Problem with science today and the war on reason

[gleem]

I am saying that people generally don't care about your reasons when your action is to take money or power from them. The objection is to the action and not to the reason.

... or make them do something that they do not want to do. So within the context of this thread there is no war on reason as such.

Is this "war" just a noisy refusal of a previously silent minority to accepts some facts from a few sciences that do not agree with their world view?

The noise being turned on by the implementation of unpopular policies.

Could it also be that public attitudes in general have not changed appreciably but have been amplified through social media? It is now much easier to find support for almost any view.

 

[russ_watters]

... or make them do something that they do not want to do. So within the context of this thread there is no war on reason as such.

No, it is still a war on reason, it's just that reason is the secondary target or perhaps even collateral damage. "I'm selfish so I'm not going to support funding carbon reduction" is not a viable debate stance even though it may be the truth. So instead they argue something external to avoid taking resopnsibility: "There is no AGW."

The anti-reason/science stance is just a means to the real end.

Could it also be that public attitudes in general have not changed appreciably but have been amplified through social media? It is now much easier to find support for almost any view.

Could be, yes. That may be hard to measure though. You would need an issue that predates social media and has been hammered on it, but hasn't changed for other reasons.

 

[Ryan_m_b]

Ok, then I think we do agree. It seems to me in medicine there are a lot of probability based statistics/predictions that don't have clear-cut cause-effect relationships or have a lot of factors that may influence the probability significantly, making it hard to isolate the impact of one in particular. I don't think your example is necessarily a great one, but I do agree the issue exists.

Those examples are ones where I'd say there are so many confounding factors with major influences that they often aren't useful...except as motivational talking points during a doctor's visit.

Epidemiology is more useful than just motivation (I'm pretty sure you'll agree but feel like someone needs to post it anyway). Whilst all the variables of a specific patient's life over a significant period of said life are difficult-impossible to determine (and not all the variables are known) it's still useful medical advice to work off of the statistics that have been validated. For example: if controlling for all known variables it's observed that smokers are 90% more likely to contract lung cancer than non-smokers it's sound medical practice to advise against smoking, even if you can't be completely sure that your patient would fall into the 10% cancer-free smoker group.

As a general note there have been posts in this thread disparaging the social sciences, and even medicine, as not science. I'm a little surprised and quite dismayed at that. If scientists disparage each other's fields without acknowledging how scientific theory is being applied what hope does the layman have of figuring out what is legitimate?

 

[jack action]

No, it is still a war on reason, it's just that reason is the secondary target or perhaps even collateral damage. "I'm selfish so I'm not going to support funding carbon reduction" is not a viable debate stance even though it may be the truth. So instead they argue something external to avoid taking resopnsibility: "There is no AGW."

This is a logical fallacy that would certainly convince me to NOT follow people advocating for actions against global warming. If this is considered an argument, than it is a really bad one.

It is not that people necessarily don't want to support funding carbon reduction. Because where the money goes is of real importance. Where science is failing in this particular debate is that the burden of proof about which methods are valid has not been met. Heck, nobody is even sure about what are the true repercussions of global warming.

Sadly, people with pecuniary interest in these solutions have tendencies to say that we must invest in their solutions or prove they are wrong. That is not how it should work: You must prove, beyond any doubt, that your method will give results. Especially if you will make money out of it. Especially if that money is government funded.

Last night, a telemarketer tried to sell me windows; I told him I didn't own a house and he hung up. It doesn't mean it was true. I'm not going to waste time evaluating the validity of his offer or check his credentials. Anything to get those leeches of my back.

Imagine if I wasted my time investigating every person that claims having the solution to save the world.

 

[russ_watters]

This is a logical fallacy that would certainly convince me to NOT follow people advocating for actions against global warming. If this is considered an argument, than it is a really bad one.

It isn't an argument, it is an example (and I accidentally picked one you care a lot about...). Maybe you oppose anti-AGW funding for better reasons, but not everyone does. And you must be aware that those who oppose, including those who oppose for non-factual reasons, tend to be those with a bias toward the status quo (Republicans). By the way, you can count me as being on the same side as you on that debate. But again - just an example.

If you don't like that one, try another one: Social Security. Most people are aware it will go bankrupt in the 2030s. Most people are also aware that the longer we wait to do something about it, the worse the issue (the more expensive the fix) gets. But there are people who argue that nothing needs to be done now or even at all. And looking at the demographics tells you they are predominantly those who are currently getting or soon to be getting the benefits.

Those who do/will benefit from the flawed program support the status quo and those who expect not to support changes. Greed vs greed. Heck, it doesn't even matter if one happens to be on the right side of the facts: do you really think a large fraction of today's young people won't flip their positions when they get close to retirement age?

 

[StatGuy2000]

If you don't like that one, try another one: Social Security. Most people are aware it will go bankrupt in the 2030s. Most people are also aware that the longer we wait to do something about it, the worse the issue (the more expensive the fix) gets. But there are people who argue that nothing needs to be done now or even at all. And looking at the demographics tells you they are predominantly those who are currently getting or soon to be getting the benefits.

