We need two different science classes.
My hero Elizer Yudkowsky says that the fundamental question of rationality is why do you believe what you believe. All truth seekers should ask themselves this question frequently. Ideally, you will practice with mundane beliefs and hone the art of analyzing your beliefs before you examine more sensitive views.
Usually, the answer is that we know what we know because people we trusted told us. Sure, we may have some factoids or one-word explanations that don’t actually explain*.
Such is the case with Will and evolution. This makes me sad.
Don’t get me wrong, believing true things is preferable to disbelieving true things. It’s just that it has an emptiness about it. Believing in something without understanding how it works and why we know has consequences.
If Will’s daughter had asked him why giraffes had long necks before his realization**, he might have said something like “because evolution” along with some other curiosity-halting words that might have had meaning to an informed scientist who already knew the answer, but to him mean nothing. And he might not have even realized that they meant nothing until well after he learned these things in high school and asked how he knew what he knew. I think it’s likely that most people don’t do that.
When people talk about problems with science education, they are generally are talking about science as a body of knowledge. If you know certain facts from within that body of knowledge, you know science.
There is another definition of science though. And it is antithetical to the body-of-knowledge definition. Richard Feynman’s description of it follows: (Also see Seth Roberts.)
[D]oubt that what is being passed from the past is in fact true, and to try to find out [about it] again from experience what the situation is, rather than trusting the experience of the past in the form in which it is passed down. And that is what science is: the result of the discovery that it is worthwhile rechecking by new direct experience, and not necessarily trusting the [human] race[‘s] experience from the past. I see it that way. That is my best definition.
This is so different than the body-of-knowledge definition, that a different word is required. For lack of a better word, I’m going to call it Testing and Discovery. Science is a collection of facts. Testing and discovery is a process that destroys belief in false facts.
The facts of science do need to be learned as a body of knowledge. And science classes should probably continue to do that since that’s what everyone already thinks science is. Don’t repurpose words with established public meanings unless you want to be misunderstood and confuse the public.
But we need a different class with different teachers to teach Testing and Discovery. Body-of-knowledge science is about building assumptions. Testing and discovery is about breaking them. These are two different attitudes, and therefore must be kept separate. Otherwise, testing and discovery will be swallowed up by “science” as just more facts to memorize.
What would a testing and discovery class look like?
Students read a biologist’s words describing élan vital and how that is why alive things are alive and dead things are dead. The instructor will then lead the students in a discussion about what they would expect to happen if élan vital is real. The instructor will stress finding experiments that should provide one result if élan vital exists and a different result if it doesn’t.
From these expectations, the teacher could select which suggestions are testable and perform the test the next class period. (It is likely that the same test could be performed each year if the instructor knows how to lead the discussion in the right direction.)
Students will treat discarded and valid theories alike. They will suggest experiments to disprove the existence of phlogiston (and succeed). And they should similarly attempt to disprove the existence of oxygen (and fail). Similarly, discarded as well as retained notions about light, electricity and magnetism, and the elements will be tested.
I am personally ashamed that the reason I believe the solar system is centered around the sun with planets circling in elliptical orbits is solely that other people who I assume know what they are talking about have told me so. I vaguely understand that I ought to be able to figure it out with a telescope, detailed record keeping, time, and mathematics. Wouldn’t this be awesome thing to try to figure out from scratch though? Get the students to take some of the measurements, so they know how that part is done, but then give them the rest of the data so they don’t have to wait years. Then they can actually use the analytic geometry they learn in math class. Even if this took high school students a semester, it would probably be a worthwhile exercise.
I see two critiques of my proposal:
- “Science classes already do this.”
No, they don’t. I do not think that students are aware of what they know for reasons they can explain themselves and what they know simply because they read it somewhere for a test. Yes, students do lab work in science classes, but these are not experiments that are owned by the class. Since students play no part in their formulation, they perform the experiment simply to see whether they can get the right answer, not to test a theory. I think the average student probably thinks the reason their class has lab work is for them to learn how to use the lab equipment. - “It is easy to say what should be included. What are you going to kick out?”
Much time could be recovered by eliminating non-experimental lab work from the regular body-of-knowledge science classes. Even if we took away real instruction time from regular science classes, I think it’d be an improvement. Yes, students would know less of the body knowledge. They might not get to find out what the steps of the Krebs cycle are, but they were just memorizing that anyway without any notion of how they could figure it out for themselves.
* Why do magnets attract iron? Answer: Magnetism.
If you can’t explain the word magnetism in a way that explains why aluminum is not attracted, it’s likely that your answer is empty, and you might as well have said “magic” for all you know about magnetism.
Photo credit: Wikimedia commons, Flickr users LEOL30 and Maia Weinstock
Really, does anyone have an explanation for magnetism, or any of the fundamental forces, that doesn’t basically boil down to “magic?” At the subatomic level, nothing makes any sense at all.Report
Obligatory (NSFW):
http://www.youtube.com/watch?v=OvmvxAcT_Yc&feature=youtube_gdata_playerReport
Also
Report
Quantum mechanics has rules; they’re just completely bizarre and unintuitive. Anything above that, like electricity, magnetism, gravity, etc. is described by approachable math, like calculus and differential geometry, and entirely accessible to the properly trained intuition. Even things like the twin paradox make sense if you work them through rather than pointing at them and saying “weird!”Report
Sure, but those are models, not explanations. Particles attract or repel each other without touching. We can define the conditions under which this will happen, but that doesn’t mean we know why.Report
Fair enough. You explain everything in terms of something else until you get to the bottom. But the bottom is still described by mathematics, or at least statistics, not capricious, which is why I wouldn’t use the word “magic”.Report
But you don’t teach high schoolers quantum mechanicsReport
Both of my kids were taught some elementary quantum mechanics in high school chemistry. Not the wave equation, of course, but spdf orbitals, as opposed to being told that electrons orbit the nucleus like tiny planets.Report
For what it’s worth, I remember spdf from high school.Report
But you don’t teach high schoolers quantum mechanics
Why not, given that they spend so much time every day with quantum mechanical devices? Every cell phone in the world is stuffed full of flash memory, all of which depends on floating-gate transistors and electron tunneling to function. If you’re going to teach them about electrons at all, you might as well explain something about just what strange little beasties they are.Report
From Ambrose Bierce’s Devil’s Dictionary:
GRAVITATION, n. The tendency of all bodies to approach one another with a strength proportion to the quantity of matter they contain— the quantity of matter they contain being ascertained by the strength of their tendency to approach one another. This is a lovely and edifying illustration of how science, having made A the proof of B, makes B the proof of A.
MAGNET, n. Something acted upon by magnetism.
MAGNETISM, n. Something acting upon a magnet.
The two definitions immediately foregoing are condensed from the works of one thousand eminent scientists, who have illuminated the subject with a great white light, to the inexpressible advancement of human knowledge.
Which is fine, so far as it goes, but if you can’t tell the difference between the organized collection of data and inferences that constitute physics and the fuzzy, often self-serving rationalizations that make up economics, you’re not trying.Report
I remember high school teaching circular orbitals plus some hand-waving about how it’s not really that simple, but you’ll learn that in college. But also calculations about how the energy required for a hydrogen atom’s electrons to jump from one circular orbital to another matches its spectrum, which could be called very, very light QM.Report
If the best way you can explain something is “there’s 16 dimensions, but 12 of them are really tiny and folded up inside each other, and there’s no way we can measure or perceive them”, then I’ll opt for “magic” as the simpler and explanation. At the least, it’s no more nonsensical.
