Mmmm, Bacon

I think my dialogue with Sam is beginning to reach the point of diminishing returns, so here I will just assert some things, correct others, and so forth. There is nothing systematic here.

That being said…

What Is The Scientific Method? 

Sam asks for clarification about the scientific method. It is true that most academic establishments dedicated to the “science” of education do not do science at all. Given that this is the circles I understand Sam to have been moving in, I certainly sympathize with his cynicism about “science.” It is a sad and startling fact of our contemporary world that most educated people, including journalists and scientists, simply don’t understand the basics of epistemology. If they had, we wouldn’t have had the New Atheists and writing 98% of articles with a headline that include the words “Study Says” would get you fired.

When I refer to “science” I mean Baconian science (Francis Bacon, don’t worry about the pictures), i.e. what is most commonly referred to as the scientific method.

The scientific method is a means for producing reliable, non-obvious predictions about the future through controlled experimentation.

For roughly two thousand years, “science” meant what Aristotle had defined it as: rerum cognoscere causas, i.e. a form of philosophy, an abstract thinking about the causes of things. Starting in the High Middle Ages and the Renaissance, exemplified most famously by Galileo and best theorized by Francis Bacon, science became an experimental engineering discipline. Baconian science is “just” “seeing what works.” Just ask Zombie Feynman. Modern science does not have as its goal knowledge, its goal is engineering. (Bacon, who was a true genius and a talented administrator, on top of being less than six degrees of separation away from anyone in Hollywood, understood this very clearly; he even understood that scientists would have to be misled about this, because if scientists are under the impression that their goal is knowledge, they will work harder.) The scientific method eschews speculation about the ultimate causes of things, and instead focuses on making reliable, non-obvious predictions about the future; its tool for doing so is controlled experimentation. Experimentation allows scientists to test hypotheses, and in doing so empirically validate (or invalidate) these hypothesis; when enough hypotheses are validated or invalidated, you can make reliable, non-obvious predictions about the future.

Experiments are controlled, i.e. they are designed to isolate all potential causes of an event except the one that is to be tested, and to test a hypothesis.

Galileo’s famous experiments about the fall of objects is famous for a reason, because they embody good science. When Galileo wanted to test whether a heavier object would fall faster than a lighter one, he didn’t, contrary to urban legend, just drop two balls from the tower of Pisa. He actually constructed elaborate setups to isolate other potential causes of the speed of fall, such as air resistance, shape, and so on. By designing his experiment so as to isolate other variables, he was able to meaningfully test his hypothesis regarding the fall of objects. And once his experiment was performed, and replicated, Galileo was able to make a reliable, non-obvious prediction about the future: [given X parameters], heavy and light objects fall at the same speed. It is non-obvious: Aristotle, most of humanity for millennia, and most small children today if you ask them, think the opposite; in fact, I think most of us today, even if we know this intellectually, if we see somebody on a balcony ready to drop a basketball and a bowling ball, our minds would instinctively expect to see the bowling ball fall faster. It is reliable: given normal conditions, heavy and light objects will always fall at the same speed, today as in the 16th century.

Of course, there are many other things that go into describing the scientific method: Popper’s work on paradigms, the sociology of science, and so on. But the basic gist is the Zombie Feynman gist: “ideas are tested by experiment.” This does not allow you, nor can it, nor should you ever think that it has anything to do with it, to define the Good or the Beautiful or decide if God exists. It does allow you, however, to build airplanes, and vaccines, and internets, and plenty of other cool shit. For more on this you can read (and definitely should if not everything I’ve just said was old hat to you) Jim Manzi’s book Uncontrolled, which contains the best treatment of basic epistemology for a general audience that I’ve ever seen. It’s also a wonderful book that deals with many other things.

So when Sam decries curricular innovations proposed by social scientists in educational departments, he is absolutely right. A social “science” of education is impossible for reasons that should be immediately obvious if you’ve been paying attention: a properly-designed experiment isolates variables that are not the one you want to test; but the causal density of a social science environment is so high that this is impossible to do non-experimentally. To say it in English: let’s say you have a finding “kids age 6 who took curriculum X read better than other kids age 6”. Does this prove something about your curriculum? Or does it mean that the kids who took curriculum X were somehow different than the other kids, perhaps in terms of socioeconomic background or culture or even the very fact that a new curriculum was tested on them. (And if you’re a social scientist ready to type into the comments that you can manage that problem with controls and regressions–no, sorry, you can’t.) These approaches do often have value as theory-building, but they do not, in fact, produce reliable, non-obvious predictions about the future. Evidenced by the fact that once the curriculum “tests well”, and you roll it out, it “stops working.”

“Ideas are tested by experiment.” It’s a very simple, very powerful idea. It was also never taken seriously for most of the history of human civilization, and still is not taken seriously enough. (Most people who are “pro-science” but do not understand epistemology view it as a kind of magic, which is also not taking it seriously.)

