Leibniz, Hangnails & the Best Possible World

In a brief Twitter conversation it was suggested to me that a world with one less hangnail would be a better world than the world we live in, and thus the world we live in cannot be the best possible world (BPW). Take that, Leibniz!

Let’s call our world up to 2014 W1. Now let’s go back in time to Billy Bob in 1956 when he suffered a hangnail and intervene ever so slightly so he avoids suffering a hangnail. Let’s call this new world W2. Now, fast-forward again to 2014. Are W1 and W2 identical except that in W1 in 1956 Billy Bob suffered a hangnail? I don’t see how we can make such a claim, and in fact we can easily imagine how by 2014 W1 and W2 could very well be radically different worlds. How so? Well, in W1 after suffering a hangnail, Billy Bob went to get medical treatment. He met a nurse at the hospital; they hit it off, eventually got married, and had 4 children. When we consider the enormous number of different events that take place in W1 as a result of Billy Bob’s hangnail it should be fairly obvious that by 2014 W1 is very different from W2.

But is it different for the better? Maybe, maybe not. Maybe for a time it gets better, after which it gets horrendously worse. Or maybe it’s horrendously worse for a time but ultimately gets far better. Maybe Billy Bob’s kids turn out to be wonderful philanthropists, and humanity reaps all manner of boons. Or perhaps they end up a family of serial killers and bring misery to hundreds. But regardless of how they turn out we still can’t tell if W1 is ultimately better or worse than W2, since there are further multiple consequences of both scenarios which we, as creatures of limited intelligence and insight, simply cannot trace.

But isn’t a world with one less hangnail – W2 – better than a world – W1 – which is identical in all respects except there is one additional hangnail? Again, this isn’t obvious. It relies on a certain understanding of “better” which we needn’t agree with. If we understand the chief purposes of God for the world to lie in the maximisation of pleasure and the minimisation of pain then perhaps W2 is “better.” But of course in Christian theism this is not God’s sole or ultimate purpose. In any event, for the sake of argument let’s grant that W2 would indeed be better than W1. But, now we must ask: is W2 a feasible world? It’s not clear that it is. Take Billy Bob again. Perhaps every world in which he suffers the hangnail turns out differently from all the worlds in which he doesn’t, such that there simply are no two worlds which are identical in all respects except that Billy Bob suffers a hangnail in one of them. And this should not surprise us as such cases – even relatively minor ones like a stubbed toe – are always part of the matrix of events. Such pains have a number of uses in the grand scheme of things. Firstly, that such pains are of biological value is well documented. Secondly, as Swinburne argues, the experience of pain is an intrinsic part of any world where creatures are to have significant moral responsibility and freedom to do good or evil. Thirdly, such experiences form part of our “epistemic environment,” contributing to the background against which we reflect on the world and form opinions about good, evil, value, and about God and the nature of ultimate reality. Fourthly, they play a role in our development as moral creatures, as we respond daily to the aches and hurts that accompany daily living.

Thus, even relatively minor hurts like stubbed toes and hangnails play a role in life, even a significant one such that a world with one less hangnail – W2 – might have over-ridding deficiencies that make W1 preferable.

The principle behind this has been named “the butterfly effect.” The basic idea is that something seemingly unimportant – a butterfly fluttering around some flower – can potentially set in motion a chain of events (or play a small but crucial role in the “events matrix”) that leads to something massive – a hurricane off the coast of Florida; and we have no way to predict or trace it.

The same idea appears in the movie Sliding Doors. In this movie the lead character is hurrying to catch a train. The film then branches off into 2 strands or “mini-movies.” In one of these worlds she catches the train, while in the other events conspire to cause her to miss it. The movie then plots how her life goes in two completely different directions as the result of this one seemingly benign and insignificant event (which of course was itself dependent on millions of prior contingent events either occurring or not).

