I eat chicken and not beef because I don't believe in eating something I could not personally kill. Also, I try to only buy free-range chickens, which I think minimizes the suffering. It is too bad that PETA turned evil.
Methane lasts in the atmosphere for 12 years. So a cow today is merely "replacing" the methane emitted by a cow in 2009. Which was replacing the methane from a cow in 1997. Replacing 1985. 1973. Etc.
Therefore eating beef today is not adding to greenhouse gas impacts on global warming, as long as the total number of cows is fairly level. It has been decreasing in my country for the last few decades so I feel no moral imperative whatsoever.
I would welcome arguments against this position. It seems to me like we expect everything to be net zero except cows, which must lead to reduced greenhouse gases.
This will sound like a contrarian troll comment but I mean it quite sincerely
Why does the calculation for the net moral cost of beef vs chicken not factor in the utility gain/loss from each of those in your diet?
I can't speak for other people but I can say personally that I would be experiencing quite a bit of suffering personally if I could never eat beef again, but I would not be experiencing that level of suffering if I could never eat chicken again.
I am in general resistant to utilitarian reasoning like this, for reasons I have expounded on at length in your various comments sections throughout the years. By framing everything in terms of this, people (or, at least, rationalists) are put in a position where they feel that their own preferences are invalid unless they can be justified on utilitarian grounds. It's not that hard of a jump from there to "my preferences don't matter" or even "my preferences don't exist"
But they do exist! And they do matter! It is fundamentally meaningful to factor in the cost _in human suffering_ of dietary restrictions when you're running these numbers.
(Further, I would remind you all that each and every one of us in this space are likely 5-sigma deviants from normalcy on at least a few critical dimensions. I imagine most of your instinctive reactions to my position is some variant of "come on, don't be ridiculous, nobody cares that much about their diet". I think most people do, and it's us here who are unique in that we don't care as much)
This seems like a great use case for having a functioning government. If there are really these kinds of moral and ecological costs to individual consumption choices, rather than push off the calculations and offsets of exactly what they are to individual consumers who are purchasing thousands of distinct items a year, reflect those costs directly in the price of goods via taxation that internalizes externalities like these.
Precision isn't accuracy. The numbers cited for ghg are laughably inaccurate, though stated with precision. This is why economists are always so wildly mistaken. They take numbers someone pulled out of their butt and run them through a formula that is utterly unrepresentative of the system.
This is an argument against Australian wheat specifically, but more generally there are clearly harms to animals from row-crop agriculture....as a vegetarian I don't quite know how to feel about this. (if you are an omnivore using this as a whataboutist argument, I hope you are already avoiding farmed meat other than grass-fed beef!)
I think all of this makes a lot more sense if you add in "and of course, you're already spending some amount of money to do good in the world, however large or small that amount may be".
Assuming you're being efficient with that money, it kind of obviates the Yog-Sothoth problem. You have already chosen to donate/spend/forgo a certain amount of money, and that implies a certain price at which you're willing to purchase goodness for the world.
Decisions like whether to eat chicken will be priced appropriately; since you're already donating to the Stop Yog-Sothoth Fund, you are indifferent about whether to donate a marginal dollar, so if eating chicken is really worth a dollar to you, it must also be worth destroying a galaxy. Because that's your BATNA! That's what you would've done with that dollar otherwise (approximately).
You get counter-intuitive conclusions when you cluster supererogatory and obligatory actions together. If eating animals is morally similar to murder, then this sort of utilitarian offsetting is morally puzzling. Can I commit a murder and then donate to prevent murders? Commit an assault and donate enough to offset it? What if I donate enough to save 10 lives through effective charities, can I then murder someone and be considered a morally good person? It seems like morality does not work that way.
I don't understand why I should take the concept of superrationality seriously. To me, it seems based on an obvious category error. As per the wikipedia article linked in the post:
"The idea of superrationality is that two logical thinkers analyzing the same problem will think of the same correct answer."
The category error is in the statement that a game like the prisoner's dilemma has a correct answer and has wrong answers.
Perhaps I am leaning too much on the "correct/incorrect" wording used on the Wikipedia article, though.
Unless I'm missing something, wouldn't you want to consider the price difference between beef and chicken? I'm sure it depends where you live, but generally I expect to pay something around 20% more for beef than chicken at a restaurant, and more than that at the grocery store. Without calculating, I expect the savings from purchasing chicken will be significant and likely enough to offset the more expensive moral cost of chicken (assuming you use the savings for that purpose).
I think the assumption that causing a chicken/cow to be killed is worse than causing that chicken/cow to never come into existence in the first place needs to be examined. If the opposite were found to be true, we might instead conclude that we need to eat as many chickens as possible in order to maximize the number of chicken lives lived.
If we are doing a utilitarian calculus and incorporating animal suffering and death, then it is not clear to me that hastening the demise of human life is not a net increase in utility. Currently, around 3 billion animals are killed daily and billions more live in inhumane conditions. So in 3 days, there have been more animals killed than humans alive. If climate change accelerated the death of all humans, then couldn't this plausibly be a utility increasing good thing if it ended factory farming?
I do not believe in utilitarianism and I am not vegan but I am sympathetic to these arguments. However, when you put them together, it seems like it is better to kill all human beings alive. Can someone explain to me the error in my reasoning if it exists?
Assuming that the choice is "factory-farmed chicken" or "some kind of beef, I dunno, normal beef I guess" weakens this argument, although it also helps keep it simple. The 30-days-of-horror life lived by a factory chicken is indeed so horrible that I (an omnivore) will go out of my way to avoid contributing to it. The life lived by the fancy free-range bug-foraging high-welfare heritage chickens sold by my local butcher is… I mean, I don't know all the details, but as far as chicken needs go I think it's pretty ok up until the final butchering scene.
So with that calculus in mind, you can reduce carbon emissions AND reduce suffering AND get to eat a much nicer grade of chicken. (You can do 2 of the 3 by buying fancy high-welfare free-range grass-fed beef, too, and sometimes I do that.)
Does America really have like 99% factory chicken or has free-range made some inroads since I left there 10+ years ago?
Pasture raised meat is even cheaper than Beyond Burgers. An ethically raised meat animal gets a happy life and a better death than most wild animals. There's just no reason to continue worrying about meat alternatives. We just need to stop doing the horrifyingly awful thing and get back to doing the actually great thing.
"A factory farmed chicken lives about thirty days, usually in extreme suffering."
Unless I missed it, the above seemed to be as close as you get... but why not touch on the *difference* in conditions than factory-farmed cows vs. chickens are raised in? My understanding - which I think is reflected in Brian Tomasik's table here: https://reducing-suffering.org/how-much-direct-suffering-is-caused-by-various-animal-foods/ - is that chickens suffer much more in captivity than cows.
I'd be interested in a "Carbon offests: much more than you wanted to know" type post that investigates the situation in consumer-level carbon offsetting. I've bought some in the past and couldn't quite shake the feeling that it was, uh, somewhat imaginary.
Can someone explain how the offsetting transaction works? Party A gives $ to Party B to offset X. In exchange for $, what does Party B do that offsets suffering or carbon emissions or whatever?
The mechanics seem super hand savvy to me. It looks like the old Catholic indulgence where parishioners paid for their relatives to spend less time in purgatory. Both indulgences and offsets sound unmeasurable and unenforceable.
"Meanwhile, if you don't eat some chickens, those particular chickens don't get eaten."
*Surely* this isn't case? Those chickens will still get sold to other people and then eaten, possibly at reduced price if your refusal to buy them affects demand. The real calculus will have to be how the reduced demand from your refusal to chicken affect prices and supply in the medium to long term.
Personally, I care about my own health more than I care about chicken vs cow suffering. Can we express these numbers as "chickens saved per expected day of decrease in the eater's lifespan"?
Chicken and Cows are largely carbon neutral. All of their carbon emissions are part of the normal cycle. This is similar to why burning new-growth wood is carbon neutral.
I think that offset prices are still meaningful in the presence of "market failures" as long as the offsets themselves actually work. In other words, as long as it is, in fact, possible to stop Yog Sothoth from consuming a galaxy for $1, I think it's correct to say that eating a chicken (and thus arousing the hunger of Yog) has a $1 moral cost.
If it's actually possible to abate the hunger of Yog with a $1 donation, every $1 purchase made is a missed opportunity to spare a galaxy of intelligent life from unimaginable suffering. Therefore the only reason that Yog should be consuming any galaxies is because every person with dollars has decided not to forgo $1 worth of consumption in order to save a galaxy, _including_ the person eating the chicken. Spending $3 on a coffee in this world is exactly 3X as bad, in terms of destroyed galaxies, as consuming one free chicken.
And likewise, if it's actually possible to abate a ton of carbon with 33 cents, then it's genuinely possible to trade 3 tons of carbon for one galaxy. Every person in this world who gives any amount of money to the Stop Yog fund and the Clean Energy Initiative (or whatever) must be willing to trade one galaxy for 3 tons of carbon. In a real sense, the moral judgment of this society is that a galaxy is as bad as 3 tons of carbon.
But how about market failures? Any traditional sense of market failure should refer to:
1) Externalities
2)
3) Information Asymmetry
The obvious one here is externalities: the costs of Yogging and polluting accrue mostly to people other than the Yogger/polluter. But in both cases, what we're concerned about is pure externality, so it's hard to think about externalities as causing more of a market failure in one context relative to another. Any internal costs and benefits of Carbon/Yog aren't really part of the moral calculus to begin with.
Market power is essentially irrelevant here. It's not particularly relevant if the stop yog fund is charging above-market prices for Yog abatement: I can still get them to Yogproof a galaxy for $1.
And information asymmetry boils down to saying that I might not really know if my donation is stopping Yog. Maybe my donation saves a galaxy, maybe it goes toward a cardboard sign that says "say no to yog" on a galaxy that was never appetizing in the first place. If I can't tell the true Yogsbanes from the grifters, maybe I keep my dollar in my pocket. But in this case, it's just not true that it costs $1 to stop Yog from eating a galaxy, because it's not true that I could give up a dollar to save a galaxy.
This sort of Singeresque utilitarian chain of logic has led me to the conclusion that very few people actually believe in catastrophic anthropogenic climate change; my thought is it’s actually a kind of character signaling, like wearing a mask after being fully vaccinated: “I recognize that doing this benefits no one, but I want to send the signal that we should take stronger collective action on similar problems.”
I struggle with these kinds of estimates of carbon cost. A cost of $10/ton where the average American releases 17.5 tons per year implies that all of America's contribution to global warming can be erased for less than half a dollar per person per day. I think if that was true it would have happened already.
Bit off topic but maybe still worth thinking about:
I think the offsets will be larger. The price of, say, 10$/ton of CO2 is low at the moment because there are currently quite a few low hanging fruits for CO2 offsetting. As soon as this will be more popular, prices will go up, not/not only because of capitalism, but also because it will be more and more difficult to find cheap ways for compensating a ton of CO2. Otherwise, I mean, let each citizen pay 17.5 * 10 = 175$ climate tax each year and the problem would be solved.
This is pure speculation, and I would much appreciate a more informed reader's comments on it.
I am endlessly perplexed (and a little vexed) that otherwise-sensible-seeming people assign moral value to the suffering of animals who cannot possibly reciprocate. What am I missing?
I would think health differences should factor in to this decision.
But also, it's hard for me to wrap my head around the mindset that one could take animal suffering or CO2 emissions from meat eating seriously enough to do this math and alter their diet or pay for offsets but not just...stop eating or rarely eat meat.
While not yet confirmed to be ransomware, let's assume it is. In the moral calculus outlined in Scott's post surely the cyberattack is a net good since the reduced cow suffering and cow related carbon emissions caused by the attack outweighs the CO2 involved in the bitcoin transaction that will be used to pay the ransom, no?
Does anyone actually care about any of this with respect to their dietary decisions. For most people these are visceral not mathematical decisions. For example, after having visited (my father’s) chicken hatcheries, production and beef packing plants (for all, read “slaughtering” houses) when I was about 9, I haven’t eaten any bird or mammal. It’s not a function of any rationalist judgment. At some point it was clear that I couldn’t consider eating certain things. For about 20 years I was a vegetarian. Then, I tried to distinguish the source of my eating habits. I realized that I couldn’t abstract from the killing process. But, there was some killing I could tolerate and would do. Namely fishing. Although I didn’t then eat fish, I had fished and gutted fish. That truth led to my starting to include fish in my diet. Not because it was more sustainable (it may not be) but because I would kill it. Therefore, I would eat it. Can I justify it by neuron count? No. Do I care? No.
When I see a hamburger I see a cow — one I would pet. When I see fried chicken I see my neighbor’s chickens whom I know by name and have held. These aren’t calculations. They’re either psychologically instinctual or ontologically driven.
To wit, most people who try to be vegan or vegetarian for environmental (CO2) reasons end up struggling and at some point either failing or internalizing a moral and visceral perspective. Again, rationalism largely fails here. It’s similar to the way that people stop exercising their new year resolution to work out by February. Either they internalize it into some ontological part of themselves, or they struggle and fail.
Given all that, the reality for most people is that mammals feel to us more akin us than do birds. We relate to them. We kiss our dogs, hug our horses, feel the run our hands over the velvety soft nose of cows. Birds seem more foreign, easier to prey upon. Again, not necessarily justifiable (as you demonstrate) but your argument is taking place in the wrong universe. That makes it irrelevant to actual dietary decisions.
Tongue only loosely in cheek, this ends up seeming like an excellent argument to donate to charities which provide contraception. Assuming that <$175/year provides funding to prevent at least one new human (which seems like a reasonable guess) you can "offset" your entire life's consumption of not just meat but all carbon-emitting goods and services for the same or less than commercial carbon offsets. Plus improve the quality of life for women in need of health care!
If you are doing comparisons of the number of suffering chickens vs. number of suffering cows, should you also multiply by their respective lifespans? (i.e. how long they suffer for) Why or why not?
Somewhat nitpicky -- but the CO2Equivalent numbers you're using for methane are on a 100-year timescale. When we consider the CO2E of methane on a 20 year time scale, it's 3x as high. (put plainly: methane emissions from cattle burps are responsible for 20% of all warming in the next two decades).