The higher-level you go in physics, the less scientific it starts to sound.Report
The higher-level you go in physics, the less scientific it starts to sound.
For example, observer dependence.
“Of course the introduction of the observer must not be misunderstood to imply that some kind of subjective features are to be brought into the description of nature. The observer has, rather, only the function of registering decisions, i.e., processes in space and time, and it does not matter whether the observer is an apparatus or a human being; but the registration, i.e., the transition from the “possible” to the “actual,” is absolutely necessary here and cannot be omitted from the interpretation of quantum theory.”
In other words, magic.Report
I remember reading a paper on causation a few years ago, and it was talking about how, at the most fundamental levels, physics doesn’t even talk about causes anymore, but states and systems and things. That, to me, is where science stops looking like science.Report
Observer dependence, I suspect, gets way overblown. We try to apply it to larger states; and I don’t think the concept translates above the quantum level.
Of course, I’ve know way to conduct an experiment and prove it, I can only imagine. And that itself is magic. The magic that drives us on to scientific discovery.Report
Chris,
That, to me, is where science stops looking like science.
What do you think it does look like? Something older? Something completely different and new?Report
zic,
Here’s something to whet your appetite:
The many-worlds interpretation is an interpretation of quantum mechanics that asserts the objective reality of the universal wavefunction and denies the actuality of wavefunction collapse. Many-worlds implies that all possible alternative histories and futures are real, each representing an actual “world” (or “universe”).Report
Ah yes, Stillwater; there are as many zics as there is quantum potential of zics. I read Zelazney, I learned the concept from him.
But that’s exactly my point; there either are as many zics as are potential, or there is a vibration of potential, celestial music, that settles into the zic that is; I suspect both are true; one at the quantum level, and one aggregated to my level of space time.
An analogy might be that I’m me; an individual. But I’m a colony, too, of all the living things that make me; the flora and fauna of zic. Both are true at the same time. I can easily step up to the level were I am part of a larger living ecosystem, fauna of Earth, and I will be recycled just as I’m recycled matter of spent stars.Report
Still, I dunno. I don’t know enough about how stuff works at that level. The math is beyond me. If causality is discarded down there, though, I don’t know what we’re dealing with. I assume it still makes predictions, the way science is supposed to, but are those predictions just based on systematic probabilities without mediating causes? I don’t know what that looks like. Where are the particle physicists in the house?Report
http://en.wikipedia.org/wiki/Arrow_of_time
Weak force establishes causality.Report
are those predictions just based on systematic probabilities without mediating causes? I don’t know what that looks like.
Not mediating causes. Mediating fields. Sorta goes like this. Electromagnetic forces are mediated by photons. The photon carries the field effect. The quick metaphor is a game of Tag. You’re “It”. You reach out, touch Mike, he’s now “It”. You’ve exchanged a virtual particle.
Different forces need other intermediaries. The forces holding the nucleus of atoms together is mediated by mesons. Gets very interesting thereafter.Report
The problem is the definition of “explanation”. If I were to present you with f=ma as a sovereign fact rather than an explanatory formula, I’d have to describe force, mass and acceleration. This formula, Newton’s First Law of Motion, is sufficiently explanatory for getting vehicles into orbit around distant planets.
But Newton’s First Law can’t be applied at very small scales. Did it stop being true? Force is simply discarded at those scales. It’s only a derivative of time and momentum. Other formulas can be applied, equally true, just phrased differently in terms of the conservation of energy and momentum. Thus they can be more generally applied.Report
This formula, Newton’s First Law of Motion, is sufficiently explanatory for getting vehicles into orbit around distant planets.
I think this tells us everything we need to know about “explanation”. Science itself has value because it allows people to do cool and interesting things. If the information you provide allows the listener to do those cool and interesting things, then you’ve accomplished much of your goal.
Of course, that is part of why the body-of-knowledge subject should be retained. Having knowledge of true things can be useful even if you don’t believe those true things for reasons other than an authority telling you.Report
I think this tells us everything we need to know about “explanation”.
In one sense, yes. The law explains why objects in motion act as they do. But the fact that objects in motion act as they do doesn’t explain the law itself, right?
Depends on what we’re trying to explain, it seems to me. When Brandon says that some apparently irreducible property is invoked to explain other phenomena but that property doesn’t admit of any analysis, we’re not explaining that property at all. So calling it “magic” starts to make some sense. Not complete sense, of course, since it’s possible the reason we can’t explain magnetism is our current ignorance.Report
Are you asking for why the universe obeys laws? There is no other “sense” to any of this.
My son once scornfully told one of his mother’s friends, one of these nu-age types: “Science tries to understand the world. Magic attempts to control the world. That’s why science will always win.” He was about eight or nine at the time.
The universe doesn’t exist for our benefit, with a handy bill of construction materials and an assembly manual. The closest we’ve come to any descriptive vocabulary is mathematics. If you want to ask the universe a question, you might start with this one: “Why do numbers serve so well to describe you?”
We live in a world of magnets and miracles are happening all around us, in the quantum foam, oceans of particles, winking into existence in pairs, annihilating each other. The universe may be expanding because they do, giving rise to time and space, both.
The great telescopes in orbit, our probes pass beyond the final eddies of our own star’s exhalations, into unexplored seas. In the centre of our galaxy, a cloud of gas approaches its titanic heart, the black hole is soon to feed. Scientists wait expectantly for an event which happened 27,000 years ago.
We are witnesses and our every attempt to derive the Whys are endlessly postponed while we answer the more-pressing questions: What, Where and When. They will occupy us forever. In many senses, Why is the stupid question — or at least the last question. The Who we know. The Who is us. Intelligent life may be the universe’s crowning achievement, an attempt to understand itself.Report
The Feynman link (which is highly recommended anyway) includes an anecdote in which Feyman asks his father why a ball in his cart appears to go backward when he pulls the cart forward and goes forward when he stops the cart. His father answers “no one knows, but scientists call it momentum” and then explains what momentum is. So, he is providing an explanation but acknowledging in what ways it is incomplete.Report
Well, scientists can call it “momentum” all they want, but naming the phenomenon isn’t an explanation of it any more than naming a Birch tree as a “Birch” explains what Birch tree is.
Some folks concepts admit an analysis in scientific terms to give them a precise meaning. Momentum might be one of them. But when we say that a heavy object in motion will destroy a light object in it’s way, we’re not explaining momemtum. We’re giving an example of the property that accounts for why the light object was shattered.Report
Among the other egregious typos, the second paragraph should begin with “folk-concepts”.Report
@blaisep
The Who is us.
Sorry? I was off smashing a guitar.Report
At some level, characterizing a phenomenon — and our conceptual characterizations are really just metaphors, with the real work being done by the mathematics — is a full explanation. Anything beyond that is outside the scope of science, because in order to extend an explanation further we would have to posit unobservable or non-measurable phenomena. There is no scientific reason why this would be necessary. There may be extra-scientific reasons why it would be, but that doesn’t mean science itself is saying, “And then… magic.” Science is saying, “This is the whole of it from my perspective. Anything else has nothing to do with me.”Report
The Who is us.
