Aquinas and Aristotle were some of the premiere thinkers of their respective times, and both thought that human reason was an important potential path to knowledge. This is shared with the contemporary sensibility – we tend to think that reason can help us to create a better picture of the world.
What was the big mistake of Aquinas and Aristotle? Not realizing the frailty of human reason. To work, it helps that human reason be tested, and tested, and tested again. Something might seem correct and clearly so to our faculty of reasoning. We might look at every step in the chain of reasoning and it might seem obviously correct. We might be some of the best human reasoners in the history of the world, such as Aquinas or Aristotle. Yet, we still might end up with an incorrect conclusion.
So, three ways to buttress human reasoning are first, as mentioned, to figure out how to test it. I.e., ‘how can I test this to see if it’s right, beyond just looking at the theory?’ One simple way to do this that is often overlooked is to just check in the world. For example, you can reason about what is or is not in the fridge, but the best way to check a theory so derived is to open up the fridge and look.
A second way to buttress human reasoning is to look for convergent lines of evidence. Consider a line of evidence that leads you to a 90% estimate of its veracity, so 10% chance of being incorrect. If you have two lines, each at 90%, then the chance of being incorrect is 1% (10% * 10%, making certain simplifying assumptions).
A third way is to look for different ways of interpreting the evidence. So, ‘Could this evidence be put together in a different way?’ For example, this is often useful in figuring out causal relationships, where cause and effect can be backwards from what is posited, say. Another useful question is ‘For this to be put together in this other way, what else would have to change?’ Consider the considerations for a geocentric or heliocentric model of the solar system, and what considerations seemed to mitigate against the latter historically.
So, this is why it’s usually a good idea to have a certain epistemic humility about reasoned-to conclusions (while simultaneously having epistemic courage in trying to figure out how a certain conclusion could work, how the evidence could be reassembled in a different way and make sense, or more sense).