The Case for Teaching Ignorance | Hacker News Reader

The Case for Teaching Ignorance

nytimes.com

56 pointstrevordixon26 comments

There's a lot to be said for this, especially in biology, where there's so much we don't know, and textbooks don't make this clear. Brain science is even worse. We still don't know how memories are stored, either the mechanism or the format.

Economics remains badly understood. The predictive power of economic models is low. Psychology and sociology are even worse.

I think the point here is not that the limits of knowledge are not know, but that the focus of course work often stops at what is know. If you are training future experts you should spend the vast majority of the time focusing on the interesting questions that don't have answers right now. The realm of the 'textbook' known is so small for many scientific domains that in an ideal world it would be relegated to summer reading (yes, I'm implying one should read most if not all of that 800 page textbook, or many curated sections from it).

"The realm of the 'textbook' known is so small for many scientific domains that in an ideal world it would be relegated to summer reading"

It seems like you're implying that the 4-8 years graduate students spend, starting in undergrad, learning what's already known about their field could be relegated to a single summer? That's fairly outrageous, given how many thousands of scientists have been building increasingly complex theories and categorizations of knowledge over the past few hundred years.

I'm exaggerating a bit but mostly because my experience is that the kinds of questions that are asked on tests a lower levels of science education are things that you simply know at higher levels and not because you were tested on them but because you have to use them and apply them every day. Better to have courses that expect you to have read all of Molecular Biology of the Cell in advance and present material with that expectation. That way students can see the real applications of the knowledge that we build on and not just see it as some stupid thing to be memorized for the next certificate. I get no prize for knowing IUPAC nomenclature, but I couldn't even start to do my work without it. The expectation should be as such, you get a prize for thinking creatively about how one could answer questions or for coming up with answerable questions that haven't been asked before.

I actually think that formal courses are an excellent way to learn 'textbook' material, but they are far more useful when you know you are going to use that knowledge in the future. The question for me is whether, at the highest levels, we should be focusing our courses, and rewarding people for the equivalent of learning how to walk when we need them to fly.

This sort of course isn't for medical school. Maybe undergrad, but to-be doctors have enough on their plate without focusing on the blank areas of scientific knowledge. Once you at a professional school the time for fostering curiosity is long gone. If you;ve dedicated much of your life to a subject then you are already curious.

I hesitate to say such a course could be useful at the highschool level. It would certainly encourage independent leaning and humble many subjects. But it would also become political. It's a fine line between teaching the limits of biological knowledge and preaching intelligent design.

Many MDs go into research.

yeah, not sure that last line fits at all

in the 20th century lots of new material

its importatnt to study the room to grow in every subject

and to constrast the positive space of what is know

with the negative space of what is not

most expertise is simply the ignorance of the unlearned

who hold certain people on pedastals because they don't

know better...think of doctors in the pre-civil war days

would the be considered "experts" today? highly questionable

yet they were the best resources that time/place could muster

every startup is exploring negative space in the market...that's growth potential

a new biz might be intelligent design, but that has nothing to do with evo bio

etc...

doctors are also scientists and biz people and human

And an 18th century doctors is no less a scientist because of his ignorance;

it doesn't make darwin a 'creationist' because he doesn't understand relativity or nuclear physics

obviously absurd example of course

There are plenty of medical practices that are used base on historic use and expert opinion (e.g. Trendelenburg position). Doctors should be aware of what the limits of our knowledge are about the efficacy of treatments.

If someone hadn't gotten curious in 2005, we might still be using the Trendelenburg position and potentially harming patients.

I think curiosity should always be encouraged, especially in professional school. I don't think it should be a separate class though, I think we should be teaching the ignorance in every class alongside the knowledge.

There is not a fine line between teaching ignorance and teaching intelligent design. Not at all.

