9 ms·

It's easy if you're deciding between two masses for an electron. In that case you've already decided that some theory which describes electrons is the one for

It's easy if you're deciding between two masses for an electron. In that case you've already decided that some theory which describes electrons is the one for you--so in that context one is a fact that the other is just plain false. But it gets tricky in cases where you're deciding between facts whose enclosing theories are incompatible. Is gravity a force, or is it an acceleration? Each is a fact in its own theory--the more relevant question has to do with your situation: which theory is more useful? Your epistemology appears to allow for the absoluteness of facts to be itself an absolute fact. Mine doesn't. From your perspective I must appear to be out of touch with reality, and from my perspective you appear to be resting an awful lot on circularity. People have been having this debate for millennia, I don't think we're going to resolve it this year. I think that a more fruitful question would be this: If one highschool uses textbooks that acknowledge that the philosophy of science is not itself a science--and presents the various competing views, and another highschool uses textbooks that assert the absolute truth of scientific facts without justification--which highschool produces more parents that opt their children out of vaccinations? The nature of facts meta-epistemic, and therefore can't be a matter of knowable truth--in that domain, utility is all we have. The vaccination debate, however (I just picked an example from a hat, feel free to substitute your favorite pseudoscience) is a clash between two theories, one with great justification, and one with hardly any. It only persists because we don't educate people to evaluate competing theories, so low-effort alternatives appear to wide audiences as viable contenders.

pdonis 8y ago
> it gets tricky in cases where you're deciding between facts whose enclosing theories are incompatible. Is gravity a force, or is it an acceleration? This isn't deciding between facts. It's deciding between different theoretical interpretations of the same facts. The facts of gravity didn't change when Einstein discovered General Relativity. All that changed was the theoretical model. > Your epistemology appears to allow for the absoluteness of facts to be itself an absolute fact. I don't know what you mean by "facts" if facts aren't absolute. Once you do an experiment and record certain results, those results don't change. That's why we call them facts. If you are claiming that those facts can change when we change theoretical models, I have no idea why you believe that. If you're agreeing that those facts don't change when we change theoretical models, then at least we agree on that, and then it's just a matter of me thinking your use of language is highly idiosyncratic. > If one highschool uses textbooks that acknowledge that the philosophy of science is not itself a science--and presents the various competing views, and another highschool uses textbooks that assert the absolute truth of scientific facts without justification--which highschool produces more parents that opt their children out of vaccinations? You're assuming that opting children out of vaccinations is necessarily a bad thing. But it might not be. Consider: when you vaccinate someone, there is a balance between two risks. If the person is vaccinated, there is some probability that they will suffer harm due to the vaccination. If the person is not vaccinated, there is some probability that they will suffer harm from catching the disease the vaccine protects against. In the world of half a century ago, when I was getting all the usual shots as a child, the second risk was demonstrably much higher than the first, because all of those diseases were common enough that the probability of catching them if you weren't vacccinated was significant, while the probability of suffering harm due to the vaccine was small (but not zero). But in the world of today, where all of those diseases are extremely rare--precisely because of such a huge fraction of the population getting vaccinated in the past--the probability of catching the disease if you aren't vaccinated is much lower. So now even a much lower probability of some kind of harm from the vaccine itself can justify not getting vaccinated. And since probabilities that low are very hard to assess anyway, it is by no means unreasonable for some parents to be unwilling to vaccinate their children without having to have any wacky beliefs. For example, if a parent has a concern about the mercury in thimerosal (a common preservative used in vaccines), even if the risk of that harm is low, it might still be enough for a reasonable parent not to take the risk. Note that this logic also allows parents to adjust to further changes, such as the recent increase in incidence of a number of diseases like measles. Such an increase pushes the tradeoff back towards vaccination even if there are low probability risks from the vaccine that haven't been fully quantified.
- Majromax 8y ago
  > But in the world of today, where all of those diseases are extremely rare--precisely because of such a huge fraction of the population getting vaccinated in the past--the probability of catching the disease if you aren't vaccinated is much lower. So now even a much lower probability of some kind of harm from the vaccine itself can justify not getting vaccinated. I think this is also a good example of the exact opposite point, however, that there is a decline in authority. What you describe here is a rational assessment of risks, but it is also a collective action problem: the individual risk of infection is low precisely because everyone else vaccinates their children. A Nash equilibrium generates a suboptimal outcome, where disease prevalence increases to the point where the perceived risks are equal. Worse yet, this has negative externalities for others, such as those too young to take a vaccine or who are allergic, since they can no longer rely on herd immunity. Collective action problems like this are usually resolved with some kind of coordinating authority. In some environments that authority can be normal social suasion, but in others it needs to be explicitly regulatory.
  - pdonis 8y ago
    > the individual risk of infection is low precisely because everyone else vaccinates their children Yes, this is true. However, this is easy for individuals to take into account as the percentage of vaccinated individuals changes: if fewer people get vaccinated, the incentive for a particular individual to do so increases. So I don't see this in itself as a collective action problem. > A Nash equilibrium generates a suboptimal outcome, where disease prevalence increases to the point where the perceived risks are equal. Not quite until perceived risks are equal; until perceived costs are equal. In other words, the equilibrium (which can change over time as the percentage of vaccinated individuals changes--see above) is where the expected cost of getting vaccinated exactly equals the expected cost of not getting vaccinated. This in itself does not mean the outcome is suboptimal. However... > this has negative externalities for others, such as those too young to take a vaccine or who are allergic ...this does, since the externalities are not considered by individuals when making their risk-cost calculations. > Collective action problems like this are usually resolved with some kind of coordinating authority. In some environments that authority can be normal social suasion, but in others it needs to be explicitly regulatory. The main coordinating authority in our society seems to me to be schools requiring kids to be vaccinated. That sets a hard ceiling on the percentage of non-vaccinated children, since parents who do not want to vaccinate have to either home school or pay for some form of private tutoring. The question would be whether this needs to be a hard requirement, vs. simply an assessment of how many kids are actually vaccinated in the population and whether that percentage is low enough to have a significant negative impact on herd immunity. But even that's not really the question, since, as I noted above, individuals can factor in how much herd immunity there is when making their own vaccination decisions. The only real coordination issue is due to the negative externalities: some percentage of people can't get vaccinated and so herd immunity is the only thing protecting them. So the real question is whether the percentage of people vaccinated because they voluntarily choose to get vaccinated after considering their individual costs vs. benefits is low enough that the risk to the people who can't get vaccinated is unacceptably high because of decreasing herd immunity. I don't know if anyone has actually tried to analyze the issue this way.
vharuck 8y ago
>Is gravity a force, or is it an acceleration? Each is a fact in its own theory--the more relevant question has to do with your situation: which theory is more useful? I would call that a model, not a fact. And, as the saying goes, all models are wrong. If it's impossible to do an experiment that shows one model to be better than another in a situation, then the choice is meaningless. Both are fine.
- aeternus 8y ago
  I would like to see a tool that lists all modern scientific hypotheses and theories along with the published experiments that support or refute the theory/hypothesis. The theories should be linked to show what builds upon what. Wikipedia is probably the closest we have to this right now. As science gets more complex, we either need to create a way for people to perform this bottom-up reasoning themselves, or we need to appeal to authority. Appeal to authority is dangerous and anti-scientific. IMO, people are right to reject that.
  - pdonis 8y ago
    I would love to see this too. Unfortunately, at present I don't know if we even have an institution that everyone would trust to provide this information in an impartial way. Also, there are still judgments involved even with this task as you describe it. How is it decided which experiments support or refute a given theory/hypothesis? How widely do we cast our net for what theories/hypotheses get included? Do we impose requirements on the experiments themselves, and if so, what?
    - aeternus 8y ago
      If I were designing it, I'd have the experimenters determine support / refute. In order to be included, they would have to have a clear write-up of the experimental setup, detailed results, etc. All theories could be included, the better the theory, the more confirming experiments it should have.
      - 8y ago
DoofusOfDeath 8y ago
__MatrixMan__: I'm going to lose karma for having a meta-discussion on HN, but I'd like to spare you some misery... In my experience, the HN crowd will heavily downvote comments that question the metaphysics assumed in other readers' posts. Without regard to the validity or applicability of such a comment, it's likely to cost you karma. I don't think this forum will entertain the kind of discussion you're trying to have, sadly.

