Absence of Evidence is Evidence of Absence

[The Atheist]

My dice example, or coin example, or book example, et al. pertains to things that potentially exist now and would potentially leave evidence for itself existing.

You're getting there. That is another admission that the hypothesis only works with concrete examples where there is already knowledge. It is therefore an almost-useless axiom.

You show that again here:

I guess it's my fault for not clarifying this in the OP, but I thought these assumptions were rather implicit. So, for full disclosure, here are the relevant details:

1. Something potentially exists now, or existed/happened in the past.
2. Something that would/could leave evidence for itself

Any talk of "future" events/things is not relevant, nor are things that we know leave no evidence for their existence,...

Nice moving of the goalposts. If you'd said that at the start, I probably would have agreed with you.

[MorpheusSandman]

You're getting there. That is another admission that the hypothesis only works with concrete examples where there is already knowledge.

As opposed to abstract examples where there is no knowledge? Such as?

It is therefore an almost-useless axiom.

Well, there's two different axioms being discussed: one insists AOE IS EOA, and the other insists that AOE IS NOT EOA. The former is a useful axiom in that it aligns itself with Bayes' Theorem and can teach people how to think in terms of expectations given their beliefs or hypotheses; the latter is useless in that the only things it pertains to are things we don't fuss about to begin with.

Elaborating further:

Nice moving of the goalposts. If you'd said that at the start, I probably would have agreed with you.

It's not INTENTIONALLY moving the goalposts. If you want to blame me for not stating those things explicitly, then fine, I'm guilty as charged; but I figured that those elements were already implicit in the phrase since it would patently obvious that for things we KNOW leave no evidence the absence of evidence wouldn't matter one way or another. Yet, I've never heard anyone use the phrase as it pertains to things for which we KNOW leave no evidence, like what Joe Blow ate 30 years ago. Rather, the phrase is always used pertaining to things for which we would EXPECT there to be evidence for (or, at least, potential evidence) if they existed; then, after finding no evidence, instead of people admitting that such absence of evidence is evidence of the thing's absence, they insist that it's not.

[YesNo]

(1) Incorrect. What you're talking about gets into the nature of "what is evidence?" rather than if evidence exists for non-existing things. It's true that certain information can LOOK like evidence for non-existing things, but because these things are non-existent, it's our interpretation of the information that's at fault, because it's not really evidence for the thing we think it is. That's why it's important to have a good grasp/understanding of what evidence is before one starts talking about whether such-and-such piece of evidence is evidence at all.

As I recall, the evidence for the existence of WMDs was satellite images. That seems like valid evidence for the existence of WMDs, however, when confronted with on the ground inspections that satellite evidence should have been dismissed.

In general, I don't think it makes sense to use something like "absence of evidence is (or is not) evidence of absence" too strictly. If you do, you should reject many worlds since there is no evidence for it. All this phrase is is a reminder that we could get things wrong if we don't have adequate evidence.

I do think it is useful to use Bayesian analysis where appropriate, but I don't think it is very useful if we can't determine those probabilities with much confidence.

[MorpheusSandman]

As I recall, the evidence for the existence of WMDs was satellite images.

It's been too long and my memory is too fuzzy to get into too deep of a debate, but, IIRC, most all experts at the time said that the satellite photos were not even decent evidence that there were WMDs in Iraq because there were simply far too many other possible explanations behind what the images were. This is in contrast with, eg, the data that was built-up/developed about Bin Ladin's whereabouts when Seal Team 6 went in; even then, the experts could only say it was maybe a 66/33 chance he was there, which was far higher than the likelihood of those images being evidence of WMDs.

If you do, you should reject many worlds since there is no evidence for it.

You mean there's no evidence for Copenhagen, since it's the interp making extra assumptions present nowhere in the math. MW is just taking the math at face value, so the math IS the evidence for MW.

I do think it is useful to use Bayesian analysis where appropriate, but I don't think it is very useful if we can't determine those probabilities with much confidence.

Really, the essence of Bayes is how to properly reason. We have beliefs, we experience new evidence; how should this new evidence effect our beliefs? Bayes gives us a formula for how evidence affects beliefs. Yet we all believe things for which we can't determine the probabilities of our confidence level, and we all experience new evidence with which we can't influence our confidence levels; so, what is one to do? You might as well apply what you've said to reasoning, in general, because if we can't determine such probabilities with confidence to our beliefs or evidence, we can't really reason at all except in extremely nebulous ways.

I think Bayes can be used practically. As I've repeated many times, I use it in my profession of poker. I don't have EXACT probabilities of how often players make plays with certain hands. Yet, I can always remind myself that the question "given they just did this, what is the probability they have that?" is best answered with two others: "given they have that, what is the probability they would've done this; and given that they don't have that, what is the probability they would've done this." Online, most sites give you 30 seconds for each decision, so you can't do high level math in your head; you basically just have to learn how to intuitively ask these questions and come to rough answers based on very rough data. Yet... it works. Despite the inevitable inaccuracies due to my limited human brain, Bayes can work even without being able to determine probabilities with a high level of confidence. It's much better than nothing at all.

