Extraordinary claims require extraordinary evidence, the discussion.

[Dinwar]

And what would a supposed alien artifact look like? What material would it be made of?

If it was obviously artificial (ie, had buttons or a display screen or something) and was made of a material we cannot manufacture on Earth at present (say, clear aluminum or polymerized iron or something), I'd at least be willing to accept the posibility of it being an alien artifact.

Ordinary claim: I show you a video I recorded on a nature hike that is grainy, blurry, and ambiguous but claim the subject is of a bear. Would you demand more evidence than that or accept it?

Extraordinary claim: I show you a video I recorded on a nature hike that is grainy, blurry, and ambiguous but claim the subject is of a bigfoot. Would you demand more evidence than that or accept it?

You've mislabeled these examples, but let's let that slide.

It depends entirely on the video, and the conditions involved. If the video is too blury, I'm going to dismiss either interpretation. Neither is valid scientific evidence, however, so as a researcher I would be forced to dismiss both videos and both interpretations. Check out the controversy over the ivory-billed woodpecker a few years ago.

Claiming that you found a new species of snail, even if it is visually similar to an already known species of snail, isn't an extraordinary claim and if someone claimed that they found one I would accept it on their word alone. No photographs or body needed.

You quite obviously are not discussing science. I would have thought that my referencing the International Code of Zoological Nomenclature would have demonstrated that I was. If you wish to discuss nonscientific implications of Sagan's statement, a loose and fuzzy interpretation of "extraordinary" is perfectly fine; vernacular is always loose and fuzzy.

Also, thanks for the chuckle. Claiming to find new species of snail can EASILY be an extraordinary claim in science (using the loose and fuzzy definitions you are implying [odd how no one can define the word, not even the advocates]; it's perfectly ordinary by strict definitions). Ever hear of a range extension? Anyone would say that finding a T. rex alive today would be extraordinary (er, the dinosaur one, not the leach), yet I've seen range extensions for mollusks and other invertebrates that are FAR greater than 65 million years. And believe me, NO ONE just "takes them at their word". There are specific protocols that must be followed.

Someone claims they find evidence of an actual chubacabra, on the other hand, and not even photographs will do it. There would need to be a corpse or a living specimen for me to regard it as being true and I imagine it would be the same for you as well.

There's a reason I specifically mentioned a type specimen. Without a type specimen no proposed new species, regardless of how mundain or extraordinary, can be accepted. I italicized that part because it's rather foundational to the science of taxonomy. You're right--I would demand a body. You're wrong in implying that I would limit that demand to only chubacabras and other fantastical creatures. I'm not allowed to not make that demand; it's literally in the rule book. My copy of said rule book has a green cover with gold lettering, and is writen in both English and French; modern copies tend to be online.

 

[barehl]

Some people view Sagan as a prophet of Science, who's statements must be poured over and analyzed to find hidden nuances.

That description might be partially true for someone like Jules Verne. I don't see how it would apply to Sagan.

The reality is, he was a scientist speaking to a non-science audience; he was making a pithy statement that people would remember, but was not discussing the acutal mechanisms of scientific investigation or debate. His statement works at that level--someone who doesn't know what science is beyond what they learned in high school--but it falls apart when you attempt to apply it as a scientific standard.

Right, which is why the examples I gave were outside of any scientific specialty. I believe his statements applied more to the fringes than the center. In real science, extraordinary claims without evidence don't tend to make it through peer review.

 

[Dinwar]

That description might be partially true for someone like Jules Verne. I don't see how it would apply to Sagan.

I would think that too--but given how much processing power people devote to trying to figure out what Sagan meant, you and I are apparently in the minority.

I believe his statements applied more to the fringes than the center.

That has nothing to do with whether you're looking at this from a scientific perspective or not. I specifically illustrated how fringe ideas can be approached from a scientific perspective, and where they fail (it isn't because their evidence is too ordinary!). Science doesn't HAVE a center; for the purposes of my statement it's an approach to learning about the universe. What I was saying was that Sagan's statement may apply to a very different approach, but not to the scientific one. It's actually a pretty damning statement coming from me, since most woo attempts to overturn science and therefore the scientific approach is the appropriate one.

In real science, extraordinary claims without evidence don't tend to make it through peer review.

Peer review isn't the be-all, end-all of science. Claims without evidence are dismissed as irrelevant by researchers. Claims that don't explain the known evidence (a different set for each researcher, but it generally overlaps with numerous other researchers; there's a shared body of knowledge in each field, which we all expand upon as we see fit) are also dismissed as irrelevant. This all happens way before any peer review process kicks in. Such claims typically don't get beyond the "having a drink at the bar" stage.

