How our reason and senses work together

Contrary to philosophical debates, we need reasoning and experience equally for knowledge

A debate between Plato and Aristotle as depicted in Raphael’s fresco “School of Athens”

One of the oldest philosophical debates is between the empiricists and the rationalists. It dates back, at least, to a clear difference between Plato and Aristotle on the source of human knowledge. Is knowledge built out of our experiences and perceptions? Or should we trust our minds and thinking abilities instead? Philosophers have been taking sides ever since. To simplify, philosophers from Ancient Greece through to (and including) Descartes, tended to be rationalists - arguing that our minds are the ultimate source of knowledge. They were notably joined by the ‘German Idealists’, such as Kant and Hegel. Alongside the founding of modern science, Locke, Hume and most analytic philosophers in the twentieth century were empiricists, who argued that the source of knowledge was our experiences.

Interestingly, the two systems of human thought explained in my previous post, and the way we saw Sherlock Holmes solve cases, provide a useful perspective on this old debate. To see this, it will help to understand the shape of the debate, at least in an over-simplified way, since Aristotle made an empiricist critique of Plato's rationalist philosophy.¹

For the rationalist, reason is "the chief source and test of knowledge".² There are two broad reasons for this. One is that there is knowledge, like mathematics, that we can know simply by thinking about it clearly. The second is that we all know that our senses get confused, tricked and often make mistakes - so it makes no sense to think our experiences are primary in building knowledge. Today there is a whole genre of visual illusions you can find online that illustrate this well.

To the empiricist, "true knowledge or justification comes only or primarily from sensory experience and empirical evidence".³ For them, any talk of rationality as the primary source of knowledge clearly puts the cart before the horse. We cannot have any knowledge without there being something that we have knowledge of, and (outside a few particular cases) we can only access any of those things we have knowledge of through our experiences. So all our knowledge has to be built up out of what we have experienced.

Two systems of thought

You may have noticed something familiar in the two sides of this debate. They bear more than an accidental resemblance to the two different thinking systems we covered in the previous post. The rationalist clearly prioritises the theory-building aspect that is typical of the left hemisphere of the brain; while the empiricists take the outwards looking, right hemisphere approach to be foundational. Assuming the two systems account is correct (or even just plausible), both the empiricists and rationalists have prioritised only one of the two systems necessary for thought and knowledge. How these work together is interesting and illuminates where both empiricists and rationalists get things right and wrong.⁴ We'll use Sherlock Holmes as a case study again, although could just as easily look at scientific practices or a number of other ways we build knowledge.

As we saw in the previous post, Sherlock Holmes' method is always to start by carefully gathering evidence - typically via closely interrogating the person who brings the case to him. This seems to put him in the empiricist camp - knowledge comes from experiences. However, what he does once he has the evidence is notable. He doesn't keep gathering more evidence and focus on building knowledge via his experiences. Instead, he stops and thinks hard - normally in good Victorian style by staring into a fire, smoking his pipe, playing a violin or going for a long walk. He takes a decidedly rationalist approach in trying to figure out what happened. He always tries to find a clear (preliminary) solution to the case purely by thinking about what would have had to have happened for the evidence to hold.

Of course, this doesn't end the story, because Holmes shifts back into an empiricist mode as he tests his rationally constructed theory against further evidence. But, crucially, this is a purposeful hunt for evidence that seeks to confirm or disprove the theory (or theories) he is considering. He considers the consequences of his theory for the world and goes searching for the evidence that should be there if his theory is correct. This is a targeted, deliberate form of empirical investigation that is profoundly determined by the rational theory he has constructed. The rationalist, or theory-building system, determines where he pays attention empirically and what he looks for. In this approach, knowledge is anchored in experience but not fully determined by it.⁵

So, on the one hand, rationalism is right in the way it highlights the crucial role that our reason plays in constructing knowledge - sometimes with very little to go on. But our knowledge is, in the end, grounded by our experiences. It doesn’t come directly from reason or thinking (except perhaps in some limited situations).

