A paper was published in November last year in the Journal for General Philosophy of Science by Roberto Fumagalli.1 The author, who is Senior Lecturer in Philosophy, Politics and Economics at King’s College London, contends that the probability of the fine-tuning (FT) of the physical properties of the universe to be conducive to complex life is significantly higher on the presumption of non-chance-based explanations than it is on the presumption that it arose purely by chance. As such, the author argues that the fine-tuning of the laws and constants of our universe is real and requires a non-chance-based explanation, whether that be design, a multiverse, or some yet to be discovered principle of physics. The paper offers a systematic discussion and detailed appraisal of objections to fine-tuning arguments that insist that it requires no special explanation.
Fine-Tuning Arguments: Analysis
Fumagalli begins by formalizing the core structure of fine-tuning arguments (FTAs) with a view towards clarifying what they do and do not claim and what would be required to undermine them. Fumagalli schematizes FTAs as follows:
FT: The universe is fine-tuned for intelligent life (i.e., only a very narrow range of parameter values permits such life).
Pr(FT∣∼CHANCE)≫Pr(FT∣CHANCE) — in other words, the probability of fine-tuning conditional on the falsity of the chance hypothesis is significantly greater than the probability of fine-tuning conditional on the chance hypothesis.
Therefore, FT calls for a non-chance-based explanation.
Fumagalli acknowledges ongoing debate among physicists as to the extent of fine-tuning, with some regarding the fine-tuning as being less numerically impressive than sometimes claimed. Nonetheless, it is widely acknowledged, even by those critics, that at least some fine-tuning is real and non-trivial. The debate, therefore, is about the extent of fine-tuning, not about whether it exists at all.
Fumagalli also acknowledges that some critics attempt to cast doubt on fine-tuning by noting that we do not know what other forms of advanced life are possible beyond carbon-based organisms. In a different universe, perhaps other forms of intelligent life could flourish. Nonetheless, “pointing to our current ignorance concerning what forms intelligent life may take does not per se cast doubt on FTAs’ assumption that this universe is fine-tuned for intelligent life.” Indeed, “no precise definition of the notion of intelligent life is required to justifiably infer that only highly specific combinations of parameter values would permit the chemical complexity required for intelligent life.”
Fumagalli also correctly notes that “substantiating FTAs’ claim that FT calls for explanation requires one to substantiate the claim that the probability of FT conditional on ∼CHANCE is much higher than the probability of FT conditional on CHANCE, but does not require one to ascribe quantitatively precise probabilities on the propositions figuring in FTAs.” This is in fact a common mistake of critics of fine-tuning arguments and indeed arguments for theism more broadly. It order to make a successful Bayesian inference to design, one does not need to assert precisely how probable the feature of nature is (in this case fine-tuning) conditional on design, nor does one even need to argue that the feature is particularly probable (or more probable than not) conditional on design. Rather, all one needs to argue is that the feature is much more probable conditional on design than on the falsity of design.
Objection from Unjustified Probability Ascriptions
In the next section, Fumagalli tackles the objection that fine-tuning arguments fail because their probability claims are not adequately justified: “The objection proceeds as follows. Assessing the extent to which this universe is fine-tuned for intelligent life requires one to specify the range of fundamental parameter values that permits intelligent life and the probability distributions for these fundamental parameter values.” But, so the objection goes, “the range of physically possible values of various fundamental parameters appears to be potentially unbounded, leading to an infinite space of physically possible parameter values.” Fumagalli offers two responses to this objection.
First, the objection overstates what fine-tuning arguments require. They do not require probability distributions over infinite, unconstrained parameter spaces. Indeed, proponents of fine-tuning arguments “can draw on both theoretical reasons and empirical evidence to identify defensible bounds on the range of physically possible parameter values…And on most proposed identification of such bounds, the range of parameter values falling within the identified bounds is wide enough to make the FT constraint significant.”
Second, Fumagalli argues that “proponents of FTAs may support FTAs’ assumption that FT is extremely unlikely to occur purely by chance even in cases where they are unable to identify defensible bounds on the range of physically possible parameter values.” He discusses non-standard probability theories (where “the real number line standardly used to express the values of probability functions” is extended to include infinitesimal values). Fumagalli explains,
Nonstandard probability functions can assign infinitesimal probabilities to each possibility such that the probability of each possibility is >0 and the sum of the probabilities over all possibilities is ≤1 (e.g.. Vallentyne 2000, 276; also Benci et al. 2018). These probability functions provide the proponents of FTAs with a mathematically tractable basis to support FTAs’ assumption that FT is extremely unlikely to occur purely by chance even in cases where they are unable to identify defensible bounds on the range of physically possible parameter values (e.g., Pruss 2021a, 777–780; also Koperski 2005, 306–311, on the possibility of supporting such assumption in measure-theoretic terms).
Fumagalli also addresses the concern that fine-tuning arguments rely on arbitrarily chosen probability measures for life-permitting physical constants, since the actual probabilities are presently underdetermined by the data — thus making the case dependent on subjective judgments. Fumagalli responds to this objection by noting that proponents can appeal to various strategies, including imprecise probability frameworks, and to theoretical and empirical constraints that restrict the range of realistic probability assignments. He emphasizes that “various such probability ascriptions support FTAs’ assumption that FT is extremely unlikely to occur purely by chance.” Indeed, “most authors (including many critics of FTAs) agree that, for all we know, the values of at least some fundamental parameters are extremely unlikely to fall within the range that permits intelligent life.” He concludes that “This, in turn, provides prima facie convincing support to FTAs’ assumption that FT is extremely unlikely to occur purely by chance.”