The linked guesstimate (https://www.getguesstimate.com/models/10897) in the EA post you cited for the chicken estimate (https://forum.effectivealtruism.org/posts/9ShnvD6Zprhj77zD8/animal-equality-showed-that-advocating-for-diet-change-works, "But what if it were chicken?") seems to say that it actually costs $0.82 to spare a chicken, not $6; it gives $6.64 for the cost to avert a chicken year. But the EA post itself says that it costs "$5.70 per chicken spared (90% interval: $0.71 to $32) and $50 per chicken year (90% interval: 6.3 to 280)". I'm confused. Maybe either the EA post or the guesstimate were updated without the other also being updated?
Possibly too tangential, but: Did anyone ever figure out why Yudkowsky is so confident that most animals, including chickens and cows, don't have qualia? (Referenced but not spelled out here, as well as some other places: https://rationalconspiracy.com/2015/12/16/a-debate-on-animal-consciousness/.) I'm very curious what his (presumably partial) solution to the hard problem of consciousness is.
What about the price of the meat itself? Chicken is significantly cheaper per calory than beef, leaving me with a budget surplus that I could then use to pay for the chicken offset. Or effects on your person health for that matter, my impression is that chicken is healthier than beef, and being healthier would allow me to work more overtime, earning more money once again.
I think the model here is far too simplistic, which is my issue with effective altruism in general - reality is complex and I could easy choose a different set of effects to focus on to argue for the opposite conclusion.
Does marginal spending on carbon offsets actual cause more carbon to be offset? I worry that we have, say, 1 billion tons of carbon offsetting capacity right now and if more people want to carbon offset, the price just goes up without the capacity increasing. Now increasing the demand for carbon offsets would increase the incentive for people to build out carbon offsets and that's plausible, but does it happen at a 1:1 rate? I'm sure economists have tried to quantify stuff like this and I'm just ignorant of the terminology.
I think this mostly shows how the "offset" concept is flawed. It's not what consequentialism is trying to get at - the offset donation is not causally connected to eating the meat - and it's used as a 'noble lie' to get people to donate at least something to charity despite being reluctant to give an altruistic amount, but like all noble lies it's incompatible with clarity of thought. Trying to staple it together with economics and its (also fictional) perfectly rational actors is a case of making two simplifying assumptions that contradict each other. Ultimately, the answer has to be to do the math on "direct animal suffering" vs. "risk of apocalypse / human suffering from global warming", and only consider money if actually making a charitable donation. The former is probably bigger, and certainly easier to quantify, which feels to me like it should carry more weight somehow.
For my part, I follow a principle of not eating mammals, which I adopted a long time ago because it's a relatively bright line while leaving me able to eat *something* in most restaurants / social settings. For myself I buy fish. In principle beef causes less suffering than chicken, but I've kept to the deontological rule in this case, though perhaps I should change that.
(On neuron counts - it's not obvious to me whether the function should be steeper than linear. Killing a bunch of individual neurons doesn't seem as bad as killing a single animal with all of them connected... but, OTOH, linearity is stable with regards to fusing brains together, so maybe it's necessary to have it as the first approximation.)
To me the moral calculus seems even more lopsided than that. I don't really care about the deaths of animal who don't make plans for the future, I care about their suffering and quality of life. On that basis chickens have utterly terrible lives and while cows experience some suffering in their lives, particularly at the end when they get shoved onto feedlots, but it seems that on net they probably have lives worth living? I think I'd rather be reincarnated as a cow than face non-existence.
> "Meanwhile, if you don't eat some chickens, those particular chickens don't get eaten."
Well, it will depend on the price-elasticities of the chicken supply curve and the chicken demand curve, right?
E.g. on the supply side: Perfectly elastic supply means refusing to eat 1 chicken causes 1 fewer chicken to be eaten (and the price will stay the same); perfectly inelastic supply means refusing to eat 1 chicken causes 0 fewer chicken to be eaten (and the price will go down). (And then vice-versa for demand.)
In reality I assume the elasticities are somewhere in between, so on average every chicken that someone refuses to eat saves somewhere from 0 to 1 chickens on average.
Anyone have any empirical data on the elasticities of the chicken supply and demand curves? Or for other animals, for that matter — I guess all else equal, you can do more good the more elastic the supply and the more inelastic the demand for that animal.
> (it's still totally fair to eat chicken and donate $1 to the Stop Yog Sothoth Fund, it's just catastrophic if you abstract away the step where you donate the dollar)
that doesn't really seem like a perfect utilitarian calculation. of course one option is to eat a salad and not donate, and compared to that eating the chicken and donating seems fair. but another equally plausible option is to eat a salad and also to donate, and compared to that the expected utility of eating chicken and donating is terrible (one whole world lost).
> Either carbon offsets, animal suffering offsets, or both could be market failures like this. In fact, if you believe that we're currently not doing enough to fight climate change or animal suffering, I think you almost have to believe there are market failures in offsets.
how about believing that there are no market failures, but that there are low-hanging fruit, and the current offsetting prices are determined by changes on the margin, whereas average offsetting prices – if we tried to reduce animal product consumption completely to zero, say – would be much higher.
This is missing two important points. 1. Chickens have much shorter lives than cows and 2. The typical farmed chicken's life seems more miserable than the typical cow's.
The dilemma between decreasing CO2 but saving the lives of chickens vs cows is very much like Chappelle's genius bit about comparing Bill Cosby to a superhero that saves lives but rapes. "That's the dilemma for the audience. Because he rapes, but he saves a lot of lives. And he saves way more than he rapes, and he only rapes to save. But he does rape.”
"Meanwhile, if you don't eat some chickens, those particular chickens don't get eaten." Not sure I agree with the argument here - the marginal chicken you decide not to eat is already dead and will now land in the bin. Your individual actions may have no impact but that doesn't absolve you from your moral responsibility.
So this is vaguely related, and hopefully someone here can help me.
I've been trying to find information on the "ideal" diet from a climate perspective. This article is combining climate and ethical considerations (which I think is totally worthwhile), but if, for a moment, one completely ignored the ethical considerations of eating meat and _only_ cared about the climate impacts of one's diet, what would that diet look like?
From general reading/information floating around, I _feel_ like the "climate optimal" diet would a) involve a lot less meat consumption than your average American diet b) would probably involve very _very_ little beef but c) would not be entirely vegan/vegetarian.
However, I can't find any good sources that talk about this, explicitely.
For context, I ask because, given where I live and my living situation, I have the ability to get all or nearly all of my meat in a way that is, to my personal satisfaction, ethical. But, from a climate only perspective, I'm not sure how I should be allocating my diet.
The simplest answer I suppose is "don't worry about it, just buy offsets", which _does_ sort of solve my individual level problem, but I'm sort of interested in the long term societal options for how to feed everyone with the most climate-netural calories possible.
As a final note, I have been saying "climate impacts" this whole time, but that was mostly as shorthand. I am also interested in more general environmental impacts like water use, fertilizer runoff, etc. I know that that makes it more complicated so I'd accept sources that talk _only_ about climate impacts (CO2 equivalents etc.), but anything that _does_ incorporate more general environmental impact information would be great.
Yes, but to further complicate things - assume my charity budget is fixed. I donate x% of my income. Right now it goes all to givewell, but I could divert some of that to a carbon or chicken offset program. Now we have to include the opportunity cost of the human suffering by african children not having malaria nets, for example.
quick correction to your math in the second paragraph:
>switching from an all-chicken diet to an all-beef diet saves 60 chicken-equivalents per year.
by your own estimates, it should be (80 chickens - 0.5cows*20chickens/cow) = 70 chickens. Or another way of looking at it is that you would reduce the animal suffering cost by 7/8
If we're trying to account for global warming/climate change, shouldn't you also be thinking about (1) the cost of feeding cows versus chickens (I'm not an expert but I think it's way more carbon intensive to feed cows, especially if you need to fertilize any pastures, and that's one of the biggest environmental reasons to avoid eating beef), and (2) a lot of modern farming involves transporting cattle between different locations depending on where they are in their lifecyles and there is a lot of CO2 emissions in moving them?
Well, I roasted a chicken yesterday and I'm thinking of boiling some beef tomorrow. I can't work out if I'm part of History's Greatest Monsters or Doing My Bit To Fight Carbon Emissions 😀
As far as moral concern goes, I think it's right to act your rational conviction, but I can't honestly surmount my own doubt that it makes sense to care about animal wellbeing. Animals are just a significant degree less interesting than humans. If I really am to say that chickens have moral worth, I don't see any easy spot to get off that train between chickens and insects. It annoys me when animal rights activists act like it's SO OBVIOUS to care about chickens but don't have an extremely OBVIOUS reason to ignore the potential moral worth of mosquitos or flies. The human race has deployed a gene-drive edited mosquito population in Florida to eradicate a particular kind of mosquito, we're literally doing very-intentionally-engineered genocide on some mosquitoes and while there are detractors of this scheme their criticism seems mostly centered on gene-drives being potentially dangerous and hard-to-control technology with unpredictable outcomes. As far as I've followed the issue, I haven't heard the perspective that we ought not genocide this population of mosquitoes because they have moral worth.
I can, however, contrive a reciprocity test. Which is that cattle are very kind creatures. Chickens are not. Pigs are not. Insects are not. You would not leave a young child in alone in a barn with chickens or pigs as you would be concerned that they would get eaten. You would not have the same concern with cows, apart from getting accidentally kicked or stepped on. I think this makes sense and is not-hypocritical. It's roughly the same logic we apply to humans under liberalism: as long as that human would not harm me, I have no right to harm it. But if we have evidence to believe that a given human WOULD likely bring harm to others, we lock them up.
I think a similar heuristic extends to animals reasonably well, and it leads me to eat chickens and not cows.
Something about the offset pricing seems wrong. If it's true that the average American produces 17.5 tons of CO2 per and equivalent per year, and that a ton of carbon can be effectively offset for $10 per ton, then an average American could offset his entire carbon output for $175 per year? And the entire country's annual CO2 output could be offset for $58 billion per year?
That's objectively a lot of money... but also not that much. Before we consider economies of scale, it's about five times as big as the Pentagon's fuel bill. That's less than twice the Department of Energy's budget. But all this is orders of magnitude cheaper than the "Green New Deal."
If we set annual costs aside and look at renewable energy projects, this also doesn't seem to make sense. The Topaz Solar Farm costs $2.5 B to build and provides power to 160,000 homes. That's orders of magnitude more expensive. To be very conservative, assume a 5% rate of return to private capital, no operating costs, and that a third of an American's carbon emissions are due to home electricity.... you're still looking at a capital cost of $125 M per year to offset the emissions of 60,000 American households, or $1,000 per person offset in capital cost alone.
If the costs involved in offsetting carbon emissions were really so low, this wouldn't be a major geopolitical issue. You wouldn't need the US and China to agree to any climate treaty, or agree to do anything at all, the EU with its strong Green Parties could simply fix global warming by spending 1% of its GDP.
This logic always seems weird to me because if I choose not to order chicken at a restaurant or buy chicken at a grocery store, I'm not actually keeping a chicken from being eaten, I'm exerting extremely subtle market forces that have some small chance of leading fewer chickens to be raised for food in the future. This goes much more so for cows. I'm not sure how sensitive these markets are to differences in demand.
How do you feel about keeping dogs and cats as pets? I have a large dog who eats several pounds of chicken and beef per day. The net meat consumption of my household is out of control, totally because of my dog.
I feel pretty hopeless about changing my own behavior, given that anything I do will hardly make a dent in my household meat consumption.
Does anyone know what fraction of meat consumption in the US comes from household pets?
The cows replaced the buffalo, who numbered around 50-60 million. There are currently around 94 millions cows in the U.S. .So, you could probably subtract the methane that would have been produced by the buffalo we exterminated from what is currently being produced by the cows to judge the actual impact of beef and dairy operations on climate change.
I find referring to "adjusting for market failure" in this context to be rather odd. "Adjusting for market failure", to me, implies that the market price is a valid first-order approximation, and we're taking into account higher order effects. But climate change isn't priced into meat, and even less so is animal welfare. Calling it a "market failure" when you have a moral agenda that the market is ignoring is really stretching the term. Suppose you believed that using contraception is a moral evil, and you value each act of birth controlled sex at -$100. Clearly, that amount of money isn't priced into birth control pills. Is that a "market failure"? Or is that just the market ignoring your moral preferences?
Low carbon investor here. My assumed social cost of carbon is more like $100 than $10 (off topic, but when I look at the studies that come up with a low cost, their scope is always way too narrow). However, I still prefer beef to chicken, for the animal welfare reasons you set out. Even better, there is an incredible interest among younger cattle raisers in sustainable and regenerative farming practices, and restoring lost carbon to the soil. There are business models emerging whereby you can make good money by adopting such practices. So I expect beef to start moving pretty dramatically towards zero carbon, or even carbon negative, within this decade.
A humanely raised free range chicken is still pretty good sometimes.
I don't understand the underlying assumption that you're responding to, stated but not really proven, that the price of a good (or the cost of an action) has anything to do with its moral value. There's a lot of work being done by this one sentence: "on the grounds that the market has priced its moral cost at lower than chicken's moral cost". What does this actually mean? The market hasn't priced a moral cost at all, and I'm not sure why that would be a default presumption.
In the carbon offset case, the price of an offset is driven by
1. The cost of producing it, eg setting up a biogas digester or hiring guards to make sure a rainforest doesn't get cut down.
2. What people/companies are willing to pay.
3. An incredible thicket of UN and national policy.
As I think another commenter has pointed out, there's a completely separate concept of the social cost of carbon (https://en.wikipedia.org/wiki/Social_cost_of_carbon#Estimates) for which a good estimate might be $100/ton. Unfortunately because this is an externality, and much of it is borne by future generations, there's no mechanism by which it gets incorporated into the demand for offsets and therefore their price, unless many people are actually altruistic and knowledgeable.
For animal welfare, it's even simpler: this isn't the "market price" of anything, it's just the cost of an intervention.
You go on to conclude, I think, that we shouldn't use these prices to determine the ethics of our actions when we don't offset because of "weird market failure" but again I'm not sure where the presumption comes from that a functioning market would set prices equal to the moral costs.
