Most ideas can only be post-rationalized

Ideas can be categorized into one of two buckets:

• Ideas that can be pre-rationalized. This is when you have the elements to determine whether the idea will be successful before trying it.
• Ideas that can only be post-rationalized. This is when you can only determine whether the idea is successful after trying it.

While we’d all love to live in a world where most ideas can be pre-rationalized with data and logic, this is not only possible, because:

• We live in a chaotic system, which makes it extremely hard to draw a straight line from our actions to their results.
• All the data we rely on comes from the past, not the future. A “black swan” event can and will throw off all the odds.

This means that most ideas cannot be pre-rationalized; they can only be post-rationalized. We try them, and if they are successful, we construct a new theory of success around them, at which point others will follow in our footsteps. This is done best when you have strong experimentation guidelines.

This is often what ends up happening anyway, even when we think we are pre-rationalizing: our attempts at pre-rationalization often work only because the act of predicting the future is creating the conditions for the future to happen.

When done well, this is the only way for humanity to move forward and create the future we want to see. However, we should also keep in mind that post-rationalization can lead to survivorship bias.

References

• The Sarumans and the Radagasts
• Alchemy
• Foresight as Activism: Researching vs. Making Culture