pdonis 8y ago

> it gets tricky in cases where you're deciding between facts whose enclosing theories are incompatible. Is gravity a force, or is it an acceleration? This isn't deciding between facts. It's deciding between different theoretical interpretations of the same facts. The facts of gravity didn't change when Einstein discovered General Relativity. All that changed was the theoretical model. > Your epistemology appears to allow for the absoluteness of facts to be itself an absolute fact. I don't know what you mean by "facts" if facts aren't absolute. Once you do an experiment and record certain results, those results don't change. That's why we call them facts. If you are claiming that those facts can change when we change theoretical models, I have no idea why you believe that. If you're agreeing that those facts don't change when we change theoretical models, then at least we agree on that, and then it's just a matter of me thinking your use of language is highly idiosyncratic. > If one highschool uses textbooks that acknowledge that the philosophy of science is not itself a science--and presents the various competing views, and another highschool uses textbooks that assert the absolute truth of scientific facts without justification--which highschool produces more parents that opt their children out of vaccinations? You're assuming that opting children out of vaccinations is necessarily a bad thing. But it might not be. Consider: when you vaccinate someone, there is a balance between two risks. If the person is vaccinated, there is some probability that they will suffer harm due to the vaccination. If the person is not vaccinated, there is some probability that they will suffer harm from catching the disease the vaccine protects against. In the world of half a century ago, when I was getting all the usual shots as a child, the second risk was demonstrably much higher than the first, because all of those diseases were common enough that the probability of catching them if you weren't vacccinated was significant, while the probability of suffering harm due to the vaccine was small (but not zero). But in the world of today, where all of those diseases are extremely rare--precisely because of such a huge fraction of the population getting vaccinated in the past--the probability of catching the disease if you aren't vaccinated is much lower. So now even a much lower probability of some kind of harm from the vaccine itself can justify not getting vaccinated. And since probabilities that low are very hard to assess anyway, it is by no means unreasonable for some parents to be unwilling to vaccinate their children without having to have any wacky beliefs. For example, if a parent has a concern about the mercury in thimerosal (a common preservative used in vaccines), even if the risk of that harm is low, it might still be enough for a reasonable parent not to take the risk. Note that this logic also allows parents to adjust to further changes, such as the recent increase in incidence of a number of diseases like measles. Such an increase pushes the tradeoff back towards vaccination even if there are low probability risks from the vaccine that haven't been fully quantified.

