The consonant inventory

Lesson opening

This is the first part of the lesson. The rest, the exercises and the reviews are in the app.

The consonants as a system

You know the individual letters already: you have read the alphabet, tamed the special č ć š ž đ dž, learned to keep the pairs apart. Now let's look at the consonants not one by one but as a system — see how they line up in voiced/voiceless pairs, how they are distributed across the mouth, and, above all, how this arrangement differs from what you may expect. It is the view "from above" that explains both the commonest foreign mistakes in pronunciation and why Serbian has exactly 25 consonants — not thirty-five, as its big cousin Russian does.

The good news first: nearly the whole set is familiar. Most Serbian consonants are sounds English has too, and they come to you free of charge. What needs work is understanding how the system is built — and then both the alternations waiting ahead and the un-learning of old habits fall into place.

What comes for free

Seventeen Serbian consonants sound just like English ones — you recognize them with no training at all. Here they are, by familiar sounds:

That leaves eight: č and dž (the hard "ch" and "j" — nearly English too), the genuinely new ć, đ, lj, nj, j, and the special Serbian h. We covered how each of them sounds in the lesson "The special letters č ć š ž đ dž"; here the task is not to repeat their shapes but to place these sounds into the system.

Voiced and voiceless: seven pairs

The heart of the system is the voiced/voiceless pairs. In each pair the two sounds are made the same way and in the same place, differing only in the vocal cords: for the voiced one the cords vibrate, for the voiceless one they are silent. Seven such pairs:

Three consonants have no pair: c (ts), h and f — voiceless loners with no voiced partner. And a whole row of voiced sounds manages without a voiceless mate — the sonorants m n nj l lj r j v: they always sound "with voice" and never join the pairing game. This picture is not decoration: future voicing assimilation runs exactly along these pairs, when a voiced and a voiceless sound meet at a morpheme seam and adjust to each other. That gets its own lesson, "Voicing assimilation".

Where in the mouth a sound is born

The second axis is the place of articulation: at which point of the vocal tract the air meets its obstruction. No need to memorize it, but the general map is useful — it explains why certain sounds "keep company" and slide into one another easily. Roughly, the consonants split into four zones:

Cheat sheet

The cheat sheet — everything essential from the lesson on one page. In the app it opens at any time.

The consonants as a system

25 consonants (30 letters minus the 5 vowels a e i o u). Three axes: voicing, place and manner of articulation. A fourth, hard/soft axis does not exist: a Serbian consonant has one value. A consonant is suglasnik, a vowel — samoglasnik.

Seventeen come free

The rest: hard č/dž, the new ć, đ, lj, nj, j and the light h.

Seven voiced/voiceless pairs

Unpaired: the voiceless loners c, h, f and the sonorants m n nj l lj r j v (always voiced, never join the pairs). Voicing assimilation at morpheme seams runs along these pairs.

Place of articulation

Everything new for an English speaker sits in the palatal zone; the velars k g h "slide" there under palatalization: ruka → ruč-, bog → bož-.

Manner of articulation

Place plus manner give a sound its exact address: c, č, ć are all affricates, but c is dental while č and ć are palatal. In alternations a sound moves "a short trip": k → č.

Consonants do not soften

Softness distinguishes no words, and there are no "soft twin" consonants — only the inherently palatal ć, đ, lj, nj, j. That is why there are 25, not 35.

A voiced ending is never devoiced

grad = "grad" (not "grat"), Beograd, led = "led" (not "let"), nož = "nozh", zub = "zub". Serbian assimilation works inside the word, at morpheme seams, and shows in the spelling: težak → teška.