---
A second point, and this may be nitpicking, but I'm not sure what you mean by "Nobody agrees on exactly how much it costs to offset a ton of carbon". There are many sellers of offsets and they all charge different prices! There are some legitimate ways to do it for $4/ton, and other legitimate ways to do it for $15/ton.
You do have to be a bit careful with those prices because the products are pretty different. On the one hand, some of them are perceived to be extremely low-quality (meet the certification but are unlikely to be doing much of anything). Some of them have "co-benefits": replacing coal plants improves air quality, protecting rainforest helps biodiversity. And also some of them are actually prices on the secondary market, where people are forecasting the "how much demand will there be fore these credits before they expire".
In any case I would agree that $10/ton is a fair placeholder value for the category.
I have the feeling this math only makes sense if you eat a wide variety of cuts of beef and chicken. If you only eat beef fillet, for example, you're responsible for killing a cow every single meal, granted that you are sharing that moral burden with everyone that eats the other cuts (and it will be a lot of people, sure).
The same can be said about eating chicken hearts, which are delicious, but I stopped eating them because just one small skewer with those are like 5-10 dead chickens worth for what can be considered a light snack.
The crucial assumption here is that carbon-offsetting actually works in practice. Not surprisingly, a quick search reveals corporations claiming it does, and environmentalist websites claiming it doesn't. This study commissioned by the EU (https://ec.europa.eu/clima/index_en), sides with the environmentalists:
TLDR is that they found that 75% of supposed "offsetting" activities were "unlikely to have resulted in additional emissions reductions" in the sense that they were not "additional" and would have gone ahead anyway, and only 2% had "high" likelihood creating additional emissions reduction.
Dishing out money to charity instead of reducing your carbon footprint seems a lot like buying indulgences from the church -- if someone says you can do bad stuff as long as you pay them afterwards, you should probably be skeptical.
What bothers me is that the only scale of measurement is dollars--as if my throwing green paper at something actually solves anything. I know that we can do things with green paper, I just don't know that it can be used to solve moral and/or ethical problems.
Yeah, but chickens are our historical enemies, descended from dinosaurs, and they're nasty filthy vicious profoundly stupid creatures, so I'm totally OK with their suffering in the service of my appetite.
Of course, in the event it happens that I eat a vegan diet, for my own health. But I would be totally OK with eating chickens if I weren't. Nasty little creatures.
I think this thought experiment is missing the largest point. Cows have best friends, you can tell the difference between a 'happy' cow and a 'sad' cow. Cows can show affection. To my knowledge, chicken's show none of this. So you could say 1 cow consciousness is worth 1,000 chicken consciousness. Or even more than that if you'd like. For instance, almost no one puts moral weight on killing a fly. There seems to be a cutoff point in which something isn't conscious. Where are chicken's and cows on that spectrum? It's certainly not as easy as comparing the number of neurons.
Both the issues of climate impact and animal welfare vary significantly depending how the animals are raised.
Free range farm animals arguably suffer much less than most wild animals. Plentiful food for life then a quick painless death. One could potentially argue that it's a moral obligation to eat these animals to ensure more are brought into existence.
Cows raised on fresh pasture that are 100% grass-fed (or close to it) also have a much lower impact on climate change than feedlot cattle. They can also be raised on land that isn't suitable for crops, negating the argument that it's inefficient use of agricultural land.
The catch of course is that it's probably not possible to produce current levels of beef and chicken under these humane and sustainable conditions.
You're doing more harm than good here. There are plenty of cold hard facts we can use in order to make these decisions. By injecting sophistry like "consciousness scales with the number of neurons,"you're simply muddying the waters and casting doubt on whatever conclusions we draw.
The market price of carbon reduction is so broken that if you buy a certificate to save 1 ton of CO2, a good approximation is that you safe 0 tons. I would argue strongly against basing any arguments on market prices for CO2 emissions.
How do you safe CO2 in this system? You ask the owner of a forest if he promises to not cut down the forest in the next 20 years. A "yes" is not legally binding, you hope that he feels morally obliged to stick to it. The carbon reduction price is the lowest amount of money that someone is willing to pay for a "Yes". If you have even one traitor then this price is ridiculously low. And note that buyers (think of an airline compensating their CO2) may have an incentive to encourage traitors.
I think this is exactly what happens. I don't even think there are mechanisms for preventing the owner to sell the same piece of forest over and over again, generating CO2 "reductions" out of the same forest many times. And then still change his mind a year later and cut down the forest. This are mechanisms against this in some countries, but as long as it works in some other countries, it will determine the market price.
Dedicated people try to have certificates of better practices. Those try to find out whether the owner is actually serious about it. So they ask "Are you REALLY SERIOUS about not cutting down the forest?", and if he still answers Yes, then he gets the money. Despite sounding cynical, I think this helps a bit.
There are also people (on the level of governments) trying to fix the system. Perhaps we will have a working system in 5 or 10 years. But currently, it is so completely broken that any money you pay now will likely have 0 returns. (There are reasons to use it nonetheless. You can speculate that using the system now makes companies get used to it, and that they can't easily withdraw when/if the system actually starts working.)
Is it true that farm animals are still given antibiotics as a matter of course to help them grow big? Can you put a dollar value on contributing to antibiotic resistance?
I think people are emotionally detached from farm animals vs pets and I think it would be useful to include common pets in these questionares as a sort of a control.
for example neurally and in terms of capacity for suffering and moral cost I'd expect pig or a cow to be as valuable as a dog, yet I suspect that most people that had no contact with farm animals disagree, as shown by the western outrage towards dog consumption elsewhere in the world.
What about donating to completely unrelated charities? E.g. you switch to eating beef only but instead of giving that $22 to carbon offsets, you give it to some charity, which you think does even more good. That way, you should come out net positive in the moral calculus.
I try to do a version of this (though I also try to eat less meat in general) and one of the main challenges in practice is knowing whether those $22 really are additional. If you were already giving to charity prior to making this commitment, then whatever sum you donate annually is sort of arbitrary. Say you donated $X last year and want to give a little more this year but haven't decided exactly how much yet. But you also want to offset your meat consumption this year, which is an additional $22. Whatever amount $Y you decide to donate this year, it's hard to tell whether you wouldn't have chosen the same $Y without the moral offsets. If you give to carbon offsets, that's a cleaner psychological separation. If you've made a Giving What We Can pledge or something similar, it's easier again.
It's kind of a stupid argument, because it's really just a psychological effect, but for the reasons Scott described above, it's important because offsetting really only works if you do the offsetting.
Still, I think giving to the absolute best charity (or a donor lottery or whatever) you can think of and using the carbon offsets as a lower bound of how much you should give is a good idea; at least if you're mostly utilitarian like me.
Red meat like beef is considered more carcinogenic than white meat like chicken, I believe. So I have a huge selfish motive to eat less beef than chicken.
One could argue that consciousness actually scales *exponentially* with neuron number, since there is some network effect going on there. If that is the case, eating 0.5 cow is much worse than 80 chickens, morally speaking.
> We calculate it out and find that eating chicken only causes the equivalent of $1 in damage (since Yog Sothoth eats only one extra galaxy, and we can save one galaxy for a dollar). The implied moral is "Go ahead and eat chicken, it only does $1 worth of damage and that's not much". But actually, it causes an entire galaxy full of intelligent beings to die.
Your error is in assuming that $1 is the market price for a galaxy. That may or may not be the price of saving the marginal galaxy, but the market is obviously not clearing at that price: more galaxies would gladly pay $1 not to be eaten than are doing so. This construct can only exist temporarily, such as if the charity were not yet well-known.
To extend the example to your chicken-suffering-versus-beef-offsets, the problem is worse: you're not valuing the direct value of the benefits. You're finding a value about how much other people would altruistically pay to stop a harm that (especially in the case of the chicken) does not directly affect them. To put it cinematically, "your money or your life" has a different weight than "your money or *his* life."
Now, assuming all of these offset programs are above-board, you as an individual can balance the moral scales (in some sense) by in fact making the offsets. In that case, you're comparing like against like, and you're at a Pareto optimum: neither the rate of chicken suffering nor the carbon content of the atmosphere would increase.
In using altrusim prices to make the argument, however, you are no longer operating in a Pareto framework. Something will be made worse for your actions. And those market prices do not reflect the real value of mitigation or the subjective value to the victims, but the altruistic value to third parties. Don't be a status-eater.
> Second, most likely anything you personally do to prevent global warming won't matter at all; either very large-scale actors like states and corporations will fail and there will be various disasters, or the large-scale actors will succeed and we will escape most problems.
This is not in fact true. As far as we can tell, the human cost of climate change is increasing on the margin. The incremental harm of going from 4 to 4.1 degrees of warming is much worse than the harm of going from 2 to 2.1 degrees of warming.
If you think that your emissions would not be replaced on the margin, then reducing them is *more* beneficial if you think that the world will not effectively coordinate. The bigger trouble is in evaluating whether emissions will be replaced, since if you are making this decision in a market framework (e.g., taking an unnecessary drive and filling up the gas tank as a result) then prices will change causing others to change their behaviours.
I'll note that carbon offsets do not yet have this problem, since for now the area appears to be demand-constrained with little offset-price-effect on the margin.
I don't at all understand why the relationship of consciousness to cortical neuron count is assumed to be linear. My prior would be that it is anything but. A network of ten neurons can assuredly not be conscious in any meaningful sense, no matter how you adjust the connection weights. Saying that ten billion such isolated systems have the same consciousness measure as an adult human seems obviously silly.
The Gödel Escher Bach approach of relating consciousness to self-modeling would also suggest this. Some threshold of neural complexity needs to be passed for such modeling to become a realistic possibility. It’s not going to be a discontinuous jump in the “consciousness function”, but it might be a sigmoid-like curve. The question then becomes: What is the slope of that sigmoid? Where on the neuron count scale is the mid point located?
I have no idea. But I don’t think you can just assume chickens and cows are both on the slope, rather than stuck at “basically 0” or “basically 1”. Let alone that they’re both on the slope, and it’s shallow enough that you can just fit a linear function to it.
With our current limited knowledge of consciousness, I don’t think there’s a good answer to this. But where I’d start is looking at actual animal intelligence experiments, rather than neuron count, and try to figure out a likelihood ratio between chickens being conscious and cows being conscious from that.
I’m far from an expert on animal intelligence, but I was under the impression that cows tend to perform far more impressively. A likelihood ratio of 1000+ in favour of the cow doesn’t seem totally ridiculous to me. Which would tilt the math to eating chicken being better.
Interestingly, while this discussion is about chickens and cattle, where you can make some pretty decent arguments about suffering either way, the case where it seems absolutely clear-cut would be swine vs. cattle. Not only do the pigs weigh less, they're _also_ more cognitively advanced and presumably more capable of suffering.
So if you want to go on a meet selection crusade, definitely start by getting rid of pork - that's the no-brainer under this line of argument.
One further (albeit a bit complicated) argument in favour of chicken over beef: it's the best choice in the possible states of the world that are most unfavourable to meat-eating, so the best response to uncertainty.
TL,DR; the beef over chicken policy is the best idea in worlds where the differences in moral value between different animals are smaller, which is also the world where the difference in moral value between humans and other animals is smaller, which is the world where animal lives matter the most.
Longer:
Dan Dennett, Eliezer Yudkowsky and Descartes could be right that moral value more or less 0 below the point where first person experience occurs and a constant value above it, and first person experience isn't very common in animals - so Chimps, Dolphins, babies past a certain age all matter about the same and no other animals do. In which case, eat whatever you want.
OTOH, Jeremy Bentham and Peter Singer could be right that first-person experience isn't the key thing and in fact it's capacity to feel and react to pain that matters, in which case there's a sliding scale of moral value that we can scale with neuron count or something similar, such that a Human doesn't matter millions of times more than a chicken, but maybe only tens to thousands of times more, so there are no sharp declines of moral value with brain complexity.
What's key is that between these two, is a VERY wide range of possibilities. We basically know that human lives matter most, and the question is how the curve looks 'behind' us - does it drop off almost straight away after Chimps or decline completely smoothly, reaching near 0 only at sea sponges and nematode worms?
So there a range of graphs over hugely different orders of magnitude, all the way from a completely flat increase of moral value where the most and least complex animal life aren't all that different (so my life only matters 10 times more than an insect) to a very hard 'exponential' increase where my life, or a chimps, matters a million times more than a dog. Not all of these are equally plausible, but the best argument for caring about animal lives is that considering how horrible factory farming is, most plausible settings of the numbers that aren't just 'animal lives don't matter at all' say it's a bad idea, and it's not that plausible that animal lives don't matter at all.
Here's the bit that's relevant to chickens vs cows - the more carnivore-friendly models of the world are ones where animal lives matter much less than humans - these are ALSO ones where there's a much steeper drop-off past humans because the moral value of animal lives depends on much more sensitively on brain complexity.
In a world where a cow's life matters 1000x more than a Chicken's life, so the climate maths doesn't work out in favour of eating beef over chicken, it's probably also true that a cow's life matters >1000x less than a human life, and so eating meat isn't much of a big deal in general.
Whereas if you think that a Cow's life matters only 2x as much as a chicken's, so the climate maths works out well for eating beef over chicken, that's probably a world where the decline in moral value with capacity to feel pain/brain complexity is much flatter IN GENERAL, so eating meat is morally much worse in general.
I shall explore this somewhat further when I am not also working, but I do wonder if there is a role for a component that honestly weighs heavy in the decisions I make about what to eat:
Is the animal a jerk?
Chickens? Jerks. They'll peck each other to death and stomp on the corpses.
Young male lambs? Jerks. My godfather takes a shovel in with him when dealing with his ram so he can hit it in the head. The ram ignores anything less, and is quite aggressive.
Steers? Often dicks, though not always.
Pigs? Not necessarily dicks. In fact, can be downright pleasant.
I think, all else being equal, it is definitely better to eat nasty animals that are mean to others than to eat ones that are nice and pleasant to all other creatures. Though on another level, should we let the nice ones have a brief, pleasant existence before we eat them?
The reasoning seems spurious to me. Chicken, cattle and pork make up a huge proportion of the animal biomass on the planet specifically because they are bred for food:
Chicken make up 3x the biomass of all wild birds combined, and cattle make up almost twice the biomass of humans and 14x more than all wild mammals combined.
