Hacker News Reader

Comment by zyxzevn | Hacker News Reader

While skeptical, he did not have much skepticism against mainstream theories.

I think it needs another item in the list: For any theory/ hypothesis: how well does it stand against the null-hypothesis? For example: How much physical evidence is there really for the string-theory?

And I would upgrade this one: If there’s a chain of physical evidence (was argument), every link in the chain must work (including the premise) — not just most of them

And when breaking these items do not mean that something is false. It means that the arguments and evidence is incomplete. Don't jump to conclusions when you think that the arguments or evidence is invalid (that is how some people even think that the moonlanding was a hoax).

View on HN

> While skeptical, he did not have much skepticism against mainstream theories.

That's tautological. The definition of a "mainstream theory" is one that is widely believed. And while, sure, sometimes scientific paradigms are wrong (c.f. Kuhn), that's rare. Demanding someone be "skeptical" of theories that end up wrong is isomorphic to demanding that they be a preternatural genius in all things able to see through mistakes that all the world's experts cannot. That doesn't work.

(It's 100% not enough just to apply a null hypothesis argument, btw!)

Really that's all of a piece with his argument. It's not a recipe for detecting truth (he didn't have one, and neither do you[1]). It's a recipe for detecting when arguments are unsupported by scientific consensus. That's not the same thing, but it's closer than other stuff like "trust".

(And it's 100% better th an applying a null-hypothesis argument, to be clear.)

[1] Well, we do, but it's called "the scientific method" and it's really, really hard. Not something to deploy in an internet argument.

> And I would upgrade this one: If there’s a chain of physical evidence (was argument), every link in the chain must work (including the premise) — not just most of them

We still use Newtonian physics plenty, despite bits of it not working due to relativity.

>We still use Newtonian physics plenty, despite bits of it not working due to relativity.

Absolutely. And since no one else has trotted this bit out yet, I guess it will be me:

All of our science is based on imperfect models of how the universe works. Every single one is wrong.

However, the models we use today are less wrong than those we used in the past. We know this because we (as you pointed out about Newtonian physics) can more accurately describe the universe than we were able to do previously.

That doesn't mean we've found "the truth." Nor does it mean that we have all the answers.

This is an important concept for baloney detection as those who are peddling baloney will (often, but not always) purport to know "the truth."

Anyone who makes such claims is either knowingly attempting to mislead or is lying to themselves and others.

Which, IMHO, is a pretty big red flag in baloney detection.

> For any theory/ hypothesis: how well does it stand against the null-hypothesis? For example: How much physical evidence is there really for the string-theory?

That's an unfortunate choice of example - the problem with string theory is that there is no null hypothesis. We know that our other theories are not self-consistent when unified, but we don't have a theory that is self-consistent, that could serve as the null hypothesis.

> And I would upgrade this one: If there’s a chain of physical evidence (was argument), every link in the chain must work (including the premise) — not just most of them

From The Demon Haunted World:

"In the middle 1970s an astronomer I admire put together a modest manifesto called “Objections to Astrology” and asked me to endorse it. I struggled with his wording, and in the end found myself unable to sign—not because I thought astrology has any validity whatever, but because I felt (and still feel) that the tone of the statement was authoritarian. It criticized astrology for having origins shrouded in superstition. But this is true as well for religion, chemistry, medicine, and astronomy, to mention only four. The issue is not what faltering and rudimentary knowledge astrology came from, but what is its present validity.

...

The statement stressed that we can think of no mechanism by which astrology could work. This is certainly a relevant point but by itself it’s unconvincing. No mechanism was known for continental drift (now subsumed in plate tectonics) when it was proposed by Alfred Wegener in the first quarter of the twentieth century to explain a range of puzzling data in geology and paleontology. (Ore-bearing veins of rocks and fossils seemed to run continuously from Eastern South America to West Africa; were the two continents once touching and the Atlantic Ocean new to our planet?) The notion was roundly dismissed by all the great geophysicists, who were certain that continents were fixed, not floating on anything, and therefore unable to “drift.” Instead, the key twentieth-century idea in geophysics turns out to be plate tectonics; we now understand that continental plates do indeed float and “drift” (or better, are carried by a kind of conveyor belt driven by the great heat engine of the Earth’s interior), and all those great geophysicists were simply wrong. Objections to pseudoscience on the grounds of unavailable mechanism can be mistaken—although if the contentions violate well-established laws of physics, such objections of course carry great weight."

Reply on news.ycombinator.com