Mar 21, 2018 20 min read Systems

Obliquity by John Kay - Summary and Book Notes

It was summer. The beach was crowded. My friends and I had just finished college, ready to head into the real world. But first we had to fight for an open patch of sand.

The warmth hugging the bottoms of our feet. The sting in the hand after a Frisbee catch. The grit of sand in potato salad. All a great distraction from the uncertainty facing us.

As the sun set, we lit a bonfire. The conversation turned toward the future.

"What's your five-year plan?" I asked.

My friend shook her head and laughed.

"That's the most ridiculous prep-school thing I've ever heard. You don't need a five-year plan."

My entire life up until that point had been according to plan. Elementary school, middle school, high school, 4 years of college. Of course I can just keep making five-year plans.

I scoffed at her and dropped the subject. "Let's see in five years," I thought.

I should have listened.

The next five years looked nothing like I had planned.

Review

In "Obliquity", John Kay argues against my naive five-year plan.

The world is too complex for our best planning. Trying to approach nebulous targets like profit or happiness directly is a fool's path.

A better path: approach the target obliquely. Take small steps and iterate, course-correcting along the way.

Kay's writing is clear, and makes for an easy read. He shows plenty of examples of the failure of direct planning, and the sucess of obliquity. This book does get repetitive, especially if you already buy the main premise.

Highly recommend for:

The fresh college grad, unsure about which path to take
The central planner who thinks they know how to solve everything
Anyone making real-world strategic decisions

If you already believe the book's main premise, you could probably skip this one.

Buy Obliquity on Amazon

Quick Summary

We often make up reasons to justify decisions we've already made (Franklin's Gambit)
The principle of obliquity: Goals are often best achieved without intending them.
Happiness, profit, and other real-world objectives are easier to acheive when approached obliquely.
Complex objectives tend to be imprecisely defined. We learn about the nature of the objectives and the means of achieving them during a process of experiment and discovery
Oblique approaches to high-level objectives should not be equated with unstructured, “intuitive” decision making.
Centralized planners lack local knowledge. Decentralized communities lack global knowledge
We always have the problem of whether to treat it as new data about the parameters of the model or new data about the relevance of the model
We find intentionality and design where there is only chance and improvisation
Obliquity is the best approach whenever complex systems evolve in an uncertain environment and whenever the effect of our actions depends on the ways in which others respond to them.

Recommendations for Further Reading

Book Notes

The following are rough notes I took while reading. These are mostly paraphrased or quoted directly from the book.

Much of our revenue was derived from selling models to large corporate clients. If these models were helpful, why did we not build similar models for our own decision making? The answer, I realized, was that our customers didn’t really use these models for their decision making either.

They used them internally or externally to justify decisions that they had already made.

“So convenient a thing is it to be a reasonable creature, since it enables one to find or make a reason for everything one had a mind to do.” -- Franklin’s gambit, Benjamin Franklin

Hindsight rationalization

If people are predictably irrational, perhaps they are not irrational at all: Perhaps the fault lies not with the world but with our concept of rationality.

I call it the principle of obliquity: Goals are often best achieved without intending them.

Chapter 1: OBLIQUITY—Why Our Objectives Are Often Best Pursued Indirectly

"Visionary companies pursue a cluster of objectives, of which making money is only one—and not necessarily the primary one. Yes, they seek profits, but they’re equally guided by a core ideology—core values and sense of purpose beyond just making money. Yet paradoxically, the visionary companies make more money than the purely profit driven companies." — Jim Collins and Jerry I. Porras, Built to Last

Obliquity describes the process of achieving complex objectives indirectly. In general, oblique approaches recognize that complex objectives tend to be imprecisely defined and contain many elements that are not necessarily or obviously compatible with one another, and that we learn about the nature of the objectives and the means of achieving them during a process of experiment and discovery. Oblique approaches often step backward to move forward.

The hope that rational design by an omniscient planner could supersede practical knowledge derived from a process of adaptation and discovery swept across many fields in the course of the twentieth century. This approach was generally described as modernism.

Oblique problem solvers do not evaluate all available alternatives: they make successive choices from a narrow range of options. Effective decision makers are distinguished not so much by the superior extent of their knowledge as by their being aware of its limitations.

