Page 1 of 21

Soft Sciences Vs. Hard Sciences

Posted: Wed Jan 16, 2019 8:07 pm
by NonZeroSum
teo123 wrote: Sat Dec 29, 2018 8:46 am
Let them hear both sides! Evolution AND creation! Civil rights AND white nationalism! No way that can go wrong...
Well, what usually goes wrong is that people think they hear both sides of the story, but they actually don't.
Think of it this way, the Flat Earth Society Forum makes you think you hear both sides of the story, but actually you don't hear a single one (the most active Flat-Earther member on that forum, JROA, is almost certainly a troll).
Or, think of the web-page where I presented my alternative interpretation of the Croatian toponyms:
http://flatassembler.000webhostapp.com/toponyms.html
I tried to explain what the mainstream linguistics claims and why I think it's wrong. Did I accurately represent the mainstream linguistics? To my best knowledge, yes, but it's quite possible that I unintentionally misrepresented it.
It's almost impossible to actually make people hear both sides of the story (none of them being misrepresented), even if you honestly want them to.
If people actually heard what creationists actually have to say (the conspiracy theories about the measurements of the parallaxes of the stars being falsified to make it look like stars are further away than they actually are, the conspiracy theories that scientists know that the half-time of some radioactive isotopes isn't constant but hide that...), I am pretty sure that even less people would believe in creationism.
teo123 wrote: Sat Dec 29, 2018 8:46 am
Rarely. And less and less as time goes on and the knowledge and resources required for advancement grow.
More often these people just waste time and resources, and do harm.
Well, yes, most scientific theories are wrong. But the same is true for the theories made by "qualified" people. And whether or not you are "qualified" is impossible to tell.
carnap wrote: Science is often pushed by those outside of the mainstream and credentials are irrelevant to scientific practice.
brimstoneSalad wrote: Rarely. And less and less as time goes on and the knowledge and resources required for advancement grow.
More often these people just waste time and resources, and do harm.
teo123 wrote: Well, yes, most scientific theories are wrong. But the same is true for the theories made by "qualified" people.
Red wrote: You haven't really learned anything at all from this experience, have you? brimstone went over this with you about 20 times in the Flat Earth thread.

No scientific theory has been disproved in about 100 years or so. Do you know what a scientific theory is? Can you list some theories in recent memory that have been disproved?
brimstoneSalad wrote: There are different magnitudes of wrongness and plausibility.

Being a little off about how electrons interact when the details are still fuzzy is different from being off on the shape of the Earth.
When theories created by actual scientists are wrong, they are mistaken hypotheses trying at an understanding we don't yet have. A simple mistake which, based even on good knowledge, seemed plausible. When pseudoscientific theories are wrong it's a very different kind of fractal wrongness.

Saying they're both usually wrong is deceptive because it fails to account for degree.

Also: Lay people rarely create actual theories, but hypotheses that fail to be precise enough to be testable (that is typical of pseudoscience).
teo123 wrote: Actually, I can. Remember Lysenkoism? Up until the late 1950s, the Mendelian genetics was not accepted by a significant number of biologists, who instead believed in Lysenkoism (which is now considered pseudoscience). And the government of some countries tried to implement science-based agriculture, taking Lysenkoism to be respectable science. And it had devastating results.
Red wrote: No, that was not considered a scientific theory, and was well known to be pseudoscience, even among the scientists who came out in support of it. It was promulgated by political activists, not scientists (biologists in particular) who knew the consequences that would ensue.

Firstly, this took place in the Soviet Union (so not a global consensus, and not representative of the entire community) under a dogmatic and oppressive government where they basically forced the scientists at gunpoint to discard the accepted theory of genetics in favor of a new one, and some were imprisoned or killed for not obeying orders.

So basically, it was not a scientific theory in any sense of the term, and was never honestly considered one.
teo123 wrote: Tue Jan 01, 2019 5:58 am That certainly played a role, but I don't think that negates my point. The truth is, up until the middle of the 20th century, many biologists didn't accept Mendel's genetics, and instead believed in something like Lysenkoism. And when the government tried to implement science-based agriculture (as was recommended by Karl Marx), it was misled by that.
True, the consequences would be less severe if the government protected free speech.
But, you are not arguing for a government that protects free speech, you are arguing for a government that censors "pseudoscience", right?
Also, the consequences would be even less severe had the government just left agriculture alone.
Red wrote: Tue Jan 01, 2019 9:28 amYou completely ignored everything I wrote and insist on citing this irrelevant nonsense and shoving words into my mouth.
Please, go back and read what I wrote, which addresses what you said. First, get an idea of what a scientific theory is, then see my response to you.

