CRISPR AND THE
TREATMENT/ENHANCEMENT
DISTINCTION
On Vagueness, Borderline Cases
and Germline Genome Editing
Ellen M. Svensson
CRISPR OCH DISTINKTIONEN MELLAN
BEHANDLING/FÖRBÄTTRING
Om vaghet, Borderline fall och ärftlig genredigering
Abstract
In this thesis, I argue that the treatment/enhancement distinction that is central to the ethical debate concerning germline genome editing and CRISPR is too vague to be ethically and normatively guiding. The problem of vagueness is twofold, being both a semantic and epistemic issue. This vagueness creates borderline cases, cases that cannot be properly
defined as either treatment or enhancement, I call this The Borderline Cases Argument. These borderline cases enable a slippery slope towards eugenic practices, radical enhancement and dangerous applications of CRISPR. The distinction therefore fails to be action guiding as it cannot distinguish treatment from enhancement as well as failing to correspond to what is genuinely morally problematic with germline genome editing and not, I call this The
Argument of Missing the Point. In using the treatment/enhancement distinction we therefore risk losing control over how CRISPR is used and for what purposes.
Keywords
Index ...
1. Introduction ... 1
1.1 Disposition ... 3
2. The Vagueness Problem ... 3
2.1 The Borderline Cases Argument ... 4
2.2 The Slippery Slope ... 8
3. Arguments in Favour of a Distinction ... 10
3.1 The Difference Between Treatment and Enhancement ... 11
3.2 Unwanted Consequences and Human Nature ... 16
3.3 Eugenics and Disability Rights ... 19
4. Arguments Against a Distinction ... 23
4.1 Limited Utility ... 23
4.2 What “Human Nature”?! ... 27
4.3 Unwanted Implications ... 28
5. Conclusion ... 29
1. Introduction
CRISPR is a gene editing tool often referred to as a “gene scissors” since it can cut, insert, delete and replace any given sequence of DNA.1 But like most tools it is simply this: a tool. It only tells us that DNA can be edited, not that it should be edited or why. Therefore, the ethical discussion concerning gene editing and CRISPR needs a way to distinguish between ethically acceptable and ethically unacceptable applications of this tool. Such a distinction should, ideally, be action-guiding and represent a demarcation between unproblematic (and generally permissible) and problematic (and generally impermissible) gene edits. This distinction should not be vague and as such it should be distinguished by three specific features: First, it must be able to draw a sharp line between the two so that we know which is which. Second, this line must correspond to what is genuinely morally problematic and not. So, that those interventions that fall on one side of the demarcation line are ethically unproblematic and the interventions that fall on the other side of the demarcation line are ethically problematic. Third and finally, it must enable a reasonable demarcation against slippery slopes towards eugenic practices and radical enhancements, where radical enhancement would include significant improvements of our current attributes and abilities that would go far beyond what we, as human beings, are currently capable of.2
The most viable candidate for this role is the treatment/enhancement distinction, where gene edits made with the purpose of treating diseases are deemed ethically acceptable and
unproblematic, while gene edits made with the purpose of enhancement are deemed ethically unacceptable and problematic. The treatment/enhancement distinction is intuitively
reasonable since it would allow for gene editing to prevent severe, chronic and painful diseases like cystic fibrosis or Huntington’s disease. At the same time, it would deem gene edits with the purpose of, for example, creating a person that is three meters tall with double the muscle mass of today ethically unacceptable and problematic because of the difference in intent. In extension, the treatment/enhancement distinction is useful because it is something that we, as citizens, can understand and, therefore, something that lawmakers and
policymakers could rely on when regulating the use of gene editing technologies like
1 Jennifer A. Doudna and Samuel H. Sternberg, A Crack in Creation: Gene Editing and the Unthinkable Power to Control Evolution (Boston: Mariner Books, 2018), 100–1
CRISPR.3 The distinction, in this sense, is intelligible and as such useful. For example, a recent survey made by the UK’s Royal Society found that 76% of people in the UK were in favour for using germline genome editing to correct genetic diseases but very few were in favour for using the same tool for non-therapeutic purposes: 69% were against edits to change eye or hair colour and 60% against edits to enhance intelligence.4 So the
treatment/enhancement distinction corresponds well with how we currently view germline genome editing and its applications.
