User:Armondikov/Now/Jan12

From RationalWiki
Jump to navigation Jump to search
Icon bullshit.svg
Thought of the Now

...half-baked ideas from the mind of someone who is almost certainly dumber than they think they are.

Archive


29th January 2012[edit]

Facetious response.

You often hear "if atheists don't believe in God, why do they talk about Him so often?" though you can very similarly say to many fundies that would be like that "you don't believe in gay marriage, why do you keep going on about it?"

Fuck these people.

26th January 2012[edit]

Theatre conundrum. You have Constantin Stanislavski, Bertolt Brecht and Adolf Hiter in the same room. But only two bullets. Who do you shoot?

23rd January 2012[edit]

Privilege (II)

Okay, where was I... okay, expanding on some points made previously.

1) Focusing on white and male privilege distracts from the general idea of selection bias and how it applies to human reasoning, and instead seems to focus a type of blame and guilt toward categories of people. Categories that, may I add, that the people have no choice in being added to.

Selection bias is one of the most powerful biases, not just because it can be cognitive but because it can also be systemic - though this latter feature actually makes it easier to notice and easier to correct for. But this only works when we know the mechanism of selection, and whether this is systemic (as in a poll conducted informally on a website) or due to experience ("I have never seen X, therefore X does not exist"). As a bias, it is something that aspiring rationalists need to account for and control carefully. The important thing about such biases is that they apply equally, we all experience them and all suffer from the drawbacks that they produce unless we're aware of them.

We can easily look at what experience and bias we have due to living in certain parts of the world. We could have a "first world privilege", for instance, due to living in a first world country in Europe or Australia. If "white privilege" is what hidden boosts are caused simply by a lack of experience of racism, then surely the benefits afforded by "first world privilege" must dwarf those on absolute terms. Our experience telling us what is bad and what isn't bad is formed around the fact that this "first world privilege" means we don't, for example, live in a theocracy, or live in a place ravaged by war. There is a certain "privilege" attached to class and where people are born, and how rich their parents are - again, I'm sure these would, as measurable effects, dwarf sexism and racism in our modern environment when like-with-like is properly compared. Our experiences are molded around such things, and our attitudes adjust accordingly. This isn't to dismiss these effects and non-existent, but to show that what is being called "privilege" is actually part of a wider class of more applicable biases. Biases work on individuals, not on broad groups.

By taking a specific type of bias and separating it out by symptom, rather than cause, we may hinder the general rules we can learn to undo our biases. The general idea of "privilege" appears to be along the lines of "white people don't experience racism, therefore their opinions on racism in society are skewed", and is almost the perfect definition of an observation selection bias - the bias that changes your opinion based on what evidence you're able to observe from your perspective. However, we can validly reverse this phrasing and instead say "black people do experience racism, therefore their opinions on racism in society are skewed". Skewed in this sense is not to say "wrong", it merely says that our experiences determine how we view the world - and we certainly wouldn't call the latter "privilege" because there are no positive connotations to it. Yet mechanistically, both are the same thing; limits of experience determining someone's view. Two people, someone who does and someone who does not experience a particular form of discrimination, will both have their views of the world shaped by their experience. Both, therefore, have observation selection effects telling them what to think. By focusing only on one side of this and giving it a name and new category for it, we deny that observation selection effects may occur equally to everyone, or even that they are simple biases at all.

But here's the thing about giving it a name: you're now saying it's special and unique, and because the category exists other factors will begin to be bolted on to it. It now becomes a "thing". No matter how many times you try to say "privilege is just descriptive of the facts" or "privilege doesn't mean you're being blamed for racism", it doesn't stop those connotations being there because you have given it a name, you've painted a large group of people with this "thing". Biases are assessed and judged on an individual basis - on the other hand, labeling one of them as "privilege" (especially with such an emotive term that conjures up images of posh people born into plenty of money, too) draws a ring around groups such as "white people" and "men". In generating this label you cease to determine and judge biases based on individual experience and instead based on, at best loose categories and at worst just on stereotypes. "Privilege" then becomes a lazy mental shortcut to an idea, which while useful (when addressing symptoms) runs the risk of abuse (where you attempt to address causes).