Why are you bringing up free speech? The Soviet Union had little of such a thing, and Stalin (who supported the campaign) was always quick to oppress. I assume you just want to get your anarchist/libertarian foot in. You are right though; I do support censoring of pseudoscience by the government, since the harm done otherwise is much too immense.
teo123 wrote: Tue Jan 08, 2019 5:10 amHow is that irrelevant nonsense? The truth is, the Mendelian genetics wasn't accepted by most of the biologists until the early 20th century. And, some biologists didn't accept it even until as late as the middle of the 20th century. And Stalin and Mao were misled by that.
Red wrote:It's completely irrelevant to the original point you were making. Have you forgotten what that was already?

I said there haven't been any scientific theories disproven in the past 100 or so years, you brought up Lysenkoism and how most biologists accepted it, I explained how it was not actually accepted by biologists who were forced to accept it at gunpoint, and how it was never considered a scientific theory by the community abroad. You did not address my point or my arguments at all. Go do that now.
Appreciate the cliff notes of how you misunderstood the conversation teo, but you said it all in your opening post. You want to blame your linguistic research failings on lack of communication in academia, so you have to keep up this fatalist conspiracy ridden imagining of how scientific theories are developed. It's just really sad you would think using the most blatent hoax in history, Stalin needing to build a cult of personality with such feats as miracle crops would qualify as evidence of how mainstream scientific consensus can be more wrong than you striking upon a linguistic theory.

If you really need to scratch an itch for discovering something, fly to the rainforest before Bolsenaro cuts it down and discover a new species, apparently that's still fairly easy to do within a few days/weeks and a keen eye. Otherwise join a citizen science group and take pride in documenting evidence that someone with millions of pounds of labratory tech and 10s of years studying the field can use to build on scientific theories. And work to get there yourself one day.

Anyone who wants to talk to you has to segment their posts and remind you a billion times to respond to the train of discussion. I get some of it is you just not understanding, but a lot of it is over-confidence, and it's really not fair to expect everyone visiting the forum to have to do this to get you to stay on point.

Re: Which Sciences Are the Most Useful?

Posted: Wed Jan 16, 2019 8:39 pm
by Red
Thanks for the post NonZero, this was definitely one of Teo's more irritating stupidities in this thread.

Re: Which Sciences Are the Most Useful?

Posted: Thu Jan 17, 2019 12:27 am
by teo123
I've never seen such a blatant misrepresentation from someone
If I actually misrepresented you, now it should be trivial to point out where.
Honestly, it seems to me now you are using a dishonest manner to make an echo-chamber here in which everybody agrees with you not only about food ethics and about religion (if you actually care about those things, and you don't seem to), but also about free speech being harmful and some weird assertions about the philosophy of science.
Perhaps what you want is to make vegans look "normal", and you think this is a way to do that. Perhaps in America those beliefs are actually considered normal. Here in Croatia, they aren't. Here, most of the people, including students, use the term "political correctness", if they use it at all, pejoratively. And many professors at the university I attend are openly right-wingers, including one who happens to also be a vegan.
You should read the conversation about social sciences Red and I had earlier in the thread.
OK, I skimmed over it a few times. Here are some things it seems to me you agree on, correct me if I am wrong.

