I remember a close friend of mine worrying that in marrying for love he felt his powers of logic were clouded. Should he wait until, in effect, he wasn’t so much in love anymore? “There is a tide in the affairs of men,” I almost found myself shouting out loud. Letting that dramatic Shakespearean inclination sweep by, I found and equally dramatic part of me, far from within my own married state, wanting to cry out, “Why would anyone in his right mind logically choose to marry?” In effect, both partners must suffer a kind of logical self-impairment to make the commitment. A marriage is creatively destructive of both partners’ cherished notions of themselves. Despite the initial hopes of perfection, what one partner wants will not occur; what the other partner wants will not occur. Both are left with the actual marriage: a radically new conversation that is built on the razed foundations of their former identities. 48
Whyte suggests that there are three key marriages in most lives: marriage in the normal sense, to another person; marriage to our work or employment; and marriage to our selves—the relationship and understanding we have of our own selves as individuals.
In the best case, these marriages are all relatively healthy, and they are integrated with one another. Deficits and struggles in one will inevitability bleed into the others and damage them. Attempts to wall them off from one another are bound to fail.
So, what’s his advice? How can we succeed in these three marriages, and build a good life? As far as I understand the book, it seems he’s advocating for a somewhat Buddhist approach to life.
We must accept that suffering is inherent in life, realize that suffering is caused by our attachments to our own preferences, and to change the situation we need to accept reality as it is, and realize that there’s something deeper in our experience than our preferences and anxieties.
In marriage with a person, we accept the person as they are in reality. We accept that both peoples’ preferences and visions for the marriage will not match what actually happens. We accept that the other person is not responsible for our other marriages to work and our selves.
In work, I’m not sure what his advice is exactly. We should find work that matches our interests and ability so we enjoy the work itself beyond the money or status or stability it might provide. Good work, if you can get it.
For the relationship with the self, it’s to notice your life. Notice your connection to things, and what moves and motivates you. To pay attention to what actually interests you, and to pursue that. To seek answers to questions that you in particular find interesting, not that are promoted or popular. To recognize that there’s more going that just the surface level of pain or questioning or anxiety, that there’s a way to travel through life enjoying things despite the guarantees of difficulty.
There are some great quotes from this book, thoughtful and thought-provoking, and beautifully written. I enjoyed the exploration of the ideas through biographical sketches. I feel the advice on traditional marriage seems to be the most actionable or straightforward (you must accept the person, not your idea of the person; you must accept the marriage, not your fantasy of the marriage; you, the other person, and the marriage will change, so get used to reapplying these rules).
For work and the self it doesn’t seem as easy to summarize, and it’s not a down-in-the-weeds career advice book.
Review
Idea Density – medium
Related Books ?
Recommend to others: Maybe for people planning to get married, or thinking of marriage generally?
Reread personally: No
Quotes
I remember a close friend of mine worrying that in marrying for love he felt his powers of logic were clouded. Should he wait until, in effect, he wasn’t so much in love anymore? “There is a tide in the affairs of men,” I almost found myself shouting out loud. Letting that dramatic Shakespearean inclination sweep by, I found and equally dramatic part of me, far from within my own married state, wanting to cry out, “Why would anyone in his right mind logically choose to marry?” In effect, both partners must suffer a kind of logical self-impairment to make the commitment. A marriage is creatively destructive of both partners’ cherished notions of themselves. Despite the initial hopes of perfection, what one partner wants will not occur; what the other partner wants will not occur. Both are left with the actual marriage: a radically new conversation that is built on the razed foundations of their former identities. 48
To feel abandoned is to deny the intimacy of your surroundings. Surely, even you, at times, have felt the grand array; the swelling presence, and the chorus, crowding out your solo voice.
