Venetian (the Italo-Romance language spoken in the area of Italy roughly corresponding to the Veneto region) has a weird sound which is usually called l evanescente (evanescent l). It varies geographically (from a completely normal /l/ to mute), but its most distinctive form has been described by Lepschy in Fonematica veneziana (pg 18-19) as (my translation)

an articulation where the air passes through a depression in the central part of the top of the tongue, lifted towards the palate, while the two sides of the top of the tongue are in contact with the sides of the crown of the upper teeth

The pronunciation of toła by ness1 in Forvo provides a fairly characteristic realization.

Ferguson, in his A Linguistic History of Venice claims it has no IPA representation, so I assume it is a rare sound. He also reports it has been described as «an unrounded prevelar semi-vowel».

Q: Is there any other language containing this sound?

  • This is my first question here, so I hope it is on topic and answerable. Feb 26, 2019 at 17:31
  • 1
    @jknappen What was the reason for removing the tag italic? A user subscribing to or searching by that tag would most likely be interested in this question. Feb 27, 2019 at 21:21
  • 1
    And indeed the accepted answer is mostly about Italic. Feb 27, 2019 at 21:22
  • 2
    @AdamBittlingmayer Actually, the most specific tag is romance-languages, given that Italic is a wide family covering all pre-roman languages spoken in Italy and their descendants (with the exception of Celtic, which is a separate branch). But both the question and the answer seem to me more about synchronic phonetics rather than historical linguistics. Anyway I'm not an experienced user, so I'll let the rest of the community to sort it out. Feb 28, 2019 at 6:56
  • 2
    @AdamBittlingmayer I removed it, because Italic refers to the sister languages of Latin like Faliscan, Umbrian, and Oscian. Modern Venetian is clearly Romance and specifically Italo-Romance. Do not confuse it with the ancient Venetian that is conventionally assumed to represent a separate branch of Indogermanic. Feb 28, 2019 at 11:20

1 Answer 1


The Wiki page on Venetian proposes that it may be phonetically [ɰ], also Ø, but that does not match the Forvo sample which is closer to [j]. In lieu of a solid phonetic study, it's not really possible to assert what this sound "is", so comparing the properties of that sound to ones in other languages is not really possible. You can compare two pronunciations of łéngua, one with no initial C and the other with initial [l], likewise łuna and ciàcołe. We would need to have examples from the same individuals with [l,j,Ø] in comparable environments as we find "ł", to see if this is an individually-distinct sound, as opposed to a sound that may have formerly been phonetically distinct. This grammar (in Venetian) might be useful.

Addendum: based on the minimal pair provided in the comment, from the same speaker, if this pattern is sustained (not a quirk of tokens), this seems to be a somewhat problematic sound. The measured F1,F2 in the area corresponding to the glides are impałà = 614, 1535 and impajà = 416, 1970. This closely matches the F2-F1 difference of [ɯ]; but F1 is more like [ɛ] and F2 is more like [ɶ]: subjectively, the 75 msc in the middle of "ł" is like [ɛ]. One problem with the letter "ɰ" is that it covers a very wide range of sounds. Ladefoged provides an IPA reference performance, which does not sound like this "ł". The Wiki article on the voiced velar approximant doesn't give any clear examples (i.e. there are numerous disclaimers like "some speakers", "certain dialects", "optional"). IMO this is a novel sound, deserving of an in-depth phonetic investigation.

  • Thanks for your answer! I had seen that wikipedia suggestion, but as a native speaker of Venetian, the wikipedia sample for [ɰ] doesn't match my perception of the sound. I'd say that the tongue in ł is more towards the front of the palate than in [ɰ]. Regarding [j], I think there are minimal couples distinguishing it from ł, I'll try to hunt them down. That said, there are certainly regional variations and some speakers pronounce ł as [l] (the "conservative" pronunciation) and some instead leave it mute. The sound I'm interested in is somewhat an intermediate stage between the two. Feb 26, 2019 at 19:54
  • Oh, and if it helps, the ł sound usually exists only in intervocalic positions between back or center vowels (or maybe in word-initial position, but not if the previous word begins with a consonant), so for example ciacołe is not a good example (I at least perceive the two pronunciations on forvo as /tʃakole/ and /tʃakoe/ respectively) Feb 26, 2019 at 19:57
  • 4
    Sorry, last comment and I'll stop, but I found the minimal pairs: impałà and impajà (with the same speaker, even!) Feb 26, 2019 at 20:05

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.