-1998-finite-state-automata-arabic-writing/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/fanton-1998-finite-state-automata-arabic-writing/) [home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [1998](https://md.research.elamur.ai/free/en/papers/1998/) / Finite State Automata and Arabic Writing # Finite State Automata and Arabic Writing **Authors:** Michel Fanton · **Year:** 1998 · **Venue:** Computational Approaches to Semitic Languages · **Licence:** [CC BY-NC-SA 3.0 (ACL Anthology, before 2016)](https://creativecommons.org/licenses/by-nc-sa/3.0/) **Links:** [Publisher's page](https://aclanthology.org/W98-1004/) · [PDF](https://aclanthology.org/W98-1004.pdf) · [OpenAlex](https://openalex.org/W2068845298) **Citations:** 4 citations (OpenAlex, 2026-10-14) ## Summary Arabic script poses two recurring problems for software. Each letter must be rendered in the correct positional shape (isolated, initial, medial, final), and the letter hamza must be written on the correct carrier. Fanton argues that both can be handled with finite state machines. The key step is a linguistic analysis that collapses the alphabet into a few relevant classes, so the automaton can be written directly instead of being compiled from contextual rules. For shape selection, letters are reduced to "binding" versus "non binding." A small nondeterministic Moore automaton then maps these classes to shape symbols, and a Prolog program verifies it by enumerating short inputs. For hamza, the input is reduced to consonant, hamza, sukūn and short or long vowels. A larger Moore automaton outputs one of five carrier types. This mainly matters at generation time, when vowels are known, for example in conjugating verbs whose roots contain hamza. Computing the spelling lets the system avoid listing every form in the dictionary. ## Key points - In 1983 ASMO adopted one code for all shapes of a letter. Hamza was the exception: its carriers received distinct codes, because short vowels are normally unwritten and the carrier cannot be inferred for display. - The contextual-analysis automaton uses source alphabet {#, n, b} and target alphabet {#, In, Ib, i, m, fn, fb}, with 8 states. - A Prolog test program generates every input of up to 6 symbols, word boundaries included, and prints the resulting shapes. - Hamza is handled in three steps: - a 21-state generating automaton describes the vocalized source language; - a 36-state translation automaton maps it to carrier symbols: hwc (no carrier), hoa (on alif), hua (under alif), how (on waw), hoy (on ya); - final short vowels are treated as sukūn, matching modern usage. - Madda is deliberately left out and handed to the derivation and conjugation stage. Short vowels and shadda are absorbed as self-loops on the current state. ## Interesting ideas - Before building a rule machine, partition symbols into the few classes the rule actually depends on (binding/non-binding), which shrinks the automaton enough to write by hand. - Split output symbols when the same label behaves differently by context (fn vs fb, In vs Ib), so that the target language itself encodes the constraints. - Verify a hand-written automaton by exhaustively generating every input up to a small length and inspecting all outputs. - Compute orthographic variants at generation time from known underlying information instead of storing every inflected form in a dictionary. - Push phenomena that belong to another level (madda arising from derivation) out of the current module instead of patching them in. ## For Mova How these ideas could serve Mova, our small open language model — its files, stages and tests: - **I1** (tokens): Morphology and tokenizer rules in crate en could be stated over small feature classes (for example vowel-final or sibilant-final stems) rather than raw characters, making each rule small and explainable. - **I2** (back): Generation could use context-split surface tags (for example a/an or contraction variants) so the realizer's legal sequences are checked by a simple grammar. - **I3** (learn): Small rule components such as agreement or inflection could be tested by enumerating all short feature combinations, as a complete check alongside sampled evaluation. - **I4** (back): The text → graph → text path could derive English irregular spelling changes (doubling, y→i, e-dropping) by rule from lemma plus features, keeping the lexicon small. - **I5** (tokens): Keep morphological processes separate from spelling adjustment in crate en, so each layer stays a clean, explainable step. *Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.* ## Institutions - [Institut National des Langues et Civilisations Orientales](https://md.research.elamur.ai/free/en/universities/institut-national-des-langues-et-civilisations-orientales/) — University, Paris, France · 5 works · 243 citations 📚 Shelves (5): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 18,614 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 17,965 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 17,450 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 17,296 works · [🧮 Computational Theory and Mathematics](https://md.research.elamur.ai/free/en/shelves/computer-science/computational-theory-and-mathematics/) · 321 works 🧭 Topics (1): [Semitic languages](https://md.research.elamur.ai/free/en/topics/semitic-languages/) · 3,066 works ## Cite Cite as: Michel Fanton. 