Problem solving is iterative and adaptive rather than direct. Good decision makers are not identified by their ability to provide compelling accounts of how they reached their conclusions. The most complex systems come into being, and function, without anyone having knowledge of the whole. Good decision makers are eclectic and tend to regard consistency as a mark of stubbornness, or ideological blindness, rather than as a virtue. Rationality is not defined by good processes; irrationality lies in persisting with methods and actions that plainly do not work—including the methods and actions that commonly masquerade as rationality.

PART ONE: The Oblique World: How Obliquity Surrounds Us

Chapter 2: FULFILLMENT—How the Happiest People Do Not Pursue Happiness

Flow experiences seem to contribute greatly to long-term well-being.

happiness is not simply the aggregate of happy moments.

Chapter 3: THE PROFIT-SEEKING PARADOX—How the Most Profitable Companies Are Not the Most Profit Oriented

The company that put more emphasis on profit in its declaration of objectives was the less profitable in its financial statements.

“Shareholder value is the dumbest idea in the world.” -- Jack Welch, former CEO at GE.

Elaborating his thought to Business Week a few days later, he explained: The job of a leader and his or her team is to deliver to commitments in the short term while investing in the long term health of the business. . . . Employees will benefit from job security and better rewards. Customers will benefit from better products or services. Communities will benefit because successful companies and their employees give back. And obviously shareholders will benefit because they can count on companies who will deliver on both their short term commitments and long term vision.

Chapter 4: THE ART OF THE DEAL—How the Wealthiest People Are Not the Most Materialistic

The richest men are not the most materialistic. Nor has it ever been otherwise.

Only in modern times has the principal route to great wealth been the development of a successful business. For most of history, and in much of the world today, the route to wealth was political and military power rather than business success.

Aristotle’s account of Thales of Miletus: it is easy for philosophers to become rich, if they want to; but that is not their objective in life.

I wondered why senior executives insisted on paying themselves so much. The remuneration was necessary, I realized, to sustain their sense of their own importance.

A corporate culture that extols greed cannot, in the end, protect itself against its own employees.

The motives that make for success in business are commitment to, passion for, business, which is not at all the same as love of money.

Everyday experience tells us that while greed is a human motive, it is not, for most, a dominant one.

Most people work not only for material rewards but also for the satisfaction of the job and the respect of friends and colleagues. Greed is not generally an overriding motive even for the very wealthy. For them, money is a mark of status, a register of achievement—or the by-product of a passion for power or for business.

Chapter 5: OBJECTIVES, GOALS AND ACTIONS—How the Means Help Us Discover the End

Daniel Nettle suggests that there are three broad senses of the term happiness.

The lowest—the basic level—comprises the momentary feelings that make us happy.
The intermediate level is typically a state of mind rather than a physical response, a sense of satisfaction and well-being.
Eudaimonia is a high-level concept, a measure of quality of life, of flourishing, of fulfilling one’s potential.

To function, we have to break a high-level objective—such as living well—into goals and actions.

The injunction “maximize shareholder value” is not a useful guide to executive action. “That’s not a strategy that helps you know what to do when you come to work every day.”

The problem seems to be determinate—we know the objective. The problem seems to be closed—we can define the possibilities open to us. The problem seems relatively simple—we understand the relationships between the different components.

If we know enough about such a problem—its objectives, its possibilities, its interactions—we do not have to worry about sharing our high-level objectives with those who are chiseling the stone.

That is what Lenin, the modernist architects and the business process reengineers believed. But they were wrong.

Direct approaches make a distinction between means and ends that often does not exist in reality.

The successful completion of the cathedral required the simultaneous pursuit of interacting basic actions, intermediate goals and high-level objectives. The stonemasons worked better because they understood that they were engaged in a great endeavor and that God was glorified not just by the magnificence of the cathedral but by their own dedication.

In a business environment—or any other environment with these characteristics—decision making cannot proceed by defining objectives, analyzing them into goals and subsequently breaking them down into actions.

those who claim to, like Le Corbusier or Lenin, have immense capacity to damage the complex systems they attempt to plan.

Chapter 6: THE UBIQUITY OF OBLIQUITY—How Obliquity Is Relevant to Many Aspects of Our Lives

The computer is very good at solving the problem we have specified and asked it to solve, but less useful when we are not quite sure what the problem is.