 

[GnaGnaMan]

Sorry, I am a new member and I cannot post URLs

Can you guys help critique it?

It's the typical empty propaganda that comes from these people. It is an attempt to convince by carefully selected quotes and misrepresented quotes.

Chris French has also stated he believes that ESP has been proven to scientific standards. However, he does not accept that those results should be accepted by science until the results have been replicated by skeptical scientists. These results have been replicated by parapsychologists. What French is saying is that replications are not valid unless the researcher has a certain philosophical beliefs. That is unheard of in any other branch of science.

Demanding independent replication is absolutely normal in science. It is revealing that the author claims that belief in ESP is a philosophical belief. How can it be a philosophical belief when at the same time it is supposed to follow from the data?

Skeptic Ray Hyman in his 1995 article Evaluation of Program on Anomalous Mental Phenomena admits that the results of remote viewing and Ganzfeld experiments have demonstrated statistically significant effects. However, he implies there could be unknown methodological flaws in the experiments. This demonstrates a double standard because in no other field of science would a positive experimental result be criticized because there might be sources of errors that no one can think of. It is impossible for any scientist to defend his work against that type of criticism.

This is a very common lie among psi believers.
Well established theories or facts are rarely found wrong. Otherwise they would not be well established. Errors of one kind or another are common, however. It is a simple fact that results which seem to contradict a well established theory are more often wrong than the theory.
This is well known in science. It is exactly the standard of any area of real science.

Here for a recent physics example:
http://en.wikipedia.org/wiki/Faster-than-light_neutrino_anomaly#Reception_by_the_physics_community

Recently, journalist Steven Volk was surprised to discover that leading skeptical psychologist Richard Wiseman has admitted that the evidence for telepathy is so good that "by the standards of any other area of science, [telepathy] is proven."

This is a plain lie. The original quote comes from the Daily Mail, a british newspaper of ill repute.
The original quote is this:
“I agree that by the standards of any other area of science that remote viewing is proven, but begs the question: do we need higher standards of evidence when we study the paranormal? I think we do.

“If I said that there is a red car outside my house, you would probably believe me.

“But if I said that a UFO had just landed, you’d probably want a lot more evidence.

“Because remote viewing is such an outlandish claim that will revolutionise the world, we need overwhelming evidence before we draw any conclusions. Right now we don’t have that evidence.”
http://www.dailymail.co.uk/news/article-510762/Could-proof-theory-ALL-psychic.html

Wiseman has since said that he was misquoted, as one would expect from the Daily Mail. He has clarified so:
“It is a slight misquote, because I was using the term in the more general sense of ESP — that is, I was not talking about remote viewing per se, but rather Ganzfeld, etc as well. I think that they do meet the usual standards for a normal claim, but are not convincing enough for an extraordinary claim.”
That's still nonsense but I'm content with exposing the misrepresentation of the article for now.

(Info lifted from here.)

If, then, parapsychology and modern science are incompatible, why not reject parapsychology? ...The choice is between believing in something "truly revolutionary" and "radically contradictory to contemporary thought" and believing in the occurrence of fraud and self-delusion. Which is more reasonable?

So, here we have two skeptics in effect admitting that if this were any other field of inquiry then the experimental data would have carried the day by 1950.

The issue of Rhine is very revealing. Rhine worked in the first half of the 20th century. Like some people before and after him, he has provided results that cannot possibly be a fluke or due to ordinary error.
However, these results do have all the signs of "fraud and self-delusion".

The simple fact is that Rhine's work lead nowhere. People tried to build on his work but the only result were some high-profile fraud cases.
In science, promising results are extended. New and/or better theories are developed. Applications are found. None of that has ever happened in parapsychology.
What happens in parapsychology is the MO of religions. You are asked to accept Psi like you are asked to accept Jesus. Accepting Jesus does not simply mean accepting the historical evidence for a historical Jesus as being sufficient. It means joining a group.

But neither Alcock, Hebb, Wiseman nor French ever bother to explain how the claims of parapsychology "stand in defiance" of science, or how "physics and physiology say that ESP is not a fact." Indeed, it is rare for a skeptic to ever back up this claim with specific examples. As I show in my new book Science and Psychic Phenomena, on those rare occasions they do, they invariably invoke the principles of classical physics, which have been known to be fundamentally incorrect for more than three quarters of a century.