I don’t wanna be Keith.Report
At some level, characterizing a phenomenon … is a full explanation. Anything beyond that is outside the scope of science,
See, I don’t think you really believe that. My history of science is shabby, but when Darwin proposed his theory, there was an implied mechanism by which natural selection could take place but for which there was no explanation. I don’t think anyone would have been happy with an explanation which which restricted itself to “the thing which causes differentiation is species” or whatever.
Let’s try it this way. There are all sorts of observable correlations to be had out there. Some of them admit of a causal analysis, and some don’t (or don’t clearly anyway). Of those that admit of a causal analysis, simply describing the causal relation isn’t necessarily an *explanation* of the causal relationship, since quite often those causal relations are analyzable in more basic terms. Insofar as they are, then providing an account or analysis of the causal properties in terms of those more basic properties constitutes an explanation of the property and its causal efficacy. In the cases where it can’t be explained in more basic terms, then the property is viewed as irreducible (for now!) and fundamental. But merely asserting that it’s irreducible can’t be equated with an explanation of that property. It’s tacit admission that we can’t explain it! That’s where magic rears it’s head.
But I don’t want to imply that functional explanations aren’t valid in a context or useful. They are. I just don’t think they’re scientific explanations. They seem to fall more on the side of engineering of something.Report
Still, what you say is true, but that’s not really what I mean when I talk about reaching a point where further explanation is beyond the scope of science. There are two possibilities for what you describe: (1) we have fully characterized a phenomenon or set of phenomena for the purposes of prediction, which is to say, all of the behavior directly resulting from it can be predicted at the highest level of precision possible given our current limitations (in instruments, say), but there may in fact be unobservable phenomena that, when our technological or theoretical sophistication improves, we will be able to observe, or (2) we either recognize that we haven’t observed all of the relevant features of a phenomenon or set of phenomena, or we perhaps have observed all of it but haven’t been able to fully characterize it. The Darwin scenario falls somewhere in (2), and most of the scientific universe falls somewhere in (1) or (2).
Electromagnetism, and most of basic particle physics, with the exception of gravity, may fall in (1), but as far as the Standard Model is concerned, with a few exceptions, it’s pretty much done. Of course, there’s a third possibility, that some new observation will cause us to realize that the Standard Model isn’t, in fact, a more or less complete model of subatomic forces, but is in fact a special case of some more general model. This is different from the “and then, magic” accusation, though. This is more Patrick’s point: science is never finished. It doesn’t deal with truth, it deals with data and competing models of that data. Or the way I usually put it: science is always wrong, but it’s getting less wrong every day.Report
Chris does an excellent job of sorting out, here. I’d like to add one more category: (3) Predictions extrapolated from applying the (1) category. The black hole, case in point: predicted long before it was detected. The mathematicians are the prophets of science. They predict.
As for Darwin, he’s dropping from the exalted ranks of (2) down to (1) these days. In saying so, I’d like to note Darwin’s exact statements have been largely superseded by other biologists — just noting we’ve observed sufficient relevant features of the phenomenon to discard the previous theories.
And let’s not be led down strange alleys, trying to come up with a “thing” which differentiates the species. It’s not as if the species are trying to differentiate. Some, like the great white shark, haven’t evolved much in millions of years: they didn’t need to. Low-theta DNA. Some, like the viruses and bacteria, are evolving so fast it’s hard to keep up. They’re evolving because we’ve been making war on them. Thanks to Stupid People who won’t take their full courses of antibiotics, they have acquired resistance to our Final Solutions for their kinds. When a proverbial tree falls in a forest, tell you who hears it — all the plants which thrive in a freshly-opened patch of sunlight. The jellyfish just love us, we’ve made a perfect world for ’em.
Phenomena aren’t their own explanations but the numbers explain them perfectly. As for the birches, they “birched”. We can compose a fractal to “birch”, form up as would a birch tree. They didn’t have any option, any more than you did. Unless you’ve been at the Clairol bottle, why is your hair that colour? What explanation is there for you, or me, or Chris? The answers are pretty obvious: our parents.Report
does anyone have an explanation for magnetism
I’m not a philosophy of science guy, but I’d say that something qualifies as an “explanation” if after hearing it you can make predictions about similar situations that differ in certain respects. So, if after hearing an explanation you then expect certain materials to be attracted and others not to be based on whether they fit the explanation or not, then the explanation you got is real. Or at least it is valuable. You, of course, won’t really know why it works that way and not some other way, but you at least have an explanation that you can take with you and later be surprised when someone shows popcorn is magnetic.Report
Right, if we have a well-tested model of how something works at the most basic level of explanation we can recognize as necessary (in the case of electromagnetism, the Standard Model does all o’ that), I think we can reasonably say that we have an explanation of that thing. To further ask “why?” is to misunderstand what science is and how it works. It’s either asking for some sort of teleological explanation (it works this way because God wanted it to, or because it was necessary for it to do so for us to be here to explain it, or something like that) or to get caught in an infinite causal regress (the Standard Model is what it is because of this even more fundamental cause, which is what it is because of this other, still more fundamental cause, and so on, and so forth, ad infinitum).
We don’t need to invoke magic. We just have to recognize that this is as far as the data takes us, at least for now, and therefore we currently are not in need of any more fundamental level of explanation.Report
but I’d say that something qualifies as an “explanation” if after hearing it you can make predictions about similar situations that differ in certain respects.
I think in a case like you’re describing what we’re essentially doing is identifying certain types of regularities and explaining them in terms of a different property. That is, we’re explaining other phenomena in terms of a concept which we haven’t yet provided an account or analysis of.
I mean, it seems to me that you’re suggestion is that the property of magnetism is explained by identifying the states of affairs in which the property is realized. Which strikes me as a strange way of understanding the word.Report
the property of magnetism is explained by identifying the states of affairs in which the property is realized
I think that is probably a fair way to characterize my definition. I agree that it is strange. I could probably be easily convinced of a different definition if someone is willing to provide a better one.Report
Don’t repurpose words with established public meanings unless you want to be misunderstood and confuse the public.
Yeah, stop saying “coercion” when you mean “deception”.Report
I really need to do the series of posts on why yudkowsky, even if he is a very smart person just gets almost everything wrong. Here is why he is wrong about Science. Science is a social enterprise not an individual enterprise. It is very easy to understate the role testimony plays in Science. No one has the time to test every bloody thing. Even as a practicing scientist you have to take a lot of things for granted. What allows us to have a high confidence in the veracity of the body of scientific knowledge is that the institution as a whole is set up to weed out beliefs based on bad evidence etc. It does not require each scientist to heroically question everything and be perfectly Bayesian. Even practicing scientists don’t re test everything to re confirm for themselves whatever they have read in textbooks and in journal articles is actually true and not merely anomalous. If people didn’t take things on trust, there would be no way to build on the existing body of knowledge. But as I have noted above, this trust is not a blind faith, but a reasoned trust. Too many improbable things would have to happen to seriously threaten the existing body of scientific knowledge.
