Questions Considered

Notes on thinking, learning, decision making, and occasionally running. Simple ideas, mostly obvious.

Epistemic Dead Ends

Learning and intellectual growth is about accretion of knowledge and deepening understanding. It is also about recognizing the myriad ways, where you might come to a stall or a stop in your journey.

Those occasions can sneak up on you, such that it feels like you have actually rather arrived at a reasonable place. The best traps are after all those that you do not even notice yourself to be in.


Here are just a few examples to give shape to this idea.

  1. Categorization as premature closure. We live in a complex world and are ever in need for tools to help us find our away around in it. Categorization, labeling is useful to group things together, reason about them, distinguish them from those. Where it simplifies reasoning, it can also short-circuit it, e.g. she voted Republican, so she must be X; he is a Christian, so he must be Y; I am over thirty so I cannot Z.
  2. The map at the expense of the territory. The map is about but not equal to the territory. We create a model to describe and reason about a reality. The model is useful to the extent that it does that well. When we stop comparing, stop paying attention to the discrepancies between the map and its territory, we risk becoming an expert of the model without understanding (and possibly divorced from) the reality it is supposed to model.
  3. Insufficiently questioned reliance on generative artificial intelligence. With generative AI and large language models we have powerful tools to engage in conversational experiences and receive AI-generated responses to our inquiries. We can use this to be more productive and to learn things. It is easy for AI-generated answers to look plausible, but also be partly or wholly wrong. If we do not pay enough attention and/or converse well outside our circle of competence, then it can likewise be easy to not notice substantive issues and mistake plausible for correct. Answers do not equal understanding – and having the former does not mean also having the latter.

All of these examples individually deserve much deeper treatment and they are merely listed here to illustrate the point. Each of these can be epistemic dead ends, not because you necessarily arrive at the wrong conclusion, but because they can halt further inquiry.

The harm, as so often, is not the tool itself, but rather what you choose to do with it.


Perhaps you think of dead ends as those no-exit streets that seem to end quite abruptly. On the other hand, when you hear cul-de-sac you might think of a road that ends in a circular turnaround, often in a small town or suburban setting.

That cul-de-sac may seem somehow better or more desirable, than the dead end. Perhaps the houses look nicer and the turnaround makes navigation a little easier. That neighborhood seems perhaps a bit more inviting.

That would be a risky deception to the extent that you feel better about just hanging out, where you find it, rather than recognizing it for what it is.

Unless this is where you want to stay, the only way to get to new places, is to turn around or back out, i.e. leave, then find a road that offers more choices.


Dead ends do not lead anywhere.

Discover more from Questions Considered

Subscribe now to keep reading and get access to the full archive.

Continue reading