Some brief thoughts on knowing things

[andrewducker]

Over on Twitter WidgetFox replied to my link about Americans and Evolution (and that only "15% are right") by pointing out that the existence of God isn't testable, and (presumably) therefore saying that the people who were claiming God's involvement were wrong isn't reasonable.

After a teeny bit of conversation I offered to bring the conversation over here, because Twitter is awful for actually laying out thoughts. So I'll try to lay out what I think about Truth, Testability, Belief, etc. Those of you who have already studied epistomology can hit page-down now, and spare yourself my garbled explanations.

People make statements about the universe all of the time. These statements are usually made in absolute terms: "It is hot today.", "Arsenal are rubbish.", "2+2=4", "Evolution is true.", and are then reacted to as if the person who stated them believes 100% that they are absolutely true, with no doubt whatsoever.^[0]

So the first thing to mention is that very few people have 100% absolute belief in anything they state. Most people, if you speak to them reasonably, and they don't feel that they're backed into a corner, or that acknowledging doubt will cause them harm (physically, socially or otherwise) will admit the possibility that their belief is not true^[1].

There are multiple classes of beliefs about the external world that can have their fact levels debated:

• There is belief that is based on logical systems that are self-contained (like mathematics) - "Given the rules of mathematics, adding two and two produces the answer four."
• There is belief that is based on experience of the world - "I saw the car, it was red." This can be further subdivided into things that other people can verify "I took a photo of the car, and as you can see, it is clearly red."^[2] and things which they cannot such as "I feel happy"^[3].
• There is belief that is based on extrapolations from our experiences of the world. "I only told Dave about Jane's surprise birthday party, and now Jane knows, therefore Dave told her."
• There is belief that is based on other people's extrapolations and experiences of the world "My physics book says that the formula for gravity is Gm₁m₂/r².", "My physics teacher told me that Newton was hit on the head by an apple, and that's how he came up with Gravity.", "Pythagoras says that irrational numbers are impossible." And, again, these can be divided into ones that can be verified ("Julie's PhD shows how she achieved her conclusions, includes all of her data, and her methods chapter tells you how to repeat her experiments."), and ones that can't ("Joseph Smith says he translated The Book Of Mormon from a golden book given to him by an angel. But then he gave it back, so you can't see it for yourself.")

There are also beliefs that don't need any justification to the outside world at all - beliefs about internal states, like "I don't like cheese.", "I feel good when I go to Church.", "I am aroused by tentacle porn" , "I'd rather people didn't strangle puppies." and "I enjoy listening to Nirvana more than One Direction." - all of these are purely descriptions of how a person feels, and as such need absolutely no reasoning behind them. It's only if people try and claim that these internal states are actually statements about the outside world, and that other people should be bound by them that they need any kind of discussion. Personally, I prefer a world in which nobody strangled puppies (and am more likely to vote for people who are pro-puppy), and also one without the government forcing (or denying) cheese consumption. It's very much up to individuals to decide how much their morals affect their actions, and how far they are willing to go to make the world into what they see as better. I do find that phrasing these beliefs as personal feelings, rather than statements about the outside world helps a lot ("I don't like cheese" is much easier to deal with than "Cheese is bad and wrong".)

Going back to beliefs about the outside world:
If someone is merely passing comment on their belief system "I believe that the primeval cow licked the first god out of a salt stone.", then one can happily take that as a statement about their beliefs and exchange views with them by telling them that "I believe that Earth gave birth to Sky, who then fertilised Earth and gave birth to some deities who overthrew their father, and then gave birth to different deities who overthrew them in turn.", and then further explanations can be fascinating as you delve into the wildly disparate beliefs you both happen to hold.

