Via Negativa: How to Improve Your Life and Work by Subtraction

What is via negativa?

Via negativa means getting better by removing things rather than adding them. Instead of asking “what should I start doing?”, you ask “what should I stop doing?”

The logic is an asymmetry of knowledge. You often cannot be sure a new habit, feature or rule will help, because adding something to a complex system creates side effects you did not plan for. But you can often be quite sure that a specific thing is hurting you: the meeting nobody needs, the smoking habit, the feature that confuses users. Removing a known harm is a more reliable bet than adding an uncertain benefit.

Via negativa does not say subtraction is always right. It says subtraction should be your first check, because people skip it by default.

Where it comes from

The phrase is Latin for “the negative way” and comes from apophatic (negative) theology. Medieval theologians argued you could not say what God is, only what God is not, so the path to understanding ran through negation.

Nassim Nicholas Taleb moved the idea into everyday decision-making in Antifragile (2012). His argument: in medicine, finance and policy, interventions often cause harm the intervener did not foresee (he uses the medical term iatrogenics, harm caused by the healer). Removing things, such as a toxin, a debt or an unneeded rule, tends to have fewer surprises than adding something.

There is also modern evidence that people overlook subtraction. In a 2021 paper in Nature, Gabrielle Adams, Benjamin Converse, Andrew Hales and Leidy Klotz ran a series of experiments where people were asked to improve an object, a piece of writing or an itinerary. Most participants added things, even when removing would have worked better, and they were more likely to notice the subtractive option when they were explicitly cued to consider it.

The same spirit shows up in design. Antoine de Saint-Exupéry wrote in Wind, Sand and Stars (1939) that perfection is reached not when there is nothing more to add, but when there is nothing left to take away.

How to use via negativa

1. Write down the goal, not the solution. “Ship faster” or “sleep better,” not “buy a new project tool” or “take a supplement.” A goal leaves room for removal; a solution already assumes addition.
2. List everything currently in the system. Every meeting, subscription, feature, commitment or habit that touches the goal. You cannot subtract what you have not listed.
3. Mark what you know is harmful. Ask which items you are confident make things worse. Confidence matters here: via negativa works best on known harms, not hunches.
4. Check the fence before you tear it down. For each item, ask why it exists. This is Chesterton’s fence: if you do not know what a rule or feature is for, find out before removing it.
5. Remove one thing, then watch. Subtract the biggest known harm first, wait long enough to see the effect, then repeat. Only after the subtraction pass should you consider adding anything.

A useful pairing is the Pareto principle: if a small share of your activities produces most of your results, the rest are candidates for removal.

3 real-world examples

Business

Say a software team keeps missing release dates. The instinct is to add: a new project tool, a daily standup, another manager. A via negativa pass asks what to take away. The team finds three recurring meetings that produce no decisions, a review step that duplicates automated tests, and a half-built feature that keeps breaking the build. Removing those frees more time than any tool would add, and nothing new had to be learned. (This is a hypothetical example.)

Personal life

Someone who wants more energy might look for a new morning routine, supplements or a fitness app. The subtractive version is usually more effective and more boring: stop drinking caffeine after noon, stop scrolling in bed, cut the second glass of wine on weeknights. Each removal targets something widely understood to hurt sleep, which is a safer bet than any single addition whose benefit is uncertain.

Investing

Many investors try to improve returns by adding: more positions, more strategies, more research subscriptions. Via negativa asks what to remove instead. High fees, frequent trading and leverage on positions you do not fully understand are well-known drags on long-run returns. Removing them is a guaranteed improvement in a way that picking the next winning stock is not. Taleb’s own view is similar: avoiding ruin matters more than finding the extra edge.

Common mistakes

• Subtracting what is actually missing. If the problem is that something essential is absent (no savings, no tests, no customer feedback), cutting more will not help.
• Removing without understanding. Tearing out a process because it looks pointless, before learning why it was put there, can reintroduce the problem it solved.
• Treating it as a one-time purge. Systems accumulate clutter continuously. Subtraction needs a recurring slot, like a quarterly review of meetings and subscriptions.
• Keeping things because you paid for them. The sunk cost fallacy is the main reason obvious removals do not happen. What you spent is gone either way.
• Confusing it with austerity. Via negativa is not about having as little as possible. It is about removing harm. A full life with no known harms is fine.

Via negativa vs inversion

Both models start from the negative side, which is why they are often confused. Inversion is a way of thinking: you ask how a plan could fail. Via negativa is a way of acting: you improve by removing.

In practice they work as a pair: inversion finds what is causing failure, and via negativa removes it.

The Mentarium take

Via negativa is the most underused move in the Mentarium Decision Loop. In the Map step, most people list options that add something: hire, buy, launch, learn. We think every Map should include at least one subtractive option, written out as explicitly as the others: “stop doing X,” “cancel Y,” “drop Z.” It often turns out to be the cheapest option on the list and, because removal is usually reversible, a low-risk first experiment.

Our rule of thumb: before you add anything to a system you already run, remove one thing from it. If you cannot find anything worth removing, you have learned the system is leaner than you thought, and the addition is more likely to be justified. If you can, you might not need the addition at all.

Via negativa also pairs well with antifragility. Taleb’s broader point is that systems which avoid known harms and keep their downside small are the ones that survive long enough to benefit from volatility. Subtraction is how you get there.