The discussion that the scientists in Nature and Science called for should remain in realism, not go on to superhumans

Just over a week ago, prominent scientists in Nature and Science called for a ban for DNA modification in human embryos. This is because the scientists presume that now it actually would be possible to alter the genome in a human embryo in order to treat genetic diseases. Consequently, this would result in modified DNA in germ cells that would be inherited to future generations. The scientists wish to have a full ethical, legal, and public discussion before any germ-line modifications will be made. Furthermore, issues of safety are of importance.

The scientists’ statement is of utmost importance and hopefully this ethical, legal, and public discussion will emerge. However, the discussion on germ-line DNA modification is at danger if the debate will be taken to the level of science fictional superhumans, as already has happen. Not only can such discussion cause unnecessary public worry, it also leads the deliberation away from the actual and urgent questions.

The traditional genetic therapy for somatic cells implies genetic modification of target cells. By knocking out or adding a gene in the target cells the functions of the cells can be altered and this has a therapeutic effect: for example, in inhibiting the growth of a cancer cell, or making the cell produce a lacking enzyme. These modifications take place only in the target cells or tissue. However, if the genetic modification will be made to the embryo, the modification will be inherited to all the cells of the future human being – including the germ cells. Consequently, the modification will be passed to potential progeny.

The scientists highlight that despite the existence of the relevant techniques (such as the of new CRISPR-Cas9 technology), it is not certain that they could be used to human genome in a safe and efficient manner. Controlling the exact level of modifications is difficult and there is the risk of inducing untoward cuts in the genome. The scientists emphasize that the precise effects might be known only after birth, and even then, potential problem might arise not until after some years. It is important to understand that the existence of some technique does not mean that this technique could be used in a totally controlled manner. Apart from the safety issues, the scientists wish a framework for open discourse for the ethical, legal, and societal implications of potential germ-line modifications. The scientists call for a careful discussion about these issues – not only because the issues at hand really need to be deliberated, but because they worry that potential mistrust or protest in the public will result as a powerful objection to these new techniques as such that offer a great potential of therapeutic development.

At the dawn of the recombinant DNA era, the most important lesson learned was that public trust in science ultimately begins with and requires ongoing transparency and open discussion. That lesson is amplified today with the emergence of CRISPR-Cas9 technology and the imminent prospects for genome engineering. Initiating these fascinating and challenging discussions now will optimize the decisions society will make at the advent of a new era in biology and genetics.

Despite the careful formulation of the scientists in Nature and Science, many further writings are taking the discussion completely to another direction. For example, The New York Times writes that the biologists fear that the new technique is so effective and easy to use that some physicians may push ahead before its safety can be assessed, and they want the public to understand the ethical issues surrounding the technique, which could be used to cure genetic diseases, but also to enhance qualities like beauty or intelligence. International Business Times, for its part, reports the piece of news with a title saying “Alarm over genetic editing of human embryos that opens door to designer babies and superhumans”, reformulating that the biologists fear that germ-line engineering takes mankind on the dangerous path to superhumans and designer babies, for those who can afford it. The biggest newspaper in Finland also goes with the title “Scientists fear: manipulating genes will lead to making superhumans”.

Reading the original articles under the microscope, the biologists said nothing about superhumans. The new era in biology and genetics means the possibility to make germ-line modifications to genes we know cause some diseases. However, taking the mentioned examples of beauty and intelligence, there is neither conceptual unanimity on what these characteristics actually mean nor reasons to believe there might a genetic way to effectively influence them. What is at stake is the possibility to modify particular genes that cause a certain phenotype. Highly polygenic and multifactorial characteristicts remain outside this scope.

If the scientists’ urge to discuss the matter is now translated into a fierce debate about what is beauty, what is intelligence, what would superhumans be like, and where the exact genes for beauty, intelligence, and superhumanity lurk, the task has failed. First, this would take the debate away from realism and possibly increase extensive fears against such techniques carrying a huge therapeutic potential – the genome-editing technologies may offer a powerful approach to treat many human diseases, including HIV/AIDS, haemophilia, sickle-cell anaemia and several forms of cancer (somatically). Putting together genome-editing technologies and superhumans as a matter of fact does not do any good for the discussion.

