Acting as though you believe something is true.
All the way back in Essay #5 we said that, according to the Enlightenment Vision of science, science had to be certain: Any ‘science’ worthy of the name had to provide knowledge. And any ‘knowledge’ worthy of the name had to be certain.
In 1877, William Clifford wrote what has since become known as ‘Clifford’s Principle’:
“It is wrong always, everywhere, and for anyone to believe anything on insufficient evidence.” 
Of course, “wrong” here, does not connote factual error. It is entirely possible to believe something on insufficient evidence and, by happy accident, to believe something true. (For example, “I believe the earth orbits the sun because the voices in my head told me so”). No; “wrong” here connotes moral error. According to Clifford’s Principle, one cannot — one must not — believe a thing for which one does not have sufficient evidence. To do so would be a moral failing.
At the time and place that Clifford was writing, the Enlightenment Vision of Science reigned supreme. It was believed that certainty in knowledge was both necessary and possible. To believe something that you did not know was lazy, slipshod. Any negative consequences arising from avoidable ignorance were as morally bad as those from actions done deliberately and knowingly because (so the argument went) there is no need for ignorance.
“Yes officer, I shot him in the head. But I thought the gun wasn’t loaded.”
“Did you check it wasn’t loaded?”
“Then you are responsible for his death because you did not check that the chamber was empty. You did not know.”
That seems reasonable.
Unfortunately for this argument, as those of you who have been following this series of Essays will know, the Enlightenment Vision did not make it through the 20th Century unscathed. (Or, more strongly, as discussed in Essays #6, #7, #8, #10, #11, the Enlightenment Vision of science is now known to fail utterly in almost every respect.)
By the end of the 20th Century, Richard Feynman could write,
“It is impossible to find an answer which someday will not be found to be wrong.” 
Michael Polanyi expressed the same sentiment as Feynman when he insisted,
“We must commit ourselves to the risk of talking complete nonsense, if we are to say anything at all.” 
The aim of the next few essays is to consider what we should then do. If Clifford says that we should not believe without sufficient evidence, and Feynman says we can never have sufficient evidence, what shall we conclude?
— Is Clifford wrong? Should we believe even on insufficient evidence?
— Is Feynman wrong? Can we ever know we are right?
— If they are both right, should we resolve to never believe anything?
— If we never believe anything, on what basis shall we act?
Over the next few Essays we shall find that (spoilers!) not only are scientists permitted to believe things on insufficient evidence; they are required to believe things on insufficient evidence. More than that: they cannot keep these beliefs to themselves as private ideas; they are required to act on those beliefs.
For this essay we shall start by looking at the situation in which we find ourselves when we try to not believe things.
The third option listed above — to refrain from believing anything — seems to provide a relatively simple way out.
At first glance, this seems to be the way science works. Maybe my detector pings because the wave-function collapses. Maybe it pings because of a many-worlds style branching universe. In the absence of definitive evidence one way or the other, I shall remain agnostic. I do not know and, if asked, I shall say that I do not know. I do not need to know. And I cannot know. Surely intellectual integrity forces such a stance on me.
In this view, we can still say things. For example, we can say that, “According to the Copenhagen interpretation of quantum mechanics, the measurement caused the wave-function to collapse.” But we do not need to believe anything. No commitment is necessary.
This also seems to work in religion. Maybe there is one God who sent His son to die. Maybe there are many gods with their own concerns and battles. In the absence of definitive evidence one way of the other, I shall remain agnostic. I do not know and, if asked, I shall say that I do not know. And I do not need to know. And I cannot know. Surely intellectual integrity forces such a stance on me.
Again, we can still say things. “According to the Christian worldview, God loves me very much.” But we do not need to believe anything. No commitment is necessary.
Agnosticism in action
Not committing to anything looks very nice on paper. But it rapidly hits major difficulties when you need to put agnosticism into action. Let us draw on a nautical illustration.
You are in a boat, far out at sea. You have had a pleasant voyage so far, but the weather turns and you are hit by a ferocious storm. The wind buffets the boat this way and that. Waves crash across the deck. You run to the lifeboat and get ready to lower it into the foaming waters.