From his poem Everything Is Waiting For You, 72
What was strange about it was that I realized it wasn’t my own question and it didn’t belong in there; it was someone else’s question that I had taken on, thinking it was important to me because it had been important to others. The question was “Is there one God, or are there many gods?” “People have been killing one another for centuries over this one,” I said to myself. It was a supposedly important question that I was just discovering in the poem, but it was not in the least bit important to me, at least in the way it was asked. 87
To take on someone else’s conversational style and to keep repeating other people’s questions as if they were our own is to exhaust ourselves. It doesn’t matter if it is the thoughts of Socrates or Sunsan Sontag. Read and admire, but then go back to first principles and ask the question yourself, in your own way. 89
Stevenson’s courage lies in the specific recognition of a specific woman in a specific place with no other life than her own, children and all. 176
Since one of the core competencies of human beings, beside constantly worrying, is giving unasked-for advice to others, each of us on the threshold of commitment is surrounded by a swirl of warnings, advice and rules to follow around the public act of marrying. 183
Only those who put more energy into self-pity than into paying attention are truly marooned. 191
There is nothing personal about death taking us away; it is simply the ebb and flow of a tide beyond human understanding. 196
To feel a joy in life is also to know it is fleeting and will pass beyond our grasp. 196
[he] had no idea how much his defensive posture reinforced his notions of the world, what he had driven off or what had not even come out of hiding to be driven off in the first place. … How man other possibilities in life had hidden or run the other way at his appearance? To him the world constantly withheld itself from him and that was another piece of evidence pointing to the equally awful way the world was made. 223
Once we have renounced the need to live without suffering, to be special, to be exempt from the losses and doubts that have afflicted all people since the beginning of time, we can see the difficulties of others without being afraid ourselves. Our fearful, disappointed surface face starts to fall away. We can welcome other people into our lives because no matter their fears, they do not make us afraid. Suffering is the natural cyclical visitation that comes from being alive. 234
There is no way of living without anxiety, but there is a way of holding ourselves that is larger than any particular worry and allows our constant sense that something is wrong to fall into a natural hierarchy of experience. 234
We know we have the right vocation and are happily married to a work when we get a song in our hearts simply from doing the work itself, as much as from its rewards and its fruits. 287
My refusal to follow my vocation to please my spouse will only result in my demanding of her, at emotional gunpoint, all the qualities I cannot garner myself through my work. She will only be imprisoned by my frustrations and the very sacrifices she may have asked me to make. The refusal to participate fully in any of the three marriages, to make them overt and speakable, causes endless friction in a relationship. 315
I may have put all of my eggs in this basket of relationship because I am actually afraid of asking about the greater dimensions of my work or afraid of getting to know myself in a more honest and intimate way. It can be much easier to ask someone else to give these commodities to me and then punish the person in large and small ways when I see that they have not been delivered to my door. 317
The authors define noise, discuss sources/types of noise, show that it’s widespread problem, make suggestions for reducing noise, and respond to arguments against noise reduction.
What is noise?
In this context, noise is defined as unwanted random variability in the judgements people make.
If employees at the same insurance company provide different estimates for a particular car’s value, that’s noise. If there was no noise, all the valuations of the various employees would be identical. They would not necessarily be correct (the could be biased, always far below the appropriate market rate), but they would not be noisy.
If two judges in the US court systems give different sentences for the same case, that’s noise. If all judges gave the same sentence for the same crime, there would be no noise. The unanimous sentence may still be too harsh or lenient, but it would no longer be noisy.
If a particular judge gives five years for certain crime, and two months later the same judge gives seven years for the same crime, that’s noise. The random variation of the same person responding differently to the same case is noise.
Why is noise a problem?
If a car company charges too much due to a noisy valuation, they will lose a customer who doesn’t want to pay high premiums. If they charge too little due to a noisy evaluation, they will lose money when the client makes a claim after an accident.
If two judges can give different sentences for the same crime, how can the justice system be considered fair, rule-bound, or non-arbitrary?
If a particular judge gives different sentences for the same crime on different days, how could we consider that judge to be competent, qualified, or rational?
