The Trouble With Objectivity

Not that long ago I wrote that:

Knowledge is a peculiar concept. It's a form of belief, really. We believe in things and some things that we believe in, we feel so sure about that we call it knowledge.

This is a pragmatist perspective on truth. Elsewhere I've written:

Facts have careers. They're born, they ascend, they decline, and they die. This isn't to say that facts don't exist. It's just that facts about the world don't merely sit about waiting to be 'discovered'. Facts are constructed to fill a need, and the dimensions of the world that these facts reveal to us will reflect less the world, and more the need.

My point is that, in most cases we never really have access to the truth. There's a mundane explanation for this: that we're not learned enough in a given subject to get at the truth, so we just trust the word of specialists. But this isn't really what I mean either. I'm talking about the fact that, as humans, we literally do not have access to the truth. We don't know what dark matter is, or what's beyond the event horizon of a black hole, because our perceptions and the tools we build to augment our perceptions simply can't handle those aspects of the world.

Instead, what we do is we map what we do know about the world based on what we need to achieve in the world. Facts are not really truths, but reflections of our worldly needs.

Not everyone agrees with me though. In fact, there's a lot of people who are rather obsessed with getting at 'objective' truth. I think that, for the most part, these people are confused about what they're doing. Let me tell you why.

¶The objectivity obsession

A great example of truth-seekers are those attracted to the Effective Altruism movement, or the newest iteration of the rationalist movement fronted by sites like LessWrong or Astral Codex Ten. To quote the LessWrong about page:

LessWrong is an online forum and community dedicated to improving human reasoning and decision-making. We seek to hold true beliefs ... the more rational you are, the more likely your reasoning leads you to have accurate beliefs

In theory, this is a pretty sensible motivation. Of course we'd like to be more accurate than less accurate. This particular concern is age-old. I've noted before that:

Plato described emotion and reason as the horses that pull us in two directions. Isaiah, the biblical prophet, encouraged us to come together and reason, rather than act without thinking.

Fifty years ago we had movements like post-modernism facing off against movements like radical behaviourism. More recently, we have popular books like The Scout Mindset, which speaks of the difference between thinking like a 'soldier'---a mindset in which we are motivated to defend our beliefs rather than being open to changing them---and thinking like a 'scout'---elevating curiosity, and unbiased truth-seeking. Or the endless pop-science books telling us to get out of the amygdala, and into the frontal lobes. Less formally, we have the constant calls for fact checking and the battle against an 'increasingly post-truth' world.

My issue here is that there are two distinct concerns in all this that are frequently confused.

The first is that many people have an intuition that there is a correspondence between our thinking and behaving and some objective external reality. This correspondence might be more-or-less aligned with that reality, but there is a reality to which we can align.

The second is that our internal biases and emotionality often has us make decisions that are rash and impulsive. That, with enough self control, we can get past these distracting passions and make more collected and sensible decisions.

Many people capably concern themselves with one or the other---philosophers, scientists, and other people with the time on their hands to explore these ideas. But for the average person, there is this sort of confused combination of the two which leads to the misguided idea that there is:

some idealised version of ourselves that lives emotion free---flesh made automata---that we're trying to achieve.

And because we have this kind of sense, we're terrified of being wrong. It might indicate that our emotional biases have carried us away from this ideal. And so we overrate the search for 'truth'. But we don't really spend the time to figure out just what that truth is.

¶What is truth

There's certainly room to optimise for accuracy. We might have to build something that requires physical facts about the world for example. But in many practical cases, this emphasis on corresponding to some external fact about the world isn't what's important.

Maybe the example of this that we encounter the most frequently in our day to day life is when we're doing things that require us to assess the conscious experience of others. There's no way to produce facts about the lived experience of someone else. No one will ever be able to tell me if I experience green in the same way as you do. Despite the thousands of words spilled on the issue, we have to take the experience of others on faith. Importantly, it doesn't actually matter what, if any, the objective truth of 'green' is. What matters is only what the other person thinks it is.

You see, we very, very rarely need to engage with the question of whether there is or is not some objective external reality. It's usually irrelevant. Let me give you another example I like. To quote me:

The world that humans live in is a very different world to the one birds live in. Humans are trichromats: because our eyes have three kinds of photoreceptor, we have three colour channels. There are, thus, three dimensions to our colours. We can see light on wavelengths that correspond to red, green, and blue. Birds are tetrachromats. They have four dimensions to their colour vision. They can, for example, see ultraviolet light. We cannot.

Being able to see ultraviolet---or rather having four-dimensional colour vision---is not quite the same thing as being able to see a colour you've never seen before. We can get a better flavour of this if we add our sense of time to our colours. Rather than pink, we might also have 'fast pink' or 'slow pink'. Something like this better approximates the world that birds live in.

Is the bird's perception of the world more true? Is it useful for me or you to model ultraviolet as a dimension of visual perception when it doesn't serve this purpose in our lives? In some circumstances, yes. If you need to know things about birds. But day-to-day, probably not so much. There might well be some objective truth to be had regarding colour, but usually, working out what are 'true' colours and what aren't is going to be an exercise that depends very much on just what it is you're planning to do with that information.¹

¶Outro

I think, for most people, the solution is a simple one. Split the concerns and focus on what each actually means.

When we're talking about being 'rational', stop confusing it for seeking out 'truth'. Really, what you're doing is trying to make less narrowly-biased decisions associated with passionate emotions and more expansively-biased ones associated with dispassionate emotions.

When we're talking about truth, stop confusing it for aligning with some kind of objective, external reality. Really, what you're doing is orienting yourself towards some particular need. There's no reason that this truth has to be unbiased. In fact, it's almost certainly going to be. A fact about the world that reflects some of it, and ignores the bits that aren't relevant.

If we do this, then objectivity stops seeming like such a desirable goal. It's not that we're accepting a "post-truth world". It's an argument for being clear about what the truth we're orienting ourselves towards is concerned with---the thing we're trying to achieve. Because this, more than anything else, is what matters.