You can't "save" chicken by offsetting, because there is no wild population of chicken to save, and if there were, it would not be in anywhere near the volume needed for effective offsetting. So unless we start devoting even more space for chicken or cattle as ornamental pets (presumably at a cost orders of magnitude higher than battery-farmed chicken or cattle), the supply of offset credits would simply limit to a tiny elite of virtue-signaling rich people.
The above is a training video from the Australian cattle industry about how to humanely process cattle for slaughter. It is worth a watch. Obviously, includes a stungun firing a bolt into a cow's head, and a cow getting it's throat cut.
Having watched it, I am fairly confident that the cow likely doesn't know what's going on and is dead quickly and humanely. I wouldn't want to work in a slaughterhouse, but if you made me watch the slaughter of every animal I eat I probably could.
I would suggest it for people. I definitely learned something about myself watching it.
This problem reminds me a lot of Alasdair MacIntyre's critque of 19th Century Utilitarianism in _After Virtue_ Ch. 6 "Some Consequences of the Failure of the Enlightenment Project". Somewhat more than most things, it's probably a bad idea to jump into the middle of that without knowing the context within the larger argument and work itself (Utilitarianism isn't even the focus of the book). But, to throw in a piece anyway: "For different pleasures and different happinesses are to a large degree incommensurable: there are no scales of quality or quantity on which to weigh them. Consequently appeal to the criteria of pleasure will not tell me whether to drink or swim and appeal to those of happiness cannot decide for me between the life of a monk and that of a soldier. / To have understood the polymorphous character of pleasure and happiness is of course to have rendered those concepts useless for utilitarian purposes... it follows that the notion of the greatest happiness of the greatest number is a notion without any clear content at all. It is indeed a pseudo-concept available for a variety of ideological uses, but no more than that.(MacIntyre - After Virtue, Third Edition, Ch. 6, p.64)
One I always wonder about with this kind of domestic animal problem though: what's the proposed alternative for these particular animals that are not getting eaten? In the short term, I suppose a bunch could just be set free into the wild. Maybe at least some of them would survive for a while in a sufficiently abundant area, maybe there would even be some feral herds that linger. I'm doubtful there would be the same number of said animals in existence in the long run though, and if that is the case, the question is was avoiding suffering so paramount that existence existence or at least life should be forgone? Perhaps we could consider that all the matter that made the chicken or the cow would still exist, and from one view that's equivalent, but then it's hard to figure out why that set of atoms existing as dirt or suffering as a plant rather than as a chicken is particularly better. That's not to say we shouldn't care about suffering, but I think there's a lot more questions to be asked and answered on this kind of problem, and perhaps the framework needs to be scrutinized more.
It looks like nobody has brought up the huge issue with these offset numbers, which is that "saving X chickens" actually means "convincing people to eat X less chickens". This of course has significant costs to those people.
Offsetters are simply shifting the burden of vegetarianism, plus adding the costs of charity operation (including targets watching depressing/disturbing videos).
Offsets only work if you actually consider ALL effects of the charity, not just the metrics they advertise.
What surprises me the most is how cheap it actually is to offset for Co2 emissions... I really like internal combustion engine cars, and I dream about one day owning a really cool ICE car, and if this is the price I have to pay to offset the effects of Co2 emissions I don't even see how it ever became an issue.
What I like most about his post is that it has always bothered me that we have religions that won't eat pork or beef but none, so far as I know, that won't specifically avoid chicken. Could this be the birth of one?
The chicken vrs beef question is an interesting one, but in view a less important one than the standard vrs organic/free range one for the reason that I would think that more people will be inclined to switch to organic rather than choose exclusively chicken or beef
Central to the standard vrs free run benefit / cost analysis are two key considerations: 1) Do animals have "standing" in calculations over welfare and 2) The challenges in quantifying suffering and utility generally
Assuming you agree that animals have standing, then the next question is do some animals have more standing than others and if so, how do you determine these weights (as Scott identifies).
Cost and purchasing power also come into it, as arguably there is more moral imperative for someone who is wealthy to switch to free run, than someone on low income.
If you care about the suffering of, say, cows, why? Are they your mother?
Some people don’t care at all about the suffering of others. These people are generally called psychopaths. If you are solipsist, it’s logical to be a psychopath. If others don’t exist except in your imagination, you should use them as you wilt.
If you are not a psychopath and care about the suffering of other humans, or even other animals, or even vegetables, why? It’s probably because you view them as other beings, like yourself. You empathize, or at least sympathize with these other beings.
Why do you care about the suffering of other beings? It’s because you consider yourself a being, right? You care about them because they are in the fraternity of beings, a club which you belong to.
What makes you both beings? You experience shit. Who experiences shit? You do. And they do.
Now, are you separate entities, you and the chickens? It’s all semantics from here down, but my macro view is beings are these things made of matter and the entity must exist within the matter. Seems more likely matter itself has qualia than that it’s an emergent quality, because it's a simpler explanation.
Scaling moral worth by (cerebral cortex) neuron count seems like a poor measure to me. By that line of reasoning we'll have to conclude whales are worth as much as a human, using Wiki's data. Orcas are worth 3 humans.
That's without getting into the fact that a lot of things are determined by neuronal connections rather than neuron count. Should we determine moral worth using the number of synapses in the cerebral cortex instead?
Scott, I think a more interesting thread might include a definition of suffering and some down-the-rabbit-hole speculation on why avoiding is preferable. It seems to me that vegans and other holier-than-thou types are striving for a kind of purity that cannot exist in the context of natural life. I mean, let's face it, if you want to minimize the suffering of animals you should develop the skill of marksmanship and become a big game hunter who only takes head shots. Wild animals don't have retirement communities. They either die of starvation (or freeze to death, or bleed out) or they get eaten (often eaten alive) by predators.
I believe that natural life is, at least in large part, suffering. Anyone who denies suffering is actually denying life in the natural world. Sorry vegans, but that combine harvesting the mega monocrop is chopping up everything from terrified fawns to snakes and birds.
People should embrace suffering. We're all gonna die anyway.
From these considerations, really by far the most important thing <a href="https://www.unz.com/akarlin/animals/">is to just avoid pork</a>. Or get probably a OOM-tier reduction in cognition-adjusted animal suffering by going pescetarian.
PS. Personally I mostly consume chicken and turkey as animal protein.
Shouldn't these calculations take into account wild animal deaths? There are diffuse causes that are hard to pin down, like global warming, pollution, and land converted for growing feed, but cattle ranchers kill a lot of things directly (wolves, bears, mountain lions, snakes), sometimes in really unpleasant ways (prairie dogs). Ugh and all the deer etc that get tangled in barbed wire. I guess it's grimly nice that there is less wild animal suffering per beef produced as rangeland is defaunated.
I appreciate the reasoning, if you're reluctant to eat creatures, and some of them are one meal, and others are thousands of meals... the math is pretty stark.
Maybe pilot whales are the result of a timeless evolutionary process, where they adopt decoy cneurons to defend against time-traveling rationalists who prefer whale meat over insect-based protein.
Cheap jokes aside, I do worry about moral weights scaling cleanly with cneurons. And this is something I've worried about for a long time.
In fairness, I'm also very worried I'm getting this wrong, given the amount of cycles you've spent on this too. I absolutely clicked through to read all the times you've previously discussed it, and find the surveys fascinating and challenging.
So I want to say pretty cautiously, I think I have an idea why the surveys come out that way, while not being particularly normative. But I'm still committed to thinking about this in more detail.
I'll list a few problem areas, then a possible explanation for why those emerge, and a few alternatives (none of which are particularly satisfying, but might prompt someone smarter than me to move the ball further forward.)
My initial concern is that the cneuron:moral weight analysis produces weird results at the margins. Does a whale's worth of mosquitos have any moral weight at all? Is there a Jupiter sized swarm of (galactic space faring) mosquitos we should sacrifice humanity for?
For digital intelligence, is suspending a process on my CPU similar in kind, but just not degree, to ending an AI with agency, identity, goals, a way to contextualize its own history?
Then at the other extreme, how much more moral worth does a post-singularity paperclip maximizer hold over you and me?
The extremes are relevant, because when we're talking about animals at 1-2% of our cneurons, then we are already at some intuition breaking extremes. People report feeling moral responsibility towards robots that so much as mimic human facial expressions, with the processing power of a toy. People generally feel more moral responsibility towards "cute" animals relative to "ugly" animals. They aren't great at surveys of moral weight at extremes.
I also worry about this for close comparisons. Do smarter humans hold greater moral weight than others? I can imagine lifeboat ethics situations where intelligence is the most important skill to ensure the survival of the greatest number. But as a universal principle? I can imagine somewhat more fanciful scenarios where the best person at growing soybeans, juggling chainsaws, or mixing strong drinks that appease angry aliens are the most critical to everyone's survival.
Are we sure we want to put down that dogs are *roughly twice* as morally valuable as cats, even though there's enough ambiguity in that *factor of two* difference that owners can still endlessly debate relative intelligence?
So close comparisons and extreme marginal cases both pose some problems. Yet a clear scale seems to emerge in surveys, and track our rough intuitions, with more confidence on one end.
What could cause that?
Maybe the scale is right, but there are just hard cases. A lot of hard cases clustered in those domains.
Another natural way to get those results, and those same problems, is to have a sorites paradox about a binary condition. Any bright line feels arbitrary, so you get a smooth-ish distribution in polls, with more certainty at the ends. But you're still measuring essentially a binary (or at least a very steep) condition.
To break a sorites paradox you need to think carefully about what constitutes that threshold. (Or bite the bullet and go with an arbitrary line). You'll still get probabilistic arguments near that threshold, and that's fine, but it will allow you to consider a much narrower range than "anything with any processing power whatsoever."
If reducing the firing of nociceptors is your main overriding concern, if pain signals buttress your entire morality, then ok, nevermind, this doesn't need any touch ups.
But maybe harm reduction is "mostly right," in the way that Newton was "mostly right." And now we're at the end of the paradigm runway, where new models that provide additional resolution are necessary (while sure, preserving most of the previous results).
Imagine a weird lab that ran current through isolated nociceptors. Great tragedy or no? If not, why not?
I'd propose that neurons in petri dishes have no moral weight because of larger factors, like, (a) they can't contextualize pain, or, less likely (b) they aren't part of a moral community that can engage in exchanges of moral behavior, or, Sam Harris's (c) they don't contribute to the flourishing of humanity, or, the more Stross-ian (d) they are unlikely to contribute meaningfully to maxing out the instructions per second per gram in our galaxy. A proxy for (d) might be likelihood the species' descendants will experiment with genomic editing or interplanetary travel.
None of those are problem-free, I'll admit. But any of them provides a framework you can extend to build a cutoff that falls short of humans and isn't completely speciesest, providing opportunities for promising high intelligence species like corvids, but without going all the way down the chain. This is useful if you do not believe that mere sentience, plus sufficient mass, becomes more important than sapience.
But maybe I'm wrong, and should alter my consumption based on the p() of confidence in this counterargument, which isn't stupendously high...
My personal intuition is that there is a cut off-- like chicken suffering doesnt matter 1/6 as much as cow suffering it just doesnt matter, because for suffering to matter you have to reach the complexity of a dog, I think in retrospect i mostly just wanted to eat chicken though.
I won't say the 'suffering of chickens vs cows' question is worth *nothing* but if I'm being honest I think the ecological impact question trumps the suffering question by a ratio so large we can safely round it up to infinity.
Trying to gauge the moral gravity of fewer numbers of relatively larger beasts vs greater numbers of simpler beasts (and let us note in passing that none of them are ever going to feel grief, or write histories; and their slaughter-as-opposed-to-living-out-their-lives-naturally will not leave any lasting impact on the world whatsoever) seems like exactly the kind of question a philosopher might pose as a reductio of utilitarianism because *obviously* it has no answer.
There’s also another benefit though not all of us will see it that way. Eating a lot of red meat could result in earlier mortality, reducing social security and Medicare costs to the government.
I just love your style m8. It doesn't matter wrong or no. That's besides the point. The themes you write on a hard per se to get right. But your style of writing is nice and good and funny in a non-funny way. Glad I have your blog to make my day!
I eat chicken and not beef because I don't believe in eating something I could not personally kill. Also, I try to only buy free-range chickens, which I think minimizes the suffering. It is too bad that PETA turned evil.
Interesting analysis, though.
Another great argument for general vegetarianism when possible -- avoiding shaky math.
Methane lasts in the atmosphere for 12 years. So a cow today is merely "replacing" the methane emitted by a cow in 2009. Which was replacing the methane from a cow in 1997. Replacing 1985. 1973. Etc.
Therefore eating beef today is not adding to greenhouse gas impacts on global warming, as long as the total number of cows is fairly level. It has been decreasing in my country for the last few decades so I feel no moral imperative whatsoever.
I would welcome arguments against this position. It seems to me like we expect everything to be net zero except cows, which must lead to reduced greenhouse gases.
Don't forget that absent human consumption of chicken (and eggs) and beef, cows and chickens go extinct, at least their domsticated versions.
This will sound like a contrarian troll comment but I mean it quite sincerely
Why does the calculation for the net moral cost of beef vs chicken not factor in the utility gain/loss from each of those in your diet?
I can't speak for other people but I can say personally that I would be experiencing quite a bit of suffering personally if I could never eat beef again, but I would not be experiencing that level of suffering if I could never eat chicken again.
I am in general resistant to utilitarian reasoning like this, for reasons I have expounded on at length in your various comments sections throughout the years. By framing everything in terms of this, people (or, at least, rationalists) are put in a position where they feel that their own preferences are invalid unless they can be justified on utilitarian grounds. It's not that hard of a jump from there to "my preferences don't matter" or even "my preferences don't exist"
But they do exist! And they do matter! It is fundamentally meaningful to factor in the cost _in human suffering_ of dietary restrictions when you're running these numbers.
(Further, I would remind you all that each and every one of us in this space are likely 5-sigma deviants from normalcy on at least a few critical dimensions. I imagine most of your instinctive reactions to my position is some variant of "come on, don't be ridiculous, nobody cares that much about their diet". I think most people do, and it's us here who are unique in that we don't care as much)
This seems like a great use case for having a functioning government. If there are really these kinds of moral and ecological costs to individual consumption choices, rather than push off the calculations and offsets of exactly what they are to individual consumers who are purchasing thousands of distinct items a year, reflect those costs directly in the price of goods via taxation that internalizes externalities like these.
