Naturalism, Fine Tuning and Flies: A Response

Before reading my response, readers should address their attention first to Aron Lucas’s work Naturalism, Fine Tuning and Flies. While a brief summary follows, it is always worth consulting the original work in order to make sure that it is fairly represented by the critic (in this case me).

Possible Universes

The crux of Aron Lucas’s argument is a very straightforward objection: The Fine Tuning argument posits a narrow class of possible universes, restricted to only those consistent with our laws of physics. The set of possible universes on naturalism includes those with different laws of physics, not just different constants and quantities. This set is inscrutable and, thus, on naturalism, the probability of life permitting universes itself is inscrutable. I think one important question here what we mean by “alternate laws of physics”. We can think of this in a number of ways:

  • Alternate functions representing existing laws
  • Removal of existing laws altogether
  • Completely new laws
  • Completely new fundamental substances

Possible Objections

There seem to be three rebutting objections to this response to the argument from fine tuning, which I will list here and then elaborate on each.

  • Two can play at this game
  • We are warranted in ignoring these possible universes (those governed by different laws)
  • These possible universes are not, in fact, inscrutable (or at least to the degree at which it is meaningful)

Two Can Play at this Game

One of the key goals of Aron Lucas’s objection is to extend the possible universes in a way that makes it impossible to infer the likelihood of life permitting universes among those possible universes. By opening up the floodgates of an infinite number of possible universes with wholly different laws, within which life permitting universes may be plenteous (or not). Well, can the proponent of the FTA make a similar move that multiplies his or her probabilistic case? I think so.

Aron Lucas refers to the current state of possible universes in the FTA as a “narrow class” insofar as the possible universes are limited to those with the same laws (albeit different constants and quantities). It is true that the class is restricted, but removing that restriction does not necessarily improve the case of the objector to the FTA. While Lucas can argue that we should allow possible universes with different laws altogether, the proponent of the FTA could, for example, extend the list of non-life permitting universes by considering different arrangements of particles within that universe.

Take one of the most common fine tuned constants, the expansion rate of the universe. Imagine now that rate is increased such that it immediately results in the heat death of the universe (immeasurable distances between each subatomic particle or photon). Now, imagine that same universe, but every particle is shuffled into a slightly different position. Now shuffle it again, and again. Each universe represents a different, possible, conceivable universe that is not life permitting. And, since we can use a similar methodology to extend all possible universes within the “narrow class“, we can strengthen the FTA ad infinitum.

Now, I don’t think that this is a move that the proponent of the FTA should make, but I think it is every bit as reasonable as the move made by Aron Lucas. If we are to consider possibilities outside our extant physical laws, it seems to me equally fair to consider all possibilia within our extant laws including varying more than just the constants and quantities, but the possible outcomes given those different constants and quantities. Ultimately, I think this exercises is silly because in both cases we seem to just throw infinities at the problem without really giving any justification as to why they warrant consideration.

Warranted Ignorance Towards Non-Compliant Possible Universes

Should we consider non-compliant possible universes as Aron Lucas suggests? Let’s explore this concept with an analogy.

Imagine you have a bag of 100 marbles. You reach in and pull out 10 marbles, 7 are red, 2 are blue and 1 is white. You place the marbles back in the bag, give it a shake, and grab another handful. This time you have 8 red, 1 blue, and 1 white. You place the marbles back in the bag, give it a shake, and grab another handful. This time you have 6 red, 3 blue, and 1 white.

With simple statistics, we can make a predictions about (1) the ratio of red, blue and white marbles in the bag and (2) the ratio of red, blue and white marbles you will retrieve in each handful. One of the conclusions of these formulas is that white marbles are rare.

