30 May 2021

Subtract – Leidy Klotz

Introduction: The Other Kind of Change

What these three have in common is that they tapped into the power of subtraction. Sue Bierman subtracted a freeway to create one of the most visited places in the world. Leo Robinson sparked the financial subtraction that brought down apartheid. Elinor Ostrom subtracted wrong ideas to give humanity a better approach to our common future. All three made positive change because of their thought, courage, and persistence in taking away. And all three made change because they saw opportunities everyone else had missed.

Do your resolutions more often start with “I should do more of…” than with “I should do less of…”? Do you have more stuff than you used to? Do you spend more time acquiring information – whether through podcasts, websites, or conversation – than you spend distilling what you already know? Do you spend more time writing new content than editing what’s there? Have you started more organizations, initiatives, and activities than you have phased out? Do you add new rules in your household or workplace more often than you take rules away? Do you think more about providing for the disadvantaged than about removing unearned privilege? Are you busier today than you were three years ago? If you answered yes to any of these questions, you’re not alone. In our striving to improve our lives, our work, and our society, we overwhelmingly add.

sometimes more is better. And sometimes taking away brings delight.

Donating money to anti-apartheid rebels is helpful, but it doesn’t remove the harmful system’s power. Taking money away from apartheid does.

The breakthrough came when I figured out that what I am interested in is not simplicity, or elegance, or any other form of “less is more.” Subtracting is an action. Less is an end state.

subtraction is the act of getting to less, but it is not the same as doing less. In fact, getting to less often means doing, or at least thinking, more. Removing a freeway is far more challenging than leaving it alone or than not building it in the first place.

There is plenty of advice that works because it brings us to less: computer scientist Cal Newport preaches digital minimalism; chef Jamie Oliver distills recipes down to five ingredients; and the tidying savant Marie Kondo declutters homes. Each of these gurus guides us to specific ways we can subtract to improve. And their counterintuitive advice brings joy. But why does this advice remain surprising?

It’s been five centuries since Da Vinci defined perfection as when there is nothing left to take away; seven centuries since William of Ockham noted that it is “in vain to do with more what can be done with less,” and two and a half millennia since Lao Tzu advised: “To attain knowledge, add things every day. To attain wisdom, subtract things every day.”

Part I: Seeing More

we neglect subtraction as a way to change things


  • Lego bridge height
  • Grid
  • Miniature golf
  • Lego structure holding a brick
  • Grid with reduced bandwidth

Your task is to make the patterns on the left and right sides of the dark middle line match each other, so that if either one were lifted up and placed directly on top of the other, there would be a perfect fit. Your challenge is to do so by making the fewest possible changes.

Just 20 percent of participants were more likely to subtract to transform the grids.

Figure 1: One of Andy’s grid patterns

Also thanks to Andy’s grids, we learned that all of the adding was not due to varying effort required to add and subtract. It can be kind of hard to pull apart Legos. And, while the soup recipes were just a list on a screen, perhaps participants were imagining how much of a hassle it would be to physically remove tomatoes from the rest of the mixture. On the grids, however, it was the same click of a mouse or touch of a screen whether turning a square from shaded to white or the other way around.

One way to know what people are thinking is to ask. After completing the grid tasks, we asked participants to indicate their approach to transforming the grids: either “I added squares until they were symmetrical,” or, “I removed squares until they were symmetrical.” These self-reported responses confirmed that people were thinking about adding as adding and subtracting as subtracting.

