Dental consonants, which involve the corona of the tongue contacting the teeth (typically the upper teeth) are known to be rare throughout the world’s languages. More specifically, phonemic distinction between dental and alveolar consonants of the same manner of articulation is extremely uncommon. English is very lucky to have two consistently dental phonemes (/ð/, which is my favorite sound in the language, and /θ/) and a few others that can be realized dentally (especially /l/, /s/, and /z/). The languages that do have a dental element to their coronal plosives are usually considered to have denti-alveolar consonants rather than truly dental ones.

I have seen this rarity explained in thee ways: first, that dental sounds are hard to produce; second, that they are hard to distinguish from alveolar consonants (which doesn’t really explain why one place would be preferred over the other); and third, that dental fricatives are acoustically faint in fast speech.

However, I feel that there is another possible explanation for this: the teething of babies. A key commonality cross-linguistically (I’ll refrain from calling it “linguistic universal”) and a basic necessity for language acquisition is that young children can easily learn every given language. Thus, the first words they learn have sounds that are easy for them, like open vowels and nasal or bilabial consonants. Mama- and papa-type words become part of a child’s repertoire extremely early, with other vowels and coronal consonants coming later. Babies don’t have teeth for some months, and they grow in gradually. Also, the teeth they grow will eventually be replaced by different ones.

Does this have an effect on the rarity of dental sounds? Babies can’t learn them if they don’t have teeth (other than linguolabial or pre-alveolar variations), and they might not have the initiative to when the teeth are new.

  • 6
    note that it's pretty common for children to still be unable to properly articulate certain consonants (e.g. laterals, or some consonants with coarticulations) for several years - long after their teeth come in. If those consonants can exist and be reasonably common I don't think babies' lack of teeth could be significant
    – Tristan
    Commented Jan 17, 2023 at 9:49
  • /s/ and /z/ "dentalized" are /θ/ and /ð/, respectively. I think you meant /t/ and /d/?
    – No Name
    Commented Jan 18, 2023 at 10:55
  • 2
    @NoName Nope, [θ] and [ð] are from a different class of sounds than [s] and [z]. The latter two are sibilants while the former are not. You can have a sibilant produced with your tongue tip at either your top or bottom teeth ([s̪] and [z̪]) as well as a non-sibilant produced with your tongue at the alveolar ridge ([θ̠] and [ð̠], found in Icelandic). The /s/ and /z/ phonemes actually have quite a bit of variation in English with regards to alveolar vs dental, apical vs laminal, tongue tip up vs down, and other little details. /t/ and /d/ can also be dental.
    – Graham H.
    Commented Jan 18, 2023 at 12:43

1 Answer 1


There is no evidence that dentals are rare per se – they exist in many languages, for example many Indic languages, Finnish, French and other Romance languages. What is rare is a contrast in front linguals – dentals vs. alveolars, and primarily among stops. In this article by Jongman & Blumstein, it is found that there isn't a clear acoustic signature of the difference that reliably works across languages. Within Malayalam, which is one of the few languages with a robust contrast, the distinction is signaled by a lower release burst amplitude on dentals, compared to alveolars. Amplitude differences are not a robust cue for signalling short-term segmental differences. On the other hand, in Mapudungun, the difference is signaled by lower F2 on the vowel near dentals. But in Wubuy, the difference is found in F1 and F3, and not F2. In other words, it is a difficult distinction to perceive, which means that it is hard to create such a contrast (turning a single front lingual into two), and hard to maintain it once you create it.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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