What does the tongue do to discriminate sibilant vs. non-sibilant?

Question

My native language is Korean. And including me, probably most Korean people often confuse /θ/ with /s/, and /ð/ with /d/. The problem is, I don't know enough how sibilant fricatives and non-sibilant fricatives are discriminated. I know that they are discriminated by whether the airflow goes between the teeth or not, but such difference are made by the tongue, isn't it? So how different is the tongue?

With enough practice to speak English, I can now speak /θ/ and /ð/ properly, but I still don't get /ɹ̝/, /t͜θ/, etc.

Answer 1

In /s,z/ the sides of the tongue are slightly raised, creating an u-shaped channel or groove, known as sulcus. You probably already do that unconsciously when you say a Korean /s/ (unless you have a lisp). Sulcalization (grooving) concentrates the airflow, making it sound more strident as it hits the top front teeth.

In /θ,ð/ the tongue is comparatively flat, without sulcalization; this makes the airflow spread, resulting in a softer sound. An easy way to do that is to make the tongue tip peek out beyond the teeth. Many native speakers do that (if you pay attention, sometimes you can spot a tongue tip in a singer's /θ/ in a music video). Depending on the dialect and phonetic context, other speakers don't pop out the tongue at all; they can produce /θ,ð/ sounds with a flattish tongue entirely inside, tongue tip behind the top teeth.

Answer 2

The tongue does not discriminate, the ear does (the tongue articulates). The sibilants of English, [s,ʃ], are formed with airflow hitting an obstruction at a sharp angle, and the non-sibilants (f,θ) are formed without such an obstruction. Actually, the term "sibilant" isn't a technical term in linguistics, so it's impossible to say what the "sibilants" versus "non-sibilants" are except that we know that s, ʃ are sibilants. The theoretical linguistic distinction (of dubious empirical validity) is between strident fricatives (s,ʃ but also f) and non-strident (φ,θ). In the production of θ, the tongue impinges on the teeth further forward, so that the air more or less passes straight down the tongue and through the teeth (creating very little turbulence), but in producing s, the tongue is retracted so the air hits the roof of the mouth around the alveolar ridge, then takes a turn down the teeth, which creates a lot of turbulence.