Notes for using the Jaghbub -> Arabic converter

There are two variants, one which creates a text with vowels, and one that creates an unvoweled Arabic text. The latter is probably more useful. If you type faster in English script than with an Arabic keyboard, you can use this tool to enter Arabic text: Type the Arabic in transliteration, save and convert, and presto: an Arabic text file.

As we know, Arabic transliteration isn't compeletly unambiguous. You will need to look through and correct any errors or misinterpretations that the converter has done. In order to speed up this process, it may be useful to know some points about how the setup works (in the no-vowels version):

• All short vowels are deleted, so you can enter or not enter them as you like; except for the set phrases and words listed below, where correct vocalization is assumed; so it is better to have a vocalized text. Double consonants are written with shadda, so note in "fiqhihi" you must enter the vowel between the two h's to avoid it becoming a single h/shadda. All uppercase letters are converted to lowercase, so both may be used freely in the text.

• Kh, th, dh, are ambiguous: kaf-ha' or kha'. The converter assumes kha'. Thus, to enforce kaf-ha to split, you must force them apart with a short vowel, or if you find that distasteful, an apostrophe: mad'hash becomes m-d-h-sh (the apostrophe is deleted in the conversion. Do notice that e.g. madhhab does not fall in this category.) The "Germanic" single characters (d and t with line under, h with curve under) are recognized.

• Long a is always rendered as alif, also at the end of words. For alif maqsura, one must enter either á (the LC standard) or à. For pure convenience, since the accented a's often require more typing, I have also set e to produce alif maqsura. Either of these three can be used. Long i will give ya', with dots under.

• -a at the end of words are also ambiguous: verb form or noun with ta marbuta? The converter assumes that all final -a's are ta marbutas. Thus all verbs must be written "pausally", write dakhal and katab, not dakhala or kataba. Do remember it, e.g. with -ka suffixes; write -k (kitabak). The converter will recongnize ...at al- as ta marbuta, but otherwise a final short a may be used always to indicate a marbuta.

• For some particles, this will be hard do remember. Thus, exceptions are made for huwa, hiya, inta, inna, anna, dhalika, ayna, kayfa, lima, mima, 'inda, ma'a, fawqa, tahta, bayna, ba'da, qabla and wa when written in full like these; they will be converted as a group. Other special conversions are Allah, lillah, Rahman, hadha, hadhihi.

• The converter will attempt to recognize the various elision forms for al-, thus fi 'l-, fi'l, fi-l-, li'l-, li-l- etc. with normal apostrophe, "smart-quote" or hamza. However, smart quotes, n- and m-length hyphens and other special characters are to be discouraged, as they may not always be converted correctly (single "smart" opening quote is converted to 'ain, closing to hamza). Elision forms like at-Tahrir, ash-shams are not recognized and must be avoided. Similarly, elisions like wa'ttisal will not be recognized. Write wa-ittisal

• Also, notice that you must write the dash after al- (not walkitab) for the converter to know that it is the article (alkitab without a prefix, will, however work).
• The carriers of hamza is a very complex issue, and correct conversion cannot be guaranteed at all. The converter does try to follow the rules that initial 'a and 'u gives alif/hamza above, 'i alif/hamza below, and that the internal carrier is determined by the combinations of adjoining vowels. For these cases, vowels thus will help correct conversion. Hamza-long a will give madda, so when this is desired, type this combination. In general, a Jaghbub hamza will trigger an Arabic hamza, while a long a will trigger an alif without hamza, so you can use this to steer the desired use of of hamza or not. 'ukhra will be written with hamza on the alif, ukhra with initial u will be written with alif, but without hamza. (While it may sometimes be desired to indicate e.g. an initial i by hamza below, this must also cover cases of hamzat al-qat', where hamza is not to be written. Hence it is preferred never to insert initial hamza unless specifically indicated.) But the user is advised to look carefully at this in the resulting text.

• Another ambiguity is final long u, whether it should have a suffixed alif or not. The converter assumes not. You can force this either by typing a long a (which always gives alif) after the final u, or a star *, long u+* is converted as long u+alif.
• A few extra symbols you may use: The @ symbol is converted to a "lengthener", & will always be converted to alif, and $ to alif with hamza below, if you must force these two characters.
• The general policy of this converter is that one types the same transliterated diacritic characters that one is familiar with, as this will normally go more quickly. However, some may prefer to use some of the English characters that Arabic does not need (c, v, x) rather than typing option-characters. For those, these non-Arabic English characters are converted thusly: c converts to 'ain, g to emphatic h, s to emphatic s, v to emphatic d, o to emphatic z and p converts to emphatic t

These are some indications that may help speeding up the process of text entry and conversion. However, as indicated, any text carried through such conversions must be thoroughly proofread at the end, as some conversion mistakes are unavoidable.

The converter is part of the "Jaghbub Convertors" package, which you can download by clicking here.

Knut S. Vikør.
28.9.96