Now imagine a skeptical stranger comes and says to you that this is not true. The problem is that we are only considering the color. It is possible that there are only 20 marbles that are large enough to handle in the bag, and that 80 are so small that they always slip through your fingers. These 80 are all white (or some other color, we just don’t know). Or maybe they are uniform in size, but the white marbles are sentient and are good at evading fingertips, so there are actually mostly white marbles in the bag, we are just only capable of retrieving a couple at a time. Or maybe all the marbles are white, but some of them have a unique property where when introduced to sunlight they project a distortion around them to make them look red or blue to the human eye, while remaining white underneath. In fact, all the marbles are white. Or maybe the bag has 100 marbles but the marbles are constantly popping in and out of existence at random, changing colors every time. Despite our observations the first 3 times, they could be anything in the future.

We can construct an infinite number of these possible scenarios which would undermine the initial observations and conditions used to predict the marble colors. The skeptical stranger is correct, in a sense. Unless we pour out the contents of the whole bag, we can’t know with certainty that our sampling is accurate. However, I think we would all be hesitant to believe the skeptic’s concerns. In fact, if we were to accept the skeptics concerns, we would be unable to discern anything by inference because one can construct an infinite number of scenarios which undermine the assumed uniformity of key properties regarding the possibilities.

The question which seems key here is when are we warranted in assuming uniformity of certain properties of the possibilities and when are we not.

Perhaps there are some rules of thumb we could establish to help guide us as to when we should and should not consider possibilities.

  1. Vary as little as possible compared to what is known
  2. Vary that which we can model or conceive

Let’s explore each of these briefly. The first rule of thumb seems to be obvious. If we are considering alternate possibilities, we are safest considering possibilities that are similar to what we already know is possible. The current formulations of the Fine Tuning Argument do just this. They vary constants and quantities without changing the underlying formulas. More importantly, changing the fundamental laws themselves individually is a non-starter. The relationships defined by the laws of physics are intertwined. We could not, for example, replace the law of gravity based on mass with a law of gravity based on color, without changing every other law. There is no modest variation available, it is all or nothing.

The second rule of thumb is just an exercise in pragmatism. If the possibilities you wish to consider are completely inscrutable and wildly undefined such that they cannot even be conceived, much less modeled, perhaps we are warranted in ignoring them. For example, we can model and conceive of universes with different constants and quantities. However, it is terribly difficult to model or conceive of universes that have completely different laws. It is important here to recognize that by laws we mean relationships between particles and forces. We can conceive of different relationships (which we modify by changing constants and quantities), but to conceive of different laws would be more than just saying a relationship is no longer present (for example, there is no relation between mass and gravity, as that can be zeroed out in our current fine tuning models). However, conceiving of a relationship where gravity relates, for example, objects based on color seems at face value to be inscrutable if not non-nonsensical.

Given these rules of thumb, it seems to me that the proponent of the FTA should not be worried about non-compliant possible universes as Aron proposes. Our confidence that these non-compliant universes are actually possible is questionable.

Non-Compliant Possible Universes are not Inscrutable

Once again, I outlined some examples of possible universes that Aron Lucas may be referring to…

  • Alternate functions representing existing laws
  • Removal of existing laws altogether
  • Completely new laws
  • Completely new fundamental substances

Let’s assume for a second that my first two responses fail and that we are left with Aron’s objection that the naturalist should consider this broader class of possible universes. Assume even further that these non-compliant universes are in fact possible. Even if we grant this, I am not confident that they are wholly inscrutable. Let’s look at each one…

  • Alternate functions representing existing laws:
    Perhaps the formula we use for gravity is different in another possible universe. This seems to be fairly plausible relative to the other proposed non-compliant universes. But what impact would this have on, for example, the fine tuning of the weak nuclear force? Even if our formula for gravity were radically different, if the weak nuclear force were off by 1/10^100, there would still exist the problem of star formation. And even if there were a completely different formula for the weak nuclear force, if gravitational constant were modestly altered, it would prevent star formation. In either case, the change of existing laws would not change the fact that other fine tuned constants and quantities would still be sufficient to undermine the formation of life. Aron has not provided any reason to believe that such new formula would negate existing fine tuning parameters.
  • Removal of existing laws altogether:
    In many cases, this is already simulated as relationships are zeroed out by the modification of constants and quantities. Moreover, the same response I provided above seems to be relevant. Removing one law or multiple laws, would not necessarily mean that the other fine tuned constants and quantities would still not have the same impact on the universe. Aron has not provided any reason to believe that removing some existing laws would negate existing fine tuning parameters.
  • Completely new laws:
    Once again, a new law would not necessarily mean that the other fine tuned constants and quantities would still not have the same impact on the universe. Aron has not provided any reason to believe that new laws would negate existing fine tuning parameters, unless we imagine ad hoc laws that intentionally necessitate the parameters as they are (but, of course, we would then ask for an explanation of that law)
  • Completely new fundamental substances:
    This seems to me to be the hardest of the possibilities to deal with just because it is so hard to conceive. This would violate nearly every modal intuition we have, so I think Aron needs to provide a reason for us to do so if this is what he means by a possible universe beyond what is considered in the FTA.