“Ideas are tested by experiment.” If you do this smartly enough over a long enough period of time, you will be able to make reliable, non-obvious predictions about the future. This is what Doctor Maria Montessori did, over several decades. If you put a 3-year-old–any 3-year-old–in a room with such and such activities with an educator with such and such instructions (mostly: get out of the way)–within two years they will know how to read, write and do basic arithmetic; they will have more self-control than other kids of their age, more concentration, they will enjoy working and learning and collaborating more, and so on. These are non-obvious predictions; that, say, tracing sandpaper letters with your fingers leads to better learning of writing is not obvious. And they are reliable. Children at Dr. Montessori’s first Children’s House were from the poorest, most illiterate quarters of late-19th century Rome, and they beat the s@#$ out of the country’s brightest children in a national test, including the richest and most “exquisitely”-educated. It was education’s “Galileo Moment.” Montessori schools reliably produce these outcomes. Not as reliably as, say, the Third Law of Thermodynamics produces its outcomes. But more reliably than chance, more reliably than any other method, as far as we can tell. Apart from Montessori educators, I am not aware of anyone who can make reliable, non-obvious predictions about the future in the field of education.

This is why I went on a rant about the platitudes in the Church’s theology of education. A platitude is a reliable, obvious prediction about the future. “If you are a loving father who models a good example for their children, all else equal, your children will do better than otherwise.” I’m totally fine in saying that this advice is reliable. It is also obvious. It is pre-scientific. It is the difference between a horse and a car. The world has no need for more platitudes, and neither does the Church.

Again: understand me, I am not saying that applying the scientific method to education is the end-all-be-all. But it should certainly be the start.

Perhaps the right analogy here is medicine. Like education, medicine was the field of charlatans for the vast majority of human history. Then, medicine became a scientific discipline. Because human beings are enormously complex, medicine remains a less reliable science than, say, chemistry. But it still makes reliable, non-obvious predictions about the future. In 98% of cases, this pill will cure your flu; ok, in 1% of cases, it might kill you, and in 1% of cases it might do nothing. It is not 100% reliable. It is still reliable and non-obvious.

In other words, there is, still, an art to medicine, and we shouldn’t forget that. But it remains primarily a science. Nobody would say that with pluck, and good judgement, you can be a great doctor regardless of whether you learn the science of medicine or not. Nobody publishes heartwarming stories of bright recent college graduates with no medical education who turn around failing inner-city hospitals, curing cancer through the powers of dedication and common sense. It’s possible for someone who has never learned education to be a better educator than someone with countless degrees in education; it’s not possible for someone who has never learned medicine to be a better doctor than even a lousy doctor. That’s the difference between a scientific discipline and a non-scientific discipline.

The other parallel with medicine is that even as doctors readily draw social prestige from the scientific character of their discipline, they often forget it. It feels more valorizing to be an artiste than a mere technician, which is what happens when your field becomes technological. So, for example, medical doctors, who are supposed to be paragons of scientific values, nonetheless brutally resist the trend towards evidence-based diagnosis, causing, out of sheer pride, countless unnecessary deaths. For example, the medical establishment had to be dragged screaming to recognize the science of the link between smoking and lung cancer; not because they were all in the pocket of Philip Morris mind you, just because pride and original sin. Science advances one funeral at a time–even most purported scientists hate science, when it makes them uncomfortable. The educational establishment just might be the most anti-scientific group of people on the planet, after Richard Dawkins & friends (*rimshot*).

Science has not obviated the need for people like Sam. Philosophers will be the last people to be obsoleted by technology, and the more technology we have the more we need them. Science makes reliable, non-obvious predictions about the future, but it cannot tell us what to do once we have these predictions. It cannot make value judgements, which is always essential but particularly in a field such as education.

But education has had its Galileo Moment. It happened in a weird sort of limbo where everybody pretends it hasn’t happened. But it has. It happened in a remote corner of the Church. It’s time for the Church to grab hold of what it has fostered and ride it to the sunset.

En Vrac: Responses to Sam

(Sorry if I’m brutal. You’re a big boy, Sam, you can take it.)

…the fact that a model works does not by itself show that reality is structured like the model. This sensible idea is an elementary ingredient of scientific practice…

Yeah. I am not talking about models. I am talking about empirical, experimental investigation. Models can be useful, but they are theories. What shows whether reality is structured like the model is experiment.

Consider PEG’s prickly claim that “Catholic Tradition paints a pretty reliable picture of what virtue and holiness is,” for instance. Holiness is union with God, sure. Theosis, right. But that picture is not a moral or pragmatic

For the record, I meant the seven virtues. It’s a sketch, sure, but the sketch is pretty obvious in what it depicts.