Maybe we choose adding because we like things better that we have built ourselves – the IKEA effect. Perhaps we choose adding because to take away is to admit that previous additions are sunk costs. Or maybe we choose not to subtract, because we assume that, if something exists, there’s a good reason for it. Or because losses loom larger than gains.

how we change a situation depends on our conscious choices, but those choices depend on what comes to mind quickly and easily;

If adding was more accessible than subtracting, we hypothesized that three approaches would reduce this difference: searching our minds more deeply for ways to change the situation; bringing subtraction specifically to mind; and devoting more mental bandwidth to the change effort.

we add more than we subtract, that doing so causes us to miss out on good options, sometimes because we don’t even think of taking away.

participants in the reduced bandwidth condition were instructed to press the F key on the keyboard every time the number 5 came across the screen.

less bandwidth meant less subtracting. Along with cues and deeper searches, the bandwidth affect gave us a trifecta of evidence that the first reason we neglect subtraction is because we don’t even think of it.

There are surely other psychological forces, and cultural, and economic ones too, that explain why subtracting is neglected.

No matter how beneficial an act of subtraction is, it’s not likely to leave as much evidence of what we’ve done.

oversight isn’t the only way we neglect subtraction. As we’re about to see, our biology (this chapter), our cultures (chapter 3), and our economies (chapter 4) not only contribute to this oversight, they also lead us to reject perfectly good subtractions even when we do think of them.

Adding too much and then not subtracting enough may seem silly in experiments, but this same behavior turns sad when it is ruining real lives. Just as stress is linked to overeating, Preston has found that stress correlates with adding objects.

even before we learn math, we perceive quantity. As with touch, sight, and smell, we have a sense for less and more.

Raising the headphone volume from one to two bars sounds like more change than raising it from eight to nine. The change we sense depends on the initial state. An instinct for relative, as opposed to absolute, change would have been a helpful evolutionary behavior. When we rely on our instincts, the change between one and two is bigger than the change between eight and nine.

Does it favor more? Does it disadvantage less?

more than three million years of our ancestors spent their time subtracting from rocks, whether by instinct or choice. So, it seems, not all our inherited behaviors excuse our modern subtraction neglect.

When we sleep, our brain cells shrink, which leaves space for microglial cells to come in and clean up unused connections between neurons. Synaptic pruning is the name neuroscientists have given this automated subtracting.

To make better use of less, we can gain inspiration from nature. At ecosystem, species, and cellular scales, natural selection works with both hands. We may have instincts to add. But we’re surrounded by life that has been transformed by both adding and by subtracting.

Life was a high-stakes camping trip you were born into and never went home from. To survive was to be mobile. There were no home additions to build, obsolete newspapers to hoard, or bags of chips to binge on. Our ancestors had a built-in check on physical adding. When you need to carry all your stuff everywhere you go, you tend to have less of it.

Antoine de Saint-Exupéry, observed: “Perfection is achieved, not when there is nothing more to add, but when there is nothing left to take away.”

Modern analysis suggests that monumental architecture may have been the catalyst for civilization. (...) Scholars had long believed that no matter how eager a group was to begin adding temples, they had to first learn to farm and live in settled communities. But Schmidt said it was the other way around; the long-term effort to build the temple was the impetus for farming. In this way, adding brought civilization.

Writing both showed adding and enabled it, releasing the capacity to accumulate information from the confines of individual minds.

our view of the field has roots in cultural evolution. To see how, imagine yourself in this study: first, you are shown a pile of sand, which has been formed in the shape of the letter S. Then, you are shown two more situations: 1) a pile of sand, and 2) glass pieces that have been formed in the shape of an S. You are then asked, “Which of these second two situations is most similar to the first situation you saw – the sand shaped like an S?” (...) In the S-shaped sand study, American participants were more likely to say that the S made of glass was more similar to the S made of sand. Japanese participants were more likely to choose the pile of sand. (...) some people see the field (the sand), whereas others see objects (the S). Different people can see the exact same situations in predictably different ways.

just because people overlook subtraction, I am not contending that it is always the best choice.

an anaphora, one of the oldest rhetorical tools. An anaphora is made by repeating the same word or phrase at the beginning of successive clauses, for emphasis.

