Archive for the ‘atheism’ Category

What would constitute a proof of God?

Monday, October 19th, 2009

One argument that theists sometimes bring up in arguments with atheists is "What proof would convince you of the existence of God?" the implication being that the existence of the universe and all the wonderful things in it is proof enough.

Well, if you define God as just being the things we don't understand (currently the big bang and the mysteries of the human mind) then that's a valid argument, but the resulting God is just the "God of the gaps" who has been shrinking rapidly as science improves. Pretty much everything else in the universe (excepting only a few relatively minor details) we have good working scientific theories about. I think eventually we'll come to understand scientifically both the human mind and the big bang as well (in fact, I think it will not be possible to understand either without the other).

Most theists don't seem to believe in just a God who created the universe at the beginning and then left it alone - they believe things like "praying works". This suggests a simple test involving praying for (for example) heads in a coin toss and then seeing if there is any statistically measurable effect. Once an effect is found, the experiment could be refined to determine which religion and sect has the most effective prayers. Theology would become a science. Theists will usually claim that prayers don't work that way, but ultimately they either work or they don't, and if they do work then that effect can be observed and experimented on. I understand some such experiments have been done, and have shown no statistically significant effects with the possible exception of medical patients who know they are being prayed for. This can be attributed to the placebo effect.

Another example of such a possible proof comes from the observation that "if God is so great, why does he keep needing money to fix church roofs?" I would find it a very compelling piece of evidence towards God's existence if consecrated buildings did not suffer the same kinds of wear and tear that unconsecrated buildings do.

What does supernatural actually mean?

Thursday, October 8th, 2009

I was discussing philosophy with a theist friend recently and the argument "this only applies to natural things and God is supernatural so this doesn't apply" came up. I've seen this argument in other debates as well, but I have to confess that I don't completely understand it. What does "supernatural" actually mean? The dictionary definition that seems to best apply is "unexplainable by natural law or phenomena".

There's two possible meanings to that. One is "unexplainable by the laws of physics as they are currently known" and the other is "unexplainable by the laws of physics even if we knew them all".

The existence of supernatural things of the first type is not denied by any sufficiently well-informed scientist - it's no secret that the laws of physics are incomplete. One possible example might be the details of the event horizon around a black hole - at small scales this requires a theory of quantum gravity, which we don't yet have. I think we'll eventually eliminate all such supernatural things by having a complete theory of physics.

I suspect theists would therefore prefer the second definition. But does this definition even make any sense? What would it mean for there to be phenomena in our universe for which no physical theory could be described to explain? Well, supernatural phenomena of that sort of can be said to exist too - when a quantum variable is measured and the wavefunction collapses, the result isn't necessarily determined by anything in the universe. But I don't see any theists suggesting that God acts on the universe by deciding how each and every wavefunction collapse occurs, despite the apparent omnipotence that such power would grant. I suspect this is because this would eliminate any possibility of human free will - quantum wavefunction collapse is part of all physical processes, so controlling quantum wavefunction collapse would mean controlling all our thoughts and actions. There would be no will except the will of God possible, making all religion (and indeed everything) rather pointless.

It seems to me there are philosophical reasons to reject the concept of a non-random supernatural process - if something is non-random there is some sort of (at least partially) predictable pattern to it, which means one could come up with a law of physics to describe that pattern, which means it's no longer supernatural. "Predictable" only means predictable in principle, though, not predictable in practice. Chaitin's constant (the probability that a random program will halt if run for long enough) for example isn't random but is uncomputable in the sense that only a finite number of its digits could be determined by any finite algorithm. Curiously, this number could be thought of as omniscient - it encompasses all mathematical knowledge (since it can be used as an oracle to solve the halting problem) but a number (even a real, uncomputable one) doesn't seem like it could match the theists' descriptions of God as having certain properties like compassion and goodness.

As well as the gaps in our knowledge of physics and the gaps caused by quantum improbability, there are also gaps due to the fact that there are some real world phenomena which we just can't do experiments on for one reason or another. We can't do experiments on UFOs because we can't predict when and where they will show up (though I'm sure that if one did show up in a suitably equipped science lab, laws of physics could be found to describe it).