Precision isn't accuracy. The numbers cited for ghg are laughably inaccurate, though stated with precision. This is why economists are always so wildly mistaken. They take numbers someone pulled out of their butt and run them through a formula that is utterly unrepresentative of the system.
What do you feel about the mouse plague argument against vegetarianism (particularly relevant since there's a mouse plague currently ongoing in Australia)? https://www.abc.net.au/radionational/programs/scienceshow/mice-the-biggest-losers-w-vegetarianism/4660498
This is an argument against Australian wheat specifically, but more generally there are clearly harms to animals from row-crop agriculture....as a vegetarian I don't quite know how to feel about this. (if you are an omnivore using this as a whataboutist argument, I hope you are already avoiding farmed meat other than grass-fed beef!)
I think all of this makes a lot more sense if you add in "and of course, you're already spending some amount of money to do good in the world, however large or small that amount may be".
Assuming you're being efficient with that money, it kind of obviates the Yog-Sothoth problem. You have already chosen to donate/spend/forgo a certain amount of money, and that implies a certain price at which you're willing to purchase goodness for the world.
Decisions like whether to eat chicken will be priced appropriately; since you're already donating to the Stop Yog-Sothoth Fund, you are indifferent about whether to donate a marginal dollar, so if eating chicken is really worth a dollar to you, it must also be worth destroying a galaxy. Because that's your BATNA! That's what you would've done with that dollar otherwise (approximately).
This is another problem I have with utilitarianism. I described it in a piece I wrote earlier. (https://parrhesia.substack.com/p/utilitarian-action-coupling)
You get counter-intuitive conclusions when you cluster supererogatory and obligatory actions together. If eating animals is morally similar to murder, then this sort of utilitarian offsetting is morally puzzling. Can I commit a murder and then donate to prevent murders? Commit an assault and donate enough to offset it? What if I donate enough to save 10 lives through effective charities, can I then murder someone and be considered a morally good person? It seems like morality does not work that way.
I don't understand why I should take the concept of superrationality seriously. To me, it seems based on an obvious category error. As per the wikipedia article linked in the post:
"The idea of superrationality is that two logical thinkers analyzing the same problem will think of the same correct answer."
The category error is in the statement that a game like the prisoner's dilemma has a correct answer and has wrong answers.
Perhaps I am leaning too much on the "correct/incorrect" wording used on the Wikipedia article, though.
Chicken suffering has no moral weight, on the grounds that chickens are assholes and deserve it.
Unless I'm missing something, wouldn't you want to consider the price difference between beef and chicken? I'm sure it depends where you live, but generally I expect to pay something around 20% more for beef than chicken at a restaurant, and more than that at the grocery store. Without calculating, I expect the savings from purchasing chicken will be significant and likely enough to offset the more expensive moral cost of chicken (assuming you use the savings for that purpose).
I think the assumption that causing a chicken/cow to be killed is worse than causing that chicken/cow to never come into existence in the first place needs to be examined. If the opposite were found to be true, we might instead conclude that we need to eat as many chickens as possible in order to maximize the number of chicken lives lived.
If we are doing a utilitarian calculus and incorporating animal suffering and death, then it is not clear to me that hastening the demise of human life is not a net increase in utility. Currently, around 3 billion animals are killed daily and billions more live in inhumane conditions. So in 3 days, there have been more animals killed than humans alive. If climate change accelerated the death of all humans, then couldn't this plausibly be a utility increasing good thing if it ended factory farming?
I do not believe in utilitarianism and I am not vegan but I am sympathetic to these arguments. However, when you put them together, it seems like it is better to kill all human beings alive. Can someone explain to me the error in my reasoning if it exists?
Assuming that the choice is "factory-farmed chicken" or "some kind of beef, I dunno, normal beef I guess" weakens this argument, although it also helps keep it simple. The 30-days-of-horror life lived by a factory chicken is indeed so horrible that I (an omnivore) will go out of my way to avoid contributing to it. The life lived by the fancy free-range bug-foraging high-welfare heritage chickens sold by my local butcher is… I mean, I don't know all the details, but as far as chicken needs go I think it's pretty ok up until the final butchering scene.
So with that calculus in mind, you can reduce carbon emissions AND reduce suffering AND get to eat a much nicer grade of chicken. (You can do 2 of the 3 by buying fancy high-welfare free-range grass-fed beef, too, and sometimes I do that.)
Does America really have like 99% factory chicken or has free-range made some inroads since I left there 10+ years ago?
Why are we converting kg into tons when tonnes are right there, ready to help us?!
At the object level, the obvious win is to eat sustainable and ethical meat. White Oak Pastures beats Beyond Burgers in terms of sustainability (https://civileats.com/2019/06/19/impossible-foods-and-regenerative-grazers-face-off-in-a-carbon-farming-dust-up/) - beef sinking 3.5kg of carbon per kg of beef produced, vs Beyond, coming in at 2kg of carbon emitted per kg of food produced.
Pasture raised meat is even cheaper than Beyond Burgers. An ethically raised meat animal gets a happy life and a better death than most wild animals. There's just no reason to continue worrying about meat alternatives. We just need to stop doing the horrifyingly awful thing and get back to doing the actually great thing.
"A factory farmed chicken lives about thirty days, usually in extreme suffering."
Unless I missed it, the above seemed to be as close as you get... but why not touch on the *difference* in conditions than factory-farmed cows vs. chickens are raised in? My understanding - which I think is reflected in Brian Tomasik's table here: https://reducing-suffering.org/how-much-direct-suffering-is-caused-by-various-animal-foods/ - is that chickens suffer much more in captivity than cows.
I'd be interested in a "Carbon offests: much more than you wanted to know" type post that investigates the situation in consumer-level carbon offsetting. I've bought some in the past and couldn't quite shake the feeling that it was, uh, somewhat imaginary.
Can someone explain how the offsetting transaction works? Party A gives $ to Party B to offset X. In exchange for $, what does Party B do that offsets suffering or carbon emissions or whatever?
The mechanics seem super hand savvy to me. It looks like the old Catholic indulgence where parishioners paid for their relatives to spend less time in purgatory. Both indulgences and offsets sound unmeasurable and unenforceable.
"Meanwhile, if you don't eat some chickens, those particular chickens don't get eaten."
*Surely* this isn't case? Those chickens will still get sold to other people and then eaten, possibly at reduced price if your refusal to buy them affects demand. The real calculus will have to be how the reduced demand from your refusal to chicken affect prices and supply in the medium to long term.
Personally, I care about my own health more than I care about chicken vs cow suffering. Can we express these numbers as "chickens saved per expected day of decrease in the eater's lifespan"?
I will take this opportunity to urge everybody to eat only happy animals whenever possible. Murder is really much more palatable than suffering.
Chicken and Cows are largely carbon neutral. All of their carbon emissions are part of the normal cycle. This is similar to why burning new-growth wood is carbon neutral.
If only the people behind the JBS cyberattack had read this post in time.
I think that offset prices are still meaningful in the presence of "market failures" as long as the offsets themselves actually work. In other words, as long as it is, in fact, possible to stop Yog Sothoth from consuming a galaxy for $1, I think it's correct to say that eating a chicken (and thus arousing the hunger of Yog) has a $1 moral cost.
If it's actually possible to abate the hunger of Yog with a $1 donation, every $1 purchase made is a missed opportunity to spare a galaxy of intelligent life from unimaginable suffering. Therefore the only reason that Yog should be consuming any galaxies is because every person with dollars has decided not to forgo $1 worth of consumption in order to save a galaxy, _including_ the person eating the chicken. Spending $3 on a coffee in this world is exactly 3X as bad, in terms of destroyed galaxies, as consuming one free chicken.
And likewise, if it's actually possible to abate a ton of carbon with 33 cents, then it's genuinely possible to trade 3 tons of carbon for one galaxy. Every person in this world who gives any amount of money to the Stop Yog fund and the Clean Energy Initiative (or whatever) must be willing to trade one galaxy for 3 tons of carbon. In a real sense, the moral judgment of this society is that a galaxy is as bad as 3 tons of carbon.
But how about market failures? Any traditional sense of market failure should refer to:
1) Externalities
2)
3) Information Asymmetry
The obvious one here is externalities: the costs of Yogging and polluting accrue mostly to people other than the Yogger/polluter. But in both cases, what we're concerned about is pure externality, so it's hard to think about externalities as causing more of a market failure in one context relative to another. Any internal costs and benefits of Carbon/Yog aren't really part of the moral calculus to begin with.
Market power is essentially irrelevant here. It's not particularly relevant if the stop yog fund is charging above-market prices for Yog abatement: I can still get them to Yogproof a galaxy for $1.
And information asymmetry boils down to saying that I might not really know if my donation is stopping Yog. Maybe my donation saves a galaxy, maybe it goes toward a cardboard sign that says "say no to yog" on a galaxy that was never appetizing in the first place. If I can't tell the true Yogsbanes from the grifters, maybe I keep my dollar in my pocket. But in this case, it's just not true that it costs $1 to stop Yog from eating a galaxy, because it's not true that I could give up a dollar to save a galaxy.
This sort of Singeresque utilitarian chain of logic has led me to the conclusion that very few people actually believe in catastrophic anthropogenic climate change; my thought is it’s actually a kind of character signaling, like wearing a mask after being fully vaccinated: “I recognize that doing this benefits no one, but I want to send the signal that we should take stronger collective action on similar problems.”
I struggle with these kinds of estimates of carbon cost. A cost of $10/ton where the average American releases 17.5 tons per year implies that all of America's contribution to global warming can be erased for less than half a dollar per person per day. I think if that was true it would have happened already.
Bit off topic but maybe still worth thinking about:
I think the offsets will be larger. The price of, say, 10$/ton of CO2 is low at the moment because there are currently quite a few low hanging fruits for CO2 offsetting. As soon as this will be more popular, prices will go up, not/not only because of capitalism, but also because it will be more and more difficult to find cheap ways for compensating a ton of CO2. Otherwise, I mean, let each citizen pay 17.5 * 10 = 175$ climate tax each year and the problem would be solved.
This is pure speculation, and I would much appreciate a more informed reader's comments on it.
I am endlessly perplexed (and a little vexed) that otherwise-sensible-seeming people assign moral value to the suffering of animals who cannot possibly reciprocate. What am I missing?
Or you can source your beef from a farm that feeds the cows kelp to reduce their methane outputs and feel confident that you've made a great decision. https://www.ucdavis.edu/news/feeding-cattle-seaweed-reduces-their-greenhouse-gas-emissions-82-percent#:~:text=A%20bit%20of%20seaweed%20in,the%20University%20of%20California%2C%20Davis.&text=%E2%80%9CThis%20could%20help%20farmers%20sustainably,the%20world%2C%E2%80%9D%20Roque%20added.
I would think health differences should factor in to this decision.
But also, it's hard for me to wrap my head around the mindset that one could take animal suffering or CO2 emissions from meat eating seriously enough to do this math and alter their diet or pay for offsets but not just...stop eating or rarely eat meat.
JBS, a major beef slaughterhouse operator recently had to cease operations due to a cyber attack (https://www.bloomberg.com/news/articles/2021-05-31/meat-is-latest-cyber-victim-as-hackers-hit-top-supplier-jbs).
While not yet confirmed to be ransomware, let's assume it is. In the moral calculus outlined in Scott's post surely the cyberattack is a net good since the reduced cow suffering and cow related carbon emissions caused by the attack outweighs the CO2 involved in the bitcoin transaction that will be used to pay the ransom, no?
Does anyone actually care about any of this with respect to their dietary decisions. For most people these are visceral not mathematical decisions. For example, after having visited (my father’s) chicken hatcheries, production and beef packing plants (for all, read “slaughtering” houses) when I was about 9, I haven’t eaten any bird or mammal. It’s not a function of any rationalist judgment. At some point it was clear that I couldn’t consider eating certain things. For about 20 years I was a vegetarian. Then, I tried to distinguish the source of my eating habits. I realized that I couldn’t abstract from the killing process. But, there was some killing I could tolerate and would do. Namely fishing. Although I didn’t then eat fish, I had fished and gutted fish. That truth led to my starting to include fish in my diet. Not because it was more sustainable (it may not be) but because I would kill it. Therefore, I would eat it. Can I justify it by neuron count? No. Do I care? No.
When I see a hamburger I see a cow — one I would pet. When I see fried chicken I see my neighbor’s chickens whom I know by name and have held. These aren’t calculations. They’re either psychologically instinctual or ontologically driven.
To wit, most people who try to be vegan or vegetarian for environmental (CO2) reasons end up struggling and at some point either failing or internalizing a moral and visceral perspective. Again, rationalism largely fails here. It’s similar to the way that people stop exercising their new year resolution to work out by February. Either they internalize it into some ontological part of themselves, or they struggle and fail.
Given all that, the reality for most people is that mammals feel to us more akin us than do birds. We relate to them. We kiss our dogs, hug our horses, feel the run our hands over the velvety soft nose of cows. Birds seem more foreign, easier to prey upon. Again, not necessarily justifiable (as you demonstrate) but your argument is taking place in the wrong universe. That makes it irrelevant to actual dietary decisions.
Tongue only loosely in cheek, this ends up seeming like an excellent argument to donate to charities which provide contraception. Assuming that <$175/year provides funding to prevent at least one new human (which seems like a reasonable guess) you can "offset" your entire life's consumption of not just meat but all carbon-emitting goods and services for the same or less than commercial carbon offsets. Plus improve the quality of life for women in need of health care!
If you are doing comparisons of the number of suffering chickens vs. number of suffering cows, should you also multiply by their respective lifespans? (i.e. how long they suffer for) Why or why not?
Somewhat nitpicky -- but the CO2Equivalent numbers you're using for methane are on a 100-year timescale. When we consider the CO2E of methane on a 20 year time scale, it's 3x as high. (put plainly: methane emissions from cattle burps are responsible for 20% of all warming in the next two decades).