Directness was the product of obliquity.

“Art,” said Picasso, “is a lie that makes us realize the truth.”

Time demonstrates, but only slowly, whether policy has gone too far in one direction or the other

Jane Jacobs, who led the reaction against such planning (especially that of Moses), explained how the richness of city life was the product of obliquity, not design.

But even these flashes of inspiration, in which a solution suddenly reveals itself, generally come to people who have been thinking about a problem obliquely for a long time.

PART TWO The Need for Obliquity: Why We Often Can’t Solve Problems Directly

Chapter 7 MUDDLING THROUGH—Why Oblique Approaches Succeed

In 1959, Charles Lindblom described “the science of muddling through.”

The oblique approach was characterized by what he called successive limited comparison. “Muddling through” was a process of “initially building out from the current situation, step-by-step and by small degrees.”

Oblique approaches to high-level objectives should not be equated with unstructured, “intuitive” decision making. Lindblom’s vision of “muddling through” is a disciplined, ordered process.

Picasso, Walton, Buffett—each made his selection from a much smaller range of options than those in principle available.

Objectives, goals, states, actions evolve together because learning about the nature of high-level goals is a never-ending process.

The distinction between means and ends, which seems obvious and important in simple problem solving, is, as Lindblom explains, not central to practical decision making. The process in which well-defined and prioritized objectives are broken down into specific states and actions whose progress can be monitored and measured is not the reality of how people find fulfillment in their lives, create great art, establish great societies or build good businesses.

Chapter 8 PLURALISM—Why There Is Usually More Than One Answer to a Problem

Most real-life problems have less clear descriptions. Our high-level objectives are loosely defined and cannot be completely broken down in advance into specific goals and actions.

High-level objectives—live a fulfilling life, create a successful business, produce a distinguished work of art, glorify God—are almost always too imprecise for us to have any clear idea how to achieve them.

Because the process of achieving high-level objectives is necessarily iterative in this sense, the path to these objectives is bound to be oblique.

“What is the highest good in all matters of action? As to the name, there is almost complete agreement, for uneducated and educated alike call it flourishing, and make flourishing identical with the good life and successful living. They disagree, however, about the meaning of flourishing.” — Aristotle

Most people would agree that Canada scores higher for human development than Sierra Leone. But if the index didn’t give us that ranking, we would change the index, not our view of Canada or Sierra Leone.

Our judgments of what is a great poem and what is meant by human development determine the measures and the weights, rather than the weights and measures determining our assessment of the greatness of the poem or the meaning of human development.

But surely business is still different from poetry, or sociology? If we must be subtle and oblique in our appreciation of poetry or our assessment of human development, we must be direct and linear in the measurement of profit. Profit is a fact. Or is it?

Very few companies headline the supposedly objective measures given by International Financial Reporting Standards (IFRS) or Generally Accepted Accounting Principles (GAAP),

But anyone who has thought hard about the matter knows that the quest for the true measure of profit is as illusory as the search for a measure of poetic perfection, and for essentially similar reasons.

In the face of such difficulties, the philosopher of ideas Isaiah Berlin wrote in support of value pluralism.

Dr. J. Evans Pritchard may be on the right track when he invites students to consider the importance of a poem’s subject matter and the perfection of its form but on quite the wrong track when he proposes to measure greatness as the product of the two.

It is not absurd to ask, “What are the characteristics of a great poem?” But it is absurd to ask whether Keats’s “On First Looking into Chapman’s Homer” is a greater poem than Whitman’s ‟O Captain, My Captain,” and if so how much greater.

Berlin’s philosophy is pluralism, its essence “the notion that there is more than one answer to a question,” in contrast to monism, “the ancient belief that there is a single harmony of truth into which everything, if it is genuine, must fit.”

Pluralism is naturally oblique in its approach to objectives, monism direct.

Great artists do not only break the rules; they redefine them. Such obliquity is a key part of what makes a painter great. The criteria of achievement are constantly redefined by great achievers.

As Solon observed, only at the end will we know whether life was lived well. And whether we experienced a good education, whether business decisions were successful or whether the poem approached perfection of form is something that will be known only after, and often long after, the event. Even then we will never know whether the life that was lived or the education that was received, the business that was created or the poem that was written was the best possible.