saying no is not subtracting. When someone rings the no-bell, it’s because they haven’t added some new activity, which is not the same as taking away an activity from what they were already doing. (...) if he really wanted a subtractive approach to productivity, he needed a stop-doing list. These lists had come to my attention by way of the management expert Jim Collins’s book Good to Great.

when we overlook stop-doings, we not only fail to streamline our schedules, we miss a chance to make ourselves happy. Thanks to such findings, I’ve gotten better at subtracting to-dos. I had lots of room for improvement, having been raised by parents who will spend thirty minutes going to pick up pizza, just to save the five dollars to have it delivered. (...) it’s challenging enough to stop-doing when it’s free. Spending money to save time is a stop-doing that you have to pay for. Time, it turns out, is worth the investment. (...) They asked more than six thousand people, from North America and Europe, whether they spent money on time-saving services like cleaning, cooking, and household maintenance. The rare few who did invest in stop-doings reported greater life satisfaction. (...) it was not about the money. Millionaires who paid to avoid the busy trap tended to be happier than millionaires who did not. And it was the same for people living on minimum wage. (...) To be more certain that spending money to save time was causing the increase in satisfaction, the researchers ran a field experiment. In it, they gave working adults two payments of forty dollars on two consecutive weekends. On one weekend, participants were randomly instructed to spend the forty dollars in a way that would remove something they found unpleasant from their schedule. On the other weekend, participants were assigned to spend the forty dollars to acquire something tangible.

Children’s bikes have been marketed as their own distinct class of bicycle for almost a century. There were plenty of design changes over that time: training wheels, fatter tires, shock-absorbing forks and seats, more and more speeds, and contraptions that connect a kid’s bike to a grown-up’s like a caboose. It’s remarkable, but perhaps not surprising given what we have learned, how long it took for someone to think of subtracting the pedals, which made two-wheeled bikes ridable for a whole new age group – and salable to their parents.

For about a decade, Nike’s less was invisible. You had to trust that the air was in there. Things really took off when Nike showed what Rudy had done. The iconic Air Max 1s featured a window on the side of the sole – to show the air. The window made the less noticeable, and the noticeable less proved marketable.

only one in sixty of the participants in our studies subtracts to improve the Lego structure.

Part II: Sharing Less

“I didn’t have time to write you a short letter, so I wrote you a long one.” Mark Twain often gets credited with this quip, and while there isn’t a record of him actually saying it, there is no shortage of similar sentiment. (...) John Locke’s: “But to confess the Truth, I am now too lazy, or too busy to make it shorter.”

Good writing illustrates this principle. Experts, examples, and research all suggest the same thing: less is objectively better. This is what Twain (maybe) and his predecessors acknowledged. This was the essence of Ernest Hemingway’s confessed practice of taking out parts of his short stories on the “theory that you could omit anything … and the omitted part would strengthen the story.” (...) Research has shown that an argument’s length is often used as a proxy for its quality. (...) teachers everywhere encourage us to add. History professors specify that an essay should be “at least ten pages.” Math instructors take off points if you don’t “show your work.”

Whereas default home organization advice is to get rid of things you don’t want, or that don’t fit, Kondo flipped that logic around – and focused it on us humans. She said we should keep what sparks joy and get rid of everything else.

One reward for those who persist in the process of improving is that this process can bring what happiness researchers call “flow” states. You are in a flow state when you are so immersed in what you are doing that time passes imperceptibly; all of a sudden, there is only a minute left in the game, the disc jockey announces the last song, or you have reached the last step in the directions for the Lego castle. In his authoritative book Finding Flow, the psychologist Mihaly Csikszentmihalyi lays out a convincing case that these states represent optimal mental experience. It happens, he argues, at the alignment between a challenge and our ability.

lazy less doesn’t spark flow because it is not a challenge. There is no transformation. Adding to get to good enough is slightly more challenging. But satisficing, by definition, doesn’t test the limits of changes or of our ability to make them. Good enough does not bring flow. Flow happens when we go beyond good enough.

