[mdpopescu]

... and, of course, the rebuttals to this were mostly of the "no true Scotsman" flavor. "It's not the fault of scientists, it's the bad media". Let's ignore http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1182327/ (Why Most Published Research Findings Are False) and many other similar results - we have to defend scientists because!

Look - I'm an engineer. I use a sort-of Bayesian reasoning: I started from what I learned as a child and then kept adjusting probabilities. You (plural "you") wouldn't be surprised if I told you I believe most politicians are liars. Well, I got to "most scientists are wrong" the same way. Sure, I didn't actually talk to most scientists, or even heard about them; those I did hear about, though, are at best misguided.

In any case, my original argument wasn't about the scientists; my argument was: most people treat science as a religion. They feel very superior to the religious people who are accepting everything they are told uncritically and then turn around and accept everything that is prefixed with "science says". Ten years ago, I used to make fun of people for believing the Earth revolves around the Sun - I have yet to find a single one who could do better than "that's what scientists say". That is arguably worse than those who accepted the Ptolemaic model - for one thing, they could at least see that the Sun was rising and setting.

(I can rant on this subject for hours but I don't think this is the proper forum.)

[megaman22]

> ... and, of course, the rebuttals to this were mostly of the "no true Scotsman" flavor. "It's not the fault of scientists, it's the bad media". Let's ignore http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1182327/ (Why Most Published Research Findings Are False) and many other similar results - we have to defend scientists because!

It seems like the worst offenders are in biology/pharmacology/genetics/biochemistry. I'm not trained in any of those fields, but it seems pretty obvious that organic life should be orders of magnitudes more complex than mechanics, materials science, aeronautics, or any of the other man-made scientific fields. They are also comparatively young subjects. We don't really understand enough about how all the pieces fit together, or even what all the pieces are, in biological processes to make much in the way of definitive statements.

But seriously, if I hear another nightly news story about how food X prevents cancer based on a study of a dozen people or the current consensus on whether eggs are good or bad for you...

[jleyank]

Careful. Human biological response is sufficiently well understood to permit the design of antibodies (which are grown in fermenters) that treat the symptoms of disease (RA, for example). There are decades of successful pharmacological research based on receptor targets, signaling pathways, etc.

However, your point is valid in that knowledge of, say, digital engineering is easier to understand and investigate as the systems are simpler. Those who are successful in the tech field should keep this in mind when they talk of disrupting fields involving wet-ware.

[RodericDay]

I'm an engineer too, and I fully agree with you.

My particular peeve is engineers who kowtow to economics, putting it in a much higher pedestal than sociology or anthropology, because it is "mathy". The implicit assumptions are usually along the lines of "the assumptions seem obvious enough, and look at all those formulae in the transformations thereof!"

[gizi]

Science id defined by its method: to set up experiments in order to look for counterexamples for a theorem. It is pretty much impossible to set up experiments in economics. While I agree that sociology and anthropology do not satisfy the definition of science, economics does not either. Where are the experiments? Without experiments, no science.

[tpeo]

Do astronomers perform experiments?

This is the example used in methodological discussions of economics. I don't know if astronomers do actually perform experiments on some limited scale, but it seems that classicaly they relied on "luck" to test predictions e. g. testing general relativity through observations of eclipses. In social sciences, there is a somewhat similar thing called "natural experiment", which is the main test for the validity of predictions. In both cases, the researcher takes a more passive role in testing his predictions.

This can be dauting to researchers, but overall it's for the best. There are indeed experiments in economics, specially in game theory. But the majority of possible experiments would cross the boundary of what is morally acceptable, and well into morally abhorrent e. g. randomly restricting a sample's access to education to estimate it's effects on wages.

[marcosdumay]

It helps that astronomy is a branch of physics. In a sense, astronomers do perform experiments, on different conditions but with the same theory. (By the way, that's what bothers me most about cosmology, it seems to not agree with any small scale experiment performed.)

Well, certainly, natural experiments can lead to knowledge. But that's a much slower and more dangerous path, artificial experiments are better in every sense, except that they are mostly not available for economics. Anyway, I don't think the use of natural experiments are the bottleneck currently holding economics back.

[JoeAltmaier]

That's an antique notion. There are plenty of rigorous experiments being done in all of those fields.

[gizi]

For every theorem in sociology, for example, can you show us what experiment they set up, and what results they obtained and how we can repeat their experiment in order to find counterexamples for their theorem? From there, I will acknowledge the scientific status of that one, single theorem. I will still consider every other statement in sociology to be non-science. Seriously, they must produce such data for every single one of their theorems. Physics does that. Chemistry does that. Why would sociology be entitled to the same scientific status without putting in the same effort?

[darkmighty]

You're confusing science with mathematics. There are no theorems in science, it's all experimental. We have certainty on a few basic principles or models and make inferences based on those models.

