Misunderstanding critical thinking

philosophy
epistemology
Have your feet on the ground.
Author

Norman Simon Rodriguez

Published

21 January 2026

It’s always annoyed me a bit that critical thinking is often understood in academia as a skill that has to do with spotting problems in the way arguments are formulated, that is, as a skill in finding cracks in language games. In other words, critical thinking isn’t seen as truth-seeking, only as consistency-seeking. The risk with that approach is that many theories, especially in the humanities, end up being elevated to a status of respectability only because they are well formulated, regardless of whether they are true or not. But because assessing the inner coherence of a complex theory is hard work, many theories end up getting their ‘respectable’ badge without passing through a truly rigorous QA process. The result is that theories get promoted and accepted only because they seem to be internally consistent. This is already two rungs below the ideal outcome. Ideally, widely accepted theories should be true, but critical thinking is demoted to only caring about internal coherence. That’s one rung below. But then, academic gatekeeping can only handle the appearance of internal coherence. That’s two rungs below.

I remember feeling uneasy about this in my first economics classes a while back. Neoclassical economics was taught as this failproof, mathematically elegant way of solving the economy as though it were a giant equation. All I could see was a massively simplified picture that looked more like a toy than a tool for responsible policymaking. But the alternative coming from the intellectuals just opposite the Economics department, namely the neomarxist approaches, seemed similarly inadequate. Their theories, again, looked so plausible, so intellectually satisfying. And just like neoclassical economics, these approaches had also developed an astonishingly complex and tightly condensed corpus of doctrine that made sense within itself. But I was very aware that the neomarxists’ way of seeing reality was equally narrow and simplistic and, of course, that History had proven them wrong again and again (and again).

I have a different opinion on critical thinking. For me, critical thinking is the disciplined business of picking and refining mental models based on how well they actually hold up when they collide with themselves and with the real world.

It’s best to think of a belief not as some isolated thought floating in a vacuum, but as a component of a larger mental model. Models are structured representations of how things work, and they’re incredibly useful because they generate expectations and guide our actions. From my perspective, critical thinking isn’t just about judging whether an argument sounds clever or rhetorically ‘fair’. Instead, it’s about taking the containing competing models and running them in full against reality to see which one breaks first. It’s less like a courtroom drama and more like a stress test in a laboratory.

In other words, a model earns its keep by being right about what happens next. If it’s going to have any clout, it needs to anticipate observable outcomes and do so with a decent level of reliability. Think of it like this: if you’ve got two economic models designed to tackle inflation and one consistently predicts what’ll happen after a policy shift while the other is erratic, the first one is clearly superior. It doesn’t matter how elegant the second model’s maths might be; if it fails to survive contact with reality, it’s got to go.

A corollary is that belief is always provisional. We should only stay committed to a model until a better one comes along and starts outperforming it. It’s a bit like upgrading your phone; you don’t do it because the old one stopped existing, but because the new one actually does the job better.

I’m not saying of course that internal coherence isn’t relevant. Assessing that coherence is still part of critical thinking. But I do think that external evaluation criteria should be taken into account. Defining what those criteria should be is a subjective choice, however. As humans, we decide which parts of the world are worth our attention based on what we’re trying to achieve. An engineer will naturally prioritise safety constraints, whereas a business owner is going to be more interested in market trends and profit margins. These choices are pragmatic and goal-relative rather than a search for some abstract, context-free truth. Once you’ve set the rules of the game, however, the results of the model comparison aren’t arbitrary at all. In my view, critical thinking should be like a benchmarking exercise with pre-defined comparison metrics.

Truth, in this sense, is about what works given a subjective/sociological choice of what the goal of the comparison is. A model is perfectly acceptable if it performs at least as well as its rivals relative to the goals we’ve set.

To put it simply, critical thinking is the disciplined practice of choosing between mental models by looking at: