Tuesday, April 26, 2011

My least favorite question

It's a little uncomfortable for a free thinking skeptic to realize that there are questions he hates. That seems like a terrible attitude. Questions ought to be my bread and butter. But then again, I also hate the phrase "bread and butter," so where does that leave me?

Maybe it has to do with my interest in language and writing. In literature, we scour words for meaning and significance. Meanings piggyback on other meanings and we often pick up messages that aren't explicitly stated in what we read. Sometimes those meanings really aren't there at all, and we end up reacting to literary pareidolia. Still, even those erroneous reactions are important, because the reactions are real even if the message isn't. This is why I hate "Why?"

"Why?" is such a fundamental question. In many ways, it is the prototypical Big Question. Unfortunately, this tiny word hides a huge number of false, or at least unnecessary, assumptions, and to answer it may require the acceptance of falsehoods.


"Why" is often a stand-in for a slightly more complicated question. "Why won't my dog eat his food?" is my question of the day, but what I'd really like to know is more detailed. I'd like to know if his medication is affecting his appetite, or if he's in too much pain from his surgery despite the medication. Maybe he's just nervous because my schedule has changed in the last few days and he isn't sure what's happening. Maybe when I chose to move his food bowl over by his bed so he wouldn't have to walk so far, that threw him off. In this case, "Why" isn't that great an offense, because I'm still talking about a living mammal with a working brain, who can make decisions in some way. Dante is an agent who can have reasons for his actions. When I ask "Why" about something more abstract, the question loses the usefulness of an agent, but I'm still begging the question by implying agency.

Say I ask a child's question about the world, "Why does the sun shine?"* It's no wonder we have this stereotype of the repetitive "Why" from children. Any sensible, useful answer will give a description of nuclear physics and gravity, and it should be little surprise that this answer will often be met with, "Why does it work that way?" This is a perfectly valid question, in that theoretical physicists pursue the underlying laws of nature by asking it, but science escapes from the question of "Why" by looking at "How" and "In what way?"

"Why" gives a Mindy-like infinite regress when we examine nature, because there is no agent. There is no dog deciding not to eat for a reason, and physics are not a rule book. "Why" leads us to think that the laws of physics are the way they are for a reason. Why is there something rather than nothing? That's a nonsense question, because you can't even show that nothing is an option. Why are the laws of physics the way they are? They are a description of processes that we have made through observation and testing, not a constitution of reality we all have to obey. "Why" holds the hidden assumption of agency, and even when there is an agent, it's probably not the question you want answered.

It's not just the interpretation of the grand scale of reality where "Why" falls apart. I see it every day in retail. Here are two recent conversations, one with one of my employees, one with a customer.

Scenario 1 - This took place over the radio. "Alice" was at the register while I was in the office doing paperwork.

Alice: Can I have a manager to the front? I have a rewards card that doesn't come up by email, but when I look it up by phone number, I get a card number different from the card the customer has. That card has the same email I tried to look up. Why is the card doing that?


Me: That's not something we can fix in the store. That has to be a database error. The customer probably tried to merge two accounts online and screwed something up.


Alice: Why would there be two cards with the same email?


Me: Because computers aren't perfect, and computer users are even less perfect. I have no idea. Just give the customer the 800 number for customer service.


Alice: Can you come up to the register and explain to the customer why this is happening?


Me: No. I have no idea why. Because something screwed up in the card database. I wasn't there when it was screwed up, so I have no possible way of answering that question. Besides, the answer won't fix anything. Just give them the phone number, because if I come up there, all I'll do is give them the phone number. I can't fix it here.

Scenario 2 - This time, I was on register. A customer was buying a pack of Magic: the Gathering cards.

Customer: Why are your cards so expensive?


Kyle: I've never seen them any cheaper than this. Where did you see a lower price?


Customer: I dunno, I just wonder why you charge so much for them.


Kyle: Um...because that's the price we set on them. I don't know what you really want to know here. The price is set by the corporate office based on how much they cost us.


Customer: I just don't like paying so much.