I'm not going to defend sociology specifically because I don't know the field or it's state, but in every science (e.g. chemistry), you make a model for some restricted case to fit some data and you make assumptions on the scope of the model. Otherwise you couldn't say pretty much anything about anything: "We tested that proton and that one has a mass of X units. We can't say anything about this proton, thogh." -- the evidence piles up that the model has widespread vality. In the same sense, we can make assumptions on the scope of models. Of course, I imagine we have no hope for the time being of uniting the basic physical laws to that of sociology, simply because we don't have the power to understand the human brain just yet. I don't think refraining from modelling behavior is useful, and it will probably even help us better model the brain and generalize it's behavior towards "better" intelligence (AI).

[darkmighty]

Try telling that to this guy:

https://www.youtube.com/watch?v=gjVDqfUhXOY

[JoeAltmaier]

So we think that is a dangerous thing? Believing the best, tested results of experiments (science) is not superior to random religious tracts written thousands of years ago?

The critical difference between treating science as a religion, and treating religion as a religion, is that the science people have a good chance of being right.

Its rational and reasonable to be uncritical of scientific results - the scientist has already been critical for you. That's kind of the whole point.

[mdpopescu]

Just look at this statement:

_Its [sic] rational and reasonable to be uncritical of scientific results_

I do not believe that this person is sarcastic; I say that because I have talked to hundreds of people with the same opinion. People actually believe this. My authority is better than yours so you uncritically accepting your authority is wrong, me uncritically accepting mine is not only right but obviously so.

Ignoring the absurdity of the statement, and the mind-numbing appeal to authority - this also shows a lack of knowledge of history. When people used Aristotle as an authority 500 years ago, they did it using this same rationale: Aristotle had already done all the thinking, who are you to claim to know better?

This is just painful.

[nitrogen]

It's not an argument from authority, just a statement of the implied Bayesian priors. One who gives greater weight to scientific papers than to religious pronouncements is entirely rational in doing so. That is the entire well-argued thesis of the Relativity of Wrong essay we are discussing.

[RodericDay]

Unthinking faith in "the institution of science", particularly the denial that it can be co-opted for certain policy goals, and influenced by funding, and lack the foresight to control for important environmental variables, has and will continue to lead into stuff like eugenics, craniology, and austerity.

I'm still reeling from "Its rational and reasonable to be uncritical of scientific results"- smh.

[nitrogen]

Nobody used the words "unthinking" or "faith", or anything remotely resembling them. Nobody said it was ideal to be uncritical of scientific results, only that it was rational. Rationality is not binary, at least as it appears to be used in a charitable interpretation of the commenter's position. It can be more rational to be uncritical of scientific results (which we can define for the sake of argument any way we like, but probably involving peer review and replication) than to be uncritical of unscientific pronouncements.

Given the available time and energy (or lack thereof) for most of us to examine every scientific result in detail, it can even be more rational (as in best allocation of personal resources) to accept some scientific results uncritically than to study them in detail. I'll note that I would argue for tentative acceptance, with growing confidence over time myself, but the original statement isn't wrong. It's just being interpreted through different vocabularies.

P.S. This next part is not a response to you specifically, just a general comment. I'm really disappointed by the polarizing tone of this thread, the uncharitable readings of others comments, the implications of guilt by association, and the emotionally motivated downvote brigades. HN used to be better; we've discussed this essay before and it wasn't nearly so bad. So I'm disappointed, and I'm disappointed in myself for participating, but feel I must because of how much I like the essay and how important it is for people to understand the concept of the relative wrongness of science.

[voronoff]

He's right. It's an low cost way to be mostly right. There are much worse tradeoffs to be made.

Assuming that science news corresponds to scientific results will get one in trouble though.

[gizi]

The essence of science is its method: Setting up experiments in order to look for counterexamples for a scientific statement. Therefore, anybody who is uncritical of scientific results and does not desire to find counterexamples, is not engaging in science but in anti-science, which in turn is a dangerous and counterproductive ideology. The science-as-religion people do not have a good chance of being right. They are always completely wrong.

[JoeAltmaier]

Advocating positions shown likely true by science does not require a scientist at all. We don't all have to be in the lab, testing every conclusion. We have people for that.

The rest of us will do very, very well to believe the results of science. Because its the only game in town that's actually trying to get it right.

[jmaygarden]

This is painful to read. It leads to a statement like the following.

"The foremost scientific authority claims that the Earth is flat. There is no need to be critical of this scientific finding. All opposing viewpoints must be shunned, and proponents of those contrarian ideas ridiculed and ruined."

[nitrogen]

Science has an ongoing process for correcting itself. This is a false slippery slope ("leads to").

[jleyank]

Check out the referred-to video on Youtube re: accepting things without reflection. Here's a discussion of it:

http://www.brainpickings.org/2014/04/24/jacob-bronowski-asce...

Beware of dogma.

[cinquemb]

> In any case, my original argument wasn't about the scientists; my argument was: most people treat science as a religion. They feel very superior to the religious people who are accepting everything they are told uncritically and then turn around and accept everything that is prefixed with "science says".

