Some countries watch foreign films dubbed almost universally, while others read subtitles as a matter of course. The division was set nearly a century ago and has proved remarkably stable.
Sound created the problem
Silent films crossed borders easily, since intertitles could be swapped for a translated set at minimal cost. Recorded dialogue removed that flexibility overnight.
Studios first tried shooting the same film several times with different casts, which was expensive and produced inconsistent results. Neither dubbing nor subtitling was yet established.
Once dubbing became technically viable, each market had to decide which route to take, and the decision was made under very different conditions in each country.
Market size drove the economics
Dubbing carries a fixed cost per film covering studio time, translation, casting and performance. Subtitling costs a fraction of that.
In a large language market the dubbing cost spreads across a big audience and becomes negligible per viewer. In a small one it can exceed what the release will earn.
This is the clearest single predictor. Large-population languages generally dub, smaller ones generally subtitle, with the exceptions explained by history rather than economics.
Language policy played a part
Several governments in the early sound era restricted the exhibition of films in foreign languages, which made dubbing effectively compulsory for any imported title.
Those rules were bound up with national language promotion, and in some cases with censorship, since a dubbing script is far easier to alter than a subtitle sitting beneath the original audio.
The regulations lapsed long ago, but by then the infrastructure existed and audiences had formed expectations that outlived the policy entirely.
Habit becomes preference
Reading subtitles is a learned skill, and audiences raised on them process the text without conscious effort. Audiences unaccustomed to them report the experience as tiring.
Dubbing audiences develop attachments of their own, with particular voice actors permanently associated with particular performers. Changing that voice provokes genuine objection.
Each system therefore produces the preference that sustains it, which is why surveys in dubbing and subtitling countries return firmly opposed results.
Streaming offers both and changes little
Platforms now ship multiple audio and subtitle tracks with a title, making the choice a menu setting rather than a national decision.
Usage data shows people largely retaining inherited habits, though subtitle use has risen among younger viewers everywhere, partly for reasons of clarity and viewing environment.
The lasting change is in supply rather than demand. A dubbing industry that once served a national release now produces tracks for a global catalogue, which is far more work.