Why don’t schools teach debugging, or, more fundamentally, fundamentals?

If you find this piece worthwhile, consider the Go Fund Me that’s funding ongoing cancer care.

A story from Dan Luu, from back when he “TA’ed EE 202, a second year class on signals and systems at Purdue:”

When I suggested to the professor that he spend half an hour reviewing algebra for those students who never had the material covered cogently in high school, I was told in no uncertain terms that it would be a waste of time because some people just can’t hack it in engineering. I was told that I wouldn’t be so naive once the semester was done, because some people just can’t hack it in engineering.

This matches my experiences: when I was a first-year grad student in English,[1] my advisor was complaining about his students not knowing how to use commas, and I made a suggestion very similar to Luu’s: “Why not teach commas?” His reasoning was slightly different from “some people just can’t hack it in engineering,” in that he thought students should’ve learned comma usage in high school. I argued that, while he might be right in theory, if the students don’t know how to use commas, he ought to teach them how. He looked at me like I was a little dim and said “no.” 

I thought and still think he’s wrong.

If a person doesn’t know fundamentals of a given field, and particularly if a larger group doesn’t, teach those fundamentals.[2] I’ve taught commas and semicolons to students almost every semester I’ve taught in college, and it’s neither time consuming nor hard. A lot of the students appreciate it and say no one has ever stopped to do so. 

Usually I ask, when the first or second draft of their paper is due for peer editing, that students write down four major comma rules and a sample sentence showcasing each. I’m looking for something like: connecting two independent clauses (aka complete sentences) with a coordinating conjunction (like “and” or “or”), offsetting a dependent word, clause, or phase (“When John picked up the knife, …”), as a parenthetical (sometimes called “appositives” for reasons not obvious to me but probably having something to do with Latin), and lists. Students often know about lists (“John went to the store and bought mango, avocado, and shrimp”), but the other three elude them.

I don’t obsess with the way the rules are phrased and if the student has gotten the gist of the idea, that’s sufficient. They write for a few minutes, then I walk around and look at their answers and offer a bit of individual feedback. Ideally, I have some chocolate and give the winner or sometimes winners a treat. After, we go over the rules as a class. I repeat this three times, for each major paper. Students sometimes come up with funny example sentences. The goal is to rapidly learn and recall the material, then move on. There aren’t formal grades or punishments, but most students try in part because they know I’m coming around to read their answers.

We do semicolons, too—they’re used to conjoin related independent clauses without a coordinating conjunction, or to separate complex lists. I’ll use an example sentence of unrelated independent clauses like “I went to the grocery store; there is no god.”

I tell students that, once they know comma rules, they can break them, as I did in the previous paragraph. I don’t get into smaller, less important comma rules, which are covered by whatever book I assign students, like Write Right!.

Humanities classes almost never teach editing, either, which I find bizarre. I suspect that editing is to debugging as writing is to programming (or hardware design): essential. I usually teach editing at the sentence level, by collecting example sentences from student journals, then putting them on the board and asking students: “what would you do with this sentence, and why?” I walk around to read answers and offer brief feedback or tips. These are, to my mind, fundamental skills. Sentences I’ve used in the past include ones like this, regarding a chapter from Alain de Botton’s novel On Love: “Revealed in ‘Marxism,’ those who are satisfying a desire are not experiencing love rather they are using the concept to give themselves a purpose.” Or: “Contrast is something that most people find most intriguing.” These sentences are representative of the ones first- and second-year undergrads tend to produce at first.

I showed Bess an early version of this essay, and it turns out she had experiences similar to Luu’s, but at Arizona State University (ASU):

My O-chem professor was teaching us all something new, but he told me to quit when I didn’t just understand it immediately and was struggling. He had daily office hours, and I was determined to figure out the material, so I kept showing up. He wanted to appear helpful, but then acted resentful when I asked questions, “wasting his time” with topics from which he’d already moved on, and which I “should already understand”.

He suggested I drop the class, because “O-Chem is just too much for some people.” When I got the second-highest grade in the class two semesters in a row, he refused to write me a letter of recommendation because it had been so hard for me to initially grasp the material, despite the fact that I now thought fluently in it. My need for extra assistance to grasp the basics somehow overshadowed the fact that I became adept, and eventually offered tutoring for the course (where I hope I was kinder and more helpful to students than he was).

Regarding Bess’s organic chemistry story, I’m reminded of a section from David Epstein’s book Range: How Generalists Triumph in a Specialized World. In his chapter “Learning, Fast and Slow” Epstein writes that “for learning that is both durable (it sticks) and flexible (it can be applied broadly), fast and easy is precisely the problem” (85). Instead, it’s important to encounter “desirable difficulties,” or “obstacles that make learning more challenging, slower, and more frustrating in the short term, but better in the long term.” According to Epstein, students like Bess are often the ones who master the material and go on to be able to apply it. How many students has that professor foolishly discouraged? Has he ever read Range? Maybe he should.

Bess went on:  

Dan’s story also reminds me of an attending doctor in my emergency medicine program; she judged residents on what they already knew and thought negatively of ones who, like me, asked a bunch of questions. But how else are you supposed to learn? This woman (I’m tempted to use a less-nice word) considered a good resident one who’d either already been taught the information during medical school, or, more likely, pretended to know it.

She saw the desire to learn and be taught—the point of a medical residency— as an inconvenience (hers) and a weakness (ours). Residency should be about gaining a firm foundation in an environment ostensibly about education, but turns out it’s really about cheap labor, posturing, and also some education where you can pick it up off the floor. When I see hospitals claiming that residency is about education, not work, I laugh. Everyone knows that argument is bullshit.    

We can and should do a better job of teaching fundamentals, though I don’t see a lot of incentive to do so in formal settings. In most K – 12 public schools, after one to three years most teachers can’t effectively be fired, due to union rules, so the incentive to be good, let alone great, is weak. In universities, a lot of professors are, as I noted earlier, hired for research, not teaching. It’s possible that, as charter schools spread, we’ll see more experimentation and improvement at the ˚K –12 level. At the college and graduate school level, I’d love to see more efforts at instructional and institutional experimentation and diversity, but apart from the University of Austin, Minerva, the Thiel Fellowship, and a few other efforts, the teaching business is business-as-usual.

Moreover, there’s an important quirk of the college system: Congress and the Department of Education have outsourced the credentialing of colleges and universities to regional accreditation bodies. Harvard, for example, is accredited by “The New England Association of Schools and Colleges (NEASC).” But guess who makes up the regional accreditation bodies? Existing colleges and universities. How excited are existing colleges and universities to allow new competitors? Exactly. The term for this is “cartel.” This point is near top-of-mind because Marc Andreessen and Ben Horowitz emphasized it on their recent podcast regarding the crises of higher education. If you want a lot more, their podcast is good.

Unfortunately, my notions of what’s important in teaching don’t matter much any more because it’s unlikely I’ll ever teach again, given that I no longer have a tongue  and am consequently difficult to understand. I really liked (and still like!) teaching, but doing it as an adjunct making $3 – $4k / class has been unwise for many years and is even more unwise given how short time is for me right now. Plus, the likelihood of me living out the year is not high.   

In terms of trying to facilitate change and better practices, I also don’t know where, if at all, people teaching writing congregate online. Maybe they don’t congregate anywhere, so it’s hard to try and engage large numbers of instructors.

Tyler Cowen has a theory, expounded in various podcasts I’ve heard him on, that better teachers are really here to inspire students—which is true regarding both formal and informal education. Part of inspiration is, in my view, being able to rapidly traverse the knowledge space and figure out whatever the learner needs.

Until we perfect neural chips that can download the entirety of human knowledge to the fetal cortex while still in utero, no one springs from the womb knowing everything. In some areas you’ll always be a beginner. Competence, let alone mastery, starts with desire and basics.

If you’ve gotten this far, consider the Go Fund Me that’s funding ongoing care.


[1] Going to grad school in general is a bad idea; going in any humanities discipline is a horribly bad idea, but I did it, and am now a cautionary tale for having done it.

