About this page
This page collects my replies to questions that I'm occasionally asked. I titled this page "Frequently asked questions" because that's the standard name for this type of page, but not all these questions are frequently asked; some have only been asked once or twice.
In most cases I made up the text of the questions to generally express the gist of what I'm asked. In some cases I've quoted specific questions from readers with their permission.
As this page grows over time, I might split sections off into their own subpages.
Food animals
Are you vegan? If not, why not?
Until 2005, I was an omnivore because I hadn't thought about animal suffering as a serious moral issue and didn't even know if non-human animals were conscious. After reading a few Peter Singer articles in late spring 2005, I became roughly vegan, though I still ate cheese until fall 2005 when I became fully vegan. I didn't eat any meat, eggs, or milk except if there were leftovers that would have otherwise been thrown out. I remained fully vegan until ~2008, at which point I decided to add some animal protein—namely dairy—back into my diet as a health precaution. In my opinion, dairy is roughly the least bad form of animal protein in the standard Western diet (Tomasik "How Much Direct ..."), because a cow produces so much milk per day. (For this reason, I find it unfortunate that animal-rights activists often focus on opposing dairy in their messaging.) As far as greenhouse-gas emissions, dairy is comparable to most other non-beef animal products.
Reddit ("Why isn't ...") is a Reddit question asking why I'm lacto-vegetarian rather than vegan. The main answer I gave on that thread was as follows:
Originally it started out as a combination of saving money on protein powder (plant powders were more expensive) and hedging one's bets from a health perspective in a way that's relatively low-impact from an animal-suffering perspective. Now I continue consuming dairy because it contributes to feeling good in a way that just tofu, nuts, lentils, etc alone don't seem to offer, although maybe I'm just addicted or could transition to feeling the same way with a vegan diet. Combined with the fact that the sign of the overall impact is unclear in light of wild-animal suffering, I haven't prioritized attempting to change this particular practice compared with other things (though as a lazy human, I of course spend some time on less useful things than this!). I don't see the "vegan" line as particularly special, since there are always so many ways to make a difference. Earning a few hundred dollars less per year that you then can't donate to veg charities presumably causes more harm than eating dairy yourself (assuming dairy causes net harm). Different people have different self-imposed guidelines about what moral compromises they're willing to make for the sake of convenience/selfishness. For example, I used to go running in the woods but stopped partly to avoid crushing tons of bugs; many vegans who would never consume dairy are ok with walking in the woods or fields.
For exploration of how various animal food products might affect wild-animal suffering, see Tomasik ("How Does Vegetarianism ...").
Despite consuming dairy products (mostly cheese), I still strictly avoid eggs, because they cause so much suffering for a given amount of food (roughly a day of hen suffering per egg). As of 2020, I also try to avoid
- honey (see the "Honey" section of Tomasik ("Applied Welfare Biology ..."))
- candies containing confectioner's glaze (Tomasik "Insect Suffering from ...")
- rice (see the "Deaths in rice production" section of Tomasik ("Vegans Should Not ..."))
- almonds (because they require a lot of farmed bees and a lot of irrigation (Tomasik "How Irrigation ..."), though I haven't yet looked into the details of almond farming).
These rules are relatively easy for me to follow, whereas if you find it hard to give up almonds but easy to give up cheese, then that's what works for you. I haven't done a rigorous review of which foods cause more total suffering than which other foods, so any "do not eat" list that I might come up with will be somewhat random and arbitrary.
Wild-animal suffering
Do you think we should prevent the existence of all wild animals on Earth?
It's important to make a distinction between what I think is morally best relative to my values, versus what I would advocate in practice given that most of the world doesn't and likely will never share my values.