Losing one hundred dollars, Tversky and Kahneman showed, feels more disappointing than gaining one hundred dollars feels satisfying. They called this finding – that the response to losses is stronger than the response to gains – “loss aversion.”

Those who sell things use our loss aversion to their advantage. Car dealers urge us to take that no-strings-attached test-drive, because the more we feel like we have the car, the more value we assign to it. Amazon.com gave me unlimited free two-day shipping for a year. I wasn’t going to pay the annual fee to get the service, but I now pay that same fee not to lose it.

Her verbs reveal, clean, and carve are gentler alternatives to subtract. They do not invoke a negative valence, and they do not activate loss aversion.

Kondo inverts subtraction too. Just as giving my students a pencil makes them value pencils more, when Kondo advises, “Before you start, visualize your destination,” she’s giving me a decluttered living space, at least in my head. With that tidy vision in mind, a Lego room that sparks joy, rather than the toys I’ll need to part with to achieve it, becomes the loss that looms larger than gains.

Daniel Kahneman put it this way: “Lewin’s insight was that if you want to achieve change in behavior, there is one good way to do it and one bad way. The good way is by diminishing restraining forces, not by increasing the driving forces.” Lewin’s “bad way” was to add – whether incentives for good behavior or punishments for bad behavior – because this increases tension in the system.

Kurt Lewin was not the only scholar (or Kurt) from the Gestalt school to gift us subtractive wisdom. Kurt Koffka, in between being married four times to the same two women, originated the cliché about high-performing systems: “The whole is greater than the sum of its parts.” Koffka had found an uncomplex way to say that, no matter how much we learn about parts of a complex system, we still can’t predict its behavior. The platitude above, however, while favored by sports announcers and motivational speakers, turns out to be a subtraction-blind mistranslation of what Koffka actually wrote. His original – and more accurate – wisdom was: “The whole is something else than the sum of the parts.” Koffka was miffed by the “is more than” misinterpretation. He knew that the whole can also be less than the sum of the parts. As he repeatedly clarified, to no avail: “This is not a principle of addition.”

Subtracting a part to enhance the overall performance of a complex system remains counterintuitive.

we played with nine players, two fewer than the eleven we were allowed in games. That shook us out of our suboptimal equilibrium. Only after we began to function as a system of nine did the coaches add two players back on. We were a transformed system. We played better than we had all year

as decision aids for complex systems, abridged lists work better than comprehensive ones. To understand why lists with less win, we need to understand working memory. This is the cognitive system that temporarily holds the information we have available for processing. In other words, working memory holds the ideas we can quickly bring to bear on our change efforts.

We cannot both represent every detail of a system and also expect to use the information. (...) our working memory has a severely restricted capacity, often below seven things, has been shown repeatedly.

Thinking in Systems emphasizes finding the goals of the system. As Meadows put it, we discover these goals by asking, “What is the system trying to achieve?”

Subtracting unnecessary detail is how we clarify places and ways to intervene.

after you have subtracted detail to find the essence of the system you wish to change, consider subtracting first, as in Jenga. Then persist to noticeable less. Last but not least, don’t forget that you can reuse your subtractions.

Doughnut holes provide a memorable illustration of this step in the subtracting process. As with the transition from solid blocks to Anna Keichline’s K-brick, it took a long time for someone to realize that fried dough could be improved by removing from it. (...) Removing a ball of dough from the center of the doughnut lets it cook more evenly – and provides more surface area for cinnamon sugar. (...) Less is literally more. (...) It would take more than a century for the holes to turn from functional void to salable solid. As we now know, those little bits of subtracted dough have plenty of appeal on their own. Whether you prefer Dunkin’ Donuts Munchkins (1972) or Tim Hortons Timbits (1976), reusing the subtraction has made for another stream of income.

