In Defense of the Fine Tuning Design Argument (2001)
The Internet Infidels have long been interested in publishing a theistics defense of the fine-tuning argument against our criticisms, which are collected among our many essays on the Design Argument in general. James Hannam is the first to oblige, and the following essay is his defense of the Fine Tuning Argument for a creator. The Internet Infidels will respond in a forthcoming essay by Richard Carrier.
Of the traditional arguments for God, the teleological or design argument has recently enjoyed a new lease of life after many years out of favour. The classical formulation comes from Natural Theology, a book by William Paley published in 1802, in which the author asks us to compare living things to a watch he has found on the common. Because we can all agree that a watch is purposefully designed to tell the time, we should be equally able to conclude that plants and animals are purposefully designed to survive in each of their particular habitats. The argument is far older than that and appears in different forms in the work of both St Thomas Aquinas and Rene Descartes. Charles Darwin put paid to this formulation of the argument in 1859 with his theory of evolution which explained how living things came to be. All was quiet on this front for some time, but today teleology is back in fashion as an argument from the fundamentals of physics and, less successfully perhaps, biology. This essay outlines the new argument from design based on the fine tuning of the laws of physics.
One thing a design argument must not do is look for a ‘God in the Gaps.’ That is to say, we must not try and find theological explanations for scientific phenomena that science itself has not yet explained. The classic example of this is the origin of life on Earth. Many Christians have been beguiled by the idea that because life is so improbable, God must have actually stepped in to have created it. I reject this for two reasons–the first is that I do not think God, having created the universe for living creatures to live in, would have done such a bad job as to make it near impossible for such creatures to arise. Secondly, we are held hostage by the possibility (in my opinion, probability) that one day a science will create life in a test tube. Hence I feel Christians should avoid the origin of life in apologetics especially as I believe that many efforts in this field are misinformed.
Notes on probability
Design arguments are often based on and challenged by assertions about the nature of probability. It is therefore worth saying a few words about it without delving too far into the complicated mathematics. The trouble with talking about probabilities of life or of a finely tuned universe is that we presently have a sample of one–our Earth and our Universe. Can we make any valid statements about probability from this? Yes, we can, but we must be careful not to overstate our case or understate our assumptions. The most useful of the probability theorems for the argument from design is Bayes’s Theorem. This is a mathematical formula that links different probabilities but when explained in words it begins to sound like little more than common sense. Essentially it says the following:
- If we know an event has happened and want to investigate the likelihood of other events we can ask whether these other events make what we observed more or less probable.
- Thus we have seen a volcano erupt. We want to compare two theories about what is under the earth–either it is full of molten rock, or it is hollow. We ask which of these hypotheses makes a volcanic eruption more likely.
- Seeing the lava pouring out of the volcano we might consider that this observation makes our hypothesis that the earth is full of molten rock more likely.
Note that Bayes’s Theorem stated like this is evidential, as it does not deal in proofs. Even the strict mathematical form is plagued by the difficulty in assigning numbers to its various terms. We can also argue about which hypothesis makes our observed data more likely and essentially the whole argument can boil down to subjective opinion. However, it does allow us to make some probabilistic observations about what we do not know from something that we do.
Next, two fallacies related to probabilities need to be examined as they often impact on arguments from design. I have called these the specific fallacy and the lottery fallacy.
An example of the former is when we are asked why we are not amazed by the incredible odds that we were born given the number of eggs and sperm produced by our parents. Is it not amazing that I am not someone else? This is fallacious because we have no way to distinguish between any of the results (the different people we could have been). However, if all those sperm harboured a genetic illness that I did not in fact inherit, then we could tell after the fact that something odd had happened. We might consider that one sperm had spontaneously mutated so as to not have this genetic illness and I had happened to have been born from that sperm. We might consider the postman. But we would certainly start looking for explanations for this remarkable occurrence.
