In case 1, I don't think you're making the relevant comparison. The question is not of the child the woman would have had before she stopped drinking versus the child she has after she stopped drinking; it's of the child she has after she stopped drinking versus how that same child might have turned out had she not stopped drinking. We're not comparing past and present, but two different possible presents.
I honestly do not understand how anyone can object to this. Sure, there are your religious pro-life people with the "god maybe wanted you to have that schizophrenic baby," given their complete lack of empathy for women carrying issue doomed to die shortly after birth, but preventing human suffering, if possible, is a net good.
But even this isn’t an argument against polygenic selection. It’s an argument against IVF in general, which usually involves production of more embryos than the couple intend to bring to term.
I think that there is such an argument, if you accept the premise. Polygenic selection makes IVF more popular, so more people will do it. This is the reason that we make it illegal to pay for murder — it makes murder more popular. You still have to accept the non-selection is homicide premise to get to that conclusion, but the structure of your argument was that you were momentarily accepting it for the sake of your argument.
> It’s an argument against IVF in general, which usually involves production of more embryos than the couple intend to bring to term.
It's also more crisply totally incompatible with PGT-M. This line of argument doesn't allow discarding an embryo which inherited Huntington's Disease, where a person could live a couple healthy decades first.
I don't think the "life begins at conception" people honestly have thought through these cases carefully so I don't want to say it's a reasoned argument against PGT-M, but the two can't really coexist.
I must be way out of the loop, because I cannot imagine someone who is not opposed to IVF based on “life begins at conception” grounds but is opposed to embryo selection because people should be born with diseases (and it’s somehow bad to prevent that). I don’t know anyone like this and my theory of mind is also failing me here
My old anatomy textbook from college mentioned that a fairly large percentage of fertilized embryos pass without implanting. (Just looked it up again and Google says 10%-40%.) I think it you're asking the question, "when does a woman have a moral obligation to a fertilized egg?" it doesn't make sense to set the bar before implantation. Otherwise, you'd need to worry about embryos that naturally fail to implant. But you'd never even know about them anyway, most of the time.
i think schizophrenia and autism, two highly heritable conditions, are going to be targeted first by polygenic selection (OK, real talk, thre are ppl who have already done this...)
Polygenic selection seems like a great idea, but I do have a question on it (unrelated to this post, which is about an argument I find ridiculous).
If widespread polygenic selection leads to selecting embryos that match the comparatively few characteristics we think are good, isn't it quite possible to face unintended negative consequences from lack of genetic diversity? In a book I read, 'Seeing like a State', European foresters thought it was a great idea to replace wild forest with monoculture plantations of trees they thought were good, but in retrospect they didn't have a good enough understanding of forestry science and it turned out to be a mistake.
You say: "But once you’re already doing IVF, selecting the embryos based on some criterion, like low schizophrenia risk, doesn’t make this issue any worse."
The criterion in the schizophrenia case is selecting for non-schizophrenic genes, whereas in the other cases it was more or less random, with respect to genes. . In the schizophrenia case, it might help to eliminate the schizophrenia genes from the gene pool. But some people might still object to this. If they were able to select for alcoholic genes and eliminate those, some might argue that is bad because alcoholics might tend to be more creative than non alcoholic. (For example, a lot of writers are alcoholics.)
Or if you could select for gay genes, some might say we shouldn't be trying to eliminate gays.
The schizophrenia complaint at the beginning has a much simpler rejoinder - by selecting any of the embryos, they are necessarily not selecting 9 others. In this framing, all possible selections, regardless of schizophrenia genes, have equal moral weight. (Whether one thinks this is a positive or moral weight is beside the point; the point is that they are all equal)
The correct answer here is "no one", since it was never proven to be an effective practice in the first place.
Other than that, congratulations for rediscovering the arguments of the late-90ies "is-IVF-morally-okay" debate, followed closely by the somewhat less dumb early-2000s "is-PIGD-murder" debate. If you keep this up, we'll have an "eugenics-will-save-us-all" post in a few months, followed by a "are-we-the-bad-guys" realisation in late summer.
Pardon the sarcasm, but if you've dug yourself into a hole, stop digging. Every respectable medical and scientific society spoke out against this previously, and the data are not coming in to change that.
It was a real comment, not pedantry, so thanks for engaging! I think another natural parallel is abortion for the purpose of sex selection, which many people (including pro-choice people) find very off-putting but is very common globally. We certainly don't call it "preventing femaleness". Specific to IVF, some countries (Germany) knowing the sex of your IVF embryos is illegal to avoid sex selection via IVF.
Fwiw I do feel there is a spectrum, and that while selecting embryos to avoid conditions like "female baby" or "probably shorter than my other embryos" does seem immoral to me, I would never advise a woman to transfer an embryo with (say) Edwards syndrome. Schizophrenia is in between those two.
Eugenics: fine on an individual free-choice level, bad on a societally imposed level. Does this seem like common sense? Is it a strawman of Scott's position?
Is it just mood affiliation/fear of the slippery slope to eugenics? (a bullet I'm happy to bite: why would you not want to give your child the best genes possible, regardless of the historical connotation)
People make the argument against polygenic selection because it forces them to think about the influence of genes on various life outcomes. This in turn forces them to acknowledge that not everyone is born equal. But acknowledging this will get you cancelled in certain circles, so you have to try and construct an informal taboo against anything connected to the subject.
As a firm "life begins at conception" pro-lifer, I think Scott conflates two questions in his discussion of our perspective. Namely, "Does polygenic screening cause additional harm?" and "Does polygenic screening prevent schizophrenia?"
It's fair (if horrifying) to say that if you're already killing 9 out of 10 embryonic human beings, there's no additional harm in making sure that you kill the ones with various undesirable conditions. But this doesn't translate into it being fair to say that this process "prevents schizophrenia". You're just killing the schizophrenics.
Suppose a Nazi death camp has a doctor assigned to select certain quota of Jews to be spared the gas chamber and put to work in a labor camp instead. And suppose he makes sure to pick the strongest and healthiest ones without conditions like schizophrenia. The Nazi doctor is not, strictly speaking, making the situation worse. But he's also not "preventing schizophrenia".
I don't think you can make any strong arguments in this granular case. It would be most fair to evaluate the rule and think of more refinements, genetic therapies, and other items that dip more and more into modifications, and on a larger scale, and perhaps less voluntary as far as buy in goes, or perhaps expected.
One consideration I think this leaves out: it can be a heavier burden on both parents and children when parents are more the “authors” of their children through trait selection/abortion rather than the “receivers” of their child, open to uncertainty.
“The introduction of a choice reshapes the terrain on which we all stand. To opt out of testing is to become someone who chose to opt out. To test and end a pregnancy because of Down syndrome is to become someone who chose not to have a child with a disability. To test and continue the pregnancy after a Down syndrome diagnosis is to become someone who chose to have a child with a disability. Each choice puts you behind one demarcating line or another. There is no neutral ground, except perhaps in hoping that the test comes back negative and you never have to choose what’s next.”
Can you save someone from a traffic accident? After all, the person you saved was by definition not a victim in the traffic accident. You merely preferred to go down the branch in which the person was unharmed rather than the world branch you would share with the traffic accident victim, you terrible ableist.
--
I think conscious beings should get special treatment here. Their preferences regarding if and how they would continue to exist should be respected (within reason). They should not be euthanized and replaced with other conscious beings, unless that is what they want. Thus we do not have to debate if healing a mental or physical illness will just replace an ill person with a healthier (plus or minus scare quotes) one, simply follow the patient's preference.
(There might be some exemptions for young children and psychiatric patients reduced to the cognitive capabilities of young children. Basically, they get to have preferences, but sometimes their preferences may be overridden because of their cognitive limitations.)
So saving the traffic victim is the right choice, because presumably that is what their preferences are. By contrast, plants, insects and embryos do not get that protection, they are morally interchangeable. (At least until you add time travel, traveling into the past and preventing your enemy from being conceived by gifting their parents two tickets for the opera is probably evil.)
> But even this isn’t an argument against polygenic selection. It’s an argument against IVF in general, which usually involves production of more embryos than the couple intend to bring to term... But once you’re already doing IVF, selecting the embryos based on some criterion, like low schizophrenia risk, doesn’t make this issue any worse.
I think that depends on for what reason one chooses to do the IVF. If the purpose of the IVF is to solely select embryos based on some (whatever) criterion (e.g. eliminate schizophrenia), then polygenic selection in this case inherit the moral argument against IVF as well hence the anti-abortionist do have a case here.
I think the strongest objection may be from people who are, not necessarily anti-abortion or anti-IVF, but pro-schizophrenia. When I asked my schizophrenic friend what he thought about this situation he said, "I think we're doomed as a society if people start doing that."
His argument is essentially (a) that schizophrenic people have access to perspectives and abilities that others don't have and that those things are valuable, and (b) that society is narrowing down so that only a very narrow set of characteristics can actually thrive in it, and that THAT is a bad thing, and this is both symptom and cause of the acceleration of that process. Schizophrenia outcomes are much better in societies that don't stigmatize it, so from the perspective of someone who sees value in being a way that not everybody is, I think the answer to who polygenic selection helps is "nobody, but it makes a lot of people think they've been helped, and that's a very bad thing."
I don't know if I agree with him or not overall but I have sympathy for that perspective. People without schizophrenia probably wouldn't have the same thoughts; in which case, it's exactly an example of the kind of perspective that schizophrenic people bring into society that others would not have access to without those people's existence. What do we lose when we select those people out on purpose? What if we don't even know?
I also grant that schizophrenia is a special case to him and by extension to me, by virtue of discussion - I have no idea what he'd think about fetal alcohol syndrome. I should ask...
When some people feel a vague distaste to something new they sometimes make up a line of moralistic-sounding bullshit to make this distaste count in an argument. Arguing against pre-implantation embryo selection on the grounds that it does not help an individual is just that, and it is not grounded in legitimate moral reasoning. Scott completely demolished this argument but I'd like add another nail in the coffin:
An individual human is in part shaped but not wholly defined by his genes. We are spiritual beings first and foremost, defined by our memories, desires, social relationships and other indexical features that are created as the brain develops. An embryo is not a person, it is merely a molecular machine that runs a program to create the material vessel in which our spirit gradually grows. "Scott's first baby" is a person, nestled in his unique place in the world, with a unique relationship with his parents. Substituting another molecular machine before Scott's baby was delivered into his unique place would not make it another person. It would be still Scott's first baby, just with a different set of genes.
This way of thinking might feel counterintuitive at first but it actually flows directly from realizing that the individual human existence does not start with some arbitrary molecular event, like the formation of a zygote and a new diploid set of genes. Instead, the human individual grows into being gradually as a neural network that a bootstraps itself into consciousness, many months later. If you believe that this neural stage of development is the true beginning then changing genes at an earlier stage is no longer special, it has the same ontological status as all the other events that impact future humans, be it prenatal vitamins or alcohol abuse counseling of parents.
It is therefore right and proper that we give future individuals the best sets of genes we can, so they grow into more perfect beings.
