Prediction frameworks

For years, I’ve frequently shared an idea about belief and the human experience. It often comes up when discussing the many paths to "truth" that societies have developed. I’ve tentatively called this idea “Prediction Frameworks.” This is the first time I’ve written it down, so the name may evolve.

Most of our mental energy is spent making predictions. Yet the tools we develop to make those predictions often become points of focus themselves, rather than being judged by the quality of the predictions they were designed to produce.

A classic example is the Native American rain dance. Today, the idea of dancing to “make it rain” seems laughable. Most discussions would devolve into arguments about why rain dances don’t work. But these discussions miss the core reason the ritual existed: depending on geography, it actually rained roughly 20% of the time the dance was performed.

The concept of causation versus correlation doesn’t affect that outcome. The success rate was sufficient for communities to adopt the ritual, predicting that performing it would produce a desirable result.

We make predictions constantly, often without realizing it. Consider trust: Can I trust my neighbor, friend, or colleague? If I leave a key with a neighbor, do I predict they won’t steal it? These examples touch the core problem that likely motivated early humans to develop systems of belief: we gain confidence in our predictions when we share assumptions about how people should - or shouldn’t - behave.

This suggests that concepts like respect, honesty, and kindness are inventions rather than naturally occurring traits. These inventions likely took hundreds of thousands of years to develop, across multiple species, to reduce stress and uncertainty in the environments and communities where we live and reproduce.

Religions are one form of prediction framework. Each arose over time to address the concerns of a specific population in a specific environment. A parent’s confidence, for instance, increases when they can predict how someone’s son will behave with their daughter—and vice versa. The set of predictions a person can make by adopting a religion borders infinity.

From this perspective, questions like “Is Christianity true?” or “Does God exist?” lose much of their weight. They overlook the fundamental reason these beliefs exist: to improve predictive confidence. More precise questions would be, “What predictions have troubled humans enough to develop and maintain Christianity?” and “What predictions does the concept of God allow humans to make with greater confidence?”

Science fits into this framework in a similar way, though popular culture often casts it as the antithesis or enemy of religion. Science is a process of evidence-based observation and experimentation. It is frequently confused with “truth,” leading to debates that mirror religious wars. In reality, the goal of science is predictive accuracy. We accept gravity, for instance, not because it is “true,” but because 100% of the time, when we jump, the Earth pulls us back down.

Even scientific predictions are bounded by the domain in which they are proven. This is crucial: predictive success is only meaningful in contexts where the concept has been validated.

Scientific discoveries are therefore best understood as tools localized to the problems they solve, not as absolute truths. Take astronomy: over millennia, humans developed it to predict location, direction, and season. Observers evolved from mapping positions on a flat Earth, to the sun orbiting the Earth, to understanding the Earth orbits the sun. Across this entire history, the “scientific facts” of the time were sufficient to make the predictions required. Merchant ships could navigate from Istanbul to Alexandria by reading the night sky, regardless of whether the Earth was flat, round, or at the center of the universe.

This reinforces a key principle: any framework is only as good as the predictions it enables, given the information available. Not all frameworks are equal. Science is not always the best tool, just as Islam - or any religion - is not always the best tool. In a narrow case, if you must predict someone’s aggressiveness and all you know is their birthday, astrology might actually be the most useful framework.

At their core, debates about whether “astrology is real,” “Judaism is true,” or “science is fact” are often distractions. These frameworks are all attempts to improve predictive confidence in human life. They are tool-kits for approaching life’s challenges, which are inherently context- and culture-dependent.