8 ms·

The key to the quote above is that nothing is proven right. That is the fundamental flaw with science, if you understand this, you understand science. In scie

The key to the quote above is that nothing is proven right. That is the fundamental flaw with science, if you understand this, you understand science. In science and therefore reality as we know it, nothing can ever be proven true. It is fundamentally impossible. Every single claim made by every single scientist since the beginning of science has never ever been proven true. This “flaw” is what drives most of the debate and misunderstandings about science in the cultural and political arena. For example... No scientist can ever prove global warming to be true... it is fundamentally impossible. Hence the debate rages on endlessly.

vlovich123 6y ago
Except science _can_ set error bounds. No, global warming can never be "true" from the perspective of 100% accurate with no room for error. Science is able to set error bounds though. The latest particle physics results come with "this theory is correct to within 99.99% of the theoretical model". That means that even if the theory is wrong, it can account for 99.99% of all observations you make. That is the power. Obviously climate change models aren't like particle physics so we don't expect such accurate error bounds (too many sources for error). However, the question you must ask is "are the error bars sufficient for the decision I have to make?". My conclusion thus is that unlike dietary science or pop psychology, the evidence here is very likely real, despite any doubts that may persist. I could of course be wrong. That's a fundamental truth one has to admit to oneself & other peers who understand how science works. That's not a political truth that would make sense to admit because saying "I can never know but I'm pretty sure" has been weaponized into "you don't know anything & you're wrong". Same reason when speaking among science-literate friends I say "global warming" but am careful to say "climate change" to everyone else because the warming aspect got corrupted into "you said it's global warming but winter was especially cold this year". Maybe there has been a massive social pressure within the scientific community into confirmation bias. Certainly the community does self-police & ridicule anyone who doesn't ascribe to it. Why do I think the theory is reliable & it's not a massive conspiracy (intentional or otherwise)? I trust that there's lots of very bright people who have studied the math, from within & without the field & validated the models don't have any fundamental issues of any kind (conceptual, numerical issues, computer sims are solid, etc). I trust that people outside of the field who have related degrees have validated huge swaths of it. I trust that technological advances have provided us with exponentially better sensors & monitoring and that has fed back into exponentially better simulations to validate models. And with all of that development, over the course of 20 years, the results haven't changed, no new hypothesis have borne out. The scientific community is both small & large. It's large in that there will always be some amount of bad behavior somewhere, bad or unethical science, wrong results, mistaken theories that take hold for a time, etc. It's small though in the sense that when there's a very real problem, it can be brought to light & it's really hard to suppress that knowledge. That's why the chorus about global warming has been increasing. Oil companies knew about this as a problem & suppressed their own voices to the debate because it would hurt their bottom lines. In fact, they often fund the opposition. This isn't a conspiracy theory. There are court documents showing this. A well-informed person can use all of this knowledge to make a guess about who to listen to. A less-informed person will let themselves be swayed by the opinion they want to hear & or as a "fuck you" to the scientific community for ruining their career choice.
- neonological 6y ago
  No. This is incorrect. You're talking about model accuracy and observation accuracy. I am talking about something far more fundamental. Using axiomatic logic and probability, you cannot prove anything in science. Even with observations that are 100% accurate with 100% precision. This literally has fundamental consequences on our interpretation of reality as we know it and has affected out perception of reality and our science as well. This occurs because at any point in time a new observation can be made that falsifies a theory. Let's say you have a hypothesis that all zebras have stripes. You can observe 500 zebras with 100% accuracy and see that all of those zebras have stripes. But at any point in time in the future you can happen upon a hidden island that has 2 million zebras on it that has spots instead. 500 observations is minuscule in the face of 2 million and it literally renders your initial hypothesis ludicrous. Zebras are creatures that are more likely to have spots then stripes is the complete U-turn conclusion based off of new observations. Keep in mind the new conclusion occurred regardless of how methodological and accurate your initial observations were. The accuracy of the observation is Completely and utterly irrelevant. Because at any point in time I can encounter another new island with 1 billion zebras that has stripes rendering my second conclusion completely wrong, again. This is the fundamental flaw of science. It is far more fundamental then limited accuracy in observational measurements. For example take newtons laws of motion. There is no 99% right or wrong on that model. Assuming that our observations are accurate, Newtons laws of motion are 100% percent wrong. Yes they may be accurate numerically to a certain extent but the theory has ultimately been falsified and we now know relativity is a more accurate description. However, keep in mind that even relativity is not "proven" it can Never be proven and it will always be open for a complete reversal the same way Newtonian motion was. In fact Newtons laws of motion is the perfect example. It was the ultimate example of scientific verification. All experiments pointed to the theory being completely accurate, to disbelieve the science was to disbelieve reality. It was at the time equivalent to disbelieving evolution. This is the fundamental flaw with science. Nothing can ever truly be proven. And everything even the fundamental pillars of reality we rely on today from Newtons laws to evolution can never actually be proven to be true, and is always open for a complete rewrite. This is the exact reason as to why people can pick and choose the reality they believe in, whether it be Christianity or evolution. Neither can in actuality be proven, nothing has and nothing ever will.
