Summary

Tempo is a look at decision-making that focuses on narrative as opposed to a more traditional “calculative rationality” that you see in most decision making research. Rather than focusing on how to be more “rational,” it acknowledges and embraces the idea that humans are not dispassionate calculators.

Rather, we rely heavily on narrative and story to make decisions about a world that is so complex we could not possibly engage in rational calculation even if we wanted to. Starting from this premise, it looks at how we can embrace and improve our ‘irrational’ decision-making.

The book starts by looking at the importance of timing and pace in decision-making. 

Take the example of a sleepy restaurant in a small town that is faced with an unexpected onslaught of diners from a tour bus while the executive chef is out running an errand. 

The staff starts to panic and loses its collective head. Ten minutes into the confusion, the executive chef returns, and instantly gets that there is a “situation.” He barks orders: “You! get the water boiling for the pasta. You there, man the vegetable station and get started chopping the tomatoes. We’re almost out of potatoes. You, run down to the store and get another 10 lb bag”

At the heart of the evolution of the situation is a change in the group’s collective tempo at the eleventh minute, when the executive chef turns confusion and anxiety into action. The psychological clock ticks faster, attempting to catch up with the real world. If you were to experience the episode and tell the story later, this would be a key moment in your narrative. ¹

In order to understand the tempo of a given situation and to guide our behavior, we use mental models. Specifically, we use narrative-based models. The chef had some mental model for how to handle a big order that other people in the kitchen did not.

A narrative mental model is a dynamic, unstable, and partially coherent set of beliefs, desires, and intentions held together by narratives that weave through the current realities, possible histories, and possible futures of a situation. 

Simple beliefs, desires and intentions are the primitive elements that comprise the vocabulary of thought. They are rather like the nouns, adjectives and verbs that comprise much of the vocabulary of languages.

This choice of primitives isn’t arbitrary; it turns out you can derive good accounts of nearly everything else involved in decision-making, such as planning and policies, from beliefs, desires and intentions. 

The relation between these three primitives and decision-making is straightforward: beliefs create or constrain possibilities, desires lead to preferences among them, and intentions represent commitments to specific courses of action. Each of the primitive elements can evolve in time, which is why mental models have momentum.

This view acknowledges narratives as powerful, unavoidable, and dangerous tools. Nicholas Nassim Taleb alongside many behavioral economists have argued that all narrative thinking should in fact be considered flawed. 

The critiques from Taleb and other behavioral economists are somewhat valid and are based on the observation that thinking in terms of stories leads to all sorts of biases. What critics miss, however, is that there is no such thing as non-narrative thought. It is not that rational cacluation is hard, but rather it is impossible (and, even if it was possible, it’s not clear that it’s desirable).

There are always multiple narratives at work, framing our perceptions, memories, active thoughts, decisions, and actions. 

There is no meaningful way to talk about specific decisions outside of a narrative frame and a concrete context.

Models of ‘rationality’ lie inside the mental models and narrative contexts that operate by them. To not acknowledge that is merely to fail to understand your own preconceptions.

By embracing narrative rationality, we can take stories seriously and make use of notions of truth other than scientific empiricism. 

We can meaningfully ask and obtain value from questions such as “Has he lived a fulfilling life?” or “Is this movie funny?” that only make sense within particular narrative contexts. 

The dangers of narrative remain, of course. But the best we can do to defend against them is to add an element of ironic skepticism and systematic doubt to our narrative imagination. This is tough precisely because we can only see the world through a mental model in the first place.

A simple (enough) model proposed for thinking about narrative rationality is the Double Freytag triangle which maps quite closely to Joseph Campbells’s Hero’s Journey.

You start out on some path and go through an exploration phase where you are groping around. Eventually, you figure out some organizing way to tell the story (the cheap trick) that you then spend figuring out how to craft the rest of the strategy around (sense-making and heavy lift). This eventually either works or doesn’t work (separation event) and then you reflect on it and start the process again.

The life of a person or organization is then viewed as a “Freytag staircase” of fractal narratives happening one after another.

The guiding heuristic in narrative-rational decision-making is the “most compelling and elegant story.”

Once you find a cheap trick, you can organize what you know in a very compact way. This compression and compaction create a mental model where the pieces fit together in a meaningful way and lend the model significant and coherent momentum, like a generally disturbed weather pattern coalescing into a tornado. 

This is akin to Occam’s razor: the simplest explanation of a set of facts is probably the correct one.

To make sense of a complex, ambiguous, and confusing set of facts, you should look for an organizing insight that dissolves the complexity and provides you a compelling and elegant way to look at your situation. 

To be compelling your view must be comprehensive and provide you a way to organize as much as possible, from what you know. To be elegant, the resulting mental model must be as compact as possible. In general, these models will be very local and unique to the immediate situation – you are not looking for a narrative that solves all the problems all the time, just the one you are dealing with right now.

This is a different sensibility from how most people think of “good decision making.” Rather than pretending some fully rational mode of decision-making is possible, it embraces the narrative and human element of it and looks at how to use it most effectively.