[2] Schools like Purdue also overwhelmingly select faculty on the basis of research and grantsmanship, not teaching, so it’s possible that the instructors don’t care at all. Not every researcher is a Feynman, to put it lightly.

How To Efficiently Advance Your Software Engineering Career [Guest Post]

This guest essay is by my former student Nathan Lin.

At age 30 I was able to retire from my software engineering career to pursue my interest in day trading. I had advanced rapidly as a software engineer, climbing through multiple promotion cycles. Combined with some good luck and timing, by my 30th birthday I had enough of a bankroll saved for the next 5 years.

I didn’t start my career in Big Tech, although I did end my almost decade-long career there. I’ve worked in various sized companies with as few as 50 engineers to over 30,000. What I discuss here should work regardless of the size of company.

I didn’t knowingly create this framework of expectation management at the time. Only by reflecting back on my career and identifying the recurring themes am I now able to put a name and a method to it. It’s largely inspired by Jake’s essay on How to get coaching, mentoring, and attention, which I read in college, and changed the trajectory of my academic and professional career.

My framework of expectation management is built on the same foundations, and is like a “Part 2” applied to software engineering. 

Cycles

A key part of the framework of expectation management requires you to understand cycles. Cycles continuously repeat themselves throughout your career, and it is imperative that you understand:

  1. What cycle are you at in your career? (The primary cycle)
  2. What cycle are you at in your specific job? (The secondary cycle)

Just as the seasons change and pass, so too does your location in the cycle change, and that location determines how you must manage expectation and performance. Note that the primary cycle also applies to other corporate 9 ­– 5 jobs:

  1. The novice (0-2 years)
  2. The intermediate (2-4 years)
  3. The expert (4-10 years)
  4. The guru (10+ years)

These cycles apply not only to you where you are at in your career, but also where you are at in your job. I discovered this first-hand as I followed the software engineering maxim of moving every 18 months to maximize compensation. There’s always a “rampup period” where, even though you may have a few years of industry experience, you’re given time to learn your new employer’s technology stack.

After that period, you’re expected to perform at a meaningful level. And after even more time, you’ll be expected to handle expert-level questions and solve tough problems. The cycle is compressed from years to months compared to the career cycle. I call this the secondary cycle:

  1. The novice (1-4 months)
  2. The intermediate (4-12 months)
  3. The expert (12 – 36 months)
  4. The guru (36+ months)

If you want to advance rapidly, you must know your location in both cycles, and destroy the expectations placed on you. The expectations are cycle-driven and tacitly acknowledged. The industry itself is aware of these cycles, as this is how roles and expectations are defined. However, these cycles are never really explicitly discussed, which is what makes exceeding expectations and rapidly advancing difficult. I’m trying to make tacit knowledge explicit. Most CS programs don’t include class called “the software engineer career,” but maybe they should.

The secondary cycle is reflexive, based on your location in the primary cycle. So, if you’re a novice who just graduated college as a 22-year old, you may be given 4 months at your first job to be brought up to speed. Some companies will even give slightly more. But as you progress to intermediate and expert, you’ll be expected to ramp up faster.

After all, if you’ve been in software for 8 years, why would it take you 4 months to ship your first meaningful pull request?

Expectations

Understanding the actual expectations is the last piece before getting to the framework. The framework itself is quite easy to apply once you understand cycles and expectations.

You must understand the cycle you’re in so that you figure out the expectations for both the primary and secondary cycles. They’re pretty much identical, so I’ll discuss the primary cycle:

  1. The novice – junior engineer
    1. Requires supervision and frequent check-ins
    1. Needs to be told what to do next
    1. Sees the 10 foot view
  2. The intermediate – midrange engineer
    1. Requires minimal supervision
    1. Capable of self-direction and identifying the next task in a larger project
    1. Sees the 100 – 1,000 foot view
  3. The expert – senior engineer
    1. Supervises intermediates and novices
    1. Plans entire projects to solve business needs
    1. Sees the 1,000 – 10,000 foot view
  4. The guru – staff/principal engineer
    1. Supervises experts, intermediates and novices
    1. Plans large-scale initiatives that break down into projects, coordinates with multiple teams to do so
    1. Sees the 10,000 – 100,000 foot view

Another way of viewing the cycle: at the novice level, you’ll always have someone you can ask for help. If you get stuck, ask your mentor/team lead/a senior engineer. But as you advance, you work on harder and harder problems and the number of people you can ask for help approach 0 at the expert and guru levels.

When you get to those levels, you are the one person you can approach for help, and you must ultimately figure out the answers by yourself. Note that you can (and should) consult other experts for ideas. Peers often spark new perspectives and ideas. In the end, however, you’ll have to do the heavy lifting yourself.  

The Framework

The framework is actually very straightforward:

  1. Find your location in the career cycle
  2. Find your location in the secondary cycle
  3. Exceed expectations

Now, you don’t necessarily have to exceed expectations. This really depends on your goals and how motivated you are. You don’t have to do it the way I did it, working 60+ hours a week to climb furiously. I had a goal in mind and I was determined to get after it. 

If you enjoy the 9-5 corporate life, there’s nothing wrong with meeting, instead of exceeding, expectations as the third step. The 9 – 5 corporate life can yield a great career, and if you know which companies have a healthy work life balance, you can have a very fulfilling job.

How do you exceed expectations? I only gave a high-level summary, so you may have some trouble with the details. But the rough idea can be found in Jake’s essay: prove to your team, especially your team lead and manager, that you’re here to get shit done. 

And the only way to prove that is to actually do it. This is where the rubber meets the road, and where the buck stops. Delivering results is the measure by which all top-level performers measure themselves and others by. So you must start speaking their language. The easiest way to do this is to deliver a small result as fast as possible, in your first week if you’re experienced, or your first month if you are a new graduate.

I’ve found fixing a low-risk bug, or improving test coverage especially reliable.

This is where the concept of cycles comes into play if you want to do this efficiently. You can quickly demolish expectations when you first arrive at a new job. Expectations are low, your team is giving you space to learn, so this is therefore the most logical place for you to make a first impression.

The beautiful thing is that this small result compounds. You’ve established a foothold in the codebase, and it’s now easier to work outwards from there. There may be some starting effort to establish the initial foothold, but it’ll pay off as the next pull request becomes easier. And as you continue to add code and make improvements, you’ll end up holding entire swaths of the codebase in your head.

Word on the team will travel fast. Good engineers are hard to find, and your mentor will mention your results to your manager. Your manager will begin to pay attention to your development. You may end up working on projects designed to promote you to the next level. And you may work with the heavy hitters on your team, and it won’t be a coincidence.

Skills compound over time. Mentorship opportunities open up to those most worthy of them. Life isn’t always fair, but over time, merit wins more often than not.

Promotions

If you want a promotion you must exceed expectations. At a certain point in time, you must begin signaling the attributes of the next rank above you, which requires two things. First, you must be proficient in your existing role, and second, you must begin to model the people on your team who are already at that next level.

Reverse engineer the next role, and ask yourself: “What does someone in that role do that I don’t?” So if you’re a junior engineer and you wish to become a midrange engineer, start proactively taking on tasks. Identify the next logical step to do without waiting for your mentor/team lead to give you the task. 

There’s an element of critical thinking here where you should ask, “What’s the next thing I should work on to efficiently move the project towards the finish line?”

Ship some code, then go back and refactor it, or move to the next task: “This is in a part of the codebase I was just in, I’m already familiar with these files.” 

Model the people who are already at the level you want to be at. If you think the way they think and code the way they code, then you’ll automatically begin operating at their level. They’ll see how you do it in code review and they’ll think, “Wow, she coded it the same way I would have. She’s pretty smart”. These people are also usually the ones your manager will solicit to determine if you’re ready. 

This brings me to my promotion rule: The more frequently you signal you are ready, the faster you’ll get promoted.

The Resting Rate

I haven’t talked about when to take a break. If you continue to smash expectations day after day, week after week, you’ll undoubtedly burn out (I speak from experience). When’s the ideal time to take a breather? There are actually a few places to strategically do so.