Those who do/will benefit from the flawed program support the status quo and those who expect not to support changes. Greed vs greed. Heck, it doesn't even matter if one happens to be on the right side of the facts: do you really think a large fraction of today's young people won't flip their positions when they get close to retirement age?

On a side note, I have already accepted the reality that Social Security, or its Canadian equivalent (the Canada Pension Plan, or CPP), will not be available and/or sufficient to support me financially when I'm close to retirement age. Of course, I have also conceded that I may never retire.

 

[jack action]

As a general note there have been posts in this thread disparaging the social sciences, and even medicine, as not science. I'm a little surprised and quite dismayed at that.

I think it's more about separating a scientific fact from a statistical analysis which are often used in social sciences and medicine.

What I consider "science" are statements like this: «If I drive a car at 50km/h for 2 hours, I will travel 100 km.» That is a fact that is indisputable.

For example: if controlling for all known variables it's observed that smokers are 90% more likely to contract lung cancer than non-smokers it's sound medical practice to advise against smoking, even if you can't be completely sure that your patient would fall into the 10% cancer-free smoker group.

This is a perfect example on how you can mislead with statistics. Anyone reading this automatically imagine that 9 out of 10 smokers will get lung cancer. But it actually compares risk with non-smokers. What is the risk of getting lung cancer for a non-smokers?

More real numbers (with a positive spin) are that 84% of male smokers will NOT have lung cancer; That number goes up to 90% for women smokers. Even considering "heavy smokers", 75% of them won't get lung cancer. These numbers don't say to me "if you smoke, you will get lung cancer." Far from it.

I don't mind getting those analyses such that I can make an informed decision. They may even lead us to true scientific facts about smoking one day. But I do mind having the government making decisions for me or taking my money to promote certain behaviors based on these type of statistical analyses.

 

[Ryan_m_b]

I think it's more about separating a scientific fact from a statistical analysis which are often used in social sciences and medicine.

What I consider "science" are statements like this: «If I drive a car at 50km/h for 2 hours, I will travel 100 km.» That is a fact that is indisputable.

Are you saying you'd only consider a priori knowledge to be science? Because outside of mathematics most scientific fields would fail this criterion.

This is a perfect example on how you can mislead with statistics. Anyone reading this automatically imagine that 9 out of 10 smokers will get lung cancer. But it actually compares risk with non-smokers. What is the risk of getting lung cancer for a non-smokers?

I don't think odds ratios are inherently misleading but you're right in that they can be used as such, frequently by the media that take detailed peer-reviewed articles and reduced it to headlines along the lines of "BACON DOUBLES RISK OF CANCER!"

More real numbers (with a positive spin) are that 84% of male smokers will NOT have lung cancer; That number goes up to 90% for women smokers. Even considering "heavy smokers", 75% of them won't get lung cancer. These numbers don't say to me "if you smoke, you will get lung cancer." Far from it.

I'm sceptical as to the credibility of that site. But regardless a simple pubmed search shows that the OR in large scale studies is calculated to be >100 for men and >60 for women:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3296911/

Given the percentage chance of developing lung cancer and the fact that smoking can increase this chance by orders of magnitude it's not misleading to advise against smoking on the basis of that. Unless I'm misunderstanding your point?

I don't mind getting those analyses such that I can make an informed decision. They may even lead us to true scientific facts about smoking one day. But I do mind having the government making decisions for me or taking my money to promote certain behaviors based on these type of statistical analyses.

Am I misinterpreting this or are you seriously suggesting that there is an open debate about whether or not smoking causes lung cancer...?

 

[jack action]

Maybe you oppose anti-AGW funding

If you don't like that one, try another one: Social Security.

It's not about what is my opinion about any of those subjects. It's respecting everyone's opinion. To get back closer to the thread subject, it's socialism that I don't like and I think it will hurt the trust people can have in science.

Socialism is about forcing everyone to go in the same direction. If it's the right direction, we win a lot. If it's the wrong direction, we loose it all. Statistically, we are bound to make a wrong decision. That is why I prefer letting people make their own choices where some will loose but some will also win. We need survivors. We need people who chose the right way, that we can follow, even if it was pure dumb luck.

Too many people use socialism to fund their projects. Even if there was good intentions initially, a big pile of money often attracts the wrong crowd and that increases the risks of wrong decisions.

In recent history, we showed the extend of what could be achieved with scientific facts. It appears that people now solely use the "weak" sibling of science - statistics and probability - to define their findings as "scientific", especially for social programs or laws. To me, statistics are observations, not facts. A distinction that is not clearly understood by non-initiated. And when all of these social decisions won't work, the masses may put all extents of science in the same bag. IMHO, it has already begun. I wish something would be done before someone decide to burn all the books.

 

[gleem]

That is why I prefer letting people make their own choices where some will loose but some will also win.

But isn't this anarchy or at least minarchy From here we only start moving more in the same direction.

