What 𐞚 means
It marks a voiced lateral-and-median fricative release.
Extended IPA (extIPA) is a system for writing down speech sounds. There, 𐞚 follows a consonant such as d. It signals a noisy release made while the vocal folds vibrate, with air escaping along both the tongue’s side and middle. It is a transcription mark, not separate l and z letters in a word.
The extIPA system began in 1990 for transcribing atypical speech. A revision approved in 2016 appeared in 2018; a 2020 Unicode proposal requested superscript fricative letters for unusual releases of stop consonants—sounds made by briefly blocking airflow—including 𐞚.
Is 𐞚 the same as ʫ?
No: ʫ represents the sound; 𐞚 marks its release on another consonant.
The full-size ʫ represents the voiced lateral–median fricative itself. In the extIPA chart, 𐞚 after d marks that sound as d’s release; its paired modifier 𐞙 after t marks the voiceless counterpart. The small marks modify a consonant rather than replace it.
Why can 𐞚 render as ʫ?
In CLDR 48 character fallback data, 𐞚 maps to full-size ʫ. A system using that compatibility fallback can erase the superscript cue, making the release mark look like a standalone sound symbol.
Technical phonetic notation used in narrow transcription, including clinical work.
Read 𐞚 with the consonant before it; alone, it does not identify a release.
Who writes 𐞚
Loading…
What it means when someone sends it to you
2 readingsA voiced consonant released into lateral-and-median frication.
Don't confuse it with
2Combination ideas
2Questions people ask
1Why does 𐞚 show as a box on some devices?
Because that device's fonts have no glyph for it. The character is still correct and still copies correctly — the box is a missing glyph, not a broken character.
How to type or find 𐞚
- Search it as
- extipa · ipa · superscript lz · lz digraph · lateral median release · voiced lz · 𐞚 vs ʫ
- Shows up in
- Anything that takes text · its shape is whatever your font draws
𐞚\1079ACodes and encoding1 code points
A visible symbol can contain several code points. JS .length counts UTF-16 units; UTF-8 counts bytes.