In neither of those conversations was "Why" the question they wanted answered. The only relevant information in the first one was "How can I fix this?" There is no why. The reason that something was broken was that someone or something broke it. Was it done on purpose? Highly doubtful, since that would benefit no one and hurt all parties involved. If not on purpose, it was either a human accident or a computing accident, both of which are meaningless as "why" answers. Without deliberate agency, the ultimate answer is "Because." Useless!

In the second conversation, the customer didn't really even want an answer to a question. He just wanted to spend less money on his game, but knew he couldn't. He was even using a coupon to make the cards cheaper than the list price, so there was no real reason for complaint, except that he probably spends a lot of money on those cards (I remember the days...).

"Why" is just a surrogate for other thoughts, other questions, other complaints. Unless you're talking about intelligent agency with deliberate intent, it is not only meaningless, it's counterproductive. Think before you ask why. It probably won't give you what you're looking for. If you don't know what you're looking for, then you should definitely think before you ask anything at all.




*Spot my hidden assumption in that sentence for bonus points.

Tuesday, April 19, 2011

Bad Universe, episode 3


There is a conceit in movies that always irritates me: the audience surrogate character. His whole purpose is to comment on how cool or how scary the events are as they happen. The best (worst?) example I can think of is Link. In the Matrix Reloaded, he's the operator on the computers during the big freeway chase, and his whole job is to look scared so the audience knows This Is Serious Fight, and to cheer when the heroes pull out an impossible win, presumably so that the audience knows that the writers would like you to feel thrilled now please thanks. That kind of writing is cheap, like sucking on a greasy coin or cribbing lines from Hannibal.

As a writing technique, it's awful, but being an audience surrogate can be a truly great thing in one place: education. I just watched Phil Plait's Bad Universe on Discovery, the last of three episodes testing the waters for audience interest. Discovery Channel, if you're listening, the waters are just fine. In a scene about solar coronal events, the Bad Astronomer shows a demonstration of chain reactions that I've seen before, where ping pong balls are set on mousetraps to make a violent cascade of springing traps and balls, then he shouts with glee, “That's so cool!” as if the audience needs to be told.

Here's why Phil Plait is different from Link: the audience for science really does need to be told.

We need to know how cool science is, and how to recognize what is science and what is nonsense. In education, it isn't bad writing to have an audience surrogate because it's real. He is that excited, and he wants us to be that excited too. Having someone clapping on screen and laughing for us allows us to clap along and agree. The intro to the show talks about destruction and debunking, yet it could not have a more positive message, and telling the audience when to cheer really helps.

Think about this show. It's all about what's dangerous in the universe, and how it can kill us, with little excursions on how some of these things can be prevented. The preventative portion of the show is ostensibly the positive bit, except that the answer often seems to be that either it's impossible to prevent the problem or that we've just never invested in the solution. This show about inevitable destruction and extinction is played with such undeniable fun and excitement. The best news is often, “We don't have a good answer for that yet.” To present galactic-scale problems with unknown or incomplete solutions in an exciting and excited way is certain to draw new minds to the world of science.

Bad Universe sits somewhere between Mythbusters and Cosmos, for what appear to be soundly strategic reasons. Showing passion for science like the dignified Carl Sagan really ought to be enough, in a perfect world, but let's be real here. To draw people in, you need to blow something up at least once per episode. I want to complain that the papier-maché supernova wasn't really that accurate a model, in the same way that I sometimes want to complain that a Mythbusters stunt didn't use experimental controls that made sense, but I just can't do it. I can't.

There are enough CG effects in the episode that clearly they could have gotten away with a CG supernova to demonstrate a dying star's destructive power, but they chose fun over accuracy. A computer model would have shown the audience exactly how it all works, but blowing up a paper ball with dynamite is just better. I do computer animation myself, and I can't clap for a CG supernova (much as I'd like to animate one just for its own kind of personal fun). I can clap along and laugh for that stupid paper bomb though.

Discovery Channel, please give this man a camera and more dynamite, immediately.

“What sort of evidence would you accept?”


There are a great number of things I don't believe. Discussions about these with people who do believe tend toward a familiar impasse. They can't understand my rejection of their beliefs, and I can't understand their definition of evidence. Eventually it comes to this question. This is one of a few questions I hate, the worst of which is “Why?” but more on that another day.