taylodl 6y ago
And denialists exploit the fact science a isn't process of proving things true. Scientific American recently had an essay on the Denialist Playbook that I found interesting - https://www.scientificamerican.com/article/the-denialist-playbook/ https://www.scientificamerican.com/article/the-denialist-pla...
- neonological 6y ago
  For sure. If someone can come up with a way to prove anything to be definitively true, most of these arguments would be over. It is this flaw in science that is the origin of all these controversial debates that surround science, religion, global warming and other controversial topics. Nothing can be proven and therefore reality will always be open to interpretation.
  - taylodl 6y ago
    Just to clarify, things can be proven false, they just can't be proven true. That's why we speak of a good scientific theory being falsifiable - there must be a way to show it's not true. The Scientific Method then becomes a process of incremental refinement. Newton's theory of gravity didn't become "wrong" just because Einstein's theory was more accurate - heck we still teach, and use, Newton's theory to this day in both high school and college. It's useful. We just know that all bets are off when velocities approaching the speed of light are involved. Turns out that's not usually the case in everyday situations. We suspect that Einstein's theory may not be the last word on gravity either - but the cases where the theory breaks down are getting more and more extreme. It's not like apples are going to start falling up from trees because we changed our theory!
    - neonological 6y ago
      Yeah I'm already aware of this. Almost no one else is though. Keep in mind though, Incremental refinement is not a given. There is still no way to know if each new theory is actually closer to the truth. In fact it may even be a step backwards. Additionally limited accuracy in observation tools actually make it impossible to falsify anything either. Can you trust that the critical observation was 100% accurate? But tbh this is just splitting hairs.
phkahler 6y ago
>> In science and therefore reality as we know it, nothing can ever be proven true. In engineering we take the science and use it to predict how a design will work. When things work as intended it is a confirmation that the science is useful, if not "correct". To me that is what science does. It allows us to make useful predictions that can inform decisions.
- neonological 6y ago
  No that is not science. Science is a process of falsification. You have a hypothesis and you attempt to falsify the hypothesis. What you're describing is a model that has stood the test of falsification. Someone comes up with a mathematical model that can predict the future. They test that model under science to see if they can falsify the model. If they fail to falsify the model then at best they say this model might as well be true because we couldn't determine otherwise. That is science. When you use the model to predict how something will happen in the real world for some real world application. That is more called "engineering." Engineering is application, science is best attempt verification.
  - phkahler 6y ago
    I think you're splitting hairs, but ok. You describe the process of science and how it produces models. I described how we use those models in engineering - not the science I suppose, but the output (useful models) from it. In a sense, engineering is hoping their efforts confirm the model not falsify it :-) I think there is some important stuff in this, so maybe splitting hairs to explain accurately is important.
    - neonological 6y ago
      I'm not splitting hairs. There are people who are called "scientists" who test models with scientific experiment. Then there are people are "engineers" who use the models. If our social structure splits the difference by occupation how am I splitting hairs? Literally if you didn't run an experiment with a hypothesis you aren't doing science, and people in society therefore won't refer to you as a scientist.
pdonis 6y ago
> No scientist can ever prove global warming to be true Or to be false. Focusing on "true" vs. "false" leads nowhere except, as you note, to endless debate. The real question is, do we have models that can make reasonably accurate predictions? That is the right question for ordinary citizens, concerned about what, if any, political policies they should support or oppose, to be asking about any scientific claims. In the case of global warming, the answer to that question is no; we have models, and they make predictions, but the predictions aren't very good, and they haven't gotten any better over the last few decades despite a lot of effort. That means we should be very careful putting much confidence in those models.
- neonological 6y ago
  Falsification in science is possible. If I hypothesize that all zebras have stripes then observing one zebra with spots falsifies the entire hypothesis. This is definitive. However, no amount of zebras that I observe with stripes can ever prove my hypothesis correct. I can observe 500 zebras all with stripes and at any point in time the 501st zebra can have spots. I can observe 1 billion zebras with stripes and the possibility still remains open that the next zebra I see has spots. Science is not symmetric. Falsification is possible, proof is not.
  - pdonis 6y ago
    > Falsification in science is possible. Falsification of really, really simple hypotheses (like "all zebras have stripes", or the classic "all ravens are black") is possible, yes. But no hypothesis of any real significance in science is that simple. In any scientific model of any significance, there are always ways to patch the model to account for new observations. In a model that is destined to be supplanted, the patching gets more and more cumbersome and less and less plausible over time; but there is no hard and fast breaking point at which there is too much patching, that's always a judgment call about which different scientists can disagree. Also, unless and until there is some new model available that can account for all the same observations, including the new ones, in a simpler and more intuitively plausible way, preferably also with more accurate predictions, scientists will continue to try to use the old patched model because there is no alternative.