I think attacking any of this from the angle of where life begins and what can be constituted as "replacing one child with the other" is entirely the wrong model to think about this. All of this - fetal alcohol syndrome prevention, IVF, etc - *feels wrong* to some people because it's artificial, civilized, technological interference with a process of childbirth, which *feels* natural. It feels like it should be a beautiful and chaotic and unpredictable leap-of-faith process, and if you got a schizophrenic child, well, sucks ass, that's fate.
It's the same logic that drives the discussion about "GMO foods" and "would you like tomatoes with no genes and DNA in them".
(for the record, I do not espouse any of these views, but I'm confident people do)
The big difference between polygenic selection and your three examples is that of expressing a very specific judgement. Example 3 involves no choice, examples 1 or 2 just say "I want my kid to turn out better rather than worse." But polygenic selection very specifically implies "schizophrenia sucks, my kid shouldn't have it". And I think that activates a strong anti-judging instinct in many people's heads.
Logically it is perfectly consistent to say "I'd prefer a world with no schizophrenic people, but the schizophrenic people that already exist should be treated with compassion and understanding". But thinking that is hard unless you're a super high-decoupler. For many people, hearing "my kid shouldn't have schizophrenia" subconsciously implies "I think schizophrenics are Evil" - and so they feel icky about it.
(Also, this strongly reminds me of the debate around selective abortion in cases of a prenatal Down Syndrome diagnosis. The feelings involved seem the same, even though no IVF is involved.)
Most of the arguments I've seen around Polygenic Selection center around second-order effects—they're slippery slope arguments, or Fence arguments, saying that if we start selecting against things that are debatably bad (like schizophrenia), then we're going to start selecting against things that are very clearly mixed (like mild autism), then we're going to see pressure to start selecting FOR things that are very clearly bad for the individual but good for society (like obedience to authority). How do you rate these second order effects?
I have no moral or other objections to screening embryos for general embryo health, or for the chance that they will grow up to be schizophrenic; implanting the ones with the least risk of the relevant disease; and discarding the rest. But I don't think it's quite accurate to call what you're doing preventing schizophrenia.
Let's say somebody sets up a clinic in some lawless part of the world, where they will test a couple's baby for genes strongly associated with schizophrenia and euthanize the infants found to be at high risk. In this lawless place there are also many people willing to sell their newborns, so the clinic is also able to provide the couple with a new infant, certified to be low risk. Would we call that preventing schizophrenia? I don't think so, and I am really talking about linguistic matters here, not moral ones. Seems to me the normal way to describe what's happened is that a schizophrenic-to-be was denied a chance to grow up, because their growing up included the very unfortunate feature of their becoming schizophrenic. Discarding embryos with high risk would be described the same way.
Or here’s another thought experiment: Let’s say we euthanized all people at age 65. That would undoubtedly reduce the number of deaths from cancer, heart disease, etc. It would reduce the absolute number, but also the fraction of the population that dies of one of these diseases. But you wouldn’t describe that approach as preventing a bunch of deaths from cancer and heart disease, would you?
Maybe I haven’t fully grasped what the point of your argument is. But it seems to me like a sort of convoluted effort to find a non-inflammatory and in fact positive way of describing the process of discarding embryos with high genetic risk of growing up to be schizophrenics. I don’t think it is going to work to change any public attitudes. I think you have to fall back on saying that embryos are little more than packets of genetic info. They’re not alive, not conscious, not people. Consequently, it is sensible and not evil to discard the defective ones.
Am I right in thinking we don't currently have the technology to harvest some eggs and sperm from a couple, genetically screen the eggs and sperm individually, and then make a single IVF embryo from a "winning" pair?
If we did, I'm curious to see how that would change people's intuitions and moral calculus. I expect those who believe that life begins at conception and that screening and discarding embryos is a bit like screening and killing babies would be OK with this alternative, and those who object to schizophrenia screening on disability-rights grounds would not, and those who object to it on the grounds of playing God or meddling with fate also would not.
I am vaguely in the first camp (although not militantly, and I acknowledge it can be the lesser of two evils), and I would be basically OK with the technology described above and consider it to be much better morally than making 10 embryos and discarding 9 of them, whether randomly or based on screening. (Although, secondarily, I also have some slight qualms about the future society-wide effects if people are allowed to fully customise their kids' genetic makeup).
Every argument in this post would suggest that sex-selective IVF and/or abortion is OK. These are typically seen as morally repugnant - and are serious criminal offenses where I live. Is Scott willing to bite that bullet?
Being pro-choice doesn't mean viewing a foetus as having no rights. It's perfectly possible to think a foetus has rights, but - for example - in the difficult situation where the baby is unwanted, the mother's rights have priority. Similarly, being pro-IVF in the tragic situation where a couple would otherwise be infertile doesn't mean being pro-IVF always and for any reason.
The strongest argument is that if you're doing IVF anyway, you may as well screen for schizophrenia. But if that is allowed, then couples who would otherwise conceive naturally will use IVF. So you're not "doing IVF anyway," in aggregate.
The whole post is full of the kind of analogical reasoning that Scott would normally reject. What happened to utilitarianism? Why isn't the question "Are we better off as a society if this is permitted?"
Another example situation: Natural fertilisation. When medical doctors explain that before fertilisation, only functional sperm reaches the egg, and then in the first week(s) of pregancy a dysfunctional foetus will often naturally be aborted, pretty much nobody ever thinks that this is a tragedy, that we should allow all those sperm to fertilize eggs no matter how dysfunctional they are. Or that we should try to develop technology to forcibly get a dysfunctional foetus to be carried to term despite the body trying to abort it.
I'm not sure of the detail of 1 . Aren't eggs all produced at the initial development of the ovary, not once a month? Given that the body must then choose ~1 each month to ovulate, it's plausible that it does so by ovulating them in some predefined order, so that the choice not to drink doesn't affect which egg is ovulated. A better argument would be that everything affects which sperm ends up fertilising the egg.
I think this dilemma is sharpest if you think about people wanting to select against deafness. Deaf people who are part of signing culture often believe that deaf culture is valuable and not merely an adaptation, and (congenital) deafness should be considered a variation not a disability. As such they would argue that selection is immoral. I don't know any deaf people so I have not thought about this much, but I guess the best argument would be that the decision to select in the first place is the immoral step, not that the step of selection is discriminating against individual deaf embryos. But I don't know how a deaf person would actually argue.
Selecting against schizophrenic embryos would be systematic, whereas Situations 1 and 3 here just randomly re-roll what baby ends up being made. Situation 2 is slightly selective - I assume the healthiest embryo is chosen to maximise odds of surviving pregnancy? If so I suppose that only makes sense, but if they're being selected to be taller/smarter/etc. as children/adults that makes me uncomfortable too.
I don't mind an action changing what baby is made so long as it's a random re-roll of the die, but if we start normalising the systematic prevention of certain types embryos that would've been babies due to their 'worse genes', where does it end? People will start selecting their children for all the high-status traits such as intelligence, height, and so on.
Some of this is clearly Molochian: when everyone defects and selects an embryo to be taller, we'll simply find that the old desired heights like 6 foot 1 simply become the new 'short' lower-status heights. Desiring greater height is about wanting to be taller than peers, not being taller absolutely. And height is correlated with shorter life-expectancy, so this purely harms everyone's children for the sake of not being left in the dust by everyone else's defection.
Selection for intelligence seems very risky: would speed up AGI development.
But more generally, if people start heavily selecting their children for all sorts of traits (perhaps even using CRISPR and the like for some genetic engineering) then - assuming we haven't gone extinct a while from now - will we not just select ourselves out of existence, into some new species? It reminds me of the thought experiment - I think it was on SSC - where Gandhi could gain a reward by taking a pill to become 99% as good, but he decides not to because then he'd be fine getting another reward for taking a pill that makes him 98% as good, and so on and so forth until he'd be 0% good; and 100% good Gandhi did not want to end up 0% good.
I would like it if humanity stays 100% human, and does not - via some situation where the Ship of Theseus problem meets the slow-boiling frog - slowly replace its genetics (and thus its phenotype) over a timespan too long for it to be bothered with or realise fully the consequences, until our successors have become some creature that is wholly alien to us.
I think this prospect unnerves me a little (not enough to oppose it straightforwardly), but I think I figured out what my unease is from.
At this point, polygenic selection is straightforwardly good. It gets a little less good when the technology improves enough to predict things like personality traits.
Wouldn't it be bad if parents had an outsized control over the inherent traits of their children? Hypothetically you have parents that want a child for less noble reasons (maybe they want a live in servant that is forever tied to them, plenty of abusive parents treat their kids like this).
Are you able to stop them from selecting an embryo that is more likely to be gullible and easy to manipulate? Are you able to stop them from picking an embryo born with a chronic condition that is treatable with medicine, so that the parents always have the leash with which to control the kid and train them into a kind of house slave?
Dipping into the realm of celebrity gossip, there's some fairly credible rumours that a supermodel, whose mother was a supermodel, was forced to get a rhinoplasty at 14 and also ended up with disordered eating from maintaining the body shape. While she has a very successful modelling career, it's pretty clear that she didn't really choose to be a model and mostly just got pressured into it. Imagine if this mother had access to genetic engineering. What's to stop wealthy obsessive supermodel types like this woman from engineering humans like pugs - attractive, with lots of health problems linked to the appearance based breeding, and then forcing them to do as the parent wishes?
Yes, these are really hypothetical problem right now. What can we do to stop parents from infringing on the rights of future people, when the technology becomes viable?
In conclusion - it's great to use this tech for chronic incurable health issues currently. I think we need to really formalise rights of children before it gets to the point where parents get full control of every aspect of unborn children, because it could get really bad.
I guess the counterpump is that once the precident is set then what other conditions can a couple choose to select for at IVF? The sex of the baby? What if they have a religion that considers schizophrenics to be prophets and so they want to maximise that chance? What if they're convinced the most important thing for humanity is to maximise genetic diversity and so they want to select the one with the most genetic mutation? What if technology develops to the point where we can predict the future personality of the baby and everyone always chooses the same option?
On individual levels, it's a moral problem but at scale some become societal problem because they distort the population.
The obvious answer is to have the state allow certain selections but not others, which is basically where we're at now. But I can see the case that this grants the government a problematic power to decide on the genetic makeup of the population, and that this is a dangerous road to wander down.
I think there is an assumption underlying these discussions that two embryos with different starting genes will grow up to be different people. While a reasonable assumption, I'm not sure this should be taken for granted. Consciousness is poorly understood, but it feels like a 'person' (in the sense of being a sentient entity) may be better defined by the atoms that form its nervous system than by its starting genes.
If the set of molecules that become a nervous system is more robust to genetic variation than we expect, then we wouldn't be creating entirely different people from the same starting embryos. We are only equipping the person with a different set of genes. Viewed this way, then it would definitely be an ethical good to choose the 'best' embryo (assuming this 'best' is well defined).