So by accusing someone of having "privilege" you produce, as a side-effect, this notion that race, sex and gender are the root causes of bias (hence the connotations of blame for something that one cannot control), when in fact it is limited individual experience that is the root cause. Knowing about this general class of bias is a far more powerful idea to understand from the perspective of an aspiring rationalist.

19th January 2012[edit]

Privilege.

Recently, I've been reading a few feminism and anti-racism blogs, and this one word keeps cropping up more than ever, so much so that it could make a great drinking game; privilege.

Crommunist defines it pretty well here - and I'll begin by saying he's one of the most respectable bloggers on the subject that I've spotted. In the sense used by liberal intellectuals (our experience with right-wingers using the word "liberal" with spit and venom makes it difficult to say "liberal" without sounding snide, but you should agree that those who use it are liberal), it's an explanation for the attitudes and actions of groups of people experiencing a lack of racism, or sexism, or homophobia, or transphobia. Invariably, "privilege" like this is being used in only these sorts of settings, to refer to sex/gender issues and race issues.

It's a concept that has been developed to address people who claim to not experience or be aware of sexism and racism, precisely because they're male or white and so won't see it. This much is obvious, but let's break it down into the most general and basic formulation; these people would be saying "I have no experience of X, therefore X does not exist". They say this because they have "privilege".

If we actually examine what privilege is in this sense, it's actually a far broader phenomenon than the exclusive use against "white" people or against men suggests. It might sound familiar to those experienced with rationality and familiar with heuristics and biases; it's a form of selection bias. I liken it to an observation selection effect, or an anthropic bias, where the limits of our experience alter our perception of the world. This is the route of things such as hindsight bias and the narrative fallacy because our experience is limited to what has occurred, and what has not occurred to us is completely removed from our reality - the current textbook example of this is the countless other terrorist attacks that didn't happen instead of 9/11, and is one of the key things to understand when it comes to Black Swan Theory. Knowing about the observation selection effects (for which Nick Bostrom's book is probably the go-to work for the ins and outs of it) is a powerful thing for skeptics and rationalists. This is perhaps extreme next to experiencing a lack of racism (or lack of experience of racism if you prefer it that way around) but the end result is the same: we are biased by our experience and the failure to take into account evidence that we don't see.

So white and male privilege is really just a specific term for a more widely applicable selection bias, and this is where I take issue with it: I see no compelling reason than we should circle this specific instance and generate a new category for it. If you generate a new category it should be for good reason, it should be distinct, and it should be categorised and labeled correctly. I come at this is from two perspectives. 1) Focusing on white and male privilege distracts from the general idea of selection bias and how it applies to human reasoning, and instead seems to focus a type of blame and guilt toward categories of people. Categories that, may I add, that the people have no choice in being added to. 2) Instead of framing it as a general cognitive bias, it frames a negative - namely, the lack of experience of racism/sexism/-phobia - as a positive thing that someone has. The latter being particularly problematic as you begin treating a lack of something as a real thing that can be pointed to; consider light and dark, hot and cold, theism and atheism, for instance.

This often leads to lazy reasoning that consists of nothing more than firing off the line of "you're just privileged" - something we don't want to encourage as aspiring rationalists and skeptics. I'll expand on those two points separately at a later time. For now, feel free to dismiss this as just my privilege talking.

18th January 2012[edit]

Tebow; a one-in-a-million chance.

I recently spotted this Conservapedia-inspired claim that Tim Tebow's 316 yard game (in reference to John 3:16) is a million-to-one chance, or even a 10,000 to one chance, after it was recently posted to FSTDT. Though sport bores me, and American Football more than most, pedantry about probability, mathematics and rational analysis certainly doesn't.

A true million-to-one chance is a lot more difficult than most people think. In it's most basic form you should expect that the value of Tebow's stat should be able to take any range from 1 to 1,000,000, each with equal opportunity - to visualise the probability this way better, 1,000,000 yards is about 570 miles, or alternatively, you'd expect him to hit that figure only once in every million games, given an NFL season of around 17 games, that's once every 60,000 years of near continuous seasons. That's it, once you've made your million-to-one prediction, you're saying that in more or less all of recorded human civilisation and some more, Tebow will throw for 316 yards about once. But to cut a long story short, this would also imply that throwing 315 or 317 yards would have a near identical odds ratio so isn't terribly impressive on its own.