 

[jack action]

Given the percentage chance of developing lung cancer and the fact that smoking can increase this chance by orders of magnitude it's not misleading to advise against smoking on the basis of that. Unless I'm misunderstanding your point?

First of all, you are still quoting ratios. For me, the real question is: "If I smoke, what are the odds I will get cancer?" That is a simple question to answer: You count all the smokers and you count how many got lung cancer. This is the odd I'm interested in, for smoking or any other thing I'm interested in.

Second, I appreciate the research and that some expert advises me about the subject. Though, I won't appreciate if that expert takes the decision for me. For example, I like knowing what are the odds of survival in an accident I'm getting with or without a seat belt, such that I can make an informed decision; I don't like having a law forcing me to use a seat belt.

Am I misinterpreting this or are you seriously suggesting that there is an open debate about whether or not smoking causes lung cancer...?

Can you definitely state that if someone smoke, he will get lung cancer? If not, then there are effects that you don't fully grasp and more research is needed. But the greater the odds means that you are probably on the right path.

When I hear those numbers, I tend to think about those who don't get lung cancer. Are they superheroes? How can they protect themselves from all of this smoke they are inhaling? I wish there were a lot more studies about these people. Instead, we tends to attack them because they contradict and disprove the anti-tobacco lobby.

 

[Bandersnatch]

Can you definitely state that if someone smoke drives drunk, he will get lung cancer die in a car accident? If not, then there are effects that you don't fully grasp and more research is needed. But the greater the odds means that you are probably on the right path.

When I hear those numbers, I tend to think about those who don't get lung cancer die in a car accident. Are they superheroes? How can they protect themselves from all of this smoke they are inhaling alcohol they're ingesting? I wish there were a lot more studies about these people. Instead, we tends to attack them because they contradict and disprove the anti-tobacco anti-booze lobby.

 

[jack action]

But isn't this anarchy or at least minarchy From here we only start moving more in the same direction.

Not necessarily. Let's take house insurance for example. I can choose to which insurance group I will belong. I can choose to not get insurance at all. We can still get together to do stuff. The fun thing is that I can also be part of a different insurance group for my car insurance.

It's the same thing for pension plans or unemployment insurance. I don't understand the importance of having one, mandatory, nation-wide plan. The only thing I agree as a social measure is that you give the minimum required for survival to those in needs. Other than that, let people make their own choices with their own money. That is best example of a democracy that works.

 

[PeterDonis]

I'm not sure the implied analogy here is valid.

First, I think we understand the causal factors that connect drunk driving with accidents better than we understand the causal factors that connect smoking with lung cancer. So we have more of a basis to argue that people should not drive drunk, over and above just quoting the statistics. (It also helps that the connection between drunk driving and accidents covers a much shorter time span than the connection between smoking and lung cancer, so it's easier to convince people on a gut level that it exists and get them to act accordingly.)

Second, I think it's less likely that there is a useful common factor to be discovered among people who drive drunk and don't die in accidents (other than "dumb luck") than that there is a useful common factor to be discovered among people who smoke but don't get lung cancer. So I'm not sure that an intuitive sense that there's not much to be gained by studying people who drive drunk and don't die in accidents should carry over to smoking/lung cancer.

 

[jack action]

@Bandersnatch :

Sorry, even if you meant to be funny, I totally agree with what you corrected.

Right now, we stop vehicles on the road for no reasons at all. We test the drivers for alcohol and give them a criminal record if they go over the limit (which, funny thing, can often vary according to the type of license you have, like a new driver or a truck driver). Around here, drunk people sleeping in their car are arrested and given a criminal record. The law states that you cannot have control of a vehicle while drunk. Pure insanity.

Think about it: A criminal record. Like someone who stoled or killed. What have they done wrong? Nothing. Too often, you couldn't even identified erratic behavior while driving. The crime they committed is that they chose to take a risk.

Here, we get the same kind of statistics from the anti-booze lobby: In all accidents, 40% had alcohol involved. That doesn't say what are your odds of having an accident while drunk. The real question I want to answer is: "If I drive drunk, what are the odds I will get an accident?" That is a simple question to answer: You count all the smokers drunk driver trips and you count how many got lung cancer ended up in an accident. This is the odd I'm interested in, for smoking drunk driving or any other thing I'm interested in.

To me, it's like giving a criminal record to someone playing baseball because he could of hurt someone swinging a bat. Can we wait until the person actually do harm or declare his intention to do so before considering him a criminal?

Deciding the correct behavior for everyone in all circumstances will not end well for our societies. Anyway, why would anyone want to live in a society where they don't trust others to do the right thing? That's insane. It's certainly not solidarity.

Don't get me wrong, you should get arrested by the police if you drive too fast, erratically or in any dangerous way, whether you are drunk or not. Does it require a criminal record? That seems harsh in my opinion. If someone's drunk, just force him to call a taxi and a towing.

What are the results of these harsh laws? They don't work because there are still drunk drivers. We are just getting more and more enraged about them. I refuse to live that way.

By the way, I don't drink or smoke, never did.