vlovich123 6y ago

Except science _can_ set error bounds. No, global warming can never be "true" from the perspective of 100% accurate with no room for error. Science is able to set error bounds though. The latest particle physics results come with "this theory is correct to within 99.99% of the theoretical model". That means that even if the theory is wrong, it can account for 99.99% of all observations you make. That is the power. Obviously climate change models aren't like particle physics so we don't expect such accurate error bounds (too many sources for error). However, the question you must ask is "are the error bars sufficient for the decision I have to make?". My conclusion thus is that unlike dietary science or pop psychology, the evidence here is very likely real, despite any doubts that may persist. I could of course be wrong. That's a fundamental truth one has to admit to oneself & other peers who understand how science works. That's not a political truth that would make sense to admit because saying "I can never know but I'm pretty sure" has been weaponized into "you don't know anything & you're wrong". Same reason when speaking among science-literate friends I say "global warming" but am careful to say "climate change" to everyone else because the warming aspect got corrupted into "you said it's global warming but winter was especially cold this year". Maybe there has been a massive social pressure within the scientific community into confirmation bias. Certainly the community does self-police & ridicule anyone who doesn't ascribe to it. Why do I think the theory is reliable & it's not a massive conspiracy (intentional or otherwise)? I trust that there's lots of very bright people who have studied the math, from within & without the field & validated the models don't have any fundamental issues of any kind (conceptual, numerical issues, computer sims are solid, etc). I trust that people outside of the field who have related degrees have validated huge swaths of it. I trust that technological advances have provided us with exponentially better sensors & monitoring and that has fed back into exponentially better simulations to validate models. And with all of that development, over the course of 20 years, the results haven't changed, no new hypothesis have borne out. The scientific community is both small & large. It's large in that there will always be some amount of bad behavior somewhere, bad or unethical science, wrong results, mistaken theories that take hold for a time, etc. It's small though in the sense that when there's a very real problem, it can be brought to light & it's really hard to suppress that knowledge. That's why the chorus about global warming has been increasing. Oil companies knew about this as a problem & suppressed their own voices to the debate because it would hurt their bottom lines. In fact, they often fund the opposition. This isn't a conspiracy theory. There are court documents showing this. A well-informed person can use all of this knowledge to make a guess about who to listen to. A less-informed person will let themselves be swayed by the opinion they want to hear & or as a "fuck you" to the scientific community for ruining their career choice.

neonological 6y ago
No. This is incorrect. You're talking about model accuracy and observation accuracy. I am talking about something far more fundamental. Using axiomatic logic and probability, you cannot prove anything in science. Even with observations that are 100% accurate with 100% precision. This literally has fundamental consequences on our interpretation of reality as we know it and has affected out perception of reality and our science as well. This occurs because at any point in time a new observation can be made that falsifies a theory. Let's say you have a hypothesis that all zebras have stripes. You can observe 500 zebras with 100% accuracy and see that all of those zebras have stripes. But at any point in time in the future you can happen upon a hidden island that has 2 million zebras on it that has spots instead. 500 observations is minuscule in the face of 2 million and it literally renders your initial hypothesis ludicrous. Zebras are creatures that are more likely to have spots then stripes is the complete U-turn conclusion based off of new observations. Keep in mind the new conclusion occurred regardless of how methodological and accurate your initial observations were. The accuracy of the observation is Completely and utterly irrelevant. Because at any point in time I can encounter another new island with 1 billion zebras that has stripes rendering my second conclusion completely wrong, again. This is the fundamental flaw of science. It is far more fundamental then limited accuracy in observational measurements. For example take newtons laws of motion. There is no 99% right or wrong on that model. Assuming that our observations are accurate, Newtons laws of motion are 100% percent wrong. Yes they may be accurate numerically to a certain extent but the theory has ultimately been falsified and we now know relativity is a more accurate description. However, keep in mind that even relativity is not "proven" it can Never be proven and it will always be open for a complete reversal the same way Newtonian motion was. In fact Newtons laws of motion is the perfect example. It was the ultimate example of scientific verification. All experiments pointed to the theory being completely accurate, to disbelieve the science was to disbelieve reality. It was at the time equivalent to disbelieving evolution. This is the fundamental flaw with science. Nothing can ever truly be proven. And everything even the fundamental pillars of reality we rely on today from Newtons laws to evolution can never actually be proven to be true, and is always open for a complete rewrite. This is the exact reason as to why people can pick and choose the reality they believe in, whether it be Christianity or evolution. Neither can in actuality be proven, nothing has and nothing ever will.