There are four levels of evidence for beliefs:
1) There is clear, replicable, evidence for your belief. Photos of Titans overthrowing Chronos, Zeus hosting a Tuesday-night chat-show while maintaining the form of a swan, etc. This should make persuading other people a lot easier.
2) There is personal evidence for your belief. While the chariot of Helios isn't seen drawing the sun across the sky each day, Zeus _did_ once turn up in your home dressed as a shower of golden rain. Possibly you hallucinated the whole thing, but possibly you didn't - and having experienced a god first-hand seems like a pretty good reason to believe in them^[4]. Can be upgraded to (1) by having a camera with you next time Zeus pops round for a shag.
3) Other people tell you that Apollo is real, and will help you out if you pray to him. They don't have any proof, and Apollo seems to work more through coincidence than anything else, but surely that many people can't be wrong? The problem here being that there are lots of other religious believers who believe completely contradictory things to your lot. Sure, they have a bunch of books, but so do the Latter Day Saints, the Scientologists, and John Doe from Se7en.
4) You really, _really_ want to. This one doesn't usually happen by itself. There's usually either (2) or (3) to act as a trigger, followed by (4) to hold the belief in place. Sometimes, though, people start off because of a group of people/books, and then they abandon one part of the book, and some of the people, and change their belief to be something that makes them happier. But now they don't even have "A bunch of people claim that their distant ancestors saw Zeus, back when he used to meddle all the time." as an explanation for their belief, because the Zeus they believe in doesn't match the one that the Strict Zeusists claim to have seen.

Now, these four different options aren't just applicable to religion, of course. They can equally apply to other beliefs about the world.
1) A scientist has done a high quality study, that is entirely openly published, and has been replicated by numerous other scientists all over the world. Other people are free to build on this, or try to refute it, but currently it has not been contradicted.
2) You've discovered that eating beef makes you sick. Possibly there's something going on at a biological level where you have an unusual physiology, or maybe you once ate some bad beef and now you feel nauseous when you even think of eating it. (Alternative example: You find social situations difficult and once did an internet quiz that told you you had Aspergers.) Upgradeable to (1) by getting an expert to test you, but may be useful on a personal level even without that.
3) Many people have told you that they diluted something poisonous with so much water that there was none left, and that taking it helped them with their problem^[5], despite there being a complete lack of any evidence that homeopathy works.
4) You've now defended homeopathy/the perils of beef-eating so many times that it has become part of you self-identity, and is thus hard to let go of.

In any case, my reactions to people's claims vary dramatically, depending on how well I know them, how their claims fit in with existing beliefs, what kind of mood I'm in at the time. The majority of the time that people claim that things are true without proof I just move on. I can happily talk to a Christian about how awesome Avatar:The Legend of Korra is^[6] without having to worry about whether we share a belief in how the universe began^[7]. Also, I don't see any point in jumping up and down on other people's beliefs, unless they're using them as an excuse for behaving badly towards people. I'm fascinated by why people believe - and even then I try to avoid drilling too deep, because I don't want to offend people. However, when people say they believe things that they admit are untestable, then my reaction is to wonder what it was that caused them to believe in something without any proof. Personally, I'm much happier saying "I don't know" about the gaps in my knowledge than I am making assumptions to fit into the gaps that happen to make me happy. It seems a far more honest a way to deal with the world.

[0]I'm going to avoid using the word "Knowledge", because it gets twisted around by people. I'm using "Belief" here to mean the ideas people have about how the world (and other systems) work.
[1]Personally, I'm not willing to discount the idea that I'm living in a computer simulation. Which would make all my other beliefs about the world immediately suspect. I don't think it's very likely, but I can't see any way to disprove it.
[2]Leaving aside here differences in perception, colour-blindness, etc. When struck by daylight, the car reflects light of a specific wavelength, generally named "Red" by those who see it.
[3]Well, until we can analyse the brain at a sufficiently high level.
[4]I know a couple of people who fall into this category. And much though _I_ interpret their experiences as being brain-spasms, I can at least firmly understand why they believe. Partially because I don't know what I would do if I thought I saw an angel/demon/vampire/alien one day, but nobody else did.
[5]Thanks to the wonders of regression to the mean. Basically, if you're really ill, chances are that you're about to get better, whether you take the sugar-water or not.
[6]Not as awesome as Avatar: The Last Airbender, but still pretty damned awesome.
[7]My current working assumption is "with a big bang", but I'm by no means certain. As to what happened before that, I just don't know, and I don't see any point in making guesses, except for entertainment value.

