Stoic Impressions
You cannot choose your first reaction, and the Stoics never said you could. What you can choose is the sentence you add immediately afterwards — which is where nearly all of the distress lives.
The most quoted line in Stoicism is that we are disturbed not by events but by our judgements about them.
It is also the least useful, because as advice it is unusable. Nobody has ever calmed down by being told their reaction was a judgement.
The Stoics were considerably more precise than the quote suggests, and the precision is where the practice lives.
The part that is not up to you
Start with what the popular version gets wrong. It implies you could, in principle, simply not react.
Seneca says otherwise, and at length. In On Anger he distinguishes the first movement — the involuntary jolt when something strikes you — from anger proper. The flush, the tightening, the lurch when you read the message: these arrive on their own. They are not yet passions. They are not decisions, and treating them as moral failures is a category error that produces nothing but self-blame.
The Stoic word for the raw appearance is phantasia, an impression. It arrives whether you like it or not.
What follows it is different. The impression comes bearing a proposition — this is bad, this is an injury, this means something about me — and you either sign it or you do not. That signature is sunkatathesis, assent, and it is the one part of the sequence that was ever claimed to be yours.
So the accurate version of the famous line is not you choose your reaction. It is: you cannot help the flinch; you can decline to sign what the flinch hands you.
Marcus shows the method
The clearest demonstration in the whole literature is four lines long:
Say nothing more to thyself than what the first appearances report. Suppose that it has been reported to thee that a certain person speaks ill of thee. This has been reported; but that thou hast been injured has not been reported.
— Marcus Aurelius, Meditations 8.49 (trans. George Long, 1862)
Read what he has actually done there. He has split one apparently single event into two statements:
- Reported: someone said something about me.
- Added: I have been injured.
The first is information. The second is a judgement that arrived so fast it felt like part of the information. Almost all of the pain lives in the second sentence, and the second sentence is not in the report.
That is the entire technique. Everything below is scaffolding for doing it reliably when you are upset, which is the only time it matters.
The pause
Epictetus gives the move that buys you the time to do it:
Wait for me a little, impression: allow me to see what you are, and what you are an impression of; allow me to try you.
— Epictetus, Discourses 2.18.24 (trans. George Long, 1877)
He is talking to the impression, and the address is deliberate. Naming the thing as an impression — rather than as the situation — creates a gap that would not otherwise exist. It is the ancient version of noticing that you are having a thought instead of simply being inside one.
The Stoics thought this had to be practised cold, on small provocations, precisely so it would be available during large ones. That is the right expectation. Nobody executes this for the first time during a real fight.
The exercise
Three lines. Do it after the event, not during — at least until it is habitual.
1. What a camera would have recorded
Write only what an indifferent recording device would have captured. Words said, in the order said. Actions taken. Times. No adjectives that imply intent, no motives, no tone.
This is harder than it sounds, and the difficulty is the diagnostic. If you cannot write the line without dismissively, deliberately, or again, those words are the judgement leaking into the report.
2. What I added
Now write, separately and honestly, the thing you concluded. Not a softened version — the actual sentence that was running.
She did that on purpose. He thinks I am incompetent. This always happens to me. I have been made a fool of.
Most people find the addition is a sentence about themselves, attached to an event that was about something else.
3. Is it true, and was it mine to make?
Two questions, in that order.
Is it true? Not is it possible — is it actually established by what the camera recorded? Usually you will find the judgement rests on one ambiguous detail carrying an enormous load.
Was it mine to make? Some judgements are yours by right: whether you were treated badly, whether you want to keep working with someone. Others are inferences about other people's inner states, which you cannot see and habitually get wrong.
Then decide, deliberately, whether you still assent. Sometimes you will, and that is a legitimate result.
A worked example
The event. A colleague replies to your long proposal with a single line: "Let's discuss."
1. What a camera recorded.
At 16:42 I sent a nine-paragraph proposal. At 17:03 they replied with the two words "Let's discuss." No meeting was scheduled.
2. What I added.
They think it's bad. They couldn't be bothered to read it. This is how it always starts before something gets shelved.
3. Is it true, and was it mine?
Is it true that they think it's bad? Not established. Twenty-one minutes is roughly enough time to read nine paragraphs, which cuts against "couldn't be bothered." "Let's discuss" is two words and carries no valence at all — the entire negative reading is being loaded onto brevity.
Was it mine to make? No. "What they think of the proposal" is an inference about a mind I cannot see, and I have been treating an unknown as a known for the last hour.
What survives. Someone wants to talk about the proposal. That is the whole report. What I do next — ask what specifically they want to discuss — is available now and was not available while I was arguing with an imaginary version of them.