So, noise is unwanted random variability in the judgements people make, and they can cause problems? When does noise happen? “Wherever there is judgement, there is noise—and more of it than you think.” Imagine the noise in these situations: doctors or mechanics making a diagnosis, a hiring committee grading an applicant, teachers evaluating an essay.
How can we reduce noise?
We can make rules that govern or restrict judgements, like sentencing guidelines for judges. If judges can only give sentences between one and five years for a certain class of crime, by definition it’s less noisy than a case where judges are free to give up to ten years.
We can use models, rules, or algorithms to help us make decisions. A human judgement might be influenced by the good looks or personality of a job candidate, but the algorithm doesn’t take that as input. A human’s mood or stress level might increase noise in their judgement, but algorithms don’t experience moods or stress, or hunger and fatigue, etc.
We can use relative scales. When grading an essay, it can be difficult to assign a score of one to ten, because essays have so many dimensions (grammar, vocabulary, structure, voice, organization, creativity, coherence, scope, etc.). But when you read two essays, it’s usually easy to say which essay is stronger, to rank essays relative to one another. Rank allow you to make one comparison (essay one vs essay two), instead of trying to balance and weigh all the dimensions against some arbitrary scoring metric. If teachers rate two essays one to ten, there will likely be more noise than if they ranked the two essays relative to one another.
We can break a larger judgement into smaller components, and consider the components independently to avoid contamination. A job candidate might be evaluated on intelligence, collaborative skills, independence, a particular skill set, their body of work. Consider each component separately, so not to be biased by one factor.
We can aggregate judgements, to harness group intelligence. This works when people are competent, and give their judgements independently. This allows individuals to give useful input while avoiding social contagion.
What are the objections to noise reduction?
The fact that a judge is called a judge implies that their judgement is their reason for being. Cases can never be matched perfectly for prior convictions, intention, and other circumstances, so the noise we observe is due to the specifics of the case, not due to noisy decisions by the judge. The authors respond that guidelines still allow for recognition of these case specifics while still reducing noise that might allow us to view the justice system as somewhat coherent.
Rules or algorithms that we employ to help us make judgements might be biased. This may be true, but then we can adjust the rule so it is not biased. The rule is explicitly outlined, whereas people are not, so the rule can be more easily adjusted than a person. Both bias and noise contribute to error. So even if a rule only reduced noise, leaving bias intact, it would still be better than a situation without the rule where noise persisted.
People are responsive to subtleties that rules cannot detect or respond to. The evidence shows that, in general, you are not responding to subtlety, you are just noisy. If a doctor is presented with exactly the same medical information to make a diagnosis two weeks apart, and they make different diagnosis, they are not responding to subtle cues that a rule missed, they are just adding random unwanted variability to the diagnosis.
I enjoyed the book. It’s pretty easy to read, the examples and case studies are engaging. It seems to be a major problem in many domains, they make suggestions that seemingly help. I also appreciate the idea that they may be addressing a neglected problem. Bias gets attention because it’s directional and causal, but noise is random and statistical, so perhaps it’s harder to notice or think about.