The upshot of all this is that we really can’t tell which is better – W1 or W2 – by engaging in observation and imaginative thought experiments. In fact, whilst Leibniz believed that this world is the BPW, he didn’t believe it on the basis of observation. He understood that drawing such a conclusion was impossible. His belief that this world is the BPW was a deduction from his prior belief that the world is the creation of an omnipotent and perfectly good God. If the universe is the creation of such a being then, reckoned Leibniz, there is some reason to think it is the BPW. But Leibniz would have had no time for modern atheistic arguments that run like so:

(1) If God exists our world would be the best possible world.
(2) Our world is not the best possible world.
(3) Therefore God does not exist.

Whilst Leibniz would agree with (1), he would regard (2) – quite rightly in my view – as utterly speculative. William King put it like so: “You’ll say that some particular things might have been better. But, since you do not thoroughly understand the whole, you have no right to affirm this much.”

Thus, though Leibnizian arguments might rationally conclude that this is the BPW because it has been created by an omnipotent and perfectly good God, atheist arguments to the opposite conclusion will always rely on a premise that is fundamentally unknowable.

Stephen J Graham.

See also:

My main essay on Leibniz here:


and another shorter article here:


1 Comment

Filed under Leibniz, Problem of Evil

One response to “Leibniz, Hangnails & the Best Possible World

  1. Consider this argument:

    1. If God exists, then instances of gratuitous natural evil do not exist.
    2. Instances of gratuitous natural evil do exist.
    3. Therefore, God does not exist.

    The hangnail is essentially a defense of (2). I think it’s a strong defense because it seems fairly obvious to me that there is a possible world such that (a) this world is identical to our own with regard to all axiological properties, and (b) the hangnail never obtains. At least, it seems fairly obvious *if we don’t take God as background knowledge* when deciding whether this world is a possible (or feasible) world.

    Of course, Leibniz would (again) go Moorean on this, and propose (2′) God exists. And perhaps this would be an obvious premise if we don’t take (2) as background knowledge. But to me, (2) seems like the the far better premise to place one’s stake in, and here’s why.

    Imagine that I have two breakfast options: waffles and toast. I can eat waffles today and toast tomorrow, or vice versa. Now, there’s a world W1 where the former leads to some greater good, and a world W2 where the latter does. But there’s also a W3 where things turn out the same regardless. To deny that W3 is a possible world is to imply that *necessarily*, eating toast will lead to a greater good. But that would likely break all of our intuitions about modality. Also, you can’t deny just W3. You’d also have to deny either W1 or W2. If you don’t, you have two mutually exclusive necessities (or: the greater good itself will necessarily obtain). But if you deny W3, then how could you possibly decide between W1 and W2?

    Likewise for hangnails; except that unlike toast and waffles, hangnails are an instance of natural evil (a very minor one, but still). So here’s some options for the hangnail:

    H1 – world in which the hangnail leads to a greater good
    H2 – world in which not-hangnail leads to a greater good
    H3 – world in which hangnail has no effect on whether a greater good happens

    (A) H1, H2, and H3 are all possible worlds. In this case, the hangnail is gratuitous.
    (B) H1 and H2 are possible worlds, but not H3. In this case, the greater good is necessary, and the hangnail is gratuitous.
    (C) only H1 and H3 are possible worlds. In this case, hangnail is gratuitous.
    (D) only H2 and H3 are possible worlds. In this case, hangnail is gratuitous.
    (E) only H1. In this case, necessarily, hangnail leads to greater good.
    (F) only H2. In this case, necessarily, hangnail is gratuitous.
    (G) only H3. In thiscase, necessarily, hangnail is gratuitous.

    When looked at this way, we can notice something that may be rather surprising. The claim that gratuitous evil exists is identical to the claim “The actual world is either A, B, C, D, F, or G”. The claim that gratuitous evil does not exist is “The actual world is E” (in both cases, we take some instance of evil as background knowledge). The former claim is much more modest than the latter, and thus has a higher prior probability. So, in the end, I don’t think Leibniz is justified in a simple Moorean shift.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.