But wait! Do you know the lifeboat will keep you safe? Do you know it is seaworthy? It has a sticker on saying it was checked for seaworthiness a month ago. But maybe something happened in the last month. You saw someone rowing in it just two days ago. Maybe that shows that it is seaworthy, and the sticker can be believed. Or maybe, when they rowing, they damaged it. Maybe their actions made it un-seaworthy, so the sticker cannot be believed. You do not know that the lifeboat will save you, so you rush to the life-ring.
But wait! You read somewhere that there were sharks in these waters. Maybe the sharks would kill you even if the storm didn’t. But surely no shark would be feeding in this weather, so maybe you would be safe. Until the storm died down and the sharks came back for easy pickings. Unless the sharks in this region don’t like the taste of human flesh. You do not know that the life-ring will save you, so you rush to the centre of the boat and clutch the mast.
But wait! What if the ship itself goes down? A squall snaps the mast in two. The ship’s timbers creak as they are wrung this way, then that, by the tempest. But maybe the storm will pass before the ship is inundated. Unless the ship survives structurally but capsizes and you are eaten by the sharks. You do not know that the ship will save you. So…
You hunker down, clinging to the stump of the mast, and try to think. You call to mind an image of your science teacher, and wonder what she would say. In your mind’s eye, the apparition stares down at you. “Repeatability is key!” she declares, waving an admonishing finger. “Set up three hundred identical storms. In one hundred storms, you should take the lifeboat; in one hundred storms, you take the life-ring; and in one hundred storms you must remain on the ship. Compare the number of times you survive in each set of conditions and that will tell you what you should do!”
“Great!” you shout above the roar of the storm, along with some expletives that are lost in the wind. “What do I do right now?!” The spectre of A.J. Carlson comes to your mind: “You do not know what to do. You cannot know what to do. So do not do anything. Do not take the lifeboat, for it may not be seaworthy. Do not take the life-ring, for the sharks may get you. Do not stay on the ship, for it may sink. Do not! For you have insufficient evidence.”
However hard it may be to concentrate with the sound of a storm raging about you, you know that it is not possible to take Carlson’s advice. It is not simply undesirable; it is not possible. You have to do something. If you decide against trusting yourself to ship, lifeboat, or life-ring, you could instead jump into the sea hoping to swim for it (again, without any the certainty that you will survive). But that is still doing something. But you cannot do nothing.
You may not know that the ship is a good place to be. You may not even believe that the ship is a good place to be. But if you stay hugging the mast, you act as though you believe that the ship is a good place to be.
Part of this “acting as though you believe” is committing to the course of action you have chosen. You cannot push off in the lifeboat and also cling to the mast. As you hold on to the mast and watch the lifeboat get washed overboard, you are committed to the mast. Even if you are wrong. If you decide that maybe the lifeboat was the better option after all and you throw yourself into the sea to swim after it, you commit to the action of swimming after the lifeboat. Even if you do not have decisive evidence that the lifeboat can save you. And even if you do not, in your heart of hearts, believe that the lifeboat will save you. You act as though you believe it.
What differentiates the true believer who is convinced that the ship will save them and clings to the shattered mast, from the agnostic who neither knows nor believes that the boat will save them, and clings for dear life to the shattered mast anyway? The latter may be agnostic, but he acts like he is not.
So much for the perils of being lost at sea. In the next essay we will consider how this plays out in the practice of science and the practice of religion.
 W.K. Clifford, The Ethics of Belief (1877).
 Richard Feynman, The Pleasure of Finding Things Out (1999). While Gödel and Quine provide support for Feynman's epistemic pessimism with regard to knowledge obtained by reason and observation, there is some debate about whether knowledge obtained by revelation faces the same limitations. This was discussed in Essays #23a, #23b, #23c, #23d.
The results of Gödel and Quine notwithstanding, logicians who are reading this may aver that Feynman’s claim is self defeating: If it is true that every answer may someday found to be wrong, then this, too may someday found to be wrong. I accept this argument unreservedly, but it is irrelevant. I do not claim that Feynman’s position is correct, but rather that we should accept it. If you find it strange that I should argue in favour of accepting a position which I suspect may be false, read on; the rest of the essay may clarify that position. In the words of Polanyi, the aim of this essay is to understand “why I should believe that which I might reasonably doubt.” (Michael Polanyi, Personal Knowledge, 1958).
 Michael Polanyi, Personal Knowledge (1958).
 William James, The Will to Believe (1896).