Minimal conceptions of reality

We don't need to understand what is really real to be able to build knowledge.

Photo by Javardh on Unsplash

One of the core tasks of epistemology is to find a coherent way to distinguish knowledge from belief. Our intuitions on this are fairly clear, although they can be hard to define. In some way, knowledge has to match, reflect, correspond to or capture what is real or true, while belief doesn’t. As I have put it previously, we always need to test our ideas or theories against reality before we can count them as knowledge. This, however, seems to raise questions that move us from epistemology to ontology. If the definition is a test against reality, surely we need to ask: what is actually real and what isn't?

This intuitive relationship between ontology and epistemology emerges in various philosophical considerations, such as in questions about the status of knowledge and truth about fiction and fictional characters. We all know that Darth Vader was Luke Skywalker's Father, but philosophers tend to worry deeply about how we can know something about two fictional characters who were never real.

These worries are natural if we accept common attitudes about the nature of knowledge. If we assume that we should be able to achieve epistemic certainty (i.e. we should be able to know things for certain and without any rational doubt) and that statements or theories are universally true or false, then accepting that we really know that Darth Vader was Luke's Father becomes highly uncomfortable.

If, on the other hand, we take seriously the attitudes I have been arguing for, namely that we have to be humble about what we know as we might always turn out to be wrong; and that our knowledge is always, at best, a partial and limited picture of the real world; these questions matter less. To see how this works, however, we need to be clearer about what we mean by the reality we are aiming to test our ideas or theories against.

Reality is anything you might stub your toe on

The concept of knowledge, as opposed to belief, depends on us being able to measure or test things against reality. However, the logical structure of human knowledge does not depend on any profound or ontologically robust conception of reality. All the logic requires is that we have an idea about what we think is the case, and then that we can test it. If we think about what makes something a genuine test, it is (at a minimum) that there is the possibility of both a pass and a fail. Otherwise there is no actual test.

In the case of knowledge, this means that we need something to test our ideas or theories against where our test can pass or fail. For this to hold, we just need something to test against that is independent of us. If everything depends on us, on what we think or want, then the test will always be determined by what we think should be the case, that is by the idea or theory we want to test. That is not a genuine test. In other words, there just needs to be something we might bang our head or stub our toe on (literally or metaphorically) if our ideas don’t match what is out there.

To illustrate this dynamic, it is might help to consider the way our dreams often work. When we are dreaming, we can often change what is happening simply by thinking or deciding we want something different. I can decide I want to fly and so start flying. This is clear evidence that the dreamworld is not real in any substantive sense, as it is not independent of me. I cannot go exploring in the dreamworld and build genuine knowledge of what it is like because it may all change to match what I think it should be like.

The opposite of this, that there is something there that I can’t simply shift by decision or will, is not a strong commitment to any particular account of reality. For example, it is hard to say what exactly feelings of pain or love are in any strong ontological sense. But those feelings of mine are clearly real in the sense that matters for knowledge - I can’t just switch them on and off whenever I feel like it.

This minimal conception of reality - simply anything independent of my ideas or will - may not provide any substantive ontology, but is sufficient for us to build human knowledge. Notably, it provides a coherent account of why we can know things about fictional characters.¹ I can genuinely know that Darth Vader is Luke's Father because all of us can go and watch the movies to check. If someone was arguing that this wasn't true, then we have something independent to appeal or refer to. Therefore, it is meaningful to claim that we can know this information.

Interestingly, this account echoes Karl Popper's well-known philosophical argument about the line between genuine scientific theories and pseudo-science. For Popper, a theory could only be genuinely scientific if it is falsifiable, that is, that there is (even only in theory) some evidence that could prove it to be wrong. He argued, for example, that theories like Freudian psycho-analysis and communism couldn't be genuine scientific theories because each was flexible enough to explain every possible events that could happen.² This means they could never be falsified or proven wrong.