[Ecurb]

There’s really no difference between “someone ate something” and “something existed in the past”. It seems that one is a statement about an activity, and the other is a statement about the existence of something, but someone ate something could (obviously) be translated into a statement about existence (“Joe Blow’s stomach contained partially digested food at 12:00 noon on May 12, 1211.”)

In any event, I think we are in basic agreement that an absence of the evidence that one would suspect could be found if something existed constitutes evidence of its absence. Medical studies, for example, involve falsifying the “null hypothesis” (the notion that a drug has the same effect as a placebo). The null hypothesis is equivalent to the notion that something (Sasquatches, for example) does not exist. It can be falsified by certain kinds of evidence, but not by others (because of the problems previously outlined).

The further question of what constitutes “evidence” is a touchy one. If we find Joe Blow’s fingerprints on the murder weapon, that has evidentiary value – although it hardly proves that Joe is the murderer. There could be any number of reasons his fingerprints are on the blunt object.

The same is true for “evidence” about Sasquatch (or about God, for that matter). We need not claim that the eye witness reports, the footprints and the photographs have no evidentiary value to find them unpersuasive. To return to the notion of probability – it is, in my opinion, highly unlikely that any giant, hairy, non-human hominids roam the Pacific Northwest. However, given the evidence that they do, it’s MORE likely that Sasquatches roam Oregon than that giant bipedal crocodiles wearing top-hats and spats roam Oregon. That’s because we DO have evidence (albeit unpersuasive) for one, but not for the other. Similarly, it’s more likely that Lazarus rose from the dead than that Joe Blow did (although I find both highly, highly unlikely).

[Emil Miller]

... it’s MORE likely that Sasquatches roam Oregon than that giant bipedal crocodiles wearing top-hats and spats roam Oregon. That’s because we DO have evidence (albeit unpersuasive) for one, but not for the other. Similarly, it’s more likely that Lazarus rose from the dead than that Joe Blow did (although I find both highly, highly unlikely).

The American crocodile is a declining species that is only found on mainland USA in small numbers in the south of Florida. While it might be reasonable to assume that they are not bipedal or wear top-hats and spats, absence of evidence isn't evidence of absence.

[Nick Capozzoli]

This "Absence of Evidence is Evidence of Absence" thread makes me think of a joke I heard many years ago. It goes like this:

"An archeologist, who happened to be an avid amateur radio hobbyist, had spent years exploring archeological sites in Peru. He noted that after years of excavations in the Andes, he had never found any sort of metallic wire, such as the type used in wired telegraphy. He concluded that this proved that the ancient Andean societies he was studying must have had "wireless" communications.":

This was actually a two-part joke, and I omitted the first half. For completeness, here's the whole joke:

An archeologist published a report of his many-years excavation of ancient Egyptian burial sites, in which he found artifacts made of copper wire. He concluded that this meant that the ancient Egyptians understood electricity and used it to "communicate" over wires, as in telegraphy or telephonic communication.

Another archeologist, who had conducted "digs" in ancient Andean sites noted that he had never found any sort of metallic wire, such as the type used in wired telegraphy. He concluded that this proved that the ancient Andean societies he was studying must have had "wireless" communications."

[YesNo]

You mean there's no evidence for Copenhagen, since it's the interp making extra assumptions present nowhere in the math. MW is just taking the math at face value, so the math IS the evidence for MW.

Which is more likely the existence of Sasquatch or the existence of Many Worlds?

[AuntShecky]

The further question of what constitutes “evidence” is a touchy one. The same is true for “evidence” about Sasquatch (or about God, for that matter).

I'll say! Let me give you some examples from the animal kingdom. We live in an area that is partly suburban, partly wildly rural. Couple of years ago my bitter half and yours fooly spotted a mountain lion in a field adjacent to a supermarket parking lot. When we got home, we checked the Internet, only to be informed that no mountain lions exist in the area of the U.S. where we live. But the feline we saw could have posed for the photos on the Web; that's how strong the resemblance was.

Similarly, this morning about 4:30 am, when I looked out the window onto the brightly-lit front lawn I saw a creature I'd never seen before. It was extremely fat, between 2 and 3 feet long, and covered from snout to tail with gray fur, with a few black and white markings. It didn't look at all like a skunk or a ferret or anything other than a badger. I went online to look up "badgers" and the photos depicting that mammal looked exactly like the creature I'd seen early this morning, but various websites insist that badgers are found in the Mid- and far-west of the U.S., and not in my state.

