I am interested in how to transcribe the following Arabic letters oh English. The letter "families" I Ann interested in are:

  • ع

  • غ

  • ر


  • ك

  • ق

  • ء

I think one of the first two would be a "gh", but not sure about the other, while the third would be an "r".

The fourth might be a "k", while one of the fifth and sixth a "q", but not sure about the other.

Thank you for your responses.

  • 1
    I'm voting to close this question as off-topic because it is general reference: See column letter name in the table. the answer is ayn, ghayn ra, kaf, qaf, hamza or a, gh, r, k, q, ? (a glottal stop)
    – Mitch
    Commented Oct 26, 2016 at 13:25
  • As stated, the transcriptions reported in the Wikipedia page are lossy. Someone needs to fox the problem. Commented Oct 26, 2016 at 14:50
  • Where is that stated? Here on SE or on the wiki page? What is 'lossy'?
    – Mitch
    Commented Oct 26, 2016 at 14:54
  • Please read the comments for the other answers as well. Lossy means, the transcription our transliterations provided on the page do not account for reproducing the sound of all words correctly, or at lest unambiguously. Commented Oct 26, 2016 at 14:58
  • 1) I don't see any comments to the other answers to your question. 2) 'correct' transcription is near impossible given that different languages sometimes have impossible sounds in others. There's no guarantee of the possibility of unambiguity. Are you sure that that guarantee is possible from Arabic to English?
    – Mitch
    Commented Oct 26, 2016 at 15:04

4 Answers 4


As Michau said, it depends on your use case. User6726 and Michau have provided a good overview of scholarly/lossless transcription systems. But since you mention specifically transcribing for English, there's another option.

If your goal is to have the word pronounced (something vaguely like) properly by English-speakers, or to use a name in an English work, then using common letters might be more important than being able to reconstruct the original Arabic from your transcription.

In this case, you could use a lossy transcription, merging similar letters together and using familiar English digraphs.

For example, both د and ض are very close to English "d", and it takes significant practice for English-speakers even to tell them apart. So for an English-speaking audience it can be easier just to write them both as "d" rather than introducing and explaining "ḍ".

If you want to go this route, I'd suggest:

  • ع → [nothing]. /ʕ/ is unlike anything in English, and using a half-ring or "3" will just confuse most native English speakers.
  • غ → "gh". This is fairly standard for the /ɣ/ sound, and as a bonus, the digraph is completely unambiguous in Arabic!
  • ر → "r". Nice and simple.
  • ك → "k". Same sound as in English.
  • ق → "q". Unless it's before a "u", it should be clear that this is the "foreign sound sort of like a k" one, as in "Iraq".
  • ء → [apostrophe]. Glottal stops are better known than /ʕ/, and the apostrophe is common when transcribing them in dialects (as well as in Hawai'ian and other languages).

There are several transcription systems from Arabic into Latin letters. Wikipedia provides a comparison table of several transcription systems in one place. You need to decide yourself which system serves your purpose best.


If you're looking at a spoken dialect, it's worth mentioning that it depends which dialect you're transliterating from. For example, in some Palestinian Arabic, "ق" is pronounced as an English K, but in some Saudi Arabic it can be a G-as in "great"-or Q-sort of a more guttural K, almost like if you tried to say the word "gulf" with a k in front like "kulf." In some Egyptian Arabic, it's pronounced exactly the same as ء (a glottal stop). All the other major dialects subscribe to one of those 4 pronunciations, though the instance of use and quality of the sound differs a lot.

If you're talking about MSA, Quranic Arabic, fusHa, or any other kind of "formal" Arabic, there is definitely a standardized pronunciation of those letters. In that case, I would recommend either using the transliteration table from wikipedia here: https://en.wikipedia.org/wiki/Arabic_alphabet or if you don't want to use special characters I would say: ع - 'a

غ - gh

ر - r

ك - k

ق - q

ء - '

If you're interested in "arabeezy," I would say:

ع - 3

غ - 3'

ر - r

ك - k

ق - q (or 2, g, or k depending on the dialect-but q is always safe)

ء - 2


The surplus of sounds of Arabic, relative to English, means that either some distinctions must be lost, or the standard supply of letters must be augmented. The standard (Arabist professional) conventions are that ع = [ʕ] is written as `, glottal stop (ء) is ´, ذ is dh, and so on. This article gives a convenient list. There is no conventional way to write the emphatics or long vowels without some supplementing of the alphabet, and pressing under-dot (emphatic or ħ) and macron (long vowel) into service has been deemed to be the appropriate augment. The digraph solution to the problem of ش غ ذ خ ث potentially has the problem of not distinguishing a two-consonant sequence with h as the second member (leaving out gh which is unambiguous), and one does find ش rendered as š.

These conventions have existed for over a hundred years, and they won't just change to avoid a diacritic. Theoretically you have some options for rendering these sounds, methods that I've seen exploited: emphatics are with capital letters, long vowels are doubled. There are aesthetic problems with that result, and social ones (nobody does that).

Not the answer you're looking for? Browse other questions tagged or ask your own question.