Let’s talk about “the truth”. In a world where misinformation and “fake news” are ever present in public discourse, it’s a bit of an understatement to say it’s a really important concept. Additionally, you might know that truth is very important to me so this topic is very dear to my heart. So it comes as a surprise even to myself that the definition of it seems to be less straight forward than additionally thought.
Trying to define what “truth” means is surprisingly hard. There seems to be something called “the truth” which is independent of language, perception or interpretation and prescribes the content of any abstraction. I like to call these abstractions “models” because that’s what’s used in the field of statistics to describe objects that aim to replicate the behavior of a “true” system that is fundamentally unknowable. Models can be better or worse depending on how good all of their predictions are. A prediction is good if it matches observations. Observations are pieces of information about the unknowable “true” system which we can express within the domain of our model.
And that’s where the trouble starts because in reality, we don’t have any observations of the truth. As human beings we observe everything through our senses which is already an act of modeling before it reaches our conscious mind. Thus, we only ever interact with the truth through models. It’s an important part of what we do to transcode these bits of information in various ways such as language to form new models. But all these models have one thing in common: they can never be completely accurate. This is an inherent property of models, they never fully capture the true system. So an important conclusion is that nobody can ever access the full, actual, truth. All anyone has is a model.
So if that were what we meant with the word “truth”, nobody could ever rightfully call anything “true”. We need a softer definition for what people mean when they say “that is true”, and the most straightforward interpretation would be “that fits with my model of reality”. It sounds fine at first, but then you realize that people might have different models of reality so by this definition there is not one single unique truth. This form of truth is actually subjective.
Naturally, it seems like a bit of a no-go to call truth subjective. It has the same ring to it as “alternative facts”. Everyone knows there is one single truth (the one we talked about earlier) so a definition that makes it subjective is unacceptable. But note how there’s two different concepts here that we both called “truth”. There’s the one that people mean when they talk about it and there’s the one that, as we just argued, is the only one they are capable of talking about. There’s the true truth and the model of the truth. Words are models of concepts. The word “cat” is not a cat, but because we agreed on the meaning of that word it is well suited to represent a cat in a sentence. This works for nearly any concept, but the concept of truth is a special case in which it doesn't. It describes a part of the modelling process that implies its core property (uniqueness) which would have to be present in any useful model. Yet a fundamental limitation of language, i.e., that the definition of words is never unique, prevents such a word from existing. It follows that the word “truth” is not that word.
So as much as it pains me to admit this, truth is subjective. If someone says that something is true, they are always referring to their own model of reality. A model which is never fully accurate. So when a scientist goes through the full process of forming a hypothesis, conducting an experiment and interpreting the data to conclude that something is true, their use of the word isn’t any more valid than a religious person who says that it’s true that their messiah came back from the dead. Both have their respective models of reality which support their respective claims. One might be more useful than the other, one might make more accurate predictions, but that’s not what the word “truth” is about. It is about an object that both individuals think is unique but they are using a word which actually represents one that isn’t.
The takeaway from this is a profound one: you can never use something being true as an argument. If someone asks you why you believe that the earth is round, it’s tempting to answer “because it is true”. But this is not an argument, it’s just a reformulation. The two sentences “I believe the earth is round” and “it is true that the earth is round” have exactly the same meaning. So the other person has as much of a right to say “it is true that the earth is flat”, because it just means “I believe that the earth is flat”. It’s a level playing field. Just accept that neither of you can claim the truth because it is out of your reach. Just focus on how useful and how accurate your respective models are.
Especially if you are a scientifically minded person, you might be surprised to find out how useful and accurate some models are that you’d never have considered otherwise.