Of course, million-to-one for this is a nonsense figure anyway. The range of values that Tebow's passing yards stat can take is actually far shorter than 1-1,000,000. An exceptional game appears to be anywhere above 400 yards, and the average of passing yards for professionals seems to be somewhere between 80 and 270 yards. So a more realistic range to think about is 50-400 yards, and that his stat could be any number within that range. The odds of him getting 316 yards would then be about 1-in-350, which while still spectacular (it's once in 20 years of NFL seasons) it's still considerably less than the one-in-a-million touted. Even being very generous we couldn't bring this alone up to one-in-a-million - there are a lot of orders of magnitude between 350 and and 1,000,000 that you have to account for. To illustrate, the difference is approximately the difference between tossing a coin with all heads (strictly speaking, any pattern, but let's say we made a prediction of "all heads" in advance) in a row 9 times, and getting all heads 20 times. That's just over doubling the bits of information involved if you like it in information theory terms, and that's quite a leap.

However, improbable things happen because there are always repeated attempts and the actual real-world odds, when stripped of their biases, are altered by the number of games played, the number of potential stats to look at and the potential for different players to be counted in this mystical search for Biblically relevant numbers. There are 17 games in an NFL season plus however many other fixtures you care to add, and as for what stats you can mine for data, there are lots. Seriously lots. Each one of those doesn't work against a probability by multiplying it, it adds to it. Even from a very generous base figure of 1-in-1000 - that's only getting a 316 yard stat once in 70 years of constant NFL seasons - our multiple games improve the odds rather than make it less likely. Playing 70 seasons makes a thousand-to-one chance near certain as said, so an entire season makes it closer to 17-in-1000. This cancels down, rounding off because we're only playing a rough game here, to approximately 1-in-70 across an entire season. And remember, by the "improbable things happen" law, this is only the odds based on us saying, in advance, "Tim Tebow will throw this value this season". He may have written John 3:16 on his face, but that makes it coincidental, not odds defying, and understanding the difference is key to probability. We can, in fact, presume very similar odds for him throwing 315 and 317 yards, or 314 and 318 for that matter or 313 and 319 and so on as the distribution of potential values spreads out. Each one would have a similar (im)probability, but only one just so happens to match the Bible verse and one of them has to actually occur, which is why probabilities add up to (or at least, should add up to) 1. Multiple people and teams competing effectively add more attempts, reducing the odds and making it an outright certainty that the value would occur somewhere even in just a few seasons (far less spectacular odds), but even if we do accept that we're just focusing on the Tebowster due to his taste in face paint, it's emphatically not beyond chance. It's certainly not a million-to-one shot, and if the realistic odds of 1-in-70 (over a season) is your threshold for divine intervention you probably could do with a slap.

None of this matters from a prediction point of view, of course. After an event happens, the probability is always 1. At least in so far that 1 is a probability in the real world. In order for this to be in any way remarkable, someone would have to say, in advance, that "Tim Tebow will throw 316 yards, in reference to John 3:16, this season or this specific game". Because every value is equally (im)probable, and certain to happen after the event, it's easy to make post hoc statements about miraculous odds and fall into the narrative fallacy. It only becomes interesting if predicted in advance. But even then, the remarkableness of such a prediction is stymied by the fact that there are potentially millions of sports fans and Tebow worshipers who could have said it. A million-to-one chance attempted a million times reduces to 1-in-1.

This goes for adding in the other stats that have been part of this fable, such as his throwing average and the TV viewing figures, which have been touted as being 31.6 (though any real reference to the Bible should be 3.16). Hitting these values might not be even as impressive as 316 yards, as even though they're also to three significant figures, the range that they can take and fluctuate around is considerably less than the passing yards stat. But more importantly, for every one of these statistics you add you also end up with multiple shots at it due to data mining - so for every new dimension used to decrease probability, there are as many dimensions there to increase it too. The thing about this is that even if you demonstrated that a 316 yard pass, a 31.6 yard average and a viewing figure of 31.6 was a miraculous million-to-one shot... then a 317 yard pass, a 31.5 yard average and a viewing figure of 31.8 would alsobe a miraculous million-to-one shot, ditto 312, 31.7 and 31.3, ditto 321, 30.8 and 33.1... pattern recognition isn't particularly impressive from an odds and probability perspective. Tests for statistical significance always take into account how you can find multiple variables and mine them for trends - and let's not get started on numerology, which potentially guarantees you'll find some interesting quirk in any one of the countless stats provided to football fans who like to nerd out over them.