My Highlights and Notes

• In his 1975 classic, The Mythical Man Month, noted that “the programmer, like the poet, works only slightly removed from pure thought-stuff.” Today every sort of information work is acquiring the pure thought-like characteristics Brooks first noticed in programming. In these conditions, separating work into thought and action is less useful than it used to be. Work is simply whatever we must do to get from one decision to the next.
• Momentum accumulates and degrades constantly during the dozens of micro-decisions we make every minute. The world changes so fast that most of these decisions go towards simply maintaining situation awareness: largely subconscious decisions about what beliefs to add, update or discard, a process that computer scientists call truth maintenance. 
  • Note: There is just a certain tempo to life and the more dynamic the environment you are in, the more you have to update. NYC is high tempo so lots of updating that have to happen there.
• At 30 miles per hour, we are exposed to about 1,320 “items of information” per minute. 
  • Note: Why driving is tiring.
• You cannot manage this process one decision at a time. Neither can you selectively override your subconscious to only make the big decisions. This is the main reason tempo is a useful notion: it varies slowly enough that you can maintain an awareness of it, and use it to modulate the blurry torrent of your life. Being aware of tempo allows you to manage the momentum of your life. Managing momentum also means that when you do choose to slow down and pluck out individual consequential decisions for conscious and detailed processing, you won’t be the victim of conjuring tricks pulled on you by the mob of unruly autopilots – what Marvin Minsky called the “Society of Mind” – that exists just below conscious awareness. The more you consciously manage tempo, the more creative and realistic your options during the big decisions. 
  • Note: You have to make all the everyday decisions but you need to feel the tempo of where you are at. Part of why I like doing my weekly review on Friday/Saturday historically is they feel lower tempo and so I feel that I can be more reflective and thoughtful.
• If you learn to consciously manage tempo, however, you will be able to achieve the miracle of staying ahead of the fast-paced world, while slowing down.
• For bigger, more consequential decisions, we will adopt a framework based on synthesis, design, metaphor, and storytelling, rather than selection among predefined options. So rather than treating college as a “which college should I attend, and what major should I pick?” decision, we will treat it as the creative process of continuously retelling and enacting the most compelling College story you can.
• I define tempo as the set of characteristic rhythms of decision-making in the subjective life of an individual or organization, colored by associated patterns of emotion and energy.
  • Note: You can feel the emotion and energy of certain tempos and what decision-making headspace you are in. This is important for planning stuff, you need to feel your headspace. Part of the promise of psychedelics is being able to better sense this internally.
• A sleepy restaurant in a small town is faced with an unexpected onslaught of diners from a tour bus while the executive chef is out running an errand. The staff starts to panic, and lose its collective head. Ten minutes into the confusion, the executive chef returns, and instantly gets that there is a “situation.” He barks orders: “You! get the water boiling for the pasta. You there, man the vegetable station and get started chopping the tomatoes. We’re almost out of potatoes. You, run down to the store and get another 10 lb bag.” This is an example of a pattern of tactical decision-making that we’ll call scan-to-task.* You look around and rapidly assign every open resource you see to an open problem. … But at the heart of the evolution of the situation is a change in the group’s collective tempo at the eleventh minute, when the executive chef turns confusion and anxiety into action. The psychological clock ticks faster, attempting to catch up with the real world. If you were to experience the episode and tell the story later, this would be a key moment in your narrative. 
  • Note: This is an OODA example, the tour bus got inside the traditional OODA loop because it increased the tempo, you have to match the tempo of your environment to be effective.
• This represents a pinnacle of artistry where your mind is in what I like to call the clockless clock state.† Keep this phrase in mind because we will glimpse more of the clockless clock in domains such as stand-up comedy and improvisational theater. 
  • Note: good management is about controlling the tempo, it’s knowing where everyone is at and managing the tempo to keep things moving at the right pace, not too fast or too slow, with the right amount of variation. This is also kind of the idea of a morning ritual or ritual in general, you are controlling the tempo at a specific time and in a specific way that makes sense for the context. I like having a morning ritual and no calls in the AM because I can sort of ease into more deep thinking.
• One exercise I find useful is to hold on to different levels of tension in your body, from complete relaxation to spikes of frozen alertness. By consciously holding on to your bodily reactions to environmental tempo changes, for a little longer than they would normally last, you can train yourself to be more aware of them. This turns your entire body into a tempo barometer. 
  • Note: You want to feel the tempo of where things are at, be in your body in this sense.
• Ask someone who has just changed jobs, or traveled to a different country, about their impressions. If a change of tempo was involved, chances are that will be the first thing they mention. “Things are much more relaxed there,” or “it is a very exciting and fast-paced lifestyle.” In fact, the decision to switch jobs or pick a particular vacation destination was likely motivated by the allure of a different tempo.* “I wanted a change of pace,” your friend might say. 
  • Note: Life has a certain tempo that varies, this is what the Arthur Schopenhauer quote about vacillating between anxiety and boredom is about, your tempo is just moving around. This is perhaps what causes the happiness U-curve?
    
    Parents also at peak earning power are working at a very fast tempo in 30s/40s then after that kids grow up, money becomes less of an issue and the tempo slows down. This is also kind of a change your location, change your life. Like moving to Vietnam, NYC, Austin all at different times felt like I was matching the tempo to the season of my life.
• The most basic decision-making skill is adapting to the tempo of your environment and setting your own pace within it. It is significantly harder to become a pace-setter or pace-disruptor: somebody who can actually influence the tempo of the environment. The collection of behaviors involved in managing tempo is what people mean by the phrase sense of timing, and there is a lot going on beneath the sublime moments in comedy or stock trading that they have in mind when they use the phrase. 
  • Note: Obviously very influenced by OODA, the goal there is to modulate your tempo to match what the situation calls for. That’s not always moving faster, it can be slowing down or oscillating.
• Emotion results when you force yourself, or some part of the environment, to operate at a faster or slower tempo than it likes. To change a tempo you must add or remove energy by applying a force. The image to keep in mind is a child on a swing: the simple sort comprising a tire hanging on a rope from a tree branch. Depending on when you push, and in which direction, you can make the swing go higher, dampen it, change its direction, or change its phase.
• Dealing with any rhythmic process, such as a regularly scheduled meeting, or a daily work-life routine, involves similar dynamics. Drive up the tempo too much, and calm yields to excitement, and then to anxiety and panic (which is why driving up-tempo is a big part of creating FUD, Fear-Uncertainty-Doubt, in adversarial situations; “shock and awe” is an extreme example). Add too much damping, and emotions swing the other way: calm yields to impatience, frustration, anger, and finally, depressed acceptance. 
  • Note: This really is a key to management, it’s about managing the tempo of your team. You have to have fingerspitzengefuhl and einheit and know when to push and when to slow down.
• The interplay of time and emotion is deeper than you might think. Subjective time might well be the more fundamental sort of time. Einstein’s well-known quote was not a throwaway remark: “When a man sits with a pretty girl for an hour, it seems like a minute. But let him sit on a hot stove for a minute and it’s longer than an hour. That’s relativity.” 
  • Note: Totally, the tempo being too high drives me into anxious weeks. Having breaks like taking Sundays off, limiting calls in amount, and when they are and the like are all helpful ways to manage tempo in a way that is sustainable and keeps me emotionally excited and engaged without getting too anxious. This is the real-time management skill.
    