 

[gleem]

@jack action
Please stop posting I am having a hard time trying to keep up with your controvertible opinions. I haven't the time to respond. and I think a lot of this, although stimulating , is getting off topic.

 

[Bandersnatch]

First, I think we understand the causal factors that connect drunk driving with accidents better than we understand the causal factors that connect smoking with lung cancer.

How would you qualify 'better', though? DNA damage by carcinogens and the subsequent repair errors leading to development of tumorous tissue is as much of a fact as is impairment of attention, motor skills, and decision making when inebriated. Is one science better than the other? The only major difference I see is that the latter is easier to understand intuitively, since most of us have repeatedly experienced intoxication one way or another, while very few have such intimate familiarity with cellular processes.

On the second point, I also disagree. Both involve 'dumb luck'. In case of cancer it's pure chance which part of DNA will get damaged and what copying errors will pop up in the process of repairing the damage. Both also involve individual biology as a factor (e.g. better functioning repair mechanisms in one, and better alcohol metabolism in the other). Seems equally worthwhile to study those in both (the original post also asserted that these aren't studied).@jack action are you saying that, even though you must have experienced inebriation in your life, personally or otherwise, you're unconvinced that being drunk makes you a bad driver, and only a very specific and conveniently unavailable statistical data would make you change your mind?
I don't think the thread topic covers whether or not prevention should or shouldn't be implemented by our more or less benevolent socialist overlords, so I shan't comment. How we decide what is and isn't real seems right on track, though.

 

[jack action]

In order to stay on topic, I'll restate my opinion in other words and in a more general way.

Imagine we raise a generation of kids telling them that if you drive at 50 km/h for 2 hours you will travel 100 km. We tell them it's a scientific fact. We also tell them that if you drive drunk or over the speed limit, you will get into an accident. We also tell them science proves it. The kids don't know any better and have no reasons to doubt us.

The kids grow up and begin driving cars. They can easily verify that the first statement is true. Every time they wish to test it, it will be true. But if they try to drive drunk or over the speed limit or witness someone who does, they will most likely notice that there will be no accidents. Depending on how religiously we enforced this view about drunk driving or above the speed limit, they may become upset, like they've been lied too. They may start wondering if science can be wrong. What can they trust? «I drove at 120 km/h and nothing happened, maybe I can go 200 km/h and nothing will happen either. Who knows, science is unreliable.» «Is the Earth really revolving around the Sun or is that another science uncertainty?» You can quickly loose a whole generation this way.

That is the difference between "scientific facts" and stuff like "statistical analyses" or "computer modeling". And when statistical analyses are used by a group of people to nourish their greatest fear and pass it on to the rest of the society, you open the door to people disavowing them and their ways (i.e. science as whole) when they will realize that what was predicted did not happen.

When one says «God will always be by your side» to someone and his/her child dies, he/she tends to disavow religion. Same thing will happen to science if it is used to state things that science cannot well define (yet). IMHO, it is why some people are going back to religion thinking science is no better because of stuff like statistical analyses and computer modeling that get passed on as "scientifically proven".

 

[Bandersnatch]

That is the difference between "scientific facts" and stuff like "statistical analyses" or "computer modeling".

No, that's the difference between teaching the kids the concept of probability and not doing that. That A increases probability of B may be as much of a scientific fact as C equals D.

 

[PeterDonis]

DNA damage by carcinogens and the subsequent repair errors leading to development of tumorous tissue is as much of a fact as is impairment of attention, motor skills, and decision making when inebriated.

We can directly observe and measure impairment due to blood alcohol level. We don't directly observe and measure DNA-damaged cells developing into tumors; we infer it from statistical studies. That's what I mean when I say we have "better" knowledge of causal factors in the former case than in the latter.

Both involve 'dumb luck'. In case of cancer it's pure chance which part of DNA will get damaged and what copying errors will pop up in the process of repairing the damage. Both also involve individual biology as a factor (e.g. better functioning repair mechanisms in one, and better alcohol metabolism in the other).

Hm, yes, fair point.

 

[Student100]

As a general note there have been posts in this thread disparaging the social sciences, and even medicine, as not science. I'm a little surprised and quite dismayed at that. If scientists disparage each other's fields without acknowledging how scientific theory is being applied what hope does the layman have of figuring out what is legitimate?

I made a (perhaps the) post to that effect. The fact that they're not a science shouldn't be taken as disparaging. Engineering is not a science, but it has worth. Neither is history, but it too has worth. Certain fields in medicine are more akin to engineering and the use of "best practices." Sometimes it jumps the gun: for example the Zika scare, there was/is no good evidence that it causes birth defects, but "scientists" at the CDC would disagree: https://www.cdc.gov/media/releases/2016/s0413-zika-microcephaly.html

Medical science also generally suffers because it is neither space nor time invariant. One population's outcomes are not necessarily readily applicable to another different population. HIV causes AIDS, except when it doesn't; such as, in populations with CCR5 mutations or those able to effectively create broadly neutralizing antibodies that stabilize viremia counts. Nor is it time invariant, given sufficient time of co-evolution, HIV would no more cause AIDS then any of the other retroviruses we've found in our DNA.