He has a point here. There is no proper, accepted definition of the term ESP. You can't say that any feature of ESP is contradicted by fact because no one will tell you what features ESP should have.
Well, there is one feature that believers universally ascribe to ESP: It must be impossible according to current scientific theory and fact.

Psi is certainly incompatible with the old scientific worldview, based on Newtonian mechanics and behaviorist psychology. It is not incompatible with the emerging scientific worldview based upon quantum mechanics, the neurosciences, and cognitive psychology.

The whole passage about physics has simply no basis in fact. I'm just quoting this as representative.
We should remember that Newton was an alchemist and a creationist. There is nothing in Newtonian physics that is incompatible with almost any kind of magic.

Modern physics, if anything, is a problem for magic. It puts still tighter limits on what should be accepted as possible.
Faster than light speed, for example: No problem for Newton but completely impossible in Einstein's relativity.

It is traditional for Psi believers to claim the currently fashionable physics as somehow supporting their faith: Mesmer's animal magnetism, Sinclair's mental radio and now quantum physics.
Despite all the lies in this article, your average physicist would be quite surprised to learn that quantum physics predicts psychokinesis. Reputable physics journals make no mention of such claims.
Note that Walker published his Psi article over 30 years ago. Nothing came of it, not in physics, not in parapsychology. Note also that the named physicists are all very old and in some cases dead (such as Walker). You wouldn't expect up and coming ideas to be promoted by dead and dying people.

 

[GnaGnaMan]

And I completely disagree with Carl Sagan; I feel it's a badly-worded, and introduces qualitative emotional judgments to scientific thinking. I feel that's what all Sagan-inspired skeptics are doing when they parrot the phrase, as well.

I fear that one could make the same charge against any one-line summary of the principle.
Let me give it a try:
Claims that are unlikely to be true must be supported by evidence that is unlikely to mislead before they are accepted.

It definitely doesn't roll of the tongue like Sagan's quip but it is less open to misinterpretation. Is it better putting it like that?

Ideally one would explain a bit of Bayesian statistics and a lot about scientific methods and the reliability of each under varying conditions. That would take a couple of hours.

"Extraordinary" is far too emotionally-loaded a term and it doesn't belong in a scientific discussion, frankly.

How is it emotionally loaded? Maybe you mean to say imprecise. Either way, the term is used in science. It is precise enough and can be clarified when needed.

 

[Dinwar]

How is it emotionally loaded?

How is it not? You can see it in the deifinitions implied in this thread--extraordinary means something spectacular, and spectacle always includes an emotional component.

Either way, the word is pretty much undefinable, and therefore should not be used.

Either way, the term is used in science.

Not in the same way. "Extraordinary" generally means well beyond the norm--an extraordinarily large brachiopod would be defined as one that's several standard deviations past average, for example. There is no sense in which a type specimen can be considered extraordinary evidence--it's literally a requirement. For other types of woo, the concept of "several standard deviations past average" is inapplicable. I'll grant that the scientific sense of the word is applicable to some woo (for example, ESP and precognizence can be studied statistically), but for others it's simply not. The only thing extraordinary about a fairy type specimen is the emotional response to seeing it; the actual evidence is exactly what we expect for every other species identification, and therefore cannot be justifiably called extraordinary, even if the claim is considered to be so.

Ideally one would explain a bit of Bayesian statistics and a lot about scientific methods and the reliability of each under varying conditions. That would take a couple of hours.

We can actually skip the statistics; they're unnecessary for most people, including most scientists (taxonomy, for example, doesn't use such methods--they can be re-written into whatever format you desire, but taxonomists generally do not write Bayesian justifications for their conclusions). It's trivial to teach people to read scientific papers without teaching them statistical reasoning; after all, that's how much of scientific training was done prior to computers (we couldn't do many of hte statistical analyses we take for granted now--humans don't live long enough to do all the calculations necessary for many of them).

As for the rest, yeah, it'll take some time. If someone is unwilling to put that time in, that's their problem. Science isn't the provence of the super-geniuses, but it does have a cost of entry, and understanding how to properly interpret claims is a huge part of that.

 

[Mudcat]

If it was obviously artificial (ie, had buttons or a display screen or something) and was made of a material we cannot manufacture on Earth at present (say, clear aluminum or polymerized iron or something), I'd at least be willing to accept the posibility of it being an alien artifact.