Most wild animals endure at least some moments of unbearable agony, such as if they're eaten alive. If the animals could talk, they would during such moments perhaps say that they wish they hadn't existed in order to avoid experiencing so much pain. Those horrendous moments of an animal's existence aren't "offset" by other happy moments of the animal's life, because each moment of an animal's existence is a "separate person" from each other moment, in light of empty individualism. Therefore, creating a life containing some moments of extreme involuntary suffering means forcing some individuals (namely those horrific moments) to suffer for the possible benefit of other individuals (namely the happier moments). This seems morally wrong. Even if one rejects this view of personal identity, it's obvious that some animals experience severe pain throughout most of their lives, and perpetuating ecosystems means forcing into existence some animals who during most of their lives are in sufficient pain that if they could talk they would wish never to have been born. Once again, this seems morally wrong.
Unfortunately, most of humanity apparently doesn't share my perspective on this issue, and given the frequency of "life is a good thing" sentiments among humans, I expect most people never will agree with my stance. Therefore, in practice, if one wants to reduce some wild-animal suffering in the short run, one has to find ways of doing so that are actually achievable. One approach is to look for policies that would reduce plant growth (and therefore the amount of food available to create new wild animals) that humans can get on board with for other reasons. The "Sample government policies" section of Tomasik ("Cost-Effectiveness ...") has a few ideas along these lines, though probably many more could be imagined.
Bug suffering
Does driving or walking on the side of a road cause more bug suffering per kilometer? (The question and some of the considerations I discuss below were sent to me by Brian Kateman.)
It's a great question, and the answer may depend on the details of the situation.
In terms of the amount of ground area covered, it seems cars are clearly higher, because tires are wider than feet and because tires hit the entire length of pavement rather than just whatever portion of the stride length the feet cover. That said, maybe if there are lots of cars on a road but few pedestrians on the sidewalk, then the marginal area covered by an additional car would be less significant, since if a bug is on the road, perhaps it would be squished by the other cars already, whereas the pedestrian could step on bugs that no one else would have hit. Slower-moving bugs like worms and slugs seem more likely to be hit eventually on a high-traffic road compared with faster-moving bugs like ants.
Another consideration is the number of bugs per square meter of ground on the road versus on the side of a road. This difference probably depends on what kind of ground the pedestrian is walking on. Paved sidewalks tend to not have that many bugs, although they still probably have more bugs per unit area than the middle of a road does because the sidewalk is closer to grass on the side and has cracks out of which ants may emerge. In contrast, if you're walking on grassy soil on the side of the road, the bug densities might be quite high. There can be thousands of mites and springtails per square meter of grassy soil. I think it's probably better to drive on pavement than walk on grass.
I live in a rural area of upstate New York. The roads near me don't have paved sidewalks and instead just have about half a meter of gravel on either side. On , I walked on four roads near my house in an effort to measure how many visible bugs were on the pavement and sides of the roads, in order to roughly calculate how many bugs a pedestrian might kill by walking there. My measurements weren't precise; I just carried a yardstick with me, put it near the ground, and tried to estimate how many bugs I saw per square meter of ground in various randomly chosen places. These visible bugs were almost entirely ants. I also saw one beetle, one spider, and one springtail. The following table shows densities of visible bugs that I found on pavement versus on the side gravel, where each entry in the table is an average of several different spots on the ground where I looked.
| Visible bugs per m2 on road | Visible bugs per m2 in side gravel | |
|---|---|---|
| Road 1 | 0.2 | 3 |
| Road 2 | 0.3 | 4 |
| Road 3 | 0.3 | 5 |
| Road 4 | 0.2 | 5 |
We can see that bug densities on the side of the road are 10 to 20 times higher than in the road itself, which could make up for the larger ground area hit by tires compared with feet. On the other hand, these rural roads don't have that much car traffic, so most of the time it's possible to walk on the road pavement, and you only need to veer off into the gravel when a car passes.
Using the numbers I collected, and assuming that a pedestrian would spend 90% of the time walking on pavement and only 10% of the time on side gravel, I estimated that a pedestrian who wasn't actively looking at the ground to avoid stepping on bugs would squish roughly 0.02 visible bugs per step, which translates to roughly 20 visible bugs squished per kilometer. All of my numbers reported here have been rounded to 1 significant figure to avoid false precision, though the spreadsheet I used keeps all the digits. This is only the number of visible bugs (mainly ants), and if we counted hard-to-see bugs like small springtails and mites, the numbers would be somewhat higher. 20 visible bugs per kilometer seems somewhat high, so maybe the actual number is a bit lower?
