As you know I think the philosophical burden of proof is a notion harmful to clear thinking. My first post about it was here. And here is a blog from someone who thinks very much the same way. Here I will try to score some further points by undercutting common arguments used to support the view that having a burden of proof is helpful.
But first some background, what is the conjunction fallacy? Consider this example from Amos Tversky and Daniel Kahneman copied from Wikipedia:
“Linda is 31 years old, single, outspoken, and very bright. She majored in philosophy. As a student, she was deeply concerned with issues of discrimination and social justice, and also participated in anti-nuclear demonstrations.
Which is more probable?
1) Linda is a bank teller.
2) Linda is a bank teller and is active in the feminist movement.”
If you thought 2 was more probable, then you committed this fallacy. It is a formal fallacy. And it basically says that the probability of A+B will always lower or equal than the probability of A. As we add conditions the probability can only go down never up.
Why is this tricky to some people? (apparently most people chose option 2) Given the description of Linda many people might think she sounds like a feminist. Moreover, she does not sound like a bank teller. So the first option seems to only have what people think is unlikely. Even though second option has what seems unlikely it also has what seem likely too. So the second option may seem preferable. But mathematically the second option will never be more likely than the first, because every time the second option holds true the first does as well.
When we add details to a claim the probability tends to drop. The probability that I am in front of my computer now, is greater than the probability that I am now in front of my computer wearing a shirt, which is greater then the probability that I am now in front of my computer wearing a brown shirt. This is just how logic works. I think many of us know this and build this into our understanding of the world. So what does this have to do with the burden of proof?
If you go to youtube and type in burden of proof the first video to come up will be this one:
Listen to the first 20 seconds.
What is happening? Well if you notice the person is not just asking whether “something” exists “in space” that we don’t have evidence of and whether we should believe that. Which I tend to think is true – that is I do think there is something in space that we do not currently have evidence of. So how would that help his case for “the burden of proof”? Of course it wouldn’t.
So he doesn’t just ask whether a general thing exists. Instead he gets absurdly detailed with his description. It is not something we don’t know about somewhere in space but under the surface of Pluto! And it is not just “something” its not even just some form of life, but a walrus. And of course even that is not enough. It is a “tiny” “were” walrus. But not only that it is a “psychic” tiny “wer”walrus. But no tonly that! It is a tiny psychic werewalrus that “sends them psychic messages.” But not only that, the messages come “every midnight.” And not only that but it does this while juggling skulls…..
Ok. So by now if you understand the conjunction fallacy then you will know that amount of detail alone will drastically drive down the probability. There is no need to appeal any “burden of proof” to not believe in this creature. The probability is naturally driven down just by understanding the conjunction fallacy.
It’s amazing that this video received 13,000 likes when it is really just an obvious play on this formal fallacy. Now if you are like me you will see this quite often from atheists. The other day when I said there is no philosophical burden of proof I was asked whether I believe “that there is an imperceptible penguin named “Percival” standing right across the room from me, wearing his imperceptible tuxedo and his imperceptible monocle and his imperceptible derby hat, holding his imperceptible pocket-watch.”
Now I want to point out that the proponents of a burden of proof who use these examples are not themselves committing the conjunction fallacy. But rather they just seem oblivious to its logic. It is the logic that makes the conjunction fallacy a formal fallacy that makes the probability of their examples quite low. So when you see these examples the response is not that they are committing a fallacy. Rather the response is more like yes the conjunction fallacy is indeed a fallacy so the probability of your extremely detailed scenario is low.
If you express doubts about “the philosophical burden of proof” you can expect this sort of thing. They won’t ask whether you can believe there is something somewhere we do not have evidence for. No, they will not even ask whether there is some species of life we have no evidence of yet. Rather they will ask about a particular horse that we have no evidence of. But not just a horse, but a horned horse, but not just a horned horse, but a horned horse with exactly one straight long horn coming out of its head. And often even that won’t be enough the horned horse will have to be pink.
And you might respond explaining this fallacy and ask them whether they believe there is something somewhere he/she does not yet have evidence of. I certainly believe there are things that exist that I have no evidence of. For example there very well may be a particular cricket in the southwest United States that I have no evidence of.
Which brings me to another point here. There is more than just the conjunction fallacy at work here. There are also background beliefs. There are probably crickets in the southwest United States. But when and how did a walrus get to Pluto? (forget the whole juggling psychic were bit)
Perhaps the classic example of this sort of argument is Russel’s Teapot. Russel said:
“ … nobody can prove that there is not between the Earth and Mars a china teapot revolving in an elliptical orbit, but nobody thinks this sufficiently likely to be taken into account in practice. I think the Christian God just as unlikely.”
So again, its not just something in space we don’t have evidence of. It is something between the Earth and Mars and revolving in an elliptical orbit. It is not just something but a piece of China. And it is not just China but a China teapot. So we can see the logic of conjunction fallacy at work reducing the probability.
But we also have background beliefs and evidence at work as well. Plantinga wrote:
“Clearly we have a great deal of evidence against teapotism. For example, as far as we know, the only way a teapot could have gotten into orbit around the sun would be if some country with sufficiently developed space-shot capabilities had shot this pot into orbit. No country with such capabilities is sufficiently frivolous to waste its resources by trying to send a teapot into orbit. Furthermore, if some country had done so, it would have been all over the news; we would certainly have heard about it. But we haven’t. And so on. There is plenty of evidence against teapotism.”
Another example might be if I say if we randomly pick a house in a New York suburb will there be three black horses in the Garage? You might say no. One reason would be conjunction fallacy. But also you would have back ground beliefs about what suburban houses likely have in their garage. If I asked does that house have light fixtures in it, well then you might say yes.