On both occasions, I didn't have a camera. Both times, I was completely sober.

Well, I know what I saw. As far as I'm concerned, that's "evidence" that both mountain lions and badgers can be found this far east. To quote the Marx Brothers --"Who are you gonna believe --me or your lying eyes?" The state wildlife department wouldn't even let me email them!

In the end, people are only going to believe what they want to believe. (But I still know what I saw.)

[Hawkman]

I recommend always carrying a camera and bottle of booze

[MorpheusSandman]

Which is more likely the existence of Sasquatch or the existence of Many Worlds?

They aren't really comparable; the idea that there are these giant harry ape-like things roaming the forests of certain parts of the US that nobody has been able to capture either alive or dead is just preposterous; on the other hand, like I say, MW is what you get when you take the math that models QM as real and add nothing else. MW doesn't "exist" in our world except in the QM mathematical models, so it's hardly similar to a possibly living being.

[Ecurb]

ICouple of years ago my bitter half and yours fooly spotted a mountain lion in a field adjacent to a supermarket parking lot.... .)

Where do you live? Cougar sightings are becoming more common east of the Mississippi. Cougars have always lived in Florida (The Florida Panther). I've heard about well-confirmed cougar sightings in Illinois, Georgia, and some other states (I can't remember which, but some cougars have been captured or killed in these states). However, there are no sustainable populations of cougars in the East; the cougars that have been killed or captured are all young males, driven out of their birth-lands and roaming after food. Cougars are famous for traveling as much as 30-50 miles in a day, prowling for food. Increased deer populations in urban areas have led to more cougars in towns.

Here in Eugene, cougars killed two goats and several chickens within the last two weeks, right in town. Two cougars (a mother and her year-old cub) were trapped and killed (the local newspaper reported that the cougars had been "euthanized", but it seems to me they were simply killed). In any event, if a lone, young male cougar is prowling about in Eastern states that does not mean cougar populations have returned to the area -- breeding pairs have NOT been found east of the Mississippi. (How the cougars cross the Mississippi is another question, for which I do not have the answer.)

[YesNo]

They aren't really comparable; the idea that there are these giant harry ape-like things roaming the forests of certain parts of the US that nobody has been able to capture either alive or dead is just preposterous; on the other hand, like I say, MW is what you get when you take the math that models QM as real and add nothing else. MW doesn't "exist" in our world except in the QM mathematical models, so it's hardly similar to a possibly living being.

All I was comparing was their likelihood, not Sasquatch and Many Worlds. What do you say P(S), the probability of Sasquatch is? What is P(MW), the probability of Many Worlds. Those numbers are comparable. My view is the following: P(S) > 0; P(MW) = 0. Unlike Sasquatch there really is no evidence for MW.

As we've argued in the past, my position is when "you take the math that models QM as real and add nothing else" you no longer get an interpretation of QM. What you get is an interpretation of a deterministic fantasy land that might have made some minor sense in the 19th century, but is obsolete today. So the "math" is not evidence that I would accept any more than you would accept the sightings of Sasquatch.

[Gilliatt Gurgle]

....I saw a creature I'd never seen before. It was extremely fat, between 2 and 3 feet long, and covered from snout to tail with gray fur, with a few black and white markings...I went online to look up "badgers" and the photos depicting that mammal looked exactly like the creature I'd seen early this morning, but various websites insist that badgers are found in the Mid- and far-west of the U.S., and not in my state.

On both occasions, I didn't have a camera. Both times, I was completely sober...

Aunty, that sounds very much like the notorious Badgecabra the result of cross breeding gone bad between a Badger and Chupacabra. http://en.wikipedia.org/wiki/Chupacabra

Around these parts, its the wood ape aka Big Foot that I keep an eye out for.
Absence of evidence=I just haven't run across it yet. Evidence of absence= he just ain't passed through my creek yet
but when he does I'll be ready for him with my Sasquatch Field Guide signed by Dr. Jeff Meldrum-something I picked up at the 2013 Texas Bigfoot Convention.

[Calidore]

Where do you live? Cougar sightings are becoming more common east of the Mississippi. Cougars have always lived in Florida (The Florida Panther). I've heard about well-confirmed cougar sightings in Illinois, Georgia, and some other states (I can't remember which, but some cougars have been captured or killed in these states).

Just a few years ago, police killed a cougar here on the North Side of Chicago in the Roscoe Village neighborhood. That's an upper-middle class yuppie neighborhood well inside the city and nowhere near any forest preserves or anything. The location the cougar was killed was about a mile from Wrigley Field, less than half a mile from Lane Tech High School, and just a block away from a grade school.

Sightings are reported occasionally in the northern suburbs, but having one appear in the city itself, especially a heavily populated neighborhood soon after evening rush, was pretty startling.