When I started this blog, I had plenty of ideas floating about, and knew that the only way to verify their congruence was to write them down, possibly in some recognisable order. The result is surprising, at least to me. This post is an attempt to summarise what I’ve claimed so far and to verify if the different pieces do fit in a coherent discourse.
The starting point:
In my first post, I’ve declared that I believe in three founding concepts.
1. Reality exists and is unique.
2. Knowledge and beliefs can’t be distinguished ontologically, and can always be described as “models of reality”, as such they are never perfect.
3. Striving to improve these models is inherently good.
If you strongly disagree with these premises, you may want to leave this blog straight away. It’s likely that we’ll never find a way to understand each other.
I have also declared that I’m trying to figure out how to best spend my life, or, metaphorically, that this blog represents my attempt to “write my own user manual of life“. This aim finds some justification in premises 2 and 3. My beliefs are mine, I can try to improve them, but I can only judge this improvement from my own personal perspective. The result may or may not work for me, and that’s all I can asses; whether it works for you, that’s for you to find out.
How to travel:
Trying to be rigorous, after explaining where I’m starting from, the next move is to investigate how am I planning to move. This sent me into a surprising epistemological quest, aimed at identifying the method that underlines my thoughts. This was done mostly a-posteriori, trying to figure out why I think the way I do, but I suspect that this process influenced my own method as well.
So far, it all works: I’m starting from a strictly subjective stance, minding my own business and not making any generalisable claim, am I? Well, not really, because premise 1. is generic (applies to everyone), and not surprisingly, it is the key that allowed me to move forward. The first step was to explore the limits of knowledge, and see if one can claim anything at all. At first sight, if one can’t distinguish between knowledge and beliefs, and if anyway all knowledge is certainly and inescapably incomplete and somewhat flawed, one would guess that “everything is subjective” and that claims about truth are all unjustified. To some extent, this still holds, but premise 1 allows to go a little further. This can be done after observing that all beliefs find their raison d’être inside the subject that holds them, in the form of the purpose that the belief itself serves for the believing subject. As a first approximation, I’ve claimed that each belief serves one or more purposes, and that holding one belief is justifiable if the belief can be used to bring some good to the subject. I’ve also said that when one sees beliefs as models of reality, it can be generically assumed that they all share at least one purpose: it should be possible to use any given model to make predictions.
I realise now that this is problematic. What if not all beliefs have a purpose? Or, what happens when one piece of knowledge can’t be used to make any prediction? In other words, it is possible that our brain may contain parasite information that is inherently useless? My answer is “Yes”, this is certainly possible and I will have to address this issue on a separate post.
For the time being, I have concluded that: different beliefs can be compared and contrasted, by comparing and contrasting the predictions they allow to produce. Or, if you prefer, if two beliefs seem to be justified by the same purpose, one can try them out and see which one is more fit for purpose. This is possible because of premise 1: there is only one reality that can be used to measure how fit for purpose any belief is.
The surprise so far is that I’ve been able to find why it is possible to make objective claims even if all knowledge that informs them is subjective and imperfect. This is why the title of this post mentions “Subjective Science”: even if I’m starting from the (obvious and discouraging) realisation that all knowledge is in fact opinion (is subjective, belongs to the subject that believes it is true), I hope to have shown how we can use reality to obtain some (derived) knowledge about knowledge, and that in this case, we can have some hope to use it to make objective claims. The noun “Science” in my definition is supposed to imply this aspiration towards objectivity, while also accounting for the role of the one and only reality, and hence for the value of empirical verification. On the other hand the “subjective” attribute is meant to be a clear indication of where this epistemology starts, and to clearly indicate that whatever claims this science may produce, they apply to the claimer.
This is all very Cartesian, in some way. “Cogito ergo sum” is the archetypical claim of the “Subjective Science” that I’m trying to define. As a side note, I wish to reassure my readers that this doesn’t mean that my journey will lead to Cartesian dualism. It doesn’t and I know it doesn’t because I’ve already written my thoughts on consciousness (sorry, article isn’t public yet).
Anyway, the moment I insert the word “Science” in my search for a trustworthy method of enquiry, well, all I can do is accept the “scientific method” and be happy with it, right? I wish!
Turns out that I can’t, mostly for two reasons:
a) The mainstream understanding of “scientific method” doesn’t really work. It’s invalidated by silly simplifications about universal truth, the existence of “scientific facts” and similar preposterous nonsense. Once one accepts the idea that all knowledge resides inside our brains/minds (yes, I’m implying the two are one and the same) and that therefore all knowledge is incomplete (at best), there is no going back.
b) Apparently, nobody really knows what this Scientific Method is. At least, nobody can provide a unique, rigorous, unequivocal, and universally accepted system to separate science from pseudo-science. There are theories, opinions, more or less widely accepted views, but nothing more. And hey, guess what? That’s what the theory I’ve built so far predicts!