The first is after your team ships a major project. Everyone is happy: the project is complete, the deadline has been met. On most teams, the workload lightens substantially. Don’t do what I did and take on more work to keep padding your resume. There will be plenty of that down the line, so take a breather.

The second is after advancing through a phase in the cycle. But doesn’t this look bad if you get promoted and your work output drops? No, because never in my career have I seen someone demoted or put on a plan right after they were promoted. Your manager and skip level have gone through weeks/months of work to promote you, HR is involved with your compensation increase, so demotion or similar would look terrible on all parties involved. 

Now, I’m not saying to slack off and do nothing. But take a breather. If it doesn’t get done today, that’s okay. Your career will last for decades. This is something I did not really understand. 

I thought that in order to advance, I had to signal I was at the next step every single day. I suspect this is something that plagues a lot of other high performers in tech, causing burn out.

It’s true that the more you signal you’re at the next level, the faster you’ll get noticed and promoted. But there’s also a ceiling to the amount of signal you give off. There’s a point at which all the additional signal you produce will be ignored.

This is due to factors outside of your control, such as bureaucracy or company/HR policy. So you should really push as close as you can to this ceiling, and then no further. Additional effort at this point will be wasted. If you want to make it to staff/principal level, this is really important. You don’t want to be burned down to the ground by the time you make it. 

Why It Works

Several key quotes from Jake’s essay have stood the test of time in my career and life in general:

“…professors (and others with knowledge and competence) are most inclined to help people who won’t waste their time.”

“If a student really wants to learn, the professor will usually help, but most students don’t—so the professor builds a wall between herself and her students to make sure that the only students who breach the wall are the ones who do care about learning.”

“Secondarily, your professor will often recommend reading to test your seriousness…If you go away and don’t come back, you’ve demonstrated that you would’ve wasted her time had she spent an extra hour talking to you outside of class and office hours.”

This is why I recommend you immediately fix a small bug or improve code coverage. You’re going to be asked to do this anyway. Everyone is busy, and no one has time to sit you down (unless you are a recent grad) and walk you through the codebase. If you do this proactively, you’ll have taken one more item off their plate and shown, in a small way, that you’re worth their time.

The wall to getting a top performer to mentor you might even be more difficult to breach than a professor. Every minute of time spent talking to you is a minute she could be spending on advancing her career for a big raise or promotion.

You’ll often be shooed away with a simple bug fix. And if you come back asking trivial questions that can be answered by the README or a cursory glance at the code, their guard immediately goes up. How are you worth their time if you demonstrate you’re not capable of the basics?

But if you come back with a pull request, and a follow-up question about the architectural design of the system, now you’re showing you’re worth their time. You’ve produced the result and you’ve dug deep enough into the code to ask a harder question.

You have shown you are worth their time, maybe they could learn something from you. 

“People who really want to do something… do it. Or they make changes so they can… But most people say they want to do something and then they don’t. Over time, others notice this (like me), and they start to assume that most people who say they want to do or know something are full of shit.”

This is why you must signal you are at the next level for a promotion, and why you must signal often. 

Do you know how many times a year your manager gets asked for a promotion? It happens all at once, during the annual performance review, and it won’t get brought up again until the next year. Most people will get rejected and complain to their spouse, “Ah my damn manager doesn’t want to give me that promotion”.

But if you make it a recurring theme in your 1:1s, and you ask, “How am I progressing towards the next level?”, then the question sounds very different. Don’t just ask your manager to be put on a project. Find a problem that needs to be solved, come up with a plan of attack, and then build a proof of concept.

People react in one of two ways when expectation does not match reality. When reality is worse, they’re angry and disappointed. When it’s better, they are happy and excited. And that’s why expectation management works, because you’re demonstrating that reality is better. 

Completing The Cycle

I don’t want to drag this out and turn it into career management advice, so I’ll end on a few notes on advancing as efficiently as possible. 

First, you have to advance through the cycle no matter what. You can’t remain a novice forever or you’ll get fired. At best you’ll never reach your potential. If you’ve been in the industry for 10 years but produce the results of a novice, you’ll be viewed with suspicion because you’re breaking the expectation in the wrong direction.

Second, you don’t have to push as hard as I did if you don’t want to. The key is to work backwards and figure out your goals, then match your signaling frequency to that goal. 

There will be so many more projects down the line. Don’t tunnel vision on padding your resume and working 24/7. It’s okay to rest, and relax, because this is the natural cycle of things. 

Third, you have to know when, where, and how much to signal. 

When you are fighting to make it to expert status, you will need to signal more often. You have the energy and vitality to do so when you’re young. When you reach expert status, the game is more about conserving energy. You can decide whether or not you want to reach true guru status, or stay at your current level. Both are acceptable.

Thanks for reading. I hope this helps you, both in your career, and in your life, the way Jake’s essay helped me. 

Most people don’t read carefully or for comprehension

Dan Luu has a great Twitter thread about “how few bits of information it’s possible to reliably convey to a large number of people. When I was at MS, I remember initially being surprised at how unnuanced their communication was, but it really makes sense in hindsight” and he also says that he’s “noticed this problem with my blog as well. E.g., I have some posts saying BigCo $ is better than startup $ for p50 and maybe even p90 outcomes and that you should work at startups for reasons other than pay. People often read those posts as ‘you shouldn’t work at startups’.” In other words, many people are poor readers, although “hurried” or “inattentive” might be kinder word choices. His experiences, though, are congruent with mine: I’ve taught English to non-elite college students, off and on, since 2008; when I first started, I’d run classes by saying things like, “What do you all think of the reading? Any comments or questions?” I’d get some meandering responses, and maybe generate a discussion, but I often felt like the students were doing random free association, and it took me an embarrassingly long time to figure out why.

After a semester or two I began changing what I was doing. An essay like Neal Stephenson’s “Turn On, Tune In, Veg Out” is a good demonstration of why, and it’s on my mind because I taught it to students recently (you should probably read “Turn On, Tune In, Veg Out” first, because, if you don’t, the next three paragraphs won’t make a lot of sense—and you’re the kind of person who does the reading, right?). Instead of opening by asking “What do you think?”, I began class by asking, “What is the main point of ‘Turn On, Tune In, Veg Out?’” Inevitably, not all students would have done the reading, but, among those who had, almost none ever have, or give, good answers. Many get stuck on the distinction between “geeking out” and “vegging out,” even though that’s a subsidiary point. Some students haven’t seen or dislike Star Wars, and talk about their dislike, even though that’s not germane to understanding the essay.

Stephenson says at least three times that Star Wars functions a metaphor: once in the third paragraph, once in the second-to-last paragraph—although that technically compares the Jedi to scientists, rather than Star Wars as a whole to society—and again at the end (“If the ‘Star Wars’ movies are remembered a century from now, it’ll be because they are such exact parables for this state of affairs”). Most students don’t know what a “parable” is, which also means I wind up asking what they should do if they come across a word they don’t know. It’s also not like the essay is long or using numerous complex words: it’s only about 1,300 words and it’s about pop culture, not some abstract topic.

The first few times I taught “Turn On, Tune In, Veg Out” this way, I wondered if I was getting unrepresentative samples, but I’ve done it many times since and have consistently gotten the same results. I think most high-school students, to the extent they’re being taught to read effectively at all, are being taught to skim a work for keywords and then vomit up an emotional reaction (I assign free-form, pass-fail student journals, and most take this form). Very few students seem to be taught close reading, although when I was still in grad school, I had a cluster of students who all had had the same junior or senior year high school teacher, and that teacher had drilled all of them in close reading and essay writing—and they were all proficient. She seemed to be the exception, not the rule, and I meant to send her a letter thanking her but never did. Teaching “Turn On, Tune In, Veg Out” usually takes somewhere between 45 minutes and an hour, in order to go through it and look at how the essay is constructed, how the sentence “What gives?” functions as a turning point in it, and other related topics. I tell students at the end of the process that we’ve not talked about whether they like “Turn On, Tune In, Veg Out” or not; the goal is to understand it first, and evaluate it later. Understanding before judgment: Internet culture encourages precisely opposite values, as I’m sure we’ve all seen in social media like Twitter itself.

