This issue has come up in a million threads now. We can try to keep religion out of it if it makes things simpler, but I bet it will get there.
My view: absence of unexpected evidence is not evidence of absence. Absence of expected evidence is.
For the alien example: since we do not expect that aliens far away from us in general will be doing anything we are likely to sense, absence of this evidence is not evidence of their absence. It is clearly not evidence for them either. At best, it is evidence that they are not beaming radio waves at us.
On the other hand, if your model of an alien is one who is buzzing us in UFOs, abducting random people, and crashing more often than the Sheriff in Dukes of Hazzard, the absence of any evidence of their presences over the past 50 years is evidence of their absence. If they existed, we should have seen a sign.
Example number two. You are an insurance investigator, and one of your policy holders says that a dog he had in the yard for a week during vacation was stolen or escaped. You see food for the dog, and water, and the client claims that a lot of it was eaten. However, investigating the yard yields not one bit of dog poop. What does the absence of that evidence tell you?
Clergymen in the 90s were looking for evidence of Satanic Ritual Abuse in the disappearance of some children. None turned up. One pastor said that the absence of evidence was, in fact, evidence of cult murder. Which made me wonder if just one or two bloody knives would have satisfied them that the Satanists were innocent. Lack of evidence is, simply, lack of evidence.
Absence of evidence is only useful when testing specific theories that can have specific predictions applied.
eg with the dog that you will find other evidence of the dog’s presence, because this is what you will commonly in other similar scenarios.
Its use falls down when you do not have other similar scenarios to effectively extrapolate from. Many AOEINEOA arguments do not offer the same level of predictability of what evidence should or should not be present, so its utility is greately reduced.
In that case the evidence we’d expect to see is not a bunch of bloody knives but a bunch of missing persons reports. If the satanic ritual believers were right, entire towns would have been depopulated of children. Loons can always ignore evidence against or invent bizarre explanations for lack of expected evidence. I’m not talking about proof of absence, only evidence of absence.
That’s true - if you have no idea what evidence you’d expect to find, none of this works.
I set up my dog scenario so that there is motive for faking. Other evidence you might expect and won’t find would be dead patches of lawn or holes dug in the ground. You might interview neighbors, who might or might not hear barking. There can be explanations, like a neighbor who said she entered the yard and cleaned up the poop. But, the absence of the poop is certainly evidence of absence, if not conclusive.
Up to a point, sure. It is reasonable evidence, evidence a reasonable man may consider when trying to establish what is most likely true when he knows he cannot prove absitively and posolutely true/false.
An experiment in philosophy was conducted in the dorm rooms of many of our finest universities over a period of about 70 years. It was ultimately concluded that a thing (God, just to pick something at random…) might be proven to exist, but cannot be proven not to, an absence has no qualities to measure.
Could be Heisenberg, but I’m not sure. Anyway, time to go poison the cat.
What determines how strong that evidence is, though, is how likely the evidence will be absent if the phenomenon is true compared to how likely the evidence will be absent if the phenomenon is false.
Consider: Is there an elephant in the room you’re sitting in? To answer the question, you look around, and see no elephant. That is absence of evidence for an elephant in the room. From that, you conclude that there is no elephant in your room. Because, if there were an elephant in the room, you’d be nearly guaranteed to see it. If there weren’t an elephant in the room, you’d be guaranteed (assuming no delusions or whatever) that you wouldn’t see it. So that fact that you didn’t see it indicates that it really isn’t there.
Now, is there a mouse in the room? Now, just because you don’t see it, that isn’t as strong evidence, because if there were a mouse in the room, it would be much more likely that you would miss it. It could be hidden under something, or blending in with what’s behind it. The difference is smaller, so the evidence is less strong.
And when you get to the point where the likelihood of the evidence being absent is exactly the same whether the phenomenon is true or false then it is not an indicator at all.
Aside from what has been mentioned, it’s especially strong evidence of absence when the claimed entity or phenomenon would violate accepted physical laws if it existed; which is the kind of claim that typically brings up the claim that “absence of evidence isn’t evidence of absence” in the first place.
Voyager, I agree with the distinction you draw in the OP. So, absence of unexpected evidence tells us next-to-nothing. Where I find the absence of evidence most useful is in dealing with anecdotal theories, e.g., astrology, the paranormal and homeopathic medicine. Proponents of such will insist that they’ve seen the theories work, but if one asks for evidence it’s usually thin to nonexistent. Without reasonable evidence, I conclude the theories don’t work. Which is to say, I don’t bother to add them to my toolbox for understanding the world. I understand that someone may eventually come along with reasonable evidence, at which point I would alter my conclusion.
The absence of a phenomenon is not an absence of evidence. It is a negative phenomenon. No snow is not evidence of summer. But, no snow is evidence of something.
Also, multiple iterations of non compelling evidence is not proof, but proof is a pretty narrow state to seek. Most science is never actually proven, in a mathematical sense. It is convincing, and predictions based upon it are reliable, and repeatable. However, as so many are eager to point out, evolution is a theory. Theories have characteristics that differentiate them from mere opinions, or assertions, though.
I cannot prove the sun will come up, but if there is a pool at the office on whether or not it will, I am definitely betting on the it will side of the wager. Why? The absence of evidence that it ever failed to do so is compelling. I am willing to take bets that I will be visited by aliens on any specified date. You pick the date, and we can wait together. Bring your money.