If this all sounds hand wavy, that's because it is. There are a lot of open questions, like whether what matters is if some subset of molecules form a specific part of the brain, or how many molecule difference it would take for a person to be considered 'different', how much variation there is brain development, etc. However, I think it's important to discuss given how it could drastically change our starting points.
If everyone, or even a sizeable fraction of the at-risk population, used polygenic selection to filter against schizoprenia, then within a few generations we might end up "breeding out" an unwanted trait. So arguably every individual couple who chooses that route is also lowering the prevalence of the condition at the population level by a tiny amount.
I was going to point out that scenario 1 doesn't work as a response to the argument it's being mustered against, but William H Stoddard has already done that!
Here's my problem with scenario 2:
> It’s true that embryo #5 was briefly destined to be implanted and born and grow into a human being, and that the doctor’s decision caused that not to happen. But almost nobody would consider this an injury done to embryo #5 or consider this to be impermissible meddling in the threads of Fate. Nobody would say that, once the intern had picked #5, it was wrong for the doctor to switch to #7 in the name of health.
I agree that nobody says this. But that's not because they're committed to moral consistency. It's because injuries to embryos don't matter. When similar injuries happen to adults, they are recognized as injuries.
Let's take the example of Miss Universe 2015. Voting (secretly) determined that the winner was Miss Philippines. The MC (publicly) announced that the winner was Miss Colombia. Then, shortly afterward, the MC announced that he had been mistaken the first time, and the winner was in fact Miss Philippines.
Everyone recognizes this as an injury to Miss Colombia. It's not necessarily a big injury, but the MC apologized -- and imagine what people would have thought if he hadn't! -- and faced ongoing jokes about the mishandling of the event.
Drawing an analogy to scenario 2, we see that, at minimum, embryo 5 is owed an apology by the doctor. But really, the injury being dealt to embryo 5 is 𝗳𝗮𝗿 𝗺𝗼𝗿𝗲 𝘀𝗲𝘃𝗲𝗿𝗲 than the injury that was dealt to Miss Colombia. (I hope that, even though the fact that an injury occurred at all was not immediately obvious, the fact that it is much more severe than the parallel injury to Miss Colombia is obvious!)
This is because living is more important than being named Miss Universe 2015. The difference in severity of the injury that we perceive should correspond directly to the difference in importance between being alive (vs. dead) and being named Miss Universe 2015 (vs. not).
If you believe that embryos have moral weight, then the doctor has committed a very grievous wrong against embryo #5. The reason most people do not believe that the doctor has done so is that they do not assign moral weight to embryos.
Explaining tricky issues is one of Scott's great strengths. Here, there is nothing tricky - and if anyone thought there was, the first two examples should have helped. The third one seems "ouch" to me, but I guess there were such comments. - If none of this helped, I doubt anything can.
Example 2 does not make an argument about the definition of prevention but rather the ethics of embryo selection. I think there are two important differences between your examples 1 and 3 and embryo selection.
Firstly, in your examples a different baby being born is incidental rather than crucial for the preventive effect. In Example 1, there is nothing wrong the woman's eggs, there is something wrong with her behavior, which she fixes in rehab. This reduces the risk of her baby to be ill. Similarly, in Example 3, the preventive effect works through an adjustment in behavior, not the selection of the sperm.
Secondly, in Example 1 and 3, different eggs and sperm are used to create the baby but no reasonable person considers eggs and sperm to be human beings, whereas some people do consider embryos to be human beings. Selecting on the "ingredients" for humans may be seen as prevention, whereas selecting on "humans" themselves is not. Once you already have a "human", you can no longer prevent anything, you can only cull the weak!
Let me give you two counterexamples.
1. You bake 2 cakes with two eggs. One of the eggs is rotten. You can either bake both cakes and then smash the cake with the rotten egg in it, or you can simply choose not to bake the second cake and throw the rotten egg away. I think there is a meaningful way in which we can say that the first option does not prevent a bad cake from existing, while the second does.
2. You give birth to several children. Once the youngest is old enough, you give all of them psychological tests. You kill every child with bad scores for schizophrenia. Have you prevented schizophrenia? I guess, in some sense you did. However, I can also appreciate the difference to, say, improving the children's nutrition (or whatever). Culling the weak is different from supporting the weak to get strong.
Example 2 shows that the real argument is about whether an embryo is already a person or not. If you think that it is, and that culling the weak is not really the same as prevention, embryo selection is not prevention.
A broader point on the ethics of embryo selection: Some people might like randomness more than selection when it comes to creating babies for the following reason. If you select against something, that implies that the thing you select against is BAD. This implies that people with the thing are BAD. Therefore, you should treat them as BAD people. But I think that people with thing are GOOD and should be treated as GOOD people. Therefore, selecting against them is BAD.
This type of thinking makes sense if you put things in boxes labeled either GOOD or BAD and then put everything with strong associations with the original thing also in the same box. Personally, I think this type of reasoning is BAD but common. Sometimes it can even lead to opposition to curing a disease because an identity has formed from things in the GOOD box, see e.g. the deaf, HIV positives, and transgenders.
Not that I expect this to convince anyone, but traditionally the argument that life begins at conception comes from the idea of ensoulment. My understanding of the relevant religious arguments is that personhood also flows from ensoulment rather than physical characteristics, including genes (otherwise we have much bigger problems with things like resurrection, or even just radiation). How do we know whether the person conceived was given the same soul or not?
This is only intended as half sarcastic. I really would be interested if anyone has actually thought about this aspect of the question. There is at least significant (and ancient) discussion of the idea that the Abrahamic God sometimes puts a male soul in a female body and vice versa, so it cannot be that genetics fully constrains His choice.
Disclaimer: Devil's advocate post, I entirely think that preventing schizophrenia in this way is a good thing.
This feels like it's ignoring an important point.
The *other* selection criteria are not genetic, but this one is. If we get to a point where everyone uses Bujold-esque artificial wombs, selection by stochastic processes is not going to reduce the specie's genetic variability, but selection by genetic criteria very much would.
Re: the Alabama thing, in Italy we are not allowed to dispose of two frozen embryos even though we *know* they are so badly aneuploid that if implanted they'd get miscarried within weeks -- we will just have to pay €200/year to keep them frozen until Italian lawmakers realize how ridiculous this is or an AI turns us all into paperclips, whichever is first.
Just want to note that you've got the conclusion correct from a pro-life perspective: polygenic selection isn't really any new evil over and above the standard evil that takes place in IVF.
I said to myself I'd keep off this one since it's hot button, but I had to go after this sentence:
"But almost nobody would consider this an injury done to embryo #5 or consider this to be impermissible meddling in the threads of Fate."
*raises hand* Excuse me, Teacher, but I do!
Having stuck my head up, I now skulk back into the grimy recesses of the basket of the deplorables.
Carry on, men (and women and non-binary folx).
EDIT: You know an even better, more fool-proof way of preventing schizophrenia? Voluntary (or involuntary) sterilisation! Take your example couple here: they both have strong family history of schizophrenia, they both are at risk (even if they're not schizophrenic themselves) of having offspring with schizophrenia, and indeed in the example *nine* out of the ten embryos were high-risk! Clearly the best way to prevent another generation of schizophrenia is not to allow them to reproduce. Even if the tenth embryo had been implanted, it would have grown up to still be the generation that has "strong family history of schizophrenia on both sides" and still be a risk - remember, the parents themselves were not schizophrenic but the risk was there, was expressed in the vast majority of their embryos, and you can't know for sure the tenth embryo is totally free or is just as much 'at risk' as their parents of producing schizophrenic offspring if they meet up with the wrong partner and their genetic history.
The best thing for society, for the couple themselves, and even the putative tenth embryo is simply skip the step of reproduction altogether. 100% case of schizophrenia prevention, no risk, no years of worrying about "what if something happens to our child despite all our precautions and it happens?"
I feel like Gattaca hits on the moral counter argument pretty well here. Choosing "who" is born to this extent could create an artificial and higher bar of expectations. It could untether society's expectations of the individual more than they already are.
Eventually, having a normal birth becomes "immoral" because the quality of life for a randomly born baby is that much lower than for a genetically selected baby. Not because of the baby, but because of society (e.g. the theme of prior posts on perspectives towards disability). Who is "able" enough for society? Genetic selection would take it to a whole new level.
If that happens, then what is the point of normal birth and what does it mean to be human?
When we say "life begins at conception", doesn't that mean the actual person does not exist until the sperm enters the egg? If we endow an unfertilized egg with personhood it seems we are complicating matters quite a bit.
There’s a whole philosophical literature on the “non-identity problem”. Some of it is stupid, some of it is great, but if one is interested in seeing what people have already said about this, to see if some of their ideas seem like meaningful insights: https://philpapers.org/s/non-identity%20problem
I can try to steelman the anti-screening (for whatever) argument like this: first, the pro-screening side can hypothetically respond with "tough luck, go be born to someone else", to which the anti-screening side says that, aha, but if everyone in the society does the screening, there's no parents left to be born to.
It's one thing to be one of the millions of eggs and trillions of sperms who didn't win the lottery. All your examples replace a would-be winner with another more or less random candidate, then pump the intuition that the former winner was chosen randomly anyway, and with astronomically low odds, so they don't get to complain as if they were robbed of their predestined birthright.
It's another thing when the lottery is rigged to give entire categories of embryos essentially zero chance to be born. In this case, better luck next time doesn't work, luck won't help, and that feels kind of unfair.
I personally don't approve of this way of thinking because I don't think that nonexistent humans have rights, but I suspect that most opponents are reaching for something like this intuition and you have to state and address it.
There's also people like Kevin Bird who say that saying that some traits are undesirable in embryos implies that they are undesirable in existing humans, so even if we double dog swear that we are not forcefully euthanizing existing humans, their feelings are still hurt and protecting them is worth creating lots and lots of real suffering. I disagree with that in strongest possible terms.
I personally have never understood how *so many* rationalists are strongly anti-eugenics.
Eugenics is one of the strongest "net positives" available to us, to the extent that literally *everybody* practices it strenuously, to basically the level of tech available today.
If Greg Clark's The Son Also Rises taught us anything, it's that assortative mating among elites is one of the strongest possibly optimized things that people do. For those effects to be true, assortative mating has to be one of the most important and most optimized choices people make - and indeed, if you look at the world, elite people are *extremely* selective about who they have children with.
To a first approximation, Rationalists are elites, by pay and by IQ, and by occupations. And everyone in our circles optimizes hard on the quality of their mates - what is this, but eugenics?
I don't understand why when you suddenly use some science to go a bit farther, it's suddenly verboten to most people here. You already spent *decades* trying to optimize this to the n-th degree!
The only objection that I've heard that isn't ad hominem (Nazi's did it isn't a real objection, particularly when literally everyone everywhere for all time strenuously optimizes mate quality), is that there can be unintended effects due to the complexity, or that only rich people will do it and this will lead to greater societal stratification and literal castes.
First, these are contradictory - if only rich people are doing it, the sample and the people affected is low enough that we can't really mess things up with unintended consequences.
Second, the rich ALREADY socially stratify, and are basically a caste! That's what The Son Also Rises is *about!*. It's been true for thousands of years!