The moral of the story is that probability isn't something you can just make up out of nowhere and claim is mysteriously significant. It does genuinely mean something.

But as an aside to end this, I want to consider a point about proof and evidence. If you're going to claim that something is true and state "X is the evidence", then then you must accept that a lack of evidence should be in favour of it not being true. We do this every day, the lack of a fire breathing monster when opening a door is evidence against a dragon living in my garage, for instance. So if you accept Tebow throwing 316 yards is "proof" of God, then him (or any other quarterback for that matter, given how "improbable things happen" probability actually works) not throwing a Biblically significant value should be proof against God's existence. We see people fail to make 316 yards, or 31.6 average, all the time. Because we see Biblically significant numbers far less often than we see don't see them, we must conclude, in fact, that there is no God. This isn't being facetious, it's simply the logical conclusion of someone making a statement about how their belief and evidence interact. If you don't accept that argument, then you can't use this evidence as proof of God, simple as that - otherwise your belief is true in all possible worlds, isn't conditional on evidence or a real-world interaction and, to use the technical term, is a load of old bull's bollocks.

16th January 2012[edit]

Belief

It's quite trivial to prove or disprove something. You simply ask "what do you expect to see if this is true" and set about finding it. That's not to say it would be easy, as the thousands of people working to find the Higgs Boson at CERN will attest to, but the method is at least piss simple. Observations = true. No observations = not true.

No, the trick is actually getting people to believe it. To overturn prior beliefs and accept a new one, and, and this is the worst, admit that they may have been wrong. That's difficult, and very much non-trivial.

When someone says "these are difficult questions" they most likely really mean that they don't want to face the answer. It's trivial to disprove God's existence, it's another matter entirely to get the ones attached to the belief to drop it.

13th January 2012[edit]

Right and Wrong

A quick visualisation of what "correct" and "incorrect" kinda mean. Things can be outright proved wrong because observed characteristics can conflict with or not be accounted for by our inferred model of them. Conversely, things can never be outright proved right as inferred characteristics are always a larger set than observed ones. I.e., the human body can be killed by both cancer and being ran over by a train, but for any individual we only get to observe one of them, the other remains an inferred only.

Right-wrong.png

11th January 2012[edit]

The Internet Debating Dictionary

Debate: (n);

  1. Event held by two or more individuals where they ignore each other for as long as it takes for them to get around to saying their own thing.
  2. Process of shouting at someone until they give up.
  3. Used as an excuse to put off making a firm opinion and backing it up.
Usage "let's have a debate", "we should debate this subject"

Open-minded: (adj);

  1. "Why won't you agree with me!!!"

Philosophical grounds: (n);

  1. Preface to an opinion used to fake legitimacy.
  2. Last ditch attempt to avoid saying anything relevant to the real world.
Synonyms: Making shit up, bullshitting, diverting attention.
Usage: "I'm not a racist, I just object to black people on philosophical grounds."

Rational: (adj);

  1. Description of any individual that agrees with the user.
  2. Arbitrary label usually indicating the user as being self-righteous and correct.
Antonym: Irrational

Logic: (n);

  1. A series of statements making sense to the user.
  2. A series of statements making sense only to the user.
Usage: "Logic shows that..."

Evidence: (n);

  1. Some kind of visual, audible or written clue to back up an opinion. Usually unobtainable and unrealistic when requested. Usually insufficient when presented.
  2. Something the you need to show but I don't.