After registering this first encouraging sign (my model does already produce verifiable predictions), the only way forward is trying to define the Scientific Method in a way that is compatible with my subjective premises (this has two meanings: subjective in the sense that the premises are my own, but also in the sense that knowledge itself is subjective). The result is another series of posts, that introduced my metaphor of solidity. Real science is what tries to become more and more solid, that is: tries to produce more trustworthy predictions and/or increase the precision of such predictions. All the rest is unsubstantiated fluff (or intellectual flatulence, as I like to define it).
The first destination: I like the panorama.
There are quite a few consequences of the definition above, they are worth some explicit discussion.
My favourite observation is that it is now possible to declare that all the discussions about the fallibility of science, or about the fact that “scientific models” get dis-proven and discharged, are nothing more than fluff. A model is valid if it produces trustworthy predictions. Hence, this model can be (and normally is) used to produce such predictions and that’s the end of the story. One can also predict that every model will eventually reach (asymptotically) its own predictive limits, and that when this happens, new “better” models may indeed emerge. These new models will allow to produce brand new predictions and/or significantly improve the reliability of the previous model predictions. In the second case, the old model should be discharged, but not because it has been dis-proven. It should be discharged because the new model is more useful, more fit for purpose. However, we already know that also the new model isn’t objectively “true”, and that we’ll hopefully one day find a new one that is even better.
For example, lots of people are keen to point out that Newton was wrong and that his theories have been surpassed. I can now say “so what”? If you want to predict if a project for a new car frame is good enough (the car produced will be solid enough), to produce your calculations will you use quantum mechanics? Relativity? No, you’ll use “classic physics”, even if we all know it’s “wrong”, nobody gives a toot about this “wrongness” because classic physics is good enough for the job at hand and that is all that counts.
The second observation is that I’ve been building a model of how knowledge works. And that this model applies to itself, as it postulates that all knowledge is some sort of model. Fine, so I can make predictions about the model itself. First prediction: this model is wrong. Because it is a model, it doesn’t account for all details of reality, so it can’t be expected to work always for all conceivable situations. Second prediction: someday, someone will produce a better model that can make more and more precise predictions. For now, the main purposes of this model are two: to guide me in understanding how I want to live, and to help me to discriminate between science and fluff. We already know that it does work for the second purpose, so it remains to be seen (in the following posts and in the next paragraph) if it does work for my personal quest. If it does, I’ll use it.
Third observation: my initial premises look more and more solid as I proceed. We’ve seen before that we need to postulate that reality is one and one alone, for practical reasons. If there are more interchangeable realities, and/or if the rules that govern it do change arbitrarily (as when a god produces a miracle for unfathomable reasons), then knowledge can’t exist, and all intellectual efforts are worthless, without exception. In other words, if I want to think that I can produce something other than fluff, this premise is unavoidable. On the other hand, accepting that all knowledge is a form of belief does not imply that all knowledge is worthless, as discussed before, here and below.
If the worthiness of knowledge depends on its subjective usefulness, then seeking knowledge for knowledge’s sake (my third premise) is intrinsically a good thing because it’s a direct way to find or generate new useful concepts, even before knowing or hoping how exactly the new knowledge may be useful. It shortcuts the problem of unknown goods and allows to discover new useful things even if one ignores their possibility. Furthermore, as we’ll see below, it allows to make these goods transmissible, extending their usefulness to the widest possible theoretical audience.
Finally: this model also generates a new definition of generalisable knowledge. After all, we all know that we share lots of beliefs, we all agree about the basic features of reality (I’ve never seen anyone trying to pass through a solid door because s/he expects it to be immaterial), and this makes all my claims about the non-existence of “objective truths” seem quite dubious. Luckily, this observation doesn’t really clash with what I’ve proposed so far. Yes, some models are simple, reliable, and shared amongst more or less all human beings (and plenty of animals). That’s because some features of reality are simple and (for us) easily modelled in predictable ways. We all share some beliefs because their usefulness is clear to everyone. The moment this usefulness is questionable, someone will disagree.
From this observation I can close another circle: in one of the premises posts, I’ve asked myself “why should I have something to add [to any subject]?“, and I now have an answer. Because the ideas I’ve produced are useful, they explain why and how one can productively share ideas (by making it clear and verifiable how they are useful), why and how we need a scientific method, and why and how the scientific method can produce agreements, even if all knowledge is imperfect and subjective. I have, in other words, a philosophical argument that allows me to accept and surpass cultural relativism. And if you don’t believe that this promises to be useful, well, why are you still reading?