And finally, there's literally no path for "regular" people to get to a place where they can polygenically select away genetic defects and select into any benefits UNLESS you go through early adopters, ie the rich.
Going to a consequentialist perspective, economic growth is the strongest lever and driver for eliminating poverty worldwide, and we will eventually be able to select on those things that will enable the elimination of poverty worldwide, ie IQ and conscientiousness.
The fact that somewhere along the way people will also be able to make their kids blonde and tall and healthy and strong and attractive as well (horrors! Aryan master race stuff!) is totally a personal choice for those parents, and is a GOOD thing.
From whatever perspective you come from, be it the individual parents' rights and choices, or from society's, or from the perspective of eliminating poverty and building a better future for the human race, polygenic selection is going to be one of the biggest levers we have to allow more choice, better societies, and less poverty. Why would we discard this tool, just because Nazi's did it in a really poor and biased way in the past?
While I have the same moral intuitions as you on the replace vs. cure question, I don't think any of these examples address the objections people brought up in your last post. Fetal alcohol syndrome and child abuse are different from schizophrenia in ways that make your side of the argument a lot easier.
First, they aren't genetic. That lets you sidestep any argument that embryo selection = eugenics, or will inevitably lead to genetics. Second, they don't serve any valuable social function, not even an evo-psych handwave in the direction of one. That gets you around the Chesterton's fence-style objections to deleting parts of the human genome.
I would suggest redoing your situation 3 experiment with, say, autism as the condition to be avoided. If a world without autism gives you a different gut reaction than a world without child abuse, that would get you a little closer to understanding the other side of this argument.
Decisions are not made in an abstract, timeless way. Every decision is made at a specific point in time, based on the information that is available at that point.
In our particular scenario, the mother-to-be does not have the choice between being pregnant with a child conceived in the past when she was drinking, and one conceived in the future when she has stopped---not unless she's already pregnant and is considering abortion! Her choice is to get pregnant in the future while drinking, and to get pregnant in the future while not drinking. And the characteristics of either fertilized ovum are unknown; whatever she might assume about one she can just as well assume about the other.
Open question: do you think increasing human intelligence (either at the mean or on the margin), through IVF for instance, increases the risk of human catastrophe (I'm thinking from AI) or decreases it, on net?
The objection is that gene-related issues are sensitive and you are supposed to be vague when talking about them. If you are clear and straightforward about it you violate the shroud of indeterminance to which they become attached. It forces their mind to come to conclusions they may not appreciate (which they project unto the breaker of the shroud) It could be considered psychic violence
I think the major objection to PGT-P is that it isn't ready for prime-time yet. We don't know the genes that cause schizophrenia and like most complex diseases, it's likely to be many genes. The data that current associations are based on are also flawed. Also, it's unlikely that there would be much variation in risk in a small cohort of embryos to be clinically significant (my understanding is that you need hundreds before you start to see huge differences). And you may be inadvertently selecting against some beneficial trait. Having said that, if the patient has been adequately counseled and is already doing IVF for infertility reasons and wants to do PGT-P, I think they should be able to. But at this point, I personally would not recommend someone do IVF just to do PGT-P.
The problem people seem to be gesturing at is a slightly more difficult one than you seem to acknowledge, although I generally agree that these “isolated demands for rigor” are a sign of bad faith.
The problem is the question of the rights of future people, which plagues many ethical theories, but most notoriously utilitatianism in the “repugnant conclusion” thought experiment.
It seems wrong to cause harm to occur to future people, and good to cause good things to occur to future generations. Doing good/preventing harm to currently living people is straightforward, and can in most circumstances be done consensually, since the people are alive to be consulted.
But we can’t ask future generations for permission to make modifications to their lives that we currently find desireable. Furthermore, there are questions of individualism that are not easy to answer. Can we do harm to specific future people to help the average future person? How do we deal with future people as individuals?
I think many people intuitively respond to this problem complex by leaning into passivity. It’s acceptable to passively jostle sperm because the outcome is random and unavoidable. It’s unacceptable to actively select for certain traits (say, controversially, lighter skin color) because this cannot be done without asking permission, which is unavoidable. It’s a sort of ethical default preference, or maybe a version of the naturalistic fallacy.
That said, no person or ethical theorist gets out of the problem of future people looking particularly good, so I find it hard to criticize these sorts of intuitions too much. I simply don’t share them.
When you say "the anti-abortionist doesn’t have much of a case here" (you start 'Even if', but I think the logic of your paragraph is that you don't think they have much of a case) I think you mean "they don't have much of a case outside of their general case against IVF"? You seem to me to have shown that they do have a case in total.
I find utilitarianism unsatisfactory on a number of different grounds, both epistemic and ethical. But the one that's most relevant here is that it supposes we can define a utility function for an individual human being; that is, that given any two options to choose between, we can assign each a utility score, and then the better choice for that human being is the one that attains the higher utility score---in the classic wording, the one that provides the "greatest happiness." And I simply don't believe that human brains work that way. I think that there are domains within which different values are compared regularly with each other, and thus are commensurable, and we can say that one is favored over the other; but I think there are also cases where we have to choose between values that fall into different domains, in which we do not have an established ranking, and do not even have a procedure for arriving at such a ranking. That is, I think human beings do not have global utility functions. And the choice between values that fall into different domains is inherently hard, and may be painful, as in the classic conundrum of "should I betray my country or betray my friend?" The kind of purely calculational ethics that utilitarianism offers excludes the possibility of tragedy.
And I think that, in the first place, one of the merits of virtue ethics is that the virtues it recommends may stand us in good stead in the face of tragedy, enabling us to make good choices when nothing approximating utilitarian calculation is any help. And in the second place, if we are to have any method for making choices that is generally applicable of helpful in such difficulties, that method most closely approximates the creation of works of art. And while there certainly can be a rational and even calculating aspect to art, the kind of synthesis that it calls for takes place on a different level than simple calculation; and the same may apply to the kind of synthesis that we perform when we try to lead an ethical life.
My perspective is as a lawyer, law professor, and grandparent of a lovely girl born as the result of implantation of a donated embryo. I'm sure that colors my understanding - in particular, I appreciate something that is not mentioned in any of the news coverage I've seen, which is the generosity of couples donating embryos to infertile couples, so that the embryos and the families of those couples can get the benefit of having children.
It should surprise no one that the media's presentation of the decision is inaccurate. The majority held that fertilized embryos are covered by a state statute making compensable injuries to children by including in the definition of a child a fertilized embryo. (The various judges, and the dissent in particular, have a lot to say about whether that is a reasonable interpretation of a statute passed in the 19th century long before IVF was possible. That's a hotly contested question and the majority and dissent are both worth reading on the issue.)
The facts of the case are that a clinic - in what appears to be at the least a negligent way -- allowed someone to wander into its storage area and destroy some embryos by removing them from the cold storage and drop them on the floor. Certainly this clinic should be liable to the parents. The court had to decide whether it would be liable for damaging the parents' property, violating its contractual obligations to the parents, or in tort for injury to a child. It went with the latter. There are good arguments (made by the various opinions) for and against the majority's decision, but the majority's interpretation of the statute isn't the result of backwoods judges doing crazy stuff. It's a plausible interpretation of the statute. (There is an odd concurrence by the Chief Justice that delves deeply into theology in search of the meaning of "sanctity of life" - that one isn't (to my ears) really a legal argument but it also isn't the majority opinion and so isn't what drove the result.)
Clinics are upset (tort damages are more than property or contract damages would be). They are hoping to get the AL (and other states') legislature to amend the law to give them protection from such suits. One good way to launch that campaign is to try to frame the decision as anti-IVF and portray themselves as the good guys helping infertile couples. This clinic, however, was not the good guy - it was (at a minimum) pretty careless in protecting the embryos from people wandering around its facility. (Door locks are not high tech! Freezer locks aren't either!)
Reasonable people can differ about whether the court got the statutory interpretation question right (and the members of the court do). But this isn't a decision that will end IVF treatments in AL by imposing an intolerable burden on clinics - and it doesn't seem reasonable to me for the clinics to have stopped IVF treatments in response to it, so long as they were behaving non-negligently in their creation, storage, and implantation of the embryos.
I think that "social policy decisions" are a different case from what's outlined in the trolley problem. In the specific case of five versus one, you can see the five people, you can see the one person, and you are deciding that that specific one person must die to spare the lives of the others. In a social policy decision, you inherently cannot identify the people who will be made worse off, or how much worse off they will be made; you are deciding between probabilities of distributing damage in various ways through a large population, with the consequences for any one person not fully knowable in advance, and the consequences for every one person in the population impossible to know in advance. I don't see that intuitions that are developed in the one situation can meaningfully generalize to the other.
Another thought is related to trolley problem reasoning. How interchangeable are lives. How much work can yiu put into casting the role of "this couple's first child" before its a different person?
The only slightly reasonable objection I have heard for not selectively aborting down syndrome-fetuses is the following: If we like diversity, someone have to pay the cost for making the world more divers, ie having kids with downs syndrome even though it is probably more work than having kids without such syndrome. (This is btw also predicated on the sort of empirically testable hypothesis that people with downs doesnt suffer a lot more than healthy people. I have heard that this is true and that they even report higher happiness in surveys. At least i societies and times that doesnt hide and/or torture them.)
Even if I grant this argument it comes out more like a coordination-problem: How do we make people committ to keeping downs-fetuses and having downs-kids? It does not make ME want to parent a child with downs syndrome myself. (I have three healthy kids. Maybe one with downs is better than zero.)
Then again, the norm that we shouldnt selectively abort children can be seen as exactly the coordination-function we would need to get more diversity, to the cost of the parents who get the "diverse" kids.
in defensible cases, its preventing a real risk of being unable to care for a severely disabled child.
in indefensible cases its extending the parent's ability to shape the child to follow their idea of the good life. "My son must be a 190 IQ STEM master of the universe, so i will select for that!" The parent always has tension between instilling values and letting the child live their own life; if trait selection ever gets to that point the scale will tip to one side.
even if it doesn't work. one of the worst things is the expectation to be smart. or to live up to someone else's lifestyle. imagine if dad selects you for iq to try snd nudge you to be the brilliant scientist he wants and...you aren't. to know he just didnt want to raise you, he tried to build you; that's one huge burden.
I think talking about this as though it's a given is incorrect. As Scott mentioned - it's about low risk and high risk. So it's not a matter of choosing schizophrenia or not choosing it. It's a matter of how much you want.
In case 1, I don't think you're making the relevant comparison. The question is not of the child the woman would have had before she stopped drinking versus the child she has after she stopped drinking; it's of the child she has after she stopped drinking versus how that same child might have turned out had she not stopped drinking. We're not comparing past and present, but two different possible presents.
I honestly do not understand how anyone can object to this. Sure, there are your religious pro-life people with the "god maybe wanted you to have that schizophrenic baby," given their complete lack of empathy for women carrying issue doomed to die shortly after birth, but preventing human suffering, if possible, is a net good.
