Two words at the center of the teaching: evenness is yoga. Not calm, not indifference — seeing that hasn’t been bought by the outcome. Why the win distorts your judgment as reliably as the loss does.
Of all the definitions of yoga scattered through this book, this is the shortest. Not power. Not peace. Not union. Evenness.
And notice where it sits — inside an instruction about acting. This is not a retreat-house attainment. It is a field condition, the state the work is meant to be done in.
Take the content precisely, because this is an optical claim before it is an emotional one. The even eye sees the same. The triumph and the wreck. The famous person and the invisible one. Not that it pretends they are identical — that the seeing itself is not purchased by them.
The vision does not brighten for the win and dim for the loss. It does not flatter the good outcome and flinch from the bad one. It does not resize a person according to their results.
And what the equal eye is not is where this teaching usually dies. It is not numbness. It is not the shrug. It is not caring less. The proof is the setting: this is being taught to a man who is about to be sent into a battle to fight with everything he has.
The reason I care about this one is not spiritual. It is that everything downstream of seeing depends on it.
The person who sees a failure at its actual size gets its actual lesson. The one who sees it swollen by dread gets only the dread. The equal eye is not the end of feeling. It is the beginning of measurement.
And the distortion is not random. It is patterned, which is the only reason it can be corrected at all. Losses loom larger than gains — one of the most-cited findings in behavioral science, though its size and universality are more contested now than the textbooks suggest. Taken at even half strength it means a setback arrives pre-magnified at roughly double its objective weight, and every conclusion you draw through it inherits the magnification.
Then outcome bias does the rest. Show people identical decisions with different results and they rate the decisions differently. The win retroactively blesses the process that produced it; the loss retroactively indicts it. So the uneven eye cannot even conduct an honest review. The lucky success’s flaws go unexamined and get repeated. The well-made decision that happened to fail gets abandoned.
This is why I have come to think the win is the more dangerous of the two. Nobody audits a good outcome.
The corresponding research usually runs under the name equanimity, and it is careful about the distinction that matters here. What is measured in trained people is not low arousal or blunted feeling. It is reduced reactivity of evaluation. The feeling comes in at full strength; the appraisal machinery stays unswept.4
The trained show the profile you would predict: faster return to baseline after both wins and losses, less rumination, and — the finding that matters most here — better calibrated judgment afterward.4 They extract more accurate lessons from the same events, because the events are being seen at something closer to their actual size.
There is a corroborating detail from the expertise literature that I find quietly persuasive. Elite performers’ post-event reviews are famously flat in tone. The best teams debrief a triumph and a disaster with the same instruments and nearly the same voice. That flatness is not coldness. It is a deliberate refusal to let the result grade the process, because the process is where next year lives.
Every result is now rendered at maximum size. The win as spectacle, the failure as content. Ranking culture attaches a live number to people, so the resizing of a human being by their results — the exact thing this teaching refuses — now happens automatically, publicly, in real time.
And the engagement machinery feeds on the swing itself. Triumph and disaster are the two highest-engagement emotions on every platform, which means the feed is tuned to alternate them. Your phone delivers, hourly, precisely the pairs of opposites this chapter warned about, pre-magnified, with comment sections attached.
Against that, an even eye starts to look like an instrument rather than a virtue. The people who keep one are easy to spot now, because the background has gotten so loud. They are the ones who sound the same after the good news and the bad. Interviewers find them boring.
The boringness is the visible edge of an interior discipline almost nobody models anymore.
Fig.02 · Yoga is evenness. The laboratory's translation: yoga is calibration.
I want to be careful here, because there is a version of this teaching that curdles into something cold, and I have met people running it.
Grief is not a calibration error. Injustice seen evenly is injustice tolerated. The poem asks for evenness about outcomes that happen to you, and people have quietly extended that into evenness about everything, including things that ought to produce a reaction.
The line I’d draw is this: the two days after the thing belong to the feeling. Celebrate properly, grieve honestly. What the practice protects is not the feeling, it’s the writing of the lessons — and the feeling should not hold that pen.
Note — evenness is not indifference
It is said to someone about to fight with everything he has. If a practice of evenness is reducing what you care about rather than sharpening what you can see, it has become something else — and that something else is much easier to maintain, which is exactly why it gets mistaken for the real thing.
One of these is simply writing the same questions twice.
Verse renderings in this piece are my own paraphrase, worked from the translations listed below rather than quoted from any one of them. Where I give a chapter and verse number, that is the location; the wording is mine and should not be read as a translation.
Evenness is not caring less. It is seeing that has not been purchased by the outcome — feeling at full strength and appraisal left unswept.
The uneven eye pays an itemized bill: magnified failures, unexamined successes, people resized by their results. Feel it all. Then measure it at its actual size.