Can we forget the "unearthly materials" trope her for second? What if the only alien artifact left behind from a legitimate alien contact was something remarkably similar to a PDA and was made from material slightly more advanced than available here on earth but not entirely unknown to our sciences?

You've mislabeled these examples, but let's let that slide.

No, I didn't. If I produced a video in which you could only make out that the subject was something large, moving, and dark than the more mundane answer would be that the subject is a bear, the more extraordinary answer would be that the subject is a Bigfoot, and the right answer might be that the subject would be another hiker.

Just because it is the simplest answer doesn't make it the only answer or even the right one.

 

[GnaGnaMan]

I did want to add to the statement about the nature of evidence: a scientific claim must explain the existing evidence as well as new evidence. This is where woo claims fail--they may explain new evidence ("aliens visit Earth" is a coherent explanation for certain phenomena, in that it's internally consistent and does not contradict the data in question), but they fail to explain the much greater body of previously demonstrated data ("aliens visit Earth" ignores physics, psychology, and normal human behavior--it even ignores people saying "I did that, it was a hoax"!).

That sounds a bit off. The issue you are getting at here is parsimony or Occam's razor if you prefer.
We know that some alien or UFO sightings are explained by human error, equipment malfunction, fraud or hoaxing, and delusion. The principle of parsimony says that we should not assume another cause for any of the cases where the true explanation cannot be found.
Any additional reason why people report aliens need not have anything to say on other, already existing reasons.

The bit about explaining previous data is really only relevant if you are dealing with a theory that supersedes an old one or claims to be more fundamental.

 

[Dinwar]

Can we forget the "unearthly materials" trope her for second?

It's not a trope--it's the answer to your question. If you wish to dismiss the answer, don't bother asking the question.

What if the only alien artifact left behind from a legitimate alien contact was something remarkably similar to a PDA and was made from material slightly more advanced than available here on earth but not entirely unknown to our sciences?

Then we may never know that it was alien in origin. "We can't know" is a perfectly acceptable answer in science--you have no idea how annoying and frustrating it is to have to conclude that, but intellectual honesty demands that we acknowledge the limits of our knowledge. Rationality demands that our conclusions not extend beyond what the data can support.

There may be other ways to identify it, however; materials science does not exhaust all research posibilities. If it used a power supply never before seen, or if it's configured in a way that makes no sense for humans (say, instead of a touch screen it has a hole that responds to an inserted tenticle changing shape), or something like that it would certainly be suggestive if not proof.

No, I didn't.

Yes, you did. You labeled them as evidence, then discussed claims and evidence combined. That's a mislabeling. The video is evidence; the interpretations are claims. It's a bit of a pedantic point, but it's vital that you differentiate betwen these in science (though it's one of the more difficult aspects of research).

Just because it is the simplest answer doesn't make it the only answer or even the right one.

Yeah, no kidding. Ocam's Razor is only a first approximation, and only really good at determining which avenues of research will be most profitable. I've never argued against this point, and fail to understand why you brought it up as if it counters any statement I've made.

 

[Dinwar]

That sounds a bit off. The issue you are getting at here is parsimony or Occam's razor if you prefer.

Parsimony is related to the idea I was driving at, but the implications of the word (particularly when you attach Occam's Razor to it) are entirely different. I'm saying nothing about the simplicity of the explanation; I'm discussing the completeness of it. An idea that explains some outlier, but fails to explain the overwhelming majority of the data, is quite obviously wrong. Most, or at least a great deal of, woo falls into that category.

My example may have been a bit misleading, but later examples clarified it I thought.

The bit about explaining previous data is really only relevant if you are dealing with a theory that supersedes an old one or claims to be more fundamental.

Not at all. No new theory or hypothesis can contradict established data. Established INTERPRETATIONS, sure--after all, humans are not omniscient and interpretations may be flawed. But it cannot contradict established data. The data in science are always right.

 

[JoeMorgue]

I think the issue often is that an "extraordinary" claim is really multiple claims in one.

For instance, an ordinary claim. I tell you that paint on my front door is wet.

We all already agree that things called doors exist, that they are often placed on th front of houses, that a substance called paint exists, and that at certain points in the application process of paint it is wet. So "Wet Paint Front Door" we are all already on board with all that. The only real questions are as to whether or not I have a front door and whether or not it has wet paint on it. And since I stand to gain nothing by lying about this, the statement "Joe's front door has wet paint on it" does't requires to modify any basic understand of how the world works, and such a statement would be very easy to verify, we would classify such a claim as "ordinary" and require a very low burden of proof for it.

