Bailey and Rasmussen's Infinite Error

Why Bailey and Rasmussen do not have infinite value--though they do have lots of value

Bailey and Rasmussen wrote a paper arguing that humans have infinite value. Correction, their paper just argued that if humans had equal value they would have infinite value. I disagree with the paper, and thought it would be worth explaining why. It also is a hilarious paper—well written, witty, and all around fantastic, making it a treat to respond to.

To begin, they say

Human Value: Extreme and Equal Human persons are those things to which we ordinarily refer with personal pronouns. Among human persons are those things we sometimes call ‘non-cognitivists’, ‘teenagers’, ‘politicians’, ‘foreigners’, ‘heterosexuals’, and more. You are a human person. The authors of this article are people too. So how much value does a human person have? At least some. We take it to be self-evident that you are valuable. You matter. It is good that you exist. Your value seems to surpass the value of many other things, including pugs, sunsets, twitterpation, “Black Star”, valid arguments, delicious dinners, The Keeper of the Plains, and more. It is good that those items exist; the world is better for their being. Yet, there are important senses in which your value surpasses the value of beasts, landscapes, pleasant feelings, classic Yngwie J. Malmsteen tunes, proofs, meals, or majestic steel sculptures. Your value is more extreme.4 You are no exception or outlier. The authors also enjoy this remarkable status. We all do. Value is our common human birthright. We can be more specific about the value we have in common. Though we may differ with respect to instrumental value, we do not differ with respect to what we call “final” value. In this sense we are all equals. You may well be more effective than someone else when it comes to producing lucrative dance beats, for example, and thus more valuable in that instrumental sense. Or you might have powers of persuasion that make you a potent ally in campus politics. But when it comes to final value, you don’t have more of that than anyone else. You are a properly valued for your own sake to exactly the same degree as is anyone else. It is proper that someone treat you as an end in yourself, and the same goes for everyone else in exactly the same sense and to exactly the same degree. In matters of final value, we are all the same.5 So, we each enjoy final value that is extreme and equal. Were final value a kind of wealth, the tribe of humanity would be supreme in its affluence, while also blessed with a Gini coefficient of 0.

While the idea of humans being infinitely valuable sounds appealing, it is clearly false. A few examples can illustrate this.

1. Suppose one is having a child. They can either drink a mixture which will lead to their child having very severe depression or not drink the mixture. If all humans have equal value, then the value of bringing about a person with lots of misery in their life is equal to bringing about one with much less misery. Thus, it would be morally neutral to drink the potion on this account. This is clearly false.

2. Suppose one could save either a serial killer from death today who will die tomorrow even if saved or a healthy 30 year old, who will die today unless saved. They should obviously save the 30 year old. Thus, the value of saving a life depends on facts about the life—it is not intrinsic and equal.

3. Most would agree that all else equal it would be better to save a 30 year old than an 80 year old, because the 30 year old has more years of happy life left. Thus, lives don’t all have equal value.

4. If you could either bring into existence a person who would experience pure agony for their entire life or one who would experience pure bliss for their entire life, you should obviously bring into existence the one who would experience pure bliss.

These show that human lives are not all equal in value. If two things are equal in value

A) Preserving them is equally important.

B) Creating them is equally important.

However, we have seen that this is not the case for humans. Thus, humans are not all of equal value. In most practical regards it makes sense to treat humans as equally valuable, but humans aren’t really equally valuable.

The authors go on to describe that this is puzzling. Generally, in other domains like art, very different things have different value. I agree with this.

Next they say

The first argument uses a Bayesian confirmation framework. Let Equal Human Value be this datum: we all enjoy equal and extreme final value. Finite Value Hypothesis is the theory that we are each equally and extremely finally valuable to some finite degree. Infinite Value Hypothesis is the theory that we are each infinitely finally valuable. 6 Consider now two conditional epistemic probabilities. These report, roughly, the degree to which one should expect a particular item of evidence, given a hypothesis: P(Equal Human Value/Infinite Value Hypothesis) P(Equal Human Value/Finite Value Hypothesis) These probabilities are not the same. And indeed, the first is significantly higher than the second. We shall argue for this claim from both directions. Given Infinite Value Hypothesis, we may expect that people have equal value. It would not be surprising, given Infinite Value Hypothesis, that we are all deeply and equally value. It seems to be precisely what one would expect, in fact. Infinite Value Hypothesis, furthermore, offers a simple and compelling explanation of Equal Human Value. Infinity is a big number, so to speak. So no wonder we are extremely valuable. Given Finite Value Hypothesis, by contrast, Equal Human Value is unexpected. It would be quite surprising, given Finite Value Hypothesis, that we are all deeply and equally value. For equal value would require that there be some particular finite degree of value – 61.2 axins, for example – such that we are each valuable to exactly that degree. Given our variation along other dimensions, this particular equality would be a surprising result indeed. It is possible, yes – just as it is possible that all the paintings in our parable have the same degree of value. But equal finite value across wide diversity is surprising and unexpected. On Finite Value Hypothesis, there is no simple and compelling explanation of our extreme and equal value. So we judge the probability of Equal Human Value on that hypothesis to be rather low. One might, of course, supplement Finite Value Hypothesis with various auxiliary hypotheses. One could add that selective evolutionary pressures or mysterious and powerful aliens only allow the birth and continued existence of human beings that have exactly 61.2 axins of value. And the probability of Equal Human Value on this more complicated view might well be high, even 1. But this maneuver comes at familiar theoretical price. Adding extra conjuncts to a hypothesis can always expand the evidence that hypothesis can accommodate; but it drives down the prior probability (or, simplicity, if you prefer) of that hypothesis. Proponents of Infinite Value Hypothesis need not add such auxiliary hypotheses to their theory to explain or accommodate the evidence that is Equal Human Value. We have all the ingredients we need, then, for a novel argument. It unfolds as follows: C1. Equal Human Value C2. P(Equal Human Value/Infinite Value Hypothesis) [=1] >> P(Equal Human Value/Finite Value Hypothesis) [=low] C3. If so, then Equal Human Value is strong evidence for Infinite Value Hypothesis 7 C4. Therefore, Equal Human Value is strong evidence for Infinite Value Hypothesis (from C1-C3)

