I have always had trouble with the idea — which I think of as typically Christian, though many non-Christians also subscribe to it — that our beliefs are freely chosen in the same way that our actions are, and that they are therefore something for which we are morally responsible.
One problem is that the “choices to believe” which I am supposed to be making all the time are, for this observer at least, invisible to the eye of introspection. When it comes to actions, I have a very strong subjective feeling of choosing my actions, and of being able to choose otherwise than I do. However metaphysically problematic that idea may be, it is an unshakable subjective conviction. Either humans really do have free agency, or else they are subject to powerful and inescapable illusion that they have free agency. (After some vacillation, I have decided on the former, as anyone who has been reading my recent posts will know.) When it comes to beliefs, on the other hand, I have no such subjective experience. I believe what I believe, and if I have freely chosen to believe as I do, these are choices of which I have no direct knowledge. The idea of free choices which are made without the chooser’s knowledge is obviously paradoxical.
A second troubling question — the one which this post will be addressing — is why anyone would want the freedom to choose his beliefs — what the point of such freedom might be. Freedom of action is desirable because there are many different good things any given person can do, and it is not possible for him to do all of them. Therefore, there is no one right answer to the question “What should I do?” It is necessary to choose. When it comes to beliefs, on the other hand, all possible true propositions are mutually consistent in a way that all possible good actions are not. Therefore, there is one right answer to the question “What should I believe?” — and freedom to believe otherwise is nothing but the freedom to be wrong. It is a worthless freedom which can do only harm. Furthermore, the freedom to believe as you choose — that is, the “freedom” to be ignorant and to have incorrect beliefs — is detrimental to the freedom that really matters, the freedom to act. If I have an accurate map — i.e., one whose content is forced on me by reality — I am free to go wherever I want. If, on the other had, I am “free” to draw my own map, unconstrained by the actual layout of the territory I wish to navigate, I lose the freedom to choose a destination and go there, gaining only the “freedom” to get lost.
Perhaps the best way to approach this is through the map metaphor just introduced — one which I have been using for years as an argument against the desirability of freely choosing one’s beliefs. While it seems obvious that a map dictated by reality is more useful than one we freely make up, it is equally obvious that in fact we do want to choose what kind of map to use, and that there is no One True Map which is objectively better than any other.
Every map is necessarily incomplete. For one thing, it has only two spatial dimensions and, while it persists in time, it does not change; the territory it represents is extended in three dimensions and changes through time. The map also has fewer “dimensions” than the territory in the sense that each point on the map has only one distinguishing characteristic (namely, color) but represents a place in the territory which has any number of characteristics (temperature, altitude, soil type, population density, etc.); therefore only one — or, with some ingenuity, a few — of these characteristics can be portrayed on any given map. Every map is also physically smaller than the territory it represents, which necessitates many omissions.
In addition to being incomplete, every map necessarily contains distortions and inaccuracies. The most inevitable of these are those which result from using a flat surface to represent the surface of a sphere — resulting in the distortion of directions and/or proportions. Other distortions may be necessary depending on what is being mapped; all information has to be “translated” into the language of colors on a two-dimensional surface, and some information doesn’t translate very well.
Translation itself offers another useful metaphor. All translations are also necessarily incomplete and distorted. A terza rima translation of Dante necessitates a great deal of semantic distortion; but a “literal” translation in prose distorts the work’s fundamental character as a poem. And of course any conceivable translation will involve the near-complete loss of the phonetic content of the original. There is no one true translation of Dante any more than there is one true map of the world. Translations and maps cannot even be objectively ranked according to how closely they approximate this unrealizable ideal of perfection. It’s not a quantitative question of how much accurate information a given map or translation conveys, but a qualitative question of what information. Which map or translation is “best” for me depends entirely on my purposes and on what kinds of information I consider most important for those purposes.
Any representation of the world in a finite mind is going to be incomplete, as inevitably as any map or translation. Perhaps our situation regarding possible beliefs is not really all that different from our situation regarding possible actions. While it may be true in principle that all truth may be circumscribed into one great whole, in practice we mortals are no more able to assent to all possible true propositions than to carry out all possible good actions. Thought, no less than action, requires time and effort, of which we have but a finite supply, and so knowing the truth in one area entails remaining ignorant in another — or believing something false because it “works” well enough for our purposes and because the truth is more complicated. It also seems highly likely that there are some aspects of reality which our minds simply cannot model correctly — just as a flat map simply cannot accurately portray the surface of a sphere.
In the realm of action, in order to do good in the only way that limited beings are capable of doing good, we have to be able to do bad. Anything good we can do will involve failing to do some other good, and often even doing something positively bad. If we are to make omelettes, we need the freedom to break eggs. Given that freedom, though, we also become free to go around breaking eggs just for the hell of it, without making omelettes. The freedom to do evil as such is undesirable, but for limited beings it is a necessary side effect of having the freedom to do good.
Something similar may be true in the realm of belief. In order to have any “true” beliefs — that is, workable approximations of truth, such as finite minds are capable of — we need the freedom to ignore and distort certain truths. (If a cartographer is strictly forbidden to depict anything untrue, he cannot draw a map at all. A translator who cannot lie cannot translate.) With that freedom, though, necessarily comes the freedom to ignore and distort even the most vital of truths — i.e., the freedom to be wrong, even disastrously wrong.
What specific truths ought we to ignore or distort, and which ones are non-negotiable? There is no one best answer to that question, since it all depends on the individuals interests and goals. Hence the desirability of freedom.
So choosing out beliefs is, after all, desirable — for essentially the same reasons that choosing our actions is desirable. There remains the question of whether and how it is possible to choose one’s beliefs. The true seems, almost by definition, to be that which, when properly understood, compels belief. If one really believed P (a given proposition) to be true, it seems that it would be impossible to consider it an option to believe not-P instead. I have recently made some headway on this question, too. (Again, choosing beliefs turns out to be a lot more similar to choosing actions than I had realized.) But that is a subject for another post.