Majromax 8y ago
> But in the world of today, where all of those diseases are extremely rare--precisely because of such a huge fraction of the population getting vaccinated in the past--the probability of catching the disease if you aren't vaccinated is much lower. So now even a much lower probability of some kind of harm from the vaccine itself can justify not getting vaccinated. I think this is also a good example of the exact opposite point, however, that there is a decline in authority. What you describe here is a rational assessment of risks, but it is also a collective action problem: the individual risk of infection is low precisely because everyone else vaccinates their children. A Nash equilibrium generates a suboptimal outcome, where disease prevalence increases to the point where the perceived risks are equal. Worse yet, this has negative externalities for others, such as those too young to take a vaccine or who are allergic, since they can no longer rely on herd immunity. Collective action problems like this are usually resolved with some kind of coordinating authority. In some environments that authority can be normal social suasion, but in others it needs to be explicitly regulatory.
- pdonis 8y ago
  > the individual risk of infection is low precisely because everyone else vaccinates their children Yes, this is true. However, this is easy for individuals to take into account as the percentage of vaccinated individuals changes: if fewer people get vaccinated, the incentive for a particular individual to do so increases. So I don't see this in itself as a collective action problem. > A Nash equilibrium generates a suboptimal outcome, where disease prevalence increases to the point where the perceived risks are equal. Not quite until perceived risks are equal; until perceived costs are equal. In other words, the equilibrium (which can change over time as the percentage of vaccinated individuals changes--see above) is where the expected cost of getting vaccinated exactly equals the expected cost of not getting vaccinated. This in itself does not mean the outcome is suboptimal. However... > this has negative externalities for others, such as those too young to take a vaccine or who are allergic ...this does, since the externalities are not considered by individuals when making their risk-cost calculations. > Collective action problems like this are usually resolved with some kind of coordinating authority. In some environments that authority can be normal social suasion, but in others it needs to be explicitly regulatory. The main coordinating authority in our society seems to me to be schools requiring kids to be vaccinated. That sets a hard ceiling on the percentage of non-vaccinated children, since parents who do not want to vaccinate have to either home school or pay for some form of private tutoring. The question would be whether this needs to be a hard requirement, vs. simply an assessment of how many kids are actually vaccinated in the population and whether that percentage is low enough to have a significant negative impact on herd immunity. But even that's not really the question, since, as I noted above, individuals can factor in how much herd immunity there is when making their own vaccination decisions. The only real coordination issue is due to the negative externalities: some percentage of people can't get vaccinated and so herd immunity is the only thing protecting them. So the real question is whether the percentage of people vaccinated because they voluntarily choose to get vaccinated after considering their individual costs vs. benefits is low enough that the risk to the people who can't get vaccinated is unacceptably high because of decreasing herd immunity. I don't know if anyone has actually tried to analyze the issue this way.

vharuck 8y ago

>Is gravity a force, or is it an acceleration? Each is a fact in its own theory--the more relevant question has to do with your situation: which theory is more useful? I would call that a model, not a fact. And, as the saying goes, all models are wrong. If it's impossible to do an experiment that shows one model to be better than another in a situation, then the choice is meaningless. Both are fine.

aeternus 8y ago
I would like to see a tool that lists all modern scientific hypotheses and theories along with the published experiments that support or refute the theory/hypothesis. The theories should be linked to show what builds upon what. Wikipedia is probably the closest we have to this right now. As science gets more complex, we either need to create a way for people to perform this bottom-up reasoning themselves, or we need to appeal to authority. Appeal to authority is dangerous and anti-scientific. IMO, people are right to reject that.
- pdonis 8y ago
  I would love to see this too. Unfortunately, at present I don't know if we even have an institution that everyone would trust to provide this information in an impartial way. Also, there are still judgments involved even with this task as you describe it. How is it decided which experiments support or refute a given theory/hypothesis? How widely do we cast our net for what theories/hypotheses get included? Do we impose requirements on the experiments themselves, and if so, what?
  - aeternus 8y ago
    If I were designing it, I'd have the experimenters determine support / refute. In order to be included, they would have to have a clear write-up of the experimental setup, detailed results, etc. All theories could be included, the better the theory, the more confirming experiments it should have.
    - 8y ago

DoofusOfDeath 8y ago

__MatrixMan__: I'm going to lose karma for having a meta-discussion on HN, but I'd like to spare you some misery... In my experience, the HN crowd will heavily downvote comments that question the metaphysics assumed in other readers' posts. Without regard to the validity or applicability of such a comment, it's likely to cost you karma. I don't think this forum will entertain the kind of discussion you're trying to have, sadly.