I agree that having equal value would be much more likely if humans had infinite value. However, I don’t think that humans have infinite value, so this doesn’t matter. Furthermore, there are devastating arguments against humans having infinite value.

Presumably headaches don’t have infinite disvalue. However, if human life has infinite value then shortening a human life would have infinite disvalue, much like removing an infinitely valuable art piece from an art collection a second earlier would have infinite disvalue. If this is true then shortening a human life by 1 second would be worse than thousands of non lethal headaches. Additionally, this would mean that a quadrillion animals being tortured to death would be less bad than one human having their life cut short.

If a human life has infinite value then bringing one extra human into existence would be worth all sacrifices that don’t eliminate human life. This is very implausible.

Additionally, if human life had infinite value, given that 2 times infinity = infinity, this would mean that saving one human would be just as good as saving two humans. This isn’t plausible.

When responding to the earlier objections to human life being equal, Rasmussen and Bailey could argue that humans have equal intrinsic value, but different instrumental value. So while all humans have the same intrinsic value, if one will have a better life, it’s more important to bring them into existence. However, if they hold that humans have infinite value, then they can’t hold this, because any non human intrinsic value would be infinitely dwarfed by the intrinsic value of humans.

This also runs afoul of their second argument, which says

Our second argument unfolds as follows:

I1. We are all equally overall valuable

I2. We differ with respect to instrumental value

I3. If so, then either we differ with respect to final value, or we are all overall infinitely valuable

I4. Therefore, either we differ with respect to final value, or we are all overall infinitely valuable (from I1-I3)

I5. We do not differ with respect to final value I6. Therefore, we are all overall infinitely valuable (from I4-I5)

This leaves them susceptible to the earlier objections to infinite value. On this account, drinking chemicals prior to pregnancy that will lead to a future baby having a worse life, would be morally neutral. After all, both infants that would be had would have infinite value.

They address one objection that I think doesn’t work, so I won’t bother quoting their response to that objection. Next, they address my basic objection (Albeit a very abridged version).

Another objection: the arguments presuppose that human beings are all equally and extremely valuable. This premise, however, is not a datum; rather, it is false. Here are three reasons. First, there is little reason to suppose we are all equally finally valuable. This equalvalue idea is a fiction; it is perhaps useful for modern liberal democracies – but hardly a sober truth. Human beings differ in all sorts of ways, including their final value. Second, it is unclear what final value could even come to here. It is probably best interpreted as (to borrow Peter Singer’s words) “obscurantist mumbo-jumbo.” Third, there is no reason to elevate human beings above the rest of nature as done in the arguments at hand. We are not angelic beings distinct from natural reality; we are parts of it, and subject to the same finitudes as the other animals (Singer could probably be recruited in defense of this final reason as well).

We reply: we have no argument for the premise in question and we will not attempt to refute these three promising objections. But we do have two comments. First, the equal and extreme value of human beings can do much to explain a wide range of ethical phenomena. That we are extremely valuable nicely accounts for the deep wrongness of murder, for example. That we are equally valuable nicely accounts for the wrongness of inequitable treatment. For this reason, we think the equal and extreme value of human beings is a hypothesis worth testing and extending, as our article does. Second, our arguments are of some value even for those who reject their starting premise. For our arguments uncover this interesting conditional: if we are all equally and extremely valuable, then there is very good reason to think we are infinitely valuable. Some may take this as a reductio of the antecedent or as a reason to accept the consequence; such is philosophy. But either way, our arguments shed new light.

I don’t think that this is able to explain things very well. Equitable treatment doesn’t follow from equal value. Murder’s badness can be explained on a utilitarian account. I think that their theory, even if able to explain two instances of bad things, is not very satisfactory. After all, nearly all theories can explain infinite bad things. This is thus a fairly shoddy criteria. I think their second point is fruitful and am in agreement. Because humans are not of infinite value, we cannot be of infinite value.

In terms of the argument attributed to Singer, one could adopt a much more modest principle to defend the view. To hold the view that humans, unlike animals, are infinitely valuable, they’d have to hold that torturing 1000000000000000000000000000000000000000000000000 animals in order to produce one extra human would be good overall, producing infinite value at a cost of merely finite value.

Finally, if human lives have infinite value, then we should ban nearly all risk taking. After all, we wouldn’t allow people to risk ruining the infinitely valuable painting. Heaven is often supposed to have infinite value—but we wouldn’t risk heaven being destroyed for trivial benefits. Thus, if humans have infinite values we should never take risks that could jeopardize our infinite value. We should never cross the street or drive, merely to have fun with friends. This would be risking infinite value for a finite value.