    Golf has been good for me because it is a natural tempo dampener. I think people have a sort of natural tempo and part of life and your career is finding the right tempo for you. Most entrepreneurs I know tend to do better in a faster tempo. Investors at a slower tempo. Even w/in investors there is maybe a trader/investor dichotomy.
• The world moves faster than we can hope to keep up, and we adapt by separating change into a background of predictable rhythms, and a foreground of unusual rhythms and non-rhythmic elements. Without this separation into foreground and background, the sheer cognitive load of keeping up with change, the process of truth maintenance, would overwhelm us. So, rather efficiently, we hold the compact belief that ‘the sun rises and sets every day” and relegate that belief to the background, rather than precisely tracking the sun at all times. 
  • Note: Very reminiscent of predictive processing and free energy
• Developing situation awareness is primarily the process of getting attuned to the dynamics of the background. Anything left over and unexplained is the foreground: raw material for active engagement.
• The workplace has its own base rhythms. In the United States, everything happens to the drumbeat of quarterly reporting and annual planning. Higher levels of the reporting hierarchy beat out deep, slow rhythms via large communication meetings, that act to synchronize the activities throughout the organization. A frontline team might have a heartbeat as short as a day. Software development teams that use a collaboration model known as Scrum usually adopt the ritual of the daily “stand-up meeting.” Operational divisions sometimes adopt sunrise and sunset meeting rituals. 
  • Note: a key part of management is managing the tempo of the organization. Not too slow, not too fast. Speeding up at the right times and slowing down at the right times. Going to scrum during a product launch ups the tempo.
• Suppose you have some new information about a competitor that you need to share. How do you proceed? Going with the flow means using established venues – your next team communication meeting for instance – to share the news. Alternately, you could set up an urgent cascade of emails that seeks out the impacted parties within your organization and triggers the setting up of an exceptional meeting. Here’s the curious thing about this choice: whether or not you decide to go with the flow will actually drive the urgency of your organization’s reaction to the news more than the content of the news itself. Using a go-with-the-flow model can create a dangerously complacent reaction to critical news. Invoking exception-handling models for trivial reasons will cause over-reactions. 
  • Note: Tempo is a more effective way of communicating information. There’s a Marshal McLuhan angle here where the tempo is the message a la the medium is the message.
• Pace-setting is the art of harmoniously driving the natural tempo of your environment away from its current state and towards your preferred state – slower or faster – in non-disruptive ways. 
  • Note: This is a lot of management, at critical times you need to drive the pace up. At non-critical times, you need to drive it down so people don’t get burnt out. volatility clusters so you need to take advantage of it in both cases by pace setting.
• Today I have a stable pace-setting routine that I use whenever I start cooking. I move clean dishes from the dishwasher to the cabinet and move dirty ones from the sink to the dishwasher. Remember the metaphor of the child on a swing? This is an example of adding a jolt of energy to what you might call the ‘kitchen cycle” to lift its emotional state. Dishes in the cabinet are like an energized swing at its highest point. Dishes in the sink are like a swing that has slowed to a stop. 
  • Note: This is a good example of solar flaring, like starting with some small wins and getting some momentum. Definitely helps in the gym, start with some easy, light warm-up work, and usually, you can set a new pace.
    
    The same can be true of an organization and ties into the idea of momentum, when things haven’t been going well then you need to slow down and get some easy wins. When things are going really well then push your advantage a la OODA and blitzkrieg.
• Script 1 took almost 24 hours to run its course. During this period, seven emails were exchanged, of which only two had any relevant information. The actual processing time for the task (finding and returning the report as an attachment) took three minutes in script 1 and four minutes in script 2. 
  • Note: Example of multiple emails back and forth over 24 hours to get something done vs. 1 time only. I think this is a good example, you want to increase the tempo of the organization in general as much as possible.
    
    SOPs let you increase the tempo by making things more efficient. This is also why it’s so advantageous to have people that can handle a whole function, you increase tempo by reducing the communication overhead.
• This behavior – interrupting and talking over others – can be so disruptive in fact, and so often the result of a disruptive temperament rather than a situational need for disruption, that it is nearly always viewed as an unpleasant and abrasive personality trait. But used judiciously, interruption and talking over others is how you, as a socially situated decision-maker, can arrest the momentum of developing group-think and assumed consensus. If you do not develop the thick skin to occasionally interrupt, and allow yourself to be interrupted, you will help enable pathological decision-making cultures wherever you go.
• If your team has a weekly meeting, should you schedule it on a Monday when situation awareness is low but energy is high, or on Wednesday when the opposite holds, or on Friday, when distraction rules, but you have the momentum of whatever was accomplished between Monday and Thursday on your side? 
  • Note: Yes, there is a tempo to the week and the year and you want to match it up with what is best there. I think team meeting early in the week (Mon/Tues) is good to set the tempo and more strategic calls later in the week (Thurs/Fri).
• Should you ask your manager to drop in once a month or once a quarter to provide the right level of encouragement, boosting, and feedback? Should this be an exceptional review meeting, or just before, during, or after a weekly meeting? Each choice has distinct effects, and interval logic helps you see a full range of options.
  