Comments

[redbird]

It seems from here that believers who assert that the existence of God isn't testable are ceding most of the argument to us atheists: they're saying that a universe containing their god(s) is indistinguishable from one that isn't. If that's true, there doesn't seem much point to most religion. Claims about reward or punishment after death don't change this, as far as I can tell, because someone who is claiming that the existence of God is untestable is saying we live in a universe where we can't tell whether there's an afterlife, let alone what if any behaviors will affect what happens to us after death.

[andrewducker]

Yup, I totally agree.

[naath]

I think most of the "point" of religion is that it provides a place you can go every-so-often to hear someone speak on inspiring/moral topics, have a bit of a communal sing-song/recitation/dance, and then have a natter over tea and biscuits; with occasional excuses for a party and occasional periods of required restraint. Religion provides a to the day, the week, the year; and also a structure of Rules For Living By. It provides social occasions, and gives people a common interest/background.

And of course it is fiercely painful to be forced into a social structure that marginalises/oppresses you; so all of the stuff that for some people makes religious organisations comfortable-feeling places to be can also make them horrible places to be for others.

[redbird]

Patterns and such, yes. One of my partners is part of a Jewish congregation, largely because she values the community and some bits of ritual. But that's separate from the argument that somehow jumps from "there is no way to test the existence of [a sufficiently uninvolved] God" to "therefore you must follow these rules even though there is no evidence or argument that they benefit people." If there's no way to test the existence of God, there's no way to back the claim that God wants people to do, or not do, certain things.

The point of religion for some people is that it gives them those things. The point of religion for other people is that it's a source of power or money or a get-out-of-jail free card (the latter include religious leaders who get to keep molesting children or get away with stealing from the congregation because it would be "bad for the church/community" if they were exposed and removed from their jobs. If keeping the sabbath helps you, fine; the problem is finding a way to provide that which doesn't lead to parents being ostracized because they reported that the rabbi, priest, minister, or imam raped their children.

[innerbrat]

I find it interesting that in the second example you kind of use 'being an expert' as something that fits level (1).


Of course, it depends on the type of test the expert is doing: finding no evidence of beef-digesting bacteria in your gut, followed by a faecal transplant which removes the problem, could be taken as (1) level evidence, but 'this person has a professional qualification and agrees with me' is still personal evidence.

Though that's probably exactly what you mean anyway.

[innerbrat]

Also, the first 8 episodes of Korra > the first 8 episodes of AtLA. I'd even argue that the first third of Korra > the first third of AtLA.

So there.

[andrewducker]

I think the plot and dialogue is better written than at the start, but I'm not feeling the same love I felt for the characters in TLA, or the same urgency at the doom the world is undergoing.

[andrewducker]

Yes, sorry, I don't mean "See someone who has a certificate", I mean "See someone who is following methods produced after significant testing, feedback and verification."

[veronica-milvus.livejournal.com]

When you say people don't usually believe things 100%, I generally find that religion is the exception. It's 100% or you are a sinful backslider who will burn in hell, so it is 100%.

[innerbrat]

Religion isn't a singular belief, though, but an umbrella of a series of intersecting beliefs, all believed to varying degrees by the people holding them.

[andrewducker]

There are massive numbers of Catholics who use contraception.

And the Church of England seems to be built on people who kinda believe, a little bit, when they don't think about it too much.

[anef.livejournal.com]

It's a very good point that believers don't believe 100%. Faith fluctuates from day to day, from hour to hour, and that's why the Catholic Church has a programme of continuous reinforcement known as taking part in the sacraments (going to mass, etc). Nevertheless, believing in God is not the same as obeying the rules of the church. A person may be a very firm believer in the Christian God, but not accept the Catholic Church's teachings on contraception or women priests.

[bemused-leftist.livejournal.com]

Well, we might distinguish 'I'm 100% certain' from 'true of 100% of cases, no exceptions'. I'd agree that most simply stated opinions (of whatever degree of certainty) do tacitly allow for exceptions.

[lizzie-and-ari.livejournal.com]

Interesting - I too took exception to the wording in that article, but actually because it mixes these kinds of beliefs:

1) 46% of Americans believe in Creationism
As a statement, this is a level 1 belief of the poll takers, and anyone looking at the poll result. It relates to the level 3/4 belief of a proportion of people.

