How Do You Know Why Anything Happens?
A question in a cold classroom
In 1250, a student in Paris might have heard this: “Stars twinkle. Planets do not. Why?” The teacher was not looking for a guess. He wanted a proof — the kind that would let anyone, anywhere, see the reason, not just the fact.
Aristotle (384–322 BCE) had the answer. Stars are so far away that their light trembles when it reaches us. But here is the tricky part: the twinkling does not cause the distance, the distance causes the twinkling. If you argue, “Stars are far away because they twinkle,” you have the logic backward. You know that they are far away, but you do not yet know why.
That distinction matters. Aristotle believed the highest kind of knowledge — what he called scientific knowledge — is not about collecting facts. It is about seeing the causes that make those facts unavoidable. For centuries, the book where he laid this out, the Posterior Analytics, baffled almost everyone who tried to read it. When people finally unlocked it, they started a fight that lasted 400 years.
The proof machine Aristotle imagined
Aristotle designed a kind of mental machine for science. It was built from syllogisms — arguments with two premises and a conclusion, like:
- All humans are mortal.
- Socrates is a human.
- Therefore, Socrates is mortal.
But a real scientific proof is not just any old syllogism. To give genuine understanding, the middle term — the idea that links the premises — must reveal a cause. If your middle term just restates the fact in different words, you have not done the job. A good proof shows why the predicate must belong to the subject.
Aristotle thought a perfect science would start from first principles: truths that are necessary, universal, and cannot be proved by anything more basic. From them, all the other true claims in that science would flow like water down a hillside. Mathematics was his model. When you prove a triangle’s angles equal two right angles, you do not rely on slippery observation — you see it once and for all.
But here is the catch. The world we live in is full of things that do not behave like triangles. Fire burns, dogs bark, angry people shout — but only for the most part. What about eclipses, which happen rarely but predictably? What about wounds, which heal more slowly when they are circular? Real nature seemed too messy for Aristotle’s neat syllogisms. The medieval thinkers who took up the challenge had to face that mess head-on.
The bishop who turned a book into a mirror
Robert Grosseteste (1175–1253) was one of the first to truly use Aristotle’s difficult text. He wrote a commentary around 1230 that treated the Posterior Analytics as a demonstration — a proof about proofs. He claimed Aristotle was building a science of science itself, starting with definitions and systematically deducing all the properties a perfect proof must have.
Grosseteste did something else clever. Many Christian thinkers of his time were suspicious of Aristotle, because his approach seemed to leave God out of knowledge. Grosseteste stitched the two together. He suggested that before the Fall, humans could have seen the true forms of things directly in God’s mind. Since we lost that ability, he said, we now need demonstration — the slow, step-by-step path of reason and evidence — to piece the world back together.
His most practical contribution was a method for discovering causal laws. First, you might notice that a certain herb, scammony, seems to purge red bile. Your senses form an aestimatio — a hunch about a causal connection. To test it, you remove every other possible cause and see if the effect repeats. If it does, your reason can conclude that this is a real causal power of the herb. Grosseteste thought this was how we could build up genuine natural science without needing a direct vision of God’s ideas.
The definition fight that split a tradition
After Grosseteste, the big question was: what exactly makes the highest kind of demonstration — the demonstratio potissima — work? The fight centred on the middle term, the engine of the syllogism.
Albert the Great (ca. 1200–1280) argued that the middle term should be the causal definition of the attribute — the property you are trying to prove. Suppose you want to explain thunder. Thunder is a noise in a cloud. The causal definition goes deeper: thunder is a noise-in-a-cloud-produced-by-the-extinction-of-fire. That definition includes reference to an external cause (fire). It gives you real new information, not just a label. For Albert, if the middle term merely repeated the subject’s essence, the proof would be empty — a fancy way of saying the same thing twice.
His most famous student, Thomas Aquinas (1225–1274), disagreed. Thomas thought the middle term in the highest proof should be the real definition of the subject. For a human being, that would be “a rational animal.” Thomas believed the essence of a thing produces its necessary properties — its attributes — by a kind of natural outflow. Once you grasp what a thing really is, you can see, by intellectual vision alone, what its basic powers must be. You would not need to look outward at external causes.
Thomas’s view sounds elegant. Trouble is, it can look circular. If the middle term is just the subject’s definition, and the conclusion links that subject to an attribute, have we discovered anything new? Giles of Rome, a follower of Albert, pressed this hard. He said a proof whose middle term merely restates the subject’s essence begs the question. The only way out, he argued, was to use a middle term that obliquely refers to things outside the thing itself — as the definition of an attribute does.
The debate rumbled on for decades. It might sound like a dusty technicality, but it mattered deeply: it was really a debate about whether human minds can learn anything genuinely surprising about the natures of things, or whether all real discovery has to point outward, to causes and effects we bump into in the world.
Ockham pulls the roof down
William of Ockham (ca. 1287–1347) was not satisfied with either side. He accepted that the middle term should be the subject’s definition — but only if that definition referred to real, distinct parts within the subject. For Ockham, that meant real definitions only worked as middle terms in mathematics. A triangle is defined as something composed of lines, and those lines are real parts. From that, you can prove properties that actually occur.
In the natural world, Ockham thought the situation was much less tidy. Take fire. You might define a concrete instance of fire as “a hot thing” and argue that hot things are capable of heating. But this only gives you a capacity — not an actual event. A real natural demonstration, he thought, would have to use temporal conditions. You could prove, “When there is no opaque medium, the moon will be illuminated by the sun,” but that is a hypothetical claim, not a categorical one like “All triangles have 180 degrees.” For Ockham, the highest sort of demonstration was simply not available for most of what we care about in nature.
He also rejected any proof based on final causes — purposes or goals — because he did not think purposes had causal power of their own. Only efficient causes (pushes and pulls, heatings and coolings) could do the work. With that move, Ockham cut away the grand vision Grosseteste had built and left natural science in a humbler, more empirical place. Observation and experience, not intellectual vision, would have to guide us.
So what is a real proof worth today?
When you learn science today, you do not squeeze it into syllogisms. But the question Aristotle raised is still alive. Does a good explanation just predict what will happen, or does it show you why it must happen — the inner mechanism, the hidden cause?
The medieval debate sharpened this. Grosseteste wanted to test causal laws by experiment. Aquinas trusted in the power of the intellect to see essences. Albert demanded that definitions point outward to causes. Ockham warned that perfect proof might belong only to mathematics and logic, while the natural world gives us something less — probable, conditional, always open to revision.
Next time you learn why a circuit works, or why a disease spreads, or why a friend acts the way she does, you are walking a path these thinkers laid. Are you satisfied with a pattern, or do you push for the cause beneath the pattern? The fight over the highest sort of proof is not just a medieval fossil — it is your fight, too.
Think about it
- If a friend says, “I know my dog is happy because she wags her tail,” is that a scientific explanation in Aristotle’s sense? What cause is missing?
- Can you ever be absolutely certain why another person acts the way they do, or only certain that you have seen that behaviour before?
- Imagine you learn that a certain plant always cures a headache. Is that enough, or do you need to know how it works before you trust it? What if Grosseteste and Ockham disagreed?