• Your calendar – and I mean here the actual grid of dates and times you see in something like Microsoft Outlook – as a canvas for artistry. Your calendar is not an empty container. It is a landscape of invisible energy and emotion associated with all the things that are going on in your life. A good breakfast and an appreciative email leading to an energized morning could cause a great meeting. Do you normally juggle your calendar so a tough conversation is scheduled for the energized morning?
• Momentum, a basic property of mental models that helps us conceptualize a great deal about decision-making. Momentum is the reason behaviors such as the fait accompli, brinkmanship, second-guessing, passive aggression, and time-outs have the effects they do.
• Mine: A mental model is a dynamic, unstable, and partially coherent set of beliefs, desires, and intentions held together by narratives that weave through the current realities, possible histories, and possible futures of a situation.
• A mental model is a dynamic, unstable, and partially coherent set of beliefs, desires, and intentions held together by narratives that weave through the current realities, possible histories, and possible futures of a situation.
• Michael Bratman in his classic: Intention, Plans and Practical Reason. 16 Within this framework, simple beliefs, desires, and intentions are the primitive elements that comprise the vocabulary of thought.
• This choice of primitives isn’t arbitrary; it turns out you can derive good accounts of nearly everything else involved in decision-making, such as planning and policies, from beliefs, desires, and intentions.
• Beliefs create or constrain possibilities, desires lead to preferences among them, and intentions represent commitments to specific courses of action. Each of the primitive elements can evolve in time, which is why mental models have momentum.
• A part of your active mental model is becoming real or enacted, as time progresses. While your possible worlds might be fantasies, enactments must obey the laws of physics. Your behaviors and their consequences, which constitute the enactment, create a unique reality. Your understanding of the ongoing enactment, however, will be consistent with many possible interpretations. Each is a special kind of possible world, made up of a history and an expectation. A history is a possible world that you claim might be true, while an expectation is a possible world that might become a history, in part as a consequence of your actions. … when you switch between a Plan A and a Plan B, momentum shifts from one bundle of possible worlds, to another. This is rarely a matter of turning one switch off and another one on. The clean fork-in-the-road shift is the exception; the dissipation of momentum associated with one course of action, and the cohering of another, is the rule. Enactments generally cannot be turned on a dime. Complex enactments containing a lot of momentum can be as hard to steer as an aircraft carrier. 
  • Note: You have to be conscious of setting the long-term vision and getting everyone’s mental models on the same page so that they enact in concert, this is einheit. This is why planning and communication are important. Good meeting schedules and agendas shape and guide the tempo of the organization.
• How might the emotional quality of the future and past affect the tempo of the present? Consider an example: you are driving to work and you are debating whether or not to stop at your favorite coffee shop. There are two expected possible futures: “stop at coffee shop” and “do not stop at coffee shop.” Until you decide, the tempo of the enactment will be influenced by both possibilities. If you are dreading that pending report you need to finish at work, and are slightly excited by the prospect of seeing that barista on whom you have a crush, the emotional quality of the tempo you experience will be a mix of dread and excitement. The pending report might be draining momentum, while the prospect of coffee might be energizing you via anticipation. The rhythm of your driving, evident in whether or not you are riding the brakes or accelerator, and in the smoothness of your turns and lane changes, will be affected by these thoughts.
  • Note: Woo woo speak version of this is the “energy” but it’s real, have to be aware of how narratives are affecting you and adjust accordingly. E.g. Don’t make big decisions when you have bad tempo/energy.
• As with Newton’s laws, to change momentum, you need forces. The forces that act on mental models arise from internal stresses, other mental models, and interactions with reality data (particularly through the addition and falsification of beliefs). 
  • Note: mental models have a momentum to them. if you want to change them then you need to have some break in place. Sometimes this is exogenous shocks, but it could also be creating space like a weekly review.
• A lot of this momentum is tied to the one common feature of all your experiences: you. Your mental model of yourself is a self-archetype. More generally, your mental models of people are archetypes. 
  • Note: You need to take actions that re-in force that you are who you say you are – you exercise, eat healthy, set boundaries. I think one reason exercising helps me eat healthier is it reinforces the self-archetype that I am healthy.
• Archetypes operating in conversations modulate the tempo of our decision-making: if you believe you are a can-beat-the-system guy and the prosecutor is Machiavelli, your conversation tempo defaults might get set to energy: high, emotion: assertive, rhythm: quick. Archetype assessments and tempo-control variables are part of the small set of variables we mentioned at the beginning of this chapter, which allow us to control real-time decision-making more efficiently. 
  • Note: Sit up, breathe slowly, speak confidently – all these things reinforce an archetype.
• Your most stable beliefs, the ones that actually modulate your behavior, aren’t about life purposes; they are about momentum management. You are more likely to switch religions than to switch from an impatient to a patient temperament.
• I am most attracted to Archilocus’ approach: relating archetypes to animal symbols. Animals often exhibit exaggerated and hard-wired versions of characteristic human behaviors and spark much useful introspection. My favorite examples of animal archetypes are from the British children’s classic The Wind in the Willows.19 Another British writer, Philip Pullman, has written an entire science-fiction/fantasy series, His Dark Materials,20 based on an alternate universe where individuals are associated with explicit animal archetypes called daemons.
• Every universe of archetypes emphasizes some aspects of human behavior at the expense of others. Be aware of what your favorite doctrines and archetypes leave out, and avoid locking yourself into limiting self-perceptions, or consigning others permanently to pigeonholes.
• But to build even simple unique models of others, you need data about their characteristic patterns of behavior. One of the best ways to find data is to notice and occasionally mimic repeated phrases and pet assertions used by those who are important in your life. … try to notice the positive patterns of behavior you see arising from people’s strengths. This isn’t just me preaching a positive-values philosophy. If you are being smart and surrounding yourself with effective people, this will actually lead to more accurate archetypes, since their strengths are likely more developed than their weaknesses.
• Mimicking someone right in front of them is a must if you want to actually validate that you are observing a valid pattern. If it is a positive pattern, the other will appreciate it being noticed, or if they are unaware of it, be grateful to you for helping them with some self-awareness. If you made a mistake, your subject will challenge your perceptions.
• Narratives, especially cradle-to-grave life narratives, are powerful, unavoidable, and dangerous tools. The dangers led one of my favorite writers, Nicholas Nassim Taleb to argue, in The Black Swan,23 that all narrative thinking should in fact be considered flawed. An entire chapter of the book is titled The Narrative Fallacy. Other thinkers in the decision-making tradition that Taleb represents (behavioral economics) adopt an even stronger position against narrative thought. The critiques are valid and are based on the observation that thinking in terms of stories leads to all sorts of biases. 
  • Note: There is no such thing as non-narrative thinking. We can only think in narratives, we just had to develop some meta-rational awareness of what is going on.