You say, about the anti-abortion position:
But even this isn’t an argument against polygenic selection. It’s an argument against IVF in general, which usually involves production of more embryos than the couple intend to bring to term.
I think that there is such an argument, if you accept the premise. Polygenic selection makes IVF more popular, so more people will do it. This is the reason that we make it illegal to pay for murder — it makes murder more popular. You still have to accept the non-selection is homicide premise to get to that conclusion, but the structure of your argument was that you were momentarily accepting it for the sake of your argument.
> It’s an argument against IVF in general, which usually involves production of more embryos than the couple intend to bring to term.
It's also more crisply totally incompatible with PGT-M. This line of argument doesn't allow discarding an embryo which inherited Huntington's Disease, where a person could live a couple healthy decades first.
I don't think the "life begins at conception" people honestly have thought through these cases carefully so I don't want to say it's a reasoned argument against PGT-M, but the two can't really coexist.
I must be way out of the loop, because I cannot imagine someone who is not opposed to IVF based on “life begins at conception” grounds but is opposed to embryo selection because people should be born with diseases (and it’s somehow bad to prevent that). I don’t know anyone like this and my theory of mind is also failing me here
My old anatomy textbook from college mentioned that a fairly large percentage of fertilized embryos pass without implanting. (Just looked it up again and Google says 10%-40%.) I think it you're asking the question, "when does a woman have a moral obligation to a fertilized egg?" it doesn't make sense to set the bar before implantation. Otherwise, you'd need to worry about embryos that naturally fail to implant. But you'd never even know about them anyway, most of the time.
i think schizophrenia and autism, two highly heritable conditions, are going to be targeted first by polygenic selection (OK, real talk, thre are ppl who have already done this...)
Polygenic selection seems like a great idea, but I do have a question on it (unrelated to this post, which is about an argument I find ridiculous).
If widespread polygenic selection leads to selecting embryos that match the comparatively few characteristics we think are good, isn't it quite possible to face unintended negative consequences from lack of genetic diversity? In a book I read, 'Seeing like a State', European foresters thought it was a great idea to replace wild forest with monoculture plantations of trees they thought were good, but in retrospect they didn't have a good enough understanding of forestry science and it turned out to be a mistake.
Are you familiar with the non-identity problem, in philosophy? That’s relevant to this discussion. See the following for more:
https://plato.stanford.edu/ENTRIES/nonidentity-problem/
https://www.amazon.com/Non-Identity-Problem-Ethics-Future-People/dp/0199682933/
You say: "But once you’re already doing IVF, selecting the embryos based on some criterion, like low schizophrenia risk, doesn’t make this issue any worse."
The criterion in the schizophrenia case is selecting for non-schizophrenic genes, whereas in the other cases it was more or less random, with respect to genes. . In the schizophrenia case, it might help to eliminate the schizophrenia genes from the gene pool. But some people might still object to this. If they were able to select for alcoholic genes and eliminate those, some might argue that is bad because alcoholics might tend to be more creative than non alcoholic. (For example, a lot of writers are alcoholics.)
Or if you could select for gay genes, some might say we shouldn't be trying to eliminate gays.
The schizophrenia complaint at the beginning has a much simpler rejoinder - by selecting any of the embryos, they are necessarily not selecting 9 others. In this framing, all possible selections, regardless of schizophrenia genes, have equal moral weight. (Whether one thinks this is a positive or moral weight is beside the point; the point is that they are all equal)
The correct answer here is "no one", since it was never proven to be an effective practice in the first place.
Other than that, congratulations for rediscovering the arguments of the late-90ies "is-IVF-morally-okay" debate, followed closely by the somewhat less dumb early-2000s "is-PIGD-murder" debate. If you keep this up, we'll have an "eugenics-will-save-us-all" post in a few months, followed by a "are-we-the-bad-guys" realisation in late summer.
Pardon the sarcasm, but if you've dug yourself into a hole, stop digging. Every respectable medical and scientific society spoke out against this previously, and the data are not coming in to change that.
It was a real comment, not pedantry, so thanks for engaging! I think another natural parallel is abortion for the purpose of sex selection, which many people (including pro-choice people) find very off-putting but is very common globally. We certainly don't call it "preventing femaleness". Specific to IVF, some countries (Germany) knowing the sex of your IVF embryos is illegal to avoid sex selection via IVF.
Fwiw I do feel there is a spectrum, and that while selecting embryos to avoid conditions like "female baby" or "probably shorter than my other embryos" does seem immoral to me, I would never advise a woman to transfer an embryo with (say) Edwards syndrome. Schizophrenia is in between those two.
Eugenics: fine on an individual free-choice level, bad on a societally imposed level. Does this seem like common sense? Is it a strawman of Scott's position?
Is it just mood affiliation/fear of the slippery slope to eugenics? (a bullet I'm happy to bite: why would you not want to give your child the best genes possible, regardless of the historical connotation)
People make the argument against polygenic selection because it forces them to think about the influence of genes on various life outcomes. This in turn forces them to acknowledge that not everyone is born equal. But acknowledging this will get you cancelled in certain circles, so you have to try and construct an informal taboo against anything connected to the subject.
As a firm "life begins at conception" pro-lifer, I think Scott conflates two questions in his discussion of our perspective. Namely, "Does polygenic screening cause additional harm?" and "Does polygenic screening prevent schizophrenia?"
It's fair (if horrifying) to say that if you're already killing 9 out of 10 embryonic human beings, there's no additional harm in making sure that you kill the ones with various undesirable conditions. But this doesn't translate into it being fair to say that this process "prevents schizophrenia". You're just killing the schizophrenics.
Suppose a Nazi death camp has a doctor assigned to select certain quota of Jews to be spared the gas chamber and put to work in a labor camp instead. And suppose he makes sure to pick the strongest and healthiest ones without conditions like schizophrenia. The Nazi doctor is not, strictly speaking, making the situation worse. But he's also not "preventing schizophrenia".
I don't think you can make any strong arguments in this granular case. It would be most fair to evaluate the rule and think of more refinements, genetic therapies, and other items that dip more and more into modifications, and on a larger scale, and perhaps less voluntary as far as buy in goes, or perhaps expected.
One consideration I think this leaves out: it can be a heavier burden on both parents and children when parents are more the “authors” of their children through trait selection/abortion rather than the “receivers” of their child, open to uncertainty.
The Atlantic had a good piece on Down Syndrome screening (https://www.theatlantic.com/magazine/archive/2020/12/the-last-children-of-down-syndrome/616928/)
“The introduction of a choice reshapes the terrain on which we all stand. To opt out of testing is to become someone who chose to opt out. To test and end a pregnancy because of Down syndrome is to become someone who chose not to have a child with a disability. To test and continue the pregnancy after a Down syndrome diagnosis is to become someone who chose to have a child with a disability. Each choice puts you behind one demarcating line or another. There is no neutral ground, except perhaps in hoping that the test comes back negative and you never have to choose what’s next.”
Can you save someone from a traffic accident? After all, the person you saved was by definition not a victim in the traffic accident. You merely preferred to go down the branch in which the person was unharmed rather than the world branch you would share with the traffic accident victim, you terrible ableist.
--
I think conscious beings should get special treatment here. Their preferences regarding if and how they would continue to exist should be respected (within reason). They should not be euthanized and replaced with other conscious beings, unless that is what they want. Thus we do not have to debate if healing a mental or physical illness will just replace an ill person with a healthier (plus or minus scare quotes) one, simply follow the patient's preference.
(There might be some exemptions for young children and psychiatric patients reduced to the cognitive capabilities of young children. Basically, they get to have preferences, but sometimes their preferences may be overridden because of their cognitive limitations.)
So saving the traffic victim is the right choice, because presumably that is what their preferences are. By contrast, plants, insects and embryos do not get that protection, they are morally interchangeable. (At least until you add time travel, traveling into the past and preventing your enemy from being conceived by gifting their parents two tickets for the opera is probably evil.)
> But even this isn’t an argument against polygenic selection. It’s an argument against IVF in general, which usually involves production of more embryos than the couple intend to bring to term... But once you’re already doing IVF, selecting the embryos based on some criterion, like low schizophrenia risk, doesn’t make this issue any worse.
I think that depends on for what reason one chooses to do the IVF. If the purpose of the IVF is to solely select embryos based on some (whatever) criterion (e.g. eliminate schizophrenia), then polygenic selection in this case inherit the moral argument against IVF as well hence the anti-abortionist do have a case here.
I think the strongest objection may be from people who are, not necessarily anti-abortion or anti-IVF, but pro-schizophrenia. When I asked my schizophrenic friend what he thought about this situation he said, "I think we're doomed as a society if people start doing that."
His argument is essentially (a) that schizophrenic people have access to perspectives and abilities that others don't have and that those things are valuable, and (b) that society is narrowing down so that only a very narrow set of characteristics can actually thrive in it, and that THAT is a bad thing, and this is both symptom and cause of the acceleration of that process. Schizophrenia outcomes are much better in societies that don't stigmatize it, so from the perspective of someone who sees value in being a way that not everybody is, I think the answer to who polygenic selection helps is "nobody, but it makes a lot of people think they've been helped, and that's a very bad thing."
I don't know if I agree with him or not overall but I have sympathy for that perspective. People without schizophrenia probably wouldn't have the same thoughts; in which case, it's exactly an example of the kind of perspective that schizophrenic people bring into society that others would not have access to without those people's existence. What do we lose when we select those people out on purpose? What if we don't even know?
I also grant that schizophrenia is a special case to him and by extension to me, by virtue of discussion - I have no idea what he'd think about fetal alcohol syndrome. I should ask...
When some people feel a vague distaste to something new they sometimes make up a line of moralistic-sounding bullshit to make this distaste count in an argument. Arguing against pre-implantation embryo selection on the grounds that it does not help an individual is just that, and it is not grounded in legitimate moral reasoning. Scott completely demolished this argument but I'd like add another nail in the coffin:
An individual human is in part shaped but not wholly defined by his genes. We are spiritual beings first and foremost, defined by our memories, desires, social relationships and other indexical features that are created as the brain develops. An embryo is not a person, it is merely a molecular machine that runs a program to create the material vessel in which our spirit gradually grows. "Scott's first baby" is a person, nestled in his unique place in the world, with a unique relationship with his parents. Substituting another molecular machine before Scott's baby was delivered into his unique place would not make it another person. It would be still Scott's first baby, just with a different set of genes.
This way of thinking might feel counterintuitive at first but it actually flows directly from realizing that the individual human existence does not start with some arbitrary molecular event, like the formation of a zygote and a new diploid set of genes. Instead, the human individual grows into being gradually as a neural network that a bootstraps itself into consciousness, many months later. If you believe that this neural stage of development is the true beginning then changing genes at an earlier stage is no longer special, it has the same ontological status as all the other events that impact future humans, be it prenatal vitamins or alcohol abuse counseling of parents.
It is therefore right and proper that we give future individuals the best sets of genes we can, so they grow into more perfect beings.
