We paid $38,000 to select the embryo most likely to become an unusually intelligent person. Our son Felix is now four, and he is dumb as shit. I appreciate that four-year-olds are not generally at the frontier of human cognition, but we purchased an advantage over other four-year-olds, and those are the people I am comparing him with. Laura says it is unfair to compare children. The clinic compared eight children before any of them existed, and she was comfortable authorizing the transfer.
The intervention was polygenic embryo selection: rank our eight available embryos using a score associated with cognitive ability, then implant the highest-ranked one. We were explicit that our objective function was intelligence. We did not ask the clinic to maximize charm, enthusiasm, or the probability that a preschool teacher would describe him as “a lovely little boy.” Felix is a lovely little boy. So is the neighbor’s son, who was conceived for free and understands that you press the walkie-talkie button before making the noise you want the other person to hear. Felix has spent most of the week trying to transmit a fart retrospectively. Laura points out that he now says “over,” but procedural fluency should not be confused with comprehension of the underlying system.
The clinic has explained that a polygenic score is a probabilistic predictor, rather than a guarantee about an individual outcome. I understood this when we signed the forms. What I did not understand was how heavily the burden of probabilistic uncertainty would fall on the person paying $38,000, while the person receiving it would be paid with certainty. Their website described an expected cognitive gain. Felix’s preschool assessment places him within the normal developmental range, which the counselor regards as reassuring. I could have had a normal son without commissioning a ranking. I am asking where the gain appears in the assessment, and everyone keeps answering a different question about whether I love him.
There is a further limitation involving within-family variance. Felix was the highest-scoring embryo we could produce, a qualification the genetic counselor has begun emphasizing since meeting him. Laura says this means the technology cannot supply traits that neither parent possesses. I have a master’s degree and can hardly be expected to accept that as a complete account of the situation. More concerningly, if Felix is the upper end of our available distribution, seven children even dumber than a box of rocks remain cryopreserved at the clinic. This has strengthened my support for screening. People keep asking whether the procedure was worth it, but they are only seeing the successful candidate.
I am also investigating gene–environment interaction. A favorable genetic predisposition still requires an environment in which it can express itself, and Felix’s preschool appears to have no strategy for this. They use a play-based curriculum and place him among children whose parents did not even request a score. At the last meeting, I asked whether peer effects might be attenuating his initial advantage. His teacher said he was having difficulty following the same instructions as the other children. I asked whether she had tried giving him more demanding instructions, since insufficient stimulation can also impair performance. She said the instruction was to put his shoes on. I have requested that she document whether the difficulty persists when he is given a reason to find the task intellectually engaging.
Laura thinks I am retrospectively redefining intelligence to exclude everything Felix enjoys. She notes that he chose the walkie-talkie project himself, persisted through repeated failure, and recruited another child to stand in the kitchen and report whether the fart had arrived. These are recognizable components of scientific inquiry, but so is updating when the evidence becomes overwhelming. He has been running the same failed protocol for six days. When I suggested pressing the button earlier, he told me I was doing it wrong. I am prepared to credit him with intellectual independence without treating being a fucking idiot as a necessary stage of it.
The counselor recommends waiting until he is older before assessing the intervention. This creates an identification problem. By then, any apparent gain could be attributed to the money we have spent on his education, and any disappointing result could be attributed to something Laura and I did in the meantime. We need a credible counterfactual. Laura initially welcomed my interest in our remaining embryos because she wants a second child and thought I was finally discussing our family without making it carry the entire long-term future. I asked the clinic to recover the original rankings so we could consider a lower-scoring healthy embryo as a control.
A sufficiently wide predicted gap would make the comparison more informative. Both children would receive equivalent educational opportunities, and I would preregister the principal outcome measures before either could affect the analysis by becoming someone I liked. Laura says she will not have a baby so I can determine whether Felix is comparatively less of a fucking idiot. I agree that this would be an inadequate scientific objective: the question is whether the ranking contains a useful signal, not whether we can produce someone who makes the first purchase look better. She wants to know whether I would still want the second child if the clinic could not recover the scores. I have said that we should obtain the available information before making an irreversible decision, which is the position she agreed with when we selected Felix.
My daughter is in his class. The children are all taking their shoes off because the school asks them to change into indoor shoes, and Felix gets distracted talking to her while they do it. She also forgets. I would prefer not to have her categorized as an environmental exposure in correspondence to the teacher. She is four and would like him to come to her birthday party.
I paid for embryo screening too, and the most useful thing our clinic did was insist that we discuss what decisions we would make differently after receiving the rankings. Your proposed comparison doesn't seem to have a decision attached to it. If the second child were less intelligent, you would say the first intervention worked. If the second were more intelligent, you would have two children and an argument with a clinic. In either case the second child would eventually be able to read this.
The relevant comparison is not the neighbor's child. It is the child you would have had from a randomly chosen embryo, whom you cannot observe. That makes the product difficult to evaluate for an individual purchaser; it doesn't make ordinary development evidence that selection failed. I do think the clinic owes you a clear account of how much information its score contained within your particular family. That is a question about what they sold you, and it can be answered without involving Felix.
I have asked for that account. They sent the same confidence interval we received before implantation and said that Felix's development does not justify narrowing it. If the intervention delivers no information about its own performance after four years of use, I think it is reasonable to classify that as a product limitation. The counselor objected to 'use,' though I was referring to the predictor rather than to Felix.
The discussion of your master's degree is doing more work than you acknowledge. Within-family selection doesn't choose the brightest person in a population; it chooses among combinations available from two parents. You seem to accept this only when it supports your account of the seven remaining embryos. When it might say something about you, you introduce educational attainment as if it settles the question. Were the parental scores supplied to you?