• What critics miss, however, is that there is no such thing as non-narrative thought, free of possible worlds and ongoing enactments. There are always multiple narratives at work, framing our perceptions, memories, active thoughts, decisions and actions. The idea that there is always a narrative at work is one aspect of the decision-making philosophy in this book, which is a situated decision-making philosophy. It is based on the assumption that there is no meaningful way to talk about specific decisions outside of a narrative frame and a concrete context, any more than it is possible to talk about physics without reference to a specific, physical coordinate frame (the basic idea in Einstein’s relativity). For this reason, I think of situated decision analysis as a sort of theory of relativity for rationality. By relative rationality, I mean that there is no privileged, narrative-independent model of decision-making that can be labeled absolutely rational. Models of rationality lie inside the mental models and narrative contexts that operate by them. This sort of philosophical relativism is a strong position.
  • Note: What metaphor are you having the discussion in? You are always in a narrative and you need to have one.
• The dangers of narrative remain, of course. But the best we can do to defend against them is to add an element of ironic skepticism and systematic doubt to our narrative imagination. This is tough precisely because we can only see the world through a mental model in the first place.
• Brief interludes, known as liminal passages, between the waning of one important life story and the waxing of another. You may remember an evening of introspection between high school and college, or between college and your first job. Between liminal passages, we live through a special kind of enactment, which I will call a deep story. Unlike ordinary enactments, a deep story is an episode of creative destruction that is significant enough to transform you. The transformation is a rebirth of greater or lesser magnitude. 
  • Note: You need a structure of life, mini-retirements, or sabbaticals where you have a liminal passage before the next deep story.
• Our modern ideas of decision-making derive from the ideas of Clausewitz’s contemporary, Antoine-Henri Jomini. Jomini’s process-oriented approach (represented today by thinkers in the business realm such as Michael Porter) is what I call calculative rationality. Its dominance is clearly illustrated by this quote from The Behavioral Revolution, an article by David Brooks in the New York Times (Oct 28, 2008): Roughly speaking, there are four steps to every decision. First, you perceive a situation. Then you think of possible courses of action. Then you calculate which course is in your best interest. Then you take the action. Brooks is describing textbook calculative-rational behavior. The first step is what we have called developing situation awareness. The second and third steps together constitute option generation and planning. The fourth step is what we usually call execution. Note a critical missing piece: there is no mention of mental models. Since perceiving a situation requires a mental model, calculative rationality generally assumes the existence of an appropriate mental model, rather than encompassing the creation of one. If you adopt Porter’s approach to business strategy, for instance, you might use his “five forces” model of competition as your mental model. This simple view is not wrong, it is just limited to simple situations that fit one or more of your existing mental models very well. In complex situations, planning based on such models is merely a training exercise to sample the space of possible worlds, get a sense of the complexities involved, and calibrate your responses appropriately. This is what Eisenhower meant when he said, “Plans are nothing, planning is everything.” Marc Andreessen, creator of the Web browser Netscape, described the idea more clearly: The process of planning is very valuable, for forcing you to think hard about what you are doing, but the actual plan that results from it is probably useless. 
  • Note: The problem with first principles is it ignores there is some mental model going into it. We can only think in stories and metaphors so the first step in making a decision is to understand the metaphor or narrative in which you are operating. decision making For planning, this has to get incorporated in some way. You need to understand that your plan is operating within a narrative that you are likely blind to.
• Narrative rationality is the ability to think, make decisions, and act in ways that make sense with respect to the most compelling and elegant story that you can improvise about a developing enactment.
• The simplest pattern that satisfies this constraint is the rise-fall structure of simple stories, such as fairy tales, which begin with one stable state (“Once upon a time… “), proceed through a climax, and recede to a different stable state (“… happily ever after”), where something has changed (the prince and princess are united, for instance). This basic structure is captured in a classic model known as the Freytag triangle. The Freytag Triangle Storytelling is a uniquely human mode of cognition but shorn of its meaning, this characteristic rise-fall structure seems rather like the biological stress response that is common to most animals. It is reasonable to speculate that the pattern is in fact the stress response, modulating cognition (though my approach does not rest on this conjectured biological justification). Later, we will informally interpret the y axis as entropy, or a measure of the disorder in your developing mental model. 
  • Note: Narrative structure relates to the stress response. Maybe the reason narrative evolved is because people like other animals need to go through periods of stress followed by rest to trigger hormesis and narratives mentally facilitate us living lives like that. Had the same thought re: entropy, we can only handle so much entropy in our lives.
• The Double Freytag triangle represents a tractable level of complexity. It is neither as simple as the basic Freytag triangle nor as arbitrarily complex as models of individual deep stories can get (which need only begin and end with a liminal passage). The Double Freytag Triangle We will adopt the Double Freytag triangle as our canonical example of the structure of a deep story.
• In narrative rationality, the cheap trick is a more fundamental mental event than goal setting. Where a calculative-rational decision-maker might consider fair costs and actual values of various desirable goals in a new environment, the narrative-rational decision-maker looks for steals and bargains.
• This is why the “most compelling and elegant story” is the guiding heuristic in narrative-rational decision-making. Once you find a cheap trick, you can organize what you know in a very compact way. This compression and compaction create a mental model where the pieces fit together in a meaningful way and lend the model significant and coherent momentum, like a generally disturbed weather pattern coalescing into a tornado. If you have some familiarity with the philosophy of science, this should remind you of Occam’s razor: the idea that the simplest explanation of a set of facts is probably the correct one. To make sense of a complex, ambiguous, and confusing set of facts, you should look for an organizing insight that dissolves the complexity and provides you a compelling and elegant way to look at your situation. To be compelling your view must be comprehensive and provide you a way to organize as much as possible, from what you know. To be elegant, the resulting mental model must be as compact as possible. In general, these models will be very local and unique to the immediate situation. 
  • Note: If you go through life making decisions based on the most compelling and interesting narrative then you are maximizing your compression progress and thus your velocity along some vector that likely goes somewhere compelling as it provides novel insights about organizing the world driven by compression progress optimize for interesting.
• “All models are wrong, some models are useful.”
• As writers like to remark, books are never finished, they are merely abandoned. This acceptance of necessary expediency leads to the increasing doubt and anxiety characteristic of the last hours before the first significant encounter with reality: the separation event.
• From the point of view of decision-making, the future and the past differ primarily in that the future is more disorderly and complicated.
• What happens when you view your life as a string of deep stories, with each successive liminal passage being, on average, a little higher than the previous one? The difference between the initial and final liminal passages can be interpreted equally well as doctrinal growth, or decay. Since we are interpreting the vertical axis as entropy, we will adopt the latter interpretation. This upward path is the Freytag staircase, and its pattern of rise and fall is the fundamental tempo.
  