Social science suffers from the same problems, but to an even greater extent. A study of one population has no bearing on another. The time scales that the "knowledge" of one study is applicable is appreciably short, at least when compared to even medical science. I also believe many social scientist take offense when you call them "not a science" because they depend on an argumentum ad verecundiam to give their studies weight.

Social science also has a greater capacity to influence policy and human life, partly because it invokes the authority of "science". Look at lie detectors? How many people lost jobs or ended up in prison due to pseudoscience? How long before we have fMRI "lie detectors"? fMRI's can tell you one thing, blood oxygen levels in various parts of the brain, but there is all sorts of nonsense out there that extrapolates this to mean something it can never actually tell you. Is there any actual science in the social sciences?

 

[Rive]

... The fact that they're not a science ...

Sorry, but it's no fact but only a (rather narrow-minded) opinion.
And jumping to conclusions based on opinions is just totally not scientific.

Some sciences follows different kind of rule sets (other than natural science) to seek their own truths which will be valid only within the boundaries of that science, but to deny the title is just like saying that Germany is not a 'country' because they don't have any Washington and White House.

I really do hope that people would rather find the common values under such topic - especially under such topic! - than try to covertly excommunicate whole other branches from the scientific community.

This whole 'science or not' thread within the topic is just destructive and invalid.

 

[Ryan_m_b]

Don't get me wrong, you should get arrested by the police if you drive too fast, erratically or in any dangerous way, whether you are drunk or not. Does it require a criminal record? That seems harsh in my opinion. If someone's drunk, just force him to call a taxi and a towing.

What are the results of these harsh laws? They don't work because there are still drunk drivers. We are just getting more and more enraged about them. I refuse to live that way.

Its a crime because the activity of drunk driving has been shown to increase the chance of an accident that can seriously injure/kill innocent third parties. Given that we don't have policemen in every car (or perhaps we should by your logic) that can instantly stop a person the moment they begin to drive unsafe we have prescriptive laws.

In order to stay on topic, I'll restate my opinion in other words and in a more general way.

Imagine we raise a generation of kids telling them that if you drive at 50 km/h for 2 hours you will travel 100 km. We tell them it's a scientific fact. We also tell them that if you drive drunk or over the speed limit, you will get into an accident. We also tell them science proves it. The kids don't know any better and have no reasons to doubt us.

That's not science. That's an a priori logical statement, it's no more scientific than "all bachelors are unmarried men". Empiricism is not required nor present in any way to explain or verify the statement. Again you seem to be claiming that the only real science is mathematics because anything else is just statistics. Is radioactive half-life not scientific? How about quantum tunneling? Enzyme kinetics? Statistical analysis and probabilistic phenomenon pervade science and they're entirely compatible with empirical philosophy.

As for your "thought experiment" with teenagers the issue is with your hypothetical teacher, not with science. The teacher should have properly explained scientific theory and probability rather than giving the blanket statement of "all drunk drivers will have an accident".

Can I just ask: what do you do Jack? What field of research do you work in?

 

[PeroK]

Imagine we raise a generation of kids telling them that if you drive at 50 km/h for 2 hours you will travel 100 km. We tell them it's a scientific fact. We also tell them that if you drive drunk or over the speed limit, you will get into an accident. We also tell them science proves it. The kids don't know any better and have no reasons to doubt us.

They would doubt it because they are human beings with a human brain. If there is one aspect to this whole issue it is how difficult it is to say exactly how and why people think what they think and do what they do. This problem does not succumb to the type of analysis that deals in basic certainties like this.

 

[Ryan_m_b]

Medical science also generally suffers because it is neither space nor time invariant. One population's outcomes are not necessarily readily applicable to another different population. HIV causes AIDS, except when it doesn't; such as, in populations with CCR5 mutations or those able to effectively create broadly neutralizing antibodies that stabilize viremia counts. Nor is it time invariant, given sufficient time of co-evolution, HIV would no more cause AIDS then any of the other retroviruses we've found in our DNA.

This is not an argument that the field isn't science, this is an argument that the field has many complex variables for researchers to discover and take into account. You seem to have a simplistic view of medical research in that biomed scientists throw out absolute statements oblivious to the exceptions. That's incorrect, we're well aware that there are many other factors to take into account and when designing a new therapy or advising on policy these are taken into consideration.

Social science suffers from the same problems, but to an even greater extent. A study of one population has no bearing on another. The time scales that the "knowledge" of one study is applicable is appreciably short, at least when compared to even medical science. I also believe many social scientist take offense when you call them "not a science" because they depend on an argumentum ad verecundiam to give their studies weight.

Social science also has a greater capacity to influence policy and human life, partly because it invokes the authority of "science".

Social scientists:

- Observe phenomena
- Hypothesise over the cause of said phenomena
- Rigorously work to characterise the phenomena
- Design experiments/studies on these phenomena
- Create predictive models of the phenomena
- Publish their findings so that others can test their models

Thus, they are scientists. Just because it's a much harder field of science with many variables and large scale experiments are often impractical/unethical doesn't make them less of a science.