1. Social sciences usually have something to do with ethics and politics.

2. In social sciences, it's easy to pass a fraud and remain undetected for a long time.

To me, those seem like insane assertions. Here is how I'd try to rebut them:

1. That's simply not true.
What seems to happen much more often is that people who communicate social sciences, that is, journalists, ascribe political motivation to the research. An obvious example of that occurs on the Croatian Toponyms thread, when AGuyFromBalkanee claims I deny the "obvious Serbian etymology" of some place names so that I can claim they rightly belong to Croatia.
Similarly, economics is often communicated by politicians, rather that economists who don't necessarily have political agenda.
Also, the politically corrupted fields in social sciences seem to be disproportionately more widely known. An excellent example of that is macroeconomics. You know, the Austrian School, the Keynesian School, the Chicago School... Those are, for all practical purposes, unfalsifiable and politically motivated pseudosciences. Yet, they are unfortunately much more widely known than actual economics (that is, microeconomics) is.
Besides, we cannot pretend there are no people in natural sciences who are being dishonest about scientific issues because of their political opinions. Patrick Moore, for example, is a PhD ecologist, yet he is completely untrustworthy on environmental issues.

2. I seriously doubt that. When Greenberg, at that time a relatively respected linguist, published his Amerind hypothesis, and cited many entirely spurious forms supposedly existing in some little-known languages as evidence of it, his idea was (partly because of that, partly because of not following the accepted methodology at all) almost universally rejected.
And making up an obviously incoherent hypothesis is going to get you caught even on the TFES forum, as you can see here (It's a thread in which I, just to see what would happen, claimed without evidence that the cuneiform writings had been wrongly transliterated, and that the correct pronunciation of the signs for the syllables containing the Hittite laryngeal sounds actually contain semi-vowels. I was forced to admit I was wrong just a few posts later.).
Appreciate the cliff notes of how you misunderstood the conversation teo
Thank you. I thought it would make it easy to point out exactly what I misunderstood.
you have to keep up this fatalist conspiracy ridden imagining of how scientific theories are developed
I am not sure what you mean. Where did I ever here assert a conspiracy?
using the most blatent hoax in history, Stalin needing to build a cult of personality with such feats as miracle crops
You think that Stalin made that all up by himself? That he was even educated enough to know about Lamarckism? There had to be somebody quite educated about the topic, such as Lysenko, behind it.
as evidence of how mainstream scientific consensus can be more wrong
I've said that multiple times in my last post, and I will say that again: I am not saying most biologists ever believed in Lamarckism or Lysenkoism. I am saying enough of them believed that and they were vocal enough to mislead the policy-makers into thinking it was the mainstream science. You see the difference?
My point is that scientific consensus being reliable is useless for ordinary people and even policy makers, because there is no oracle machine that will tell you what is the current scientific consensus on some issue and (even worse) whether there is one.
than you striking upon a linguistic theory.
Why exactly are you implying my theory is wrong? Sorry, guys, but that's just insulting. I've put thousands of hours of research in it, and published three papers in peer-reviewed journals about it, and you are dismissing it without arguments, without showing even a basic understanding of it, without even trying to understand it. Why don't you simply admit you know nothing about it and that it's possible I've actually discovered something?

Re: Which Sciences Are the Most Useful?

Posted: Thu Jan 17, 2019 2:00 am
by brimstoneSalad
@NonZeroSum Thanks for the recap. Hope you can get teo to stay on track.
teo123 wrote: Thu Jan 17, 2019 12:27 am My point is that scientific consensus being reliable is useless for ordinary people and even policy makers, because there is no oracle machine that will tell you what is the current scientific consensus on some issue and (even worse) whether there is one.
Surveys are done of scientists for just that purpose.

The most recent/popular example being on anthropogenic climate change. Depending on the wording etc. there's some margin for error, but it's clearly an overwhelming majority.

Lacking relevant information about what consensus is, though, of course it would be difficult to use that as a guide.
teo123 wrote: Thu Jan 17, 2019 12:27 amWhy don't you simply admit you know nothing about it and that it's possible I've actually discovered something?
I don't know much about linguistics, it could be a field with low hanging fruit left. Though its utility it pretty limited unless you've come up with a new way to code automatic translation that's more intelligible/natural or less prone to creating misunderstanding.

Not sure what that has to do with anything, though. There are a few areas (typically the softer sciences) where science fair level contributions can discover things too, but they're not that dramatic. Psychology is very ripe for discovery, maybe linguistics is too, particularly for less studied languages.

Re: Which Sciences Are the Most Useful?