The atheist attacking the design argument will try to make it look like a case when we cannot tell the results apart. He will claim that there are vast numbers of ways we could come to be, on many different planets and maybe in many different universes. He will point to the possibility of non-carbon-based life forms and say that if they were us we would no doubt be just as amazed. In fact, the design argument is a case where we can clearly distinguish between results. It does not depend on the kind of life we are or which galaxy we appeared in, it depends on the existence of life at all. The only results it considers are that life does not exist, which, if the argument works, is unlikely, and that it does exist, which is in fact the case. Thus we may postulate however many different kinds of life are conceivable, but they are still a near-miraculous occurrence compared to the overwhelming likelihood that life would not exist. Or in the immortal words of Richard Dawkins, “however many ways there are of being alive, it is certain that there are vastly more ways of being dead.”
The lottery fallacy is when we have such an enormous sample size that the unusual event must happen at least once so we should not be surprised when it does. To use the lottery fallacy against the fine tuning argument we must postulate a suitably large array of universes for which we have no other evidence at all. This isn’t a fatal problem because we are also postulating a Creator but persuading an atheist that he is standing on the same metaphysical ground as his theistic opponent can be rather hard.
The fine tuning of the physical constants
That the laws of physics are such that life can appear is indisputable. Let us also assume for the moment that the vast majority of possible laws of physics are not so hospitable. The only objection at this point is that there is only one possible set of physical laws, but as we can easily perceive alternatives and even construct models of them this objection seems to me to fail. As we are investigating why the laws of physics are as they are, the answer ‘because they are’ does not seem to take us very far forward and indeed, begs the question.
Let us consider a reason why the laws of physics might be as they are, using our common sense version of Bayes’s Theorem. Let us call our hypothesis X: the hypothesis that the conditions for life were intelligently arranged rather than otherwise. Under what conditions can we consider this hypothesis? When we conclude that the laws of physics are more likely to be as they are if it is true than if it is not. Although this hypothesis introduces a host of other questions, we are, in the spirit of scientific enquiry, taking one step at a time. After all, there were those who disputed with Newton because he could tell us nothing about what gravity actually is, but I would not be among them.
Another X we might consider is that this universe is the present result of a line of cosmic evolution where universes bud off each other after subtly mutating and that somehow their propensity to reproduce themselves is related to their propensity to be a home for life. One example of this idea suggested by Lee Smolin is that of universes being born through black holes. Black holes require heavy stars and various other conditions to exist which are also responsible for the creation of those heavier elements that are needed for life. As we shall see later, I fear that cosmic evolution, although it might explain some physical laws, falls well short of improving the odds for the finely tuned laws we see to an acceptable level.
The last alternative sometimes suggested is that a multitude of universes exist and we got lucky by being in the right one. This is becoming increasingly popular and suggests that more and more atheistic scientists such as Martin Rees in Just Six Numbers are at least accepting that fine tuning is real and are considering a non-theistic explanation.
The usual atheist response to fine tuning is to try and subject it to death by a thousand questions. They will ask how we know the universe could be any different from what it is, how we can possibly know that this is the only universe intelligent life could exist in, and why we are so arrogant to assume that only our sort of life is possible. A lecture on how we do not understand probabilities is also likely at this point as is a demand that we prove that the creator means the God of the Bible. I will deal with the various objections to fine tuning in something like the order that they might be expected to arise in any debate, but like all arguments for God, this one does not seem to convince anyone who does not want to be convinced.
Is not ours the only possible universe?
It is certain that our universe is the only one of which we have any knowledge but we cannot from this reasonably postulate that ours is therefore the only one possible. Indeed, an argument would have to be constructed to show that any other universe was logically impossible and this argument would then have to be shown not to apply to ours. Aristotle did insist that only our world existed and this was part of his worldview that Christianity successfully overthrew in the march towards modern science. For the atheist to claim that no other universes are possible needs to be backed up with rather more than the observation that this is the only one we can see. A more philosophically-minded individual might claim that the fact we cannot know anything about these other universes (with whom I would agree) is tantamount to saying they cannot exist. But this is a mistake. It would be bold enough to take the view that only what actually exists can possibly do so, without going further out on a limb to proclaim that only what is known can exist.