• As you accumulate transformative experiences, your doctrine starts to occupy increasing amounts of room in your head, limiting the capacity for open-ended thinking.
• If I were to approach you with a proposal for a collaboration after you have read this book that interested you, but you were unable to commit at that moment, you might react in one of two ways. The calculative rational way would be to say, “I am overbooked right now, but I’d like to talk next year, maybe in January.” The narrative rational way would be to say, “I am in the valley of a deep story right now, how about I connect when I get to my next liminal passage?” Yes, that sounds theatrical, and I doubt my language will catch on, but it helps to think in terms of narrative time. 
  • Note: This correlates with controlling how much entropy is in your life, you don’t want too much. A good consideration for planning.
• The relationship between risk, learning, and information seems deceptively simple: every decision is based on what you know (information), and risk assessments associated with what you don’t know. Learning helps you increase usable information and lower risk. Calculative rationality focuses on risks (specifically, risks that can be modeled a priori), learning, and information primarily because you can bring a great deal of very sophisticated mathematics to bear. Empirical probabilities derived from historical data can be used to estimate the likelihood of different futures. Sophisticated discounting models can roll up the value of future payoffs into the net present value of likely options. And finally, a host of automated learning mechanisms can crunch raw data about outcomes into information about future expectations and even automated decisions. So why is this picture of comprehensibility deceptive? The reason is that despite the awe-inspiring mathematical machinery, such models are still based on an assumption that runs through almost the entire discipline of statistical decision-making like a giant fault-line: the future is like the past. The models are only as good as the instincts of the model builders in tuning the parameters, the accuracy of the model with respect to the past, and your level of faith in future-is-like-the-past assumptions in a given situation. The models can be viewed as a way of systematically leveraging or amplifying intuition. But intuition amplified through calculative-rational models, while necessary, is not sufficient for true risk management. 
  • Note: You need some high modernism in decision making but then you need a narrative overlay. This is like discretionary quant, you need to look at past models but don’t rely on them exactly. Goes back to ergodicity – you need to think about history but then extrapolate other bubbling narratives into how the future will be different.
• Future-is-like-the-past assumptions are a symptom of what is technically known as closed-world models: models in which pre-defined things exist, and only events that have been modeled can happen. The real world, by contrast, is an open world that requires that we constantly update our mental models to accommodate phenomena that we haven’t encountered before. The open world is a world that includes what Donald Rumsfeld called “unknown unknowns” and Nicholas Nassim Taleb calls “black swans” (rare, highly consequential events). This necessarily requires accommodation of periods of high entropy in mental models, while fundamentally new and unexpected information is being incorporated.
• As we age, we become more doctrinaire and less capable of open-world learning. Narrative-rational decision-makers necessarily age over time.
• The game keeps getting more complicated, and there are always more different ways to play.
• If an important decision is to be made [the Persians] discuss the question when they are drunk, and the following day the master of the house… submits their decision for reconsideration when they are sober. If they still approve it, it is adopted; if not, it is abandoned. Conversely, any decision they make when they are sober is reconsidered afterwards when they are drunk. – Herodotus
• These metaphors arise from two features shared with our ape cousins that shape our experience of the world: opposable thumbs and rich, binocular, color vision. We are defined by our hands and eyes the way dogs are defined by their noses and planning. Tactile manipulation has become our metaphor for all action, even though as social creatures, our tongues are at least as important a mode of action. So we grapple with difficult problems and deliver touching eulogies. Similarly, we have been shaped by our eyes to such an extent that seeing has become a metaphor for all perception. When you hear somebody make a point (a visual concept) in a conversation, you are more likely to respond, “I see” than “I hear you.” 
  • Note: The metaphors most used by humans are visual and tactile. What does that imply? Metaphors We Live By
• Learning helps us increase our store of usable information and lower risk. In most situations, you can learn a lot faster by doing than by watching, but unfortunately, action also exposes you to more near-term risks than watching. 
  • Note: apprenticeships are good because you get both.
• How can you act meaningfully when you don’t know much about the situation, and do not have relevant, well-developed mental models? The answer, as we saw briefly in Chapter 4, is that you cannot. You must act meaninglessly, or in other words, randomly. Unstructured learning behaviors are random to a lesser or greater extent depending on the maturity of the active mental models.
  • Note: as long as it won’t kill you just do something even if it’s wrong, that’s the fastest way to learn.
• Yerkes-Dodson law. This law posits that the ability to learn requires the right level of arousal. Too little, and you get boredom and disengagement. Too much, and you will react too quickly with hide and watch (or even fight or flight, the best-known natural behavior) to situations that might yield more learning with bolder exploration.
  • Note: This fits in with my diagram of not too much stress and not too little. There is a dose-response curve to stress.
• Humans are most comfortable in complex but legible environments. As a rough approximation, for example, we tend to prefer lightly wooded environments over either dense, impenetrable forest cover or barren deserts. That particular balance between being able to see and avoid being seen was perhaps the right one for our species’ needs in our early evolutionary environment. Similar principles apply to urban environments.
• A conceptual metaphor is a systematic structuring of meaning in one domain in terms of our understanding of another domain. Here is a simple example: “He gathered his thoughts.” Thoughts, literally, are patterns of neural firings. You cannot “gather” them. But the phrase structures our understanding of thought in terms of our understanding of tactile manipulation of a collection of solid objects. 
  • Note: Universal tactics are built of conceptual metaphors.
    
