Default positions and the increased burden of proof

[Scott]

In this post on the thread "Belief: the difference between I DO & I DON'T" I showed that logically a rational person's belief regarding any given proposition must fall into one and only one of the following three categories:

A) "I believe X AND I do not believe -X."
C) "I believe -X AND I do not believe X."
D) "I do not believe X AND I do not believe -X."

If you dispute that, please read and respond to that post in that thread. In this thread, we will take that as an undisputed premise.

***

Sometimes people might say things like, "It takes more evidence to believe a positive proposition than a negative one." In other words, they appear to be taking a default position. I agree with this. But I don't think it's explained well in any common way. Whether or not a proposition happens to be worded with negative words is irrelevant to whether it is what people really mean by 'negative' when they such a thing. I think it is much more clear to call it a default position rather than a negative anything.

What exactly do I mean by default position? I mean that for some propositions we choose position A instead of position D when the evidence for or against it is equal or especially when there is no evidence at all.

This invariably raises the questions: Why do we do this? Is it reasonable or justified? Does it make sense? I think it is reasonable and justified, and I think it is unreasonable not to have default positions.

How much evidence it takes for one to believe a proposition can be called a standard of belief. Some people have lower standards of belief; some have higher ones. For example, epistemological skeptics and nihilists have such a high standard of belief that they do not think we can know anything and some may not believe anything at all at least not with confidence. In contrast, some people have a such a low standard of belief that they may believe in Big Foot, alien abductions, angels, ware-wolfs AND the claims of alleged psychics. Though different people may have different standards of belief, it's not necessarily irrational if they use it consistently. In other words, if Joe and Bob both think there is an equal amount of evidence for Y and Z and Joe believes Y and Z and Bob does not believe Y or Z, they might both be equally rational but just have different standards of belief. It would only be irrational for either of them to believe Y or Z but not the other in terms of standard of belief since they think there is an equal amount of evidence for each one. Do you follow me so far? If not, I recommend you post your questions or concerns in the thread about standards of belief.

Anyway, this gets more complicated when we take into account default positions. In other words, if X is the default position, then we can say it takes less evidence to convince a reasonable person to believe X than to convince the person to believe -X. Insofar as a person agrees that there are default positions, we could say that their standard of belief needs to contain two standards one for default positions and one for non-default positions. In the examples below, I think I will demonstrate that it would be very unreasonable for a person to not take a default position at least on some propositions.

Examples of Default Positions

Consider the proposition, 'my neighbor works at a Starbucks.' For the sake of argument, let's say we are talking about a specific neighbor who lives in one of the apartments next to mine but who I have never meet or seen. I have no direct evidence one way or the other about where he works. Before considering such an example, some people may say that D is the default position; in other words that if I have no direct evidence one way or the other that if I am reasonable I shall neither believe X nor believe -X. But doesn't that actually seem unreasonable? That doesn't appear to be what I do. It seems that insofar as I consider it without any evidence I actually disbelieve it. In other words, I believe that my neighbor does not work at Starbucks. Sure this isn't a very confident belief, but I at least act roughly as if it is true. In fact, it would obviously be unreasonable to say there is a 50/50 chance that he works at Starbucks. I would not treat that bet like a coin-toss. If someone gave me 2-to-1 payout on a coin-flip bet, I'd jump at it, but I wouldn't bet 100-to-1 my neighbor works at Starbucks.

Consider the proposition, "A unicorn lives on Mars." I don't think I have any significant, direct evidence against that proposition or for it. Just as with my neighbor being a Starbucks employee, I would have to say I not only do not believe it is true but I believe it is not true. The proposition, "A unicorn does not live on Mars," is the default position.

Consider Mary is murdered. Now consider the proposition, "Joe is not innocent of the murder." In the absence of evidence either way, I would believe that statement is false and believe Joe is innocent of Mary's murder. That's the default position. I think this is a good example because it shows that whether a sentence happens to be worded with negative words (such as no, not and none) is not what makes it the default position. What makes it a default position must be the actual meaning of the proposition not the way it is worded.