So Plantinga brings up evidence. And I think that may be a good way to describe what he is doing. But sometimes there are things that just don’t accord with our background beliefs. Now maybe these background beliefs are rational maybe they are not. Sometimes it may be difficult to completely understand or articulate why a certain claim seems true or untrue. Is there something beyond the known universe? In any case we have far more beliefs than we may even consciously realize.
For example, I never realized I believe plaid cats do not exist until last month. But I suppose I always believed it. (that is I was inclined to act in a way that accorded with the statement “there are no plaid cats”) This may be a sad fact of this kluge we call our brains. But it does not justify erecting some imagined burden of proof. That is just piling error on top of error. When we consider 1) our background beliefs, 2) evidence and 3) the logic of the conjunction fallacy there is no reason to appeal to the burden of proof.
If you are so inclined I would love it if people would post a link to this blog in the comments of the above youtube video. Perhaps we can get a discussion going and deepen everyone’s understanding of this issue.
True and Reasonable 
Hey Joe,
I agree with the probabilistic framing, but I’m not sure if people are actually intending the term ‘burden of proof’ to be understood as if it were some objective law which guides the conduct of rational argumentation. Couldn’t somebody saying “you have the burden of proof” also just be understood as using shorthand for “I perceive that I am in the majority when I say that a proper accounting of the available intersubjective evidence yields a low prior probability that your claim is true. You need to overcome that low prior if you want me to believe you.” Seems reasonable that a shorthand for that kind of concept would arise, even if its users haven’t taken the time to unpack it.
Hi Travis it is good to hear from you thanks for the comment. I am glad you agree on the probability part.
As for what people intend the term “the burden of proof” to mean in a philosophical context I think it is far from clear. That is actually one of problems I address in the first blog on the burden of proof. What are they even saying?
I don’t think people mean only what you say, but it is possible. Lets break down what you say.
1) I perceive I am in the majority in believing that your claim is unlikely to be true
2) You need to overcome that low prior if you want me to believe
(I left out the bit about “inter-subjective” evidence I would think they would take all the evidence into account when assigning it at low probability. If that seems problematic let me know and we can discuss it further.)
So is one important? That is it important that the person to be convinced feel his view is in the majority. If most people say believe in God in the world would and that person knows it would that change whether he feels the other has a burden of proof? It would seem not. But if it does would it be the majority of people in the world or just one community? How would that change whether 2 is is still the case? I mean if he is in the majority it would still seem that I would need to overcome that low prior if I want to convince him regardless of whether he thinks his view is in the majority or not right?
Now lets look at 2. Is it not just a truism? Well maybe not. Because the person might come to believe the same thing I do even if *I* don’t overcome the low prior. For example he may do his own investigation and overcome his low prior. But to the extent is it not a truism that is the part that is problematic. Why is there some burden on me for this other person’s beliefs? i guess I want to be generally charitable with what I have learned but it just strikes me as odd that people would suggest other people have some burden to help shape their minds to the truth.
Now if you want to do away with the whole notion of my really having a burden when people say I have a burden of proof then ok. Then you are just saying if you want to be the one who convinces me then you have to convince me to fulfill that desire. Sure that seems essentially a truism and I of course agree. But I don’t see why we need to talk about a notion of burden of proof to express that. And indeed the term burden of proof seems misleading. Because no one really has a burden.
But I get the sense people really do think there is some sort of burden on one side or another. And if they don’t meet that perceived burden then somehow their view at least lacks a certain epistemic luster if not being down right irrational.
But yes it is far from clear what the consequences of not meeting this burden are other than the other person feeling smug.
Lets take an example. I talk to someone and no one else hears the conversation and no one recorded it. He tells me X. He tells you he said not X. I tell you he said X. You ask me to prove it. Well sometimes you can prove something like this sometimes you can’t. In a legal sense I understand implications of not being able to prove it. I also understand that if whatever I say is not convincing to you then you may not be convinced. But I don’t understand how i or anyone has a burden of proof outside of court.
I don’t see how any of my claims outside of court are any different you should believe in God. You say prove it. I might give you some of the reasons I believe and you may or may not be convinced. But I don’t see how I have a burden. Of course the same is true of your beliefs. You might say I think you should not believe in God. And I can say prove it. And you can tell me why you think that and I may or may not be convinced. How does inserting the idea that either of us owes any sort of burden to the other help the conversation. I think it only makes it convoluted.
The inter-subjective part was there because I feel like burden of proof statements aren’t made against claims which lack public access Or do not involve widespread knowledge. I don’t pretend that my shorthand definition applies to all, or even
Oops. Sorry, accidentally hit send. Continuing on…
I don’t pretend that my shorthand definition applies to all, or even most claimants, but if I was asked to translate what I think most people really mean when they make a burden of proof claim, that’s what I would suggest.
If that is what they meant then I would ask them the questions in my response to your comment and be curious what they said.
I think those who invoke “the burden of proof” want it to serve them better than to just say “if you don’t convince me then you wont convince me.” But that is just a truism and any purpose beyond that is in my opinion not legitimate.
Also I have not heard that the burden of proof should be restricted to claim that involves public access. By that do you mean all the evidence of the claim must be equally accessible to the public? That is of course not the case in issues of whether God exists or how the burden of proof is used in law. Some people have experienced miracles (or what they believe to be miracles) directly while others have not and must rely on accounts.
I’m not saying that the burden of proof should be anything, only that it appears to be used in a certain way. The “public access or common knowledge” statement is an observation that the burden of proof is rarely raised against claims for which the person has no means of evaluation.