So what do we mean by truth?

Humility about knowledge should change how we think about truth.

If you have been following along here, you might be getting bored with my repeated conclusion that knowledge is hard and it is easy get it wrong. My ambition has been to chart a course between epistemic skepticism (knowledge is not achievable) and epistemic certainty (we have achieved genuine and reliable knowledge).

The close connection between knowledge and truth - it doesn't make sense to say you know something but that thing is false - means it is worth considering what this mid-way position of epistemic humility means for how we think about truth. To stick closely to my focus on humility, this post sets out my current, work-in-progress thinking about the concept of truth - which may be significantly revised over time. Your comments and feedback will help this progress.

Common correspondence

Amongst philosophers, the most common approach to explaining truth is known as a 'correspondence theory'. The core idea is simple: a sentence or statement is true exactly when what it says corresponds to or matches reality. In other words, to say that a sentence like "Humble Knowledge is an insightful Substack to read" is true means precisely that Humble Knowledge really is an insightful Substack to read.

Correspondence theories of truth neatly capture something of what we mean by the concept of truth - to say that something is true means that the world really is like what that thing says. Problems arise however when we try to apply it in any meaningful way. For this approach doesn't help us discover or know what is true. Some philosophers have taken this to mean that the concept of truth doesn’t really express anything meaningful.

A core problem is that correspondence (and similar) theories really just push the interesting questions about truth back a level. In practical terms, if we ask either whether a sentence X is true, or whether what X says really is the case, we are looking for the same answer. Importantly, it is often a difficult question because we don't have direct and unfiltered access to reality.

If I were God, for example, and could see reality directly as it is, then questions about truth would be straightforward and a correspondence theory of truth would likely be sufficient to explain the concept of truth. However, reality is much harder for us humans to understand and therefore truth is much harder to find. The core insight of a correspondence theory remains valid, something is true if what it says is the case, but it doesn’t answer the important questions: how can we decide or know what is true? In other words, the correspondence is a necessary but not sufficient condition on a theory of truth.

A more complete theory of truth?

So a correspondence theory articulates what we mean by the concept of truth in the abstract, even if that isn’t something easily achievable for humans in practice. What would a better theory of truth then look like?

For a start, it is important to note that a complete theory of truth will not provide us with a complete description of everything that is true - that is a different question. It also cannot give us a definitive method or procedure for deciding whether any particular claim is true. To adapt a phrase, truth is hard and it is easy to get it wrong. Instead, it is more important to focus on how we, as humans, come to see anything as true. This means a good theory of truth overlaps significantly with our accounts of knowledge.

Given that, it is worth recapping the conclusions reached so far. To start with, the basic units of human knowledge are not facts, observations or data, but rather they are theories, worldviews or other abstracted (plural) representations of the world. Alongside this, there is no neutral or independent way of deciding which theory or abstraction is correct, but rather we have to (in the end) see how each one stacks up as a description of reality by its own criteria. Importantly our knowledge is more of a series of sketches that only partially describe reality, rather than being detailed and definitively correct; including because our representations of the world can create internal artefacts that cause them to be different from how the world is.

So what does this mean for how we humans interact with and think about truth? For one, the importance of abstracted representations in knowledge reinforces the importance of a concept of truth. These representations, whether they be theories, sentences or pictures, are something separate from reality and it matters for us to know when they match reality. In other words, to say something is true is a meaningful assertion.

A second consequence is that truth is rarely just an individual relationship between one statement (or any one thing) and reality. Our statements and theories almost always have meaning and describe reality within a broader context of language, other theories, localised place and context. The correspondence theory implies that you can map true sentences to the part of the world where what they say is the case. Instead, the broader theory and context of a sentence can determine whether it is true or not.

The complicated relationship between statements, theories or other abstracted representations and reality means that deciding whether something matches reality, whether it is true, is often not easy. This is exacerbated by the way that all our representations are only partial sketches of reality - none can be perfectly precise and universal. Very different partial sketches of a scene can sometimes all be accurate representations, but are also more often the subject of significant debate and disagreement.

Truth is no different. There is often a question of judgement involved in deciding whether a theory or statement is true or not. Practically this means that there are many situations where intelligent, well-informed and insightful people disagree about what is true.¹ This doesn't mean that there isn't a truth of the matter, or something objective to measure truth against, just that truth is hard and it is easy to get it wrong. This is why there many cases where reasonable people disagree.

So what does it mean when someone asserts that something is true?

The first point is that whatever thing it is that is claimed to be true, that thing is some kind of abstracted representation of the world. It might be expressed in language, mathematical theory, a story, a picture or something else. In any case, it is something that provides a model or picture of what some part of reality is like - necessarily in far coarser detail.

By claiming something is true, what we are asserting is that reality matches or corresponds to what we are claiming to be the case. However, given our limitations as humans and our habitual lack of direct access to reality,² this translates to something more complex in practice.

Asserting that something is true should be more accurately read as a multiple part claim that reflects our best efforts to know what is the case. I would parse this as the following: in my (or our) judgement, the relevant abstracted representation, when understood correctly, faithfully portrays the aspect of reality it purports to represent within the limitations of the representation; and I (or we) are convinced that anyone else who looks at the situation in sufficient detail would come to the same judgement.³

Adopting a position of epistemic humility leads naturally to a two part explanation of the concept of truth. What we claim when we use the concept (correspondence to reality) is different from how we come to and justify that claim. Human limitations mean that we don’t have direct access to reality and mediate knowledge and truth through abstracted representations. While this doesn’t produce a clean or mathematically elegant explanation of the concept truth, it hopefully rings more true to real life. Truth is contested, hard and often we turn out to be wrong.

1. Try reading the content of any academic journal to see this in action!

2. We are leaving aside direct first person experiences, like pain, for the sake of simplicity.

3. Or at least any reasonable person acting in good faith.