Human Capital is Real, and Some People Are Smarter Than Other People
until we acknowledge that, there can be no coherent discussion of education
Years ago I pointed out that a New York Times article didn’t make much sense. The article was about school choice within the New York City public school system. Like pretty much every experiment in school choice, the results have been underwhelming. More importantly here, the article reflected a simply incoherent conception of what we want from schools in the first place. Consider two excerpts:
Sean P. Corcoran, an associate professor of economics and education policy at New York University, has researched the choice process and how students match. He said that the best option is for students to reach for the best possible school for which they are qualified, and indeed, most students get one of their top choices. But in many cases, students reach either too low or too high.
“The average kid has to be able to get a good education, because most people are average,” Ms. Lewis said. “It’s great that the highfliers are succeeding, and they deserve the chance to succeed. But so do the average kids.”
So we have an implicit admission here, of a sort, that there are some kids that are more academically talented than others. Right? The concept of reaching for a school that’s “too low or too high” is an interesting manner of saying this; it’s a suggestion that there are schools that are better suited for students who are less talented and schools that are better suited for students who are more talented. But how is school quality defined? It’s defined according to the summative outcomes of how students perform. I’ve been arguing for a decade that perceived school quality is simply a product of the selection effects of how a school’s population was assembled. We also are presented with the idea of a good education for average kids. But what is the definition of a good education, conventionally, if it does not refer to one which produces above-average results? How would we know that an average kid really had received a good education? If we acknowledge that not all kids can be stars, how can we define success simply in terms of the best possible outcomes across populations? Despite our endless education debates, these elementary issues remain undertheorized.
The same article laments that some students “fall through the cracks,” which means that some will struggle. But if we acknowledge a range of student ability, and we also think that students should be sorted into schools based on their ability, what does it mean for a student at the bottom of the distribution to fall through the cracks? If we acknowledge that there is a bottom of the distribution, how can we insist that the existence of people at that bottom is an indictment of the system or any constituent part thereof?
I think about this all in light of this Michael Powell story about the heavily-Asian populations of New York’s elite public high schools. There are the usual omissions in discussing entrance examinations; for example, Powell is the latest to note concerns about people getting tutoring for the tests without mentioning that grades, which replace the tests, are in fact more often and more deeply affected by tutoring. More frustratingly, his article includes the most common element of all of these discussions, which is placing blame earlier in the educational chain.
Again and again, the conversation returned to the broader problem. The elementary and middle schools must prepare more students to compete at the highest level.
This is standard fare; when I was working in CUNY, of course, the buck was frequently passed down to the high schools. It's never quite coherent, though. For one thing, students show up to kindergarten with profoundly differing levels of ability. But aside from that, what confuses me is that Powell’s article talks plainly about some students being better at school than others; the exam schools are discussed as being for and producing an academic elite. But if differences in ability are not a priori a problem in high school, why would they be in elementary and middle schools? If you want academic excellence, shouldn’t you expect a distribution - that is to say, a range from very good to very bad - at every stage of development? And what can “the highest level” mean if we dramatically expand the number of people within it?
Like I said. We produce oceans of commentary on education in this country, and yet the most elementary ideas about what we want and how we define success go missing. I think ultimately it’s because most people understand intuitively that some students are more naturally gifted at school than others, but this idea is politically untenable and impolite.
As is his habit, Eric Adams said the quiet part out loud awhile ago, when he talked about low skill vs. high skill workers. And there was a whole bullshit news cycle where people pretended that they didn’t know that the market values some skills far more than others, and that these different valuations are derived in large part from how hard they are to acquire. I think everybody deserves material security and comfort and I want a vastly more socially and economically equal system than we’ve got. But why on earth would I pretend that an aerospace engineer’s talents are as easy to come by or acquire as that of a skilled barista? I want to fight for equality in full view of reality, please. I suspect that most everyone knows that some jobs are realistic occupations for a small number of people and thus more remunerative, but feels it’s impolite to say so. Adams’s real sin was saying out loud what polite liberal society has decided should be understood but not said.
You already know the outline of my thoughts: most people settle into a band of academic achievement very early in life and stay in that same relative band throughout schooling. Data gathered early in K-12 schooling provides strong predictive information about how people will perform in college. And given the large and growing corpus of population genomics research, the most parsimonious explanation is that some kids are more genetically inclined towards academic success than others, while systematic differences in environment contribute to some of the variation. But even setting aside that potential explanation, one thing we know about education is that people within it are unequal in their abilities, such inequality has existed since the dawn of schooling, and we have no tools that can dramatically change the relative performance of students at scale, despite ungodly amounts of time and treasure being spent in that endeavor and a never-ending media hype cycle about the next big thing.
When I set out to write my book, I knew the idea of intrinsic or inherent academic talent, an innate predisposition to succeed or fail, would be controversial, and was prepared for that controversy. The repeated reassurances that the book rejected race science, which annoyed some readers so deeply, were in part an attempt to ward off deliberate misunderstandings of what I was saying. (That is, that individual talents can vary thanks to genetics without that implying that group differences are genetic.) What I was consistently surprised by, though, was the number of people who responded to my book by insisting that there is no such thing as a summative difference in intelligence or academic ability - that is, that not only are there no inherent predispositions towards being good or bad at school, no one even becomes better or worse, no one is smarter than another. There are no measurable differences in what we know or can do intellectually. Or, in some tellings, no one knows what smart is, it’s some sort of ineffable quality we can’t pin down, or the very idea of “smart” is a racist Western imperialist hegemonic heteronormative con.