Look at lie detectors? How many people lost jobs or ended up in prison due to pseudoscience? How long before we have fMRI "lie detectors"? fMRI's can tell you one thing, blood oxygen levels in various parts of the brain, but there is all sorts of nonsense out there that extrapolates this to mean something it can never actually tell you. Is there any actual science in the social sciences?

I have never known a social scientist to advocate lie detectors. In fact I'm unaware of their use outside of the USA, which I think is the only country where they are officially considered to work. Lie detection is also not a social science topic, it's a biological and psychological one.

 

[UsableThought]

I'm going to check back into this discussion with what I hope will be a responsible & restrained comment. Back when the thread was new, I took exception to comment # 17 from @Student100 ; however, my initial response was much too ad hominem & over-the-top. Plus, when asked to explain, I realized that I was so stupid with fatigue that I wouldn't be able to do so; which meant I shouldn't have made the initial comment either. I deleted as much as I could & said I might be back. Let me apologize again to @Student100, as he was correct to ask for clarification of what was an overly aggressive statement to begin with.

I still can't really write anything sensible at any length. The issues are deep and my knowledge of them is shallow. I do want to just say a few things:

1) The contentious debate over "what deserves to be called science" goes pretty far back - a Wikipedia survey article on Hard and soft science suggests that some early distinctions were being drawn as far back as the 1800s, with contemporary interest accelerating in the 1950s and 1960s. Plus, it's a deep topic in that more than one scientific community is involved. We can assume multiple literatures/events/conversations, some of the conversations cutting across fields, but many not.

2) Building on point 1 above, when something with this obvious a background pops up in a thread, I think it's a always a good move to acknowledge the background - i.e. that an issue has a lengthy history & isn't just ad hoc. As a parallel example: when a poster starts a thread with an off-the-cuff comment or question on something inside physics, but with no cite, the usual response is to ask for cites and not just speculation or assumptions based on personal experience. I think the same could be done here: given this is a well known issue within the history & politics of science, if folks really want to get into the question of "what is science," we could do some research in addition to offering our own personal opinion. I'm not saying it's a "must," just a "nice to have."

3) Getting past just our only personal opinion might also help avoid the danger that I believe @Rive may be pointing to in his comment #92 above, and that others have pointed to as well - i.e. it's risky to say "field X is not really science" if we ourselves don't practice within X; we may be harboring a belief we know more than we actually do. So some judicious restraint might be useful (something I'm still working on, obviously) along with maybe doing some reading or a lit search – if this is something we really care about. I'm not saying a lit search is required; just that the fact it is even possible suggests the depth of history involved & thus some reason for caution before making overly broad statements.

4) History aside, the comments I see going on at the moment have brought up a lot of the nuances I think are important. I especially like the dialogue developing out of comment #91 from Student100 & the corresponding responses from Rive (#92) and @Ryan_m_b (#95); some really good points are being made. Hope the thread keeps on in a productive vein.

 

[brainpushups]

Social scientists:

- Observe phenomena
- Hypothesise over the cause of said phenomena
- Rigorously work to characterise the phenomena
- Design experiments/studies on these phenomena
- Create predictive models of the phenomena
- Publish their findings so that others can test their models

Replace 'social scientist' with 'astrologer' or 'creation scientist' and your criteria still arguably all apply. Further clarification of criteria in your set may help, but since the line demarcating science from non-science isn't 100% agreed upon we probably won't reach consensus here. These criteria may be necessary (or maybe not) for something to be called science, but they are not sufficient.

Perhaps an issue that brings this point back to an earlier discussion in the thread is to recognize that, within any scientific discipline exist sub-fields that could be considered non-scientific. This is probably necessary since there aren't any fields of science that are complete (and may never be) – new information comes in, theories and models are augmented, modified, or discarded. If there is a public distrust of the overall scientific method then the only way to combat it would be through education.

Rather than debating whether or not a given field or sub-field can be classified as science the question should be how do you know that X is true? I wonder what the results of a study in public opinion about how to answer this question would be? Would there be a large number of strict rationalists? I doubt it. Instead I would expect that most would be more on the side of empirical evidence being the acid test for the truth of a claim. If this is indeed the general perception of the public then perhaps this is more of a scientific age then there has been in the past. However, I would speculate that most people have a very limited understanding of the interplay between rationalism and empiricism that is essential for growth in science because non-scientists or science educators haven't worked with theory in enough detail.

 

[Ryan_m_b]

Replace 'social scientist' with 'astrologer' or 'creation scientist' and your criteria still arguably all apply. Further clarification of criteria in your set may help, but since the line demarcating science from non-science isn't 100% agreed upon we probably won't reach consensus here. These criteria may be necessary (or maybe not) for something to be called science, but they are not sufficient.