By contrast, empiricism rightly highlights the importance of our experiences but overshoots with the claims that knowledge only comes from them. Instead, knowledge comes from us thinking about, reflecting on and theorising based on our experiences. We can see these dynamics in a range of different contexts.

For one, many of the conclusions in our best scientific theories are rational postulates that produce a theory that works. Gravity is a classic example - we can't see, hear, touch or smell gravity. And yet, by rationally deciding to treat it as a force that acts over distance according to a particular mathematical formula, we can explain how important parts of the world function. Our knowledge of gravity is neither made up of our experiences nor purely rational.

Another example is the way that different cultures and languages have different sets of names of things that often don't quite match up. A famous example is that the Eskimo have about 50 words for snow. They have a set of concepts and names that are critically important for them, but less so for almost everyone else. All cultures have landed on a set of concepts that work for them and are grounded in their experience of the world. However, experiences don't fully determine the concepts and so different cultures have made different choices.⁶

In all of this, there is a constant interplay between the two systems of thought, or between empiricist and rationalist approaches to knowledge. Neither of them is primary as each cannot produce knowledge by itself. Empirical investigation of the world, or experiences, can only provide a collection of insights or ideas that are hard to communicate or generalise - and don't build anything we'd recognise as knowledge. But rational thinking or theory-building has no content to work with without our experiences.

Uncomfortable conclusions for many philosophers

Historically, both the schools of empiricism and rationalism had a clear focus on identifying the real source of human knowledge. Their aim was to provide a reliable basis to knowledge that would provide us with a certainty, or near certainty. This requires a solid foundation for knowledge, and therefore a clear source. However, if knowledge is built from an ongoing interplay between the two systems of thought, certainty becomes unachievable. Our knowledge is always (at least in theory) somewhat provisional - more evidence could arise that would require us to change our theory or knowledge.⁷

This is, in my view, a strength of this approach. Certainty about knowledge has been, as I have argued previously, a multi-millennia long preoccupation of philosophers that has never succeeded. Understanding human knowledge as an ongoing duet between empirical and rationalistic methods, or between right and left brain hemispheres, provides an account of knowledge that matches how humans operate and allows us to build confidence in knowledge without requiring us to achieve epistemic certainty.

A second consequence of this account of knowledge is that it means knowledge is not fundamentally propositional - we don't start with a series of true statements or facts and build up what we know from that. Instead, we construct whole theories or other representations of the world that (i) get judged as a whole and (ii) depend on pre-existing knowledge or other theories. Sherlock Holmes is a classic example of this - one of the reasons that Holmes can solve many of his cases is that he has an encyclopedic knowledge of how a wide array of things (from social dynamics to technology) work in the real world. Some of these are his own theories, some are more well known.

1. If you want to go into far more technical depth, check out the Stanford Encyclopedia entry on “Rationalism vs Empiricism”: https://plato.stanford.edu/entries/rationalism-empiricism/

2. This neat definition comes from https://www.britannica.com/topic/rationalism

3. This time, Wikipedia has a nice definition: https://en.wikipedia.org/wiki/Empiricism

4. As always here, the analysis is very high level and over-simplified - but the big picture pattern is useful. There are many different nuances in the literature that won’t be covered.

5. This pattern of thought or problem solving matches standard scientific practices. Scientists observe something in the world, build a theory to explain why that happens, then run experiments (or collect data) based on what the theory says should happen.

6. Almost always, the different sets of concepts can map onto the same set of objects, phenomena and experiences in the world - so different languages are not creating different worlds. But they construct the map in a different way and so the world might look different from the inside of different cultures or languages.

7. In fact, Conan Doyle did just that with his Sherlock Holmes stories. He had tired of writing Sherlock Holmes stories and so, in The Adventure of the Final Problem, Holmes dies by falling down a waterfall in combat with his mortal enemy, Moriarty. About ten years later, Conan Doyle published The Adventure of the Empty House in which we learnt that Holmes’ death was in fact an elaborate ruse.