At the end of class, I ask again, “What is the main point?” and get much better answers. I’ll sometimes do the same thing with other argumentative essays, and often the initial answers aren’t great. I posit that most students aren’t being taught close reading in high school, and part of that theory comes from me asking them, individually, what their high school English classes were like. Many report “we watched a lot of movies” or “nothing.” Sure, a few students will have taken “nothing” from excellent classes and instructors, but the answers are too uncomfortably common, especially from diligent-seeming students, for me to not see the pattern. In high school, few students seem to have looked closely at the language of a given work and how language choices are used to construct a story or argument. To my mind, and in my experience, doing that is a prerequisite for being a proficient writer, including on topics related to “social justice.”

It’s not just “turn On, Tune In, Veg Out;” when I assign Orwell’s “Politics and the English Language,” I’ll get strange responses from students about how it’s so totally true that these days the English language is being used poorly. After enough of those kinds of responses, I began to open class by asking students to take 20 or 30 seconds to write down when “Politics” was written. In case you think this is a trick question, “1946” is displayed in huge font at the start of the version I’m using, and it’s repeated again at the very end. In the text itself, Orwell cites a Communist pamphlet, and he mentions the “Soviet press,” and such choices should be clues that it’s not contemporary. Nonetheless, if a third of a given class gets in the right ballpark—pretty much anything between “1930s” and “1950s” is adequate enough for these purposes—that’s good, which implies two-thirds of a given class hasn’t done the reading or hasn’t retained what I’d call an elemental idea from the reading. Students routinely guess “2010s” or “2000s.”

Right after college I taught the LSAT for two years, and the LSAT is largely a test of reading comprehension. I worked for an independent guy named Steven Klein, who’d started his company in the late ‘80s or early ’90s, before Kaplan and Princeton Review became test-prep behemoths. He and his business partner, Sandy, would marvel at the students who had 3.8, 3.9, sometimes 4.0 GPAs in fields like sociology, communication, English, or “Law, Society, and Justice” but who couldn’t seem to understand even simple prose passages. The students would get frustrated too: they were college grads or near college grads, who were used to being told they were great. The LSAT experience made me a sympathetic reader of the book Paying for the Party, or Beer and Circus: How Big-Time College Sports Has Crippled Undergraduate Education, both of which describe how most colleges and universities have evolved vast party tracks that require minimal skill development and mental acuity, but reliably deliver high grades. I think of those books when I read about the massive, $1 trillion and growing amount of outstanding student loan debt. Many college and university students would be better served with apprenticeships and vocational education, but as a society we’ve spent 40, if not more, years disparaging such paths and exalting “college.” Articles like “41% of Recent Grads Work in Jobs Not Requiring a Degree” are common. We have many bartenders and airline stewards and stewardesses and baristas who’ve obtained expensive degrees: I’m not opposed to any of those professions and respect all of them, but a four-year degree is a very expensive way of winding up in them.

The LSAT is a standardized test, and many schools still like standardized tests because those tests aren’t changed by how rich or connected or otherwise privileged a person is. Some Ivy-League and effectively-Ivy-League schools are doing away with the SAT, in the name of “diversity,” but that usually means they’re trying to give themselves even more discretion in “holistic” admissions, which tends to mean rich kids, with a smattering of diversity admits for political cover. “Race Quotas and Class Privilege at Harvard: Meme Wars: Who gets in, and why?” is one take on this topic, although numerous others can be found. The students who had gotten weak degrees and high GPAs were flummoxed by the LSAT; when they asked what they could do to improve their reading skills, Steven and Sandy often told them, “Read more, and read more sophisticated works. The Atlantic, The New Yorker [this was a while ago], better books, and do it daily.” I’d sometimes see their faces fall at the notion of having to read more: they were hoping to learn “One Weird Trick For Improving Your Reading Skills. You Won’t Guess What It Is!” When I’ve taught undergrads, they often want to know if there’s a way to get extra credit, and I tell them to do the reading thoroughly and write great essays, because I will grade based on improvement. This seems particularly important because many haven’t been taught close reading or sentence construction. I also see the disappointment in their faces and body language, because they think I’m going to tell them the secret, and instead I tell them there is no real secret, just execution and practice. A lot of school consists of jumping through somewhat ridiculous, but well-defined, hoops, and then being rewarded for it at the end, but real learning is much stranger and more tenuous than that. Sarah Constantin argues that “Humans Who Are Not Concentrating Are Not General Intelligences,” which is consistent with my experiences.

Many, if not most, English and writing professors also seem strangely uninterested in teaching writing or close reading. I get peculiar looks when I talk about the importance of either with other people teaching writing or English; one woman at a school I taught at in New York told me that social justice is the only appropriate theme for freshman writing courses. I know what she meant, and grunted noncommittally; I didn’t really reply to her at the time, although I was thinking: “Isn’t developing high levels of skill and proficiency the ultimate form of social justice?”

This is a long-winded way of saying that poor reading comprehension may be closer to the norm than the exception, and that may also be why, as Dan observed, very few bits of information trickle down from the C suites in big companies to the line workers (“I’ve seen quite a few people in upper management attempt to convey a mixed/nuanced message since my time at MS and I have yet to observe a case of this working in a major org at a large company (I have seen this work at a startup, but that’s a very different environment)”). I’d imagine the opposite is also true: if you’re a line worker, or lower-level management, it’s probably difficult or impossible to tell the C suite people about something you think important. Startups can disrupt big companies when a few people at the startup realize something important is happening, but the decision makers and the BigCo don’t.

I’ve also learned, regarding teaching, a message similar to what the MS VPs had learned: not much goes through, and repetition is key. One time, my sister watched me teach and said after, “You repeat yourself a lot.” I told her she was right, but that I’d learned to do so. Teachers and professors repeating themselves endlessly made me crazy when I was in school, but now I understand why they do it. I’ll routinely say “Do [stuff] for Thursday. Any questions?” and have someone immediately say: “What should we do for Thursday?” There’s a funny scene in the movie Zoolander in which the David Duchovny character explains to the Ben Stiller character how male models are being used to conduct political assassinations. He goes through his explanation, and then Zoolander goes: “But why male models?” The David Duchovny character replies: “Are you stupid? I just explained exactly that to you.” Derek Zoolander is a deliberately stupid character, but I think inattention is probably the most relevant explanation in the real world. Big tech companies like Microsoft probably have very few stupid people in them. Most students aren’t stupid, but I think many haven’t been effectively challenged or trained. It’s also harder for the instructor to teach close reading than it is to have meandering discussions about how a given work, which has probably been at best skimmed, makes students feel. I’ve written on “What incentivizes professors to grade honestly? Nothing.” There’s a phrase that floats around higher education about a rueful compact between students and teachers: “They pretend to learn, and we pretend to teach.” Students, I’m sure you’ll be shocked to know, really like to get good grades. I of course grade with scrupulous honesty and integrity 100% of the time, just like everyone else, but I have heard rumors that there’s temptation to give students what they want and collect positive evaluations, which are often used for hiring and tenure purposes.

Politicians appear to have learned the same thing about repetition and the limits of the channel: the more successful ones appear to develop a simple message, and often a simple phrase (“Hope and change,” to name a recent one: you can probably think of others) and repeat it endlessly, leaving the implementation details to staff, assuming the politician in question is elected.

When Paul Graham confronts readers mis-reading his work, he’ll often ask, “Can you point to a specific sentence where I state what you say I state?” It appears almost none do. Even otherwise sophisticated people will attribute views to him that he doesn’t hold and hasn’t stated, based on the mood his essay creates. In Jan. 2016, for example, he wrote “Economic Inequality: The Short Version” because he saw “some very adventurous interpretations” of the original. In April 2007, he wrote “Microsoft is Dead: The Cliffs Notes” because many interpreted his metaphor as being literal. I often teach a few of his essays, most notably “What You’ll Wish You’d Known,” and some students will report that he’s “arrogant” or “pretentious.” Maybe he is: I’ll ask a version of the question Graham does: “Can you cite a sentence that you find arrogant or pretentious?” Usually the answer is “no.” I tell students they could write an essay arguing that he is, using specific textual evidence, but that never happens.