Must not all universes be like ours?
In his dreadful book Creation Revisited (out of print), Oxford chemist Peter Atkins suggests that the laws of nature might be reducible to logical principles that must hold true. In other words, any universe must have the laws of physics that ours does. Leaving aside that it is impossible to even reduce mathematics to a set of logical rules, Atkins would need to derive every physical constant and rule from, as he puts it, elegantly reorganised nothing. He uses the least time theorem of light propagation as one neat example but declines to attempt anything more difficult. His proof about light amounts to showing that if it has the properties it does then it will act in the way that it should. Why it has those properties is rather more of a mystery.
Although we cannot show any reason why our universe should have the physical laws that it does we have no mechanism as to how they could be any different either. This comes into play when one wants to postulate multiple universes with differing physical laws so that there is scope for one of them to be just right for us. There is a distinction between asking why our universe has the properties it does while not accepting that it had to be this way and asking how the postulated multiple universes all came to have the variations of those laws so that they can be an explanation for fine tuning.
Is the universe really so finely tuned?
This used to be the first line of defence and there has been a good deal of discussion of exactly how fine the tuning needs to be. Given that we cannot make our own universes we are not in a position to be certain that most sets of physical laws would produce a barren creation. However, we can model universes and physical laws and here the evidence becomes much more persuasive than a simple argument from incredulousness. Using this universe as our base case, it does appear that small changes in constants result in rapid breakdown of many of the chemical and physical processes required for life without any indication that something else is taking their place.
This is best illustrated by considering a very simple universe that we can both analyse fully and where we can compare a full array of different physical laws. In chapter 7 of his magnum opus, Darwin’s Dangerous Idea, Daniel Dennett explains the workings of John Horton Conway’s ‘Game of Life.’ This involves a very simple two-dimensional universe which is governed by simple rules that can generate a huge degree of complex and interesting structures. But what is fascinating about this is that the particular set of rules used is the only one that does anything remotely exciting. All the other possible rules produce dead and boring universes. By extension, the much more complicated life in our universe would seem to be the product of a much more exact balancing act of laws and constants. The latest book on the question is Martin Rees’ Just Six Numbers and this does suggest that the tuning is really quite exact. As I have mentioned, Rees is an atheist who does not really consider that the universe might be designed but instead moves straight on to the multiple universes hypothesis.
We cannot know that our particular set of physical laws and constants are the only ones that will produce a viable universe. There may be others and other possible life supporting universes. What we can say, however, is that the number of viable universes compared to the number of dead and desolate ones is probably infinitesimally small. Most thinkers recognise this and call the various factors that are just right for us to exist the anthropic coincidences. So although these coincidences represent a highly unlikely set of circumstances, we do not need to claim that ours is the ONLY possible life supporting universe.
Another point put forward by the atheist is that we cannot imagine what strange life forms could exist in an alien universe and hence we cannot make any judgement on whether or not they are likely. This is simply an appeal to ignorance (something theists are often accused of) and one can respond to it by saying that the fine tuning argument is valid according to the current state of scientific knowledge. We are all happy to modify our position if more information becomes available. After all, that is the only way to reach conclusions from the evidence.
Victor Stenger, an atheist physicist from Hawaii has tried to demonstrate that many different constants produce viable universes while simultaneously insisting on the multiple universe hypothesis as a ‘sensible’ alternative to design. If seriously challenged over his computer program “Monkey God” he would no doubt claim it was just a bit of fun. However it makes a serious polemical point for him that needs refuting.
- First, it assumes exactly the same laws of physics that we have but with different numbers. Most fine tuners would say that the laws themselves have been finely tuned and so cannot be taken as read.
- Second, he has examined only a tiny number of the effects that the constants he uses have and in particular he makes no effort to consider chemistry.