Objection from Observation Biases
In the next section, Fumagalli turns his attention to another objection — that from observation biases. That is to say, it is argued that fine-tuning arguments “fail to show that FT calls for explanation on the alleged ground that FTAs derive specious plausibility from various observation selection effects.” Fumagalli summarizes the objection as follows:
Since we exist, the values of this universe’s fundamental parameters must have fallen within the range that permits intelligent life (e.g., Carter 1974, 291–298). Moreover, we are guaranteed to observe that the values of the fundamental parameters have fallen within the range that permits intelligent life (e.g., Manson 2009, 274–278). As a result, the probability of observing that the values of this universe’s fundamental parameters have fallen within such range does not vary depending on whether one assumes ~CHANCE or CHANCE (e.g., Sober 2009, 77–78), and our discovery of FT does not favor ~CHANCE over CHANCE (e.g., Sober 2009, 84; also 2003, 41–47).
In response to this objection, Fumagalli notes that the fine-tuning argument is not premised on the fact that our universe’s physical parameters are in the life-permitting range (of course, this is entailed by the fact we exist). Rather, the premise upon which the argument rests is the surprising observation that only an exceedingly narrow range of parameter values is conducive to life, especially advanced life. This finding is disconfirming of the chance hypothesis. It is, after all, quite conceivable that scientists had discovered that a wide range of parameters supported life.
Objection from Causal Ramification
Fumagalli proceeds to address the objection from causal ramification, which runs as follows:
The alleged improbability of FT does not per se imply that FT calls for explanation. For “massively low probability events occur all the time” (Worrall 1996, 11), and we take only few such events to call for explanation (e.g., Callender 2004a, 205). Hence, “if we have any reason to believe that [FT is not] a brute fact, it must be something other than, or at least in addition to, its improbability” (Baras and Shenker 2020, 17, emphasis added; also Harker 2012, 247). As Juhl puts it, “practically any non-microscopic [phenomenon] in the universe is causally ramified [i.e.] causally depend[s] for its existence on a large and diverse collection of logically independent facts. […] Yet one does not observe [FTAs] from the existence of a pebble in one’s back yard” (Juhl 2006, 271, emphasis added).
This objection too fails, however, since the alignment of the parameter values with the life-permitting range for life is not just any set of values but carries a special significance.
Objection from Mistaken Explanatory Demand
The objection from mistaken explanatory demand “holds that FTAs fail to show that FT calls for explanation on the alleged ground that the occurrence of FT is a more appropriate explanatory stopping point than explanatory posits such as multiverses, cosmic designers and hitherto unconceived physical laws/mechanisms.” Fumagalli summarizes the objection thus: “FTAs infer that FT calls for explanation by pointing to the alleged fact that FT is extremely unlikely to occur purely by chance. However, any explanation has to take some facts as brute, since ‘every explanatory theory will feature some set of unexplained explainors’ (Grünbaum 2004, 598; also Worrall 2004, 66).” But the mere fact that some things are brute facts that require no further explanation is hardly a reason to think that fine-tuning is one of those things. Moreover, as Fumagalli observes, “the extreme improbability of this match [between the range of life-permitting values and the actual physical parameters of our Universe]…makes it epistemically inadequate to simply regard FT as a brute fact. For FT would be a highly improbable coincidence given CHANCE. And ‘there is pressure for scientific or philosophical theories to avoid [highly improbable] coincidences’ (Bhogal 2020, 677; also White 2005, 7).”
Objection from Lack of Testable Explanatory Alternatives
The final objection considered is that fine-tuning arguments “fail to show that FT calls for explanation on the alleged ground that the hitherto proposed ~CHANCE fail[s] to satisfy minimal requirements of empirical testability.” In particular, “appeals to multiverse hypotheses, cosmic design hypotheses and future physical theories currently lack empirical support.” In response to this, Fumagalli notes that “the critics of FTAs have hitherto failed to offer convincing reasons and/or evidence to think that ~CHANCE’s alleged lack of empirical support reliably indicates that ~CHANCE is false.” I would add to this that the cosmic design hypothesis can be tested in the Bayesian sense that the fine-tuning of our universe for life (especially complex life) is rendered far less surprising on the hypothesis of design than on the falsehood of design. There are also reasons, independent of cosmic fine-tuning, to think that there exists an intelligent mind behind life and the universe, which is positively relevant to our assessment of the prior probability of a mind also being the explanation of the fine-tuning of the physical laws and constants.
Multiverse, Design, or Unknown Physics?
Fumagalli does not in the end state whether he himself lands on the multiverse hypothesis, cosmic design, or unknown physics as the best explanation of cosmic fine-tuning. If unknown physics is the explanation then, it seems to me, this unknown physics would itself need to be finely tuned (since it would be a remarkably fortuitous coincidence that it constrains the values of our universe in such a manner as to be life-permitting). The multiverse theory also faces problems of its own, not least of which is that the mechanism for generating the multiverse would itself require fine-tuning. As previously stated, there are also independent reasons to believe that an intelligent mind lies behind nature. The best explanation, I am persuaded, therefore, is that the fine-tuning of our universe is the product of cosmic design.
Notes
1] Fumagalli, R. The Universe’s Fine-Tuning Does Call for Explanation. J Gen Philos Sci (2025). https://doi.org/10.1007/s10838-025-09734-8.