Now consider Big Foot, alien abductions and the Loch Ness Monster. Let's assume you feel like I do: there's a little evidence for each of these things, but mostly specious or not credible. In any case, it's not enough to convince me and meet my standard of belief. But I wouldn't merely say I hold position D on these things. In other words, I not only do not believe Big Foot exists, alien abductions have occurred on Earth or the Loch Ness Monster is real; but I also positively believe Big Foot is not real, alien abductions have not occurred and the Loch Ness Monster is not real. These are my default positions. In the absence of enough credible evidence to be convincing either way, I believe these things do not exist rather than merely not believing either way. This is particularly interesting because it shows that even when there some evidence is found to support the existence of Big Foot, it might at best convince me to treat it as having a 50/50 chance of being true or convince me from position D into position C. That's shows a significant difference in default believability .

Here are some more examples with the propositions that I believe by default in blue:

2,000 years ago exactly a butterfly froze to death.
The proposition, '2,000 years ago exactly a butterfly froze to death,' is false.
An alien named Josh with brown eyes lives on a planet within 1 million light-years of earth.
No aliens named Josh with brown eyes live on any planet within 1 million light-years of earth.
An invisible spirit follows each of us around from another realm who we can't see or observe in anyway and reports about us in that realm's equivalent of a newspaper.
There is no invisible spirit following each of us around from another realm who we can't see or observe in anyway and reports about us in that realm's equivalent of a newspaper.
We live in a computer program and everything we observe is a simulation in parallel to The Matrix.
We do not live in a computer program with everything we observe being a simulation in parallel to The Matrix.

Do you believe the statements in blue are true? I think I do. If you don't, do you at least believe that the statements in blue are more believable than their opposites? If you do not, then I think we need to settle this before we move on. If you do, then let's move on:

Qualities of a default position

So what causes one of two mutually exclusive propositions to be the default position? What makes a default position so much more believable than its opposite? Is it reasonable to consider this default position more believable than its opposite and if so why? This is my main question, so if you agree with everything I've said leading up to it, please try to answer it.

I will also try to answer it:

For one, I think what makes a default proposition more believable than its opposite is that its opposite is actually asserting something exists. Regardless of whether or not its worded with negative words such as no or not, a default position is a singular denial whereas its opposite is actually supposing real things. The premise from which we are concluding the defaults are more believable is that there are countless if not infinite things that don't exist that seem possible to us and of which we have no evidence of their non-existence BUT there's only a limited number of things that do exist. Consider the proposition, "Big Foot exists." If you were to believe that, then if you are rationally consistent you would likely believe that most if not all the other cryptids from the Loch Ness Monster to the Yeti also exist. Even if you were to say it is just as likely that Big Foot exists as that it doesn't, then that would imply that 50% of all the cryptids exist; That means there is 75% chance that either Big Foot or the Loch Ness Monster exist let alone calculating in all the others. If you believe it's just as likely that a unicorn lives on Mars as not and that it's just as likely that we are stuck in The Matrix as not, then that means you think there is a 75% chance that at least one of them is the case. If we consider it even half as likely that--without any evidence to indicate such--my neighbor is a Starbucks employee, then that would irrationally imply he has 2/3rds of all possible jobs. Rationality is all about consistency. It's irrational to assume that some hypothetical thing actually existing or some known thing having some specific quality is just as likely as not unless you are willing to believe that half of such hypotheticals are true. It's more reasonable to think no magical creatures exist on Mars then to think half of or even any significant fraction of possible imagined creatures exist there. For us to be rational, the probabilistic possibility that a unicorn exists with no evidence on Mars has to be seen as about the same for all other creatures we could possibly imagine being on Mars and for any thing we could imagine. In contrast, simply disbelieving it does not force us to believe a bunch of other hypothetical things exist or are likely to exist.

Similar to the above quality, another quality that I've noticed non-default positions have in common is that if true they confirm more information. For instance, the proposition, "I have a blue-eyed son named Tristen," is not a default position in part because it asserts so much. If we were to accept it without evidence, then we are simultaneously accepting all of these propositions it entails: "I have a child," "my child is male," "my child's name is Tristen," "I'm almost certainly not a virgin," "my child has blue-eyes," etc. As you can see, in the absence of any evidence either way, we are making less claims by taking the default position, "I do not have a blue-eyed son named Tristen." Even if that is false, any combination of those other statements could still be true. Thus believing it is false entails less immediate assumptions than otherwise. This is perhaps a definitive difference between a denial and an assertion: in a denial we are only denying that the specific group of propositions that all have to be true for the non-default proposition to be true are not all true; but some still can be in various combinations. When I say there isn't a unicorn on Mars, I'm only making a very specific denial; there could be some other kind of fairytale animal on there or something else we can imagine as possible. Whereas believing there is even a significant small fraction of a chance of a unicorn on Mars would also entail more liklihood for any entailed propositions such as "there is an animal on Mars," "there is life on Mars," "unicorns exist" etc. where any number of those could be true without the unicorn on Mars being true but they all have to be true for the unicorn on Mars to be true. When you assert a default position, you are inherently committing to a lot less than even suggesting the non-default has a significant fraction of a chance. In contrast, when you assert a non-default position, then you assert all of the necessary causes of what you are asserting.

