by Ken MacVey
One argument for the existence of a creator /designer of the universe that is popular in public and academic circles is the fine-tuning argument. It is argued that if one or more of nature’s physical constants as mathematically accounted for in subatomic physics had varied just by an infinitesimal amount, life would not exist in the universe. Some claim, for example, with an infinitesimal difference in certain physical constants the Big Bang would have collapsed upon itself before life could form or elements like carbon essential for life would never have formed. The specific settings that make life possible seem to be set to almost incomprehensible infinitesimal precision. It would be incredibly lucky to have these settings be the result of pure chance. The best explanation for life is not physics alone but the existence of a creator/designer who intentionally fine-tuned physical laws and fundamental constants of physics to make life physically possible in the universe. In other words, the best explanation for the existence of life in general and ourselves in particular, is not chance but a theistic version of a designer of the universe.
Here I will discuss problems as to why I don’t think the fine-tuning argument works. Before I start, one caveat. You don’t have to be a non-believer in theism to find the argument doesn’t work. In fact, particularly trenchant technical arguments for rejecting versions of the argument have been presented by evangelical Christian analytic philosophers Timothy McGrew and Lydia McGrew.
Another caveat. I am not a professional philosopher, physicist, cosmologist, biologist or mathematician. I am a litigator and trial lawyer. But one of the fun things about being a litigator and trial lawyer is you get to depose experts in depositions and cross exam them at trial. Sometimes when I give presentations to other lawyers on how to examine experts, I point out that lawyers don’t have to be experts in the relevant field but they should aspire to be experts on how to deal with experts. That means not being cowed by jargon, arrogance, or credentials while being ready and able to challenge the logical and factual foundation of the expert’s opinion.
Here I will list a series of problems for the fine-tuning argument that individually and cumulatively challenge the logical and factual foundation of the argument. This is only a sample and is not meant to be definitive. For easier reading, I have used headings to identify these problems.
The Evidence/Explanation Problem
Most empirical explanations involve an explanation supported by independent extrinsic evidence. For example, a coroner may conclude after an autopsy the probable cause of death was a heart attack. The fine-tuning argument is not like that. The evidence for the explanation is the thing to be explained. There is no independent evidence of the supposed fine-turner to explain fine-tuning. Rather the fine-turner’s existence is inferred from the alleged extraordinary fact of fine-tuning, and this fine-tuning is the thing to be explained. This alleged fine- tuning is in turn explained by the theism hypothesis, which supposes the existence of an unobservable supernatural being causing the unobserved act of fine- tuning.
Explanations which are solely supported by the thing to be explained can be problematic. They can be circular. “Oh wise one, why is there a drought? Because the gods are angry. Oh wise one, why do you say the gods are angry? Well, there’s a drought, right?” They can invite speculation based on no independent evidence. Why did the same guy get to win millions in the lottery three separate times? The lottery had to be rigged. Or the Luck Fairy must have blessed him. What’s your evidence? The fact that the guy did win the lottery three times, which means he was unbelievably lucky. It can get worse. Explanations only supported by evidence for the thing to be explained can result in gross miscarriages of justice. A woman was wrongfully convicted in the UK of murdering her two children each of whom was originally reported to have died from sudden infant death syndrome. An expert testified that the odds of a child dying from SIDS was one out of a little over 8500. He went to say the odds of having two children die from SIDS would therefore be 8500 times 8500, or close to 73,000,000. There was no independent extrinsic evidence that she murdered her children or consideration that multiple SIDS events could have a common cause other than homicide (such as genetics). This has become known as the prosecutor’s fallacy. Philosopher of science Colin Howson has argued the fine-tuning argument is in part based on this fallacy.
The fallacy is that you can’t conclude a hypothesis is probably true just because the outcome predicted by the hypothesis was highly likely if the hypothesis were true and the predicted outcome did in fact happen. Go back to the three-time winner lottery situation. Yes, the hypothesis that the owner had been blessed by the Luck Fairy to win three times if true would make that outcome highly likely to occur. But that does not mean the Luck Fairy hypothesis is probably true simply because the guy won the lottery three times. The Luck Fairy hypothesis had a very low probability to start with.