1998. Finite State Automata and Arabic Writing. Computational Approaches to Semitic Languages. The paper itself, not this page. BibTeX from the library's record of the work. ```bibtex @inproceedings{fanton1998finite, title = {{Finite State Automata and Arabic Writing}}, author = {Michel Fanton}, booktitle = {Computational Approaches to Semitic Languages}, year = {1998}, doi = {10.3115/1621753.1621759}, url = {https://doi.org/10.3115/1621753.1621759} } ``` - [← previous Towards a linguistically motivated computational grammar for Hebrew](https://md.research.elamur.ai/free/en/papers/wintner-1998-linguistically-motivated-computational-grammar-hebrew/) - [next → Rationality, Cooperation and Conversational Implicature](https://md.research.elamur.ai/free/en/papers/lee-1998-rationality-cooperation-conversational-implicature/) - [↑ back to papers of 1998](https://md.research.elamur.ai/free/en/papers/1998/) ## References 10 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet. 1. A. V. Aho and J. D. Ullman. 1972. The Theory of Pars. ing, Translation and Compiling, volume 1: Parsing. Prentice-Hall. [find](https://scholar.google.com/scholar?q=A%2E+V%2E+Aho+and+J%2E+D%2E+Ullman%2E+1972%2E+The+Theory+of+Pars%2E+ing%2C+Translation+and+Compiling%2C+volume+1%3A+Parsing%2E+Prentice%2DHall%2E) 2. Arab League Arab Organization for Standardization and Metrology (ASMO), 1982. Data processing 7 bit Coded Arabic Character Set for Information Interchange. [find](https://scholar.google.com/scholar?q=Arab+League+Arab+Organization+for+Standardization+and+Metrology+%28ASMO%29%2C+1982%2E+Data+processing+7+bit+Coded+Arabic+Character+Set+for+Information+Interchange%2E) 3. Arab School on Science and Technology 1st Fall Session Rabat Morocco. 1983. Applied Arabic Linguistics and Signal Information Processing, P.O. Box 7028 Damascus Syria. [find](https://scholar.google.com/scholar?q=Arab+School+on+Science+and+Technology+1st+Fall+Session+Rabat+Morocco%2E+1983%2E+Applied+Arabic+Linguistics+and+Signal+Information+Processing%2C+P%2EO%2E+Box+7028+Damascus+Syria%2E) 4. Arab School of Science and Technology 7th Summer Session, Zabadani Valley - Syria. 1985. Informaties and Applied Arabic Linguistics, P.O. Box 7028 Damascus Syria. [find](https://scholar.google.com/scholar?q=Arab+School+of+Science+and+Technology+7th+Summer+Session%2C+Zabadani+Valley+%2D+Syria%2E+1985%2E+Informaties+and+Applied+Arabic+Linguistics%2C+P%2EO%2E+Box+7028+Damascus+Syria%2E) 5. R. Blach~re and M. Gaudefroy-Demombynes. 1952. Grammaire de l'arabe classique. G.P. Maisonneuve & Larose, 3" edition. 1985. Computer Processing of the Arabic Language. April 14-16, 1985 Kuwait. [find](https://scholar.google.com/scholar?q=R%2E+Blach%7Ere+and+M%2E+Gaudefroy%2DDemombynes%2E+1952%2E+Grammaire+de+l%27arabe+classique%2E+G%2EP%2E+Maisonneuve+%26+Larose%2C+3%22+edition%2E+1985%2E+Computer+Processing+of+the+Arabic+Language%2E+April+14%2D16%2C+1985+Kuwait%2E) 6. M. Fanton. 1997. L'~criture arabe : du manuscrit h l'ordinateur. La Tribune Internationale des Langues vivantes, (21), mai. [find](https://scholar.google.com/scholar?q=M%2E+Fanton%2E+1997%2E+L%27%7Ecriture+arabe+%3A+du+manuscrit+h+l%27ordinateur%2E+La+Tribune+Internationale+des+Langues+vivantes%2C+%2821%29%2C+mai%2E) 7. J. E. Hopcroft and J. D. Unman. 1979. Introduction to Automata Theory, Languages and Computation. [find](https://scholar.google.com/scholar?q=J%2E+E%2E+Hopcroft+and+J%2E+D%2E+Unman%2E+1979%2E+Introduction+to+Automata+Theory%2C+Languages+and+Computation%2E) 8. K. Lagally. 1992. ArabT~c~X a system for typesetting arabic user manual version 3.00. Technical Report 1993/11, Universit~t Stuttgart, Fakult~it Informatik, Breitwiesenstrafle 20-22, 70565 Stuttgart, Germany. Document ~lectronique fourni avec le logiciel. [find](https://scholar.google.com/scholar?q=K%2E+Lagally%2E+1992%2E+ArabT%7Ec%7EX+a+system+for+typesetting+arabic+user+manual+version+3%2E00%2E+Technical+Report+1993%2F11%2C+Universit%7Et+Stuttgart%2C+Fakult%7Eit+Informatik%2C+Breitwiesenstrafle+20%2D22%2C+70565+Stuttgart%2C+) 9. A. Lakhdar Ghazal. 1983. L'alphabet arabe et les machines. In Applied Arabic Linguistics and Signal Information Processing (Ara, 1983), pages 233-258. [find](https://scholar.google.com/scholar?q=A%2E+Lakhdar+Ghazal%2E+1983%2E+L%27alphabet+arabe+et+les+machines%2E+In+Applied+Arabic+Linguistics+and+Signal+Information+Processing+%28Ara%2C+1983%29%2C+pages+233%2D258%2E) 10. W. Wright. 1859. A Grammar of the Arabic Language. Cambridge University Press, 3 edition. [find](https://scholar.google.com/scholar?q=W%2E+Wright%2E+1859%2E+A+Grammar+of+the+Arabic+Language%2E+Cambridge+University+Press%2C+3+edition%2E) *Updated 2 November 2026*