Addition 01 Jun 2021: It appears that I am mistaken as to the central metaphor and comparison of this post that is referred to a bit further down. It turns out that the people in the Middle Ages knew perfectly well that the Earth is not flat. This doesn't change the fact that in most of the popular consciousness they did not. Please continue reading as if they did not know that. The central claim still stands.

The most accurate critique of rational thinking is not in the Bible. It's not in the works of Kant or any other empiricist philosopher. It it is, instead, found in the most unexpected of places.

It is hidden in the flavour text of an Android: Netrunner card The All-Seeing I.

It's a simple quote - within the Android universe taken from a mystical source known as the Playbook, which seems to often paraphrase Sun Tzu's The Art of War. The intent of this quote ought to be immediately evident to anyone familiar with the world of Android - it's the fundamental underpinning of the megacorp NBN. NBN is the media-advertising-surveillance conglomerate, the ISP that watches everything you do, the social network that monitors the pages you subscribe to, the feed that filters the information you do not want to hear, the AI network analysing everything you do in order to make the most optimal decision for what to serve you next.

It strives to retrieve perfect information at the expense of people's right to privacy and free choice.

However, when one inverses the quote mentioned above, the flaws of rational thinking quickly start to show themselves.

It is self-evident that for most practical purposes we can almost guarantee that in any given situation we do not have all the information. No human is omniscient.

A well known example from history of astronomy comes to mind here - the reason why for a number of years the Church and the broader opposition of Copernicus and Galileo were able to reasonably and scientifically doubt the heliocentric theory.

Aside from the fact that the Ptolemaic geocentric model was grounded in quite a lot of evidence and calculations, each proposed proof of the Copernican heliocentrism at the time did not include a reasonable explanation for the perceived lack of stellar parallax. The argument went as follows: since the Earth supposedly moves around the Sun within a year, there should be some differences in positions of stars in the summer and in the winter. After all, when one looks at a tree standing a few metres to its left, it is relatively in a different place than if one were looking at it standing a few metres to its right.

The solution to this problem is known to us now, but it was impossible to observe with the instruments of the time — and it seems painfully simple. The matter lies with the precision of the instruments and the assumption that the stars appeared a lot closer than they in fact are. Once more precise instruments were developed and stellar parallax could be measured - orders of magnitude smaller than initially expected - the heliocentric theory could finally be accepted as true.

This story shows two very big problems with rational reasoning. Based on missing information it is very simple to dig your heels into a theory that's wrong - that's what the opponents of heliocentrism showed. And there are certain assumptions which have to be discovered empirically because they do not infer from anything obvious.

Let's leave the problem of eternal uncertainty aside for a moment and move into the realm of perfect information. The realm of chess.

While chess is an extremely complex game with trillions of permutations for pretty much every possible move, it is still significantly less complicated than the real world, the times in which we live and the decisions we have to make every day. With the perfect information, a chess grandmaster has to think many moves ahead — a feat many inexperienced players are incapable of.

And that is all in a game with a strict structure, repeating models and simple rules. How could anyone hope to achieve that level of analysis for any practical application in the infinitely more complex real world?

That's why, personally, I prefer poker. Poker is not entirely rational, even though it also has a strict structure, repeating models and simple rules. In the words of Martin Amis, "on a chessboard, the properties and powers of a bishop are fixed. In poker, it's all wobbled through the prism of personality." And besides, it's a whole lot more fun to watch than chess.

Another example, perhaps slightly more relatable to the programming part of those reading this blog might be the failure and the move away from the waterfall model of project management.

When examined superficially, waterfall seems like the most reasonable and rational system for creating a software product. There are certain assumptions about the needs of the market, then a project is rationally planned based on these assumptions and ran through various groups of professionals to completion: stakeholders, designers, programmers and testers.

What waterfall - and often also rationalism - does not take into account is change and imperfection of design. Assumptions made in the beginning change. Certain decisions taken over the lifetime of the project are deemed to be imperfect and change also. Eventually, one of two things happen: either the project is delivered sub-par or it is paralysed because of constant change which was not factored into the timeline and not delivered at all.

I am by no means suggesting that we should abandon reason forever in all circumstances - a dull knife is still useful to spread butter even if it cannot be used to cut bread. And if a dull knife is all you have then it makes sense to still use it, especially when it brings evident progress.

What I am proposing though is to abandon the everyday cult of reason, the notion that everything can be simply deconstructed and planned. Rationalism is a tool rarely within the grasp of ordinary people and useless for ordinary applications. It can be used either for extremely trivial things - those for which we are able to obtain perfect information and develop a perfect strategy - and for extremely abstract things - those for which we make up the assumptions ourselves or are absolutely certain about their validity.

It's a good tool for the purpose of science - as science is not afraid of demolishing its structure and scaffold supporting it when the assumptions do not seem to be valid and building them up anew. We saw it with geocentrism, we saw it with calculus, we saw it with non-Euclidean geometry.

However, applying reason directly to everyday situations by everyday people who are by no means disposed to the level of analysis proper reasoning requires. We are prone to attachment (which in pretty much every single major religious and philosophical movement is posed as an obstacle to enlightenment. In Christianity it's in the story of the rich in the kingdom of God) that often results in us digging our heels into perilous situations.

The problems faced by each of us in the modern world are something that rational approach is ill-fitted for. Perhaps there needs to be a shift in thinking, perhaps we need to explore a yet undiscovered branch of reason or completely reconsider the way we approach these new, difficult problems. Much like Newton had to invent calculus to solve the problems of his mathematical work, or how the discovery of the New World by Columbus upset the theory of flat Earth.

Much like in the medieval times walking off the edge of the flat Earth constituted a danger of falling off it, so too nowadays wandering off too far away from rational thought is dangerous because it leads to madness. However, much like the sea route to India is nowhere to be found on the flat Earth, so too the practical application of rationalism is nowhere to be found in a rational mind.

So perhaps a rational mind too needs shifting to maybe not necessarily find solutions, but at least acknowledge its own shortcomings much like it needed shifting 500 years ago to acknowledge that the Earth is, in fact, not flat.