Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

A scientist must go where the evidence leads

blogs.scientificamerican.com

Post reply on HN
Print view View original

51–60 of 103 posts

Re: A scientist must go where the evidence leads

#51

Post by jonmc12 »

> Rather than softening our interpretation of a conflict between our models and new evidence, we should always seek the simplest explanation and be willing to abandon failed models.

Sure, but what is exactly is "evidence"? The word is used as if observational data has no ambiguity or subjectivity is the way it is defined.

The book, Range (Epstein), had a great example of this in discussing the Challenger explosion in the last chapter.

The premise: several engineering teams in Nasa had a working theory/model about O-ring susceptibility to cold whether. However, their model was not backed up by the evidence available. Nasa made the call to launch based on data alone.

Why? the evidence collected for decision-making was incomplete and poorly presented. Range presents a thesis that over-reliance on a quantitative thinking can lead to optimization of the wrong thing and actually inhibits learning.

Related, "The Challenger: An Information Disaster" https://www.asktog.com/books/challengerExerpt.html - Tufte’s re-examination of the Challenger disaster.

Re: A scientist must go where the evidence leads

#52

Post by chousuke »

ta1234567890 wrote: post #25
chousuke wrote: post #22

Earlier quoted context omitted.

I'm not quite sure what it is exactly that you're trying to say, so apologies if I misinterpreted it, but to me you seem to suggest that science and religion are somehow complimentary; two different ways of seeing "truth". Two different "domains" of truthiness, in essence. If you're just saying that people can use religions to make themselves feel better by believing in things that are false, then yes, that's evident…

This is a philosophical issue, and it's also an ineffable issue. What is truth? Something that is the same for everyone at the same time? How could you ever prove something is true if you are not omniscient? (ie. you can only experience reality from your own point of view, but can never experience it from someone else's). To complicate things further, we are trying to use words and spoken language as a way to communi…

In the philosophical sense, I'd say truth is what reality is. In the practical sense, truth is what the evidence indicates, and anything else either remains unknown or is false if it contradicts the evidence.

We're limited to never knowing the complete truth of most things, which is why we have science so we can make our practical truth approximation ever closer to reality.

Re: A scientist must go where the evidence leads

#53

Post by UnFleshedOne »

wellpast wrote: post #29
chousuke wrote: post #22

Earlier quoted context omitted.

I'm not quite sure what it is exactly that you're trying to say, so apologies if I misinterpreted it, but to me you seem to suggest that science and religion are somehow complimentary; two different ways of seeing "truth". Two different "domains" of truthiness, in essence. If you're just saying that people can use religions to make themselves feel better by believing in things that are false, then yes, that's evident…

I mean this with no offense but I think you're reflecting the deficit in thinking I mention. E.g. here: > If you're just saying that people can use religions to make themselves feel better by believing in things that are false, I assume you're using science's rhetoric of false here. Which is really just can it be observed using the tools & vocabulary of science. > I have serious problems with this sentiment that ther…

Re: your last statement, you can say it is not science that gave people[0] their morals, and I would agree. Science didn't create humans, so it can't be blamed for human's properties.

But it is a job of science to describe how those morals emerge and why they are the way they are: yadda yadda, evolution, group selection, basic biological realities, some other natural mechanisms I haven't heard about. Thousands of man-years of research to be done yet I'm sure.

It is also a job of science to extend and improve existing moral framework where it fails due to internal contradictions. Because that framework was created by many millennia of biological and cultural evolution, mostly in small tribe hunter gatherer context, once you move people out of that context things start to work less well.

Things like scale insensitivity and loss aversion affect our moral intuitions. Depending on how the question is posed you can judge exact same outcome to be a good or bad, moral or immoral (no wonder philosophers spend all their time running more and more elaborate trolley experiments). For me that indicates that morals are more like evolved heuristics for cooperating in tribal environment than a set of divine rules bestowed by a triple-omni deity.

[0] - Both atheists and religious people get their core morals from the same place. What that place is, is being contended.

Re: A scientist must go where the evidence leads

#54

Post by mrmonkeyman »

ta1234567890 wrote: post #25
chousuke wrote: post #22

Earlier quoted context omitted.

I'm not quite sure what it is exactly that you're trying to say, so apologies if I misinterpreted it, but to me you seem to suggest that science and religion are somehow complimentary; two different ways of seeing "truth". Two different "domains" of truthiness, in essence. If you're just saying that people can use religions to make themselves feel better by believing in things that are false, then yes, that's evident…

This is a philosophical issue, and it's also an ineffable issue. What is truth? Something that is the same for everyone at the same time? How could you ever prove something is true if you are not omniscient? (ie. you can only experience reality from your own point of view, but can never experience it from someone else's). To complicate things further, we are trying to use words and spoken language as a way to communi…

Ricky Gervais said it quite eloquently: if you burn all religious fiction, in a thousand years none of it will return in the exact same form. If you burn all of science, all of it will be back in the exact same form. That is what reality is, it is not philosophical.

Re: A scientist must go where the evidence leads

#55

Post by dabbledash »

wellpast wrote: post #29
chousuke wrote: post #22

Earlier quoted context omitted.

I'm not quite sure what it is exactly that you're trying to say, so apologies if I misinterpreted it, but to me you seem to suggest that science and religion are somehow complimentary; two different ways of seeing "truth". Two different "domains" of truthiness, in essence. If you're just saying that people can use religions to make themselves feel better by believing in things that are false, then yes, that's evident…

I mean this with no offense but I think you're reflecting the deficit in thinking I mention. E.g. here: > If you're just saying that people can use religions to make themselves feel better by believing in things that are false, I assume you're using science's rhetoric of false here. Which is really just can it be observed using the tools & vocabulary of science. > I have serious problems with this sentiment that ther…

I agree that there are claims (normative, aesthetic, etc.) that are not amenable to scientific inquiry. But most religious people would claim that 1) a god exists and 2) that god has observable effects on the material world.