I think attacking any of this from the angle of where life begins and what can be constituted as "replacing one child with the other" is entirely the wrong model to think about this. All of this - fetal alcohol syndrome prevention, IVF, etc - *feels wrong* to some people because it's artificial, civilized, technological interference with a process of childbirth, which *feels* natural. It feels like it should be a beautiful and chaotic and unpredictable leap-of-faith process, and if you got a schizophrenic child, well, sucks ass, that's fate.
It's the same logic that drives the discussion about "GMO foods" and "would you like tomatoes with no genes and DNA in them".
(for the record, I do not espouse any of these views, but I'm confident people do)
The big difference between polygenic selection and your three examples is that of expressing a very specific judgement. Example 3 involves no choice, examples 1 or 2 just say "I want my kid to turn out better rather than worse." But polygenic selection very specifically implies "schizophrenia sucks, my kid shouldn't have it". And I think that activates a strong anti-judging instinct in many people's heads.
Logically it is perfectly consistent to say "I'd prefer a world with no schizophrenic people, but the schizophrenic people that already exist should be treated with compassion and understanding". But thinking that is hard unless you're a super high-decoupler. For many people, hearing "my kid shouldn't have schizophrenia" subconsciously implies "I think schizophrenics are Evil" - and so they feel icky about it.
(Also, this strongly reminds me of the debate around selective abortion in cases of a prenatal Down Syndrome diagnosis. The feelings involved seem the same, even though no IVF is involved.)
Most of the arguments I've seen around Polygenic Selection center around second-order effects—they're slippery slope arguments, or Fence arguments, saying that if we start selecting against things that are debatably bad (like schizophrenia), then we're going to start selecting against things that are very clearly mixed (like mild autism), then we're going to see pressure to start selecting FOR things that are very clearly bad for the individual but good for society (like obedience to authority). How do you rate these second order effects?
I have no moral or other objections to screening embryos for general embryo health, or for the chance that they will grow up to be schizophrenic; implanting the ones with the least risk of the relevant disease; and discarding the rest. But I don't think it's quite accurate to call what you're doing preventing schizophrenia.
Let's say somebody sets up a clinic in some lawless part of the world, where they will test a couple's baby for genes strongly associated with schizophrenia and euthanize the infants found to be at high risk. In this lawless place there are also many people willing to sell their newborns, so the clinic is also able to provide the couple with a new infant, certified to be low risk. Would we call that preventing schizophrenia? I don't think so, and I am really talking about linguistic matters here, not moral ones. Seems to me the normal way to describe what's happened is that a schizophrenic-to-be was denied a chance to grow up, because their growing up included the very unfortunate feature of their becoming schizophrenic. Discarding embryos with high risk would be described the same way.
Or here’s another thought experiment: Let’s say we euthanized all people at age 65. That would undoubtedly reduce the number of deaths from cancer, heart disease, etc. It would reduce the absolute number, but also the fraction of the population that dies of one of these diseases. But you wouldn’t describe that approach as preventing a bunch of deaths from cancer and heart disease, would you?
Maybe I haven’t fully grasped what the point of your argument is. But it seems to me like a sort of convoluted effort to find a non-inflammatory and in fact positive way of describing the process of discarding embryos with high genetic risk of growing up to be schizophrenics. I don’t think it is going to work to change any public attitudes. I think you have to fall back on saying that embryos are little more than packets of genetic info. They’re not alive, not conscious, not people. Consequently, it is sensible and not evil to discard the defective ones.
Am I right in thinking we don't currently have the technology to harvest some eggs and sperm from a couple, genetically screen the eggs and sperm individually, and then make a single IVF embryo from a "winning" pair?
If we did, I'm curious to see how that would change people's intuitions and moral calculus. I expect those who believe that life begins at conception and that screening and discarding embryos is a bit like screening and killing babies would be OK with this alternative, and those who object to schizophrenia screening on disability-rights grounds would not, and those who object to it on the grounds of playing God or meddling with fate also would not.
I am vaguely in the first camp (although not militantly, and I acknowledge it can be the lesser of two evils), and I would be basically OK with the technology described above and consider it to be much better morally than making 10 embryos and discarding 9 of them, whether randomly or based on screening. (Although, secondarily, I also have some slight qualms about the future society-wide effects if people are allowed to fully customise their kids' genetic makeup).
Every argument in this post would suggest that sex-selective IVF and/or abortion is OK. These are typically seen as morally repugnant - and are serious criminal offenses where I live. Is Scott willing to bite that bullet?
Being pro-choice doesn't mean viewing a foetus as having no rights. It's perfectly possible to think a foetus has rights, but - for example - in the difficult situation where the baby is unwanted, the mother's rights have priority. Similarly, being pro-IVF in the tragic situation where a couple would otherwise be infertile doesn't mean being pro-IVF always and for any reason.
The strongest argument is that if you're doing IVF anyway, you may as well screen for schizophrenia. But if that is allowed, then couples who would otherwise conceive naturally will use IVF. So you're not "doing IVF anyway," in aggregate.
The whole post is full of the kind of analogical reasoning that Scott would normally reject. What happened to utilitarianism? Why isn't the question "Are we better off as a society if this is permitted?"
Another example situation: Natural fertilisation. When medical doctors explain that before fertilisation, only functional sperm reaches the egg, and then in the first week(s) of pregancy a dysfunctional foetus will often naturally be aborted, pretty much nobody ever thinks that this is a tragedy, that we should allow all those sperm to fertilize eggs no matter how dysfunctional they are. Or that we should try to develop technology to forcibly get a dysfunctional foetus to be carried to term despite the body trying to abort it.
I'm not sure of the detail of 1 . Aren't eggs all produced at the initial development of the ovary, not once a month? Given that the body must then choose ~1 each month to ovulate, it's plausible that it does so by ovulating them in some predefined order, so that the choice not to drink doesn't affect which egg is ovulated. A better argument would be that everything affects which sperm ends up fertilising the egg.
I think this dilemma is sharpest if you think about people wanting to select against deafness. Deaf people who are part of signing culture often believe that deaf culture is valuable and not merely an adaptation, and (congenital) deafness should be considered a variation not a disability. As such they would argue that selection is immoral. I don't know any deaf people so I have not thought about this much, but I guess the best argument would be that the decision to select in the first place is the immoral step, not that the step of selection is discriminating against individual deaf embryos. But I don't know how a deaf person would actually argue.
Selecting against schizophrenic embryos would be systematic, whereas Situations 1 and 3 here just randomly re-roll what baby ends up being made. Situation 2 is slightly selective - I assume the healthiest embryo is chosen to maximise odds of surviving pregnancy? If so I suppose that only makes sense, but if they're being selected to be taller/smarter/etc. as children/adults that makes me uncomfortable too.
I don't mind an action changing what baby is made so long as it's a random re-roll of the die, but if we start normalising the systematic prevention of certain types embryos that would've been babies due to their 'worse genes', where does it end? People will start selecting their children for all the high-status traits such as intelligence, height, and so on.
Some of this is clearly Molochian: when everyone defects and selects an embryo to be taller, we'll simply find that the old desired heights like 6 foot 1 simply become the new 'short' lower-status heights. Desiring greater height is about wanting to be taller than peers, not being taller absolutely. And height is correlated with shorter life-expectancy, so this purely harms everyone's children for the sake of not being left in the dust by everyone else's defection.
Selection for intelligence seems very risky: would speed up AGI development.
But more generally, if people start heavily selecting their children for all sorts of traits (perhaps even using CRISPR and the like for some genetic engineering) then - assuming we haven't gone extinct a while from now - will we not just select ourselves out of existence, into some new species? It reminds me of the thought experiment - I think it was on SSC - where Gandhi could gain a reward by taking a pill to become 99% as good, but he decides not to because then he'd be fine getting another reward for taking a pill that makes him 98% as good, and so on and so forth until he'd be 0% good; and 100% good Gandhi did not want to end up 0% good.
I would like it if humanity stays 100% human, and does not - via some situation where the Ship of Theseus problem meets the slow-boiling frog - slowly replace its genetics (and thus its phenotype) over a timespan too long for it to be bothered with or realise fully the consequences, until our successors have become some creature that is wholly alien to us.
I think this prospect unnerves me a little (not enough to oppose it straightforwardly), but I think I figured out what my unease is from.
At this point, polygenic selection is straightforwardly good. It gets a little less good when the technology improves enough to predict things like personality traits.
Wouldn't it be bad if parents had an outsized control over the inherent traits of their children? Hypothetically you have parents that want a child for less noble reasons (maybe they want a live in servant that is forever tied to them, plenty of abusive parents treat their kids like this).
Are you able to stop them from selecting an embryo that is more likely to be gullible and easy to manipulate? Are you able to stop them from picking an embryo born with a chronic condition that is treatable with medicine, so that the parents always have the leash with which to control the kid and train them into a kind of house slave?
Dipping into the realm of celebrity gossip, there's some fairly credible rumours that a supermodel, whose mother was a supermodel, was forced to get a rhinoplasty at 14 and also ended up with disordered eating from maintaining the body shape. While she has a very successful modelling career, it's pretty clear that she didn't really choose to be a model and mostly just got pressured into it. Imagine if this mother had access to genetic engineering. What's to stop wealthy obsessive supermodel types like this woman from engineering humans like pugs - attractive, with lots of health problems linked to the appearance based breeding, and then forcing them to do as the parent wishes?
Yes, these are really hypothetical problem right now. What can we do to stop parents from infringing on the rights of future people, when the technology becomes viable?
In conclusion - it's great to use this tech for chronic incurable health issues currently. I think we need to really formalise rights of children before it gets to the point where parents get full control of every aspect of unborn children, because it could get really bad.
I guess the counterpump is that once the precident is set then what other conditions can a couple choose to select for at IVF? The sex of the baby? What if they have a religion that considers schizophrenics to be prophets and so they want to maximise that chance? What if they're convinced the most important thing for humanity is to maximise genetic diversity and so they want to select the one with the most genetic mutation? What if technology develops to the point where we can predict the future personality of the baby and everyone always chooses the same option?
On individual levels, it's a moral problem but at scale some become societal problem because they distort the population.
The obvious answer is to have the state allow certain selections but not others, which is basically where we're at now. But I can see the case that this grants the government a problematic power to decide on the genetic makeup of the population, and that this is a dangerous road to wander down.
I think there is an assumption underlying these discussions that two embryos with different starting genes will grow up to be different people. While a reasonable assumption, I'm not sure this should be taken for granted. Consciousness is poorly understood, but it feels like a 'person' (in the sense of being a sentient entity) may be better defined by the atoms that form its nervous system than by its starting genes.
If the set of molecules that become a nervous system is more robust to genetic variation than we expect, then we wouldn't be creating entirely different people from the same starting embryos. We are only equipping the person with a different set of genes. Viewed this way, then it would definitely be an ethical good to choose the 'best' embryo (assuming this 'best' is well defined).
If this all sounds hand wavy, that's because it is. There are a lot of open questions, like whether what matters is if some subset of molecules form a specific part of the brain, or how many molecule difference it would take for a person to be considered 'different', how much variation there is brain development, etc. However, I think it's important to discuss given how it could drastically change our starting points.