Idea Density pretty ok
Related Books Thinking Fast and Slow
Recommend to others: if the topic is compelling, not broadly i guess
Reread personally: no
Quotes
If one of the team members took another shot, we would know very little about where it is likely to hit. 4
A general property of noise is that you can recognize and measure it while knowing nothing about the target or bias. 5
Wherever there is judgement, there is noise—and more of it than you think. 12
First, judgement is difficult because the world is a complicated, uncertain place. … Second, the extent of these disagreements is much greater than we expect. … Third, noise can be reduced. … Fourth, efforts at noise reduction often raise objections and run into serious difficulties. 21
System noise is a problem of systems, which are organizations, not markets. Disagreements make market. But if one of those traders is randomly chosen to make that assessment on behalf of her colleagues in the same firm, and we found out that her colleagues in the same firm would produce very different assessments, then the firm faces noise, and that is a problem. 28
In other words, we cannot measure noise in a singular decision, but if we think counterfactually, we know for sure that noise is there. Just as the shooter’s unsteady hand implies that a singe shot could have landed somewhere else, noise in the decision makers and in the decision-making process implies that the singular decision could have been different. 37
But our inability to observe variability [in a singular decision] would not make the decision less noisy. …strategies that reduce noise in recurrent decisions should also improve the quality of singular decisions. 38
Judgement can therefore be described as measurement in which the instrument is a human mind. Implicit in the notion of measurement is the goal of accuracy—to approach truth and minimize error. 39
Judgement is not a synonym for thinking, and making accurate judgements is not a synonym for having good judgement. 40
But in every case, a reduction of noise has the same impact on overall error as does a reduction of bias by the same amount. For that reason, the measurement and reduction of noise should have the same high priority as the measurement and reduction of bias. 56
…knowing the true outcome adds nothing at all to what was already known about noise in forecasting. 58
…the benefit of reducing noise is the same, regardless of the amount of bias. 64
Judgement is like a free throw: however hard we try to repeat it precisely, it is never exactly identical. 93
The question that drove Golberg’s research was how well a simple model of the judge would predict real outcomes. … Predictions did not lose accuracy when the model generated predictions. They improved. In most cases, the model out-predicted the professional on which it was based. 118
In short, relacing you with a model of you does two things: it eliminates your subtlety, and it eliminates your pattern noise. … You may believe that you are subtler, more insightful, and more nuanced than the linear caricature of your thinking. But in fact, you are mostly noisier. 120
…noise, which is variability in judgements that should be identical. 140
The invisibility of noise is a direct consequence of causal thinking. Noise is inherently statistical: it becomes visible only when we think statistically about an ensemble of similar judgements. 219
Bias is an error we can often see and explain. It is directional: that is why a nudge can limit the detrimental effects of a bias, or why an effort to boost judgement can combat specific biases. It is also often visible: that is why an observer can hope to diagnose biases in real time as a decision is being made.
Noise, on the other hand, is unpredictable effort that we cannot easily see or explain. That is why we so often neglect it—even when it causes grave damage. For this reason, strategies for noise reduction are to debiasing what preventive hygiene measures are to medical treatment: the goal is to prevent an unspecified range of potential errors before they occur. 243
The second opinion is not independent if the person giving it knows what the first opinion was. And the third one, even less so: there can be a bias cascade. 258
Regardless of diversity, aggregation can only reduce noise if judgements are truly independent. …must welcome the disagreements that will arise when team members reach their judgements independently. Eliciting and aggregating judgements that are both independent and diverse will often be the easiest, cheapest, and most broadly applicable decision hygiene strategy. 272
The Apgar score exemplifies how guidelines work and why they reduce noise. Unlike rules or algorithms, guidelines do not eliminate the need for judgement: the decision is not straightforward computation. Disagreement remains possible on each of the components and hence the final conclusion. Yet guidelines succeed in reducing noise because they decompose a complex decision into a number of dimensions. 282
…interviewers rate the candidate separately, before they communicate with one another. Once more—aggregation works, but only if the judgements are independent. 307
decomposition breaks down the decision into components 307
Most people do not love rigidity in the abstract, but it might be the best way of reducing noise and eliminating bias and error. If only general principles are in place, noise in their interpretation and enforcement will follow. 348
…in hindsight, corrected predictions will inevitably result in some highly visible failures. However, prediction is not done in hindsight. You should remember that outliers are, by definition, extremely rare. The opposite error is much more frequent: when we predict that outliers will remain outliers, they generally don’t, because of regression to the mean. That is why, whenever the aim is to maximize accuracy, (i.e., minimize MSE), corrected predictions are superior to intuitive, matching predictions.
The Digital Sauna 