Many people are very willing to champion the importance of reason and rationality, but this can be very similar to praising motherhood and apple pie: few people defend unreasonable or irrational behaviour. I hear people advocate for rationality - but in my experience, it is quite rare to hear them say whether our actions should be based on reason alone, or if reason should be used alongside something else - and, if so, what.
Ayn Rand famously championed 'absolute reason', and she is not alone; it is not hard to find people and groups prepared to argue for the sole primacy of reason. To give just one example: "Knowledge is lucid and can only be formed by the use of reason. There is no other path. Reason is absolute."  Romanticism was a reaction to the emphasis on science and rationality produced by the Enlightenment - the Romanticists rejected the popular belief in reliance on reason alone.
This issue has been explored in many works of fiction. In one of the best-known examples, Mary Shelly's Frankenstein, the creature is produced by science, but we see both the creature and its creator being driven by passion. And in Metropolis, the classic film by Fritz Lang, we see a society ('the Hands') ruled by reason ('the Head'), and the result is efficient but brutal. The film's message, "The Mediator Between the Head and the Hands Must Be the Heart", has been dismissed as trite - but 'trite' simply means that the message is obvious and uncontroversial. For many people, it is obvious that feelings are also important, that reason must be moderated by compassion and morality.
Guided by Science?
In talking about the way the UK government responded to the Covid-19 pandemic, the Prime Minister repeatedly claimed that "at all stages, we have been guided by the science". This claim is not only false, it is a dreadful and absurd idea. It is also a common technique, used by people to avoid taking responsibility for their decisions. Science does not tell us what to do, it only tells us what can be done. "We are simply following the facts" is a manifest impossibility: facts are statements about the past, about what has been, but a plan tells you how to create something in the future. You need an intention to build or a desire to possess something which does not currently exist: what you intend to create is not yet a fact.
Neither science nor reason can guide anything, from individual choice a national strategy. If you can provide an objective, a context, a set of resources and a metric, then science and reason can (possibly) tell us the best way to achieve the objective. Science can tell us the fastest (or the cheapest, or the most energy efficient) way to get from A to B, but it cannot tell us whether we should go there. It can tell us the likely consequences of an action, but it cannot tell us if the action is good, kind or loving. Reason can often tell us how to do something, but it cannot tell us if we should do it. Science and reason were vital tools in working towards the Millennium Development Goals , but setting and agreeing the goals was mainly a combination of morality and politics.
Science, reason and rationality deal with facts, with what is, but they cannot give us a single moral principle; in addition, they cannot give us a goal to aim for, or a purpose to live for. As the saying goes, "you cannot derive an 'ought' from an 'is'." And the inability of science and reason to speak about any of the things that matter to us as human beings is not just an issue about the origin of morality, value and purpose - reason can't even be used to justify Rand's fixation on selfishness: to make selfishness a reasonable strategy, you have to add in a whole bunch of beliefs about the world, about the functioning of social organisms, and about human nature. And if you want to promote your choice of a particular set of beliefs, as Rand does, you need a stronger argument than 'they seem reasonable to me'.
Reason and ... what?
The Framework used on this web site is helpful here, distinguishing between particles, people and purpose. As we say, each of these areas requires the use of a distinct set of tools and disciplines.
- When we are talking about particles - and things like them - then science and reason are the tools to use. Science gives us laws which - with an astonishing level of precision and regularity - tell us how the physical world operates and enables us to predict thousands of years into the future.
- But when we are talking about people, and living creatures in general, things are different. Our bodies follow the laws of science, but what we do with them is another matter. There are common topics which we recognize; the basics of survival, sustenance and sex are fairly universal, but these common concerns work themselves out in uncountable and frequently unpredictable ways. Social animals care about things like status, reputation, respect, cooperation and intimidation. We can study the behaviour of living creatures and seek to understand them, in subjects such as biology, psychology and sociology, but this understanding has to be expressed in terms relevant to living creatures (such as 'safety' and 'status'), which simply cannot be expressed in terms of the activity of atoms and molecules.
- And when we come to consider issues such as purpose, things are different again: the disciplines which work when we are seeking to understand the observable activity of living creatures are not able to help us here. Ideas and beliefs are the subject matter of philosophy and religion - which does not mean that reason needs to be thrown out of the window, only that it operates in a different way when dealing with a different subject, and things it deals with are not the subjects which are considered in the physical sciences.
In each of these areas, reason alone is insufficient to tell us anything: reason always needs material to work with. In physics, reason deals with things like mass and energy; in biology, reason deals with things such as risk and mate selection; in philosophy, reason deals with things like existence and morality.
