Skip to content

Ethics of Editing the Book of Life

It’s got Nobel Prize written all over it. The scientific innovation, CRISPR, which enables accurate ‘editing’ of DNA (compared to current techniques where a viral vector introduces the DNA at random), has had one team member “jumping out of my skin with excitement”. Still at basic science level, it has already been hailed as a potential new treatment for Huntington’s Disease, HIV and other disorders. It is apparently so easy to use that Professor Mello, who was involved in the project, has said, “a total novice in my lab got it to work”.

One possible application that has been suggested is ‘correcting’ the germline: changing the genetics of sperm, eggs and embryos, to eliminate diseases not just in individuals, but in future generations. The designer baby is in production.

In some ways, germline therapies are redundant. We already have a technique- pre-implantation genetic diagnosis (PGD) which can address single gene disorders. This involves IVF and genetic diagnosis of a range of embryos, selecting those which do not carry the mutation. Whilst replacing PGD with CRISPR might bring some benefits (for example for couples who do not produce multiple viable embryos during IVF), it is likely to also carry some risks- we don’t know this yet, and may not for some time.

Robin Lovell Badge warned:

“Although remarkably efficient compared to other techniques, the genetic changes introduced by the CRISPR technique are not always as perfect as designed and on occasion it could introduce problems that are just as worse as the one being corrected. Moreover, there are problems of “off-target” hits, where the method alters DNA at other (similar) sites elsewhere in the genome and not just the gene being targeted. We can deal with this in mice or fish by breeding the animals and segregating away the mutations at these other sites, but not with patients and these off-target mutations could be bad.”

So it seems that genetic selection (IVF and PGD) are superior to genetic engineering (CRISPR). Or is it?

For some, this technique will be irresistible. It will remove the ethical arguments against PGD which inevitably discards embryos because of one faulty gene. Doctors could produce one embryo and correct its genetic defects, if it has one. For pro-lifers, this would avoid the need to produce excess embryos, inevitably discarding some.

More importantly, genetic engineering using the latest technique (CRISPR) is curing rather than choosing. It is also permanent. With PGD, offspring may still be carriers of a disease (ie if an embryo is produced with one copy of an abnormal gene but not two). Now the offspring will also be relieved of having to go through the same difficult decisions and emotionally taxing procedures themselves if they want to have children.

But there is another advance to engineering over selection. Selection involves choosing between different embryos and different possible lives. It does not benefit the embryo, because if the selection had not occurred, then a different embryo and child would have been brought into existence. This is what Derek Parfit called the Non-Identity Problem. According to the Non-Identity Problem, failing to select a healthy embryo does not harm the embryo with the mutation, unless the mutation renders life not worth living. In virtually all cases, genetic disease is not so bad that it renders life not worth living. So failing to select using PGD does not harm the future child.

Genetic engineering is different. If you fail to employ it, you make a future child worse off than he or she would otherwise have been. Failing to develop and use genetic engineering techniques like CRISPR harms future children.

So there are both personal reasons (for those opposed to destruction of embryos) and philosophical reasons to view genetic engineering as being superior to genetic selection. Of course, if there are risks with one rather than the other, that may tip the balance. But genetic engineering (such as with CRISPR) benefits people in a way that genetic selection (PGD) does not.

Ethics is thus vital to the evaluation of these new genetic engineering technologies.

More importantly, those pioneers who opt to try germ line genetic engineering (and there are sure to be some, even though it is illegal) will inevitably be guinea pigs for a technology that will be vastly more attractive to people unaffected by single gene disorders. Once genetic engineering is safe and effective for single gene disorders, the desire to improve polygenic dispositions to common disease, and bad genes in general, will be irresistible. Through their experiences of these pioneers who seek to cure their single gene disorders with genetic engineering, we will go on to learn more about this technique and what can be achieved. Robin Lovell-Badge envisages a usage to introduce resistance to infections, such as HIV (this is currently illegal in many countries). There are many more possible applications of this, some of which will raise as many ethical issues as the technique solves. As our knowledge of genes and their effects increases, so will the possibilities.

The spectre of enhancement looms large with these new technologies.

I have argued that we have a moral obligation (although it may be overridden by other obligations) to select the children with the best chance of the best life, and that parents should have discretion over what traits they select for. One common objection to this has been to point out that this does not improve the child’s life as such- it just means a different child is born, and the other child or children who were not selected do not get any shot at life at all. Using CRISPR would overcome that objection. But is there in that case an even stronger moral obligation to edit our children?

Share on

7 Comment on this post

  1. I have argued that we have a moral obligation (although it may be overridden by other obligations) to select the children with the best chance of the best life, and that parents should have discretion over what traits they select for.

    Really? Can you please restate it here so that I can try to understand what you are after?

      1. Thank you for the link. ‘t was an enjoyable reading. Two remarks:

        1. It is not clear at all that Parfit’s Simple Case example (the potential mother with rubella) is enough for supporting the Principle of Beneficence. For instance, here is a principle that supports Parfit’s Simple Case but falls short as far as the Principle of Beneficence is concerned:

        (Min) Our actions should all other things being equal minimize the occurrence of properties that individually will make a negative contribution to anybody’s well-being, actual or yet-to-become-actual.

        (Min) supports the Simple Case because delaying the pregnancy is an instance of preventing the occurrence of a certain disease in an individual whose well-being it would make a negative contribution to.

        But of course, your Principle of Beneficence requires more than (Min); it requires, among other things:

        (Max) Our actions should all other things being equal maximize the occurrence of properties that collectively will make a positive contribution to anybody’s well-being, actual or yet-to-become actual.

        together with the claim that:

        (*) Selecting genetic features of yet-to-become-actual individuals can succeed or fail with respect to (Max).

        Since (Max) is not supported, the argument looks incomplete.

        2. (Max) seems to entail some absurdity unless it is restricted to actual individuals. Indeed, if unrestricted, (Max) seems to create duties whose satisfaction conditions include possible individuals, that is, individuals such that, possibly, they will never exist. (Max) seems to create those duties since it implies rankings between sets of properties that collectively make a difference to a well-being, and the assessment of that collective contribution cannot but depend on how those properties will interact with other properties of the individual as well as with its environment.

        (Min) does not have this problem since it excludes undesirable properties without bothering whether those properties will actually occur or how they would collectively contribute to a well-being.

        But if so restricted, (Max) of course cannot justify your Principle of Beneficence.

        1. A sentence is missing at the end of the one before the last paragraph: “But if the collective contribution of the properties to the possible individual’s well-being can be thus assessed, then there is no reason to stop there and not consider the contribution of these properties to the well-being of the possible individuals they could give birth to, etc. In the end, (Max) seems to imply duties toward not only possible individuals that will be actual, but possible individuals that will not be actual — merely possible individuals.

  2. Yes, but who determines what is the best chance for the best life? People have radical differences on what constitutes a good life.

  3. “I have argued that we have a moral obligation (although it may be overridden by other obligations) to select the children with the best chance of the best life, and that parents should have discretion over what traits they select for.”

    It could be argued that what we really need is a new model of social responsibility for humanity’s future, and actually make a distinction between those having kids for selfish reasons of their own, and those prepared to take part in a breeding programme actively aimed at improving the species. Only the latter would have access to genetic engineering. They would be vetted to weed out religious types, fascists etc, and their discretion would be moderated by an obligation to follow a core genetic recipe determined by experts. Although this idea makes sense to me, it would require a popular enthusiasm for positive eugenics which doesn’t seem to be much in evidence at this stage.

Comments are closed.