Posted: Thu Jan 17, 2019 2:11 am
by NonZeroSum
teo123 wrote: Thu Jan 17, 2019 12:27 am
using the most blatent hoax in history, Stalin needing to build a cult of personality with such feats as miracle crops
You think that Stalin made that all up by himself? That he was even educated enough to know about Lamarckism? There had to be somebody quite educated about the topic, such as Lysenko, behind it.
Nope, didn't say that. Just your incredibly poor imagining of evidence. Darwin's theory of evolution was rejected in favour of other theories for a long time because of christianity's influence. This is the point at which you paint a picture of it's best to throw in the towel, there must be no stronger mainstream theories than newer ones just being worked out where they fit in. It's like when conspiracy theorists learn about an actual phenomenon called the mandela effect, but it just encourages them to believe more that it's proof of alternative realities. A one in a million conspiracy to subvert scientific evidence happens under a dictatorship fresh out of a devestating civil war where the leader has every interest to make himself out to appear to be a god, there must be a conspiracy over-emphasising the validity of all mainstream theories.

Climate change was the latest one to have funding pushed onto scientists to do everything they could to come up with alternatives to man-made climate change, was even 1% of the scientific community swayed by that funded research? Nope. Has journalism and public literacy into these matters improved dramatically since only 10 years ago? Yes.
teo123 wrote: Thu Jan 17, 2019 12:27 am
as evidence of how mainstream scientific consensus can be more wrong
I've said that multiple times in my last post, and I will say that again: I am not saying most biologists ever believed in Lamarckism or Lysenkoism. I am saying enough of them believed that and they were vocal enough to mislead the policy-makers into thinking it was the mainstream science. You see the difference?

My point is that scientific consensus being reliable is useless for ordinary people and even policy makers, because there is no oracle machine that will tell you what is the current scientific consensus on some issue and (even worse) whether there is one.
There are scientists mouths lol, journals, articles, magazines and snazzy videos to condense it down for people, with a vast open source community holding respectable publications to the facts.
teo123 wrote: Thu Jan 17, 2019 12:27 am
than you striking upon a linguistic theory.
Why exactly are you implying my theory is wrong? Sorry, guys, but that's just insulting. I've put thousands of hours of research in it, and published three papers in peer-reviewed journals about it, and you are dismissing it without arguments, without showing even a basic understanding of it, without even trying to understand it. Why don't you simply admit you know nothing about it and that it's possible I've actually discovered something?
I honestly thought you acknowled this to be the most likely case in a previous post. I only have your presentation of your situation to go on, if my remarks aren't useful to you, sorry about that, such is the advice of a stranger.
teo123 wrote: Fri Jul 20, 2018 9:46 am A few days ago, I was on an airplane for the first time in my life. I travelled to Germany. I must comment, there is a lot of "vegan" food in supermarkets that isn't available in Croatia. I will try it, though I've heard it doesn't taste good. It's expensive, but almost everything is more expensive in Germany than in Croatia. In Germany, people have higher wages, but the prices are also higher.
Anyway, since I stopped believing that airplanes didn't exist, I've convinced myself of quite a few other false things. I was thinking that perhaps, when I thought that the Earth was flat and that airplaned didn't exist, I wasn't actually thinking with my own head. I thought that maybe I wasn't actually thinking about those things, but that I was just thoughtlessly repeating what other conspiracy theorists were saying, just in a different context. So, I've spent more than two years familiarizing myself with the mainstream Croatian etymology, and then I've made my alternative interpretation of the Croatian toponyms. When I was satisfied with it, I made a web-page about it and contacted two quite prominent Croatian linguists, Mate Kapovic and Dubravka Ivsic, about it (e-mailing them a link). I've published what Dubravka Ivsic said in response on the web-page. In short, they think my interpretations are almost certainly wrong.
I don't know what to say. If they are right, why it is that I always get everything wrong? Why it is that, more effort I put into researching something, I seem to get further and further from the truth? Other people are so intellectually lazy, they just look up the etymologies in a dictionary, and assume those etymologies are right. Is it possible that I, as someone who puts by orders of magnitude more effort into researching those etymologies, somehow get things right less often than they do? What is that magic that steers me in the wrong direction, no matter what I am researching? Why is it that things that seem obvious to me are usually false? How is it possible that Hrozny, who didn't study Indo-European languages at all, correctly deciphered Hittite, yet I've studied relevant things for over two years, and the things I thought I've discovered about Illyrian are false? How it is that Ventris deciphered Linear B script with basically no knowledge of the things relevant to deciphering it? They were making wild guesses based on questionable premisses. Why is it that their guesses turned out to be right, yet whenever I make a guess, no matter how reasonable it seems, I turn out to be wrong? It's really depressing.