When an argument rests solely on a calculation to explain the thing to be explained, it flashes in red: proceed with caution. Indeed, the tables can be turned on the fine-tuning argument by a different probability calculation to show the opposite that the universe is not fine-tuned for life. That’s next.
The Definition/Measurement Problem
One key question is how fine-tuning is to be defined and measured. The fine-tuning argument relies on a micro approach focusing on physical constants at the subatomic level to define and measure supposed fine-tuning. There are problems with this approach we will get to. But right now let’s consider a macro approach as an alternative to define and measure fine-tuning.
The fine-tuning for life argument cannot escape from the situation that it rests only on two established facts : the existence of only one universe and the existence of only one planet with life. We know there is at least one universe, namely the one we are in. And to date we know of only one planet in that universe that has life, namely the one we live on, earth. So was the universe made to be fit for the existence of life or was life by natural process an adaption to its immediate surroundings ? (To borrow from Bertrand Russell, was the nose made for glasses or glasses made for the nose?) How do we answer this question? How do we go about defining and measuring the degree of fine-tuning of life and determining the direction of life’s “adaptation” to the universe?
Scientists usually can make multiple observations of recurring events under natural or controlled circumstances to test hypotheses. Here, scientists can’t experiment and create new universes to test hypotheses. Nor can they directly observe any fine-tuning or any supernatural being—any claim of fine-tuning is totally based on inference, not direct observation. We just have to go with what we can infer from what we can observe. What observational data do we have? As already mentioned, all we have is just one instance of a universe (ours) and just one instance of a planet with life, earth.
Looking at it on a macro basis with these limitations in mind, the proposition that the universe was fine-tuned for life appears weak. Estimates vary but cosmologists have estimated that there may be several septillion planets in the universe (a septillion is the number one followed by 23 zeroes). Right now the only confirmed evidence for life is one planet out of several septillion. That is not strong evidence that the universe is fine tuned for life.
True, it is possible there is life on other planets but to date none has been confirmed. Recently, the media covered the claim by two scientists that they detected biosignatures supporting the existence of extra-terrestrial life. But other scientists have rejected the study and others have argued the biosignature approach generates false positives. And some astrobiologists support the “rare earth” hypothesis that holds biological sentient intelligent life should be extremely rare in the universe.
The fact remains we actually know of only one planet out of potentially septillions of planets that have life, i.e. earth. This is hardly compelling evidence that this incomprehensively vast universe is fine-tuned for life. But the key take-away here is that how fine-tuning is to be defined and measured is a matter of choice, not necessity, which takes us to the next problem.
The Speculation Problem
For some there is something compelling about the scientific aura surrounding the fine-tuning for life argument. The argument rests on certain counter factual manipulations of the mathematics of physical constants to demonstrate life is extremely improbable without fine-tuning of physical constants. The late physicist Vic Stenger contested these calculations in a number of articles and books. He contended, for example, that certain calculations arrive at their results by changing only one constant leaving the others unchanged to show that stars and hence life could not form. But if other constants are allowed to vary as well, Stenger alleged they could compensate so that stars would form. (He created a computer model you could play with to make this point.) University of Michigan physicist Fred C. Adams in an exhaustive analysis published in 2019 concluded that the constants supposedly set for life could vary significantly and make room for life as we know it in a wide range of varied physical constants.
Are these physicists right or wrong? I am absolutely unqualified to say. And so are the NY Times columnists and many non-scientifically trained philosophers who promote a version of the fine-tuning argument. But one thing for sure is if we had to go to trial on the fine-tuning argument, the lawyers for the argument would have no problem lining up an expert to opine that the universe is remarkably fine-tuned for life, and I would have no problem finding an expert opining the opposite. This is because experts will have ample room to use criteria and methodology of their picking to define fine-tuning in a way that supports or rather, “fine-tunes,” their respective and diametrically opposed reasoning and conclusions. This is a result of the fact that the fine-tuning argument is a peculiar kind of argument in which the evidence for the explanation is the thing to be explained and the explanation is based on an unobservable event (the act of fine-tuning) caused by an unobservable agent.