Now if I were to tell you that I was abducted by aliens... well that requires us all to agree that aliens exist, that they travel to earth, and that they abduct people. We would have to agree on all those things, many of which would require a fundamental shift in how we view the world, before "Was Joe abducted by aliens?" really becomes the question.

And I think that is the core difference. All statements really at their core require the same amount of evidence, but the question really is how much of the question are we already agreeing on when we go into the discussion.

 

[GnaGnaMan]

How is it not? You can see it in the deifinitions implied in this thread--extraordinary means something spectacular, and spectacle always includes an emotional component.

I can see how most examples are spectacular. However, that does not imply that this is implied in the definition.
An exception would be Bigfoot. The only reason you might think of Bigfoot as spectacular is because it would be extremely surprising if, after all these years, it turned out to be real. However, a priori, Bigfoot is not any more extraordinary than a gorilla.

Either way, the word is pretty much undefinable, and therefore should not be used.

: very unusual : very different from what is normal or ordinary
: extremely good or impressive
-Merriam-Webster

There is no sense in which a type specimen can be considered extraordinary evidence--it's literally a requirement.

It is extraordinary because it is virtually conclusive evidence. Something that a single piece of evidence rarely is.

Indeed, a single piece of evidence will rarely be sufficient to sway the mainstream of science.

We can actually skip the statistics; they're unnecessary for most people, including most scientists (taxonomy, for example, doesn't use such methods--they can be re-written into whatever format you desire, but taxonomists generally do not write Bayesian justifications for their conclusions).

You misunderstand. Bayesian inference is the mathematical formulation of how one should assign a probability to a hypothesis being true.
I'm not talking about the statistics used in papers (which are rarely Bayesian).

 

[Dinwar]

-Merriam-Webster

Check out how dictionaries define words, then get back to me and tell me that their definitions are definitive. I'll be you can't do it with a straight face.

When used in these types of discussions, the word is plastic to the point of being meaningless. That's the important point.

It is extraordinary because it is virtually conclusive evidence.

So you would call a type specimen of a new species of sea slug extraordinary evidence? Last weekend I was at a mollusk conference where three new species were described. No one batted an eye.

I find it baffling that someone can call a requirement "extraordinary". Such a claim demonstrates my assertion that the word is plastic to thepoint of meaningless.

You misunderstand. Bayesian inference is the mathematical formulation of how one should assign a probability to a hypothesis being true.

No, I understand you perfectly. I simply disagree with you. Teaching that mathematical formula is a waste of time. Not even the researchers use it, so it's quite obviously not critical to understanding the research. No one I know has ever run a Bayesian analysis of a hypothesis in any paper before accepting or rejecting it. Considering the fact that this includes a rather remarkable number of researchers, I'm confident in saying that this proves Bayesian analysis is not necessary for doing so. It may be useful to some people, but if so they can learn it on their own. When teaching a general audience, it's a waste of time.

 

[GnaGnaMan]

Parsimony is related to the idea I was driving at, but the implications of the word (particularly when you attach Occam's Razor to it) are entirely different. I'm saying nothing about the simplicity of the explanation; I'm discussing the completeness of it. An idea that explains some outlier, but fails to explain the overwhelming majority of the data, is quite obviously wrong.

I don't get what you are driving at.
If a theory is either complete or wrong then there is only one true theory: The theory of everything. Which we don't have.

When you say "fails to explain", do you mean "makes a false prediction"?

My example may have been a bit misleading, but later examples clarified it I thought.

Not to me.

Check out how dictionaries define words, then get back to me and tell me that their definitions are definitive. I'll be you can't do it with a straight face.

I happen to agree with the dictionary definition. You have a different one? Maybe "eliciting an emotional response"?

When used in these types of discussions, the word is plastic to the point of being meaningless. That's the important point.

You can play games with any word. That doesn't mean that discussion is useless.

So you would call a type specimen of a new species of sea slug extraordinary evidence? Last weekend I was at a mollusk conference where three new species were described. No one batted an eye.

Why would they? To you extraordinary implies something emotional. Well, not to me. To me it simply means uncommon, which conclusive evidence certainly is.

No, I understand you perfectly. I simply disagree with you. Teaching that mathematical formula is a waste of time. Not even the researchers use it, so it's quite obviously not critical to understanding the research. No one I know has ever run a Bayesian analysis of a hypothesis in any paper before accepting or rejecting it. Considering the fact that this includes a rather remarkable number of researchers, I'm confident in saying that this proves Bayesian analysis is not necessary for doing so. It may be useful to some people, but if so they can learn it on their own.

