To make a philosophy explainer video, reconstruct the argument before you write a word of script: state the conclusion, list the premises in numbered form, and decide which premise the strongest objection attacks. Then write one scene per premise, one for the conclusion, one for the objection and one for the reply, roughly 450 to 750 words for a three to five minute video. Draw the argument, not the philosopher: a premise as a box, an inference as an arrow, an objection as a line striking one box. Paste that script into a document-first tool like openCanviz, keep your wording, and check that every premise on screen says exactly what the text says. In philosophy a paraphrase that drifts is a different argument.

Why most philosophy videos explain nothing

Search for any philosopher on YouTube and the first ten results share a format: a painted portrait, a slow zoom, a calm voice listing "key ideas". You come away knowing that Kant cared about duty and Nietzsche disliked Christianity. You could not defend or attack either one in a seminar.

That is because the video describes the philosopher's views instead of showing why anyone should accept them. Philosophy is assessed on the why. A tutor marking your presentation wants to see three things: that you can state the argument fairly, that you know where it is weakest, and that you have a view about whether the weakness is fatal.

A video built around that structure is shorter, easier to draw and much harder to fake. It also happens to suit drawn video well, because an argument has a shape. Premises sit side by side, the conclusion sits below them, an objection arrives from the side. A viewer can see the shape while the narration fills it in.

Reconstruct the argument first

Before any script, write the argument in standard form. This is the step that does the philosophy. Everything after it is production.

Find the conclusion. One sentence, the thing the author wants you to believe at the end. If you cannot find one, see the section below on texts without a tidy argument.

List the premises the conclusion depends on. Usually two to four. Include the unstated ones: the assumption the author needs but never writes down is very often where the argument breaks.

Check the inference. If every premise were true, would the conclusion have to follow? If not, a premise is missing.

Name the premise the best objection targets. Every serious objection denies a specific premise or the step between them. "I just disagree" is not an objection.

Worked example: Singer's drowning child

Peter Singer's 1972 paper "Famine, Affluence, and Morality", written during the famine and refugee crisis in East Bengal, is set on almost every first-year ethics course, and it reconstructs cleanly.

Notice what the table does. The pond is not the argument, it is evidence for Premise 2: Singer uses it to show you already accept the principle. A weak video spends two minutes on the pond and ten seconds on the principle. A good one makes clear that the pond is doing a specific job.

Here is how the narration for the Premise 2 scene might read, in full:

"Singer's second premise is the one doing the work. If you can prevent something bad, and doing so costs you nothing of comparable moral importance, you ought to do it. Notice how modest that sounds. It does not ask you to give until it hurts. It asks you to give up only what matters less than what you prevent. Singer's bet is that you already believe this, and the pond is how he proves it."

That is 75 words, about 30 seconds at 150 words a minute, and it is one scene.

Notice also that Singer offers a weaker version of Premise 2 (we ought to prevent something very bad from happening if we can do so without sacrificing anything morally significant) and says even that one is enough to demand far more giving than most people do. If your seminar is on the paper, that weaker version deserves its own scene.

Texts without a tidy argument

Singer is easy because he wants to be reconstructed. A lot of seminar reading does not. Félix Guattari's Chaosmosis (1992, translated into English in 1995) is the classic case: it is a book about how subjectivity gets produced, and it does not march from premises to a conclusion. Students asked to present it often freeze, because there is nothing to number.

The method changes from reconstruction to mapping. Instead of premises, pick the handful of terms the chapter leans on and show how they relate. In Chaosmosis Guattari organises his model around four domains: material and energetic Flows, machinic Phyla, existential Territories and incorporeal Universes. A video for a seminar can draw those four as regions, then spend one scene on each, with one concrete example per region taken from the text, and end with the question your seminar is actually discussing.

Three rules make this work for difficult texts.

Quote, do not paraphrase, the key sentence. Put one short quotation on screen per scene and attribute the page. A paraphrase of Guattari is very easy to get confidently wrong.

Say plainly what you are unsure of. "I read this as saying..." is honest and is exactly what a seminar wants to argue about.

Keep your own view for the last scene. The map is the text. Your reading of it comes after, clearly labelled as yours.

There is a longer treatment of turning concepts with no obvious picture into drawings in how to illustrate abstract ideas, which uses Deleuze and Guattari's tree and rhizome as one of its examples.

What to draw for each part of an argument

Whiteboard style suits arguments because the diagram builds up as you reason, and the finished frame is the whole argument at once. Painted style suits a history-of-ideas video where the setting matters, such as the Enlightenment salons or Athens in the trial of Socrates.

Make it

1. 1

   Write the argument in standard form

   Conclusion in one sentence, then two to four numbered premises, including the unstated one. Check that the conclusion really follows.

2. 2

   Pick the objection and the reply

   Choose the strongest objection you know and name the exact premise it denies. Find the author's reply, or write your own and say it is yours.

3. 3

   Write one paragraph per part

   One paragraph for each premise, the thought experiment, the conclusion, the objection and the reply. Use the author's exact key terms every time, and quote the crucial sentence with a page number.

4. 4

   Paste it into openCanviz and keep your wording

   Choose Keep my wording so the narration says exactly what you reconstructed, set a length of three to five minutes, and pick whiteboard so the argument diagram builds as you speak.

5. 5

   Check each premise on screen against the text

   Pause on every scene. If a label has shortened 'comparable moral importance' to 'importance', fix it in the editor. Small wording changes are big philosophical changes.

6. 6

   End on a question

   Close with the premise you think is weakest, stated as a question for the seminar. It gives the discussion somewhere to start.

Using a video in a seminar or for revision

Presented at the start of a seminar, a four-minute argument video does what a handout does, but with the diagram built in front of everyone. Then you stop, and the room argues. Say that you made it and which tools you used, and check your department's rules on AI tools first. The general approach to graded videos is in how to make a video for a school project.

For revision, a set of short argument videos (one per set text) is a quick way back into a course. But watching is the easy part. The test is whether you can pause before the conclusion scene and say it, then pause before the objection and state it. If you cannot, rewatching will not fix that. Writing the reconstruction again from memory will.

Common questions

Do I have to agree with the argument? No. A fair reconstruction of a view you reject is worth more than a weak version of one you hold. Make the argument as strong as you can, then attack it.

How long should a philosophy explainer be? Three to five minutes for one argument, which is 450 to 750 spoken words. A whole philosopher does not fit in one video; one argument does.

Can I use a famous thought experiment without the argument around it? You can, but then it is a puzzle, not philosophy. The trolley problem matters because of what it is used to test (whether killing and letting die differ morally). Say which premise the case supports.

What if the text is in translation? Name the translation you used and quote it consistently. Different translations of Kant or Heidegger use different English terms for the same word, and switching between them on screen confuses viewers.

Is it acceptable to let a tool draft the reconstruction? For a graded piece, write the reconstruction yourself. That is the assessed thinking. A tool drawing and narrating your reconstruction is a different matter, but check your course rules.