This argument is an invitation to engage in wholesale speculation. Case law and evidence codes have developed criteria over the last century for screening out what is sometimes called “junk science.” If the fine-tuning claim were to go to trial, the standard Kelly-Frye and Daubert case law tests for the admissibility of expert scientific testimony in court (which looks at whether the opinion is based on methodologies generally accepted in the scientific community or various factors such as whether the methodologies can be tested for reliability) a reasonable judge would be compelled to reject the admissibility of such testimony as speculation.
The Target Problem
When fine-tuning arguments are made sometimes it is not clear what the supposed target of the tuning is. Is it life in general (including cockroaches or the 99% of species on earth that are now extinct)? Is it intelligent conscious biological life (specifically us)? In any case it appears many versions of the fine-tuning of the constants really focus on what the constants have to be for carbon as an element to exist. This is then conjoined with the fact we don’t know of any form of life that isn’t carbon based. Fine-tuning for carbon is transformed into proof of fine=tuning for life.
But how do we know the presupposed designer/creator really intended? We don’t have any direct observation or verifiable communication with the designer—if we did we wouldn’t need the fine-tuning argument. David Hume in his 1700’s classic Dialogues Concerning Natural Religion pointed out we know only what a designer or agent’s intentions are by experience with that designer or agent or other similar designers and agents. We have no experience with the presupposed designer of the universe so we are not in a position to really know what the designer intended.
If the universe is fine-tuned for carbon, does it mean it was fine-tuned for life? Why not instead that it was fine-tuned for diamonds? Diamonds are almost exclusively made of carbon. According to Japanese mathematician Toshikazu Sunada, the crystalline structure of diamonds has unique mathematical symmetrical features which account for their aesthetic properties. He said he could mathematically show that there are only two crystalline structures out of an infinite number of possible crystalline structures that would have these mathematical symmetrical properties. (He did not know if the second crystal actually existed in nature.) If Sunada is correct, then the possibility of having a crystalline structure matching the symmetrical/aesthetic properties of diamonds are two out of infinity. That makes diamonds extraordinarily unique. So how do we know that the universe was not fine-tuned for diamonds to actuate an object of extraordinary beauty and uniqueness and that life just happened to be an incidental by-product? All the mathematical calculations that are supposed to demonstrate the improbability of life also could be used to demonstrate the improbability of diamonds. (This is true even if Sunada is not correct.) The existence of diamonds is just one example of the target problem. Biologist J. B. S. Haldane once famously said that the creator appeared to have been very fond of stars and beetles. Maybe those were the targets. Or maybe life was the creator’s way of making artificial intelligence agents possible—biological life was just a bridge.
The point is, no mathematical calculation by itself is going to establish the goals and objectives of an unobservable, unique agent with characteristics beyond our understanding. We know the purposes, goals, and objectives of agents by experience and observation.
The Who, What, How and Why Problem
The fine-tuning probability calculations tell us nothing about the who, what, how or why of the fine-tuning of the laws and constants of the universe. We use these laws and constants as our tools in explaining things. But we have no tools to explain the tools. We have no tools to explain what the fine-tuner is made of; who the fine turner is; how the fine-turner does the fine-tuning; or why the fine-tuner does the specific tuning.
What is the fine-tuner made of? Is it something physical? No because what is physical is defined by the constants that are supposed to have been fine-tuned. Is it incorporeal as many claim? How can something incorporeal do any physical work? Does the fine-turner consist of something else? What could that be?
Who is the fine-turner? Maybe a committee of supernatural bureaucrats? Maybe an infant deity like Hume playfully postulated? Maybe a mad scientist in another universe? The mathematics on the supposed improbability of life gives no guidance on how to answer this question.
How was the fine-tuning done? To say the fine-turner was omnipotent or very powerful is not an explanation any more than explaining sleeping pills work because they have soporific properties. How could omnipotence possibly work? Are there laws for omnipotence? Are there any mechanisms? What are they? Where did they come from? What are they made of? You might as well say the constants were tuned by magic. Magic is not an explanation. Invoking “magic” as an explanation is another way of saying we don’t have an explanation.
