Tuesday, November 19, 2013

Being Uncertainly Certain

Job had a pretty posh life, if the accounts are to be believed. He had lots of stuff (mostly the kind of stuff that eats grass and poops, but it made him happy, so don’t worry about it). He had lots of kids. He had everything a guy in the Ancient Near East could want, including a loving wife and a cadre of faithful friends.


Then, bad stuff started happening to him. He lost his stuff. He lost his kids. The wife and friends he got to keep, but they stopped being so nice to be around. Then things got bad. Job got some sort of gross disease that made him break out in boils that hurt and itched. He felt like crap and no one brought him chicken soup and ice cream to make him feel better.


Instead, Job’s friends and wife brought him their certainty. They threw it into Job’s face and used it as a club to beat him while he was already down. Good thing he got to keep his friends and wife in the deal.


But what is certainty? Why do we want it so much, that is until we’re in an uncertain situation and then we don’t want certainty?


If you’ve ever been to a funeral you’ve heard the unwanted certainty come out. “I’m sure God had a plan for all this.” Or maybe you know someone who lost a job and heard, “Whenever God closes a door he opens a window.” Or someone miscarries and they hear, “God needed another angel in heaven.”


That kind of certainty is the last thing that suffering people want to hear. It pours salt into wounds, ignores the pain, and makes a mockery of the person. Not good.


How is it that crave that certainty though? We do. We want to know, for a fact, that things will turn out in a certain way. We want a recipe for success. We want a formula for faith. We want to know, to know, that everything will work out.


The feeling of certainty is, in part, a very useful thing to have around. If nothing else, it lets us get off the couch and get something done. Even flawed certainty is more useful than complete uncertainty.


Think about plopping down to watch something on Netflix. Your queue will be accompanied by some suggestions for movies you might want to watch. Sometimes Netflix is spot on with the recommendations -- but usually it’s pretty far off base. The only way for Netflix -- or any other recommendation service -- to get better at recommending things to you is to fail at it. A lot.


Online recommendations use machine learning and algorithms to figure out what it is you want. You don’t need to understand exactly how it all works, save to know that the algorithms are nothing without input and the only way to determine the output is to run the algorithm.


So, if you get onto Netflix for the first time ever and ask it or a recommendation it won’t be able to give you any because it has no input. Not only that, but you can’t call up someone at Netflix headquarters and ask them to figure out what the algorithm would recommend based on movies you’d like to watch. They don’t know. The only way to know is to enter the information into the software and let it crunch through everything, compare it to what other people think, and come back with a result.


So, what does this have to do with Job? I’m glad you asked. Job’s wife and friends were certain that the reason Job was suffering was due to a sin he’d committed. They knew it. Without a doubt. They told Job to just give up, admit his sin, and God would stop smiting him. Not really the most compassionate thing to say to a guy scraping his sores with broken pottery, but there they were.


They were certain of their conclusions based on the input they’d received and the way their brains processed that information.


So, what happens when you host a party and friends come over with kids who sit and watch your Netflix all night? Exactly! You get recommendations for “My Little Pony” and “Rainbow Bright” the next time you sit down to watch. The input was bad so Netflix gave you bad results.


Job’s wife and friends bought into the idea that doing the right things would force God to bless and doing bad things would force God to punish. They had bad input and drew the logical conclusion from that input: Job must be a sinner because of all the bad stuff that happened to him.


Job rejected that conclusion. He knew that he hadn’t been doing all the bad stuff his wife and friends thought he’d been doing. The current input was at odds with his conclusions.


Netflix might recommend that I watch some “My Little Pony,” but that doesn’t mean it’s right. It may have taken into account what’s happened in the past, but it can only determine what happened, not why. Our brains often work in the same way. They observe what’s happened in the past and draw conclusions about what is happening now and what will happen in the future. Most of the time those conclusions are great and they help us to live normal, productive lives. But in extreme circumstances, our input is limited and our conclusions are suspect.


Think about two people walking down the street. They both hear a loud bang and see a flash out of the corner of their eyes. One drops to the ground and the other keeps walking without flinching. They are both certain about their action, but for vastly different reasons.


One is an event photographer, used to random flashes of light and loud noises. They are certain that the flash and bang is nothing but another photographer tripping over a folding metal chair. The other person just finished a tour of duty finding and disabling improvised explosive devices. They are certain that the flash and bang are an explosion and people are about to die.


Both of them are certain based on their past experience. But neither of them has control over their certainty or their reaction in the moment. Just like the Netflix algorithm, the only way we can really determine our certainty -- or uncertainty -- is through experience. We have to live it, and sometimes get it wrong, to test our conclusions.


Job’s wife and friends were completely certain that Job was a bad person, but they were completely wrong.

Discussion:
What’s the worst online recommendation that you’ve gotten? How do you train computers to come up with better recommendations for you?


Have you been on the giving or receiving end of the general platitudes about purpose in suffering? What motivated you to say them? How did it feel to hear them?


Where do you think the two people who saw the flash are? What would it mean for them to be on a battlefield? What would it mean for them to be at a trade show?

Have you ever been certain about something and ended up being wrong? How did it feel? What did you do about it?

No comments:

Post a Comment