10th January 2012[edit]

The Adventures of Mo... Part 92,554

Another day, another group gets its arse hauled for depicting Mohammad. In this case it seems to be a case of immovable objects and unstoppable forces when it comes to the rights and wrongs of the action: we can't say that the image is offensive, and yet we can't accuse Muslims for being violent and forceful (indeed, it appears to be a Union level action, rather than a private complaint by a Muslim grouo). Of course, putting up such an image this atheist society knows exactly what they're doing - it's going to provoke, it might not be Mohammad fucking a pig but they know exactly what they're doing. You can't argue unintended consequences here.

But this is all a bit of a red herring to the point.

What we're looking at is one arbitrary group enforcing their standards of behaviour, morality, ethics and commandments (again, arbitrary) upon another group. That's what it is, if we stop pussy-footing around for a moment with talking about offence and not being a dick. This is wrong.

We believe in freedom, not just of speech and expression but in general existence. We deem this to be good because it promotes the widest amount of happiness and property for all. From this we derive that people encroaching on the freedom of others to be bad - and simply put the mere fact that a mere image can be torn down for an arbitrary reason is an encroachment on such freedom. In an ideal world, such a request would never be made and it would be a non-issue. That's indeed what I would like to see; for it to be a non-issue and for Muslims to grow up and get over themselves with regard to this. I'm not a Muslim, I'm not going to hold myself to arbitrary standards set forth by Islam any more than I'll hold myself to standards put forward by Scientology.

Effectively, we'll stop depicting Mohammad when the pressure not to is lifted.

9th January 2012[edit]

9 Days New Year's resolution not to swear on Facebook lasted 9 days. Not bad, I think.

6th January 2012[edit]

Shamelessly stolen
  • Type your username: Armondikov
  • Type your username with your elbows: sadrmoiknsdikpov
  • Type your username with your nose: srmoknbn sdikv
  • Now slam your head against the keyboard: In the beginning God created the heavens and the earth. Now the earth was formless and empty, darkness was over the surface of the deep, and the Spirit of God was hovering over the waters. And God said, “Let there be light,” and there was light. God saw that the light was good, and he separated the light from the darkness....

2nd January 2012[edit]

Belief and Unbelief. (II)

So, if lacking a belief in something an having a belief that the something doesn't exist effectively say exactly the same thing, what is the difference, and why would we get precious about the distinction?

One potential distinction is that the lack of belief is trying to cover the trappings of implicit atheism. This is where one doesn't believe in God simply because they either haven't considered the question or aren't aware of what "God" actually is. This is the root of the BHA's "Don't Label Me" campaign in reference to young children - they're too young to have proper awareness, and are implied to be atheists because they simply lack the concept much in the same way that they'd implicitly not be fans of Tetris, or Meat Loaf or implicitly not Marxists, or Objectivists. We implicitly don't do or don't like a near infinite number of things, however, the implicit nature of the lack of belief disappears as soon as you become aware that you've drawn a box around what it is you don't believe in. Once you become aware of it, you're no longer not believing by implication (because you're simply unaware of it) but because you explicitly don't do it. As a result, anyone with sufficient awareness to say "I don't follow a religion", or even "I know what a religion is" is no longer merely implicit in their belief. It's like once we've drawn that box around the god-shaped cluster of things, we're given the choice whether to label it as "real" or "imaginary", and so our opinion is explicit.

But the idea of implicit atheism still holds a strong draw for atheists for many reasons. Think of it as "I simply don't acknowledge religion as important" and you can see that implicit atheism would be a pretty powerful Fuck You to religion. You would be telling religion that it literally means nothing to you - unfortunately it would be a lie if you were aware of what religion is. In another sense, it's also an enticing idea because it's not a "faith position". Implicit atheism is like the default setting, it requires no assertions, no knowledge, no conscious decision to make a solid statement about what you think. Again, this plays nicely to the idea most atheists have that religion is a matter of faith, and therefore completely irrational - otherwise why would we have to say there's "probably" no God, or hide behind terms like "agnostic atheist". It's just a shame that how you practice your belief can't reflect this probability statement; you either act as if there's a god of a religion there or you don't, you can't assign a probability to your actions.

So when we talk about a lack of belief, we're really trying to get across this idea of perhaps being more open-minded, and less in a "faith position". Indeed, if pushed for why they'd prefer to say "lack of belief", most atheist are honest enough that this is their intention, although probably not honest enough to admit that it's just bullshitting.