When the state of California subtracted $11 billion out of apartheid South Africa, that was $11 billion they could invest elsewhere. Just because a subtracted bit was holding back one system, that doesn’t mean it can’t be useful somewhere else.

we now have a checklist that gives us room to act and adapt:

  • Subtract before improving (e.g., triage)
  • Make subtracting first (e.g., Jenga)
  • Persist to noticeable less (e.g., Springsteen’s Darkness)
  • Reuse your subtractions (e.g., doughnut holes)

These four steps can direct our expertise. We can keep the steps in our working memory as we go to work. Let’s call them the lesslist. (...) Nor do these four items summarize the first six chapters. (...) we need to bring our newfound subtracting skills to the lesslist.

One way to make invisible and distant changes more noticeable is by imagining what they will look like. In experiments, people who interact with aged images of themselves save more money than those who do not.

GDP does what it was set up to do – but no more. It measures production, not welfare. As a result, it misses some useful less. GDP counts the cost of the hospital visit but not the patient outcomes. (...) No matter how much practical value we derive from internet searches, as long as our use of Google doesn’t cost us anything, this act is excluded from GDP calculations. So is open-source work, like contributions to Wikipedia. (...) Not only does this metric miss useful less, it counts harmful more. (...) So do hurricanes, oil spills, prison construction, and inefficient government spending. Many of the things we want, from security, to healthy environments, to nurtured children, simply do not factor in this blunt national statistic – and may even lower it.

Modifying measures requires that we ask difficult questions about the public good. Alternatives, including the Human Development Index put forth by the United Nations, are also not as simple to calculate as GDP. To measure human development, nations need to keep track of and then compile numerous indicators like life expectancy, education, and per capita income. It is challenging to collect all of that information, but at least the indicators are agreed upon.

Selecting relieves the tension between a wealth of information and poverty of attention.

The simplest selection filter is that, if you can’t use it, it’s definitely not information.

the value of the time you have invested in this book is worth many times what you paid for it. To harness the benefits of our information privilege, we have to take ownership of the costs – both producer- and user-incurred.

If you’re like me, weeding out a single book, even one you will never read, is traumatic. (...) Whether in our bookshelves, in-boxes, or brains, intentional and regular subtraction of information is far better than the alternative. If we don’t sleep, which is when our synapses get pruned, our brains get overloaded and slow down. (...) The good news is that when we subtract information from our mental storerooms, our processing speeds up like a computer after closing a memory-intensive program that has been running in the background. Working at full capacity, we can create new knowledge – and perhaps even distill it into wisdom.

No matter the subject, we build mental models as we learn. We take ideas and the relationships between them, and we use them to represent reality. To do so most effectively, we need to add and subtract. (...) Of course, we need to add detail to our mental models. But it is by stripping away elements of the system that we find essence. Weeding less useful ideas allows the indispensable ones to flourish. (...) One of my favorite less-is-more productivity tips is to write down fewer notes. This is the information version of the closet-cleaning tip in which you get rid of anything you haven’t worn for a year. In both cases, you are forced to filter important content from inessential content.

Chuckle all you want at four-year-olds and cult members, but we all resist shedding wrong ideas.


  • Invert: Try less before more. Subtract detail even before you act, as with triage. Then, once you are ready to make changes, put subtracting first – play Jenga. And remember, just because we now appreciate that less is not a loss, that does not mean that your audience and customers do. So, tell them about this book and, in the meantime, don’t “subtract.” Instead, clean, carve, and reveal. Add a unit of transformation.
  • Expand: Think add and subtract. (...) don’t forget to zoom out to see the field, because stop-doings and negative numbers are not impossible. Plus, the field is where the tension is, and removing it is the “good” way to change systems. So sure, add diversity, but subtracting racism is the prize.
  • Distill: Focus in on the people. Bikes do not balance, but toddlers can. (...) Strip down to what sparks joy. Decluttering delights, and so does the psychology of optimal experience.
  • Persist: Keep subtracting. (...) Don’t forget that you can reuse your subtractions, like doughnut holes.