I find the absence of evidence of alien visitations convincing enough to ignore the threat of alien invasion entirely. But I know it does not prove the absence of hostile alien civilizations just biding their time, waiting for everyone else to give up waiting for them. It’s a chance I am willing to take. So little of the universe is provable that insisting on proof is pretty much a guarantee that you can never make a decision at all, about anything.
That is why “beyond a reasonable doubt” is such an important evaluation. It’s a shame about the definition of reasonable, though.
I think both this and the related (asinine) soundbite ‘you can’t prove a negative’ are somewhat distorted echoes of the impossibility to prove universal statements from (any finite number of) singular (existential) ones. So, both are true when applied to a truly universal statement – that there is no evidence for the existence of black swans does not imply that there aren’t, in fact, any, and similarly, the proposition ‘there are no black swans’ can’t be proven. Both mantras are often erroneously held to be generally applicable, even in cases where one is dealing exclusively with singular statements – which are easily enough proven even if they come in negative formulation, and for which absence of evidence can very well be (compelling, even!) evidence of absence.
Voyager, if by “Is absence of evidence evidence of absence?” you mean “Is absence of evidence of ‘x’…evidence that ‘x’ does not exist”…then absolutely “NO”…absence of evidence is just evidence of “absence of evidence.”
Fact is, logically, nothing can be concluded about “x” from a lack of evidence of “x”…except that there is no evidence for “x.”
There is no evidence for leprechauns, for instance…or for unicorns; Flying Spaghetti Monsters; or CPA’s working on one of the moons of Saturn. But the only thing that can be obtained from that lack of evidence is that there is no evidence for any of those things. No LOGICAL case can be made that those things do not exist because of the lack of evidence for them.
A lack of evidence of “x” is in no way evidence that “x” does not exist…and I cannot imagine any working logician ever suggesting otherwise.
Lack of evidence, I will will freely acknowledge is very useful when combating an assertion that something exists for which there is no evidence. It certainly calls into question that assertion…but it is NOT evidence that it does not exist.
The “expected” and “unexpected” qualifiers does not change things in any way in that regard…and appear to me to be gratuitous in furtherance of some other agenda you may have for this question. (Can’t imagine what that could be!)
Arguing “there are no elephants in this room” is not arguing to the question of "is absence of evidence evidence of absence.”
There is plenty of evidence there is no elephant in the room I am in. To suggest that evidence there is no elephant in this room is “an absence of evidence” just doesn’t work.
I hope that can be seen without further explanation.
Evidence that something is not there is NOT an absence of evidence!
The only way this argument works is by circularity…which in logic, is a no no.
You have to say: There is no evidence that anything can “violate accepted physical laws” (and then illogically say that not only is that evidence that it cannot happen [exist], but proof that it cannot happen [exist])…therefore it can be used as part of the argument that absence of evidence is evidence of absence in the case of entities that appear to violate accepted physical laws.
Logicians don’t answer those kinds of questions. No question about existence of (ETA: possible) physical objects can be answered by logic alone. The only way to get an answer for that kind of question is by looking at the evidence.
Now, if you want dead certainty, you’ll get none either way. That’s logic, and also means you’ll have to accept less than 100% certainty on any question about physical reality.
In my opinion you’re confusing evidence with proof. Absence of reasonable evidence IS relevant - otherwise it wouldn’t make sense for you to say “[it] is very useful when combating an assertion that something exists for which there is no evidence”. It’s just not an unassailable proof.
There are all kinds of hypothetical scenarios where it’s practically impossible to get any evidence, and there are scenarios where the kind of evidence you’d expect is debatable, but for everyday assertions it works very well.
For example - I don’t know if someone owns a cat. Statistically, let’s say there’s a 50% chance they’ve got one. I can stick with that 50% or I can go to their house, look in the kitchen etc for cat food, check the house for litter boxes & bowls, look for scratch marks, check for a cat flap etc, and the cat itself. If I find none of those, it’s reasonable to assign a < 50% probability that they don’t own a cat.
I didn’t say they were trying to answer those kinds of question. I merely mentioned that I cannot imagine any logician suggesting that a lack of evidence that “x” exists is evidence that ‘x’ does not exist.
I am not asking for dead certainty. I am merely saying that a lack of evidence that “x” exists is NOT evidence that ‘x’ does not exist.
SP…if you want to think that lack of evidence of “x” IS evidence that “x” does not exist…please do so. This is a free country. I merely disagree with you on that.
There are all sorts of things that “work” for everyday assertions that should not be used in the kinds of debates in which we are engaged here. That does not impact on what I have said.
A lack of evidence for “x”…is NOT evidence that “x” does not exist.
I’ll use the example of Strinka, which I believe illustrates very well the way we assess claims in real life.
In entering a regular-sized room, you found a lack of evidence of the existence of an elephant in it. Question: Is this evidence that an elephant is absent in the room? I think so (and I would imagine, so would most of you). Now what if it’s a mouse we’re talking about, would it change your answer. I don’t think so.
First of all, one can change one’s assessment given new evidence - that has never been forbidden. If, in entering a room, you find no evidence of the existence of a mouse, then that is evidence that mouse do not exist in that room. Now, if somebody points at a live mouse hidden in a corner, that’s fine. You can always change your assessment.
Since people can still change their assessment in the presence of new evidence, I do not agree that there’s a difference in the examples you’ve given. For so long as the absence of “unexpected” evidence is not vitiated by the presence of “expected” evidence, then for me, absence of evidence (unexpected of not) would still be evidence of absence.
)