Different forms of Life

One issue I have chosen not to deal with in this response is the possibility of observers totally unlike what we know. In the current fine tuning argumentation, we go so far as to allow for random fluctuations of particles just generating an observer out of the blue. If we are willing to grant that possibility, I’m not sure we need to grant anything else.

Concluding Thoughts on the Naturalism, Fine-Tuning and Flies

Aron Lucas opens our eyes to non-compliant possible universes as potentially presenting a problem to the FTA insofar as it is a narrow class of possible universes. Unfortunately, Aron does not develop the case that…

  1. These “non-compliant” universes with different laws and/or fundamental properties are in fact possible
  2. These “non-compliant” universes would necessarily overcome existing fine tuning issues
  3. Other possible universes which dramatically increase the strength of the FTA should be excluded

Subsequently, I think the proponent of the FTA is well within their epistemic rights to follow a conservative modal intuition regarding possible universes. We are just as warranted in our belief of a Finely Tuned Universe as we are warranted in our belief that the bag of marbles will return red, blue and white in a certain proportion based on standard statistics.

1 comment

  1. Aron - October 23, 2019 7:56 pm

    Hi Russ, I’m Aron Lucas. Thanks for taking the time to put together a thoughtful response to my paper! I’m still thinking over your response, so at this stage I will just comment on your section about marbles.

    It is true that if we pull out mostly red marbles, this confirms the hypothesis that most of the marbles are red – let’s call this hypothesis R. P(mostly red|R) is high, and P(mostly red|~R) is low. But wait? Aren’t there are subhypotheses under ~R that entail that we would pick out mostly red balls, so under these subhypotheses, the likelihood of R is high? Yes, but as you point out, we can safely disregard these hypotheses. But it is important to understand *why*. Let’s take one of your subhypotheses – that the non-white marbles are sentient and evade our fingers – and call it S. Even though P(mostly red|~R&S) is high, P(S|~R) is low. For obvious reasons, the idea of sentient marbles violates all sorts of known facts in our background knowledge, as do you other subhypotheses under ~R which predict that we would select mostly red marbles. So the overwhelming majority of ~R’s probability space is taken up by hypotheses that don’t predict that we would select mostly red marbles. So the fact remains that P(mostly red|~R) is low. We are able to make this judgment because of our extensive experience with things like bags and marbles, and this allows us to make informed judgments about the probability of the different subhypotheses under ~R.

    This is not the case when it comes to the various sub-hypotheses under “chance” or “atheism.” Of all the subhypotheses that fall under this category, only a narrow sub class are governed by our laws. We can say that life is unlikely in this class. But we don’t have the relevant background knowledge to say that other subhypotheses involving different laws that would permit life are unlikely. How much of the atheism’s probability space is taken up by universes that permit life, either with our laws or other laws? We just can’t say. So whereas in the marble case we have helpful background knowledge, we just don’t have that kind of background knowledge when it comes to the case of fine tuning.

    I think a more appropriate analogy would be something like this: You have a bag filled with potentially infinite marbles. You select a small sample, not randomly, but from small portion of the bag you can access, and they are mostly red. Surely, you would be much less confident saying that this confirms to a high degree of confidence that the entire bag is mostly red marbles.


Have your say