The famous sounds are made in the throat and the expensive ones are not. Ayn, ha, kha, ghayn and qaf are genuinely new and they get all of the attention, because they are hard and because everybody warns you about them. The emphatic consonants are the pair of every ordinary consonant, they are produced with the back of the tongue raised, and they are the ones that decide whether a sentence sounds Arabic.
The useful trick with the emphatics is that you should not aim at the consonant at all. Aim at the vowel beside it, because that is where the difference is actually audible.
Six sounds, four of which are new
- ayn has no English equivalent. It is made by constricting the throat and voicing, and it is a full consonant that appears in extremely common words including the word for Arabic itself.
- qaf is a k made far back, further back than any English k, and it separates real word pairs from the ordinary k.
- ha is a strong h from the throat, distinct from the ordinary h, and the pair is another set of minimal contrasts.
- kha is the ch in Bach, which is the one new sound English speakers already have from other languages.
- the emphatic set: sad, dad, ta, za. Four consonants made with the tongue flattened and the throat narrowed, each with a plain twin, and they colour the vowels around them.
- three long vowels and three short. Only six vowel sounds in the whole language, which is far fewer than English, and length is contrastive.
An emphatic is a shadow over its neighbours
Each emphatic has a plain twin: s and sad, t and ta, d and dad, dh and za. Told that, a learner tries to make a heavier s, which is not the distinction and produces nothing. What actually happens is that the back of the tongue rises, and the vowels on both sides of the consonant change colour as a result, so an a next to an emphatic is a darker, further back a than the same letter elsewhere.
That is why the training target is the syllable rather than the letter. Practise the vowel, hear the difference between the two versions of it, and the consonant arrives as a consequence. A learner working on the consonant alone can produce it in isolation and lose it inside a word, because the syllable around it never changed.
The throat sounds are a different problem and a simpler one. Ayn and ha are made by constricting the throat itself, kha and ghayn further forward, and qaf right at the back of the mouth. None of them has an English neighbour close enough to substitute, so the usual substitutions produce either a different word or no word. They are also easy to check: a recording tells you immediately, and there is no ambiguity about whether you got it.
Sounds learned in isolation and lost inside a word
Here is the stall and it comes from the sounds being difficult. Because ayn and the emphatics take work, learners practise them as isolated noises, get them, and then find they cannot produce them inside a word at conversational speed. Everything is correct in a drill and nothing survives in a sentence.
The reason is that these are not segments you insert, they are properties of a syllable. An emphatic changes the vowels around it and a throat sound changes the timing of what follows. Practising the noise trains a noise. The exit is to practise the smallest real unit instead: a whole word, then a two word phrase, then the same phrase at speed, and never the letter on its own after the first week. Getting somebody to say honestly whether it landed is what a paid hour is for, in what a teacher is worth, and what you can do alone.
Four minutes, before any drilling
Which half of the problem you are actually in
Find a minimal pair in a recording of somebody else and answer honestly, before spending a month on the wrong half.
- No, or only sometimesThe ear has not split themYour mouth is producing an accurate copy of a target that is not there yet
- Minimal pairs, listening only
- The vowel beside the consonant, not the consonant
- Stop producing it aloud until the split holds
- Yes, every timeThe mouth has not caught upThe category exists and the muscle is behind it
- Whole words, never isolated letters
- A spectrogram, because the difference is visible
- Two word phrases, then the same phrase at speed
Most learners assume the right hand branch because it is the flattering one, and the emphatics in particular almost always turn out to be on the left. A sound that is wrong every single time is nearly always a category that has not formed.
Two references made of sound rather than description
Recitation recordings are the phonetics reference nobody mentions. Quranic recitation is the one register of Arabic where every sound is fully articulated and held, so the throat sounds and the emphatics can be heard without being rushed past, and there is an enormous free archive of it. It replaced the description in a textbook, which cannot convey a sound. The honest caveat matters: this is a formal recited register with its own rules, so it is a reference for what a consonant sounds like and not a model for how anybody talks.
Praat, the free acoustic program from the University of Amsterdam, is the second and it is unusually useful here. An emphatic and its plain twin look different on a spectrogram, because the vowel formants move, so a distinction you cannot yet hear can be seen in five seconds. It replaced judging your own production by ear, which cannot work while the distinction is the thing you are missing. Free and open source.
The mistake: substituting the nearest English sound and being understood
An English speaker produces a k for qaf, an h for ha, and simply omits ayn, and is understood a good deal of the time, because context repairs it. That is what makes the substitutions permanent: the feedback that would correct them never arrives, and each one is a real word turning into a different real word.
The pairs are worth knowing rather than the general warning. Qaf against kaf, ha against ha with a dot, and the four emphatics against their plain twins are the contrasts that separate actual words, and they are a short list. Work on those as minimal pairs, listening before producing, since a distinction the ear has not split cannot be copied by the mouth. That ordering is the same one that applies to any sound your first language does not have, and it is argued out in the stalls every learner hits, and the way through each.
What the sounds are attached to
The consonants you are learning are the ones that carry the root, so the sound system and the vocabulary system are the same system approached from two sides, which is learning by root instead of by word. Several of them are written with letters that differ only by a dot, so mishearing one and misreading one are the same error, in right to left, joined letters, and vowels you infer. The vowels the emphatics colour are the ones the pattern supplies rather than the page, in three letter roots, and patterns that generate the vocabulary. And several of these consonants are pronounced differently, or not at all, in the language people actually speak, which is which to learn first, and what each is used for.