I’ve told bits and pieces of this essay to friends in conversation, and they sometimes urge me to try and make a difference by making an effort to improve college teaching. I appreciate their encouragement, but I don’t run any writing or English departments and have a full-time job that occupies most of my time and attention. I like teaching, but teaching represents well under 10% of my total income, tenure-track jobs in humanities fields haven’t really existed since 2009, and adjunct gigs offer marginal pay. To really encourage better classroom teaching, schools would need to pay more and set up teaching systems for improving classroom teaching. The goal of the system is to propagate and perpetuate the system, not to disturb it in ways that would require more money or commitment. Pretending excellence is much easier than excellence. I’m okay with doing a bit of teaching on the side, because it’s fun and different from the kind of computer work I usually do, but I’m under no illusions that I’m capable of changing the system in any large-scale way. The writing I’ve done over the years about colleges and college teaching appears to have had an impact on the larger system that’s indistinguishable from zero.

SoHo Forum debate: “All government support of higher education should be abolished?”

Last night I went to a SoHo Forum debate on education, with Ed Glaeser supporting government-funded education and Bryan Caplan opposing; I already knew most of Caplan’s case, from reading The Case Against Education.

By far the most interesting piece of (then unknown to me) data came from Glaeser, citing a paper or set of papers the examine national income growth from 1960 to the present (or 2010 or thereabouts) that find education seems to explain income growth but income growth doesn’t seem to explain education. I didn’t catch the names of the authors, but that sounds like one of the better pieces of evidence against The Case Against Education. I’ve been following reviews of the book and so far the critical arguments haven’t been good; most have already been addressed in the book, and the authors just missed them—or their pro-education worldview prevents them from reading and understanding.

To be sure, I’m sympathetic to criticism of Case; having worked for a long time in education I want Case to be wrong. But I cannot find any good arguments against it, either on my own or that others have put forth. Many people don’t like abstract symbol manipulation, despite the way that particular skill is fetishized in the education system. At the very least, putting forth more intelligent apprenticeship options is a good place to start.

Furthermore, I’ve long complained to friends that most of school is tedious and boring. For a while I’ve thought it’s boring because of whining risk:

Almost no teacher gets in trouble for being boring, but a teacher can get in trouble or can get in trouble for being many values of “interesting.” Even I’ve had that problem, and I’m not sure I’m that interesting an instructor, and I teach college students. Students who complain about school being boring get told that school is supposed to be boring. Students who complain about school being interesting (or “offensive,” or whatever) get much more attention.

But if education is really about signaling regarding conformity and conscientiousness, then boredom almost becomes a feature, rather than a bug. If one is willing to conscientiously do even very boring work, that’s a great labor market signal. If Caplan is correct school has been boring and will continue to be boring because no force pushes it not to be, except perhaps for the occasional idealistic teacher.

Still, I have a pet theory that education may really be about very high achievement among elites (who become scientific or artistic innovators) more than about mass education, the intellectual results of which Caplan does show to be… dubious. I don’t know how to test this theory and it is not original to me, although it takes new salience in light of Case.

There’s an analogue to research here: most research is “wasteful,” but that’s because no one knows the answer till after it’s conducted; that’s why it’s called “research.” Most education may also be wasteful, but no one knows who is a waste to educate until it’s too late (up until I hit age 16 or so, I probably looked like a waste of scarce educational resources).

It’s always of interest to see someone in person who is only known on the page; I love, and often cite in propsals, Glaeser’s book The Triumph of the City, and in person he seems like one of those dapper titans of industry or now-extinct northeast, country-club Republicans of the ’50s.

Caplan posted his debate opening statement.

I didn’t think going in (and don’t think going out) that all forms of government support for education, higher or otherwise, ought to be abolished.

bryan-caplan-ed-glaeser

Why hasn’t someone tried to build or fund a very low-cost, very high-quality college?

As the title asks, why hasn’t someone tried to build or fund a very low-cost, very high-quality college? Or, if they have, what school is out there and has tried this?

It seems like a ripe strategy because virtually every (even slightly) selective school is pursing the same prestige strategy—high sticker prices, nominal discounts via “scholarships,” tenure-based faculty selection system, extensive administrative bloat, and so on. Yet even as schools relentlessly copy each other, news about outrageous student loan burdens is everywhere and probably affecting the choices made by students, and student openness to alternatives. At the same time, college tuition has been outpacing inflation for decades, and everyone knows it. Education is a component of the “cost disease” that is afflicting other sectors too. The number of college administrators has grown enormously (though that may not be the prime factor behind public-school cost increases). Students used to be possible to work summer jobs and graduate with little or no debt; schools in the 1960s or 1970s don’t appear to have been dramatically worse at education than schools today, and in some ways they may have been better, yet today colleges are many times more expensive. Schools trumpet their commitment to nominal education quality, the same way airlines trumpeted their commitment to passenger comfort, before deregulation forced airlines to compete on price and other metrics too, and anyone who’s been to a modern college knows that real commitment is “quality” is more rhetorical than real.

College costs and debts have soared, and at the same time the number of PhDs granted far outstrips the number of tenure-track or teaching jobs, so the workforce is available. Most universities and even many colleges care far more about research, much of which is bogus anyway, than teaching. Many universities don’t care about teaching at all, as long as the professor shows up to lecture, isn’t drunk, and doesn’t trade sex for grades. I hear many, many grad students and early professors lament the way their schools don’t care about teaching. There’s a surplus of cheap PhDs out there who’d desperately like to be professors. While professors who only teach two or three classes per semester complain relentlessly about all the “work” they supposedly have and how “busy” they allegedly are, it could be very easy to get professors to teach far more than they currently do at most schools, further reducing costs.

In short, the supply of faculty is there, and the supply of students ought to be there. With the setup above, let me repeat: why hasn’t anyone attempted to start a teaching-focused college with low tuition and extremely high-quality academics? I’m thinking of a school with a mandate to minimize the number of administrators, sports teams, and other boondoggles. One could even eliminate tenure, and thus ensure that PhDs hired today won’t still be on the payroll in 40 years. The school could highlight “proof of knowledge” over seat time as a metric; to my knowledge, there’s nothing intrinsic about four or five years of seat time. Students who study hard could and should spend less time in the seat.

Some of the situation I’m describing sounds like a community college, but I’m imagining a school that still draws from a national applicant pool and still maintains or attempts to maintain an elite or comprehensive academic character. Think of a liberal arts school but scaled up and with fewer administrators. If I were a billionaire I might try to do this; stupendously rich people loved endowing schools in the 19th Century, but that seems to have fallen out of fashion. Still, it worked then, so perhaps it could work now.

It may be that schools are really selling prestige and status, and consequently a low-cost, high-quality teaching school would be too low prestige and low status to attract students.

Still, and again as noted previously, pretty much every school, public or private, is pursuing the exact same prestige, admissions, and marketing strategy. With one or two exceptions (CalTech, University of Chicago—okay, there are a few others, but not many), they don’t even try (really or seriously) to distinguish themselves, and almost every school competes for the same BS college rankings. Such a uniform market seems ripe for alternate approaches, yet none are being tried or have taken off, though there are some small scale efforts, like Minerva.

What am I missing?

* Maybe it was easier to start colleges in the 19th Century, when regulation was nonexistent and complex subsidies of various kinds weren’t available. In the 19th Century, many colleges were also founded with the explicit intent of saving students’ souls, so perhaps the lack of religiosity in today’s billionaires and/or most of today’s students is a factor. God is an underappreciated component of many older endeavors.

* Current schools might just be too damn good at marketing for others to break in. Plus, college is a huge investment, which engenders a certain amount of conservatism in choices. Given costs, though, I think there’s room for experimentation here.