It is tempting to say that reason needs facts to work with - but in each area, what we consider to be 'the facts' change and develop with time. In science, new discoveries contradict (or, at least, modify) previous certainties: we believed Newton's laws for a long time, but they turned out to be close approximations to the truth when you restrict your interest to certain parts of reality; General Relativity and Quantum Mechanics both appear to be true, but contradict one another. What we consider to be scientific truth is only a hypothesis which has not yet been falsified. When you apply reason to mistaken ideas people wrongly consider to be facts, you are unlikely to get a good result. In the world of computers, there is an acronym which summarises this nicely: 'GIGO' - Garbage In, Garbage Out. We are less able to apply reason to the real world than most people think.
And remember: even if we were able to apply reason to the real world in a reliable way, a great many people think that compassion and morality should influence our behaviour, as well as reason.
The Limits of Reason
Reason is not only limited by our understanding of the world we apply it to; reason itself is limited.
The study of reason is called 'logic'. There are numerous kind (or 'systems') of logic, such as Aristotelian and Boolean. In general, systems of logic work very much as you might expect while they are purely abstract systems, but various problems arise as soon as you attempt to relate the logical system to the real world - or even when you attempt to create a purely abstract system which comes close to being able to describe the real world in a meaningful way.
In particular, problems arise when the system accommodates infinity (important when you are dealing with Mathematics) or self-reference (important when you use language and try to think about the meaning of words, also important in Mathematics when dealing with set theory), or the system attempts to be complete. There is also the irritating observation that an attempt to test the truth of two statements which are logically identical can be very different in the real world .
In Principia Mathematica , Russell and Whitehead attempted to derive Mathematical truth from a few basic axioms (statements they considered to be self-evidently true) and logic, while avoiding the pitfall of self-reference (easily seen in Russell's Paradox - if you construct the set of all sets which do not contain themselves, does this set contain itself?) but the attempt was not successful on either front: avoiding the problem of self-reference became unworkably complicated, and some 20 years later, Kurt Gödel demonstrated that the attempt was doomed to failure: in any consistent system, there must be true statements which cannot be proved .
It turns out that we are not good at identifying what is 'obviously' true. Around 300 BC, Euclid created the formal discipline of Geometry by identifying a few 'obviously' true axioms (or 'postulates'), and using them to prove a large number of geometric propositions. For around 2,000 years, almost everyone who thought about these things believed that Euclid's geometry described facts about the real world - facts which Euclid had proved, and anyone who studied him could prove for themselves. Then some Mathematicians started to play around with alternatives to one of the axioms (the 'parallel postulate'), and created other self-consistent geometries . Then Einstein came along, and pointed out that the universe we live in is not Euclidian, after all. None of the Geometry you learned at school is actually true in the real world - it's just a close approximation when the lines you draw are short enough.
The Use of Reason
Reason is limited. We care about things which reason can say nothing about; reason can only work with information which can be wrong or misunderstood; reason itself is flawed and incomplete; and reason fails to handle the real world, the one we are seeking to reason about. But this does not mean that reason is useless, only that it is not omnipotent.
There are many areas of life where we need to apply reason, where we need to identify the facts (as best we can) and form a rational response to those facts. Climate Change is one obvious example.
But reason and rationality are not enough. People do not deny Climate Change because the evidence is unclear or the reasoning unsound. There are probably many reasons: the probable consequences of our activity are too horrible to contemplate; they are too attached to their standard of living and cannot face the changes which will be required; it is far more comforting to listen to the voices who point out that people have been prophesying disaster since the beginning of human civilization; and so on. The reasons are emotional, and social and cultural, and probably many other things.
We need reason. But the most important thing we need is not a better grasp of reason, but a better grasp of when and how to use reason, and what else we need to bring to bear alongside reason, in order to address the massive problems this world faces.
(This article started as a response to the post about Ayn Rand and Objectivism.)
 From the article about 'Reason' in the Importance of Philosophy website (www.importanceofphilosophy.com/Epistemology_Reason.html).
 The Millennium Development Goals were agreed by the United Nations in 2000 (www.who.int/news-room/fact-sheets/detail/millennium-development-goals-(mdgs)).
 The two statements, "All swans are white" and "All non-white things are not swans" are the same, from a logical perspective, but one is a statement about all swans (which could potentially be verified), and the other is a statement about every non-white object in the universe (good luck with verifying that).
 An outline of Principia Mathematica which helpfully abandons the original notation is provided by Stanford University (plato.stanford.edu/entries/principia-mathematica/).
 The 'Incompleteness Theorem' is surprisingly easy to understand in outline (en.wikipedia.org/wiki/Kurt_Gödel).
 The 'parallel postulate' says that there is precisely one line which can be drawn through any given point which is parallel to another given line, although it can be stated in a number of (surprisingly different) ways (en.wikipedia.org/wiki/Euclidean_geometry).
[See also Reasonable Misdirection.]