Chapter 9 INTERACTION—Why the Outcome of What We Do Depends on How We Do It

Even if the pensions and health-care benefits that result are the same, the response of employees to the oblique approach is very different from the response to the direct one.

The actions of the man who buys us a drink in the hope that we will buy his mutual funds are formally the same as those of the friend who buys us a drink because he likes our company. But it is usually not too difficult to spot the difference, and the difference matters.

The planned, centralized solution failed in the forest because the planners did not possess the local knowledge held in communities that had nourished trees for thousands of years. The unplanned, decentralized solution failed in the Irish famine because the local communities did not possess the global knowledge required to anticipate, or protect against, potato blight.

Central targeting, a technique of planners everywhere, regards such decentralization as an essentially mechanical problem— But objectives are often vague, complexity extensive, problem descriptions incomplete, the environment uncertain

And most important, responses change the nature of the problem itself.

the introduction of measurement and control distorted the information needed to implement that measurement and control. This phenomenon is known as Goodhart’s law, after the British economist who observed that as soon as governments adopted monetary targets the aggregates they targeted changed their meaning and significance.

The balanced scorecard is one of the few business fads of the 1990s to have enduring value. The scorecard, which focuses attention not on one target but on many, reflecting many different aspects of corporate organization, recognizes the distorting effects of focusing on any single measure of performance

But Simon Marks and Israel Sieff realized that even multiplying the number of indicators was not enough. Complex high-level objectives could be decentralized effectively only if employees at every level absorbed the values that motivated these objectives.

Chapter 10 COMPLEXITY—How the World Is Too Complex for Directness to Be Direct

Under McNamara’s leadership, the Vietnam War had been fought by numbers.

"The bean counter wants to know all along the way how much is this, how much is that—and even before you get the genesis of the thing organized, they want to know the exact cost of each part. And you can’t think that way, so what you do is lie, fake, hide things behind the blackboard.... They can always prove they are right. They’ve got it on paper. They are right, but they are wrong." -- Frustrated Ford Employee

At Ford and in Vietnam, their colleagues responded to the misapplication of Franklin’s rule with Franklin’s gambit. They made up the numbers to support the conclusions they believed their superiors wanted to reach.

Honda has been too successful too often for accident and serendipity to provide a persuasive explanation of its success. Honda triumphed through improvisation and adaptation.

The greatest Whiz Kid of all simply did not know enough—and never could have known enough—about the people he was dealing with, the system he was challenging or the environment in which he was operating—to use the techniques of moral algebra that he and his colleagues had refined.

Chapter 11 INCOMPLETENESS—How We Rarely Know Enough About the Nature of Our Problems

Almost all real problems are incompletely and imperfectly specified, and to tackle them we have to try to close them in some way.

Chapter 12 ABSTRACTION—Why Models Are Imperfect Descriptions of Reality

Abstraction is the process of turning complex problems we cannot completely describe into simpler ones that we think we can solve. But gauging which simplification is appropriate requires judgment and experience.

The usual form of abstraction in business, finance and politics is the model. Like maps, models are selective simplifications.

The uncertainty about the bus’s arrival has two components. There is risk derived from the randomness reproduced within the model: the known unknown. There is uncertainty about the appropriateness of the model as a description of the world: the unknown unknown.

There is not, and cannot be, any analysis that shows that it is right to wait [for a bus] for twenty minutes but wrong to wait for twenty-five.

When new data arrives—in this case, the information that the bus hasn’t—we always have the problem of whether to treat it as new data about the parameters of the model or new data about the relevance of the model.

The assumption of normal distribution of returns seems to work well in times that are, well, normal. But what of abnormal times? The more sophisticated institutions test their own models against their own experience. But that experience is, of necessity, drawn from a period when the institution did not encounter the problems the models are designed to anticipate. The managers who used these models until their banks collapsed are like the people who are still waiting at the bus stop after an hour.

Why do people behave in these “irrational” ways? Only a few economists and game theorists find that question difficult. Everyone else knows that our approach to problem solving is more oblique. We are influenced by the context.