* Maybe there are efforts afoot and they’ve either failed or are too small for me to have noticed.

* Current schools are pursuing a complex price discrimination strategy, in which the sticker price is paid by a relatively small number of students, and much of the study body receives “scholarships” that are really tuition discounts. Maybe this system is more appealing to students and possibly schools than a transparent, everyone-pays-$5,000-per-year strategy.

* Students by and large pay with their parents’ money or pay with loans, so many an unbundled version of a school really is less attractive than one with lots of administrators, feel-good projects, fancy gyms, etc. Despite schools’ rhetoric to the contrary, they’re obsessed with attracting and retaining rich kids, so that market may be where all the juice is.

* Billionaires who might fund this are busy doing other things.

* The number of “good” or at least weird and different students who would try such a school is not great enough (given the current cost of college and the number of students out there, I find this one hard to believe, but it isn’t impossible).

I’m guessing number four is most likely, but maybe there are other features I’m missing.

Geek Heresy: Rescuing Social Change From the Cult of Technology — Kentaro Toyama

My review on Grant Writing Confidential is actually germane to readers of The Story’s Story, too, so I’ll start by directing you there. The book’s central and brilliant point is simple: for at least a century various people have imagined that better technology and the spread of technology will solve all sorts of social ills and improve all sorts of institutions, with education being perhaps the most obvious.

Geek_heresy2There are many other fascinating points—too many to describe all of them here. To take one, it’s often hard to balance short- and long-term wants. Many people want to write a novel but don’t want to write right now. Over time, that means the novel never gets written, because novels get written one sentence and one day at a time. Technology does not resolve this challenge. If anything, Internet access may make it worse. Many of us have faced an important long-term project only to diddle around on websites:

Short-term pleasure often leads to long-term dissatisfaction. That intuition underlies the psychologist’s distinction between hedonia and eudaimonia. Pleasure-seeking hedonism is questionable, but maybe long-term eudaimonic life satisfaction is good.

One sees these issues all over. Porn remains ridiculously popular (though some consumers of it are no doubt fine). Many people drink soda despite how incredibly detrimental soda is to health, and in my view how bad soda tastes compared to, say, ice cream. TV watching time is still insanely high, though it may be slightly down from its previous highs. There are various ways one can try to remove agency from the people watching porn while drinking soda and keeping one eye on a TV in the background, but the simpler solution is to look at people’s actions and see revealed preferences at work.

Most people don’t have the souls of artists and innovators trapped in average everyday lives. Most people want their sodas and breads and sugars and TV and SUVs and all the other things that elite media critics decry (often reasonable, in my view). Most people don’t connect what they’re doing right now to their long-term outcomes. Most people don’t want to be fat but the soda is right here. A lot of people want a better love life but in the meantime let’s check out Pornhub. Most people want amazing Silicon Valley tech jobs, but Netflix is here right now and Coursera seems far away.

And, to repeat myself, technology doesn’t fix any of that. As Toyama says of one project that gives computer access to children, “technology amplifies the children’s propensities. To be sure, children have a natural desire to learn and play and grow. But they also have a natural desire to distract themselves in less productive ways. Digital technology amplifies both of these appetites.” I had access to computers as a teenager. I wasted more time than I want to contemplate playing games on them, rather than building the precursors to Facebook. Large markets and social issues emerge from individual choices, and a lot of elite media types want to blame environment instead of individual. But each individual chooses computer games—or something real.

It turns out that “Low-cost technology is just not an effective way to fight inequality, because the digital divide is much more a symptom than a cause of other divides. Under the Law of Amplification, technology – even when it’s equally distributed – isn’t a bridge, but a jack. It widens existing disparities.” But those disparities emerge from individual behaviors. People who want to be writers need to write, now. People who want better partners or sex lives need to quit the sugar, now. One could pair any number of behaviors and outcomes in this style, and one could note that most people don’t do those things. The why seems obvious to me but maybe not to others. The people who become elite developers often say coding is fun for them in a way it apparently isn’t to others (including me). Writing is fun to me in a way it apparently isn’t to others. So I do a lot of it, less because it’s good for me than because it’s fun, for whatever temperamental reason. Root causes interest me, as they do many people with academic temperaments. Root causes don’t interest most people.

Let me speak to my own life. I’ve said variations on this before, but when I was an undergrad I remember how astounded some of my professors were when they’d recommend a book and I’d read it and then show up in office hours. I didn’t understand why they were astounded until I started teaching, and then I realized what most students are like and how different the elite thinkers and doers are from the average. And this is at pretty decent colleges and universities! I’m not even dealing with the people who never started.

Most of the techno-optimists, though—I used to be one—don’t realize the history of the promise of technology to solve problems:

As a computer scientist, my education included a lot of math and technology but little of the history or philosophy of my own field. This is a great flaw of most science and engineering curricula. We’re obsessed with what works today, and what might be tomorrow, but we learn little about what came before.

Yet technology doesn’t provide motivation. It’s easy to forget this. Still, I wonder if giving 100 computers to 100 kids might be useful because one of them will turn out to be very important. The idea that a small number of people drive almost all human progress is underrated. In The Enlightened Economy Joel Mokyr observes that the Industrial Revolution may actually have been driven primarily by ten to thirty thousand people. That’s a small number and a small enough number that the addition to or subtraction of a single individual from the network may have serious consequences.

This isn’t an idea that I necessarily buy but it is one I find intriguing and possibly applicable to a large number of domains. Toyama’s work may reinforce it.

When there are too many administrators, which ones do *you* fire?

You know there are too many administrators when even The Nation argues there are too many administrators.* More importantly, though, everyone regardless of political bent is against “administrators” in the abstract but almost no one lists which administrators should be on the chopping block. Too few articles and polemicists say, “These are the 100 positions I’d eliminate at the University of Washington.” If a school decided to fire its “Diversity” department in the name of cost cutting, The Nation would be the first publication screaming about racism and institutional indifference and the betrayal of high-need populations. Everyone rails about administrators, but no one has concrete plans to halt their proliferation.

Consider UC-Berkeley’s “Vice Chancellor’s Office for Equity & Inclusion;” perhaps UC-Berkeley doesn’t need seven “equity and inclusion” teams or 17 employees in the Vice Chancellor’s Office for Equity & Inclusion.** The staff includes several financial analysts and a graphic designer exclusive to that office. California’s public salary database shows that that graphic designer earned $75,800 in 2014. The Development Director earns $109,000. The Executive Assistant earns $91,400. The Vice Chancellor for Equity and Inclusion earns $209,000 a year. And so on. But UC-Berkeley will probably never cut this department (maybe that’s a Good Thing).

One sees this elsewhere. At Marymount Manhattan College, last week I got an email about a “Change of Title IX Coordinator.” That’s another part of one administrator’s job that didn’t exist decades ago. In addition, the email says the school “undertook an assessment of how best to comply with evolving federal and state legislation.” Which is another way of saying, “We spent a bunch of time and man hours.” Followed, since this is a large, modern organization, by numerous email followups. There were also “mandated student, faculty, and staff trainings” (emphasis added). Maybe that work is good and maybe it isn’t, but it’s still indicative of the time and energy and activities that otherwise hated administrators are doing.

(Title IX, by the way, is the subject of Laura Kipnis’s hilarious, expensive Title IX inquisition. I wouldn’t blame you if you left this somewhat dry article to read her funnier, ribald essay.)

I don’t want to pick on any particular school or even the education industry specifically. Regulatory compliance costs are increasing in virtually all industries, including the financial industry (link goes to a PDF) and many others. We rarely consider the systematic effects of regulatory compliance and instead think of each particular regulation / requirement in isolation. Nonetheless, when we get a lot of regulatory and other mandatory or optional costs together, we see the need for more lawyers, bureaucrats, administrators, and other people who all need to be paid and who have to be at least somewhat good at abstract thinking, writing, and statistics.