taylodl 6y ago

And denialists exploit the fact science a isn't process of proving things true. Scientific American recently had an essay on the Denialist Playbook that I found interesting - https://www.scientificamerican.com/article/the-denialist-playbook/ https://www.scientificamerican.com/article/the-denialist-pla...

neonological 6y ago
For sure. If someone can come up with a way to prove anything to be definitively true, most of these arguments would be over. It is this flaw in science that is the origin of all these controversial debates that surround science, religion, global warming and other controversial topics. Nothing can be proven and therefore reality will always be open to interpretation.
- taylodl 6y ago
  Just to clarify, things can be proven false, they just can't be proven true. That's why we speak of a good scientific theory being falsifiable - there must be a way to show it's not true. The Scientific Method then becomes a process of incremental refinement. Newton's theory of gravity didn't become "wrong" just because Einstein's theory was more accurate - heck we still teach, and use, Newton's theory to this day in both high school and college. It's useful. We just know that all bets are off when velocities approaching the speed of light are involved. Turns out that's not usually the case in everyday situations. We suspect that Einstein's theory may not be the last word on gravity either - but the cases where the theory breaks down are getting more and more extreme. It's not like apples are going to start falling up from trees because we changed our theory!
  - neonological 6y ago
    Yeah I'm already aware of this. Almost no one else is though. Keep in mind though, Incremental refinement is not a given. There is still no way to know if each new theory is actually closer to the truth. In fact it may even be a step backwards. Additionally limited accuracy in observation tools actually make it impossible to falsify anything either. Can you trust that the critical observation was 100% accurate? But tbh this is just splitting hairs.

phkahler 6y ago

>> In science and therefore reality as we know it, nothing can ever be proven true. In engineering we take the science and use it to predict how a design will work. When things work as intended it is a confirmation that the science is useful, if not "correct". To me that is what science does. It allows us to make useful predictions that can inform decisions.

neonological 6y ago
No that is not science. Science is a process of falsification. You have a hypothesis and you attempt to falsify the hypothesis. What you're describing is a model that has stood the test of falsification. Someone comes up with a mathematical model that can predict the future. They test that model under science to see if they can falsify the model. If they fail to falsify the model then at best they say this model might as well be true because we couldn't determine otherwise. That is science. When you use the model to predict how something will happen in the real world for some real world application. That is more called "engineering." Engineering is application, science is best attempt verification.
- phkahler 6y ago
  I think you're splitting hairs, but ok. You describe the process of science and how it produces models. I described how we use those models in engineering - not the science I suppose, but the output (useful models) from it. In a sense, engineering is hoping their efforts confirm the model not falsify it :-) I think there is some important stuff in this, so maybe splitting hairs to explain accurately is important.
  - neonological 6y ago
    I'm not splitting hairs. There are people who are called "scientists" who test models with scientific experiment. Then there are people are "engineers" who use the models. If our social structure splits the difference by occupation how am I splitting hairs? Literally if you didn't run an experiment with a hypothesis you aren't doing science, and people in society therefore won't refer to you as a scientist.

pdonis 6y ago

> No scientist can ever prove global warming to be true Or to be false. Focusing on "true" vs. "false" leads nowhere except, as you note, to endless debate. The real question is, do we have models that can make reasonably accurate predictions? That is the right question for ordinary citizens, concerned about what, if any, political policies they should support or oppose, to be asking about any scientific claims. In the case of global warming, the answer to that question is no; we have models, and they make predictions, but the predictions aren't very good, and they haven't gotten any better over the last few decades despite a lot of effort. That means we should be very careful putting much confidence in those models.

neonological 6y ago
Falsification in science is possible. If I hypothesize that all zebras have stripes then observing one zebra with spots falsifies the entire hypothesis. This is definitive. However, no amount of zebras that I observe with stripes can ever prove my hypothesis correct. I can observe 500 zebras all with stripes and at any point in time the 501st zebra can have spots. I can observe 1 billion zebras with stripes and the possibility still remains open that the next zebra I see has spots. Science is not symmetric. Falsification is possible, proof is not.
- pdonis 6y ago
  > Falsification in science is possible. Falsification of really, really simple hypotheses (like "all zebras have stripes", or the classic "all ravens are black") is possible, yes. But no hypothesis of any real significance in science is that simple. In any scientific model of any significance, there are always ways to patch the model to account for new observations. In a model that is destined to be supplanted, the patching gets more and more cumbersome and less and less plausible over time; but there is no hard and fast breaking point at which there is too much patching, that's always a judgment call about which different scientists can disagree. Also, unless and until there is some new model available that can account for all the same observations, including the new ones, in a simpler and more intuitively plausible way, preferably also with more accurate predictions, scientists will continue to try to use the old patched model because there is no alternative.