The linked guesstimate (https://www.getguesstimate.com/models/10897) in the EA post you cited for the chicken estimate (https://forum.effectivealtruism.org/posts/9ShnvD6Zprhj77zD8/animal-equality-showed-that-advocating-for-diet-change-works, "But what if it were chicken?") seems to say that it actually costs $0.82 to spare a chicken, not $6; it gives $6.64 for the cost to avert a chicken year. But the EA post itself says that it costs "$5.70 per chicken spared (90% interval: $0.71 to $32) and $50 per chicken year (90% interval: 6.3 to 280)". I'm confused. Maybe either the EA post or the guesstimate were updated without the other also being updated?
chicken are basically vegetables, so if I'm on a day I'll feel guilty I eat them instead of a mammal
Possibly too tangential, but: Did anyone ever figure out why Yudkowsky is so confident that most animals, including chickens and cows, don't have qualia? (Referenced but not spelled out here, as well as some other places: https://rationalconspiracy.com/2015/12/16/a-debate-on-animal-consciousness/.) I'm very curious what his (presumably partial) solution to the hard problem of consciousness is.
What about the price of the meat itself? Chicken is significantly cheaper per calory than beef, leaving me with a budget surplus that I could then use to pay for the chicken offset. Or effects on your person health for that matter, my impression is that chicken is healthier than beef, and being healthier would allow me to work more overtime, earning more money once again.
I think the model here is far too simplistic, which is my issue with effective altruism in general - reality is complex and I could easy choose a different set of effects to focus on to argue for the opposite conclusion.
Does marginal spending on carbon offsets actual cause more carbon to be offset? I worry that we have, say, 1 billion tons of carbon offsetting capacity right now and if more people want to carbon offset, the price just goes up without the capacity increasing. Now increasing the demand for carbon offsets would increase the incentive for people to build out carbon offsets and that's plausible, but does it happen at a 1:1 rate? I'm sure economists have tried to quantify stuff like this and I'm just ignorant of the terminology.
I think this mostly shows how the "offset" concept is flawed. It's not what consequentialism is trying to get at - the offset donation is not causally connected to eating the meat - and it's used as a 'noble lie' to get people to donate at least something to charity despite being reluctant to give an altruistic amount, but like all noble lies it's incompatible with clarity of thought. Trying to staple it together with economics and its (also fictional) perfectly rational actors is a case of making two simplifying assumptions that contradict each other. Ultimately, the answer has to be to do the math on "direct animal suffering" vs. "risk of apocalypse / human suffering from global warming", and only consider money if actually making a charitable donation. The former is probably bigger, and certainly easier to quantify, which feels to me like it should carry more weight somehow.
For my part, I follow a principle of not eating mammals, which I adopted a long time ago because it's a relatively bright line while leaving me able to eat *something* in most restaurants / social settings. For myself I buy fish. In principle beef causes less suffering than chicken, but I've kept to the deontological rule in this case, though perhaps I should change that.
(On neuron counts - it's not obvious to me whether the function should be steeper than linear. Killing a bunch of individual neurons doesn't seem as bad as killing a single animal with all of them connected... but, OTOH, linearity is stable with regards to fusing brains together, so maybe it's necessary to have it as the first approximation.)
To me the moral calculus seems even more lopsided than that. I don't really care about the deaths of animal who don't make plans for the future, I care about their suffering and quality of life. On that basis chickens have utterly terrible lives and while cows experience some suffering in their lives, particularly at the end when they get shoved onto feedlots, but it seems that on net they probably have lives worth living? I think I'd rather be reincarnated as a cow than face non-existence.
> "Meanwhile, if you don't eat some chickens, those particular chickens don't get eaten."
Well, it will depend on the price-elasticities of the chicken supply curve and the chicken demand curve, right?
E.g. on the supply side: Perfectly elastic supply means refusing to eat 1 chicken causes 1 fewer chicken to be eaten (and the price will stay the same); perfectly inelastic supply means refusing to eat 1 chicken causes 0 fewer chicken to be eaten (and the price will go down). (And then vice-versa for demand.)
In reality I assume the elasticities are somewhere in between, so on average every chicken that someone refuses to eat saves somewhere from 0 to 1 chickens on average.
Anyone have any empirical data on the elasticities of the chicken supply and demand curves? Or for other animals, for that matter — I guess all else equal, you can do more good the more elastic the supply and the more inelastic the demand for that animal.
> (it's still totally fair to eat chicken and donate $1 to the Stop Yog Sothoth Fund, it's just catastrophic if you abstract away the step where you donate the dollar)
that doesn't really seem like a perfect utilitarian calculation. of course one option is to eat a salad and not donate, and compared to that eating the chicken and donating seems fair. but another equally plausible option is to eat a salad and also to donate, and compared to that the expected utility of eating chicken and donating is terrible (one whole world lost).
> Either carbon offsets, animal suffering offsets, or both could be market failures like this. In fact, if you believe that we're currently not doing enough to fight climate change or animal suffering, I think you almost have to believe there are market failures in offsets.
how about believing that there are no market failures, but that there are low-hanging fruit, and the current offsetting prices are determined by changes on the margin, whereas average offsetting prices – if we tried to reduce animal product consumption completely to zero, say – would be much higher.
This is missing two important points. 1. Chickens have much shorter lives than cows and 2. The typical farmed chicken's life seems more miserable than the typical cow's.
The dilemma between decreasing CO2 but saving the lives of chickens vs cows is very much like Chappelle's genius bit about comparing Bill Cosby to a superhero that saves lives but rapes. "That's the dilemma for the audience. Because he rapes, but he saves a lot of lives. And he saves way more than he rapes, and he only rapes to save. But he does rape.”
"Meanwhile, if you don't eat some chickens, those particular chickens don't get eaten." Not sure I agree with the argument here - the marginal chicken you decide not to eat is already dead and will now land in the bin. Your individual actions may have no impact but that doesn't absolve you from your moral responsibility.
So this is vaguely related, and hopefully someone here can help me.
I've been trying to find information on the "ideal" diet from a climate perspective. This article is combining climate and ethical considerations (which I think is totally worthwhile), but if, for a moment, one completely ignored the ethical considerations of eating meat and _only_ cared about the climate impacts of one's diet, what would that diet look like?
From general reading/information floating around, I _feel_ like the "climate optimal" diet would a) involve a lot less meat consumption than your average American diet b) would probably involve very _very_ little beef but c) would not be entirely vegan/vegetarian.
However, I can't find any good sources that talk about this, explicitely.
For context, I ask because, given where I live and my living situation, I have the ability to get all or nearly all of my meat in a way that is, to my personal satisfaction, ethical. But, from a climate only perspective, I'm not sure how I should be allocating my diet.
The simplest answer I suppose is "don't worry about it, just buy offsets", which _does_ sort of solve my individual level problem, but I'm sort of interested in the long term societal options for how to feed everyone with the most climate-netural calories possible.
As a final note, I have been saying "climate impacts" this whole time, but that was mostly as shorthand. I am also interested in more general environmental impacts like water use, fertilizer runoff, etc. I know that that makes it more complicated so I'd accept sources that talk _only_ about climate impacts (CO2 equivalents etc.), but anything that _does_ incorporate more general environmental impact information would be great.
Yes, but to further complicate things - assume my charity budget is fixed. I donate x% of my income. Right now it goes all to givewell, but I could divert some of that to a carbon or chicken offset program. Now we have to include the opportunity cost of the human suffering by african children not having malaria nets, for example.
quick correction to your math in the second paragraph:
>switching from an all-chicken diet to an all-beef diet saves 60 chicken-equivalents per year.
by your own estimates, it should be (80 chickens - 0.5cows*20chickens/cow) = 70 chickens. Or another way of looking at it is that you would reduce the animal suffering cost by 7/8
How do the relative health benefits of consuming chicken over beef play into this calculation?
If we're trying to account for global warming/climate change, shouldn't you also be thinking about (1) the cost of feeding cows versus chickens (I'm not an expert but I think it's way more carbon intensive to feed cows, especially if you need to fertilize any pastures, and that's one of the biggest environmental reasons to avoid eating beef), and (2) a lot of modern farming involves transporting cattle between different locations depending on where they are in their lifecyles and there is a lot of CO2 emissions in moving them?
Start by not eating factory farmed chicken or anything else factory farmed or out of a feedlot for that matter.
We should also take account that there are easier way to reduce net CO2 (and methane): a revenue neutral tax on net emissions.
Now I'm curious about lamb and pork.
What about pork? It feels like it should be the best of both worlds.
Well, I roasted a chicken yesterday and I'm thinking of boiling some beef tomorrow. I can't work out if I'm part of History's Greatest Monsters or Doing My Bit To Fight Carbon Emissions 😀
Does the economic calculus depend on which bit of the animal you're eating?
Is a joint of lamb more or less elastic than haggis (in terms of the total number of animals slaughtered), for instance?
This post could not be a better illustration on why I don't perform moral calculations over my food.
As far as moral concern goes, I think it's right to act your rational conviction, but I can't honestly surmount my own doubt that it makes sense to care about animal wellbeing. Animals are just a significant degree less interesting than humans. If I really am to say that chickens have moral worth, I don't see any easy spot to get off that train between chickens and insects. It annoys me when animal rights activists act like it's SO OBVIOUS to care about chickens but don't have an extremely OBVIOUS reason to ignore the potential moral worth of mosquitos or flies. The human race has deployed a gene-drive edited mosquito population in Florida to eradicate a particular kind of mosquito, we're literally doing very-intentionally-engineered genocide on some mosquitoes and while there are detractors of this scheme their criticism seems mostly centered on gene-drives being potentially dangerous and hard-to-control technology with unpredictable outcomes. As far as I've followed the issue, I haven't heard the perspective that we ought not genocide this population of mosquitoes because they have moral worth.
I can, however, contrive a reciprocity test. Which is that cattle are very kind creatures. Chickens are not. Pigs are not. Insects are not. You would not leave a young child in alone in a barn with chickens or pigs as you would be concerned that they would get eaten. You would not have the same concern with cows, apart from getting accidentally kicked or stepped on. I think this makes sense and is not-hypocritical. It's roughly the same logic we apply to humans under liberalism: as long as that human would not harm me, I have no right to harm it. But if we have evidence to believe that a given human WOULD likely bring harm to others, we lock them up.
I think a similar heuristic extends to animals reasonably well, and it leads me to eat chickens and not cows.
Something about the offset pricing seems wrong. If it's true that the average American produces 17.5 tons of CO2 per and equivalent per year, and that a ton of carbon can be effectively offset for $10 per ton, then an average American could offset his entire carbon output for $175 per year? And the entire country's annual CO2 output could be offset for $58 billion per year?
That's objectively a lot of money... but also not that much. Before we consider economies of scale, it's about five times as big as the Pentagon's fuel bill. That's less than twice the Department of Energy's budget. But all this is orders of magnitude cheaper than the "Green New Deal."
If we set annual costs aside and look at renewable energy projects, this also doesn't seem to make sense. The Topaz Solar Farm costs $2.5 B to build and provides power to 160,000 homes. That's orders of magnitude more expensive. To be very conservative, assume a 5% rate of return to private capital, no operating costs, and that a third of an American's carbon emissions are due to home electricity.... you're still looking at a capital cost of $125 M per year to offset the emissions of 60,000 American households, or $1,000 per person offset in capital cost alone.
If the costs involved in offsetting carbon emissions were really so low, this wouldn't be a major geopolitical issue. You wouldn't need the US and China to agree to any climate treaty, or agree to do anything at all, the EU with its strong Green Parties could simply fix global warming by spending 1% of its GDP.
I notice that I am confused.
This logic always seems weird to me because if I choose not to order chicken at a restaurant or buy chicken at a grocery store, I'm not actually keeping a chicken from being eaten, I'm exerting extremely subtle market forces that have some small chance of leading fewer chickens to be raised for food in the future. This goes much more so for cows. I'm not sure how sensitive these markets are to differences in demand.
How do you feel about keeping dogs and cats as pets? I have a large dog who eats several pounds of chicken and beef per day. The net meat consumption of my household is out of control, totally because of my dog.
I feel pretty hopeless about changing my own behavior, given that anything I do will hardly make a dent in my household meat consumption.
Does anyone know what fraction of meat consumption in the US comes from household pets?
The cows replaced the buffalo, who numbered around 50-60 million. There are currently around 94 millions cows in the U.S. .So, you could probably subtract the methane that would have been produced by the buffalo we exterminated from what is currently being produced by the cows to judge the actual impact of beef and dairy operations on climate change.
I find referring to "adjusting for market failure" in this context to be rather odd. "Adjusting for market failure", to me, implies that the market price is a valid first-order approximation, and we're taking into account higher order effects. But climate change isn't priced into meat, and even less so is animal welfare. Calling it a "market failure" when you have a moral agenda that the market is ignoring is really stretching the term. Suppose you believed that using contraception is a moral evil, and you value each act of birth controlled sex at -$100. Clearly, that amount of money isn't priced into birth control pills. Is that a "market failure"? Or is that just the market ignoring your moral preferences?
Low carbon investor here. My assumed social cost of carbon is more like $100 than $10 (off topic, but when I look at the studies that come up with a low cost, their scope is always way too narrow). However, I still prefer beef to chicken, for the animal welfare reasons you set out. Even better, there is an incredible interest among younger cattle raisers in sustainable and regenerative farming practices, and restoring lost carbon to the soil. There are business models emerging whereby you can make good money by adopting such practices. So I expect beef to start moving pretty dramatically towards zero carbon, or even carbon negative, within this decade.
A humanely raised free range chicken is still pretty good sometimes.
I don't understand the underlying assumption that you're responding to, stated but not really proven, that the price of a good (or the cost of an action) has anything to do with its moral value. There's a lot of work being done by this one sentence: "on the grounds that the market has priced its moral cost at lower than chicken's moral cost". What does this actually mean? The market hasn't priced a moral cost at all, and I'm not sure why that would be a default presumption.
In the carbon offset case, the price of an offset is driven by
1. The cost of producing it, eg setting up a biogas digester or hiring guards to make sure a rainforest doesn't get cut down.
