from Hacker News

Ask HN: How to prove that astrology works?

by zipotm on 3/16/22, 9:28 PM with 55 comments

• by politelemon on 3/16/22, 9:46 PM

  Through a double blind method. Open source or publish your calculations so that anyone can derive the same personality/event results for a given set of criteria for a person. That is, without you being the intermediary, the result of the calculation performed by different people should arrive at the same result.

  The result will also need to avoid the use of Barnum statements https://en.wikipedia.org/wiki/Barnum_effect as they are subject to interpretation.

• by duped on 3/16/22, 9:59 PM

  It's actually pretty well documented that birthdates affect outcomes in society, for example in athletics (1). It's called the relative age effect (2).

  You would have to control for this if you wanted to convince anyone that it was due to the positions of stellar objects.

  (1) https://sportsmedicine-open.springeropen.com/articles/10.118...

  (2) https://en.m.wikipedia.org/wiki/Relative_age_effect

• by krapp on 3/17/22, 1:20 AM

  Come up with a scientific hypothesis to test your claims. How does the position of the stars affect human personality and events? What is the physical process involved? Is this effect transmitted by light? By gravity, By quantum mechanics? Your hypothesis will need to stand up to the scrutiny of known modern science. How do you account for the speed of light being a hard limit on the transmission of information, and stars being many light years away? How do you account for relativity, precession, and other effects when determining the relevance of position? Is Earth the only relevant vantage point? If so, why, and what does this imply about relativity's thesis that the universe has no preferred reference frames? If your hypothesis involves constellations, you need to explain how stars which just happen to appear in groups from Earth, in the illusory 2D plane of the sky, correlate effects across their arbitrary positions in spacetime.

  Since you claim to have enough knowledge to calculate the events created by astrology, an accurate list of future events would be a good place to start, as well as a formal proof of your method for such calculation, presented for peer review.

  You will also need to explain how every experiment attempting to validate astrology has found that it does not actually work, and that there is no known scientific basis by which it could even possibly work, beyond psychological effects and fraud, and why all of those experiments were wrong, and no replicable proof of astrology was possible until you and your theories came along, despite people wanting and trying to prove it true for millennia. The principles of astrology were formed at a time when human societies had a wholly supernatural worldview, before people even knew what stars were. That's going to be the difficult part, proving that ancient wizards, alchemists and soothsayers had a more valid and accurate cosmological model than modern astronomers and physicists.

  Once your results have been published, peer reviewed, experimentally replicated and validated through rigorous scientific consensus, then you can submit an article about it to Hacker News. Of course by then, having essentially overturned the fundamentals of known physics and proven the existence of the supernatural, everyone will probably already know about it.

• by cato_the_elder on 3/16/22, 9:50 PM

  I, like most people around here, don't believe in astrology. But I think I am also more sympathetic with people with strange beliefs (or I guess a lot of ordinary girls in this case).

  I think you'd be interested in knowing that Kary Mullis, the inventor of PCR (and a Nobel laureate), also believed in astrology [1].

  [1]: http://www.crawfordperspectives.com/documents/IAMACAPRICORN_...

• by KineticLensman on 3/16/22, 10:07 PM

  Use your knowledge to make some precise predictions and state them before the event. Possibly put them in the equivalent of a sealed envelope so the predictions themselves don't influence human reactions. The prediction should be unambiguous and falsifiable.

  Also, you must mean stars and planets, right? The positions of the stars take multiple human lifetimes to visibly change.

• by nullandvoid on 3/16/22, 9:44 PM

  Use the scientific method, and people here will listen. That means reproducible evidence from a reputable source.

• by curtisblaine on 3/17/22, 12:59 AM

  Write an horoscope for every sign every morning and publish them here every night, without saying which sign correspond to which horoscope. Ask users to take note of the most faithful description of their day. After a few month, publish the sign correspondence for every horoscope you've published and let users calculate the times they matched their sign to the "correct" horoscope. If they all match considerably more than 1/12, you might be onto something.

  (this ignoring the challenge of defining "night" and "morning" for an international audience)

• by ryandvm on 3/17/22, 2:34 PM

  I hate that this is flagged. HN's penchant for self-stifling interesting discussions is probably its worst feature.

  Does astrology work? Almost certainly not. (Though I will suggest that the position of ONE PARTICULAR STAR could have some measurable effects.)

  However, a discussion wherein a community of smart folks use logic and the scientific method to lead someone from a pasture of mysticism to one of rational thinking could be a very interesting discussion indeed. Especially since we seem to be in the middle of a resurgence of unscientific thinking...

• by MerelyMortal on 3/16/22, 9:48 PM

  Do astrological teachings share how the information was originally devined? Do books/gurus/etc track provenance of information?

  (Disclaimer: I do not, and will never, believe in astrology.)

• by gregjor on 3/17/22, 12:55 AM

  If astrology works for you then use that edge. Use the stars to predict lottery numbers (events) or how important people might influence events (personality). If you want to persuade other people it works show them how you got rich from your knowledge of astrology. If you can reliably predict events or the behavior of other people based on the stars you have a big advantage.

  Writing books and giving talks to profit from the gullibility of people doesn’t count.

• by CyberRabbi on 3/16/22, 10:04 PM

  1) Make a survey that classifies respondents into one of the 12 astrological signs

  2) give it to a lot of people, the more the better

  3) also collect their birthdays

  4) do a regression analysis to determine how well the sign based on birthday predicts the sign based on the survey response

  5) profit

  Good luck

• by oceanghost on 3/16/22, 9:53 PM

  If astrology "works" then it makes testable predictions.

  Everything else is religion.