• Universal tactics combine to form basic fragments of enactments which we will call decision patterns, which are analogous to sentences. These in turn combine to form complete enactments.
  • Note: Interesting taxonomy. Universal tactics are metaphors that are grounded in our concrete reality. Groups of metaphors form decision patterns and decision patterns form enactments.
    
• Generating deliberative behavior requires computing with mental models. You simulate a few possible worlds consistent with the current state of your enactment. Whichever simulation gathers sufficient momentum to take control of your behavior first, generally wins. During this race, many possible worlds are simulated in parallel, interacting and competing with each other for control. Most of the time, the momentum race solves the problem of which decision pattern to enact, with no conscious weighing of costs and rewards. This is because simulations cause a weak pre-experience of associated energy and emotion patterns, creating anticipatory attraction/aversion responses. We weigh options during the process of coming up with them. 
  • Note: You are running a simulation of which model has the right emotional energy and then going with your gut. This is part of what the weekly review or quarterly planning accomplishes, you have time to run these simulations.
• Experts, characteristically, can skip quickly to correct answers without exhaustively exploring a complex decision tree of formal contingency plans. Effective recognition-primed behavior emerges when an inefficient and time-consuming calculative-rational decision tree is reorganized as a more compact but cryptic narrative-rational one. This happens as a novice gains experience and insights that allow him or her to gradually incorporate the cheap tricks (Chapter 4) from every new experience into canned prescriptions. Experts often describe novice behavior as being “by the book.” Their own repertoire of cheap tricks and hacks can be understood, within this metaphor, as scribbled margin notes that improve a calculative-rational tactical manual
  • Note: Experts have an intuitively built out and robust decision tree like that Slack diagram, just know all the contingencies with “cheap tricks” whereas the amateurs are much more sparse and require a lot of deliberation at each point. The value is in the marginalia, knowing what rules to break and when. Fingerspitzengefuhl.
• The keys to opportunism are unconscious recognition of a situation from one enactment while being absorbed in another, and the ability to artistically organize your enactments according to interval logic (the skill we encountered in Chapter 2). The two enactments may be loosely related, such as meeting a colleague from one project in a meeting about a different project, hardly a pure coincidence. Or they may be nearly unrelated, such as finding in the seat pocket of an airplane, a copy of an obscure book that you’ve been waiting to read. Opportunism involves the same dynamic as a cheap trick in a deep story: recognition of the potential for disproportionate rewards, which is why you get a weaker version of the same “Aha!” feeling. In general, when trading off the costs and benefits of one ongoing enactment against those of another, there are no bargains to be found. Few synergies are truly powerful. A no-free-lunch condition normally holds: gains in one enactment will be offset by comparable losses in another. At the extremities of the tradeoff, where we entirely abandon one enactment in favor of another, we normally think in terms of opportunity costs. Opportunistic tactical patterns violate no-free-lunch constraints by exploiting local conditions (such as the fact that there is a grocery store on your way home). Since time is often the resource under contention, so long as there is sufficient slack in the locally prevailing interval logic, unintended consequences due to opportunism are unlikely. You would not stop to buy coffee if you were rushing to catch a plane, but a five-minute delay getting home is usually acceptable. For this reason, opportunism can often be catalyzed by introducing some sort of inefficiency into all your enactments. Temporal inefficiency (schedules with slack) is the simplest kind, but any sort of inefficiency helps. Highly optimized behaviors are also highly blinkered behaviors. 
  • Note: serendipity and fast transients come from having a database AKA fingerspitzengefuhl and then having enough slack to realize it. Don’t plan too much in a week or for quarterly initiatives so people have space to try new things.
• The central idea in OODA is a generalization of Butterfly-Bee: to simply operate at a higher tempo than your opponent. This is a subtle point. A higher tempo is not the same as higher speed, in the sense of a race car overtaking another. To think in terms of tempo means to think in terms of (narrative time) frequencies rather than speed. The primary effect of operating at a faster tempo is that you can maneuver inside the decision cycle of your opponent, disrupting his or her enactment by introducing entropy into it faster than it can be removed. In terms of the entropy-based concepts in Chapter 4, you are forcing your opponent to stall in the high-anxiety exploration epoch, or prematurely commit to the wrong cheap trick out of sheer desperation. This style is obviously effective in adversarial settings, but it is equally effective in cooperative settings, a point that many who are inspired by Boyd miss. The alert waiters at upscale restaurants, fulfilling your needs before you even recognize them, are a good example. Exceptional customer service, not just war plans, can arise from OODA thinking.
• If we are more modest in our aspirations, the vocabulary of narrative rationality allows us to craft a pair of simple but cryptic and non-prescriptive definitions. A strategy is a cheap trick. A tactic is a metaphoric mapping among primitive action concepts in two or more domains.
• Decision-making is ultimately about constantly looking for the next cheap trick in ongoing enactments, scripted in constantly shifting tactical vocabularies. You will occasionally get a temporary and local advantage, but ultimately nature will defeat you. You can, however, consistently outmaneuver those who don’t recognize that there are no universal formulas, just as you can consistently outplay someone who always plays “rock” in a game of rock, paper, scissors. The equivalent to always playing “rock” in general decision-making is always reducing the fluid, creative-destructive strategy-tactics distinction to a specific rigid and prescriptive one. Each such reductive distinction creates a particular pattern of confusion. 
  • Note: You can’t get inside the market’s OODA loop, but you can get inside the OODA loop of predictable players. This is sort of the idea with disruption theory
• Definitions are unfalsifiable constructs. The success of a new definition ultimately rests on the accuracy with which it manages to label an intuitive and useful, but unarticulated concept.
• In other words, I hope to help you develop a sensibility, not teach you an idea. I call this sensibility the clockless clock sensibility.
• The four central ideas we have encountered – timing, mental models, narratives, and metaphor-based universal tactics – allow us to look at the world from a certain perspective: everything is a mental model. This is trivially true of natural realities: we can only comprehend them through mental models. It is also true of artificial realities but in a less trivial way. The material things we create individually and collectively are externalized mental models: parts of reality arranged to conform to the structure of an internal mental model 
  • Note: mental models are externalized through human creations. This is Cesar Hidalgo’s idea of crystallized intelligence.
• Kant once said. we see not what is, but who we are.
• You could therefore describe normal human thought as a balancing act between sensing realistically, and seeing through simplified platonic abstractions. When we create, our work usually reveals a bias towards one side or the other. The more we desire control and comprehension, the greater the extent to which the realities we see are simplified by the platonic categories of our mind, before emerging as creations at the other end. The more comfortable we are with chaos, the more we act as trusting, unconscious (and possibly uncomprehending) agents of raw perception. These two extremes lead to a common pair of opposed archetypes of action in mythologies around the world: design and dance.
• We are using the word legible in a slightly deeper way than is usual here. Typically we mean something like readable. In this usual sense, illegible corresponds to something like a smudged piece of written text. The original text is the result of coherent human intent (which we recognize even if we do not understand the language) while the smudging is an accident from the human perspective. The only order in the smudge is due to the laws of physics operating on ink and paper. Our use of the term is similar but broader: a piece of physical reality is legible if it is obviously the product of coherent human agency, a deliberate externalization of a mental model. When human and natural sources of order are harder to tease apart, you get greater illegibility. 
  • Note: Good definition of legibility
• Used with adversarial intentions, Boyd’s OODA can be understood as a deliberate use of illegibility to cause failure. As Scott notes, rebellions often start in illegible parts of cities: slums. 
  • Note: OODA is about starting with the illegible areas.
• Organization is about the meaning that matters to the person whose mental model is being externalized. Many people fail when they attempt to get organized because they make the mistake of striving for legibility and meaningfulness to an external eye, by imposing conventional or received social meanings onto personal realities: They might arrange their books by the Dewey Decimal system, believing it to be the “scientific” way, instead of a more useful logic for small personal collections (such as read/unread).
• So apparent creative chaos is fine, so long as it is sufficiently legible and meaningful to you. The externalizations of your mental models only have to be legible and meaningful to others to the extent that you must share meaning with them. 
  • Note: You need to let people organize in a way that makes sense to them but you also need to have some coherent way for organizing collective action, a clear schwerpunkt. I think having weekly meetings and clearly presenting these and then letting people execute as they see fit makes the most sense.
• The tempo of an externalized mental model is the reason why you can land in an unfamiliar and strange old city, and immediately get a feel for the place. 
  • Note: location leverage
• Some fields contain a lot of momentum (in the sense of mental models, not physics), such as an assembly line.
  • Note: start writing something easy and get flowing.
• A set of systems and processes, or a system-process complex, is usually the codified embedding of the mental models of a few centralized decision-makers, working according to highly platonic logic, and driven by a great desire for legibility and meaning from a central, but external, point of view. Just as calculative rationality often fails due to its impoverished view of planning and decision-making, system-process complexes often fail because they prioritize legibility and order over effectiveness. The governing aesthetic of system-process thinking is what Scott43 calls “authoritarian high modernism.” When system-process complexes fail, it is often due to the same pathology we noted before: overweening, impoverished design crushing an ongoing, organic dance. 
  • Note: everything in processes is totally up for change and should change all the time. You don’t want to make processes too pretty, you want to make them work. SOPs.
• System-process complexes often fail because they prioritize legibility and order over effectiveness. The governing aesthetic of system-process thinking is what Scott43 calls “authoritarian high modernism.” When system-process complexes fail, it is often due to the same pathology we noted before: overweening, impoverished design crushing an ongoing, organic dance. 
  • Note: everything in processes is totally up for change and should change all the time.
• Taylorism is commonly criticized for its dehumanizing effects, but for us, the more important problem is that it is also ineffective with respect to many of the problems it attempts to solve. To understand the limits of Taylorist system-process thinking, it is only necessary to imagine applying its methods (such as time and motion studies) to a problem such as “how can I throw a great dance party?” For a simpler illustration of the limits, recall the problem of restless elevator riders impatiently pushing the call button too many times. A typical Taylorist thinker would add a feedback signal, such as a display that shows the current location of all elevators, or a sign that says “please do not repeatedly push the button.” To arrive at artistic solutions like putting a mirror across from the bank of elevators requires moving beyond the systems-process mindset. Artistic approaches to field-flow complexes exploit the natural chemistry among mental models in richer ways. The key to such approaches is a tasteful blending of preparatory design and participatory dance. Successful field-flow complexes can neither be designed nor danced into existence. They must be orchestrated. A conductor or choreographer, rather than a pure designer or dancer, is necessary.
  