We had a subject in my lab yesterday that after the experiment had a "heated" debate where he put physics vs religion with one of the volunteers who was religious. While I was in the control room tinkering with the software, I listened in on their conversation and could hear the smugness of the subject as he used science as a tool to try and "convert" the volunteer. I thought the volunteer handled the conversation quite well, but it seemed like quite the inquisition from the "scientist"…

[JoeAltmaier]

Its easy to be smug about science; its so often right. If you don't have time to research the subject yourself, I'd always advise taking the side of the scientist.

[cinquemb]

If one considers all the research papers that have ever been published and retracted, and could enumerate over them and calculate their "rightness" and notices that greater than half are "right" then that would be a fact. As of today, no one has done that, and to assume it to be "often right" is quite dogmatic.

I posit that if one doesn't have time to research the subject themself, one shouldn't form opinion about such subjects beyond that they now know such subjects exist. But I suppose asking that of human beings is too much :P

[JoeAltmaier]

You typed that on a portable device, right? And without any sense of irony about the unreliability of science.

The rise of a civilization based on technology and information is a towering testament to the successes of science. Lets not fool ourselves.

[cinquemb]

I didn't argue about the unreliability of science. Do I think such could be quantifiable by what information is available to humans today? Yes. Do I think humans without the aid of technology can say either way to the degree that science is unreliable? Maybe for tiny slivers like meta analysis papers on relatively minute subjects within science, but as a whole? I think not.

[DanBC]

That's foolish. Scientists have to work hard to avoid bias. A lot of them don't manage it. Most science isn't as rigorous as eg ATLAS at CERN at avoiding biases.

There are plenty of scientists who are fucking idiots.

[JoeAltmaier]

There are a vast majority doing good work. Your cell phone is a towering testament to this. To be fair, try making the converse argument: how many religious zealots avoid bias? Zero. Lets not condemn science on the basis of a strawman ("some scientists somewhere are not very good")

[cinquemb]

> There are a vast majority doing good work. Your cell phone is a towering testament to this. To be fair, try making the converse argument: how many religious zealots avoid bias? Zero. Lets not condemn science on the basis of a strawman ("some scientists somewhere are not very good")

Yet to uphold what people consider science today, you use one example. As if the vast majority of scientists had anything to do with the development of the cell phone (and use the word testament which has a non zero religious connotations), and give no lip service to how many religious people throughout the times have supported scientific endeavors…

[JoeAltmaier]

I leave other examples to the reader. Hint: point at anything near you. It was the result of science.

As for honoring the religious who supported science. I say, to the degree they pursued religion, they shortchanged their scientific accomplishments. Consider what else Newton might have done, had he abandoned his alchemy and astrology.

[pdonis]

> If you don't have time to research the subject yourself, I'd always advise taking the side of the scientist.

If you don't have time to research the subject yourself, it means you don't care about the answer anyway. So why should you take any side? Why not just accept that you don't know?

[nitrogen]

I think what's being argued here is that the probability of being right by blindly following scientific consensus is greater than the probability of being right by blindly following non-scientific anything. Nobody's saying that science is never wrong, only that it's less wrong than other sources of knowledge, has a process that makes it even less wrong over time, and when it is wrong, it's only relatively wrong in the way that is described by Asimov.

[pdonis]

Asimov wasn't talking about following scientific "consensus". He was talking about things that are nailed down by data. The roundness of the Earth wasn't a matter of scientific "consensus"; it was a matter of making observations that showed the curvature of the Earth's surface (vs. earlier observations which were consistent with flatness), then further observations that showed that the curvature varied from place to place (vs. earlier observations which were consistent with it being a perfect sphere), and so on.

The point is that what is "less wrong" is the understanding we build up by extending the range of our data and our ability to predict what new data that we haven't yet observed will look like. It's not anything we build up by "consensus". So a lay person, trying to figure out what to "blindly follow", should not be looking at "consensus"; they should be looking at what data we have, how reliable it is, what does it cover, and how well we can predict what we will see when we get further data. As you can see, I put "blindly follow" in quotes because you can't do all this blindly; you can't just ask what the "consensus" is. You have to actually look at the content.

[nitrogen]

By "consensus" I mean what you describe, the accumulation of data from multiple sources. But even a simple human consensus of scientists would be right more of the time than a simple consensus of non-scientists.

Edit: also, I was referring initially to the arguments from other comments, not to Asimov specifically.

[eli_gottlieb]

>Ten years ago, I used to make fun of people for believing the Earth revolves around the Sun - I have yet to find a single one who could do better than "that's what scientists say".

Wait... what? If the Earth didn't go around the Sun, why would we have seasons based on the angle of the Sun to the ground? Why would that angle continually cycle throughout the year? And what about the movements of the planets in the night sky, relative to Earth?

Sure, you could have a god who designed perversely complex features into the world for no causally sensible reason, but really, the simplest explanation for our available geological and astronomical data is that the Earth orbits the Sun.