Second, translating the scientists’ call into issues of superhumanity will take us away from the actual questions that need discussion. We need serious debate about autonomy, beneficence, harm, justice, precaution, dignity, solidarity, and like, concerning germ-line therapies. The future individual might be forever cured from a genetic disease, or forever, together with their progeny, suffering from an unexpected side-effect. What does this mean to the individual, to the society, to the humanity? There should be room for varying ethical discourses.


  • Facebook
  • Twitter
  • Reddit

2 Responses to The discussion that the scientists in Nature and Science called for should remain in realism, not go on to superhumans

  • Anders Sandberg says:

    We have a somewhat similar problem in the discussion about AI safety, exemplified by the Future of Life Institute’s open letter (“Research Priorities for Robust and Beneficial Artificial Intelligence”). The main point of the letter is to get researchers to work on maximizing the societal benefit of AI – safety and utility – and not just its capability. The letter itself is pretty agnostic about just how powerful or dangerous AI could become, but the framing in the press invariably comes with terminator robots.

    Here the issue is that superintelligence *is* a valid part of the discussion: AI safety stretches from the near-term issues of what autonomous cars and data mining systems do, over whether we should accept autonomous weapons and how to handle technological unemployment to existential risks from very smart systems. Arguing it shouldn’t be discussed because it is speculative is problematic, since autonomous weapons (or cars) are speculative until they happen. Arguing that there are lots of cognitive biases and ideology polluting analysis also applies to technological unemployment. And so on: there is no good dividing line, and as we at FHI have argued, even low-probability and rather far off existential risks have a lot of moral priority. Yet the superintelligence/intelligence explosion part of the discussion gets far more attention than the more mundane parts, *and* its weirdness is used by those who think we shouldn’t do AI ethics now to argue that the entire domain is silly (often they think we do not need it because it is silly, in a self-reinforcing loop of rejection).

    So while I do agree with the post that it is a problem that the media misrepresents what the call for moratorium actually is about, I also think we should recognize that superhumans are a part of the discussion. Saying that this is just about autonomy, beneficence, harm, justice, precaution, dignity, solidarity, and the like concerning germ-line therapies means that big issues that actually do carry heavy ethical weight are excluded from the discussion. In fact, restricting it to the traditional medical ethics domain means that by the time someone uses these methods for enhancement (many therapies can be turned into enhancement) there will not have been a proper analysis. Worse, if some of these non-therapeutic uses have sufficiently ethically weighty effects they might affect the morality of the entire enterprise yet not be allowed to be explored or taken into account.

    I fear that restricting conversations about emerging technologies to non-speculative here-and-now issues means they lose much of their foresight. Imagine “nuclear ethics” in the 1930s, discussing the issues of safety, beneficience, precaution and international solidarity in the use of radioactivity for research, medicine and energy production but explicitly renouncing speculation about nuclear weapons and their ethics.

  • Johanna Ahola-Launonen says:

    Thanks Anders for this insightful comment, I very much agree with you, and indeed hoping for “restrictions for discussion” is always kind of problematic. But while in agreement, here are my points that I’m still holding for:

    1) If the germ-line-therapy issue is taken to the superhuman level, we will not have a discussion about what is at stake right now, that is, the possibility to want to manipulate embryo genome to intervene particular genes causing genetic diseases – and all the ethical issues related to that. I’m afraid that the “superhuman-gene-fuss” would lead the discussion, while the actual dilemma would be left with less attention. Well, maybe this is a useless worry, because I guess at least the expert-led panels and discussions will stay on the subject.

    2) While I really don’t know much about AI and its potential possibilities, I know genetics, and I’m pretty certain that the superhuman & genetics – related characteristics, such as intelligence and beaty mentioned in the blog, whatever they mean, are not such characteristics that they could be effectively enhanched genetically. This is because there is no single of two genes that would make them and they are not genetically determined. So in this sense, I would say that the slippery slope from e.g. curing Huntington’s with germ-line therapy to genetically enhancing intelligence is very implausible.

    So while I have no competence on saying what superhuman-scenarios are reasonable related to AI and like, I’m pretty confident about the current scientific paradigm concerning genetics where it can be drawn that the idea of manipulating with effectivity highly polygenic and multifactorial characteristics is not reasonable to expect. Of course: obviously it cannot be falsified that it turns out that much that we know about genetics turns our to be wrong, but some scenarios are more reasonable than others.


Subscribe Via Email