Is the universe fine-tuned for life?

Many of the constants and conditions that describe our universe appear to fall in narrow ranges that permit life. The physicist Luke Barnes summarizes the claim: “in the space of possible physical laws, parameters and initial conditions, the set that permits the evolution of intelligent life is very small”^{[source 1]}. Examples include the relative strengths of the fundamental forces, the masses of elementary particles, and the cosmological constant, the energy of empty space, whose observed value is far smaller than simple theoretical estimates^{[source 1][source 3]}.

Physicists sort the examples into three kinds: constants, such as the strength of gravity or the mass of the electron; initial conditions, such as the density and smoothness of the early universe; and the form of the laws themselves^{[source 3]}. Barnes surveys cases from each, arguing that many survive careful scrutiny even when several quantities are allowed to vary at once^{[source 1]}.

The astronomer Fred Hoyle, who was not religious, concluded from his work on how stars make carbon that “a superintellect has monkeyed with physics, as well as with chemistry and biology”^{[source 2]}. The philosopher Robin Collins develops the theistic argument: a life-permitting universe is far less surprising if a designer intended one than if the constants were set by chance, so fine-tuning counts as evidence for design^{[source 5]}.

Hoyle’s own example concerned carbon, the basis of all known life. Stars make carbon and oxygen through a sequence of nuclear reactions that depends on a specific energy level in the carbon nucleus, at 7.65 MeV, and on one in oxygen. Hoyle noted that if you wanted to produce carbon and oxygen in roughly equal quantities in stars, these are just the two levels you would have to fix, and that they sit just about where they are found^{[source 2]}.

Collins frames the argument as a likelihood comparison rather than a proof: whatever one’s prior view of God, the fine-tuning evidence is much more probable on the hypothesis that a designer intended embodied life than on the hypothesis of a single universe with constants set by chance, so it raises the probability of design^{[source 5]}. He presents it as one strand of a cumulative case, not a demonstration.

One response is the anthropic principle: we could only ever observe a universe compatible with our existence. As the physicist Brandon Carter put it, our location in the universe is necessarily privileged to the extent of being compatible with our existence as observers^{[source 3]}. On its own this explains why we see a life-permitting universe, not why there is one.

Another is the multiverse. If there are many universes with different constants, as some versions of inflationary cosmology and string theory suggest, life-permitting ones are bound to occur somewhere^{[source 3]}. Critics of this response note that other universes cannot be observed, and some philosophers argue that inferring a multiverse from our one life-permitting universe commits an “inverse gambler’s fallacy”^{[source 3]}.

The “inverse gambler’s fallacy” charge comes from the philosopher Ian Hacking, who raised it against universes arranged in a temporal sequence; others have since extended it to multiverses generally. It is the mistake of inferring, from one remarkable outcome, that there have probably been many earlier trials with less remarkable ones, like someone who walks into a casino, sees a remarkable roll at the nearest table, and concludes the dice must have been rolled many times before^{[source 3]}. Defenders of the multiverse response answer that the analogy fails, because only a life-permitting universe could contain observers to notice it, and the debate over which side has the better of this continues^{[source 3]}.

Others question the premise. Victor Stenger argued that the life-permitting range is much wider than claimed when several constants are varied together^{[source 6]}; Barnes replied in detail that most cases of fine-tuning survive that objection^{[source 1]}. Some philosophers doubt that probabilities can be meaningfully assigned to constants that could take any value^{[source 3]}, and Sean Carroll argues that we know too little about which universes could support some form of life, and about what a designer would do, for fine-tuning to favor theism^{[source 4]}. Collins replies that probabilities can be assessed over the range of values physicists can actually consider, which he calls the epistemically illuminated range, and that a multiverse generator would itself need finely tuned laws^{[source 5]}.

Some critics also question what the argument shows even if fine-tuning is real. The philosophers Elliott Sober and Jan Narveson argue that we are in no position to predict what a cosmic designer would prefer, so we cannot say that design makes a life-permitting universe more likely than chance^{[source 3]}. Theistic philosophers have replied that a good God has reason to create embodied creatures like us, and the exchange continues^{[source 3]}.

That some constants are finely tuned in this sense is widely accepted among physicists, though how many and how finely is debated^{[source 1][source 3]}. What explains it is not something physics currently settles: design, a multiverse, an undiscovered deeper law and brute fact all have serious defenders, and the choice among them rests largely on philosophical arguments about probability and explanation^{[source 3]}.

Within physics itself, fine-tuning is treated as a puzzle to be solved. The related criterion of naturalness, which counts a theory as suspect if its parameters must be finely adjusted, plays a large role in contemporary particle physics and cosmology, and has motivated much of the search for deeper theories^{[source 3]}. If such a theory were found, some cases of fine-tuning might be explained, though proponents of the design argument reply that the deeper theory’s own features would then need explaining^{[source 5]}.