Another thing we can't do physics on is subjective experience, simply because it's subjective. We don't currently have any technology by which one person can experience what it's like to be another person (and even if we did, there is no objective way to be sure that it's the same experience for both people - one can't compare subjective things with objective things). All we can do is ask people to report on their subjective experiences, and a personal report isn't as reliable a piece of evidence as a repeatable experiment.

Each one of us can't even be truly sure that other human beings actually *have* subjective experiences - maybe they're just p-zombies who say they do. It's a useful working hypothesis to assume that they do, though (and the opposite assumption would be rather dangerous for all concerned).

So perhaps God is Himself a subjective experience. That certainly dovetails with some things that theists say, like "I know God is in my heart and I've experienced His love, but I have no way to prove that to you". And objective evidence of God does seem to be rather thin on the ground, to put it mildly. If this is what God is then I am a teeny bit jealous of theists for having that experience that I never had (even when I was a theist, went to church, prayed regularly etc.).

If neuroscientists are able to generate religious feelings in others by stimulation of their temporal lobes with magnetic fields then, objectively, that suggests that religious experiences (at least these ones) are "fake" in the sense that there is no "objective God" causing them. But on the other hand, why should we treat subjective experiences as less "real" than objective reality? What could be more real to someone than something they have experienced first hand? By that logic, hallucinations caused by drugs or schizophrenia should also be considered "real" in this subjective sense.

Objective reality, on the other hand, consists of the things that we can in principle (given sufficient experimental data and suitable application of logic) convince other rational human beings of - in other words the things that we can in principle (if we're honest) all agree upon. As such, only objectively verifiable things should be used as a basis for public policy.

Disproof of God

Wednesday, July 22nd, 2009

It's pretty easy to prove that God didn't create the universe, given just a couple of very uncontroversial postulates and the definitions of the words "universe" and "God" as most people understand them.

  • Define the causal closure of a point in space-time X to be X plus the causal closure of any points that could influence X or be influenced by it.
    Define the universe (which we'll also call the "L-universe") as the causal closure of planet Earth as it is today. (If you dislike the use of "planet Earth" or "as it is today" in this proof, you can substitute it for some other subset of the universe that is alleged to be created by God.)
  • Postulate that if A created B then A influenced B. This is a pretty trivial postulate - creation of something is obviously a kind of influence over that thing.
  • Postulate that creators cannot create themselves. This is also pretty trivial - the concept of creation of an X implies that there is a time before the X exists and a time after which X exists. The creator of X must exist in both of these times, but the creation can only exist in the latter.

Suppose that the universe was created by God.

  • This implies that planet Earth was created by God (planet Earth is part of the universe).
  • This implies that God influenced planet Earth (creation is a sort of influence).
  • This implies that God is in the causal closure of planet Earth (definition of causal closure).
  • This implies that God is part of the universe (definition of universe).
  • This implies that God could not have created the universe (creators are not part of their creations).
  • Which is a contradiction. Therefore, the universe did not have a creator.

This is a formalization of the common "If God created the universe, who created God?" argument but sidesteps the possibility of a creatorless God or a God created by another God by including all such Gods in the larger L-universe.

This suggests that to believe in God, one must have a different definition of "universe" (call it "S-universe") which is a subset of the "L-universe". This brings us to the real value of this proof - any argument for the existence of God that doesn't distinguish between the L-universe and the S-universe must be invalid, because to be true it would have to apply to the S-universe (for which there can be a God) but not to the L-universe (for which we have already seen that there isn't). Some examples of such arguments:

  • We don't know how the universe was created, so let's just define God to be whatever created the universe.
  • The universe seems well suited to our needs.
  • Anything of sufficient complexity must have had an intelligent creator.

Evidence against God

Sunday, August 24th, 2008

A commenter recently suggested that if I wanted to believe in God and went looking for the evidence, I would find it. Nope. Tried that. It didn't work. Back at university I engaged in some great debates with theists, and I really wanted to believe in it. But a dispassionate look at the evidence really does favour the atheist point of view.