Whenever someone asks me “What sort of evidence would you accept?” I find myself giving a clumsily qualified answer in order not to sound unreasonable. The things I disbelieve lack evidence that meets any useful definition of the word evidence. That's the primary reason that I disbelieve. In most cases, there really is evidence, and it all says no. Trying to explain that dimension to this problem leaves the believer with the impression that I'm dismissing their evidence unfairly. I asked for evidence, and they brought me some! How can I say they don't have any evidence? It's right there! OK, true believers, to make sense of this failure to communicate, I take you to a court room in session.

Judge: Is the defense prepared to enter a plea?

Defense Attorney: I'm sorry Your Honor, but I don't know what's going on here. We don't have a client.

Judge: What are you talking about? State vs. Credence Woo, I've got the papers right here. Where's the defendant?

Defense Attorney: Seriously, Your Honor, we've been trying to figure this out for a while now. There doesn't seem to be anyone named Credence Woo. Honestly, we're not even sure how the DA managed to bring this case in the first place.

Prosecuting Attorney: That's ridiculous! We have sworn affidavits that he murdered three people in broad daylight.

Defense Attorney: We've looked through those as well. None of them agree about the facts of the case, and anyway, they aren't sworn affidavits. They're anonymous descriptions. One of them seems to be mostly plagiarized off another, with a few contradictory additions.

Judge: How are we even having a trial if the defendant doesn't exist?

Prosecuting Attorney: He does! There's plenty of other evidence. Haven't you seen the photographs?

Defense Attorney: Those aren't photographs. They're drawings. And they don't even look particularly similar to one another. Oh, except that one, but that's a photo of Brad Pitt that's been Photoshopped quite badly. Anyway, that's not even the real problem. We can't even confirm that this murder actually took place at all. We do know that two of the victims named in the case did exist and are dead, but as far as we can discover, they died several years apart in different states, not at the mall downtown two months ago. On top of that, the prosecution's proposed method of murder is physically impossible.

Prosecuting Attorney: No it's not. Look, if you understand quantum mechanics, you'll see that--

Judge: OK, I'm just going to stop you right there. If I don't let you finish that sentence, it will remain in a state of superposition, so there's a 50% chance it's not bullshit. It can be Schroedinger's argument, both bullshit and not bullshit until an observer collapses it into nonsense. Now someone please tell me what we're doing here if there's no defendant and no confirmed murder.

Defense Attorney: Actually we did find evidence of a murder at the mall on that day, Your Honor. Well, it was near the mall, and took place very early in the morning. There were four victims, not three, none of them the ones named in this case, and it looks like it was done by a small group, not a single person. Tragic, but it looks like an interesting case, and no one is currently investigating it except us.

Prosecuting Attorney: No, stop dismissing me! You're just not looking at the evidence. You need to keep an open mind and really dig into the evidence I've got here.

Judge: I'm getting the impression you don't understand what evidence is. Are you even familiar with standards of evidence?

Prosecuting Attorney: Fine then. What sort of evidence would you accept?

Judge: At this point, you've demonstrated that you don't know how evidence works. There is zero real evidence in favor of your case, and the defense has uncovered a different case that you're not working on. That seems irresponsibly close to obstruction of justice if you're letting someone else get away with murder in favor of this nonsense. I'm going to throw out this whole case as a mistrial. You would not only have to bring good evidence in favor of your case, but now you would have to find evidence that contradicts confirmed facts that we've already established. Case dismissed.


If it helps you to interpret who is analogically who here, I'm the Judge. I'll let you figure out the rest.

Tuesday, April 5, 2011

Video Series

I've taken down an old post that previewed a video I was working on. This is not to say that I'm no longer working on it, but the project has grown a little more ambitious. I'm now making it a series of short videos tentatively entitled "Not Enough Science." They will cover science basics that are frequently misunderstood or poorly taught. Skeptically minded folks will recognize them as the kind of things that get thrown out as spurious arguments by the badly uninformed. Episode one will be a slightly altered version of my original idea, and will cover thermodynamics. Until the basic animation assets are finished, expect infrequent posts.

However, I am looking for anyone who would like to fact-check my science. It would be a shame for me to cause the very problem I'm hoping to solve, wouldn't it?