I’ve had the pleasure of having many contentious conversations over the past couple of years. This book dissects many of the problems and cognitive distortions present in these types of arguments and poses solutions to all of them.
All of the best techniques I’ve accidentally stumbled upon over the years (Rappaport’s Rules, understanding falsifiability, etc) are here, plus more. I particularly enjoyed its focus on epistemology: understanding why someone holds a belief gives more leeway to challenge its foundations. I also appreciated the use of scales: “On a scale of 1-10, how sure are you?”—These types of questions open room for doubt which can be explored without needing to force it. There is a lot of good phraseology and techniques for even the most advanced quarreler to learn from.
The Kahneman and Tsversky collaboration produced some of the most interesting and consequential work in the social sciences of the past century. Michael Lewis is an excellent author, and told their story with enjoyable and exhaustive narrative. The reader gets to learn not just about what their psychological discoveries were, but how and why they came to be.
I found Michael Lewis’s framing of their psychological work particularly interesting. In Moneyball, the focus is on how decision making can be made perfect and logical. In Kahneman and Tsversky’s work, the focus is on why decision making is not that way.
Shane Parrish’s website Farnam Street is one of my favorite on the internet. His collection of mental models is large, clear, and concise. It is the inspiration for me to have begun collecting mental models. When I found out he was writing a book—or, more precisely, a series of books—I was ecstatic, and preordered this months ahead of time.
It was good. He spoke through a number of the important mental models, cognitive distortions, and more. I took many useful notes. I found his organization of writing unique in that he introduced all the characters up front at the beginning of each chapter, and this was a novel technique I appreciated. If I were to find fault in the book, it would be that some of his examples don’t fully represent the models he is trying to illustrate. Still, they worked well enough.
While it is billed as a leadership book, Call Sign Chaos reads more like a narrative or memoir (from which one can glean many useful leadership lessons). I found Secretary Mattis’s exposé on the wars of the past four decades to be fascinating. You follow his rise from the lowest ranks in the Marine Corps to General and Secretary of Defense. Along the way, you learn much about warfare, the stakes, the people, and the decisions made. Like with General McMaster’s book Dereliction of Duty, one often feels acutely aware of how many of the bad decisions in warfare come from Washington. Political pressure seems a deadly phenomenon.
I have compiled the notes I highlighted from the book below into a reduced and readable “Sparknotes.” I’ve split it into four sections: leadership, organizations, warfighting, and politics.
Our education and credentialing system is a historical artifact. The information age has moved libraries from the universities to the internet. Education is more available than ever before. Yet for many fields, credentialing remains in the sole hands of colleges and universities.
It is expensive to educate in the higher education system. Why do students incur a cost which may take decades to recover from? Education is important, but you can learn in many different and more affordable ways. The secret lies with the degree. For many, college feels compulsory because it holds a monopoly on credentialing. Without a degree, you often can’t win an interview, much less get hired. We need a new system which breaks credentialing away from universities’ monopolistic grasp.
By splitting the education and credentialing systems, individuals and employers both benefit. Applicants are free to learn in alternative and more flexible ways. Apprenticeships, self-learning, trade schools, bootcamps, and more become viable options to amass knowledge and ability. Employers also get to source trusted talent from a wider pool of capable individuals. A new system can provide these benefits and more.
Properties of a Functional Credentialing System
Employers face an information gap between an applicant’s perceived value and actual value. Trusted third parties seek to collapse this gap by issuing credentials, which independently verify ability. Credentials are also an asset for an applicant who can be better compensated for proving a less risky candidate.
Closing the information gap is a multi-partite process. First, employers perceive a value of the credential and credentialing agent. Our first two properties will represent this process. Then, the perceived knowledge provided by the credential maps to the actual knowledge of the applicant. Good credentials correctly represent ability, bad credentials misrepresent it. Our third and fourth properties will represent this. Our fifth and last property ensures this formula can apply across the workforce.
1. Credentialing organizations must be trustworthy.
Credentials which are not trusted will fail to allow an employer to perceive ability in a candidate. The university system establishes trust by competing in a market of reputation. Collegiate reputation is afforded to those who graduate the best and brightest applicants.
In an alternative system, credentialing agents should compete in a similar way. However, because they need not be exclusive like colleges, credentialing agents’ reputations would not be based off what applicants they credential, but how well the credentials represent ability. Good reputation is bestowed to those who do an accurate job of measuring applicants, and the market dynamics allow for competition of methods.