I never said that understanding Bayesian inference is necessary for understanding research. In fact, I pointed out that Bayesian statistics is rarely used.

It's not even necessary to understand why "claims that are unlikely to be true must be supported by evidence that is unlikely to mislead before they are accepted", as I rephrased Sagan's quip. Most people grasp the concept intuitively as evidenced in this thread but it nevertheless provides a mathematically sound foundation. Also, there are some non-intuitive implications but that's unimportant.

When teaching a general audience, it's a waste of time.

Indeed. I would go with Sagan for that purpose.

 

[Dinwar]

When you say "fails to explain", do you mean "makes a false prediction"?

No, I mean it doesn't factor the issue into the equation at all. Darwinism sensu stricto did not factor in genetics, for example--Darwinism had no mechanism, and while it predicted the existence of such a mechanism the actual study of genetics is entirely outside of Darwinism sensu stricto.

I happen to agree with the dictionary definition. You have a different one?

Fair enough--but it's obvious that you're not applying it consistently.

You can play games with any word. That doesn't mean that discussion is useless.

True, but some words are inherently slippery and to be avoided.

Why would they? To you extraordinary implies something emotional. Well, not to me. To me it simply means uncommon, which conclusive evidence certainly is.

I can conclude from this statement that you do not have much of a background in biology (your emphasis on Bayesian analysis also suggests that--you find that far more frequently in physicists than in biologists). Type specimens are extremely common. Almost every species has one (all should, but some exceptions were made for species which had their type specimens destroyed during WWII). To call them extraordinary is to misrepresent the nature of these specimens.

There is no rational definition of "extraordinary" that includes requirements of the system. The obligatory is not extraordinary. Type specimens are required for defining a new species; therefore, they cannot be considered extraordinary. It may seem so TO YOU, but that's an emotional reaction, not a rational analysis.

I say this to point out how even with your dictionary definition, the term remains slipper. The two of us can define the same thing--the same bloody specimen!--as either extraordinary or ordinary. So it's not a valid standard. There is too much subjectivity involved.

I never said that understanding Bayesian inference is necessary for understanding research.

You said, as a replacement for pithy sentences:

Ideally one would explain a bit of Bayesian statistics and a lot about scientific methods and the reliability of each under varying conditions.

The fact that you specifically called out Bayesian analysis, you implied that you considered it necessary for interpreting evidence. You put it in par with the scientific method! I have been arguing, since that point, that the call for teaching Bayesian analysis is unwarranted.

Indeed. I would go with Sagan for that purpose.

I would not. Sagan's quote misrepresents the scientific process and can easily lead to people thinking that extraordinary evidence (a slippery term that is rarely if ever defined) is the only requirement for scientific acceptence of an extraordinary claim. It further perpetuates the quasi-religious view of science many people, including some here, seem to hold.

 

[GnaGnaMan]

No, I mean it doesn't factor the issue into the equation at all. Darwinism sensu stricto did not factor in genetics, for example--Darwinism had no mechanism, and while it predicted the existence of such a mechanism the actual study of genetics is entirely outside of Darwinism sensu stricto.

Good. That's how I initially understood it.
But it doesn't tell me what you were driving at. Could you explain that again?

Fair enough--but it's obvious that you're not applying it consistently.

How so?

I can conclude from this statement that you do not have much of a background in biology (your emphasis on Bayesian analysis also suggests that--you find that far more frequently in physicists than in biologists).

You cannot conclude that from the statement but you are right nonetheless.

Type specimens are extremely common. Almost every species has one (all should, but some exceptions were made for species which had their type specimens destroyed during WWII). To call them extraordinary is to misrepresent the nature of these specimens.

It is my understanding that classifying species, taxonomy, is only a small sub-field of biology.
I don't think it is valid to generalize from this specific field to biology as a whole.

There is no rational definition of "extraordinary" that includes requirements of the system.

I don't understand this sentence. Please rephrase.

The obligatory is not extraordinary. Type specimens are required for defining a new species; therefore, they cannot be considered extraordinary. It may seem so TO YOU, but that's an emotional reaction, not a rational analysis.

If you argue that the evidence required in a scientific discipline for acceptance of something as fact is, by definition, not extraordinary then I don't think that anything ever requires extraordinary evidence.
I doubt that Sagan meant to say that some claims require more than just scientifically convincing evidence.