A final consideration is how painful each squishing is. Cars are heavy, so maybe some of the bugs squished by car tires are flattened pretty well, which is less painful than the incomplete crushing that a footstep might cause. That said, car tires have tread patterns, so it's possible that only half of a bug would be squished by the tire. Also, some paved roads contain lots of small holes in them, meaning that even a fully flat car tire might leave some of the bug uncrushed if parts of the bug were pushed into those holes. Cars can also hit bugs in the air, and that may not fully kill them either. In 2016, I examined roadkilled bugs near my house, and I found that a decent portion of them (such as grasshoppers) weren't fully dead. Cars can also kill vertebrates like rabbits, frogs, turtles, and snakes.
People tend to drive many more kilometers than they walk, so I presume that driving has a bigger total impact than walking on the side of roads for the average American. That said, walking on bug-dense ground like grass or decaying leaves seems to me plausibly much worse than driving on pavement per unit distance traveled.
Small creatures
Should we stop playing video games to avoid causing suffering to the enemies we kill?
No, I think the moral cost of killing enemies in video games is basically negligible compared with the moral cost of other things, at least until the cognitive sophistication of video-game characters grows significantly.
My Vox interview on this topic (Matthews 2014) had the headline "This guy thinks killing video game characters is immoral", but that was a somewhat clickbaity title perhaps chosen by an editor. A more accurate title might be: "This guy thinks there's an extremely small but nonzero moral cost to violating the preferences of video-game characters including but not limited to by killing them". The fact that something has a nonzero moral cost doesn't mean it's wrong overall. Literally everything we do has a nonzero moral cost (as well as a nonzero moral benefit).
The moral cost of playing video games may be similar to the moral cost of running a comparable number of computations in other ways on your computer. In both cases there are short-lived "subroutines" whose micro-preferences get violated in various ways, including sometimes by the abrupt termination of the subroutines (Tomasik "What Are Suffering Subroutines?").
A single biological cell plausibly has much more complexity and sentience (Tomasik "Cognitive Abilities ...") than a video-game character. The death of a single bacterium in your body might be a bigger moral tragedy than the slaughter of hundreds of soldiers in a battle game. Of course, neither of these things is a very big tragedy, but I'm giving a sense of the scale of moral priorities as I see them.
Are ejaculations immoral because each one kills tens to hundreds of millions of sperm cells?
The practical answer is that even if ejaculating were slightly morally costly, it would be best to ignore that fact in order to save your mental resources for more important things. I imagine that other routine activities, like spitting or washing dirty items in your house, also kill tons of unicellular organisms, though I don't know enough about bacteria for this to be more than speculation. Washing bacteria from your hands to reduce your chance of getting sick outweighs any moral costs of doing so, assuming your life has positive altruistic value. Likewise, the mental-health benefits of ejaculation probably outweigh moral costs to the sperm cells, assuming your life has positive altruistic value.
As far as the theoretical question: it partly depends whether ejaculation increases new sperm production. I don't know if it does, and if not, then the sperm will be "born" and then die soon afterward whether you emit them from your body or not. Also, even if fewer sperm are born by not ejaculating, the food you would have eaten to create extra sperm will instead be eaten by someone else (such as bacteria), so it's not clear if the total amount of unicellular creatures is reduced after all. This illustrates why I tend to focus on net primary productivity (Tomasik "Net Primary Productivity ..."), since net primary productivity controls the total amount of heterotrophic life that ends up being created. Getting rid of your houseplants so that they can't photosynthesize any more food would plausibly reduce more unicellular suffering than fretting about how many times you ejaculate.
In my particular case, I have a vasectomy, so my sperm don't enter the ejaculate. Therefore, regardless of how often I ejaculate, I assume the sperm end up with the same fate. Of course, most of the ejaculate load remains even after a vasectomy, since only 2 to 5 percent of the volume of a normal ejaculation is sperm.