Re: Which Sciences Are the Most Useful?

Posted: Thu Jan 17, 2019 7:00 am
by teo123
Lacking relevant information about what consensus is, though, of course it would be difficult to use that as a guide.
That's true. It's very hard for me to figure out whether the bloggers who claim most social scientists agree anarchism isn't too harmful have a point or not.
Though its utility it pretty limited unless you've come up with a new way to code automatic translation that's more intelligible/natural or less prone to creating misunderstanding.
I haven't studied much NLP, to be honest. What I could do with my knowledge is perhaps make some algorithm to predict the meanings of words from their sounds given their meanings in related languages, but that's also not useful.
I mean, I have also studied quite a lot of computer programming, my greatest achievement is almost certainly making my own simple programming language that compiles into assembly. However, I am not sure if what I've learnt from that can be used in making an efficient machine translation software, the most efficient machine translation algorithms today don't even rely on AST-es, right?
Not sure what that has to do with anything, though.
Well, that would prove I have a basic understanding of how science works, right?
Darwin's theory of evolution was rejected in favour of other theories for a long time because of christianity's influence.
Partly because of Christianity, and certainly partly because the Lamarck's theory would be very useful if it were.
I honestly thought you acknowled this to be the most likely case in a previous post.
Most likely, most of the things that are currently accepted by science are false. For some things, like the shape of the Earth, there is an unanimous consensus among the experts. For most of the things, there isn't, and what most of the scientists believe is bound to change.
What is it that you expect the experts to say if you e-mail them with a relatively long web-page explaining some hypothesis that could be true, but is contrary to the mainstream science? Do you expect them to respond with "Wow, you've solved a mystery! Thanks a lot!"? Or do you expect them to respond with "Your hypotheses seem rather unlikely. Are you sure you know enough about...", as they indeed did?
The arguments I got against my theory were basically two of them:
1. That "issa", which I suggested was an Illyrian word meaning "health-giving spring", is more likely to be a suffix than a part of a compound word, because the same suffix exists in Celtic languages. My response to that is that there is no reason to assume it existed in Illyrian, since it couldn't have possibly existed in Proto-Indo-European (whose phonotactics didn't allow for geminate consonants), unless we assume Illyrian and Celtic were more closely related.
2. That I didn't include a detailed description of the sound changes that happened when a word was borrowed from Illyrian into Croatian. That was a legitimate criticism, so I studied a bit more of the Proto-Slavic phonology and included those descriptions. Mate Kapovic told me that, if I study the Proto-Slavic phonology in detail, I will realize most of my etymologies are unreasonable. I am still not sure what he was referring to. Studying the historical phonology even affirmed my beliefs.

Re: Which Sciences Are the Most Useful?

Posted: Thu Jan 17, 2019 6:45 pm
by brimstoneSalad
You can't compare that kind of historical linguistics to hard sciences like that. Figuring out where a word came from is more like putting together a crime-scene; it's using limited circumstantial evidence (particularly pre-DNA) to make plausible assumptions.

We can't exactly submit your notion that "issa" has certain roots to the same kind of experimentation to prove or disprove it like we can with theories in physics, chemistry, biology, etc.

While I do not doubt that these questions are interesting to many, it's more of an issue of history than a science.

A science would be more like asking which regions of the brain different words stimulate in a certain population and mapping their relatedness based on that. Or looking at usage and doing some kind of statistical analysis of relative syntactic positions in the body of extant work in a language. Questions of testable contemporary fact, not ancient history based on plausible guesses. These are things you may very well be able to revolutionize with your interest in computer science and linguistics. Look at the cross-over where you can evaluate linguistic concepts with hard experiment, and you'll be able to overturn classical linguistic practice.

Understanding the differences between the softer and harder sciences, and what makes them so, is paramount here.