The experimenters have tried to eliminate these subjective factors. They lock their subjects in a room where they can’t see the other person. They tell them they only have to play the game once. They are trying to create a problem with defined objectives that is closed, simple and divorced from any social or economic context.

They are substituting a problem their methods can solve for the problem we actually face. They think about their subjects as agents, or players, not people—and there is a big difference.

We operate in a world in which we may sometimes be uncertain about our own objectives and will frequently be uncertain about the other player’s objectives. Real ultimatum games only superficially resemble the experimenter’s problem. These real-life problems are neither closed nor simple.

Map showing a short walking distance between two subway stations, and a longer 4 stop route if following the subway map

If I had asked at Paddington for directions to Hyde Park Gardens, I would have saved much time: If I had tried to buy a subway ticket to Lancaster Gate, a kindly ticket clerk might have told me that this was a silly way to make that journey. But computers are replacing these clerks.

PART THREE Coping with Obliquity: How to Solve Problems in a Complex World

Chapter 13 THE FLICKERING LAMP OF HISTORY—How We Mistakenly Infer Design from Outcome

We are inclined to see history through the lives of great men. That inclination blinds us to the real complexity of politics, business and finance. So we find intentionality and design where there is only chance and improvisation

The teleological fallacy, which infers causes from outcomes, is one of the oldest mistakes people make.

Tolstoy described it: The profoundest and most excellent dispositions and orders seem very bad, and every learned militarist criticizes them with looks of importance, when they relate to a battle that has been lost, and the very worst dispositions and orders seem very good and serious people fill whole volumes and demonstrate their merits, when they relate to a battle that has been won.

Today we call these interpretations, and misinterpretations, halo effects.

The mistake is to make inferences about the relationships between outcomes and processes when we cannot observe and do not understand the processes themselves.

To say that intention cannot be inferred from outcome is not to say that intentions have no effect on outcomes. Napoleon’s influence on history may be exaggerated, and Sculley may not have been the cause of Apple’s success or failure. But Napoleon did have an impact on the shape of modern Europe and Sculley bore responsibility for what happened at Apple.

The direct decision maker perceives a direct connection between intentions and outcomes; the oblique decision maker believes that the intention is neither necessary nor sufficient to secure the outcome.

The direct problem solver reviews all possible outcomes; the oblique problem solver chooses from a much more limited set.

The direct problem solver assembles all available information; the oblique decision maker recognizes the limits of his or her knowledge.

The direct decision maker maximizes his or her objectives; the oblique decision maker adapts continuously.

The direct problem solver can always find an explanation for his or her choices; the oblique problem solver sometimes just finds the right answer.

The direct decision maker believes that order is the production of a directing mind; the oblique decision maker recognizes that order often emerges spontaneously—no one fully grasps it.

The direct problem solver insists on consistency, on always treating the same problem in the same way; the oblique problem solver never encounters exactly the same problem twice.

The direct decision maker emphasizes the importance of rationality of process; the oblique decision maker believes that decision making is inherently subjective and prefers to emphasize good judgment.

Chapter 14 THE STOCKDALE PARADOX—How We Have Less Freedom of Choice Than We Think

Stockdale Paradox. Stockdale [a prisoner of war] could remain confident of and committed to the high-level objective of survival, yet acknowledge that he had almost no freedom of choice about his current actions.

A plan of how he would survive would have been harmful, not helpful. Stockdale observed that those who died were typically the optimists—those who said to themselves: “We’ll be out by Christmas.”

One lesson from these great survivors is the realistic possibility of accomplishing high-level objectives even without knowledge of future states or control over current actions.

The illusion that we have more control over our lives than we possess, that we understand more about the world and the future than we do or can, is pervasive.

Franklin D. Roosevelt, understood very well that goals and actions must constantly be revised if high-level objectives are to be achieved. Roosevelt described his approach as one of “bold, persistent experimentation.” “Try something,” Roosevelt went on. “If it fails, admit it frankly, and try another.”

THE HEDGEHOG AND THE FOX—How Good Decision Makers Recognize the Limits of Their Knowledge

Tetlock’s most striking discovery is that although the foxes perform better in terms of the quality of their judgments, the hedgehogs perform better in terms of public acclaim.

Hedgehogs are people who know the answers. Foxes know the limitations of their knowledge. Hedgehogs create headlines for journalists, and their confident certainties attract the attention of politicians and business leaders.