If everyone, or even a sizeable fraction of the at-risk population, used polygenic selection to filter against schizoprenia, then within a few generations we might end up "breeding out" an unwanted trait. So arguably every individual couple who chooses that route is also lowering the prevalence of the condition at the population level by a tiny amount.
> An Alabama court made this argument on anti-abortion grounds recently.
I think you have the causation wrong. Instead, both the Alabama decision and anti-abortion are based on life-is-sacred-and-begins-at-conception.
I was going to point out that scenario 1 doesn't work as a response to the argument it's being mustered against, but William H Stoddard has already done that!
Here's my problem with scenario 2:
> It’s true that embryo #5 was briefly destined to be implanted and born and grow into a human being, and that the doctor’s decision caused that not to happen. But almost nobody would consider this an injury done to embryo #5 or consider this to be impermissible meddling in the threads of Fate. Nobody would say that, once the intern had picked #5, it was wrong for the doctor to switch to #7 in the name of health.
I agree that nobody says this. But that's not because they're committed to moral consistency. It's because injuries to embryos don't matter. When similar injuries happen to adults, they are recognized as injuries.
Let's take the example of Miss Universe 2015. Voting (secretly) determined that the winner was Miss Philippines. The MC (publicly) announced that the winner was Miss Colombia. Then, shortly afterward, the MC announced that he had been mistaken the first time, and the winner was in fact Miss Philippines.
Everyone recognizes this as an injury to Miss Colombia. It's not necessarily a big injury, but the MC apologized -- and imagine what people would have thought if he hadn't! -- and faced ongoing jokes about the mishandling of the event.
Drawing an analogy to scenario 2, we see that, at minimum, embryo 5 is owed an apology by the doctor. But really, the injury being dealt to embryo 5 is 𝗳𝗮𝗿 𝗺𝗼𝗿𝗲 𝘀𝗲𝘃𝗲𝗿𝗲 than the injury that was dealt to Miss Colombia. (I hope that, even though the fact that an injury occurred at all was not immediately obvious, the fact that it is much more severe than the parallel injury to Miss Colombia is obvious!)
This is because living is more important than being named Miss Universe 2015. The difference in severity of the injury that we perceive should correspond directly to the difference in importance between being alive (vs. dead) and being named Miss Universe 2015 (vs. not).
If you believe that embryos have moral weight, then the doctor has committed a very grievous wrong against embryo #5. The reason most people do not believe that the doctor has done so is that they do not assign moral weight to embryos.
Explaining tricky issues is one of Scott's great strengths. Here, there is nothing tricky - and if anyone thought there was, the first two examples should have helped. The third one seems "ouch" to me, but I guess there were such comments. - If none of this helped, I doubt anything can.
Example 2 does not make an argument about the definition of prevention but rather the ethics of embryo selection. I think there are two important differences between your examples 1 and 3 and embryo selection.
Firstly, in your examples a different baby being born is incidental rather than crucial for the preventive effect. In Example 1, there is nothing wrong the woman's eggs, there is something wrong with her behavior, which she fixes in rehab. This reduces the risk of her baby to be ill. Similarly, in Example 3, the preventive effect works through an adjustment in behavior, not the selection of the sperm.
Secondly, in Example 1 and 3, different eggs and sperm are used to create the baby but no reasonable person considers eggs and sperm to be human beings, whereas some people do consider embryos to be human beings. Selecting on the "ingredients" for humans may be seen as prevention, whereas selecting on "humans" themselves is not. Once you already have a "human", you can no longer prevent anything, you can only cull the weak!
Let me give you two counterexamples.
1. You bake 2 cakes with two eggs. One of the eggs is rotten. You can either bake both cakes and then smash the cake with the rotten egg in it, or you can simply choose not to bake the second cake and throw the rotten egg away. I think there is a meaningful way in which we can say that the first option does not prevent a bad cake from existing, while the second does.
2. You give birth to several children. Once the youngest is old enough, you give all of them psychological tests. You kill every child with bad scores for schizophrenia. Have you prevented schizophrenia? I guess, in some sense you did. However, I can also appreciate the difference to, say, improving the children's nutrition (or whatever). Culling the weak is different from supporting the weak to get strong.
Example 2 shows that the real argument is about whether an embryo is already a person or not. If you think that it is, and that culling the weak is not really the same as prevention, embryo selection is not prevention.
A broader point on the ethics of embryo selection: Some people might like randomness more than selection when it comes to creating babies for the following reason. If you select against something, that implies that the thing you select against is BAD. This implies that people with the thing are BAD. Therefore, you should treat them as BAD people. But I think that people with thing are GOOD and should be treated as GOOD people. Therefore, selecting against them is BAD.
This type of thinking makes sense if you put things in boxes labeled either GOOD or BAD and then put everything with strong associations with the original thing also in the same box. Personally, I think this type of reasoning is BAD but common. Sometimes it can even lead to opposition to curing a disease because an identity has formed from things in the GOOD box, see e.g. the deaf, HIV positives, and transgenders.
Not that I expect this to convince anyone, but traditionally the argument that life begins at conception comes from the idea of ensoulment. My understanding of the relevant religious arguments is that personhood also flows from ensoulment rather than physical characteristics, including genes (otherwise we have much bigger problems with things like resurrection, or even just radiation). How do we know whether the person conceived was given the same soul or not?
This is only intended as half sarcastic. I really would be interested if anyone has actually thought about this aspect of the question. There is at least significant (and ancient) discussion of the idea that the Abrahamic God sometimes puts a male soul in a female body and vice versa, so it cannot be that genetics fully constrains His choice.
Disclaimer: Devil's advocate post, I entirely think that preventing schizophrenia in this way is a good thing.
This feels like it's ignoring an important point.
The *other* selection criteria are not genetic, but this one is. If we get to a point where everyone uses Bujold-esque artificial wombs, selection by stochastic processes is not going to reduce the specie's genetic variability, but selection by genetic criteria very much would.
Re: the Alabama thing, in Italy we are not allowed to dispose of two frozen embryos even though we *know* they are so badly aneuploid that if implanted they'd get miscarried within weeks -- we will just have to pay €200/year to keep them frozen until Italian lawmakers realize how ridiculous this is or an AI turns us all into paperclips, whichever is first.
Just want to note that you've got the conclusion correct from a pro-life perspective: polygenic selection isn't really any new evil over and above the standard evil that takes place in IVF.
I said to myself I'd keep off this one since it's hot button, but I had to go after this sentence:
"But almost nobody would consider this an injury done to embryo #5 or consider this to be impermissible meddling in the threads of Fate."
*raises hand* Excuse me, Teacher, but I do!
Having stuck my head up, I now skulk back into the grimy recesses of the basket of the deplorables.
Carry on, men (and women and non-binary folx).
EDIT: You know an even better, more fool-proof way of preventing schizophrenia? Voluntary (or involuntary) sterilisation! Take your example couple here: they both have strong family history of schizophrenia, they both are at risk (even if they're not schizophrenic themselves) of having offspring with schizophrenia, and indeed in the example *nine* out of the ten embryos were high-risk! Clearly the best way to prevent another generation of schizophrenia is not to allow them to reproduce. Even if the tenth embryo had been implanted, it would have grown up to still be the generation that has "strong family history of schizophrenia on both sides" and still be a risk - remember, the parents themselves were not schizophrenic but the risk was there, was expressed in the vast majority of their embryos, and you can't know for sure the tenth embryo is totally free or is just as much 'at risk' as their parents of producing schizophrenic offspring if they meet up with the wrong partner and their genetic history.
The best thing for society, for the couple themselves, and even the putative tenth embryo is simply skip the step of reproduction altogether. 100% case of schizophrenia prevention, no risk, no years of worrying about "what if something happens to our child despite all our precautions and it happens?"
I feel like Gattaca hits on the moral counter argument pretty well here. Choosing "who" is born to this extent could create an artificial and higher bar of expectations. It could untether society's expectations of the individual more than they already are.
Eventually, having a normal birth becomes "immoral" because the quality of life for a randomly born baby is that much lower than for a genetically selected baby. Not because of the baby, but because of society (e.g. the theme of prior posts on perspectives towards disability). Who is "able" enough for society? Genetic selection would take it to a whole new level.
If that happens, then what is the point of normal birth and what does it mean to be human?
https://en.wikipedia.org/wiki/Gattaca
When we say "life begins at conception", doesn't that mean the actual person does not exist until the sperm enters the egg? If we endow an unfertilized egg with personhood it seems we are complicating matters quite a bit.
There’s a whole philosophical literature on the “non-identity problem”. Some of it is stupid, some of it is great, but if one is interested in seeing what people have already said about this, to see if some of their ideas seem like meaningful insights: https://philpapers.org/s/non-identity%20problem
I can try to steelman the anti-screening (for whatever) argument like this: first, the pro-screening side can hypothetically respond with "tough luck, go be born to someone else", to which the anti-screening side says that, aha, but if everyone in the society does the screening, there's no parents left to be born to.
It's one thing to be one of the millions of eggs and trillions of sperms who didn't win the lottery. All your examples replace a would-be winner with another more or less random candidate, then pump the intuition that the former winner was chosen randomly anyway, and with astronomically low odds, so they don't get to complain as if they were robbed of their predestined birthright.
It's another thing when the lottery is rigged to give entire categories of embryos essentially zero chance to be born. In this case, better luck next time doesn't work, luck won't help, and that feels kind of unfair.
I personally don't approve of this way of thinking because I don't think that nonexistent humans have rights, but I suspect that most opponents are reaching for something like this intuition and you have to state and address it.
There's also people like Kevin Bird who say that saying that some traits are undesirable in embryos implies that they are undesirable in existing humans, so even if we double dog swear that we are not forcefully euthanizing existing humans, their feelings are still hurt and protecting them is worth creating lots and lots of real suffering. I disagree with that in strongest possible terms.
I personally have never understood how *so many* rationalists are strongly anti-eugenics.
Eugenics is one of the strongest "net positives" available to us, to the extent that literally *everybody* practices it strenuously, to basically the level of tech available today.
If Greg Clark's The Son Also Rises taught us anything, it's that assortative mating among elites is one of the strongest possibly optimized things that people do. For those effects to be true, assortative mating has to be one of the most important and most optimized choices people make - and indeed, if you look at the world, elite people are *extremely* selective about who they have children with.
To a first approximation, Rationalists are elites, by pay and by IQ, and by occupations. And everyone in our circles optimizes hard on the quality of their mates - what is this, but eugenics?
I don't understand why when you suddenly use some science to go a bit farther, it's suddenly verboten to most people here. You already spent *decades* trying to optimize this to the n-th degree!
The only objection that I've heard that isn't ad hominem (Nazi's did it isn't a real objection, particularly when literally everyone everywhere for all time strenuously optimizes mate quality), is that there can be unintended effects due to the complexity, or that only rich people will do it and this will lead to greater societal stratification and literal castes.
First, these are contradictory - if only rich people are doing it, the sample and the people affected is low enough that we can't really mess things up with unintended consequences.
