Some of what we publish is video as well as text — short Reflections you can watch, and Vignettes told as illustrated scenes. This page explains how captions work on them, why there are two different kinds of caption and what each is for, and what the player does before you press play.
What you can watch
Reflections. A short, tall video built around the artwork for that day's reflection, with the text read aloud over it. These are the same videos we publish to YouTube and Instagram, which matters for the captions — see below.
Vignettes. Illustrated scenes with narration and dialogue, told as a sequence of panels.
Recordings and talks. Where a talk was filmed rather than only recorded, the video sits with its write-up.
Note
Not every reflection has a video, and nothing is hidden if it does not. Where there is no video, there is simply no player.
Captions: there are two kinds, and they are not the same thing
This is the part worth explaining properly, because on the surface it looks like we are doing the same job twice.
Captions printed into the picture. The words in a Reflection video are drawn onto the video image itself as it is made. They are part of the picture, in the same way the artwork is. This is deliberate and it is not going away: these videos are also published to YouTube, Instagram and elsewhere, and when a video travels to another platform a separate caption file does not travel with it. Printed-in words are the only captions that survive the trip.
Captions as a separate file. Alongside the video we now also publish the same words as text, with the timings that say when each line is spoken. Your browser reads that file and draws the captions itself.
They look similar on screen. They are not remotely equivalent.
| Printed into the picture | Separate caption file | |
|---|---|---|
| Can a screen reader read them | No — they are shapes, not words | Yes |
| Can you make them larger | No | Yes |
| Can you change the colour or add a background | No | Yes |
| Can you turn them off | No | Yes |
| Can they be shown in another language | No | Yes — sixteen languages |
| Do they survive being shared to YouTube | Yes | No |
The last row is why we keep both.
Why the separate file matters
Because to software, words printed into a picture are not words. They are an arrangement of light and dark. A screen reader has nothing to announce, a browser has nothing to enlarge, and a search has nothing to match. Everything a caption is for — beyond simply being visible to someone already looking at the screen — depends on the words existing as text somewhere.
Your own caption settings start working. Every operating system and browser lets you decide how captions should look: bigger text, a solid background behind them, higher contrast, a different typeface. Those settings had nothing to act on before, because there was nothing on our videos for them to act on. They work now. For anyone who relies on captions rather than merely finding them handy, this is the change that matters most — it is the difference between captions that happen to be legible to you and captions you can make legible.
And it opens the door to things we cannot do with printed-in words. The first of those is now built: captions in sixteen languages, described in the next section. A readable transcript, and being able to search for a phrase and find the moment it was said, are not built yet. None of them would be possible at all without the words existing as text.
Note
With captions switched on you will see both sets at once: the browser draws its captions at the top of the video, and the printed-in ones stay at the bottom. They are deliberately kept apart rather than stacked on each other. The ones at the top are yours — those are the ones that follow your own size, colour and contrast settings.
Subtitles in sixteen languages
Every Reflection and Vignette made from August 2026 onward carries its captions in sixteen languages: English, Arabic, Farsi, Hebrew, Urdu, French, German, Spanish, Italian, Russian, Indonesian, Swahili, Chinese (Simplified), Chinese (Traditional), Japanese and Greek.
Chinese comes in both scripts, and they are listed separately on purpose. A reader who uses Traditional characters can often work out Simplified, and the other way round — but not reliably, and not comfortably at the speed subtitles move. 繁體中文 serves Taiwan, Hong Kong, Macau and a good part of the diaspora, and it is not a nicety.
Where a video has more than one, a language menu appears next to the Captions button. Pick a language and it is remembered — every video, every visit — in the same way the on/off choice is.
The spoken audio stays in English. What changes is the text on screen. If you are reading along in a language other than English, you will see the English words printed into the picture at the bottom and your chosen language at the top.
Choosing a language costs you nothing. Only the one you have selected is ever downloaded, so a video offering sixteen languages is exactly as light to load as one offering a single language — and if you never turn captions on, none of them is fetched at all.
Important
The translations are done by machine, and we would rather say so than let you assume otherwise.
Every language other than English is produced automatically. Nobody has read them. For ordinary description they are good; for anything turning on a precise word — a line of scripture, a theological term, a name with a settled form in your language — the English is the text to rely on, and a translated line marked with an asterisk is telling you exactly that.
We took the same view here as we did with pronunciation elsewhere on the site: an unreviewed rendering warns you, it never blocks you. Hiding a language until someone has checked it would mean most people get nothing, for a long time, in the name of accuracy. Offering it with an honest label means you can read it today and judge it for yourself.
Each translated track closes with a short line saying the subtitling was done by our own tools. That is there for the same reason.
A language we have missed? Tell us which one — the sixteen above are a starting point, not a boundary, and someone asking is the most useful signal we have about which to add next. Traditional Chinese was added exactly this way.
The Captions button
Under any video that has a caption file you will find a Captions button that says whether they are on or off.
You may reasonably wonder why we added one when browsers already have a captions control. They do — and at the size these videos are shown, they hide it. A tall Reflection video is displayed in a fairly narrow column, and at that width a browser folds most of its controls away behind a small unlabelled menu. Captions that exist but cannot be found are not much better than no captions.
Your choice is remembered. Turn captions on once and they stay on — every video, every visit — until you turn them off. Needing captions is something about you, not about one particular video, and asking you to switch them on again each time would be a poor way to treat that.
Important
Captions — in any language — are only available on videos made after August 2026, and this is a genuine limitation rather than a rollout.
The timings that record when each word is spoken are worked out while a video is being produced, and older videos did not keep them afterwards. There is nothing left to build a caption file from, in English or in any other language — so those videos would have to be remade from scratch to gain one. Where there is no caption file, the button does not appear.
What the player does before you press play
Nothing. A video is not downloaded until you actually start it.
This sounds obvious and it was not previously true. A page carrying a video used to begin fetching it as soon as the page loaded, whether or not you ever watched — and on a phone that could mean several megabytes spent on something you scrolled straight past. The still image you see before pressing play is a separate, small picture, so the video itself is never needed to show you what is there.
The caption file behaves the same way: it is not downloaded unless you switch captions on. Nobody pays for a feature they are not using.
Quality adjusts to your connection
A video used to be one fixed quality, take it or leave it. If your connection could not keep up with it, the only thing the player could do was stop and buffer.
Each video is now prepared at three quality levels, and your browser moves between them as it plays — starting carefully, then stepping up once it can see there is room. On a good connection you get the same quality you always did. On a poor one the video keeps playing at a lower quality instead of stopping altogether.
There is nothing to choose and nothing to set. The player works it out from how quickly the video is arriving, and it can change its mind partway through if your connection does.
Note
The picture stays exactly the same size and shape throughout. Only the amount of detail being sent changes, and on a phone-sized screen the difference is usually hard to see — which is rather the point.
iPhones and iPads need nothing extra for this, because they understand this kind of video natively. Other browsers need a small piece of code from us, and it is downloaded only when you press play — never for someone who scrolls past.
Important
Older videos are being converted to this gradually, in the background. Until one has been converted it plays exactly as it always has. There is no point at which a video is half-converted or unavailable — it is either the new way or the old way, and both work.
Downloading a video
Where a download link appears under a video, it gives you the video file itself, with the captions printed into the picture. That is the version to share, precisely because those captions travel with it.
A note on the narration
Reflection videos are narrated, and the same care applies as anywhere else on the site: where a passage is quoted in Arabic, Farsi or Hebrew, it is written out phonetically in English letters and the narration reads that. It can be wrong, and the written text is the one to trust. Listening to the Site explains this in full, and it applies to the captions on a video just as it does to a narrated article.