I do agree with your general point that there are many aspects of scientific research that aren't strictly anything to do with the scientific method (like bioethics in pharmacology, ergonomics in engineering, linguistics in coding etc) but I don't think it's fair to say you can swap out social scientist with astrologer. Because if an astrologer or creation scientist was actually following the method then a) there wouldn't be so many non-predictive theories in either and b) when they get to the testing stage they'd reveal their models to be flawed. It's not just that those fields are bad science its that they're pseudo-science, having the veneer purely for authority purposes. On the other hand social scientists, like any other kind of proper scientist, adjust their models based on testing and new data. It might be a lot harder to gather the data on large scale populations compared to something mundane like an atom but that doesn't invalidate it.

 

[brainpushups]

Because if an astrologer or creation scientist was actually following the method then a) there wouldn't be so many non-predictive theories in either and b) when they get to the testing stage they'd reveal their models to be flawed. It's not just that those fields are bad science its that they're pseudo-science, having the veneer purely for authority purposes. On the other hand social scientists, like any other kind of proper scientist, adjust their models based on testing and new data. It might be a lot harder to gather the data on large scale populations compared to something mundane like an atom but that doesn't invalidate it.

I was comparing social science to pseudo science, not non-science. Sorry if that wasn't clear. I'm not saying that Feynman is the authority to trust here, but he is famous for arguing that social sciences are pseudoscientific. I'm assuming that some progress has been made since his death, but, as far as I know, there are no 'laws' yet. Some have argued in this thread that the statistical analysis inherent in the social sciences helps qualify them as scientific. I wonder how Feynman would respond.

 

[Ryan_m_b]

I was comparing social science to pseudo science, not non-science. Sorry if that wasn't clear. I'm not saying that Feynman is the authority to trust here, but he is famous for arguing that social sciences are pseudoscientific. I'm assuming that some progress has been made since his death, but, as far as I know, there are no 'laws' yet. Some have argued in this thread that the statistical analysis inherent in the social sciences helps qualify them as scientific. I wonder how Feynman would respond.

There are barely any laws in biology either, does that make it a pseudoscience? When you're working with complex emergent systems with a huge number of variables you're not going to be able to reduce the observed phenomena to simple laws. I'd say that Feynman was wrong here and tbh I have virtually no interest in what he may or may not think on this topic.

 

[brainpushups]

There are barely any laws in biology either, does that make it a pseudoscience? When you're working with complex emergent systems with a huge number of variables you're not going to be able to reduce the observed phenomena to simple laws. I'd say that Feynman was wrong here and tbh I have virtually no interest in what he may or may not think on this topic.

Fair enough. I only use Feynman as an example as he was outspoken and well known. It's hard to know whether we may someday be able to reduce biological systems to obeying simple laws. Maybe someday they will and then biology would be 'elevated' (note: I agree with sentiments above that just because something is not 'hard science' that it is not valuable) to 'hard science.' It happened with chemistry.

 

[Ryan_m_b]

Fair enough. I only use Feynman as an example as he was outspoken and well known. It's hard to know whether we may someday be able to reduce biological systems to obeying simple laws. Maybe someday they will and then biology would be 'elevated' (note: I agree with sentiments above that just because something is not 'hard science' that it is not valuable) to 'hard science.' It happened with chemistry.

You will never be able to reduce biology to simple laws. You're not going to get a neat little equation or absolute rule that describes cellular behaviour or pharmaceutic interactions. We may reach a point where we understand the underlying principles of chemistry so well that we can accurately model complex biological systems mathematically, but it's going to result in the same complex emergent behaviour as exists in vivo. It's not some question of limits, its a question of what biology is.

Also the idea that biology will be "elevated" to "hard science" as though the latter is some sort of special club for the better fields is laughable, no offense.

 

[UsableThought]

Sorry if that wasn't clear. I'm not saying that Feynman is the authority to trust here, but he is famous for arguing that social sciences are pseudoscientific. I'm assuming that some progress has been made since his death, but, as far as I know, there are no 'laws' yet. Some have argued in this thread that the statistical analysis inherent in the social sciences helps qualify them as scientific. I wonder how Feynman would respond.

There are barely any laws in biology either, does that make it a pseudoscience? When you're working with complex emergent systems with a huge number of variables you're not going to be able to reduce the observed phenomena to simple laws.

Something tickled my brain here while reading these two comments in order; and the word "reductionism" floated up.

Reductionism isn't quite the same as arguing that hard sciences are "better" ("more scientific", etc.) than soft sciences; but it seems awfully close. Now, I'm not trying to characterize anything that has been said in this thread so far, but in general, it seems to me there is a sort of lazy arrogance to saying "We in physics and the other hard sciences are better (morally, scientifically, whatever) than those squishy soft sciences who shouldn't even call themselves sciences"; and then, at the same time, when asked to help with the really tough problems of human society & physical environment - way tougher than cosmology, sorry - have nothing to offer because, guess what? "We say that's not science." Not, "Gosh that's tough, maybe certain aspects of the problem can be approached w/ similar techniques" but "That's not science."