I say this to point out how even with your dictionary definition, the term remains slipper. The two of us can define the same thing--the same bloody specimen!--as either extraordinary or ordinary. So it's not a valid standard. There is too much subjectivity involved.

There's no way around that, I'm afraid. Eventually scientific fact is decided by human judgment.

You said, as a replacement for pithy sentences: The fact that you specifically called out Bayesian analysis, you implied that you considered it necessary for interpreting evidence.

I consider it necessary for properly understanding how the probability of a claim being true is related both to the evidence and the a prior likelihood of the claim.

You put it in par with the scientific method!

I don't see where I did that. What do you mean by "scientific method" anyway?

I have been arguing, since that point, that the call for teaching Bayesian analysis is unwarranted.

I did not call for teaching Bayesian inference. I was not quite sure what you were arguing, so thank you for that clarification. I don't know why you would argue that but whatever.

I would not. Sagan's quote misrepresents the scientific process and can easily lead to people thinking that extraordinary evidence (a slippery term that is rarely if ever defined) is the only requirement for scientific acceptence of an extraordinary claim.

I agree that the Sagan quote can be and is misunderstood. Perhaps that should be taken as a reason not to use it but I am pessimistic if clearer formulations are really superior. I certainly don't know any that is as neat.
Properly understood (see my previous take), it encapsulates an important principle that is always obeyed.
What other requirement do you see besides solid evidence?

 

[Dinwar]

You cannot conclude that from the statement but you are right nonetheless.

Oh yes I can. I have a fair bit of experience with advocates of Bayesian analysis, and while my conclusions are necessarily not well-supported (ie, I haven't done rigorous and formal studies on the matter), certain trends have emerged. Whenever someone brings up Bayesian analysis it's a good bet they've got a physics background; I've rarely heard a biologist bring it up, and when they have they've been much more restrained about the utility and necessity of such analyses. There may be exceptions, but I haven't found any.

It is my understanding that classifying species, taxonomy, is only a small sub-field of biology

Your understanding is wrong. It's foundational to biology, and spans multiple fields.

I don't think it is valid to generalize from this specific field to biology as a whole.

True. Which is why I didn't do it. I used the example to illustrate the flaw in the "extraordinary claims require extraordinary evidence" nonsense. In that case there are literal rules for how to do it--and again, the obligatory cannot rationally be termed the extraordinary.

I don't understand this sentence. Please rephrase.

If something is required it cannot rationally be termed extraordinary. The requirement renders it ordinary, as every single example of that type of thing will have that feature. Every example of a paper discussing a new species has a type specimen--bar none (though the specimen may not exist anymore). Thus, to say presenting a type specimen is extraordinary evidence is nonsensical. You may as well argue taht a paper going through peer review is extraordinary.

Let's look at it a different way: All items in Set A have Trait Y. All of them. It is part of the requirements to being included in Set A. You are attempting to convince me that Trait Y is, in some cases, an extraordinary claim. My counterargument is basically pointing to all the rest of the cases of Trait Y which you don't consider extraordinary, and asking "Why?" There would be nothing unique about presenting a type specimen of a fairy; it would be a perfectly normal biology monograph. It would also be conclusive proof of fairies.

If you argue that the evidence required in a scientific discipline for acceptance of something as fact is, by definition, not extraordinary then I don't think that anything ever requires extraordinary evidence.

Taxonomy is rather a special case, in that there are formal requirements (the International Code of Zoological Nomenclature). The requirements are much less formal for most other fields.

That said, apparently you're coming over to my side.

There's no way around that, I'm afraid. Eventually scientific fact is decided by human judgment.

True--however, we usually attempt to remove such subjectivity as is inherent in the term "extraordinary". When two researchers can look at one datum and draw three conclusions, there's a problem.

I consider it necessary for properly understanding how the probability of a claim being true is related both to the evidence and the a prior likelihood of the claim.

I consider your claim to be contrary to the data. Hundreds of years worth of scientists analyzed the probability of hundreds if not thousands of claims being true just fine, without applying Bayesian analysis.

I don't see where I did that. What do you mean by "scientific method" anyway?

I quoted you saying it. You mentioned two things in that quote: Bayesian analysis and the scientific method. Thus, you put them on par with one another, at least in that context. If you can't see that at this point, I do not think anything I say will convince you that your statement can be so interpreted, and I will not continue trying.

Properly understood (see my previous take), it encapsulates an important principle that is always obeyed.

I don't agree. I think it drastically and detrimentally misrepresents how scientists evaluate data. I further believe that much, if not most, of woo doesn't fail by failing to provide extraordinary evidence, but rather because it cannot explain or contradicts the known data.