2) 32% of Americans believe in god-guided evolution
As a statement, this is a level 1 belief of the poll takers, and anyone looking at the poll result. It relates to the level 3/4 belief of a proportion of people.

3) 15% of Americans are actually right:
As a statement, this is a level 4 belief of the person writing the article. It undermines the level-1 believability of the poll. It would be like Julie reporting the 45% of people enjoy jogging in group, 34% of people prefer exercising alone, and 21% of people are lazy. You're suddenly introducing a different level of belief into reporting.

Believing that evolution is god-guided or that it is not god-guided is undeniably a level 4 belief because there is no _possible_ evidence either way. If you believe that a visible god, Mr Tumpty, who wears a blue coat and lives at 34 Brick Lane, Soho, is at constant work evolving the universe whilst providing an unending well of gold in Govan, it becomes a level 4+ belief - there is level 1 evidence that your belief is not true. Someone refuting that specific belief would hold a level 1 belief - they have examined a clear claim against clear evidence and come to the conclusion that it's not true.

If someone believed, however, that 14 billion-odd years ago, someone began what it essentially a very sophisticated science project wherein particles exploding from a big bang would evolve over time, and that this someone had specifically designed the way that this things would evolve and it was all going perfectly according to plan, then there is no evidence against that whatsoever, there's just no clear evidence for it either. People have to make a choice as to whether to believe in that and that choice comes down to what they want to believe.

Most people's belief in God, of course, is somewhere between Mr Tumpty and the Universe science project. They believe that there's something else going on. Not being able to truly define what that something else might be, they create stories around it, often incorporating useful rules to live your life by and interesting allegories to explain why stuff works that certainly does consistently seem to work.

Where both fanatical believers* and fanatical non-believers* go wrong is when the stories they have built up become the focus - and they mix and match stories in the wrong way:

The bible tells you a condensed version of the history of the earth* and you decide that that means that anyone who suggests anything different is going to go to hell. Conversely, you mock the communion service you think that the churchgoers really believe they are eating human flesh.

Oh dear that ended up being longer that I intended it to be and I should cut it down now but fuck it.

xxx


* encompassing religious people, homeopaths, actual lunatics, whatever
* broadly defined as people who angrily refute people's right to believe anything beyond level 1
*** First light, then a planet, then plantlife, then sea life, then animal life, then people, then evil... they got the stars in at the wrong bit but it's otherwise pretty much there





(That's without getting too far into the poll validity, which illustrates the point that all beliefs are actually to one extent or another based on other beliefs, but that's a whole different strand)

[widgetfox.livejournal.com]

I think this pretty much says exactly what I would have wanted to say, but much better. Thank you.

[lizzie-and-ari.livejournal.com]

Thanks, you're welcome x

[bemused-leftist.livejournal.com]

" there being a complete lack of any evidence that homeopathy works."

Isn't that rather an absolute statement? How could you know what evidence may exist which you haven't seen?

[andrewducker]

If there is positive evidence then I would have thought the homeopathic groups would have presented it by now. Rather than live with the ignomy of the numerous meta-analyses showing that it's no better than placebo.

[bemused-leftist.livejournal.com]

That's a lot of assumptions for an absolute statement to rest on.

[nickys.livejournal.com]

How about:
"All claims about the validity of homeopathy that have been checked so far can be explained by the well known and well documented placebo effect. Therefore unless further evidence is presented and proved it is reasonable to say that homeopathy does not work"
?

[anton-p-nym.livejournal.com]

No incontrovertable evidence has been brought into public view that shows homeopathy having any greater benefit than placebo. A few studies claim to show that, but their methodologies are considered dubious by trained scientists.

I am, therefor, comfortable with the phrase "complete lack of any evidence that homeopathy works" in casual conversation.

-- Steve will also note that there may be evidence of giant space turtles preparing to devour the planet that we haven't seen, but only the barking mad would demad we construct anti-turtle lasers without presenting such evidence.

[ggreig]

I dunno, if the consequences of not having anti-turtle lasers are sufficiently severe (planet-devouring sounds pretty bad), and the costs sufficiently low, at some point won't it become a wise precaution, regardless of its likelihood? ;-)

Of course, I'm not sure how you'd establish what would be an effective anti-turtle laser without some evidence of the turtles...