The line to modern therapy
This is the Stoic exercise with the clearest modern descendant, and the lineage is acknowledged by the people who built it.
Albert Ellis credited Epictetus as an influence on Rational Emotive Behaviour Therapy, whose ABC model — activating event, belief, consequence — is structurally the Stoic sequence of impression, assent, passion. Aaron Beck likewise pointed to Stoic philosophy as the philosophical origin of cognitive therapy.
Two things follow, and they point in opposite directions.
It means the practice has a considerably stronger evidence base behind its mechanism than most self-improvement advice — the idea that examining an automatic judgement changes the resulting feeling has been tested extensively.
It also means you should not confuse the exercise with the treatment. Cognitive therapy is a structured course delivered by a trained person over months. This is a philosophical exercise you can do in a notebook. They share an ancestor; they are not the same intervention, and one does not replace the other.
Four ways it goes wrong
Trying not to flinch. The first movement is involuntary. Aiming at it guarantees failure and adds a layer of self-criticism on top of the original upset.
Using it to overrule a true judgement. Sometimes the addition survives examination. Someone did behave badly; the proposal really was dismissed. The practice is a test, not a verdict decided in advance, and using it to talk yourself out of accurate perceptions is a way of becoming unreliable to yourself.
Running it while flooded. If your heart rate is up, the examination will not hold — you will produce a persuasive-looking analysis that confirms whatever the impression already said. Wait until the physiology comes down, then run it. Twenty minutes is usually enough.
Skipping step one. The camera line is the whole exercise. Going straight to "is this judgement true?" without first writing what actually happened means you are examining the judgement using the judgement, and it always passes.
Where it sits among the other practices
| Practice | Use it when |
|---|---|
| Dichotomy of control | You are stuck on an outcome that was never yours |
| Premeditatio malorum | One specific feared loss keeps circling |
| View from above | Something real has swollen past its actual size |
| Memento mori | You are craving or grovelling and want the pressure out |
| Stoic journaling | You want the pattern across weeks, not one moment |
| Amor fati | It is settled, it is real, and you are still fighting it |
| Voluntary discomfort | You suspect a preference has become a requirement |
| Impressions | A specific thing just happened and you are reacting to your story about it |
This is the most immediate of the six — it operates on a single event, usually within a day of it. Its natural partner is the evening review, where the same additions turn up again and again attached to different events, which is how you discover you have two or three standing judgements rather than a hundred separate upsets.
Where to start
Take the last thing that annoyed you — small is better, not larger. Yesterday's, not this morning's.
Write what a camera recorded. Then write what you added. Read the two sentences next to each other.
The gap between them is the entire subject of this practice, and seeing it once in your own handwriting does more than reading the quotation a hundred times.
Common questions
What is a phantasia in Stoicism?
A phantasia is an impression — the raw appearance of something, before you have decided what it means. It arrives involuntarily. The Stoic claim is that the impression alone is harmless; what makes it painful is sunkatathesis, your assent to the judgement that comes with it. Assent is the part that is up to you.
Did Epictetus really say we are not disturbed by things?
Yes, in Enchiridion 5: men are disturbed not by things, but by the opinions they form about things. He immediately gives the example of death, which is not terrible in itself, since Socrates did not think it so — the terror is in the opinion about death, not in death.
Can I stop myself reacting?
Not the first reaction, and the Stoics did not claim otherwise. Seneca is explicit in On Anger that the first movement — the jolt, the flush, the flinch — is involuntary and is not yet a passion. What follows it is a judgement, and that is the part open to examination. Trying to suppress the flinch produces self-blame and nothing else.
Is this the same as cognitive behavioural therapy?
It is the ancestor of it, by the founders' own account. Albert Ellis credited Epictetus as an influence on Rational Emotive Behaviour Therapy, and Aaron Beck pointed to Stoic philosophy as the philosophical origin of cognitive therapy. The Stoic sequence of impression, assent and passion maps closely onto the modern activating event, belief and consequence. It is not a substitute for therapy, but it is not unrelated to it either.
What is the actual exercise?
Write down only what a camera would have recorded. Then, in a separate sentence, write what you added to it. Then ask whether the addition is true and whether it was yours to make. Marcus Aurelius gives the model in Meditations 8.49: that someone spoke ill of you has been reported; that you have been harmed has not been reported.
Does this mean my feelings are wrong?
No. Some impressions are accurate reports of real problems, and the practice is not a device for talking yourself out of legitimate reactions. Examining a judgement sometimes confirms it — at which point you have a real situation rather than a story about one, which is a better place to act from.