Good. That's how I initially understood it.
But it doesn't tell me what you were driving at. Could you explain that again?

Any new theory must explain all the data explained by the old theory, as well as any new data, if we are to accept it. If it cannot explain the data we already have, we cannot accept the new theory. A theory that contradicts data must be discarded. I'm not sure how much more clear I can make the statement. If you could tell me which parts are problematic for you, it would help.

 

[Checkmite]

Now let's say the claim is that there is extraterrestrial life visiting earth, mutilating cattle, and abducting people and performing experiments on them. You can say that it's a subjective opinion, but most people would agree that that is an extraordinarily extraordinary claim. And the level of evidence required to support the claim would be magnitudes greater than the claim that the cattle mutilations were performed by humans or that the alien abductions were some kind of fever dreams if they weren't outright made up for attention.

No; I'm afraid I disagree. It would not take "magnitudes" of evidence. One single piece of evidence - say, an alien literally caught in the act, could prove that aliens are visiting Earth, abducting people, and so forth and such like. But the only thing that would make this evidence "extraordinary" is the fact that we already think of the claim it proves as "extraordinary"; a human getting caught in the act would be hardly remarkable at all, even though it is literally the exact same evidence in every way save for one single variable.

The idiom should be rephrased to something like "evidence proving an extraordinary claim tends to seem extraordinary by extension"; which would admittedly lend itself a bit less to sloganeering, but would be a more accurate rendition of the sentiment while removing the appearance of a certain burden of proof "above the norm" that doesn't actually exist.

 

[GnaGnaMan]

Oh yes I can. I have a fair bit of experience with advocates of Bayesian analysis, and while my conclusions are necessarily not well-supported (ie, I haven't done rigorous and formal studies on the matter), certain trends have emerged.

The statement in question was not concerned with Bayesian analysis.
Also I am not an advocate of Bayesian analysis.

If something is required it cannot rationally be termed extraordinary. The requirement renders it ordinary, as every single example of that type of thing will have that feature.

So you are saying that the statement is logically inconsistent.

The problem I have with that is that it is an obvious misinterpretation. Now if you were to argue that it should not be used because it lends itself to misinterpretations then you obviously have a point there.

That said, apparently you're coming over to my side.

So you are saying that there is no such thing as extraordinary evidence?

True--however, we usually attempt to remove such subjectivity as is inherent in the term "extraordinary". When two researchers can look at one datum and draw three conclusions, there's a problem.

Well, Bayesian inference is one such attempt and you seem to think it a waste of time...

I consider your claim to be contrary to the data. Hundreds of years worth of scientists analyzed the probability of hundreds if not thousands of claims being true just fine, without applying Bayesian analysis.

So your retract your claim that taxonomy is foundational to biology?
I mean, we got along fine without it for hundreds of years...

I quoted you saying it. You mentioned two things in that quote: Bayesian analysis and the scientific method.

Nowhere did I mention the scientific method, other than to ask you about it. I'm not even sure what that is.

Any new theory must explain all the data explained by the old theory, as well as any new data, if we are to accept it. If it cannot explain the data we already have, we cannot accept the new theory. [...] I'm not sure how much more clear I can make the statement. If you could tell me which parts are problematic for you, it would help.

You clarified that not explaining data is not the same as contradicting data. You still conflate the two which I find confusing, so I took out the sentence regarding contradicting data. I hope I am not mangling your meaning.

I do not understand why a new theory should explain anything but what it does actually explain.
To reuse your example: If a UFO is sighted, and tracked to a spot on the ground. And if we then find an alien spaceship (or a meteor crater) there then we can conclude that this explains that specific UFO sighting and perhaps other similar ones that remain unexplained. We cannot discard the alien spaceship (or the meteor) as explaining that specific UFO sighting just because it does not explain any number of other UFO sightings.

 

[GnaGnaMan]

The idiom should be rephrased to something like "evidence proving an extraordinary claim tends to seem extraordinary by extension"; which would admittedly lend itself a bit less to sloganeering, but would be a more accurate rendition of the sentiment while removing the appearance of a certain burden of proof "above the norm" that doesn't actually exist.

The problem is that this sentence does not mean what Sagan meant.
It also is not what people want to convey when they use the Sagan quote.
When someone claims to have been abducted by aliens, why would anyone want to tell them that evidence for that would be regarded as extraordinary when the whole subject is whether such evidence exists?