But more importantly, unless you have a complete picture of everything, everyone is in a way engaging in magical thinking. Why do chemical reactions occur? The thing you learn in high school with electrons in orbitals is not completely accurate. Also, if you don’t already have a good grasp of the accepted body of knowledge, you can’t even begin to intelligently ask the right questions. There is a time and place for everything. Highschool is not necessarily the best place to teach people about questioning and testing. And facile (and incorrect) accounts of the “Scientific Method” by the likes of Yudkowsky et al do not help.Report
Sorry, I left a train of thought hanging. Unless you’ve got the right explanation all the way down, you’ve just passed the buck to another “magic”Report
Well said, Murali.
I think the best thing for science courses to do is to teach people how to think scientifically, that is, in terms of the scientific process: prediction and the testing of predictions, control, replication, etc. Scientific “facts,” such as they are, are things people can acquire later, and their ability to evaluate them, and their sources, will come from that initial education.
This is, coincidentally, how psychology departments usually do it. The first psych course majors take might be an intro to psychology (though I actually never took that course), but it is immediately followed by statistics and methods courses before they get into the taller weeds of the discipline. The idea is that once they understand how to think like psychological scientists, they will be better able to evaluate the information they get in the courses that teach them the “facts” of psychological science. And the better psych upper level psych courses spend a lot of time talking about the research that produced those facts.Report
It strikes me that Murali’s point is, in general, a good way to conduct our everyday lives; to recognize that we’re operating on assumptions, and that we need to be willing to challenge and revise those assumptions. Changing your mind (waffling) in the face of new evidence is a feature, not a failure.Report
Even as a practicing scientist you have to take a lot of things for granted.
Not really.
I mean, yes, you depend on existing work being correct, but the proper way to think of this is “given that such-and-so showed this is pretty likely to be true, if we accept that it’s true and we apply this other thing, we get this, which is also pretty likely to be true.”
If such-and-so’s work is shown to be incomplete or wrong, you can revise the other things.
The idea that you’re talking about here is “in order to get anything done, scientists need to be able to accept things as true”. That’s a mathematician’s approach, it shouldn’t be a scientist’s approach.
Scientists don’t know any damn thing to be true. They only have degrees of certainty. There’s nothing wrong with that, it’s actually a pretty good way to surf the Universe.
This idea that laypeople have that science theories are true or not true is sorta missing the whole point of the exercise. It’s not just laypeople, you see lots of scientists making this error, too.
Everybody’s obsessed with certainty.Report
Given how specialized science has become, there are concentric circles of trust expanding ever outward from a scientist’s particular area of expertise, and that simply can’t be avoided. Imagine you’re a person who studies perceptual categorization. You know, read, and discuss the work of all the other people who study visual categorization (a relatively small group). Here there is a very basic level of trust, perhaps the most basic in science, which is that your colleagues, the people studying the same things you are, report their work honestly. This trust is so basic, and so important to the integrity of science, that any violation of it will likely result in the violator being excluded from professional science for life. On top of this basic trust is built the day to day practice of science: you, and experts in your field, discuss your work, evaluate each other’s work, submit it to reputable journals for peer review and publication, and so on.
Outside of your specific area of expertise, you will have a high level of knowledge of several related fields — in your case, as someone who studies perceptual categorization, you may have a high level of knowledge of higher level categorization and of higher order vision. Your knowledge of these areas may be so extensive that you can conduct original research in these areas, usually in collaboration with other researchers for whom these are their primary areas of expertise. On this circle, you may be required to have a bit more trust of the researchers doing the bulk of the work in these adjacent areas of expertise, because they are doing the bulk of the work in the first circle — the discussing and evaluating and peer reviewing, along with replicating and further testing. You can do some of this, of course, and you’re pretty good at evaluating the research once published (assuming, for the most part, that it has been adequately peer reviewed — you only read journals with good reputations, right?).
In the next circle or two, you have a broadly defined discipline or two– say vision science or cognitive psychology — and you may work on on occasional project that falls into these broadly defined disciplines bout outside of your specific area of expertise and the areas immediately adjacent to it. You’re going to rely pretty heavily on collaborators in that case, and when you’re not doing original research, you’re able to evaluate others’ research somewhat well, though you are not as familiar with the specific methods and concepts, so your trust in the experts will have to be even more substantial.
Then you get out of the broadly defined disciplines, into related disciplines, on which you may have taken some graduate-level courses, or done some serious reading, so that you can at least do fairly basic level evaluations of research (you’ll recognize glaring methodological problems, for example), but now you’re pretty much relying on a whole discipline-worth of people to do the work that you and your colleagues are doing in your own. For a person who studies visual categorization, this circle of trust may go through social psychology, clinical psychology, basic neuroscience, perhaps even some areas of biology (you might know about non-primate visual systems, e.g., and follow those literatures from a distance).
And then you get into areas in which your expertise offers very little help, except a basic knowledge of the scientific method — most of chemistry, most of physics, linguistics, etc., for the visual categorization researcher. Here, your level of trust is not that much lower than that of non-scientists.
I don’t think it’s a stretch to call much of this taking things for granted.Report
I go back and forth about this. Murali and Chris are right. Even if you want to conduct a little experiment that involves weighing something, you are then accepting all the assumptions built into the mechanics of that scale. We never get to show something from scratch, because anything you do rests on some edifice that someone else could call into question.
But it’s too way too easy to go from that and infer that no one really knows anything, which is just stupid. Practically speaking, you need to just use your scale for what you’re doing and later if you want to question the scale, run some other test (or as is almost always the case, hope that someone else has already checked).Report
I think we’re probably dancing around a feeling, more than anything else.
Your description is pretty apt, that’s true. I guess what I’m looking at myself is how we regard the things that we take for granted.
If you take them for granted for working purposes, that’s okay. You have to accept some things in order to get stuff done, even in the most positivist of disciplines. But the idea of “Truth” comes with all sorts of baggage attached to it; people stop thinking in terms of “I’m going to provisionally accept this thing for the sake of moving on, but it’s only accepted as likely” and start thinking in terms of building things based upon the axiomatic truth of something done by someone else. That leads… elsewhere.
I don’t like this “truth” business or “proof” business in science. Truth belongs in philosophy, proof belongs in math. Science is the land of probability.Report
Truth and proof are for logicians and mathematicians.Report
I really need to do the series of posts on why yudkowsky, even if he is a very smart person just gets almost everything wrong.
You do! I eagerly await.
It is very easy to understate the role testimony plays in Science.
I would agree he doesn’t address this point, but should he? Do we have a lot of people who are attempting to do science disconnected from what anyone else is doing?
Highschool is not necessarily the best place to teach people about questioning and testing.
I take it that you mean it is too early. We disagree there. Without learning that knowledge comes from evidence, all they learn is that textbooks and white coats hold truth, and then when someone hands them a book that says something different, then they won’t necessarily understand that that doesn’t mean the claims are equally valid.
—
For what it’s worth, my own views of how science works align more with Paul Feyerabend than Yudkowsky, but I still value what the latter is doing.Report
“No one has the time to test every bloody thing.”
