The case for reading while watching. Why the global film-literate audience prefers subtitles, why most dubbing is bad, and where the exception lies.
Bong Joon-ho's 2020 Golden Globes acceptance speech included the line that has, more than any other, framed the contemporary discussion about foreign-language films in English-speaking territories: 'Once you overcome the one-inch-tall barrier of subtitles, you will be introduced to so many more amazing films.' The line was delivered in Korean; the audience read the translation.
This essay tries to lay out the case for subtitles over dubbing, where each works, and what the streaming era has changed about the discussion.
Subtitles preserve the original audio performance. The actor's voice, breathing, accent, vocal inflection, and rhythm are all part of what the audience receives. The argument for subtitles is, ultimately, that watching a film is also listening to it, and that hearing the work the original actors did in their native language is, in most cases, preferable to hearing a different performance by different actors recorded in a different room months after the production.
The argument is at its strongest with dialogue-heavy work in which the original performance is doing significant character-establishing work. Parasite's class register depends substantially on the differences in how the wealthy and working-class characters speak Korean — different vocabulary, different vowel quality, different rhythmic patterns. Almost none of this survives in dubbing. The English voice cast on the Netflix dub is competent; the linguistic class-coding is, structurally, gone."],
Dubbing works well in certain specific contexts. Animation is the clearest case. The original audio in animated films is, by definition, also a voice-acting performance recorded in studio; replacing one voice-acting performance with another in a different language does not erode 'the' performance in the way replacing a live-action actor's voice does. Studio Ghibli's English dubs (often produced under Disney's licensing, sometimes featuring significant American actors) are widely considered acceptable substitutes for the Japanese originals. Many international audiences first encountered Ghibli through these dubs.
The other case where dubbing works is the European auteurist tradition of post-synchronisation. Almost every Italian film from the 1950s through the 1980s — including Fellini, Antonioni, Pasolini, Bertolucci, and many others — had its dialogue recorded in post-production rather than on set. The original Italian audio is itself, structurally, a dubbed performance recorded by the actors themselves in studio. In this context, the distinction between 'original' and 'dubbed' is more fluid than in conventional sync-sound production.
The standard live-action dub, in which different actors re-record dialogue in a different language for a film originally shot in the actors' own language, typically loses several specific things. First, lip sync: the dubbed dialogue is constrained to match approximate mouth movements rather than to deliver the same lines, which forces translators to rewrite dialogue in ways that change meaning. Second, vocal performance: the dubbed actor cannot replicate the original actor's specific breathing, pauses, and emotional register; the dub typically flattens these qualities. Third, accent and class markers: linguistic features that signal social position in the original language are usually translated into a generic register in the dub.
What's notable is that the audience often does not consciously notice what is missing — particularly if they have never seen the original. The dub feels normal because it is what they are watching. The structural loss is real, but it does not produce a felt absence the way a missing dialogue line would. This is part of why dubbing remains commercially viable in many territories: the audience is not aware that they are receiving a structurally diminished version of the work.
Netflix has, since the global success of Squid Game in 2021, invested heavily in both subtitle and dub production for its international originals. Netflix's user data reveals an interesting pattern: a meaningful minority of viewers in English-speaking countries watch Korean-language content with subtitles by default. The 'one-inch barrier' has, for the post-2020 streaming audience, become significantly lower than it was for previous generations.
Younger audiences (under 30) are particularly comfortable with subtitle-reading. The cultural assumption that subtitles are a barrier — held throughout the 1990s and 2000s by older American audiences — has weakened across the past five years. The shift is structural: subtitle-reading has become a normal part of the streaming experience.
The Criterion Channel (see our Criterion essay) has the most-curated foreign-language catalogue in English-speaking streaming services. MUBI is the other significant subtitled-cinema service. Netflix has, on a per-title basis, the largest international library but the most-uneven subtitle quality. Amazon Prime's international catalogue is significant but harder to navigate. The Criterion physical Blu-ray catalogue (or its DVD equivalent for older titles) remains the gold standard for subtitle quality.
For a working starting point, see our list of best foreign-language films. Almost every entry on that list is a film whose subtitled experience is significantly superior to any available dub.