I find this all unhelpful. Narrow down as specifically as you can and no one can persist in denying that there are differences in summative ability. Can anyone really claim that I can do calculus as well as a math professor who teaches it? Because I can’t do calculus at all! Of course people have different things they know she understand and can do intellectually. I’m not naturally talented at math. I don't like it but it's true. And easily quantifiable. If there were no such distinctions school would not exist.
But the will to obscure this fact is strong. In many fields, the academics at the top are busily abstracting and mystifying success, the better to insist that no one is bad at what you study. (My old field, writing studies, is filled with academics who believe there is no such thing as being better or worse at writing, which makes you wonder why anyone is paying their salaries.) Every day academics declare that grade are a capitalist plot, tests evil, and the very idea of assessment offensive. But there really are things that you can know and not know in life, and some of them, such as reading, are really important. And in fact we are very good indeed at creating instruments that measure whether you can read or write or do algebra. It’s just that their results are socially inconvenient.
If the concern is saying that there are attributes and abilities in life that matter that are not academic or connected to intelligence, and that they should be taken seriously and rewarded, the news is good, as this is perhaps the core argument of my book. If the concern is saying that being smart is an unhealthy obsession in our society and too essential to having material security, the news is good, as that is perhaps the other way to state the core argument of my book. But I don’t understand why we would pretend that academic or intellectual ability doesn’t exist, and act as though that attitude is a prerequisite to be a progressive person who desires equality of rights, dignity, and human value. As I never get tired of pointing out, traditional left thinkers like Marx never pretended that all of us are equal in our abilities. (“From each according to his abilities” implies the opposite!) What the left pushes for is equality of human value, including across - perhaps especially across - differences in talent. Equal value, equal dignity, and equal right to demand the minimum conditions needed for human flourishing.
We can lawyer about the concept of intrinsic ability as much as we want. (For the record, acknowledging that genes and environment both play important roles in education, and that there are complex interactions between them, does not imply that outcomes are therefore mutable.) We live in a world where some people can do things, intellectually, that are monetarily rewarded and socially valuable, and some people can’t. Our attempts to spread these abilities universally have been an abject failure. Because each of us has a nature, and while we’re all good at something, we’re not all good at the same things, and capitalism most certainly does not reward all gifts equally, and so much the worse for us. (Indeed, this is the very reason redistribution is necessary.) Yes, intelligence is multivariate and complex and exists in many dimensions. But so is love, and no one pretends that love therefore does not exist. We are already asking the impossible of our education system, expecting it to reward excellence and create equality at the same time. Let’s not burden it even further by pretending we don’t know some people are better and some at worse at school.
In grad school, while I was most despondent over my break from my field and my growing understanding that I would never really be a part of it, I got into some textual processing stuff. I would take big corpuses like the ICNALE and use various kludgy programs that researchers had put together and mine those data sets for information on (say) the diversity of vocabulary in a given text and the score it received on a standardized test. I was operating pretty much entirely on my own and not doing anything particularly interesting to people who really know this stuff. I did eventually publish a paper trying to validate a theoretical equivalence between two lexical diversity measures empirically; it was interesting but ultimately a bit of academic ambulance chasing. At some point I realized that I couldn’t do what I wanted to do with other people’s stuff. I needed to code myself. Everybody I knew in the relevant fields said Python was the way to go. So I got a book and a bunch of files off the internet and I got to work. I really wanted to do this and so I really invested the time.
And the punch line, which I’m sure you’ve predicted, is that I sucked at it. Just sucked massively. Everything took me forever. I’d do the little exercises, the programs wouldn’t work, I’d blearily stare at the code, I’d find what was wrong, I’d fix it and smile confidently, then I’d run it again and it would still be wrong. And I did it for months. Months. I never got better. I would eventually get through the exercises, but I was clearly taking far longer than was intended, and I was getting nowhere close to my actual goals. And late one night, after several months of trying, and firmly believing that the only way to salvage an academic career was to get good at this thing, I looked out the window and I said, “I can’t do this. I’m not good at this. I’ll never be good at this.”
And I was right.
My mom was a teacher and a great one but she always seemed to believe, in a flower child kind of way, that we could all be doctors or engineers if we wanted to. I think I mostly believed her until my husband and I adopted our children. Want some back up for the power of genetics? Ask an adoptive parent. Both of our kids were adopted as infants. One is social-butterfly artist that everyone likes but couldn’t make change to save her life. The other one scores in the top percentiles on everything academic and has the social skills of a lovable brick wall. I think our emphasis on education and our financial ability to give them opportunities has definitely enhanced their lives, but we have had zero to do with their skill sets.
Along the same lines, not sure if I’ve missed it but would love to read your thoughts on gender differences in education and the decreasing numbers of boys succeeding in high schools and choosing to attend college. Touring colleges with our daughter this has become really evident.
> why on earth would I pretend that an aerospace engineer’s talents are as easy to come by or acquire as that of a skilled barista?
Thank you for helping me to articulate why this viral tweet (21k likes) has been irritating the shit out of me:
"If you have a robust skincare routine you have demonstrated aptitude in many core skills of data science. I will not be taking criticism of this idea. Girls who have developed personalized skincare routines know more about multivariate causal inference than many engineers."
I want young girls to feel good about themselves. But we don't need to tell them that shopping at Sephora means they're data scientists.