And that’s why we still believe in spontaneous generation, heliocentricity, and aether flow. Because no one has the time to test every bloody thing, the science is settled, only an idiot would continue to obstinately resist the obvious truth.Report
Perhaps it would be simple to teach the kids a discarded view on some subject, as if it was true, and the walk them through how they’d go about challenging and disproving it.
Interestingly, if you read Galileo, Descartes, and other authors writing at the beginnings of modern science, they do a wonderful job of challenging intuitive notions about force, motion, weight, and other things familiar from everyday experience, explaining their reasoning step-by-step, with many analogies and simple experiments. The starting point is that you believe an incorrect concept. Then they posit ways you could test whether the concept is true, and then they carry out experiments (both in thought and in the real world) and examine the results.
It’s a given that almost all a child’s intuitive notions about motion and energy are wrong (we’re basically wired to be Aristotelians, with ideas about “oomph”). Usually we just teach the correct and modern physics in the lab, where the students time falling balls or graphic parabolas or something, skipping the all-important step of coming up with such an idea in the first place.
You could take a page from “The Matrix” and state it as “Many of the things you believe are wrong. You don’t know which ones, and you don’t know how they’re wrong, but you have to figure out a way to make the underlying system reveal itself by finding a way to make it defy your expectations. Then you’ll have shown something about how the system is really built, instead of how it appears to a casual observer.”Report
discarded view on some subject
Yes, I think it would be great. We believed in phlogiston for *forever*. There were a lot of smart adults looking at that problem, and they never figure it out. I think it would be a really powerful lesson for students to learn how those scientists thought about it and how it was ultimately shown to be wrong.Report
Perhaps it would be simple to teach the kids a discarded view on some subject, as if it was true,
Like, the Civil War was about tariffs?Report
Trickle down economics?Report
I actually recently read the Confederate Constitution, and it *does* include at least one change aimed at preventing regulatory capture, so I think it is fair to say the Civil was more than 0% about tariffs. 🙂
Still probably less than 1% though…Report
Nonsense. There is no need for two science classes; rather there is a need for all science classes to be better than they are. (Scientists, as a rule, advocate this. School boards and other political hacks who value nothing more highly than standardized test scores, do not.) Feynman’s definition of science is simply the correct definition. If it is true that “when people talk about problems with science education, they are generally are talking about science as a body of knowledge” (I’m not convinced that it is), then it is because they were taught science badly. The “facts of science” are mostly the results of investigations that are too elaborate or expensive to be carried out in a classroom; they should be taught as such. Let students discover the few facts that they, as a practical matter, can, and learn of the experiments that led to the rest. Why do you assume that experiments can be persuasive only when students design them themselves?
BTW, if Will’s daughter had asked him why giraffe’s had long necks, he should have said “so that they can reach the leaves on tall trees.” The principle of evolution is no more complicated than that observation. (Of course, the details are.)Report
Feynman’s definition of science is simply the correct definition.
That is a lost battle. If Richard Feynman wasn’t able to convince people that science is something different than just a body of knowledge, who will? If you want to communicate with people properly, you have to use words they understand the meaning to, not words that they have incorrect definitions of.
The “facts of science” are mostly the results of investigations that are too elaborate or expensive to be carried out in a classroom
Some are. Many fundamental ones are not. Many of Maxwell’s most important electricity experiments were pretty simple. The same goes for Lavosier’s in chemistry.
Why do you assume that experiments can be persuasive only when students design them themselves?
Absent any effort to design the experiment, they are following a recipe to get certain results. They won’t know what purpose the results serve. I grant that it is possible that they might given better instruction, but that is not really different than what I am saying. I am suggesting that brainstorming possible experiments that could be performed with students *is* the type of instruction that would get them to understand what their lab work is about.Report
I’m not certain two classes (I think you mean subjects) is required.
Students can see the progress of history, the world was the flat center once, after all. So much of what you’re suggesting is actually gotten at by teaching the history of science.
I sometimes think there’s another fault here: the lack of appreciation for the value of amateur scientists and the aggregation of information; that it’s some separate speciality. I could write a program to go scrape Audubon-sponsored bird-watcher sites and generate a maps bird migrations underway right now. The accuracy of that map would depend on the numbers of people recording sightings, their accurate identification, and geographic distribution. So participation in data collection of the observable world around you is one part of ‘science,’ it’s not just the hypothesis-experiment model. Another example of this would be the way they’re collecting information about radiation spread in Japan; actually distributing geiger counters to people. Science isn’t just setting up an experiment, it’s a process of recognizing where we can get the information to analyze; can we measure changes in fish stocks over time based on the reports of fish catches from the 1800’s; or carbon over time based on air bubbles in ice crystals?
There’s an ongoing process of refining and updating based on long-term observations, limited by the accuracy of our observing equipment. Think of how much we learned when we got even a flawed Hubble telescope out in space, without the atmospheric distortions of earth-bound telescopes. This is the classic conundrum of quantum physics; I believe the metaphor used by one physicist was that quantum physics was like trying to understand a bowling alley by rolling bowling balls across the lanes, with no other data coming in.
But even here — much of the existing body of work you’re suggesting glossing over is the math that science rests on, certainly in physics, and on the statistical analysis of information that’s often in form form. I’m not sure a student needs to come out of high school with an understanding of a specific formula, say measuring heat calories, memorized as they need to come out of high school understanding that there are existing rules that have been drawn up based on repeated observation, they can be learned if needed, and they can often be refined, revised, or overturned based on new information or better analysis of existing information, and most importantly, that there are many, many ways to collect information and each way has its limitations.Report
Yes, I meant subjects. Thanks for the clarification.
History of science *would* do it. But I would note that much of history of science at the K-12 level seems to be things like “Democratus thought atoms were tiny, indivisible billiard balls.” We don’t learn how anyone figured out Democratus was wrong about that. Nor do we ask what would be the consequences of Democratus having been right.
participation in data collection of the observable world around you is one part of ‘science
Yes, I would support this being included. I would classify what you are talking about as “evidence gathering”, which belongs to the testing and discovery subject rather than the body-of-knowledge one.Report
Yes, I would support this being included. I would classify what you are talking about as “evidence gathering”, which belongs to the testing and discovery subject rather than the body-of-knowledge one.
This is where we begin slipping down a rabbit hole, @vikram-bath; for ‘gathering evidence’ is typically shaped by body-of-knowledge. How to accurately measure things, how to record follows the rules often defined in the body of knowledge.
You didn’t respond to my third point; it’s not memorizing he body-of-knowledge that matters so much as being familiar that it 1) exists, and 2) where to find the portions of it needed for any given situation. I would go to computer programming for an analogy; the best programmers don’t necessarily know the details of every device protocol they might need, but they do know how to quickly find that information and how to use it once they find it.Report
How to accurately measure things, how to record follows the rules often defined in the body of knowledge.
Granted. I think this can be safely ignored until some philosophy class later on someday. It is possible that some students will ask about how evidence gathering can occur without applying some pre-existing knowledge, but that needn’t stop the whole endeavor in its tracks.
it’s not memorizing he body-of-knowledge that matters so much as being familiar that it 1) exists, and 2) where to find the portions of it needed for any given situation.