Finally, and this is again similar to and overlapping the previous mentioned qualities, I think we can say a default position tends to come with less of an expectation of evidence. It's easier to get or inadvertently notice evidence that my neighbor does work at Starbucks if he does than to get or notice evidence he doesn't if he doesn't. Even in the unicorn example there seems to be more of a potential of evidence of a unicorn than not. If we spotted one on Mars with a telescope, then we'd have evidence. But to get anything considered to be evidence against the unicorn we might have to scout the entire surface of the planet, and even then we would hardly be as sure of its non-existence as we would be of it's existence if we saw it. In another example, we expect to find more evidence of people's guilt of a murder than not. It would be unreasonable to solve every murder investigation by getting a list of every person on the planet and search for evidence for each person's innocence and thus use a process of elimination to cross out the ones who are innocent. It makes more sense to conclude Big Foot doesn't exist from the lack of evidence for his existence than to conclude he does exist from the lack of evidence that he doesn't exist because there is inherently a greater expectation of evidence of his existence. And when we are deciding between two mutually exclusive propositions when neither has any evidence for it, all we have to go by is expectation of evidence. So whichever one has less of an expectation of evidence tends to be the default positions.

I think it is also important to note the parallels of considering default positions more believable than their opposite in the absence of evidence to the way Occam's Razor deals with competing theories.

I think it's also interesting how we seem to have this ability to figure out almost a priori which of certain propositions is more believable in the absence of empirical evidence based on the qualities of their meaning. One wouldn't normally think of, for example, disbelief in Big Foot as [i]a priori[//] or a matter of rationalism, but in a way it seems it is at least partially. Perhaps this connection to rationalism and so-called a priori knowledge was overlooked historically due to the traditional rationalists tendency towards trying to prove certain religious mythology or justify faith in the truth of certain religious mythology, namely the existence of a god.

What do you think?

Great set of posts about the basics, Scott.

I think this is similar to what I've heard referred to as 'the Standard model' in epistemology, but (ironically?) that terminology is non-standard. Basically, if you're making any sort of statement that goes against the beliefs of the average person, who takes his/her subjective perceptions pretty much at face value, then you got some 'splaining to do.

While this may not be the most logically rigorous way of going about finding the Truth, it does make allocating the burden of proof much more socially practical, even in a scholarly sort of debate.

You method is a little different than this, since statements like, "If I see something in front of me, I am justified in thinking that thing exists," would not be the default position. In fact, your recommended default position would make something like this pretty hard to justify. I'd say that this is how it should be, at least in a philosophical setting, but only up to a point. We should probably stop questioning when we get to general premises we all accept, at least until we agree on the implications of those premises.

[Scott]

I also don't think my justifications for making one position the default position as such work without first assuming the validity of logic including inductive reasoning. Dismissing the belief in certain propositions or even dismissing the lack of strong disbelief in those certain propositions as irrational means little without already assuming the validity of logic and rationality.

To avoid this default position standard leading us to utter epistemological skepticism or nihilism, I think we have to accept that a certain level of credible empirical evidence trumps the default position. In other words, we will believe the non-default proposition (e.g. Big Foot exists) once we have come into contact with a certain level of credible empirical evidence. I think the trustworthiness of empirical evidence--particularly in the form of the scientific method--is justified as an extension of inductive reasoning. And the inductive reasoning and the usefulness of logic and rationality had to be assumed from the get-go, I think, to justify the use of default positions.

Now one might ask how we can trust inductive reasoning let alone all logic and reason. I of course can't see any way to justify inductive reasoning that isn't begging the question by using inductive reasoning. However, I don't find it hard or that unsettling to assume it as an axiom, not just in for this discussion but in general, namely since it is admittedly fallible. This also goes along with your advice, Alun, to not doubt fundamental premises we all already accept as premises.