- Third, he only uses four constants whereas even Martin Rees insists there are six. With more added, the number of possible universe will increase exponentially.
Could there not be multiple universes and we are in the right one?
Both Stenger and Rees are atheists and subscribe to the multiple universes hypothesis. This states that there are zillions of universes, each with slightly different properties, and we just ended up with one in which we can live. After all, says the atheist, smiling disarmingly, if the universe was otherwise we would not be here to see it.
The problems with the multiple universe theory are manifold but the most important is that we have no evidence for them whatsoever. They are not necessary as a consequence of any other physical theory and there is no theory that predicts they might exist. Also, for the multiple universe theory to help the atheist at all, the universes must all have different physical laws and no one has any idea why this might be. Finally, the vast number of universes required seems to insult every principal of scientific elegance from Ockham’s razor onwards. Sir John Polkinghorne, retired Professor of Physics at Cambridge University, is not alone in doubting the motives of proponents of the theory as they seem to be happy to suggest anything that avoids having to think about the possibility of God.
The atheist should realise that hypothesising multiple universe is metaphysics and not science. It is not a scientific theory because it cannot be experimentally verified or falsified and neither can it be considered superior just because it is naturalistic. Once we move into metaphysics the naturalistic assumption of science must be done away with as it is no longer either justifiable or useful. Indeed it is a metaphysical statement itself–as it lies behind science, it cannot be examined scientifically.
Could our universe not have evolved or formed from another universe?
I have previously mentioned Lee Smolin’s ideas about cosmic evolution and must confess to quite liking them. They do not have anything much to say about ultimate causes and neither do we have any mechanism by which this evolution would be possible. However, the idea has, I think, more merit than the multiple universes hypothesis, as it is both far more elegant and more susceptible to scientific investigation. If we learn enough about black holes to suggest that they really can give birth to new universes and that these may be subtly different from ours then this would be a powerful piece of evidence for cosmic evolution.
To complete the theory we would also need to find some connection between our own existence and the ability of black holes in this universe to produce new ones. Otherwise there would be no reason why universes good at producing ‘offspring’ would necessarily be good at producing us and this means the anthropic coincidences would be no nearer being explained. This is because ‘successful universes’ good at reproducing would not be the ones we would expect to find ourselves in. As such our existence would be even more unlikely, as the direction of cosmic evolution would be working against our appearance.
Another interesting idea was put forward by Carl Sagan when he suggested that powerful evidence that the universe had a designer would come from our learning to do it ourselves. Although we must reluctantly leave this speculation in the realms of science fiction in which it belongs, many a late night drinking session could be sustained on such musings.
Who is this creator and what is he up to?
It has been alleged that as we cannot know how the a creator went about his work or what his motives are, the design argument is useless. This would also have put paid to most scientific advances as when Darwin first proposed evolution he did not have a clue what the mechanisms for it were and when Newton explained why the planets move as they do he brought up another unknown–gravity–which he was happy to call an occult property. If the universe appears to contain design then the fact is that we can ask about the explanation for that without needing to move on to a further stage of conjecture in advance.
So does fine tuning prove that God exists?
No. The considered position of scientists is that fine tuning in the form of the anthropic coincidences is real but that a creator is by no means the only explanation. Even if intelligent design were proven it would still not necessarily mean that the Judeo-Christian God was involved. Further information may improve the strength of the argument. For instance, if we find that our laws of physics provide an easy pathway to the origin of life that requires them to be exactly as they are, this would be further evidence for an intelligent creator. So would the discovery of life on other planets with the same or a very similar genetic architecture to our own, as this would suggest that different pathways to life are not common.
We have examined the claim that the laws of physics seem to have been set up so that the conditions for life are possible in our universe. We examined some alternative possibilities but found that at the moment design remains one of the best. Richard Dawkins once told A.J. Ayer that evolution was what allowed him to be an intellectually fulfilled atheist. I would suggest that to truly reach atheist nirvana, Professor Dawkins has a good deal more explaining to do.