About his claim that “in many ways, the Church is still in a pre-Galileo age”: the Pontifical Academy of Sciences would strongly beg to differ.

Seriously, Sam, did you think I did not know the Pontifical Academy of Sciences exists? If they’re any good, they would strongly agree with me. It’s not because the Church has a committee with scientists that it takes science seriously, anymore than having an ethics committee means an investment bank is ethical.

I say moreover that you make a great, a very great, mistake, if you think that psychology, being the science of the mind’s laws, is something from which you can deduce definite programmes and schemes and methods of instruction for immediate schoolroom use.

That is not what I am saying. I am saying that rigorous experimentation is something from which we can “deduce definite programmes and schemes and methods of instruction for immediate schoolroom use.” The reason I say that is because it is has already happened.

the associative human capacity to link things together, “an intermediary inventive mind,” will always put teaching (and study) within the domain of judgement, and this domain is where the artist in all of us does its work.

I don’t necessarily disagree with this. See my doctor example. Practicing medicine is in many ways an art. Thinking that we can dispense with the scientific method in order to improve medicine is criminally insane.

Now, about the claims that it has been demonstrated that education is science, not art, and the waxing Martin Luther King Jr. There is not a single universally accepted theoretical, empirical, or otherwise, description of what the thing “education” is. None. From Plato to Augustine to Locke, Rousseau, and Kant, to Piaget and Vygosky, to Dewey to the present. No field or approach from epistemology to anthropology to psychology or neuroscience has settled the matter. There has never been a single agreed upon conception of “education” or even a single use of the term “education,” in recorded history.

Whatever. If you can use experimentation to make reliable, non-obvious predictions about it, it’s a science. Maria Montessori can make reliable, non-obvious predictions about the future. You can’t.

And again: it doesn’t mean there’s no room for philosophy, quite to the contrary–there is philosophy of biology, of chemistry, of computer science (it’s fascinating!). We need it more than ever. But we need it to start from the right starting point. A philosophy of biology that denies that it’s possible to use experimentation to make reliable, non-obvious predictions about biology is not a good philosophy of biology. All I’m sayin’.

This is what I fear PEG has fallen into, with his bravado about education and dismissal of the fact that for most of human history education has not so much failed as it has not been burdened by compulsory schooling or other institutional and social debts, some for good others for ill.

We’re agreed that Lockean schooling is bad. Lockean schooling is bad for the same reason that leeches and bloodletting are bad, though, and it’s important to grasp that.

Education remains a mystery, an enchanting entity that, like the cosmos or being itself, we all experience daily but never quite understand.

Sure. Or like physics.

STEM (science, technology, engineering, and mathematics) is all the rage for curriculum and “best practices” are the thing to do for administration and teaching.

Again, there is an obvious category error here. Most “best practices” are junk. It does not follow from that that there should be no best practices.

Scientism and psychologism are what feed and poison the mind about education

Let me rephrase that: junk science and no science are what feed and poison the mind about education.

 Both of these movements are built on so-called “science-based” research, which are usually quantitative social science research.

Right. I’ve already laid out why quantitative social science research is useless for curriculum design. The answer to junk science is not no science, it’s good science.

He can of course fix this, but it would then require that he be much, much more clear about what he means to say when he invokes science above art and empiricism above theology

(a) I do not advocate empiricism above theology, I advocate an integration of empiricism and theology. The very idea that to advocate empiricism is to put it above theology is why we can’t have nice things. Again, Aquinas is on my side, here.

(b) I hope I’ve made myself clear in this post.

muscular claims asserting […] that (real, not all the fake stuff of which we are not told how to distinguish between one or the other) Montessori is the answer.

So, just to make this point clear, because Sam is not-too-subtly accusing me of a No True Scotsman fallacy here (luv ya, Sam). The reason why I refer to “true” Montessori is because the word Montessori is not trademarked or copyrighted, so anybody can do anything and call it Montessori (which is fine, freedom’s great). Thankfully, there is a Montessori accrediting body called Association Montessori Internationale, which accredits schools, teachers, programs, and so on. This is based on a very rigorous and practical definition of what Montessori is. There is no mystery or goalpost-shifting here, it is merely a terminology issue.

Anyhoo. In a way, I understand you Sam. You come from a world, the world of the educational academy, where most people who employ rhetoric that sort-of sounds like the rhetoric I employ have been mostly bullshit artists and, as you quite rightly put it, Philistines. You’re right. I would much rather hang out with you than with them. But, like, facts matter. We can make reliable, non-obvious predictions about the future in education, and more importantly we can use experimentation to make many more. It’s a secret hiding in plain sight, and if the Church seizes it it will do a wonderful thing for the Kingdom.

Yay science. Yay philosophy. Yay theology. Yay bacon. To infinity and beyond!