Critically Minded

Here's another short post exploring an idea I'm finding useful. I've used this in a few recent posts, so I just want to set it out on its own.

People seem to think in quite distinctly different ways. The different kinds of thinking are numerous, but there are a couple which I think are often confused : critical thinking and analytical reasoning.

The analytical thinker is generally someone like a scientist, who will pick apart an idea into its component variables and explore in detail what would happen if you change any of them. They'll run with this to the nth degree, examining consequences until they really feel that they've fully understood a concept, or can make a testable prediction to check whether the whole thing holds up to scrutiny. Analytic thinking is often technical : most obviously, perhaps, when it comes to mathematics, but it can apply to other areas too. Anyone who'd ever solved a problem by testing out different aspects to destruction (who hasn't ?) has done at least some analytical thinking.

I would define this as asking the question : what if this is true ? What consequences follow ? How does this one thing affect other things ? That's the essence of analysis.

What I think people often confuse this with is critical thinking. And to be fair, this too is a major component of science. Having decided on a way to test their idea, the scientist should then actually do so, or at least consider the results from someone else's test that attempts to answer the same question. The critical thinker is not concerned with what the consequences are, they're concerned with whether the premise is correct. They don't want to speculate as the analyst does. They want to know if they're right at all.

This mode of thinking I would define as asking the question : is this really true ? Can I verify it ? Could there be another explanation ? That's what I mean by critical thinking.

The analogy I like is from programming, having learned this the hard way through direct personal experience. A really protracted debugging session will go something like this :

• First, I'll check if I have any typos, any missed commas or wrong equals signs, or something where a variable isn't being set correctly. This would be the case of the code not doing what I thought it was doing.
• Next, I'll go up a level and look at the code structure – maybe I've got a loop that's nested wrongly, so it's iterating over the wrong variable or not being terminated correctly. In this case, the code might be doing what I thought it was doing, but the way I thought things should be done was itself partially incorrect.
• Finally, I'll stop and think if the very method I've been using is actually likely to give me the correct result at all, if it's even fundamentally possible for it to work or I've built a horribly complex house of cards and need to start again.

Programming seems like a good analogy for me because it encapsulates both modes of thought. The low-level debugging is analytical : have I got the right variables, what if I change where the loop is run, is my input correct, etc. The high-level stuff is critical : is this method actually going to work if I do it correctly ? This sort of multi-level, or multi-scale, thinking blends quite nicely from one mode of thought to the other. It's something LLMs have become noticeably better at over the last few months, no longer picking over minutiae, but actually stopping to consider the premise of a question.

We can of course imagine in a four-way graph to describe this. We can have (1) those who are both critical and analytical, which is pretty much ideal for a scientist. There are (2) those who are analytical but not so critical, again a trait common in the sciences : these people are fine so long as they're given the right problem to tackle. People who are (3) critical but not analytical are less helpful, something common among loons on the internet : "dark matter doesn't exist because your ugly face, that's why" types. And finally of course we have (4), those poor unfortunates who don't do much of either of these modes of thinking.

Dedicated readers might remember my longer 2015 post about skepticism. Back then I struggled to find a good word to describe this sort of concern for the truth, and perhaps there isn't one : "critical" has the same popular negative connotations as "skeptical" in everyday use. But critical thinking, as a term, does seem to be used in this sense of wanting to find out the truth regardless of the result

Arguably there's an overlap here with curiosity, which similarly implies wanting to know the truth. The problem is that curiosity can also mean something more like a greed for more and more facts : a desire to travel for the sake of experiences, or to read more and more books to see what they contain, or an urge to run an experiment without any preconceptions as to what will happen at all, rather than to test the validity of a claim. If you follow this blog regularly you'll know I'm intensely curious about mythology, but not because I want to determine if Zeus existed or if I need to defend myself against the afanc on my next trip home. 

Critical thinking, on the other hand, seems to much more specifically capture this sense of a desire for verification. Like everything else, it's not an absolute state of mind. Someone might be extremely critical when they first learn a new fact but far less eager to test something they learned years ago, or their degree of critical reasoning might vary enormously across different subjects (I want to know if, say, dark matter exists, but I don't give a flying crap about whether celebrity X really said statement Y on social media platform Z).

Nor is it realistically possible to hold ourselves to the highest standards of critical thinking at all times. If you go down that way, you end up in a postmodernist Humean nightmare where nothing can ever be truly verified and nothing known, further progress being hampered by intellectual impotence. This is why Ronald Hutton's Pagan Britain annoyed me so, being so resolutely noncommittal that he wouldn't even venture to suggest how we could even test anything, let alone claiming that any one interpretation was actually true. That, in my opinion, is not a productive way of learning anything. Better by far to hold an opinion but be prepared to surrender it rather than never believing anything at all.

Likewise, it's also possible to be analytical to a fault, obsessively examining every detail even when they have no possibility of changing any major result (this is the fault of many a peer reviewer). So our four-way graph would be complicated, with the extreme not necessarily being the place one wants to be. Perhaps this helps with a description of wisdom. Maybe wisdom is knowing what should be done, when to apply critical and analytical reasoning and how much, when to rabidly fixate on an issue and when to let go – where exactly the balance of the different ways of thinking lies to ensure a successful, happy outcome.