To be sure, the presidents and so forth making $500,000 or more per year is obscene on its face, but those are a relatively small number of positions, and, while I agree that college presidents should behave more like part of the university and less like corporate titans, I’m not sure that a small number of overly paid people is the biggest problem. I am sure that the next time I see someone announcing that we need to first fire all the administrators I’ll send them this post and get nothing in response.


* But here’s one, alternate explanation.

** Much of this post and its research came from a friend, who gave me permission to publish it.

Why I don’t donate to Clark University, and thoughts on the future of college

I went to Clark University, and a couple weeks ago I talked to someone from their “development” department (read: they ask alumni for money) about what I’d been up to, what I thought about Clark, and then, finally, in the “Will-she-sleep-with-me” moment, whether I’d give more than $10 a year. I won’t. Even if I magically made Zuckerbergian billions, I wouldn’t give much more because while Clark is a good school, it isn’t in a position to solve the most pressing problem(s) in higher education: cost and access. Clark can be a wonderful and amazing experience for individual students but it will never be widely accessible due to cost and its model is not replicable for the same reason; the major problems in education are cost and access, which I’ll return to below.

Right now I give a little cash because of bogus rankings like those by U.S. News and World Report; here’s a good piece by Malcolm Gladwell on their bogosity. Nonetheless, despite them being bogus, people love rankings—even very bad rankings. When I was in high school, someone—the villain U.S. News again, maybe—ranked high schools simply by the number of students divided by the number of AP tests (or vice-versa). My high school came out well in that regard and parents and administrators and even the students themselves (to some extent) ran around saying “Oh wow we go to one of the best high schools in America!!” Which was bullshit to anyone who stopped to think for 30 seconds, but the meme propagated anyway and the number of people infected with the counter-meme (“Most school rankings are bullshit”) was and is much smaller than the number with the first meme.*

Maybe nothing short of a cultural change in views on college can alleviate the obsession-with-ranking problem. Some of that cultural change may be in the air: here’s one of the articles about Google’s decreased emphasis on college degrees. Maybe more firms will move in this direction. Certainly I would be more interested in assessing someone’s blog, books, or other material in hiring them than their degree. I’ve met a lot of PhDs who are morons. That is not to deny the value of education—it is easier and more pleasant for most people to learn in the context of someone who can select material, judge material, and accelerate learning. But too few teachers seem able or willing to do that. Alternate signals may emerge.

To look at one alternative to the present education system consider Western Governors University. This is one article on WGU, though there are many others. As I mentioned in the first paragraph, the major problems in contemporary higher ed emerge from rising costs, Baumol’s Cost Disease, weird cross subsidies, and related factors. Tyler Cowen’s book The Great Stagnation is good on these subjects. I obviously like and generally support Clark but I don’t think the school is the answer to the biggest problems in higher ed today. There may not be one single answer. We may be seeing the researcher-teacher hybrid model splitting back into their constituent pats as well, since, as has long been observed, someone very good at one may not be good at the other.

The “teacher” point is important too, because teaching well is expensive and difficult. It’s not clear to me that the current structure of higher education is sustainable regarding teaching. Here is one well-written and half-right, half-wrong piece about how “Teaching Is Not a Business.” In some sense everything is a business whether we want it to be or not.

Saying that teaching is not a business is another way of saying, “We can pour an infinite amount of money into this endeavor without asking what we’re getting it.” There is a magic to teaching and I’m susceptible to that feeling, but teaching is also a system and set of institutions and many other things as well. Not surprisingly most members of the guild want to retain the mystique and a lot of outsiders appalled at rising costs want to de-mystify and improve. The overall trajectory of the last two or three hundred years makes me think the latter are eventually going to win, even if the definition of winning changes and the win takes decades to play out.

This is getting far afield from the point about donating to Clark, but the biggest issue is that I don’t see how most of the current version of higher ed is rewarding teaching adequately. Some like “The Minerva Project” may be the answer. It and Western Governors University are both very consciously doing a lot of things very differently than the standard college model, which Clark follows in important ways. Clark has a high cost structure and can’t avoid that. As I said above it is a good school. If I had a kid and could afford to send them I would.

But how much does Clark cost?

Somewhere within Clark, someone has the minimum number of dollars per student the school must take in in order to stay afloat. If I had to guess, I’d guess that number is between $25,000 and $30,000, and Clark must hit it whether Joe pays $15,000 and Jane pays $40,000 or vice-versa. Every college has this number somewhere. For a few schools it’s probably zero, counting endowments. Until we get more clarity about that number, however, it’s hard to get a meaningful value for it.

This began life as an e-mail to the Clark development person. Most of the answers she gets are probably more emotional than my somewhat cerebral / systems-based thinking, but part of my dissertation is about academia and I’ve now worked in, around, and for a lot of colleges, as a student, instructor, and consultant. The inside of the sausage factory is not a pretty place and the romantic notions I may have once had regarding the college experience are now dashed. I still retain hope and even optimism—I would be teaching as an adjunct this semester if I didn’t—but the ugly reality is that relatively few existing institutions have the structure or infrastructure, literally or intellectually or politically, necessary to make real changes. Whatever spare cash I might have one day—ha!—is unlikely to go to existing providers. It’ll go to whoever is trying to augment or replace them. Right now I don’t know who that is.

It’s not you, Clark. It’s it.**


* These sorts of idiocies persist. When I was in grad school, some girl in the University of Arizona’s Rhet Comp (or “Rhetoric and Composition”) program claimed that they were “number two in the country.” Being the obnoxious person I am I asked, “As ranked by who?” She didn’t know. “As measured how?” She didn’t know and didn’t like me. To be fair I thought she was dumb and didn’t see her manifesting evidence to the contrary while I was around.

** See also “Ten Ways Colleges Work You Over;” I doubt any individuals at Clark approve of the competitive college race, but they are also relatively powerless to stop it.

What is college for? Matt Reed’s hypothetical and following the money

Matt Reed’s post “Parity” asks this, partially as a thought experiment and partially as a proposal: “What if every sector of higher education received the same per-student funding? Right now, the more affluent the student body, the more public aid money the sector receives.” He’s right. He goes on to say, “From a social-justice perspective, that’s counterintuitive.” He’s right about that too, and he eventually asks: “What is the argument for spending the most on those who have the most?”

I can’t guarantee this is the argument—and indeed there may not be one, since the higher-education system evolved by accident rather than being planned by design—but one possible answer is that the current system evolved primarily to subsidize and conduct research. If the purpose of the fiscal structure of universities attempts to maximize research rather than social justice, then it may make sense to spend the most money on universities and programs that produce a lot of research. That obviously isn’t community colleges, whatever their other merits.

The idea that universities are primarily about social justice seems to have come along later than the idea of universities as research labs. In the U.S. at least, universities have had a couple major phases: first primarily as seminaries for the clergy; then as finishing schools for the wealthy, which usually coexisted with ways of spreading knowledge about agriculture and teaching; then, during and after World War II, as research hubs; and in the last couple decades as ways of rectifying real or perceived inequality. Reed’s third paragraph starts with “From a social-justice perspective,” and that may not be the dominant perspective among legislators, whether state or national. Certainly during much of the Cold War period from 1945 – 1975, when money poured into universities per Louis Menand’s The Marketplace of Ideas, it wasn’t.

My dissertation is on academic novels and I’ve now read a huge amount of material related to the conception of universities from 1945 – the present. One persistent theme is that intelligent people in every era disagree both what universities as a whole are for and quite often on the discipline or department level what each discipline or department is for. In this respect Reed’s post is a continuation of this discussion.

My favorite answer about the question of what universities for has been attributed to various people, and here is one rendition: “a university is a happy place if the administration provides football for the alumni, parking for the faculty, and sex for the students.” Incidentally, in all three regards and certainly for the first and last, flagship public universities far outperform their Ivy League peers. It’s nice to be number one in some domains. Murray Sperber’s Beer & Circus argues that sports and sex have been central preoccupations for a very long time; perhaps nerds like me have the wrong perspective.