• by andrewfromx on 3/16/22, 9:43 PM

  Watts on this subject: https://youtu.be/83Mw-tF1AdI?t=94

• by aoeusnth1 on 3/17/22, 6:03 AM

  First question I have is, what have you seen that has convinced you that astrology calculations work?

• by PaulHoule on 3/16/22, 9:46 PM

  If sun signs (I am a Taurus) were valid wouldn’t personality tests correlate with when you were born?

• by Smaug123 on 3/16/22, 9:44 PM

  Prove it to individuals before proving it to an audience; you'll need the practice!

• by abetusk on 3/16/22, 10:30 PM

  A lot of hate for this on HN. My challenge to the community, assume this was a good faith argument, how would you convince someone to structure a test so they were convinced astrology wasn't real? If you can't structure at least a rough outline of experiments then you don't really understand why astrology can't be real. If you defer to experts without understanding why they hold this belief then you're ceding critical thinking to someone else. If presented with evidence that astrology was real based on verifiable, repeatable experiments, then you can't claim to be logical.

  Now, assuming this was a good faith question, the answer is that you need to provide a series of repeatable and testable experiments that can be independently verified by others using the same protocol [0].

  There are many confounding factors that will skew results and is one of the reasons why "blinding" experiments were invented, because experimenters that had an expectation of an outcome from an experiment could pollute results from interacting with subjects [1]. As another commenter mentioned, these confounding factors can be extremely subtle [2] so extreme care has to be taken when doing these experiments. The subtlety can even extend to just repeating the experiment over a lifetime until you get a "significant" result that can be published [3].

  If you were to do this in earnest, I would suggest starting with a deep literature review. There are many people doing research into ESP and other supernatural beliefs. They all haven't come up with anything accepted by the scientific community at large. It's good to understand what kind of experiments they set up so that you can see what you might be able to do and how you could replicate or extend those studies. My apologies, I don't have anything focusing on astrology specifically but a good starting point is James Randi [4] [5], the Committee for Skeptical Inquiry and the Skeptical Inquirer [6] [7].

  Once you're in a position to actually conduct an experiment, my suggestion would be to start with something that is as simple as possible, with an aim to reduce any confounding factors and be independently repeatable and verifiable by others. There is a long history of experimentation in this area that has provided no evidence for supernatural effects, so a claim of supernatural effect is considered extraordinary and will need extraordinary evidence to be accepted by the community at large.

  You also have to make sure to guard against people intentionally trying to sabotage your experiment as this area is full of people who are adept at lying convincingly.

  There will always be people who hold irrational belief but if you provide a real experiment that is repeatable and testable independently that can show the effect you believe is real, this will convince people of rational minds that it's real.

  I would also urge you to adopt a scientific mindset in that if you find no evidence supporting your hypothesis or with evidence that explains why you believed your hypothesis to be true without the need for supernatural effect, then you should abandon your hypothesis. Put another way, if you're asking an audience to believe you if you have evidence, then it's hypocritical to not change your views based on evidence as well.

  [0] https://en.wikipedia.org/wiki/Scientific_method

  [1] https://en.wikipedia.org/wiki/Blinded_experiment

  [2] https://news.ycombinator.com/item?id=30704665

  [3] https://slate.com/health-and-science/2017/06/daryl-bem-prove...

  [4] https://en.wikipedia.org/wiki/James_Randi

  [5] https://web.randi.org/

  [6] https://en.wikipedia.org/wiki/Committee_for_Skeptical_Inquir...

  [7] https://skepticalinquirer.org/

• by ykevinator2 on 3/17/22, 12:13 AM

  You'd be better of proving that Jesus was conceived without sperm.

• by la64710 on 3/16/22, 10:05 PM

  Make predictions. Simple

• by bjourne on 3/16/22, 9:53 PM

  Read my comment history and then tell me my date of birth!

• by squarefoot on 3/16/22, 10:13 PM

  > but how to prove it to an audience like HN?

  Sorry, you can't.

• by somehnguy on 3/16/22, 10:03 PM

  That’s an impossible task - because it isn’t.

• by rozab on 3/16/22, 10:12 PM

  There's already a wealth of parapsychology research out there which "proves" the existence of all sorts of supernatural effects with standard scientific techniques.

  https://slatestarcodex.com/2014/04/28/the-control-group-is-o...

  That is to say, there's probably no amount of proof you could show me which would convince me.

• by uberman on 3/16/22, 9:38 PM

  Perhaps you could also prove how many angels can dance on the head of a pin while you are at it.

• by matt321 on 3/16/22, 9:29 PM

  This has got to be a joke

• by epgui on 3/16/22, 10:14 PM

  As a scientist, the comments entertaining this question make me sad.

  Astrology doesn't "work". If it did, then it would be at least in part a science.

  Astrology is not a science, but a pseudoscience.

• by moltke on 3/16/22, 9:56 PM

  I have a personal theory that Astrology did actually work to a small degree in ancient times but it worked by coordinating people in a kind of mematic way that was replaced by the active coordination we do now via the internet etc. It didn't have any real supernatural power.

• by LinuxBender on 3/16/22, 9:56 PM

  In theory if we were in a simulation I might expect there to be instructions and cheat codes. Maybe even Easter Eggs. Astrology could be instructions. Sorcery could be cheat codes. Now you just need to map them out in a way anyone can use them, thus enabling scientific method assuming the simulation does not circumvent a validation process if and when it recognizes it is being tested or monitored. i.e. the VW test.

  There is one theory according to Douglas Adams that if one figures this out the universe will reset and recreate itself in a much more complex manor. There are other theories that this has already occurred.