• To understand the limits of Taylorist system-process thinking, it is only necessary to imagine applying its methods (such as time and motion studies) to a problem such as “how can I throw a great dance party?” For a simpler illustration of the limits, recall the problem of restless elevator riders impatiently pushing the call button too many times. A typical Taylorist thinker would add a feedback signal, such as a display that shows the current location of all elevators, or a sign that says “please do not repeatedly push the button.” To arrive at artistic solutions like putting a mirror across from the bank of elevators requires moving beyond the systems-process mindset. 
  • Note: This comes back to the Four Levels of Organization, Taylorism is helpful at a certain point but you need expertise and culture to make it all work.
• Artistic approaches to field-flow complexes exploit the natural chemistry among mental models in richer ways. The key to such approaches is a tasteful blending of preparatory design and participatory dance. Successful field-flow complexes can neither be designed nor danced into existence. They must be orchestrated. A conductor or choreographer, rather than a pure designer or dancer, is necessary. 
  • Note: orchestrating or composing is a great metaphor for life and management.
• At the largest scale, the deliberate design remains effectively impossible. This is the central reality that must be faced by all organizations. So how do large and coherent organizations emerge at all? They emerge as an outcome of growth and evolution processes, rather than construction processes, and as such cannot be comprehended a priori by a human designer. They only make sense with the benefit of hindsight, and even then, only make partial sense. Organizations that are grown are fundamentally less legible than those that are constructed. 
  • Note: This is the organic metaphor and why it is important. It creates something less legible but more effective.
• For these processes to be successful, the mental models that seed any generative process of growth must have sufficient levels of mutual legibility and harmonize effectively at the level of tempo. In other words, they must be shared mental models, capable of sustaining a shared situation awareness of relevant realities. Shared mental models must include organizational self-archetypes and doctrines, just as individual mental models must include individual-level self-archetypes and doctrines. For organizations that exist within a larger ecosystem that includes other organizations, such as the set of businesses competing in a market, organizational mental models must also include archetypes for other organizations. 
  • Note: You need a sufficient level of shared values as part of the original company culture to have something emerge organically that is also functional. The organic growth only works with a sufficient set of shared values/mental models.
• Where complex realities are to be grown rather than constructed, the role of the individual orchestrator is limited to catalyzing the emergence of the right shared mental models in the early stages. This includes planting the seed of an organizational self-archetype and doctrine that is appropriate to the raison d’etre of the organization. Once the process is underway, the orchestrators can do little: the organization dances itself into existence and self-awareness. 
  • Note: this is company culture, the goal at the beginning is to get everyone to have the same shared mental models and understand how the business operates and let it grow. That isn’t the most important thing a CEO/management can do.

1. To couch it in OODA terms, the tour bus got inside the traditional OODA loop because it increased the tempo, you have to match the tempo of your environment to be effective.