2. What people/companies are willing to pay.
3. An incredible thicket of UN and national policy.
As I think another commenter has pointed out, there's a completely separate concept of the social cost of carbon (https://en.wikipedia.org/wiki/Social_cost_of_carbon#Estimates) for which a good estimate might be $100/ton. Unfortunately because this is an externality, and much of it is borne by future generations, there's no mechanism by which it gets incorporated into the demand for offsets and therefore their price, unless many people are actually altruistic and knowledgeable.
For animal welfare, it's even simpler: this isn't the "market price" of anything, it's just the cost of an intervention.
You go on to conclude, I think, that we shouldn't use these prices to determine the ethics of our actions when we don't offset because of "weird market failure" but again I'm not sure where the presumption comes from that a functioning market would set prices equal to the moral costs.
---
A second point, and this may be nitpicking, but I'm not sure what you mean by "Nobody agrees on exactly how much it costs to offset a ton of carbon". There are many sellers of offsets and they all charge different prices! There are some legitimate ways to do it for $4/ton, and other legitimate ways to do it for $15/ton.
You do have to be a bit careful with those prices because the products are pretty different. On the one hand, some of them are perceived to be extremely low-quality (meet the certification but are unlikely to be doing much of anything). Some of them have "co-benefits": replacing coal plants improves air quality, protecting rainforest helps biodiversity. And also some of them are actually prices on the secondary market, where people are forecasting the "how much demand will there be fore these credits before they expire".
In any case I would agree that $10/ton is a fair placeholder value for the category.
I have the feeling this math only makes sense if you eat a wide variety of cuts of beef and chicken. If you only eat beef fillet, for example, you're responsible for killing a cow every single meal, granted that you are sharing that moral burden with everyone that eats the other cuts (and it will be a lot of people, sure).
The same can be said about eating chicken hearts, which are delicious, but I stopped eating them because just one small skewer with those are like 5-10 dead chickens worth for what can be considered a light snack.
The crucial assumption here is that carbon-offsetting actually works in practice. Not surprisingly, a quick search reveals corporations claiming it does, and environmentalist websites claiming it doesn't. This study commissioned by the EU (https://ec.europa.eu/clima/index_en), sides with the environmentalists:
https://ec.europa.eu/clima/sites/clima/files/ets/docs/clean_dev_mechanism_en.pdf
TLDR is that they found that 75% of supposed "offsetting" activities were "unlikely to have resulted in additional emissions reductions" in the sense that they were not "additional" and would have gone ahead anyway, and only 2% had "high" likelihood creating additional emissions reduction.
Dishing out money to charity instead of reducing your carbon footprint seems a lot like buying indulgences from the church -- if someone says you can do bad stuff as long as you pay them afterwards, you should probably be skeptical.
What bothers me is that the only scale of measurement is dollars--as if my throwing green paper at something actually solves anything. I know that we can do things with green paper, I just don't know that it can be used to solve moral and/or ethical problems.
Yeah, but chickens are our historical enemies, descended from dinosaurs, and they're nasty filthy vicious profoundly stupid creatures, so I'm totally OK with their suffering in the service of my appetite.
Of course, in the event it happens that I eat a vegan diet, for my own health. But I would be totally OK with eating chickens if I weren't. Nasty little creatures.
I think this thought experiment is missing the largest point. Cows have best friends, you can tell the difference between a 'happy' cow and a 'sad' cow. Cows can show affection. To my knowledge, chicken's show none of this. So you could say 1 cow consciousness is worth 1,000 chicken consciousness. Or even more than that if you'd like. For instance, almost no one puts moral weight on killing a fly. There seems to be a cutoff point in which something isn't conscious. Where are chicken's and cows on that spectrum? It's certainly not as easy as comparing the number of neurons.
One other calculation to include is amount of suffering, which seems so much higher for a chicken than a cow.
Both the issues of climate impact and animal welfare vary significantly depending how the animals are raised.
Free range farm animals arguably suffer much less than most wild animals. Plentiful food for life then a quick painless death. One could potentially argue that it's a moral obligation to eat these animals to ensure more are brought into existence.
Cows raised on fresh pasture that are 100% grass-fed (or close to it) also have a much lower impact on climate change than feedlot cattle. They can also be raised on land that isn't suitable for crops, negating the argument that it's inefficient use of agricultural land.
The catch of course is that it's probably not possible to produce current levels of beef and chicken under these humane and sustainable conditions.
You're doing more harm than good here. There are plenty of cold hard facts we can use in order to make these decisions. By injecting sophistry like "consciousness scales with the number of neurons,"you're simply muddying the waters and casting doubt on whatever conclusions we draw.
The market price of carbon reduction is so broken that if you buy a certificate to save 1 ton of CO2, a good approximation is that you safe 0 tons. I would argue strongly against basing any arguments on market prices for CO2 emissions.
How do you safe CO2 in this system? You ask the owner of a forest if he promises to not cut down the forest in the next 20 years. A "yes" is not legally binding, you hope that he feels morally obliged to stick to it. The carbon reduction price is the lowest amount of money that someone is willing to pay for a "Yes". If you have even one traitor then this price is ridiculously low. And note that buyers (think of an airline compensating their CO2) may have an incentive to encourage traitors.
I think this is exactly what happens. I don't even think there are mechanisms for preventing the owner to sell the same piece of forest over and over again, generating CO2 "reductions" out of the same forest many times. And then still change his mind a year later and cut down the forest. This are mechanisms against this in some countries, but as long as it works in some other countries, it will determine the market price.
Dedicated people try to have certificates of better practices. Those try to find out whether the owner is actually serious about it. So they ask "Are you REALLY SERIOUS about not cutting down the forest?", and if he still answers Yes, then he gets the money. Despite sounding cynical, I think this helps a bit.
There are also people (on the level of governments) trying to fix the system. Perhaps we will have a working system in 5 or 10 years. But currently, it is so completely broken that any money you pay now will likely have 0 returns. (There are reasons to use it nonetheless. You can speculate that using the system now makes companies get used to it, and that they can't easily withdraw when/if the system actually starts working.)
Is it true that farm animals are still given antibiotics as a matter of course to help them grow big? Can you put a dollar value on contributing to antibiotic resistance?
I think people are emotionally detached from farm animals vs pets and I think it would be useful to include common pets in these questionares as a sort of a control.
for example neurally and in terms of capacity for suffering and moral cost I'd expect pig or a cow to be as valuable as a dog, yet I suspect that most people that had no contact with farm animals disagree, as shown by the western outrage towards dog consumption elsewhere in the world.
What about donating to completely unrelated charities? E.g. you switch to eating beef only but instead of giving that $22 to carbon offsets, you give it to some charity, which you think does even more good. That way, you should come out net positive in the moral calculus.
I try to do a version of this (though I also try to eat less meat in general) and one of the main challenges in practice is knowing whether those $22 really are additional. If you were already giving to charity prior to making this commitment, then whatever sum you donate annually is sort of arbitrary. Say you donated $X last year and want to give a little more this year but haven't decided exactly how much yet. But you also want to offset your meat consumption this year, which is an additional $22. Whatever amount $Y you decide to donate this year, it's hard to tell whether you wouldn't have chosen the same $Y without the moral offsets. If you give to carbon offsets, that's a cleaner psychological separation. If you've made a Giving What We Can pledge or something similar, it's easier again.
It's kind of a stupid argument, because it's really just a psychological effect, but for the reasons Scott described above, it's important because offsetting really only works if you do the offsetting.
Still, I think giving to the absolute best charity (or a donor lottery or whatever) you can think of and using the carbon offsets as a lower bound of how much you should give is a good idea; at least if you're mostly utilitarian like me.
Why not just stop eating meat altogether? No need to pick the lesser of two evils if you can root both of them out.
Red meat like beef is considered more carcinogenic than white meat like chicken, I believe. So I have a huge selfish motive to eat less beef than chicken.
One could argue that consciousness actually scales *exponentially* with neuron number, since there is some network effect going on there. If that is the case, eating 0.5 cow is much worse than 80 chickens, morally speaking.
> We calculate it out and find that eating chicken only causes the equivalent of $1 in damage (since Yog Sothoth eats only one extra galaxy, and we can save one galaxy for a dollar). The implied moral is "Go ahead and eat chicken, it only does $1 worth of damage and that's not much". But actually, it causes an entire galaxy full of intelligent beings to die.
Your error is in assuming that $1 is the market price for a galaxy. That may or may not be the price of saving the marginal galaxy, but the market is obviously not clearing at that price: more galaxies would gladly pay $1 not to be eaten than are doing so. This construct can only exist temporarily, such as if the charity were not yet well-known.
To extend the example to your chicken-suffering-versus-beef-offsets, the problem is worse: you're not valuing the direct value of the benefits. You're finding a value about how much other people would altruistically pay to stop a harm that (especially in the case of the chicken) does not directly affect them. To put it cinematically, "your money or your life" has a different weight than "your money or *his* life."
Now, assuming all of these offset programs are above-board, you as an individual can balance the moral scales (in some sense) by in fact making the offsets. In that case, you're comparing like against like, and you're at a Pareto optimum: neither the rate of chicken suffering nor the carbon content of the atmosphere would increase.
In using altrusim prices to make the argument, however, you are no longer operating in a Pareto framework. Something will be made worse for your actions. And those market prices do not reflect the real value of mitigation or the subjective value to the victims, but the altruistic value to third parties. Don't be a status-eater.
As another point:
> Second, most likely anything you personally do to prevent global warming won't matter at all; either very large-scale actors like states and corporations will fail and there will be various disasters, or the large-scale actors will succeed and we will escape most problems.
This is not in fact true. As far as we can tell, the human cost of climate change is increasing on the margin. The incremental harm of going from 4 to 4.1 degrees of warming is much worse than the harm of going from 2 to 2.1 degrees of warming.
If you think that your emissions would not be replaced on the margin, then reducing them is *more* beneficial if you think that the world will not effectively coordinate. The bigger trouble is in evaluating whether emissions will be replaced, since if you are making this decision in a market framework (e.g., taking an unnecessary drive and filling up the gas tank as a result) then prices will change causing others to change their behaviours.
I'll note that carbon offsets do not yet have this problem, since for now the area appears to be demand-constrained with little offset-price-effect on the margin.
I don't at all understand why the relationship of consciousness to cortical neuron count is assumed to be linear. My prior would be that it is anything but. A network of ten neurons can assuredly not be conscious in any meaningful sense, no matter how you adjust the connection weights. Saying that ten billion such isolated systems have the same consciousness measure as an adult human seems obviously silly.
The Gödel Escher Bach approach of relating consciousness to self-modeling would also suggest this. Some threshold of neural complexity needs to be passed for such modeling to become a realistic possibility. It’s not going to be a discontinuous jump in the “consciousness function”, but it might be a sigmoid-like curve. The question then becomes: What is the slope of that sigmoid? Where on the neuron count scale is the mid point located?
I have no idea. But I don’t think you can just assume chickens and cows are both on the slope, rather than stuck at “basically 0” or “basically 1”. Let alone that they’re both on the slope, and it’s shallow enough that you can just fit a linear function to it.
With our current limited knowledge of consciousness, I don’t think there’s a good answer to this. But where I’d start is looking at actual animal intelligence experiments, rather than neuron count, and try to figure out a likelihood ratio between chickens being conscious and cows being conscious from that.
I’m far from an expert on animal intelligence, but I was under the impression that cows tend to perform far more impressively. A likelihood ratio of 1000+ in favour of the cow doesn’t seem totally ridiculous to me. Which would tilt the math to eating chicken being better.
Scott, I love your writing. Every now and then one of these posts reads like a parody of the rationalist movement though
Interestingly, while this discussion is about chickens and cattle, where you can make some pretty decent arguments about suffering either way, the case where it seems absolutely clear-cut would be swine vs. cattle. Not only do the pigs weigh less, they're _also_ more cognitively advanced and presumably more capable of suffering.
So if you want to go on a meet selection crusade, definitely start by getting rid of pork - that's the no-brainer under this line of argument.
One further (albeit a bit complicated) argument in favour of chicken over beef: it's the best choice in the possible states of the world that are most unfavourable to meat-eating, so the best response to uncertainty.
TL,DR; the beef over chicken policy is the best idea in worlds where the differences in moral value between different animals are smaller, which is also the world where the difference in moral value between humans and other animals is smaller, which is the world where animal lives matter the most.
Longer:
Dan Dennett, Eliezer Yudkowsky and Descartes could be right that moral value more or less 0 below the point where first person experience occurs and a constant value above it, and first person experience isn't very common in animals - so Chimps, Dolphins, babies past a certain age all matter about the same and no other animals do. In which case, eat whatever you want.
OTOH, Jeremy Bentham and Peter Singer could be right that first-person experience isn't the key thing and in fact it's capacity to feel and react to pain that matters, in which case there's a sliding scale of moral value that we can scale with neuron count or something similar, such that a Human doesn't matter millions of times more than a chicken, but maybe only tens to thousands of times more, so there are no sharp declines of moral value with brain complexity.
What's key is that between these two, is a VERY wide range of possibilities. We basically know that human lives matter most, and the question is how the curve looks 'behind' us - does it drop off almost straight away after Chimps or decline completely smoothly, reaching near 0 only at sea sponges and nematode worms?
So there a range of graphs over hugely different orders of magnitude, all the way from a completely flat increase of moral value where the most and least complex animal life aren't all that different (so my life only matters 10 times more than an insect) to a very hard 'exponential' increase where my life, or a chimps, matters a million times more than a dog. Not all of these are equally plausible, but the best argument for caring about animal lives is that considering how horrible factory farming is, most plausible settings of the numbers that aren't just 'animal lives don't matter at all' say it's a bad idea, and it's not that plausible that animal lives don't matter at all.
Here's the bit that's relevant to chickens vs cows - the more carnivore-friendly models of the world are ones where animal lives matter much less than humans - these are ALSO ones where there's a much steeper drop-off past humans because the moral value of animal lives depends on much more sensitively on brain complexity.
In a world where a cow's life matters 1000x more than a Chicken's life, so the climate maths doesn't work out in favour of eating beef over chicken, it's probably also true that a cow's life matters >1000x less than a human life, and so eating meat isn't much of a big deal in general.
Whereas if you think that a Cow's life matters only 2x as much as a chicken's, so the climate maths works out well for eating beef over chicken, that's probably a world where the decline in moral value with capacity to feel pain/brain complexity is much flatter IN GENERAL, so eating meat is morally much worse in general.