[bemused-leftist.livejournal.com]

Some of the assumptions this absolute statement ("complete lack of any") rests on, are indicated by 'incontrovertable', 'dubious', and 'trained.' And 'not brought into public view' does not equal 'does not exist'.

[f4f3.livejournal.com]

Evidence that Andrew has seen might possibly exist. Evidence which no one has seen does not exist as evidence.

[cartesiandaemon.livejournal.com]

I was actually thinking about something similar recently, sparked from a less wrong thread about falsifiability and occams razor.

(For the record, occams razor as it is commonly used in science, and, I think, seems to work well, not a literal interpretation of what counts as "multiplying entities")

That in many ways, falsifiability and occams razor are quite similar. But not completely so. We usually implicitly compare theories to our current best theories. But it might be clearer to always compare two specific theories. Then two theories are equivalent -- effectively synonymous -- if they agree on all possible predictions. And in some sense, if one is right, the other is just as effective, not wrong per se, but we may still have a preference (a) for the one that is simpler to EXPRESS and (b) for the simpler one, on the grounds that it's more likely to be right if we ever CAN distinguish.

This occurred to me about the less wrong sequences about many worlds interpretation of quantum dynamics, or assuming that the universe continues beyond the light cone we can ever see, that the theories that they do, or don't, are equivalent, but the one that fits more naturally is better scientifically, even if not (yet) more true.

Similarly, I think "there's an invisible unicorn no-one can detect, but otherwise the universe continues to behave like the most up-to-date theories of physics" is arguably compeltely equivalent to the physics theories, but most people intuitively see that the theory WITHOUT the unicorn is superior.

So in some sense, there's no difference between a God who doesn't affect the universe, and none at all.

Although in fact, I think most people who say that, feel God DOES make a difference, if only in morality.

[drdoug.livejournal.com]

"Personally, I'm not willing to discount the idea that I'm living in a computer simulation."

I am. At least, I'm willing to discount for practical purposes the idea that I'm living in a computer simulation that there is no conceivable way of telling from a "real", physical world. It seems like the modern version of the Omphalos/Last Thursdayism idea. (Those are variations on the theory that God created the world in 4004BC/last Thursday but in such a way that it looks like it's billions of years old.) None of them are theories that can possibly affect the world. They are not even irrelevant.

[andrewducker]

When I say "I'm not willing to discount it" I mean "I am not willing to state that I am definitely not", not "I will live my life as if I might suddenly find out that it is the case". It's not a Pascal's Wager kind of thing :->

I feel the same way about the existence of Gods (or Space Gods, or Aliens). Sure, they're a possibility, but not one I spend time thinking about.

[khbrown.livejournal.com]

"If men define situations as real, they are real in their consequences." - W. I. Thomas

[randomchris.livejournal.com]

Important distinction to make: there are two classes of belief above. There are people who believe that everything they believe is the absolute truth and anyone who thinks differently from them is simply wrong (for some people this requires showing them the error of their ways, others are content to just "agree to disagree" while secretly thinking "but I'm right, of course"). However, there's a second option; the ability to accept that others' beliefs are equally valid as long as they agree with the available evidence.

In Christianity (for me at least) this means accepting that, for example, some people see homophobia as a religious duty. I disagree entirely with placing a high priority on it, because it's mentioned about three times, but I can see the verses they quote and understand why they've come to that conclusion, while obviously coming to the opposite conclusion myself.

With homeopathy, I see people who believe homeopathy works, and I'm happy for them to continue believing that; they might be at least partly right (statistically possible). I can disagree entirely with any placement of it above methods which have much stronger proof of working, and have justification for doing so, but I can't say they're completely wrong about homeopathy.

[andrewducker]

There are two different classes of belief there:
1) Preferences/feelings. "I find gay people icky." is an emotional feeling, and as such not one I can tell people they aren't allowed to have. They have a particular feeling, and that's just how they are (although, obviously, feelings can change over time). Feelings aren't right or wrong, they just are.

2) Statements about the outside world. Homeopathy either works, or it doesn't work. If it works, then it should be testable. Take 1000 people, give half of them a homeopathic remedy, the other half get a placebo. See which half come out better. (Hint: Been done, they come out exactly the same.)