If claim 2 is true we should be able to use the usual tools to know it.

If claim 1 is true but claim 2 is false, that is practically indistinguishable from a situation where God does not exist at all.

Re: A scientist must go where the evidence leads

#56

Post by PavlovsCat »

kiba wrote: post #39
wellpast wrote: post #29

Earlier quoted context omitted.

I mean this with no offense but I think you're reflecting the deficit in thinking I mention. E.g. here: > If you're just saying that people can use religions to make themselves feel better by believing in things that are false, I assume you're using science's rhetoric of false here. Which is really just can it be observed using the tools & vocabulary of science. > I have serious problems with this sentiment that ther…

Now someone may report to be an atheist and report that they have a moral outlook that involves respect for fellow man. I guess I won't insist that its Religion; I simply insist its not Science that is giving them their moral outlook. When people claimed that god exists, then it is a claim that can be challenged. If someone claimed that Jesus Christ came to back to life, then it is a claim that can be challenged. Dit…

> If someone claimed that Jesus Christ came to back to life, then it is a claim that can be challenged.

True, but doesn't change that science ultimately deals with models we create to explain our observations. They're nowhere near as nilly-willy as, say, scientology, sure. But it's also not 100% objective, everything is filtered through us, or the models and instruments we make.

> And so in its actual procedure physics studies not these inscrutable qualities, but pointer-readings which we can observe, The readings, it is true, reflect the fluctuations of the world-qualities; but our exact knowledge is of the readings, not of the qualities. The former have as much resemblance to the latter as a telephone number has to a subscriber.

-- Arthur Stanley Eddington, The Domain of Physical Science (1925)

> Ditto for system of morality where certain aspects can be subjected to empirical testing.

I honestly have no idea what aspects of a moral system one could test empirically, could you give an example?

The way I see it, science can tell us that malnutrition in kids can affect their development and how, but it can't tell us whether it's right or wrong to let children starve, for example. It can't even tell us if it's better for people to be healthy or sick, happy or unhappy, and so on. It can at best measure physical reality accurately and even with some predictive power, but that still leaves all the moral questions (I can think of right now) to us.

Re: A scientist must go where the evidence leads

#57

Post by bookofjoe »

tabtab wrote: post #41

Even in research, when everyone else is zigging, it may be time to start zagging and pursue alternatives. Neural nets have a lot of attention, but maybe competitors will show sufficient promise once enough research is done. For example, Factor Tables: https://github.com/RowColz/AI Most of the research on early semiconductors focused on germanium. Manufacturers got pretty good at preparing it, but some started looking…

e.g. Blu-Ray v VHS

Re: A scientist must go where the evidence leads

#58

Post by krapp »

simonsarris wrote: post #24
wellpast wrote: post #18

> There are major aspects of our life that remain unexplained, leaving professional scientists with the choice of whether to accept the general notion that they will also be resolved one day by the scientific method. ... This ambiguous state of affairs explains why some of the best scientists are religious. This to me reflects huge deficit in the author's thinking and perception. It also reflects a common sentiment a…

Agree very much. It's a terrible state to be in. I think religion's job is not to explain things, as the very non-religious (and even now, most religious!) seem to think, but to leave you in wonder, to fill you with the mystery of the world. Demanding such an explanation from your religion seems like asking for the wrong thing. Like asking your cat to sub in as a human baby. You will be disappointed, and scratched. >…

>I think religion's job is not to explain things, as the very non-religious (and even now, most religious!) seem to think, but to leave you in wonder, to fill you with the mystery of the world.

Religions were created when there was no concept of a separation between a physical and metaphysical world, when everything natural was also supernatural. Religion's job is not only to explain things, but to assert itself as the only valid explanation.

Also, I disagree that religion is necessary for one to be able experience wonder and mystery in the world. One doesn't need to believe in faeries at the bottom of a pool to appreciate its beauty.

Re: A scientist must go where the evidence leads

#59

Post by dabbledash »

UnFleshedOne wrote: post #53
wellpast wrote: post #29

Earlier quoted context omitted.

I mean this with no offense but I think you're reflecting the deficit in thinking I mention. E.g. here: > If you're just saying that people can use religions to make themselves feel better by believing in things that are false, I assume you're using science's rhetoric of false here. Which is really just can it be observed using the tools & vocabulary of science. > I have serious problems with this sentiment that ther…

Re: your last statement, you can say it is not science that gave people[0] their morals, and I would agree. Science didn't create humans, so it can't be blamed for human's properties. But it is a job of science to describe how those morals emerge and why they are the way they are: yadda yadda, evolution, group selection, basic biological realities, some other natural mechanisms I haven't heard about. Thousands of man…

There is distinction between investigating how people’s moral intuitions work, which science can do, and answering the question “what is actually Good.” Many materialists would say the latter question doesn’t have an answer, because there is no such thing as morality outside of human opinion.

Re: A scientist must go where the evidence leads

#60

Post by gus_massa »

wellpast wrote: post #18

> There are major aspects of our life that remain unexplained, leaving professional scientists with the choice of whether to accept the general notion that they will also be resolved one day by the scientific method. ... This ambiguous state of affairs explains why some of the best scientists are religious. This to me reflects huge deficit in the author's thinking and perception. It also reflects a common sentiment a…

People use both science and religion to say what other people should do:

* What they should and should not eat.

* Who they should and should not marry.

* Who/What they should and should not kill.

When some people think that science backs their ideas and other people think that religion backs their ideas, then you got a discussion.

Post reply on HN