Second, the rich ALREADY socially stratify, and are basically a caste! That's what The Son Also Rises is *about!*. It's been true for thousands of years!
And finally, there's literally no path for "regular" people to get to a place where they can polygenically select away genetic defects and select into any benefits UNLESS you go through early adopters, ie the rich.
Going to a consequentialist perspective, economic growth is the strongest lever and driver for eliminating poverty worldwide, and we will eventually be able to select on those things that will enable the elimination of poverty worldwide, ie IQ and conscientiousness.
The fact that somewhere along the way people will also be able to make their kids blonde and tall and healthy and strong and attractive as well (horrors! Aryan master race stuff!) is totally a personal choice for those parents, and is a GOOD thing.
From whatever perspective you come from, be it the individual parents' rights and choices, or from society's, or from the perspective of eliminating poverty and building a better future for the human race, polygenic selection is going to be one of the biggest levers we have to allow more choice, better societies, and less poverty. Why would we discard this tool, just because Nazi's did it in a really poor and biased way in the past?
While I have the same moral intuitions as you on the replace vs. cure question, I don't think any of these examples address the objections people brought up in your last post. Fetal alcohol syndrome and child abuse are different from schizophrenia in ways that make your side of the argument a lot easier.
First, they aren't genetic. That lets you sidestep any argument that embryo selection = eugenics, or will inevitably lead to genetics. Second, they don't serve any valuable social function, not even an evo-psych handwave in the direction of one. That gets you around the Chesterton's fence-style objections to deleting parts of the human genome.
I would suggest redoing your situation 3 experiment with, say, autism as the condition to be avoided. If a world without autism gives you a different gut reaction than a world without child abuse, that would get you a little closer to understanding the other side of this argument.
Decisions are not made in an abstract, timeless way. Every decision is made at a specific point in time, based on the information that is available at that point.
In our particular scenario, the mother-to-be does not have the choice between being pregnant with a child conceived in the past when she was drinking, and one conceived in the future when she has stopped---not unless she's already pregnant and is considering abortion! Her choice is to get pregnant in the future while drinking, and to get pregnant in the future while not drinking. And the characteristics of either fertilized ovum are unknown; whatever she might assume about one she can just as well assume about the other.
Open question: do you think increasing human intelligence (either at the mean or on the margin), through IVF for instance, increases the risk of human catastrophe (I'm thinking from AI) or decreases it, on net?
This made me think of this: https://xkcd.com/2071/
Whom.
The objection is that gene-related issues are sensitive and you are supposed to be vague when talking about them. If you are clear and straightforward about it you violate the shroud of indeterminance to which they become attached. It forces their mind to come to conclusions they may not appreciate (which they project unto the breaker of the shroud) It could be considered psychic violence
I think the major objection to PGT-P is that it isn't ready for prime-time yet. We don't know the genes that cause schizophrenia and like most complex diseases, it's likely to be many genes. The data that current associations are based on are also flawed. Also, it's unlikely that there would be much variation in risk in a small cohort of embryos to be clinically significant (my understanding is that you need hundreds before you start to see huge differences). And you may be inadvertently selecting against some beneficial trait. Having said that, if the patient has been adequately counseled and is already doing IVF for infertility reasons and wants to do PGT-P, I think they should be able to. But at this point, I personally would not recommend someone do IVF just to do PGT-P.
The problem people seem to be gesturing at is a slightly more difficult one than you seem to acknowledge, although I generally agree that these “isolated demands for rigor” are a sign of bad faith.
The problem is the question of the rights of future people, which plagues many ethical theories, but most notoriously utilitatianism in the “repugnant conclusion” thought experiment.
It seems wrong to cause harm to occur to future people, and good to cause good things to occur to future generations. Doing good/preventing harm to currently living people is straightforward, and can in most circumstances be done consensually, since the people are alive to be consulted.
But we can’t ask future generations for permission to make modifications to their lives that we currently find desireable. Furthermore, there are questions of individualism that are not easy to answer. Can we do harm to specific future people to help the average future person? How do we deal with future people as individuals?
I think many people intuitively respond to this problem complex by leaning into passivity. It’s acceptable to passively jostle sperm because the outcome is random and unavoidable. It’s unacceptable to actively select for certain traits (say, controversially, lighter skin color) because this cannot be done without asking permission, which is unavoidable. It’s a sort of ethical default preference, or maybe a version of the naturalistic fallacy.
That said, no person or ethical theorist gets out of the problem of future people looking particularly good, so I find it hard to criticize these sorts of intuitions too much. I simply don’t share them.
When you say "the anti-abortionist doesn’t have much of a case here" (you start 'Even if', but I think the logic of your paragraph is that you don't think they have much of a case) I think you mean "they don't have much of a case outside of their general case against IVF"? You seem to me to have shown that they do have a case in total.
I find utilitarianism unsatisfactory on a number of different grounds, both epistemic and ethical. But the one that's most relevant here is that it supposes we can define a utility function for an individual human being; that is, that given any two options to choose between, we can assign each a utility score, and then the better choice for that human being is the one that attains the higher utility score---in the classic wording, the one that provides the "greatest happiness." And I simply don't believe that human brains work that way. I think that there are domains within which different values are compared regularly with each other, and thus are commensurable, and we can say that one is favored over the other; but I think there are also cases where we have to choose between values that fall into different domains, in which we do not have an established ranking, and do not even have a procedure for arriving at such a ranking. That is, I think human beings do not have global utility functions. And the choice between values that fall into different domains is inherently hard, and may be painful, as in the classic conundrum of "should I betray my country or betray my friend?" The kind of purely calculational ethics that utilitarianism offers excludes the possibility of tragedy.
And I think that, in the first place, one of the merits of virtue ethics is that the virtues it recommends may stand us in good stead in the face of tragedy, enabling us to make good choices when nothing approximating utilitarian calculation is any help. And in the second place, if we are to have any method for making choices that is generally applicable of helpful in such difficulties, that method most closely approximates the creation of works of art. And while there certainly can be a rational and even calculating aspect to art, the kind of synthesis that it calls for takes place on a different level than simple calculation; and the same may apply to the kind of synthesis that we perform when we try to lead an ethical life.
Before discussing the Alabama IVF decision, they should read the opinions (all 131+ pages!)
Here's a link: https://publicportal-api.alappeals.gov/courts/68f021c4-6a44-4735-9a76-5360b2e8af13/cms/case/343D203A-B13D-463A-8176-C46E3AE4F695/docketentrydocuments/E3D95592-3CBE-4384-AFA6-063D4595AA1D
My perspective is as a lawyer, law professor, and grandparent of a lovely girl born as the result of implantation of a donated embryo. I'm sure that colors my understanding - in particular, I appreciate something that is not mentioned in any of the news coverage I've seen, which is the generosity of couples donating embryos to infertile couples, so that the embryos and the families of those couples can get the benefit of having children.
It should surprise no one that the media's presentation of the decision is inaccurate. The majority held that fertilized embryos are covered by a state statute making compensable injuries to children by including in the definition of a child a fertilized embryo. (The various judges, and the dissent in particular, have a lot to say about whether that is a reasonable interpretation of a statute passed in the 19th century long before IVF was possible. That's a hotly contested question and the majority and dissent are both worth reading on the issue.)
The facts of the case are that a clinic - in what appears to be at the least a negligent way -- allowed someone to wander into its storage area and destroy some embryos by removing them from the cold storage and drop them on the floor. Certainly this clinic should be liable to the parents. The court had to decide whether it would be liable for damaging the parents' property, violating its contractual obligations to the parents, or in tort for injury to a child. It went with the latter. There are good arguments (made by the various opinions) for and against the majority's decision, but the majority's interpretation of the statute isn't the result of backwoods judges doing crazy stuff. It's a plausible interpretation of the statute. (There is an odd concurrence by the Chief Justice that delves deeply into theology in search of the meaning of "sanctity of life" - that one isn't (to my ears) really a legal argument but it also isn't the majority opinion and so isn't what drove the result.)
Clinics are upset (tort damages are more than property or contract damages would be). They are hoping to get the AL (and other states') legislature to amend the law to give them protection from such suits. One good way to launch that campaign is to try to frame the decision as anti-IVF and portray themselves as the good guys helping infertile couples. This clinic, however, was not the good guy - it was (at a minimum) pretty careless in protecting the embryos from people wandering around its facility. (Door locks are not high tech! Freezer locks aren't either!)
Reasonable people can differ about whether the court got the statutory interpretation question right (and the members of the court do). But this isn't a decision that will end IVF treatments in AL by imposing an intolerable burden on clinics - and it doesn't seem reasonable to me for the clinics to have stopped IVF treatments in response to it, so long as they were behaving non-negligently in their creation, storage, and implantation of the embryos.
I think that "social policy decisions" are a different case from what's outlined in the trolley problem. In the specific case of five versus one, you can see the five people, you can see the one person, and you are deciding that that specific one person must die to spare the lives of the others. In a social policy decision, you inherently cannot identify the people who will be made worse off, or how much worse off they will be made; you are deciding between probabilities of distributing damage in various ways through a large population, with the consequences for any one person not fully knowable in advance, and the consequences for every one person in the population impossible to know in advance. I don't see that intuitions that are developed in the one situation can meaningfully generalize to the other.
Another thought is related to trolley problem reasoning. How interchangeable are lives. How much work can yiu put into casting the role of "this couple's first child" before its a different person?
The only slightly reasonable objection I have heard for not selectively aborting down syndrome-fetuses is the following: If we like diversity, someone have to pay the cost for making the world more divers, ie having kids with downs syndrome even though it is probably more work than having kids without such syndrome. (This is btw also predicated on the sort of empirically testable hypothesis that people with downs doesnt suffer a lot more than healthy people. I have heard that this is true and that they even report higher happiness in surveys. At least i societies and times that doesnt hide and/or torture them.)
Even if I grant this argument it comes out more like a coordination-problem: How do we make people committ to keeping downs-fetuses and having downs-kids? It does not make ME want to parent a child with downs syndrome myself. (I have three healthy kids. Maybe one with downs is better than zero.)
Then again, the norm that we shouldnt selectively abort children can be seen as exactly the coordination-function we would need to get more diversity, to the cost of the parents who get the "diverse" kids.
it helps the parents.
in defensible cases, its preventing a real risk of being unable to care for a severely disabled child.
in indefensible cases its extending the parent's ability to shape the child to follow their idea of the good life. "My son must be a 190 IQ STEM master of the universe, so i will select for that!" The parent always has tension between instilling values and letting the child live their own life; if trait selection ever gets to that point the scale will tip to one side.
even if it doesn't work. one of the worst things is the expectation to be smart. or to live up to someone else's lifestyle. imagine if dad selects you for iq to try snd nudge you to be the brilliant scientist he wants and...you aren't. to know he just didnt want to raise you, he tried to build you; that's one huge burden.
I think talking about this as though it's a given is incorrect. As Scott mentioned - it's about low risk and high risk. So it's not a matter of choosing schizophrenia or not choosing it. It's a matter of how much you want.