The first piece of evidence that Christians will point towards is the Bible. "It's a historical record" they will say. Evidence it may be, but it's not good quality evidence - it's all second hand. Stories passed along from person to person, distorted by time and translation and plucked from historical context. Let's contrast this with the evidence for quantum mechanics (which seems similarly ridiculous at first glance). This is real evidence - experiments which you can do to convince yourself even if you think the original experimenters lied about their results. Experiments which have been repeated all over the world and verified many times. That's real evidence. A book (even a very popular one) is not good evidence.

The next thing Christians will try is sheer weight of support - "two billion people can't be wrong" they say. Well, of course they can - there are also two billion people who are Muslims and Hindus, who the Christians will say are wrong. And another two billion people are either not religious or believe something other than "the big 3". So whatever the truth is, at least two thirds of the people are wrong about it. And even the 2 billion Christians can't all agree on which particular sect is the right one. Number of people is not good evidence.

The third thing is "the power of prayer" and religious people have lots of anecdotes about how prayer has helped particular people. Anecdotes are not evidence though - they suffer from selection effects. When theists pray for something and it comes true it is seen as evidence for God but when they pray for something and it doesn't come through it "must not be God's will for this to come about". If you do actual scientific, statistically accurate analysis of the power of prayer it turns out to have little or no effect. Any effect it does have is not proof of God anyway - it could just be a form of placebo effect.

The fourth is "religious experiences" - people who claim to have seen or touched God, or to have had very powerful feelings of being close to a magnificent, benevolent power. I used to find this argument extremely compelling (and was quite jealous that I hadn't experienced it myself) but I have since discovered that similar effects can be induced in the brain with magnetic fields or drugs. Like the power of prayer, this is something else that is "entirely within the mind". Chemical imbalance in the brain seems to be a much more likely explanation for these experiences than God.

These seem to be the main arguments. There are a number of minor ones as well but they mostly seem to have logical flaws like taking their conclusion as an initial assumption.

There are some very compelling arguments against God as well. The main one is Occam's razor - positing the existence of God doesn't actually explain anything (not even the creation of the universe, as God Himself then has the same origin problem). So we can get a theory of the universe that is simpler (and therefore more likely to be correct) by not including God.

Another thing that I found very convincing is the geographic distribution of different religions. People tend to believe the same things as their parents, friends and neighbours. This suggests that religion is passed on from person to person like a meme rather than having any intrinsic truth. Dawkins poses an evolutionary explanation for this strange human behavior - there is an evolutionary advantage to believing what your parents tell you. This behavior started out as a way for advantage-giving wisdom like "don't eat these particular berries" to spread to one's offspring but because historically we had no way of verifying these memes, incorrect but (mostly) harmless memes like "pray to the great invisible being for healthy crops" also sprang up and took advantage of the same mechanism.

A third argument against religion is how contradictory it is. Surely if there was a God and He wanted us all to behave in a particular way, he would have made the bible very coherent and hard to misinterpret. It is none of these things, as we see from the number of people who misinterpret it to forward their own agendas constantly. In fact, religious dogma seems to be particularly effective as a tool for powerful ruling classes to exert control over the general population, which suggests that it evolved to be that way for exactly that purpose.

Atheism vs agnosticism

Saturday, August 23rd, 2008

I used to describe myself as an agnostic, but now I describe myself as an atheist. What changed was not my faith (or lack of it) but what I understood those words to mean.

I used to think that the meanings were as follows:

Atheism: I believe there is no god.
Agnosticism: I neither believe nor disbelieve in God.

But I know think that the following definitions are more accurate:

Atheism: I do not believe that there is a god.
Agnosticism: I have no opinion about the existence or nonexistence of God.

Specifically, I don't think that there is a god but would rather avoid describing myself as having any sort of "belief" one way or the other, because that word carries some implication of "belief without (or despite) evidence". While it's impossible to ever prove with certainty that there is no god (or indeed to prove with certainty anything other than "I think, therefore I am") I think that there is a great deal of evidence that there is no God. I'll take a look at some of this evidence tomorrow.