Speech

The best diagnostic instrument I have costs nothing and works before I even open my mouth to ask anything.

I listen to how the patient speaks in the first minutes, while they are still telling me why they came.

The reason is simple, but worth stating explicitly: speech runs on automated motor programs. When I ask someone to swallow, or to put their tongue "where it belongs", I get their best version, performed with attention. When they speak freely about something else, I get the real version — the one they execute thousands of times a day without thinking.

The diagnosis is already in front of me. I just have to know what I am listening for.

What I actually listen for

I do not listen for whether it sounds correct. I listen for how it is executed.

The sounds that expose the tongue most clearly are D, T, N, S and L. I sometimes ask patients to count slowly to ten — almost all of them appear there.

What I watch for:

• does the tongue show between the teeth on S or T?
• does the tip of the tongue reach up, to the area behind the upper front teeth, or does it rest low, on the lower teeth?
• does the air leave through the centre on S, or does it escape sideways?
• does the jaw move to compensate for something the tongue should be doing?

A patient can sound perfectly normal and, at the same time, execute incorrectly. A listener's ear does not catch the difference. The dentition catches it — because it receives, year after year, pushes in a direction where no pushes should exist.

Why it matters after treatment

This is the part that interests me most as an orthodontist.

When I move teeth, I build a new architecture. What does not change along with it is the motor program of speech. It was learned in childhood, against the old architecture, and it keeps running just the same.

If the tongue keeps pressing where it pressed before, that force now works against the result. A retainer holds the teeth exactly as long as it is worn. The motor program works twenty-four hours out of twenty-four.

That is why, in certain patients, speech re-education is part of the orthodontic treatment — not as a cosmetic correction of the sound, but as a correction of the mechanics.

And here I have to be honest about a practical difficulty: classical speech therapy is mainly concerned with how the sound sounds from the outside. What I need is someone who looks at the execution — where the tongue lands, with what force, on what surface. Such specialists exist, but they are not the majority. Finding the right referral partner is, for an orthodontist working in this framework, part of the practice's infrastructure.

My hypothesis: spoken languages and the maps of malocclusion

Here I step onto ground I want to mark very clearly before setting foot on it: what follows is a hypothesis of mine. It is not textbook fact and it has not yet been proven.

The observation I started from is that orthodontic anomalies are not evenly distributed across the globe. Some populations have much higher proportions of Class II; others, of Class III. The usual explanations are genetics and diet.

But they do not explain everything. Immigrant populations do not keep the pattern of their country of origin — within a generation or two, they converge towards the pattern of the country they arrived in. Genes do not change that fast. Something else does.

My proposal is that one of the unstudied variables is the spoken language.

The reasoning: every language distributes tongue contacts differently. Some languages demand contacts predominantly high — behind the upper front teeth. Others demand contacts predominantly low. A child produces tens of thousands of such contacts every day, throughout the entire growth window. It is exactly the type of repetitive, low-intensity loading already known to shape bone.

So far I have classified fifteen language groups. Every one fits the predicted direction. I have not yet found a language–population pair that contradicts the hypothesis.

That does not mean the hypothesis is true. Language, diet, genetics, climate and socio-economic conditions travel together across populations and are very hard to separate. The study that could decide has not been done yet. I have described in detail, in a professional article, what that study should look like and what result would prove me wrong.

Until then, it remains a well-argued hypothesis. Nothing more.

What stays valid even if the hypothesis falls

It is worth saying, because it is the practical part: the clinical observations the hypothesis rests on remain useful whatever its fate.

Speech remains a real-time diagnostic register for tongue position. That is true regardless.

Patients who grew up in one language and live their adult lives in another remain a group with a particular vulnerability to relapse — two competing contact programs, one learned in childhood and one layered on top in adulthood. They deserve a different retention strategy, especially in the lower front region, where the smallest muscular changes show up first.

And referral to a mechanics-aware specialist remains necessary for the patients in whom execution works against the result.

The hypothesis is the framework that ties these observations together. Even if that framework proves partly wrong, the observations it organises remain standing.

What you can notice yourself

This is not self-diagnosis — but there are things you can notice:

• does your tongue show between your teeth when you say words with S or T?
• does your S whistle, or does the air escape sideways?
• does your child speak "with the tongue between the teeth", and were you told it would pass on its own?
• did you grow up in one language and, as an adult, speak another every day?

The last question seems unrelated to orthodontics. In the framework I work in, it is not.