It is quite important for medical school. Doctors give poor care when they say, "You don't have a problem," rather than, "Science does not know how to deal with this problem." Examples of this include "complementary" medicine that is supported by science (http://www.nytimes.com/2015/08/11/upshot/labels-like-alterna...) but which doctors are not necessarily educated about, and even worse, the false certainty around prostate cancer screening which has led to urination problems and erectile dysfunction and even death for many men who would not have died from prostate cancer (http://www.cancer.gov/types/prostate/research/overtreatment).

Doctors must know the limits of their scientific knowledge or they endanger the lifespan and the quality of life of their patients.

I'd love to see a similar approach to CS-related topics. We don't know how to solve many security problems (key delivery, etc). We don't know what exactly makes a good user interface (many psychological/natural/social factors that we aren't aware). And we're yet to know how to write a correct program, let alone what "correctness" is. We don't even know what's a good way approach the problem, or even discuss it.

Actually these topics are actively discussed at CS programs in higher education - MSc and higher. At least in the department I'm working at, the specific topics you name are very active conversation topics among students and faculty alike.

It took me nearly two decades working in IT before I was in a position where I was respected, secure and comfortable enough to be able to say "I don't know" in front of the business client.

If I had said that even after five years on the job I would have been hauled into a meeting room by my manager and dressed-down for giving a 'poor image of the Technology department'.

Unfortunately that's also one of the reasons we have so much buggy code; people are afraid to express their lack of knowledge of a particular subject, so they kludge on through with some code pasted from a Google search.

I'm not sure about the philosophy of it all, but I think the concept to be conveyed is an understanding of how large a domain of knowledge is, and an understating of how little of that domain you have removed the "fog of war" from. Of course, this is a Bayesian interpretation, believing everything exists, we have but to go discover and interpret it.

Do others perceive a domain of knowledge to grow and change as one interacts with it and asks questions?

I don't buy it. Any academic presentation of a subject includes the limits of current knowledge. Maybe this is not emphasized enough, but it's a matter of extent, not a qualitative refusal to acknowledge ignorance.

A great example is that mainstream academic economists are very open about not being able to explain why monetary stimulus works. At one point (maybe this goes back to Keynes or even further) people attributed it to sticky prices. But current research cannot come up with a reasonable mechanism based on sticky prices that would explain the magnitude of the effect of monetary policy. Currently people accept that what we observe on the macro level cannot be explained on the micro level.

Medical science has limits? I thought we proved we're just finite state machines that are occasionally jostled into unfavorable states that can only balanced by external chemicals.

Can't tell whether you're being sarcastic or didn't read the article.

"We are blind to our blindness." Daniel Khaneman

"But there are also unknown unknowns – the ones we don't know we don't know." SecDef Donald Rumsfeld

I don't get why he is laughed at because of that quote. I personally find it quite elegant.

The quote wasn't spoken eloquently but it was profound. He was speaking truth but it was also amazingly ironic. Why risk and assume so much about success in Iraq with the quagmire of obvious unknowns?

It blows my mind that we don't know if Japanese intestines are longer than Westerners even though it's a popular meme in Japan.

It blows my mind that there is some myth about the Japanese still floating around in Japan may have been started by some pasty white German in the late 1800's.

But how come we can't/haven't disproved it, this to me is more the question, unless the article is incorrect and it has been disproved.

Why don't we have average sizes for body parts on record? Surly this would have been step one in Anatomy.

https://medium.com/unpublishable-elsewhere/are-japanese-inte...

Seems more like she's trying to teach a class on relativism, a subject I wholeheartedly hope stays away from science. Additionally, epistemology already exists, and is a well established field of study.

Did you even read the article before typing out your smug post? The professor is not advocating ignorance. She is advocating an honest recognition of the bounds of knowledge and ignorance versus false certainty.

When did I make that claim? I said that I think what she's promoting is based more off of relativism, and that epistemology is already an established field; hell, it's quite literally the study of knowledge. So smug!

Reply on news.ycombinator.com