This point can be rephrased in the terms I have been using. Theories cannot provide any genuine knowledge if there is no way to properly test them against something independent. If, assuming Popper’s argument is correct, every possible state of affairs is consistent with a theory, then no genuine test is possible.

Knowledge is always 'knowledge about'

This approach, however, raises a question about the nature and coherence of our knowledge, especially if we are happy to count as genuine our knowledge of fictional characters. It seems like we can easily end up with incoherent or contradictory knowledge. For example, consider a statement like "Sherlock Holmes lived on Baker St." It is (presumably³) false as a statement about historical residents of Baker St in London. But it is also obviously true as a statement about the fictional world Arthur Conan Doyle wrote about. Surely it can't make sense that the same sentence is both true and false?

This potential puzzle disappears when we take epistemic humility seriously and remember that human knowledge is closer to a series of partial sketches or pictures of reality than something that precisely describes reality as it is. One important point about a picture is that it is always a picture of something and from a particular perspective.

Initially, this might seem like a radical claim that undermines any claims to truth, but it quite closely matches standard scientific practice. All of our best scientific theories are theories of something. We have the Theories of General Relativity, and of Quantum Mechanics, and Germ Theory, and so on. And all of these theories are from the perspective of a particular discipline, which means that they don't translate easily into the other theories.

Similarly, anything we claim as knowledge is knowledge about something. It might be about the nature of the world at a sub-atomic scale, or about the processes to follow to get a form approved, or about a fictional world described in a book or a movie. Some people dream of a Theory of Everything that avoids this limitation, but we don’t even know what that might consist of.

To connect the arguments here, what some knowledge is about is precisely the reality it describes, or whatever it is that we can test or measure those claims to knowledge against. This is standard practice in all fields of knowledge. Quantum mechanics is about something different than sociology and so we take different types of evidence seriously in different fields.

Keep track of the about

As a general rule, we do not expect knowledge about different topics, fields or 'realities' to line up with each other if the 'realities' themselves aren't consistent. So knowledge about London as it appears in the Sherlock Holmes stories need not match anything historically accurate about Victorian London. Similarly, knowledge about the financial situation in a company won’t automatically be consistent with the lived experience of what it is like to work there.⁴

This means that knowledge is messier than we would ideally like it to be. But it doesn't mean that there are no genuine questions about what is true or false. For each of the different topics and fields (and each of the perspectives) there is an independent reality to test ideas against (or stub your toe on). If there is nothing independent to test against, we cannot really claim that there is anything to know.

When this messiness is unrecognised, it often creates fractious behaviour and conflict. The challenge is that the about matters a lot but we often only express the about implicitly and rely on context or implications. Whenever someone involved doesn't have or understand the context, things often go astray.

For example, the general tone of debate online and on social media tends to be people shouting very loudly past each other. They are using the same words, and sometimes even similar concepts, but regularly they are relying on different evidence and knowledge. The lack of understanding can be because they are talking about different things or (in the terms here) different ‘realities’.

For some readers, the argument here that all human knowledge is always about something - and that thing need only be an independent reality in the minimal sense that it is something that doesn’t depend on what I want - might seem fairly trivial and unsurprising. But it has more notable consequences than there is space for here. I’ll get back to this in later articles and so am very interested in questions or observations from readers in the comments (or elsewhere).

1. It also offers a neat explanation of how we can know things about logic or maths even if they don't describe things that exist. The underlying logic at play in these fields provides something independent from us and so there is something for us to stub our toes against, i.e. some way we can get things wrong.

2. Popper’s targets were often the strident followers of theories like these, rather than the original accounts. These followers could interpret any events to fit their theory, even if the original advocates offered definitive predictions that could be falsified.

3. I’ll admit that I haven’t done the research to check whether there was even a real, historical person called Sherlock Holmes who might have lived in London. It seems safe to assume not.

4. This gets can get very complicated when different fields with different types of evidence overlap, say with economics and political science, or sociology and human biology.