Better to have existed than not at all?

A reader sent me the following question about eating "happy cows". I will not be able to get to this until tomorrow or Wednesday, but thought others might like to discuss in the interim.
If I may, I was wondering if I could get your take on a common objection concerning the raising of humanely treated animals, that the alternative is non-existence, and it's better for them to exist than to never have existed. Jeff McMahan says this claim is incoherent ("Eating animals the nice way"), but I'm struggling to understand why.
Suppose the cow Betty has experienced 100 net units of happiness, and one claims it is better for Betty to exist than to never have existed: because, if Betty never existed, she wouldn't have experienced any units of happiness at all. Thus the claim makes a comparison between existent Betty and non-existent Betty, claiming that the latter would have been worse off. Does the incoherency arise because, to say that non-existent Betty would have been worse off, we must assume that she would have somehow been an existing, non-existent individual who failed to experience any units of happiness?
-- Spencer Lo
My answer--

OK, so there's a defense of benign carnivorism that says we are helping Betty (taking her from worse to better) by creating her, giving her a nice life, and then painlessly killing her and eating her. We'd actually be harming Betty if we kept her in the hell of non-existence, so to speak, and didn't let her into the actual world.

I think McMahan is saying* all that's sheer nonsense.  To help Betty, she's got to already exist.  If a being doesn't exist, we can't help her, can't make her go from a worse condition to a better condition. Likewise, we can't harm someone by creating her--she doesn't go from better to worse by coming into existence (even if her life will be full of pain).  Bottom line--we should reject all the helping and harming talk. 

This is very signficant, as we tend to think there are rather strong duties to help and even stronger duties not to harm.  So if we were helping animals by creating them, treating them nicely, and then eating them, that would create a pretty strong case for doing so. If we were harming animals by keeping them out of existence, that would create a pretty strong case for getting busy and breeding more of them.  So jettisoning all the helping/harming talk is a major step.

Now, even if we take that step, says McMahan, we're still left with some reason to keep breeding animals, treating them nicely, and eating them. That's because even if we don't help Betty by creating her, we can still say we do something good, assuming there will be a lot more happiness than misery in her life.  In fact, we do something good for Betty, since she's the one who will enjoy that relatively happy life.  We don't help her by creating her; we simply do something good, and it happens to be good that Betty enjoys.  As McMahan says (p. 3) "Since benign carnivorism by definition aims to cause animals to exist with lives that are good -- in which the good elements outweigh the bad -- it is plausible to say that the practice is good for the animals it causes to exist, even if the ultimate aim is to make them available for human consumption." 

Using terminology developed by Derek Parfit, McMahan later in the article says that our reasons to create animals are "impersonal" (p. 6). That is to say, they are not helping/harming type reasons, but just "doing good" type reasons. There's no victim if we don't create happy animals, and no beneficiary, in the sense of someone who winds up better off, if we do create happy animals. That then leads to the crux of the matter.  He says our reasons for benign carnivorism consist in "the human interest in eating meat, and whatever impersonal reasons one might have to cause animals to exist with lives that would be good for them."  He then says "In general, we assign little or no weight to impersonal reasons to cause individuals to exist." (p. 6)  We don't think we need to get to work and make more babies, for example, if we prefer to be childless. It's fair to say we'd be "doing good" if we made more people, but that doesn't make much of a case for making more people. It makes even less of a case for making more animals, since animals have less good in their lives (on McMahan's view).

My reaction to this: it's actually a major mystery why "we assign little or no weight to impersonal reasons to cause individuals to exist," so this seems like a very shaky way to respond to someone defending benign carnivorism.  Peter Singer doesn't respond to the argument this way, as he's a "total utilitarian" and does assign weight to these impersonal reasons. His very complicated discussion of this in Practical Ethics might be the thing to read next. 

Hope that was helpful!

* Article is HERE.


Wayne said...

No need to assume impossible cows here (existing non-existent cows are impossible cows). The comparison is either between utility states, or subjective states.

If its utility states, then we're comparing the world with the cow, vs without the cow, and how much utility is in that world. We're supposed to come to the conclusion that we need THINGS (not necessarily cows) to be happy, and a world with more cows that are treated humanely, is one that maximizes utility.

If its subjective states, then we're simply asking, what would the COW experience, and which is preferable. Would a life of some happiness, with some terror, be better, than never experiencing anything at all? Non-existence, isn't a subjective state at all! So in non-existence, the cow could never be harmed... since it doesn't exist. If we're about minimizing harm, then we should prefer the cow to not exist, even if its existence does cause some pain and suffering to itself.