Why was the fine-tuning done? This takes us back to the target problem. But it goes beyond that too. Even if we could determine that life was the target of fine-tuning that does not answer why life was the target.
Hume in his An Enquiry Concerning Human Understanding noted that we can infer causes from effects when we have had experience with a pattern of causes and effects. But here we have no relevant pattern to consult. We have here a claim that the universe was fine-tuned by a single and unique cause by a single and unique agent when we have no observational experience with either. No surprise then the probability calculations that the fine-tuning argument is based on are useless in answering the who, what, how, and why of fine-tuning.
The Under-Tuning Problem
At the same time, there is an under-tuning problem. Why wasn’t the university tuned for more life? Physicist Fred C. Adams in 2019 in his exhaustive analysis not only calculated that life could exist under a wide spectrum of variations of physical constants, he calculated life could also could have been much more prevalent if only certain physical constants had varied by infinitesimal amounts. If the universe is fine-tuned for life, it is not tuned to maximize life. And if it is fine-tuned for life, why is the universe so overwhelmingly hostile to it? Life as we know it cannot exist in outer space or on stars or planets like Jupiter. Cosmologists have various theories on how our universe may end but few to none contend life as we know it will survive.
On the other hand, if the target of the fine-tuning was intelligent sentient life, namely us, the initial fine-tuning appears to have been botched. The current scientific thinking is that dinosaurs died out due to a massive asteroid (or maybe a comet) hitting the earth 66 million years ago, which then allowed mammals to become more dominate. If that didn’t happen it is hard to see how we or other similarly intelligent sentient beings would be here. So did the universe need additional fine-tuning to make sure an asteroid hit earth at just the right moment?
The Necessity Problem
Why is fine-tuning for the existence of life even necessary for an all-powerful supernatural being? Such a being could create beings adaptable to any environment. This is particularly a problem for those who are substantive dualists believing the mind is separate from the physical body. If so, there is no need to fine-tune for biological life for minds to exist. They can be created as non-physical entities compatible with any environment.
The Testability Problem
The argument is presented as a scientific based empirical theory. The problem is there are no proposals for how the fine-tuning for life explanation can be tested and independently confirmed or disconfirmed. This is a consequence of the fact that the evidence for the explanation is the thing to be explained and the explanation consists of an unobservable agent (a fine-turner external to the universe) and an unobservable event (fine-tuning of the universe). This pushes the claim outside of scientific verifiability, which is based on empirical tests that can help confirm or disconfirm hypotheses. As discussed next, it also puts the explanation proffered by the fine-tuning argument outside the realm of being mathematically or scientifically modeled to test for logical consistency and empirical predictions.
The Jumping Off the Probability Space and Modeling Problem
The fine tuning argument is a mathematical/physics modeling argument. It argues that based on mathematical/physics counterfactual modeling of existing physical constants life is extremely improbable. It is assumed with little or no explanation that these constants can vary the way postulated. And from there it is concluded that the hypothesis of a supernatural agent is more probable than mere chance under naturalism.
This leap to a supernatural being is a non-sequitur. The argument relies on physics and mathematics to define the probability space and then jumps off this space to an allegedly more probable explanation that is not based on any physics or mathematical modeling. No matter how improbable an event is in a given probability space in order to compare probabilities in different spaces you need to find a way to put them in a common space to make apple to apple comparisons.
But how can you do that? How could you even mathematically and scientifically model the probability of the existence of an unobservable agent and unobservable causal event that is not subject to any known laws? Of the infinite number of off-the- probability grid explanations, including an infinite number of potential supernatural beings, how do you mathematically model them and how do you mathematically select one from that infinite array? After all, isn’t it a remarkable coincidence of all the infinite choices of gods and other explanations you could speculate on as an explanation for a supposedly fine-tuned universe happens to be the one we like? Isn’t it remarkable that all of the infinite versions of gods or other extraordinary beings that are logically possible, we got the one that happens to like our particular universe? As philosopher Keith Parsons has observed, if anything is fine-tuned, it’s the selection of the right supernatural designer out of an infinite set of possible designers. So how do we explain that fine-tuning? Do we need a fine-tuner of fine-tuners?