2. Credentialing organizations should be comparable.
Employers must be able to compare different credentials to compare individuals. The enormous number of colleges and universities make comparing degrees difficult. An alternative system should have fewer credentialing institutions. Because credentialing takes fewer resources, this could happen naturally when split from education.
Examining institutions could handle more people than educational institutions like colleges. If everyone had access to the top exams, the market would dwindle to those most reputable. For example, there are few personal training certifications in the United States. Gyms are able to know the reputation of each, compare, and hire as appropriate.
3. Credentials should be granular in both depth and breadth.
Colleges issue degrees in four main tiers: associate’s, bachelor’s, master’s, and doctorates. These degrees also represent different fields (or majors). Rather than provide trust in knowledge and ability directly, college degrees do this by using education as a proxy.
An alternative system could improve by use scoring across more subfields and would measure ability rather than certify education. Colleges themselves differentiate ability for admission based on scores. Tests like SATs, ACTs, LSATs, GREs, and MCATs provide precision in more categories than they would with four tiers of “Bad,” “Fair,” “Good,” and “Excellent.” In the current system, someone might have a bachelor’s degree in Computer Science. A new system could score across Machine Learning, Programming, Computer Vision, and more.
4. Credentials should reflect someone’s skill over time.
College degrees last a lifetime. The knowledge acquired from a bachelor’s degree in Economics, Neuroscience, or Petroleum Engineering from decades ago might be considered outdated today.
That a person with that degree is still useful in the workplace is a credit to their years of experience. An alternative system should consider changes in the field and of the individual. They could use continuing education credits, fading scores, work-reviews, or other methods.
5. Credentialing should be independent of the source of one’s education.
Some universities offer credit for work experience, but the limitations are too great. College, apprenticeships, self-learning, and work can all yield capable individuals.
An applicant might have spent decades working in a field where they have acquired doctorate-level knowledge. They may not have a degree, but credentials ought to represent ability, not the source of ability.
The Future of Credentialing
In summary, a proper system for credentialing should be trustworthy and comparable, making the system useful. It should be granular and change over time, allowing it to best map perceived to actual skill. Finally, it should be independent of education, which makes it inclusive for more of the workforce. By being useful, accurate, and inclusive, an alternate system would more efficiently place people in the roles for which they are best fit.
Envisioning single day multiple-choice tests lacks imagination. Methods for implementing substantive tests would vary. You could imagine days-long or weeks-long exams. Or, to disincentivize cramming, the process might require several weekends of testing across a year. They could have written sections, oral examinations, and lab practicals. If simple testing is not sufficient to certify knowledge, then testing procedures would evolve. Institutions with the most trusted approaches would win in the market of reputation.
Some might lament that college is useful for other things besides credentials.This is true, but a straw-man for the argument at hand. People can and will find formal higher education enjoyable. They will make friends and future business connections. Undergraduates will mature into adults and learn to live alone. All these benefits are true for an on-campus college experience, but they can also be true outside of the system. An alternative system does not mean an end to college; people will still find value in attending for these benefits. It means that the monopoly on credentialing would be ended.
This system does not directly affect how education is delivered, but the second order effects will. All the people in the country who currently cannot afford four years of college may be better able to afford individual classes. In the current system, taking one-off classes is rarely worth it as one class does not buy a credential. In a new system, people will have an incentive to continue learning while on the job. They could take night or weekend courses to supplement their work, and each increase in tangible knowledge could allow them to increase their credential. This means more and continued education for millions who might otherwise forego college. Rather than accruing debt and losing time, people can further their careers and make education a life-long pursuit.
From an employer’s perspective, new credentialing institutions could more precisely rank candidates. This is more information, less risk, and better placement of employees. For individuals, there would be a great amount of new educational and financial flexibility. Rather than incur debt, one could work in their field, make money, and earn credit for real experience.
Change is not easy, but we need not start from scratch. A new system could emerge parallel to the existing one. They could be borne from the colleges, issuing college-equivalency degrees. Existing credentialing organizations could also expand their purview, like the American Bar Association, the Educational Testing Service (GREs), or the College Board (APs, SATs). No matter how it’s accomplished, there are changes we must make. Doing so will allow employers to best select employees and will liberate individuals to educate in more diverse and affordable ways. It’s a new millennium, and our antiquated system needs change.