I didn’t respond to that because I agree. Certainly such things ought to be taught.Report
@zic
I mostly agree. My only reservation is that sometimes memorizing, or at least being well-versed in the specifics of, a body of knowledge can help people with no.’s 1 and 2.Report
So, wouldn’t it be appropriate to teach two types of literature class? One the body of knowledge (Moby Dick), the other the process? Personally, I find it appalling that people graduate from high school (or even college) with no real experience with the process. And I’m mostly a math/science guy.Report
Well, I wouldn’t find it too weird if people took both writing and reading classes.
That said, I don’t know if they need to be separated. Writing isn’t opposed to reading in the same way that I (and Feynman) claim testing and discovery is opposed to body-of-knowledge science.
—
As a former instructor, I would love it if the system produced better writers. It is needed.Report
@vikram-bath ,
When I discuss my science curriculum, I always phrase it as “Science and Inquiry”. With young children, learning process is as, if not more, important than facts. I think this extends beyond just the scientific domain. I tend to say that everything we do in a pre-kindergarten classroom is science. The children are constantly exploring, making discovery, testing, etc.Report
“Science & Inquiry” does sound appealing, I admit.
It would be nice if knowledge transfer as well as knowledge testing could be done in the same class. It probably would work with younger kids who haven’t been corrupted yet. I’m not sure that a regular science class could be modified to include what I am talking about though once the kids get older.Report
If I boil what I teach into it’s most basic component parts, I think of two things: content and skills.
Some content is of the utmost importance. Some if it isn’t really discoverable, largely because it is entirely abstract. Take the alphabet, for instance (something I’ll expand on in an upcoming education post). You can’t really discover that the three strokes used to form this symbol – A – will be called “A” and will make the various sounds that “A” makes and understand how it is used in conjunction with other combinations of strokes to form words and sounds and ideas, etc. You could learn this informally – “Hey, every time that guy writes that letter he calls it “A” and makes one of these sounds,” but you are still learning it by being taught by others. And this is because our letters are abstract; they are representational symbols.
However, most content is discoverable. And, perhaps more importantly, much of it has been discovered, at least in a part.
In a world that is both increasingly automated and accessible, I put much more weight on skills than content. Do I really need to teach kids the state capitals when they can pull this info up in a few seconds on their phones? Probably not. Sure, I could teach them a song that they memorized and which they can recite and they can then tell you that Albany is the capital of NY, but how valuable is that really? I’d rather teach them skills that help them to understand why and how Albany came to be the capital of NY and what it means for the state and country today that the seat of political power of the state is located a couple hundred miles from the biggest and most powerful city in the state, and one of the most influential in the world. That, to me, seems me important.
So I teach them to question, to analyze, to synthesize, to generalize, to notice patterns, to notice deviations from patterns, etc. I attempt to teach them to question authority, even knowledge authority. I’ll hold up the blue paint and say, “I’m going to use green today,” and when they all say, “NOOOOOOO!” I’ll say, “Wait a minute… I’m the teacher… I’m saying this is green. Don’t you listen to the teacher?” It’s silly, I know, but it shows them not to just accept knowledge at face value, especially if it disagrees with their own knowledge and experience.
I sincerely belief that if given the right skills, kids can acquire most any knowledge in a way that is meaningful. Problem is, our education system is geared primarily towards knowledge, and perhaps increasingly so. The prevalence of high stakes tests tend to reward knowledge, largely because it is easier to gauge. I can more easily assess whether or not you know what 4×4 is than I can determine whether you understand what it actually means to multiply and if you can successfully apply this skill when appropriately called upon to do so.
As for how we structure classes, I think if you were to build from the bottom up with a science/inquiry model or, more broadly, a better balance between skill and knowledge, process and content, you could develop a model that works for older children. But if you walked into a current 7th grade science class and tried to make the shift, you’d be hard pressed to.
That is why I tend to believe in a bottom up approach to education reform, in which you let children actually grow with the mode. Taking high school seniors and attempting to make a massive reform effort and expecting real change is generally a fool’s errand.Report
“Science & inquiry” still kind of opposes “science” qua facts to inquiry qua process, though. Not that it’s not right for K’s purposes, but what about “Science in Two Equal Parts: Nature & Inquiry”?Report
My first- and second-year university science classes have given me detailed experience with why this won’t work. Namely, that accurate experimentation is difficult, and involves a high degree of precision and skillful performance of each step in the experiment. Experiments done by low-level university students – and even more so, high school students – don’t fulfill these requirements. In first-year chemistry, getting the expected results from an experiment was the exception rather than the norm. In a high school class, you could very likely end up “proving” the existence of phlogiston, or “disproving” the existence of oxygen.
For an amateur in scientific procedures, there’s about a 1% chance that getting an unexpected answer means you’ve disproven established theory or made a major new scientific discovery. There’s a 99% chance that it means you screwed up the experiment at some time.Report
But isn’t there value in screwing up?Report
In what way? In general, yes, we can learn through making mistakes.
But Vikram is talking about testing the veracity of basic scientific principles through experiments performed by high-schoolers. If – or rather, when – the experiment (due to being mis-performed) produces conclusions that are contrary to basic scientific knowledge, your options are:
A) accept the experiment’s results as being accurate and teach the students a principle that is known (by the experiments of substantially more proficient scientists) to be incorrect OR
B) tell the students, “well, this is what the results show, but the actual scientific facts and records of previous experiments demonstrate the opposite, and our results are due to errors in performing the experiment”. Which is what’s currently done in science lab classes, and defeats the very point of a separate “testing and discovery” class.Report
Depends on the consequences of the screwup. If the screwed-up solution isn’t perceived to be a screwup, it takes on a life of its own, cases in point: Ptolemaic epicycles and phlogiston. That sort of Screwup Solution ties people to stakes and burns them, leads Galileo into the torture chamber and confines him to house arrest.
Not every screwup is seen to be one at the time. Einstein, feebly jabbing at the consequences of his own theories: “God does not play at dice.” To which Stephen Hawking responds: “Not only does God throw dice, he throws them where even he cannot see them.”Report
Labs are typically geared towards trying to be as difficult as can be handled by the student. The intent is not to prove or disprove theory. (I remember that in one of my labs from undergrad we made aspirin. Now that that is sort of cool, but the intent was not to test an idea.)
If you don’t design the experiment with the intent of being technically difficult though, really simple experiments can be performed. For example, it’s pretty easy to figure out whether its true whether plants really need sunlight. Or if they really need oxygen.
I do acknowledge that there is a risk that the results won’t always come out the way you expect, but those cases can be considered an opportunity and investigated further.Report
I think part of what you’re describing is true of most (all?) disciplines. There’s a conception by the layperson of x subject as being “a body of knowledge to be memorized” and there’s a conception by the specialist that the discipline “is much more than that” or “not really that in the first place.”
Take history, for example. It’s not just about memorizing dates and events and people and narratives, although I do think it’s that to a greater degree than other historians I know But it’s also about framing interpretations and narratives from the facts and engaging in a discussion with others who have done similar work. And it can also be an exploration into how we know what we know. We have to formulate, at least implicitly, theories of causation and we have to decide on how something can be “known” in the first place, especially since there is a limited number of sources, none of which being purely “primary” (i.e., unmediated).