The same combination of simplification and reiteration is found in the thinking of gurus who believe they can reengineer large corporations, ideologues who see all political events through the same lens and visionaries who project some current technological or geopolitical trend with exaggerated pace and to exaggerated extent. All lack the real imagination needed to understand the complexity of their environment.

And yet the success of obliquity remains paradoxical. Surely you must do better if you intend to achieve something than if you don’t? The metaphor of the blind watchmaker illustrates that the answer to that question is often no.

If the environment is uncertain, imperfectly understood and constantly changing, the product of a process of adaptation and evolution may be better adapted to that environment than the product of conscious design. It generally will be.

I don’t know what the answer would be, although I am confident I could develop a model to deliver any answer I thought my client wanted.

If there is a one-line explanation of the power of obliquity, it would be “Evolution is smarter than you are.”

The processes of social and economic adaptation involve learning and imitation. Such adaptive processes operate on groups—as well as on individuals

Perhaps there is a better one-sentence resolution of the paradox of obliquity—“Adaptation is smarter than you are.”

But which of these factors was causal, and which derivative? The answer is both all and none. Genes, technology and society adapted together.

Chapter 18 ORDER WITHOUT DESIGN—How Complex Outcomes Are Achieved Without Knowledge of an Overall Purpose

the outperformance of controlled and planned economies by decentralized, disorganized market systems is perhaps the greatest triumph of obliquity.

Hume understood that the social contract was a metaphor, not a historical description of how governments came about.

business is distinguished from other forms of organization by having profit as its defining purpose. Yet this agreement encompasses evident nonsense.

why would society allow such an organization to exist? Business exists to serve social purposes and enjoys legitimacy in the short term and survival in the long term only to the extent that business meets the needs of the community.

If business did not maximize profit, what did it maximize? Making the same teleologicall fallacy mistake.

"What is a tiger for?" Tigers are good at being tigers because adaptation has honed them to be well adapted to the daily life of tigerdom. There is no more, or less, to it than that.

A good oil company is good at being an oil company, just as a good university is good at being a university, etc.

The most sucessful business leaders pursued the unquantifiable but entirely meaningful objective of building a great business.

Chapter 19 VERY WELL THEN, I CONTRADICT MYSELF--How It is More Important to Be Right Than to Be Consistent

The theory of rational choice dominates economic thinking today. Obliquity contradicts the theory that has been the most influential doctrine in the social sciences for at least forty years.

Sometimes we want things that are incompatible. There is nothing irrational about wanting incompatible things.

In a complex and uncertain world, there is no objective basis for deciding whether what has changed is the options or the choices, the factors that influence the decision or the decision itself.

Consistency as a hallmark of rationality belongs to a world far more certain than the one we inhabit.

"Do I contradict myself? Very well then, I contradict myself. I am large, I contain multitudes." -- Walt Whitman

Chapter 20 DODGEY DOSSIERS--How Spurious Rationality Is Often Confused with Good Decision Making

we get angry, and reject outcomes we think are unfair, for good evolutionary reasons.

Good oblique decisions are based on information, analysis and many years of experience, not on what people routinely call "intuition"

It would be silly to ask David Beckham to give an account of why he proposed to kick the ball in that particular way. But this demand for explanation is what we do when faced with major decisions in business, politics and finance.

Chapter 21 THE PRACTICE OF OBLIQUITY--The Advantages of Oblique Decision Making

The sucess of the physical sciences has encouraged us to believe there might be a science of decision making. There is no such science, and there never will be. The world is complex and imperfectly understood, and it always will be.

In obliquity, we learn about the structure of a problem as we solve it.

When faced with a task that daunts you, begin by doing something. Chose a small component that seems potentially relevant to the task.

The models aided the practiced judgment of foresters but did not replace it. Oblique approaches rely on a tool kit of models and narratives rather than any simple or single account.

Oblique problem solving is not less rational than Franklin's rule but more so. Obliquity doesn't mean that we should stop thinking about objectives, fail to examine options or omit to seek information and understand as best we can the complex systems that we deal with.

Obliquity is the best approach whenever complex systems evolve in an uncertain environment and whenever the effect of our actions depends on the ways in which others respond to them.