Or we could take the full utilitarian route and say that the cow should exist, treated well, and shouldn't be eaten. This maximizes utility, minimizes pain (arguably...)

Anonymous said...

It depends how you set up your measure of "good" and "better". It seems to me that A is better than B if A is "more good" than B, and makes sense if we can measure the "goodness" of both A and B. If the individual doesn't not exist, I don't think we can say that anything is good (or bad) for him. Thus, nothing can be better for him. So, existence is no better (and no worse) than non-existence.

Now, some people (in particular Derek Parfit ⇒ "Reasons and Persons") believe that we can say that something is "good" even if it is not "better" than anything else. That part always puzzled me, and I cannot say that I really understands what he means there.

In general I think that measuring everything in one dimension is quite reductive. Can we say that a life with a lot of pain and lot of happiness is better than a life with a little of both? How can we reduce this to only the sum (or difference) of the two?

Also, there seems to be strong asymmetries that play an important role here (see David Benatar, http://amzn.com/0199549265 )

Spencer said...

I blogged about Jeff McMahan's paper in which he made the claim: http://animalblawg.wordpress.com/2012/09/11/whats-wrong-with-happy-meat/


I think McMahan would say we need to distinguish two claims:

1) It is better or worse for an individual cow to exist than to never have existed.

2) It is good or bad for an individual cow to exist than to never have exist.

According to McMahan, (1) is logically incoherent but (2) is not, and we can capture the moral significance of bringing a happy or unhappy cow into existence with (2). And I think we can distinguish two further claims:

3) It is better or worse for the world that an individual cow was caused to exist than to never have existed.

4) It is good or bad for the world that an individual cow was caused to exist than to never have existed.

What would McMahan say about the coherency (3) and (4)? (4) is coherent, and possibly (3). But if (1) is incoherent, I wonder if it follows that (3) is as well.

Wayne said...

Is 1 incoherent because there is no cow to compare to? If so than 3 could be coherent.... Unless he doesn't like talking about possible worlds.

Spencer said...


Yes, I think that's what McMahan is saying: the terms "better" and "worse" are comparative, so (1) implicitly involves comparing the well-being of an existing cow to one who never existed. But it doesn't make sense to talk about the well-being of someone who never existed.

Tom said...

I replied to an analogous argument just recently.

Aeolus said...

Ultimately, in the article, McMahan comes down against "benign carnivorism". He says, "The animals’ interest in continuing to live outweighs the human interest in eating them. That those who now want to kill the animals had earlier caused them to exist-–an act that was good for them-–is, at this point, irrelevant. One cannot plausibly claim that in killing them one would be depriving them only of what one gave them in the first place. That justification would allow parents to kill their children. Whatever good the practice has bestowed on animals up to this point cannot be cited as credit from which the killing can now be debited."

Spencer said...

In case anyone is interested, I blog about McMahan's article here: http://animalblawg.wordpress.com/2012/09/11/whats-wrong-with-happy-meat/

Jean Kazez said...

I added some thoughts to the post, but I should say-- I haven't had time to study the entire paper closely.

Spencer said...

Hi Jean,

You say that on McMahan's view, "we can't harm someone by creating her--she doesn't go from better to worse by coming into existence (even if her life will be full of pain)."

But I wonder if there is a conception of "harm" that might be consistent with McMahan's view that creating Betty would be bad for her. For McMahan, if Betty's life will be fully of pain, causing her to exist would be bad for her. A natural way of interpreting this claim is that causing Betty to exist harms her, specifically by causing her to experience a painful life. This notion of harm doesn't seem to involve comparing her well-being with her well-being when she never existed.

Jean Kazez said...

I think there really is a major difference here that he's trying to flag. Take "helping" first. If someone's in a bad situation--trapped in a closet for example--you help them by getting them out. "Harming" or "victimizing" is just the opposite--you force them from freedom into the closet. Helping or harming in this sense are clearly things you can only do to someone who already exist--you don't help non-existent people and you don't harm/victimize them either, by leaving them nonexistent.

Now it's true, if you create someone, they'll be either in a good state or a bad state (non-comparatively speaking). You will have done something good (for them--since they get they good) or something bad (for them--since they get the bad). If you called this "helping" or "harming" we could follow (sort of) but you'd just be creating a lot of confusion, as it is nothing at all (morally) like "helping" or "harming" in the usual case. I think that's the idea here--basically, don't conflate creating a new individual with pushing someone into a closet or pulling them out of a closet. There are much stronger imperatives telling us to help and not harm in the closet type case.