In this case, knives, statistics, and files.

I remember very distinctly the first time I was allowed a pocket knife.

My older brothers had been whittling for years, and knowing my parents it's likely that I

had to come of some arbitrary age and then pass some sort of pocket knife test before

I was allowed to join them in knife ownership.

So I was very motivated to learn the rules of proper pocket knife usage, but I remember

having some internal angst over one rule: always cut away from yourself.

I understood the reasoning behind that rule.

Your self is towards yourself, so if you cut in that direction there's a risk of cutting

yourself.

I was familiar with many safety rules of that form: "do a thing always to avoid something

bad that might happen sometimes."

Humans have been using knives for millions of years.

Statistics, on the other hand, is only a few hundred years old, maybe a few thousand if

you count the most basic concepts of probability, and risk analysis is very recent.

So for a long time, doing something always to avoid a bad sometimes was a good heuristic.

In fact, humans will go so far out of their way to do things always to avoid sometimes

that we have a long history of superstitions and obsessive habits.

We lived in a dangerous world, and before the discovery of the mathematics that would

help us asses risk, that was the best we had.

So we learn rules, repeat them, and are expected to follow them, regardless of whether or not

we know that the statistics justify the rule.

A file, on the other hand, is a tool that only works when you file away from yourself,

because of the way it's designed, with tiny saw-like teeth.

"Always cut away from yourself" isn't just a rule you should follow, it's a rule you

have to follow, or the file doesn't work, and you hurt not yourself, but the file, which

is somehow more convincing.

Whether it's better to design our tools with these rules already inside them, I don't know,

but it's something I think about a lot.

My first knife was a small red Swiss Army knife, and I could not help but think that

it was worth the risk of maybe cutting myself if only I could whittle in whichever direction

I pleased.

Each individual cut was so important to my work, and each individual cut had such low

risk of going wrong.

And what did I care, if I cut my finger a little with my tiny pocket knife in service

of my art?

Cuts heal, but my work, my beautiful clumsily-whittled pointy sticks are forever!

A risk/reward or cost/benefit calculation looks at the relationship between the chance

of something bad happening and the chance of something good happening, like the chance

of cutting yourself vs the benefit of knife direction freedom.

The tricky part is that in order to make this calculation you need to have the risk stats

and be able to put values on those good and bad things.

To me, a cut is just a cut, like any other cut.

But to my parents, a cut I get from a pocket knife they gave to me is fundamentally different

than a cut I get falling on the sidewalk.

There's extra responsibility and social cost to them, while I'm the only one who gets the

benefit of multidirectional knife freedom, so the real risk is that they'd take my

knife away.

It's really hard to reconcile differences in how much we value the things we value.

When parents value their kids' safety more than their kids do, or when we value the things

and people we know over those we don't, or when we value our now more than people a century

ago valued their future.

Still, I wanted my knife, and so I went along with this rule I knew my parents cared about,

Always cut away from yourself, at least when they were watching.

But unlike tests in school, where the "Right answers" you repeat might take 10 years to

become relevant to your actual life, the rule "always cut away from yourself" became practical

almost immediately when I cut towards myself and cut my thumb.

It was a very tiny cut, but suddenly it seemed like always cutting away from myself was just

a really easy rule to follow so why not just do it.

The freedom of being able to cut in whatever direction I want, the freedom of riding a

bike without a helmet or of not wearing a seat belt, can seem so valuable when you don't

have any experience with the down sides and doesn't know anyone who has.

It's not uncommon to hear stories of people who would always ignore a safety rule until

something went wrong, and then suddenly the equation changes.

Getting a concussion from not wearing a helmet doesn't change the overall risk, but it does

make you more likely to wear a helmet in the future.

Not just because you perceive the risk as being greater, but because suddenly it seems

stupid to value some mysterious intangible cool-factor that you used to think magically

floated around your helmet-less head.

We have the mathematical tools to deal with these kind of things when we look at statistics

in aggregate and translate to some sort of monetary cost associated with insurance or

something.

But to individuals these variables are much more variable.

It gets so much easier when we can just get rid of one of these fuzzy variables, change

it from a probability to a simple 0 or 1, always or never, nothing or everything.

It doesn't matter the risk of cutting towards yourself because cutting away costs you nothing.

And when we or someone we love do something important that comes with a very high risk,

the easiest way to justify those actions is by telling ourselves it's worth everything.

The math of this equation pushes our different values to extremes.

And it goes beyond physical harm.

How many people have stubbornly used some non-PC word or phrase out of the allure of

Freedom of Speech and Expression until they find out that someone they care about has

been hurt by it, and suddenly it just seems really easy to not say that particular word.

Or the opposite case, where someone stays attached to the simplified rule of “never

use that word” even in cases where the benefit outweighs the harm, just as a good whittler

does sometimes cut towards themselves for the sake of their art.

Of course, there's infinite complexities missing from these equations, such as that small mistakes

are worth making as long as we learn from them, which makes them not mistakes at all.

So as much as I like the aesthetic of the file, I wouldn't want the world to only have

single-purpose tools with built-in rules that prevent them from telling us anything about

the world around them.