I shall explore this somewhat further when I am not also working, but I do wonder if there is a role for a component that honestly weighs heavy in the decisions I make about what to eat:
Is the animal a jerk?
Chickens? Jerks. They'll peck each other to death and stomp on the corpses.
Young male lambs? Jerks. My godfather takes a shovel in with him when dealing with his ram so he can hit it in the head. The ram ignores anything less, and is quite aggressive.
Steers? Often dicks, though not always.
Pigs? Not necessarily dicks. In fact, can be downright pleasant.
I think, all else being equal, it is definitely better to eat nasty animals that are mean to others than to eat ones that are nice and pleasant to all other creatures. Though on another level, should we let the nice ones have a brief, pleasant existence before we eat them?
The reasoning seems spurious to me. Chicken, cattle and pork make up a huge proportion of the animal biomass on the planet specifically because they are bred for food:
https://www.pnas.org/content/115/25/6506
Chicken make up 3x the biomass of all wild birds combined, and cattle make up almost twice the biomass of humans and 14x more than all wild mammals combined.
You can't "save" chicken by offsetting, because there is no wild population of chicken to save, and if there were, it would not be in anywhere near the volume needed for effective offsetting. So unless we start devoting even more space for chicken or cattle as ornamental pets (presumably at a cost orders of magnitude higher than battery-farmed chicken or cattle), the supply of offset credits would simply limit to a tiny elite of virtue-signaling rich people.
https://www.youtube.com/watch?v=isj-IYeCbnI
The above is a training video from the Australian cattle industry about how to humanely process cattle for slaughter. It is worth a watch. Obviously, includes a stungun firing a bolt into a cow's head, and a cow getting it's throat cut.
Having watched it, I am fairly confident that the cow likely doesn't know what's going on and is dead quickly and humanely. I wouldn't want to work in a slaughterhouse, but if you made me watch the slaughter of every animal I eat I probably could.
I would suggest it for people. I definitely learned something about myself watching it.
This problem reminds me a lot of Alasdair MacIntyre's critque of 19th Century Utilitarianism in _After Virtue_ Ch. 6 "Some Consequences of the Failure of the Enlightenment Project". Somewhat more than most things, it's probably a bad idea to jump into the middle of that without knowing the context within the larger argument and work itself (Utilitarianism isn't even the focus of the book). But, to throw in a piece anyway: "For different pleasures and different happinesses are to a large degree incommensurable: there are no scales of quality or quantity on which to weigh them. Consequently appeal to the criteria of pleasure will not tell me whether to drink or swim and appeal to those of happiness cannot decide for me between the life of a monk and that of a soldier. / To have understood the polymorphous character of pleasure and happiness is of course to have rendered those concepts useless for utilitarian purposes... it follows that the notion of the greatest happiness of the greatest number is a notion without any clear content at all. It is indeed a pseudo-concept available for a variety of ideological uses, but no more than that.(MacIntyre - After Virtue, Third Edition, Ch. 6, p.64)
One I always wonder about with this kind of domestic animal problem though: what's the proposed alternative for these particular animals that are not getting eaten? In the short term, I suppose a bunch could just be set free into the wild. Maybe at least some of them would survive for a while in a sufficiently abundant area, maybe there would even be some feral herds that linger. I'm doubtful there would be the same number of said animals in existence in the long run though, and if that is the case, the question is was avoiding suffering so paramount that existence existence or at least life should be forgone? Perhaps we could consider that all the matter that made the chicken or the cow would still exist, and from one view that's equivalent, but then it's hard to figure out why that set of atoms existing as dirt or suffering as a plant rather than as a chicken is particularly better. That's not to say we shouldn't care about suffering, but I think there's a lot more questions to be asked and answered on this kind of problem, and perhaps the framework needs to be scrutinized more.
It looks like nobody has brought up the huge issue with these offset numbers, which is that "saving X chickens" actually means "convincing people to eat X less chickens". This of course has significant costs to those people.
Offsetters are simply shifting the burden of vegetarianism, plus adding the costs of charity operation (including targets watching depressing/disturbing videos).
Offsets only work if you actually consider ALL effects of the charity, not just the metrics they advertise.
>>If you eat the US average of 250,000 calories of meat per year, you can either eat 0.5 cows, or 80 chickens. <<
I never did the math, but I have thought this for years.
What surprises me the most is how cheap it actually is to offset for Co2 emissions... I really like internal combustion engine cars, and I dream about one day owning a really cool ICE car, and if this is the price I have to pay to offset the effects of Co2 emissions I don't even see how it ever became an issue.
What I like most about his post is that it has always bothered me that we have religions that won't eat pork or beef but none, so far as I know, that won't specifically avoid chicken. Could this be the birth of one?
The chicken vrs beef question is an interesting one, but in view a less important one than the standard vrs organic/free range one for the reason that I would think that more people will be inclined to switch to organic rather than choose exclusively chicken or beef
Central to the standard vrs free run benefit / cost analysis are two key considerations: 1) Do animals have "standing" in calculations over welfare and 2) The challenges in quantifying suffering and utility generally
Assuming you agree that animals have standing, then the next question is do some animals have more standing than others and if so, how do you determine these weights (as Scott identifies).
Cost and purchasing power also come into it, as arguably there is more moral imperative for someone who is wealthy to switch to free run, than someone on low income.
Utilitarianism implies a belief in Pantheism
If you care about the suffering of, say, cows, why? Are they your mother?
Some people don’t care at all about the suffering of others. These people are generally called psychopaths. If you are solipsist, it’s logical to be a psychopath. If others don’t exist except in your imagination, you should use them as you wilt.
If you are not a psychopath and care about the suffering of other humans, or even other animals, or even vegetables, why? It’s probably because you view them as other beings, like yourself. You empathize, or at least sympathize with these other beings.
Why do you care about the suffering of other beings? It’s because you consider yourself a being, right? You care about them because they are in the fraternity of beings, a club which you belong to.
What makes you both beings? You experience shit. Who experiences shit? You do. And they do.
Now, are you separate entities, you and the chickens? It’s all semantics from here down, but my macro view is beings are these things made of matter and the entity must exist within the matter. Seems more likely matter itself has qualia than that it’s an emergent quality, because it's a simpler explanation.
Scaling moral worth by (cerebral cortex) neuron count seems like a poor measure to me. By that line of reasoning we'll have to conclude whales are worth as much as a human, using Wiki's data. Orcas are worth 3 humans.
That's without getting into the fact that a lot of things are determined by neuronal connections rather than neuron count. Should we determine moral worth using the number of synapses in the cerebral cortex instead?
Scott, I think a more interesting thread might include a definition of suffering and some down-the-rabbit-hole speculation on why avoiding is preferable. It seems to me that vegans and other holier-than-thou types are striving for a kind of purity that cannot exist in the context of natural life. I mean, let's face it, if you want to minimize the suffering of animals you should develop the skill of marksmanship and become a big game hunter who only takes head shots. Wild animals don't have retirement communities. They either die of starvation (or freeze to death, or bleed out) or they get eaten (often eaten alive) by predators.
I believe that natural life is, at least in large part, suffering. Anyone who denies suffering is actually denying life in the natural world. Sorry vegans, but that combine harvesting the mega monocrop is chopping up everything from terrified fawns to snakes and birds.
People should embrace suffering. We're all gonna die anyway.
Wondering if the Straussian reading of this is that you are trolling...
From these considerations, really by far the most important thing <a href="https://www.unz.com/akarlin/animals/">is to just avoid pork</a>. Or get probably a OOM-tier reduction in cognition-adjusted animal suffering by going pescetarian.
PS. Personally I mostly consume chicken and turkey as animal protein.
The section that attempts to compare the relative moral worth of the mental life of a cow and a chicken feels extremely hand-wavy.
Shouldn't these calculations take into account wild animal deaths? There are diffuse causes that are hard to pin down, like global warming, pollution, and land converted for growing feed, but cattle ranchers kill a lot of things directly (wolves, bears, mountain lions, snakes), sometimes in really unpleasant ways (prairie dogs). Ugh and all the deer etc that get tangled in barbed wire. I guess it's grimly nice that there is less wild animal suffering per beef produced as rangeland is defaunated.
I appreciate the reasoning, if you're reluctant to eat creatures, and some of them are one meal, and others are thousands of meals... the math is pretty stark.
Maybe pilot whales are the result of a timeless evolutionary process, where they adopt decoy cneurons to defend against time-traveling rationalists who prefer whale meat over insect-based protein.
Cheap jokes aside, I do worry about moral weights scaling cleanly with cneurons. And this is something I've worried about for a long time.
In fairness, I'm also very worried I'm getting this wrong, given the amount of cycles you've spent on this too. I absolutely clicked through to read all the times you've previously discussed it, and find the surveys fascinating and challenging.
So I want to say pretty cautiously, I think I have an idea why the surveys come out that way, while not being particularly normative. But I'm still committed to thinking about this in more detail.
I'll list a few problem areas, then a possible explanation for why those emerge, and a few alternatives (none of which are particularly satisfying, but might prompt someone smarter than me to move the ball further forward.)
My initial concern is that the cneuron:moral weight analysis produces weird results at the margins. Does a whale's worth of mosquitos have any moral weight at all? Is there a Jupiter sized swarm of (galactic space faring) mosquitos we should sacrifice humanity for?
For digital intelligence, is suspending a process on my CPU similar in kind, but just not degree, to ending an AI with agency, identity, goals, a way to contextualize its own history?
Then at the other extreme, how much more moral worth does a post-singularity paperclip maximizer hold over you and me?
The extremes are relevant, because when we're talking about animals at 1-2% of our cneurons, then we are already at some intuition breaking extremes. People report feeling moral responsibility towards robots that so much as mimic human facial expressions, with the processing power of a toy. People generally feel more moral responsibility towards "cute" animals relative to "ugly" animals. They aren't great at surveys of moral weight at extremes.
I also worry about this for close comparisons. Do smarter humans hold greater moral weight than others? I can imagine lifeboat ethics situations where intelligence is the most important skill to ensure the survival of the greatest number. But as a universal principle? I can imagine somewhat more fanciful scenarios where the best person at growing soybeans, juggling chainsaws, or mixing strong drinks that appease angry aliens are the most critical to everyone's survival.
Are we sure we want to put down that dogs are *roughly twice* as morally valuable as cats, even though there's enough ambiguity in that *factor of two* difference that owners can still endlessly debate relative intelligence?
So close comparisons and extreme marginal cases both pose some problems. Yet a clear scale seems to emerge in surveys, and track our rough intuitions, with more confidence on one end.
What could cause that?
Maybe the scale is right, but there are just hard cases. A lot of hard cases clustered in those domains.
Another natural way to get those results, and those same problems, is to have a sorites paradox about a binary condition. Any bright line feels arbitrary, so you get a smooth-ish distribution in polls, with more certainty at the ends. But you're still measuring essentially a binary (or at least a very steep) condition.
To break a sorites paradox you need to think carefully about what constitutes that threshold. (Or bite the bullet and go with an arbitrary line). You'll still get probabilistic arguments near that threshold, and that's fine, but it will allow you to consider a much narrower range than "anything with any processing power whatsoever."
If reducing the firing of nociceptors is your main overriding concern, if pain signals buttress your entire morality, then ok, nevermind, this doesn't need any touch ups.
But maybe harm reduction is "mostly right," in the way that Newton was "mostly right." And now we're at the end of the paradigm runway, where new models that provide additional resolution are necessary (while sure, preserving most of the previous results).
Imagine a weird lab that ran current through isolated nociceptors. Great tragedy or no? If not, why not?
I'd propose that neurons in petri dishes have no moral weight because of larger factors, like, (a) they can't contextualize pain, or, less likely (b) they aren't part of a moral community that can engage in exchanges of moral behavior, or, Sam Harris's (c) they don't contribute to the flourishing of humanity, or, the more Stross-ian (d) they are unlikely to contribute meaningfully to maxing out the instructions per second per gram in our galaxy. A proxy for (d) might be likelihood the species' descendants will experiment with genomic editing or interplanetary travel.
None of those are problem-free, I'll admit. But any of them provides a framework you can extend to build a cutoff that falls short of humans and isn't completely speciesest, providing opportunities for promising high intelligence species like corvids, but without going all the way down the chain. This is useful if you do not believe that mere sentience, plus sufficient mass, becomes more important than sapience.
But maybe I'm wrong, and should alter my consumption based on the p() of confidence in this counterargument, which isn't stupendously high...
My personal intuition is that there is a cut off-- like chicken suffering doesnt matter 1/6 as much as cow suffering it just doesnt matter, because for suffering to matter you have to reach the complexity of a dog, I think in retrospect i mostly just wanted to eat chicken though.
I won't say the 'suffering of chickens vs cows' question is worth *nothing* but if I'm being honest I think the ecological impact question trumps the suffering question by a ratio so large we can safely round it up to infinity.
Trying to gauge the moral gravity of fewer numbers of relatively larger beasts vs greater numbers of simpler beasts (and let us note in passing that none of them are ever going to feel grief, or write histories; and their slaughter-as-opposed-to-living-out-their-lives-naturally will not leave any lasting impact on the world whatsoever) seems like exactly the kind of question a philosopher might pose as a reductio of utilitarianism because *obviously* it has no answer.
If insects are so good, why didn't the Gods demand the sacrifice of hecatombs of cicadas?
And why should the gods of postmodernity be less finicky ?
lhttps://vvattsupwiththat.blogspot.com/2021/06/it-takes-tough-devourer-of-galaxies-to.html
There’s also another benefit though not all of us will see it that way. Eating a lot of red meat could result in earlier mortality, reducing social security and Medicare costs to the government.
Isn't there a health dimension to this also? Eating red meat that is typically fatty increases cholesterol, which has an impact om CO2
Why is veganism not being discussed? It is truly the superior option if trying to reduce animal suffering and environmental impact.
How does turkey figure into this? Roughly the same moral weight as eating chicken?
I just love your style m8. It doesn't matter wrong or no. That's besides the point. The themes you write on a hard per se to get right. But your style of writing is nice and good and funny in a non-funny way. Glad I have your blog to make my day!