The Biased Selection Effect Problem
There is an implicit assumption underlying the fine-tuning which is rarely articulated or justified. It’s not the mere fact that the universe contains a unique feature that is highly improbable which drives the fine-tuning argument. No one, for example, argues that our solar system must have been fine-tuned so that earth’s moon would be exactly at the right dimension and distance to allow us to observe the sun’s corona during a full solar eclipse, a situation which is very fortuitous for earth astronomers studying the sun. What makes the fine-tuning argument seem so compelling is that we put a special value on life, especially intelligent sentient life. It seems like it just can’t be a lucky coincidence that something so valuable and so singularly and incredibly improbable exists.
Let’s see what happens if someone started with a different presumption of value. Suppose a group of anti-natalists believe life inevitably entails suffering that outweighs life’s value and therefore it would be better to never have been born. (There are contemporary philosophers who take that position.) And further suppose they also believe in the metaphysical value of beauty and that diamonds are the most beautiful objects in the universe. They could easily fall for the fine-tuning logic based on exactly the same calculations made for the fine-tuning for life argument but conclude instead that the universe was actually fine-tuned for diamonds, not life. (See my prior discussion on the target problem.) That is because they have a different set of values as to what they consider to be of ultimate value, which then tilts them to prefer a different explanation. In other words, an empirical claim is driven by non-empirical value judgments.
The fine-tuning argument in a way treats the universe like a deck of 52 cards. In any truly randomly shuffled deck of cards a specific order of cards would unlikely be repeated even after a trillion shuffles. (52 x 51 x 50 etc. results in an extraordinarily large number.) There is nothing amazing about that fact. It’s when we find a particular order of cards significant (like dealing a royal flush in poker which has only has a one out of 650,000 chance of showing up) that we marvel at the improbability of the shuffle and may even wonder whether the shuffle was rigged. That is because some orders of cards in games we play have particular significance to us, especially when money is at stake and card sharks are known to cheat. The fine-tuning argument for life treats the universe like a deck of cards where the same poker player wins the pot three times with three royal flushes in one night. Likewise, life, particularly sentient intelligent life, especially matters to us as sentient intelligent beings. We then assume that somehow life’s existence in some way had to be rigged, that somehow life must matter to some unobservable supernatural agent just like royal flushes matter to poker players and card sharks. That does not follow. We have experiences with poker players and cheats. We don’t have experiences with unobservable super beings that cause unobservable events.
Our source for saying what is valuable is actually us—Protagoras said “man is the measure of all things.” It is not surprising that we find sentient intelligent life to be remarkably valuable—in fact, no more surprising than if we encountered a narcissist who thinks he is uniquely handsome and intelligent and a product of providence. (Please don’t misconstrue my point–I believe it is appropriate to tailor our values based on the kind of beings we are.)
The fine-tuning argument is similar to p-hacking. P-hacking can occur where an experiment is run to test for statistically significant results. After an experiment is completed, the experimenter then works backwards by cherry picking and looking for statistically significant results to support a hypothesis that is also cherry picked. (The latter picking is called HARKing—standing for “hypothetical after the result is known.”) This problem culminated in what some called a replicability crisis regarding studies published in psychology and social science journals where published experimental results could not be replicated. This led to some journals requiring as a condition of publication that a study be registered in advance with a specific description of the hypothesis being tested before the study actually commences. That way hunting for a statistically significant result to be worked backwards in support of a post hoc chosen hypothesis was precluded. We can’t take similar measures with respect to the fine tuning argument. We are already here. We can’t start a universe to see what will happen. We have to start already knowing we exist. So we start with the “remarkable” (to us) fact that life or intelligent sentient life exists and work backwards to support the hypothesis of a fine-turner. The fine-tuning argument is itself fine-tuned to our liking.
This does not constitute a full survey of problems with the fine-tuning argument. But it comes down to this — no matter how forcefully the fine-tuning argument is “sung,” it can still sound off-key.
.