I wish I had a neat transition into this point, but I don’t while still thinking it important to note: tne problem or virtue with universities comes from the way all sorts of weird cross subsidies happen at all kinds of levels, to the point that I’m not sure it’s possible to disentangle what’s happening fiscally.

EDIT: Malcolm Gladwell’s article “The Order of Things, about the impossibility of ranking heterogeneous colleges in a fair or objective way, is also relevant here:

The U.S. News rankings turn out to be full of these kinds of implicit ideological choices. [. . .] There is no right answer to how much weight a ranking system should give to these two competing values. It’s a matter of which educational model you value more—and here, once again, U.S. News makes its position clear.

I admire Reed for raising the question. But it’s also important to recognize the priorities any division of resources like the one among colleges entails.

Why can’t we solve poverty, or solve it through schools?

I’m not that old, and I’ve already seen a lot of proposals for solving “poverty” come and go. Many—think Head Start—are tied up in education. The current debate around education tends to run in two directions: one group wants to improve parenting, or ameliorate poverty, or something along those lines, having seen innumerable correlative studies demonstrating that rich kids on average do better than poor kids at school. The other group—the one I belong to—tends to think that we could do a lot for schools, and especially big urban schools, through some combination of charters, vouchers, and/or weakening the power of teachers’s unions. For more on why the latter group thinks as we do, see the many links in this post.

The first group—the one that wants to attack poverty and what not—tends to say things like cjensen’s: “Statistical studies have long shown that (1) education outcomes strongly correlate with parenting,” to which I replied:

Citations are needed on this: “Statistical studies have long shown that…”

“We”—schools, society, etc.—can’t really control parenting. But we can control schools, and it is probably possible to get substantially better outcomes than the ones we’re getting now, chiefly through better teachers. At the moment, most public school teachers are paid in lockstep based on seniority—CS teachers and PE teachers get the same pay—and can’t be fired after their second or third year of teaching, and that creates a lot of perverse incentives.

Ceras replied with another fairly common sentiment: “Programs exist for this with some positive results. Here’s one from a quick Google search,” and he linked to “Nurse-Family Partnership – Top Tier.”

But innumerable small-scale programs that show limited positive results, but almost none of them scale up, for the reasons Megan McArdle describes at the link:

That pilot program has a huge administrative staff whose sole incentive is to ensure that it is meticulously carried out. In the real world, that curriculum will be put into place by an administrator whose priority list is crowded with everything from mollifying the latest lunatic on the school board[. . . ]

That pilot program is staffed with a narrow band of extremely highly qualified teachers, sifted from the best the environment has to offer. In the real world, whoever happens to be standing in front of the classroom come September 5th has to do it, even if they flunked Remedial Math four times and only got this job because the school board needed a body.

McArdle’s book The Up Side of Down is also good on this subject. Lots of small-scale Head Start programs show promise too, but the program’s effects fade out after a couple years, and on a large scale it hasn’t done anything except provide daycare and jobs. Despite the 40-year failure of Head Start to do what it was intended to do—improve life outcomes for poor, minority kids—there’s a press for it in liberal cities, only now it goes by the phrase “Universal Pre-Kindergarten” (UPK). New York City has a UPK program. Seattle mayor Ed Murray wants one, and he wants to spend a lot of money creating it.

OLYMPUS DIGITAL CAMERALet me return to Ceras’s example. Programs like “Nurse-Family Partnership – Top Tier” (NFP) already operate. I know because I’ve written numerous Health Resources and Services Administration (HRSA) Healthy Start Initiative (HSI) proposals that attempt to do just what NFP proposes. For my real, work-for-money job, I do grant writing for nonprofit and public agencies, so I see citations like the one to NFP all the time. Next time I write an HSI or similar program, I might cite NFP. Doing so isn’t going to make the specific program any better—HSI has been operating for a couple decades, under different names, and hasn’t accomplished much on a large scale, in part because of the scale-up problems described a few paragraphs ago.

Ideas like NFP sound good in the abstract, but the gap between the real world and the proposal world is quite wide. Virtually every idea for improving health, welfare, and education has been funded through some grant program or another, but most people proposing new programs aren’t aware of the old ones—and they aren’t aware of the gap between the real and proposal world. After his $100 million donation to the Newark Public Schools, Zuckerberg has evidently learned this.

So what can “we” do? The people who want to keep the existing structure of education in place usually say they want to fight poverty first. On some level who doesn’t? There are some challenges, however. Poverty is a moving target. It’s usually calculated as a percentage of income, which means that it will always be with us (barring some unforeseen technology, or extinction). In addition, from the perspective of someone in 1800 or 1700 or really anytime before about 1950, we have solved poverty, at least in a material sense. Virtually no one in the United States lacks running water, plumbing, or refrigeration. Almost no one starves to death, and the real problem among the poor is obesity. TV penetration is hovering around 98% of households, and the households without TVs are more likely to be like mine—that is, relatively well-off people who choose not to have a TV.

I’m not saying it’s great to be poor in the U.S., but it’s still better to be poor in the U.S. than to be poor in, say, Nigeria, or Brazil. Globally, there have been innumerable people trying to improve life in the developing world, and many books about why those efforts haven’t been totally effective: Why Nations Fail is good. Dead Aid is good. There are others; you’ll see them at the Amazon links. Developmental economics is an entire field devoted to this question. There aren’t easy answers, because if there were, they already would’ve been found and implemented. To quote Megan McArdle again, “The very existence of a policy issue tells you that it is difficult to solve, either politically or technically.”

Beyond measurement and definitional issues around what one means by “poverty,” consider the history of fighting it. Johnson launched the “War on Poverty” 50 years ago, and even the New York Times (at the link) calls it “a mixed bag,” which sounds charitable to me. There is a large poverty-fighting infrastructure that does some really good things (like Food Stamps, now called TANF), and some less good things. Nonetheless, if poverty could be “fought” successfully, I think it would have already been defeated. That it hasn’t should make us question our approach.

There has also been some regression in terms of culture and behaviors: that’s one important message of Charles Murray’s book Coming Apart: The State of White America, 1960-2010. Most women, for example, are better off having children with a dedicated and ideally married partner, but around 40% of all births are currently to unmarried women. There’s a political argument about why that is and what if anything should be done about it, but the behavioral and sociological changes of the last 50 years are still real.

This has a lot to do with education because, as I noted in the first paragraph, people who are relatively okay with the educational status quo tend to want to address things outside of school first. Diane Ravitch is a great leader for this group. I’ve read two of Ravitch’s books on education—Left Back: A Century of Battles over School Reform and The Death and Life of the Great American School System: How Testing and Choice Are Undermining Education—and to read her work is to respect her knowledge and erudition. She moved from a strong educational reformer who favored charter schools to someone who… I don’t know how to characterize her current position other than to say she doesn’t favor charters or vouchers. She does observe the many ways particular charter schools haven’t done very well, but in my view they haven’t been worse than the urban schools they competed with, and some have done much better.

Overall, Ravitch wants to reduce poverty, but as noted above I’m skeptical of social or government forces to do so. In Reign of Error, her most recent book—I’m not all the way through it—she says that public schools are better than they’re commonly depicted. She’s somewhat right: relatively wealthy suburban schools are okay. But that pretty much leaves urban schools (L.A., Chicago, New York, Newark) to languish, and those are the areas and schools that are most promising for vouchers.

The final thing I’ll note is that a lot of people favor “more” money for schools. Overall, inflation-adjusted funding has roughly doubled on a per-pupil basis, per the New Yorker article, and overall funding is quite high—including in screwed up districts like Washington D.C.’s. The Great Stagnation also discusses this dynamic. So while “more” money for school districts may or may not be a good thing, it’s apparent that more money does not automatically lead to better results.

This has turned into a much longer post than I meant it to be, but, to reiterate a point made above, there are no simple answers. Though this post is long it is shorter than many of the books it cites, and it is much shorter and more fun to read than many of the proposals I’ve written. The number of people who are genuinely interested in this kind of social policy minutia is probably small, as the popular support for programs like UPK shows.