For another example, take my former attitude toward accounting, something that I–a historian and non-accountant–for a very long time did not appreciate as a “real” discipline. Until embarrassingly recently–within the last 10 years–I seriously believed it was just filling numbers in columns and doing addition and subtraction exercises. I’ve since learned that it is much more nuanced than that, and it’s as much about a process about how we can “know” and identify costs and about how businesses “speak.” I still don’t understand accounting in most of its particulars, but I have a much greater appreciation for it as a discipline.
So, to your point about whether there should be two classes/subjects, say, “science” (or “biology,” “chemistry,” “physics,” etc.) and “scientific inquiry” or “history of science” or whatever, I’m of mixed minds. Part of science–maybe its essence–is indeed the scientific process, but there’s also the technical, body-of-knowledge style stuff that needs to be learned or practiced in order to understand the language in which scientists speak to each other. That would seem to argue in favor of the “two subjects” hypothesis.
But at the same time, I would hope it could be taught in the same class, so students would be able to see and understand how what they are studying is part of an evolution of knowledge produced by a tradition of inquiry.* I’m also worried a bit about how a class on “scientific inquiry” might work in practice. I’m afraid it will devolve down into “Aristotle was wrong, but for thousands of years everyone believed him because of religious bigotry. Finally, a few intrepid heroes came along and disproved him. And now we have space travel, DVD players, and i-phones.” That’s not a criticism of your idea, really, because any good idea can in practice be done poorly. Just a reservation.
*To some extent, my high school science classes did this. Most of my science class textbooks had an introductory “what is science?” chapter that tried to get the student to consider what scientific inquiry meant, and how it some times worked in practice. And my teachers usually went over this first chapter, but usually didn’t refer back to it during the rest of the course. The inquiry into the why’s and wherefores usually gave way to really needing to make sure we understood the “up-spin” vs. “down-spin” distinction, or the citric acid cycle, or PV=NRT. Note that this was before high-stakes testing, so it can’t at least in that case be blamed on testing, not that such testing stands to make things better.Report
Most of mathematics is taught historically. We start with the whole numbers, introduce zero, proceed through multiplication, division, fractions, algebra, trig — far too much time is wasted on algebra — I’d be teaching the rudiments of the calculus as soon as people had a decent grasp of the supporting algebraic concepts. By the time we get caught up to modern times, these kids are so losted and disgusted, it’s rather like trying to turn someone into a decent pianist by only teaching the classics.
For a few people, the historical route might work. For most, it won’t. Introduce people to the experiments of Galileo, show them how Newton calculated the moon’s orbit (and got it wrong), the basics of astronomy, trajectories, approximation and the basics of finite element analysis. Emphasise the uselessness of complete solutions. Good means good enough. We have enough digits of PI to cover the earth but a few dozen suffice to work with the known universe.
Solutions solve problems. What’s a problem? The answer to that Most Useful Question is where the philosophers dovetail nicely with the mathematicians and most of that union is highly useful for the rest of life. Teaching people to think logically and scientifically is the most important of all skills in life.Report
You should probably take a gander at brain development before designing math classes.
One issue I’ve seen schools run into a lot lately is trying to teach Geometry to junior high kids. Doesn’t work. It’s bad enough for 9th graders, but it’s the really rare 7th or 8th grader whose brain development is far enough along to handle geometric proofs.
Bluntly put, the kids don’t have the hardware to do math like that. That stuff kicks in around 14 to 15.
Kids brain’s are unfilled adult brains, they’re not miniature adults with giant gaps in their knowledge. Their brains are different, and bits and pieces of it change as they grow. You can’t just treat kids as little ignorant adults. Doesn’t work. The best you get is kids skilled at faking it.Report
Certain aspects of geometry are well within the scope of the child’s mind as surely as telling time is within the grasp of a first grader.
Time lapse photography of the night sky, a compass, a protractor and a clock — have the kids predict where the star will be in one hour. Two hours. Three hours. Not with numbers, just putting a point on a picture. Did that with my own children. Funny how that hourly angle change works out to exactly half the angle traversed by the hour hand of the clock, isn’t it?
Letting kids play with K’nex toys is just great. I’m not going to teach them Euclid’s Axioms, that’s not for little kids. That comes after they work out the implications of working with rulers and compasses.
I simply do not accept the idea of Brain Development as some impediment to learning. Of course you start with application and proceed to theory thereafter.
Da Vinci: Although nature commences with reason and ends in experience it is necessary for us to do the opposite, that is to commence with experience and from this to proceed to investigate the reason.
And that where we’re shortchanging children. Little kids are physicists from the minute they can pick something up and drop it.Report
“I simply do not accept the idea of Brain Development as some impediment to learning.”
Excellent. Let me know when you get a 9-month-old doing calculus.Report
A couple of anecdotes: I took calculus when I was 14, turning 15 in the middle of the academic year. I got a 5 on the AP Calculus BC test, even though the class I took only covered the AB material. For those who have no idea what I’m talking about, AB covers one college semester’s worth of material and BC covers two, and 5 is the maximum possible score.
Nevertheless, I did not understand calculus. I’d memorized the procedures for integration and differentiation, but I didn’t really understand anything. I couldn’t do word problems because I didn’t know how to set up the integrals.
Three years later, with no math classes in between, I was working on my physics homework and everything suddenly made sense. The fundamental theorem seemed obvious to me, and I could easily set up integrals for volume, mass, or surface area of a solid.
Same thing happened with pointers in C. I first tried to teach myself C programming when I was 15, and just didn’t get pointers. I could write programs that used them based on copying patterns I saw in books, but I didn’t really understand what was going on. A few years later, it just clicked all at once.
Something must have happened to my brain in those three years to make it easier for me to understand these things.
On the other hand, I really didn’t have any problems with geometry when I was 12-13. Everything made so much intuitive sense that I didn’t even bother to study.Report
I simply do not accept the idea of Brain Development as some impediment to learning.
And some people don’t accept the existence of atoms or evolution. Doesn’t change reality a whit. You’ll find an awful lot of the worst failures in education — the stuff that just flat out failed so badly that it might have been better just to literally hit the child with a book and hope the knowledge transferred — were people with exactly that attitude.
“Neuroscience — fancy-pants ivory tower elitist BS! I can TOO teach a 4 year old set theory! He won’t just memorize a bunch of stupid rules and parrot it back to me, he’ll understand math on a deep level you morons can only dream of!”
Seriously — you’re wrong. And obviously so. To be right, you have to believe that a child’s brain is developmentally identical to an adult’s brain. Which is, you know, demonstrably false. Trivially so.Report
Erm, as soon as I can get that kid to throw a ball in the front yard, I’ll teach him. Taught my own kids how to conduct just such experiments.
Equipment: tennis balls, a surveyor’s tape and another kid.
Location: any sufficiently long run where the impact can be located.
Problem: How far can you throw the ball?
Observations: Throwing the ball straight out doesn’t go as far as throwing it into the air. But throwing the ball too far into the air won’t go as far as — get this — the optimal angle. And you don’t think about that with protractors and algebra. It’s estimating.Report
How so? I’ve worked with my own kids, my wife’s classrooms full of kids, taught them all how to measure the distance of a thrown ball, how to use coordinate sets on a square field with a tennis ball machine. What on earth are you trying to say? That I haven’t done it? That children can’t master concepts once they’ve conducted a demonstration?