That's not what has been said in this thread - fair enough - but doesn't it seem that is a chasm that could be fallen into if judgement is too severe & somehow the scientific community withdraws from these more difficult fields? The Wikipedia entry on reductionism, under the subtopic of science, summarizes the argument against over-application of reductionism pretty well:

Others argue that inappropriate use of reductionism limits our understanding of complex systems. In particular, ecologist Robert Ulanowicz says that science must develop techniques to study ways in which larger scales of organization influence smaller ones, and also ways in which feedback loops create structure at a given level, independently of details at a lower level of organization. He advocates (and uses) information theory as a framework to study propensities in natural systems.[19]Ulanowicz attributes these criticisms of reductionism to the philosopher Karl Popper and biologist Robert Rosen.[20]

The idea that phenomena such as emergence and work within the field of complex systems theory pose limits to reductionism has been advocated by Stuart Kauffman.[21]Ermegence is especially relevant when systems exhibit historicity.[22] Emergence is strongly related to nonlinearity.[23] The limits of the application of reductionism are claimed to be especially evident at levels of organization with higher amounts of complexity, including living cells,[24] neural networks, ecosystems, society, and other systems formed from assemblies of large numbers of diverse components linked by multiple feedback loops.[24][25]

Nobel laureate P.W. Anderson used the idea that symmetry breaking is an example of an emergent phenomenon in his 1972 Science paper "More is different" to make an argument about the limitations of reductionism.[26] One observation he made was that the sciences can be arranged roughly in a linear hierarchy — particle physics, many body physics, chemistry, molecular biology, cellular biology, physiology, psychology, social sciences — in that the elementary entities of one science obeys the laws of the science that precedes it in the hierarchy; yet this does not imply that one science is just an applied version of the science that precedes it. He writes that "At each stage, entirely new laws, concepts and generalizations are necessary, requiring inspiration and creativity to just as great a degree as in the previous one. Psychology is not applied biology nor is biology applied chemistry."

 

[Rive]

...sort of lazy arrogance..

Right. Since metaphysics cannot define itself as science (consequently: there is no scientific way to define science, regardless of the amount of words spent), without connections to practical means it tends to revert to authority and beliefs/religion.

I think as this conversation goes now the thread got completely derailed and is on a dead end already.

 

[jack action]

That's not science. That's an a priori logical statement, it's no more scientific than "all bachelors are unmarried men". Empiricism is not required nor present in any way to explain or verify the statement. Again you seem to be claiming that the only real science is mathematics because anything else is just statistics.

When you're working with complex emergent systems with a huge number of variables you're not going to be able to reduce the observed phenomena to simple laws.

First, let me start by saying that it is all science to me. If you make observations and hypotheses about your environment, it is science.

That being said, the methods that are used to make the hypotheses are not all equals.

My example about speed, time and distance is not simple logic nor it is just mathematics: It's a physics law, i.e. $v=\frac{dx}{dt}$. To be called a "law" in science there is a repeatability criteria which always requires a well defined environment (for example the speed of light is exactly equal to 299792458 m/s ... in vacuum).

Like you said yourself, the more complex the system is, the more difficult laws can be made. It usually is because there are too many variables involved and they cannot be isolated or some values are so great that they are humanly impossible to verify (ex.: It's difficult to prove what exactly happen 1 billion year ago, as nobody was there to share their observations).

If you use statistical analyses and find that in a given environment you can observe something in 99.99% of the cases, you pretty much met my criteria of repeatability. So statistical analysis is a valid scientific tool.

But the further you get from that certainty - no matter the tool you use - the less you get to claim your observation is a "scientific fact". Though, it doesn't mean I'm not glad you didn't make the observation nor the hypothesis. It does contribute to the advancement of science; In the very least, it proves what is not true.

Here's a betting game based on scientific evidences:

If a car goes at 50 km/h for 2 hours, how much how are you willing to bet it will travel 100 km? I'm willing to bet my life on it. Why? Because there is a well defined physics law that proves it.

If there is a smoker and a non-smoker, which one do you think will get lung cancer? I'll bet on the smoker without a doubt because there are scientific studies that says that the odds are 100:1. Am I going to bet my life on it? No, because it is very plausible that the smoker will not get lung cancer regardless of the non-smoker getting lung cancer or not.

If there is a smoker, would you bet he would get lung cancer or not? I'll surely bet that the smoker will NOT get lung cancer, because the odds are at least 3:1. That is also based on scientific observations and that is why I'm confident about it.

As you can see science is a good advisor in all cases. But right now, most people will squint about my third bet because they are used to hear the message from a strong lobby trying to hide this fact that doesn't say that all smokers will get lung cancer. Not getting lung cancer for a smoker is not an exception, quite the opposite. But some think that smoking must be "evil" and it must be shown that everyone who embraces it must lose in the end. So we treat others like 2-year-old and just tell what needs to be told in order for them to act the way we want. This use of these scientific results is not OK by me because the methods used in those studies are not good enough to define clear lines between right and wrong of these complex systems.

And the more we allow politics to use science in that way, the more we are going to get people disavow science.