When it comes to softer sciences (and keep in mind there are variations in hardness and softness even within a science depending on the question being asked, it's not always fair to lump an entire field), of course most "theories" are probably false: in so far as they have many plausible alternatives, those alternatives all have about the same chance of being true. Principally because the "theories" are more hypotheses which are rarely falsifiable.
The same is simply not true of hard sciences.

Re: Which Sciences Are the Most Useful?

Posted: Fri Jan 18, 2019 2:10 am
by teo123
We can't exactly submit your notion that "issa" has certain roots to the same kind of experimentation to prove or disprove it like we can with theories in physics, chemistry, biology, etc.
In some natural sciences, like meteorology or astronomy, it's also impossible to experimentally test the theories.
Principally because the "theories" are more hypotheses which are rarely falsifiable.
If I claim that "Issa" and "Iasa" meant "health-giving springs" in Illyrian (as I do), that is an (at least theoretically) falsifiable claim: find a few toponyms, on the territory where Illyrian was spoken, in which that element occurs, and show they refer to places where there couldn't have plausibly been springs believed to be health-giving, and you have falsified my claim.
The same is simply not true of hard sciences.
I am not sure what you mean. If there is some unsolved problem in hard science, of course there are going to be multiple competing hypotheses, all of which are about as likely given the limited data we happen to have. If there are no competing hypotheses, that means either that the situation is very clear (which is rare) or that it is very understudied.

Re: Which Sciences Are the Most Useful?

Posted: Sat Jan 19, 2019 9:51 am
by teo123
Also, let's face it, this "hard science"/"soft science" division can easily be a self-fulfilling prophecy.

Re: Which Sciences Are the Most Useful?

Posted: Sat Jan 19, 2019 1:27 pm
by brimstoneSalad
teo123 wrote: Fri Jan 18, 2019 2:10 am In some natural sciences, like meteorology or astronomy, it's also impossible to experimentally test the theories.
That's a misconception, much like Ken Ham's belief that evolution is "historical science" rather than experimental science.

You should study astronomy more, and the why and how of what we know, before making claims like that.
Even meteorology involves models with falsifiable predictions and very hard science when it comes to the physics of weather phenomena.

The trouble here is that you can't understand any of that without more in depth study. The only way you can know a science is soft is by having enough study in it to evaluate it. Or... statistical analysis like this can help:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3694152/

That was an interesting way to measure it, and they confirmed it's both 1. A real predictive phenomena (and not random) and 2. A spectrum between sciences.
teo123 wrote: Fri Jan 18, 2019 2:10 amIf I claim that "Issa" and "Iasa" meant "health-giving springs" in Illyrian (as I do), that is an (at least theoretically) falsifiable claim: find a few toponyms, on the territory where Illyrian was spoken, in which that element occurs, and show they refer to places where there couldn't have plausibly been springs believed to be health-giving, and you have falsified my claim.
There's no reason to believe those toponyms exist. Even without them it doesn't suggest it's true because they could have simply disappeared or never proliferated in the same way. AND even if you did find something that looked like one it could be something else; you can't definitively show that, or show that they came from the same sources, etc.
There's always plausible deniability there.

What you have is something that is practically unfalsifiable without a time machine since it's something of a wild goose chase and may in fact be completely so, because if those toponyms have gone extinct that's not an option and again false positives are plausible.

You can always get incredibly lucky and find a Rosetta stone, but banking on dumb luck like that for some unlikely artifact to settle a dispute is not how science works. That's not what we call falsifiability.

You could say, in that light, that the divinity of Jesus is falsifiable too because Jesus or God could come down to Earth and confirm or deny it. :roll: Again, relying on improbable outside forces isn't how falsifiability in science works. That's why theology isn't a legitimate branch of science.
teo123 wrote: Fri Jan 18, 2019 2:10 amI am not sure what you mean. If there is some unsolved problem in hard science, of course there are going to be multiple competing hypotheses, all of which are about as likely given the limited data we happen to have. If there are no competing hypotheses, that means either that the situation is very clear (which is rare) or that it is very understudied.
If you're talking fringe stuff like string theory, sort of, but those things are also models and not exactly hypotheses. That's another issue.
There are cases kind of like this, but the exist along the periphery outside of modern ability to falsify them, not in mainstream theories; most accepted theories aren't wrong.