Hitting kids with books doesn’t work. Books are only recipes, not the food itself. Big difference. I wouldn’t teach a child about an atom, not before he’s seen how distilled water won’t conduct a current and adding salt to that solution will conduct electricity. What is your point? That experience can’t be extrapolated into principles until someone is 14 years old? Don’t be ridiculous.Report
Heh. Every toddler has mastered the basics of set theory. As soon as he can say the words “Mine! ” and “No!”, we can just put in a big ol’ set of Curly Braces and teach ’em to use some commas and we’re off and running. Sick of this ankle biting.Report
Wow, Blaise, talk about missing the point entirely.
I pointed out something entirely common-sense, often overlooked, and based purely on pretty well understood biology: Kids aren’t born with adult brains, and the process of turning an ‘infant’ brain into an ‘adult’ brain takes a good 20+ years — and that does NOT include the ‘filling it with knowledge of society’ bit.
Which means when it comes times to teach kids, you need to be aware of what they can and cannot learn — because it would be a vast mistake to treat a 12 year old’s brain like it was a 25 year old’s brain.
And you dispute that (apparently not a firm believer in biology, I guess) and then toss out a few anecdotes which I suppose you feel somehow prove your point (a point which I’m unsure of, except you object to the very notion that the brains of children are unlike the brains of adults and thus should be treated differently).
*shrug*. I’ll tell you the same thing I’ve told many parents: What you’ve taught your kids and what they’ve learned may or may not be the same thing. (As Brandon Berg gave a delightful example earlier, mentioning how he was capable of working Calculus like it was a mystical machine he had the proper rituals for, but didn’t actually understand until years later).
Which frankly doesn’t really matter to you, because the only people you are teaching is your own kids and nobody really cares if they can fake something through rote until they’re older and actually learn it for real and it’s not going to affect anyone’s life.
But it DOES matter when it’s school boards and people setting education curricula, because it’s a real problem when you decide to teach kids something that 95% of them lack the necessary biological hardware to grasp yet. It turns out — and this is crazy — that 95% of them won’t learn it, no matter how good your teachers are or HOW much you scream about unions and the three R’s and it’s just a real waste of everyone’s time and money.
Especially if your educational plans rely on them understanding something later that you taught them well before they could physically understand it, which has been the case a few times.
In short, Blaise — whether or not your kids understood Algebra at 3 or 8 or whenever is immaterial (heck if I even know what age is a good age for that — I am NOT an expert in childhood learning or brain development, I just happen to know that the reasoning for geometric proofs is an issue for 8th graders and that research indicates 9th grade is a stretch for many because it was a local issue a few years back and I got real familiar with the research on that specific bit only) — but idiots who think 10 year old brains function like adult brains and play educational curricula around them (those people are rarely teachers, but often consultants. Strange that) is material.Report
Who’s missing the point here, Morat? Common sense won’t do, not when you’re trying to teach a principle. Kids are born with an inherent sense of reasoning. That’s why infants are putting everything in their mouths, throwing them, playing with them. Determining the possibilities of things, forming up the frameworks of consciousness.
Biology be damned. The very idea, that you’d now get up on your hind legs to tell me about the brain. I do AI for a living. I know enough about the brain to say our understanding of that organ is vastly imperfect and that ignorant people have been trying to reify it and reduce it to metaphor for centuries now, rushing to every sort of ignorant and hasty conclusion as you are now. I stay up on neuroscience, strictly in a professional capacity.
Of course a ten year old’s brain isn’t an adult brain. Before puberty, a child can learn another language without an accent. After puberty, that child can master another language — but he’ll always speak it with an accent. The child’s brain is plastic, his neurons are still wrapped in a myelin sheath. Puberty locks much of that plasticity by removing the myelin sheath. That’s when abstraction takes over. As the brain grows older, it encapsulates ever-larger constructs. Preferences form, personality develops, until at last we’re mostly creatures of habit. We think our every thought so fine and new. We’re considerably less unique than our hubris would tell us.
Every child learns differently. That much I did learn from writing and editing my wife’s papers for her master’s degree in bilingual special education. Some kids won’t be told anything but can derive the principle from guided experiments. Some learn visually, others by example. It’s all up to the kid.
I never taught by rote. That’s for idiots. Don’t you accuse me of that, you don’t know me well enough to say such a thing. I taught, and still teach, by example. I’ve been on school boards and steering committees.
I never pretended to be an Expert on Childhood Learning. I was a parent and I am a teacher. A good deal of what I do now is un-teach, substituting pragmatism for brittle theory and classroom dogma. I teach people to code and analyse. I’ve been teaching code since the 70s and analysis since about 2000. Mostly, I teach from the problem, believing every better definition of a problem leads to an incrementally better solution. I don’t teach from principles. But some principles have emerged, long-solved problems for which there is no better solution. They’re called Patterns in my line of work.
The guy who thinks he can just go to work and do what he’s told will end up sweeping the hangar and will never build the aircraft. That’s the difference between consultants and employees. I don’t have the inclination to sweep the hangar. Did it for long enough to know it’s a dead end. You’re rewarded with promotion — out of code and into management — where you’ll never write a line of code again.Report
The very idea, that you’d now get up on your hind legs to tell me about the brain. I do AI for a living.
No point arguing with a man who just said that seriously. Because “I do AI for a living” and the implied “I know all about the human brain” are connected only in fantasy land.
Jesus, even soft sci-fi authors gave up on the belief that machine learning would be anything like the human brain decades ago.Report
Caveat: the adolescent brain is myelinating some nerves, reorganising others. Huge changes in the frontal lobes. Many of the myelin-isolated neurons are pruned down. More hereReport
You’re not arguing. You’re just contradicting. No proof for a word you’re saying. Machine learning certainly isn’t going to produce a Shakespeare sonnet. But it’s doing lots of interesting things, even here. Our spam filter operates on Bayesian logic. Sometimes it misbehaves. It learns.
As for what I know about the human brain, I’ve specifically we don’t know very much and furthermore said you’re rushing to all sorts of hasty conclusions about it. Do try to support your opinions from what few facts you can summon up and quit this sophomoric attempt at gainsaying the obvious. Mathematics in schools is in a wretched state of affairs, mostly because all the college kids who could get through Calc II didn’t go into Education.Report
Great post.
Murali and Chris captured many of my initial thoughts above, but here are a few additional points…
I would suggest the way to distinguish the two categories or disciplines is to label one “the scientific method” and the other is science. Science is usually sorted into sub topics of course.
The more I have read on the scientific method, the more convinced I am that it is neither well understood or well documented. In brief the methodology differs dramatically between disciplines and in many cases within disciplines as you look at different research traditions. Furthermore, the scientific method popular today differs from that of the 19thC which differs even more of that practiced by the Royal Academy.
In other words, one strength of the scientific method is that it adapts and evolves. I think Hayek would agree that the institution of science is beyond human design. Research traditions and methodology and paradigms have evolved over time within the various disciplines along with the nature of the problems they solve.Report
More Henri Poincaré for you. Less of these Objectivists. They’re not objective enough.Report