However, I am going to argue that this distinction is, in fact, vague since it fails to meet both the first and the second feature needed for this ethical demarcation line. The vagueness of the treatment/enhancement distinction makes it impossible to draw a sharp line between what should be considered treatment and what should be considered enhancement. Not only do interventions fall on the “wrong side” of this demarcation line, but the vagueness of the distinction also leads to borderline cases: cases that simply fall in between the two categories. I call this The Borderline Case Argument. Even more problematic, the treatment/enhancement distinction fails to recognise the most problematic features of gene editing and as such it fails to give correct ethical and moral guidance. I call this The Argument of Missing the Point. Therefore, the distinction also fails on the third feature needed: it cannot enable a reasonable demarcation against slippery slopes towards eugenic practices and radical enhancements.
The question posed in this thesis is whether it is possible to make a distinction between treatment and enhancement that separates problematic applications of CRISPR from unproblematic applications of CRISPR. I am arguing that it is not. The
treatment/enhancement distinction is too vague to be ethically or normatively guiding. As such, it cannot occupy the definitive moral or normative function it has been given in the debate. The problem, as we shall see, is that the treatment/enhancement distinction remains vague and as such always leads to borderline cases enabling a slippery slope towards eugenic practices, radical enhancement and dangerous applications of CRISPR that we will no longer be able to control.
3 Andrew McGee, “Using the therapy and enhancement distinction in law and policy,” Bioethics Vol. 34, Issue 1 (January 2020): 78
My discussion will focus on germline genome editing, henceforth GGE, since edits to germline cells are heritable and will be passed down to any future descendants. Edits done to somatic cell, instead, are not heritable and hence will not invoke the same ethical concerns.
1.1 Disposition
This thesis is divided into 5 sections. In section 2 I present the vagueness problem and The Borderline Case Argument, explaining how this enables the slippery slope. In section 3 and 4 I give an overview of the current arguments given in favour for and against the
treatment/enhancement distinction as well as discussing The Argument of Missing the Point. Finally, in section 5 I draw my conclusions and argue that the distinction remains vague.
2. The Vagueness Problem
In this section I present the vagueness problem concerning the treatment/enhancement distinction. It is important, however, to note that the problem is not GGE but rather how we, as the users of CRISPR, apply it. I am not arguing that we shouldn’t use CRISPR, because it does represent an unprecedented possibility to cure genetic disease before it occurs. But I am arguing that the treatment/enhancement distinction is the wrong way of looking at the ethical concerns with GGE since it is too vague to be ethically and normatively guiding. Therefore, the central role this distinction occupies in the ethical debate on CRISPR is problematic since it is too vague to fill the definitive moral or legal function it has been given.
The vagueness problem concerning the treatment/enhancement distinction can be illustrated with the help of the Sorites paradoxes. The Sorites paradoxes get their name from the Greek word for “heap” (soros) and the paradox is therefore also commonly known as the paradox of the heap. You should imagine that you are standing in front of a heap of sand. If you remove one grain of sand, you are still standing in front of a heap, simply because removing a single grain of sand cannot turn something that is a heap into a non-heap. The paradox relies on the premise that: “If two collections of grains of sand differ in number by just one grain, then both or neither are heaps.”5 One grain of sand cannot be the difference in between being or not being a heap. But if you let this process unfold and keep removing grain after grain of sand you will, sooner or later, be left with only one grain of sand – which, following the logic
of the paradox, seems to imply that all grains of sand are heaps or that there are no heaps of sand. If we apply this to CRISPR we can consider, for example, edits done to slow down the process of ageing with the purpose of extending our life expectancy. If we add one year to our life expectancy this seems to be a clear case of treatment, because in the end all healthcare aim to extend our life expectancy and increase our quality of life. But if we let this process unfold, as with the grains of sand, and add one year after the other to our life expectancy until we live an average of 600 years this seems like a clear case of enhancement. Just as in the Sorites paradox, where we struggle to give a decisive number as to how many grains of sand are needed for a heap, we struggle to give a decisive number as to how many years added should count as an enhancement or treatment. The line between treatment and enhancement remains vague because in adding one year to our life expectancy this seems like a clear case of treatment, but in adding 600 years this seems like a clear case of enhancement. But we still struggle to give a decisive number as to how many added years should count as treatment and how many added years that should count as enhancement. If so, the slope towards eugenic practices and radical enhancement seems to remain slippery because the vagueness of the treatment/enhancement distinction makes it impossible for us to know when we have crossed the line between the two. This is because vagueness creates borderline cases, as I shall argue in the next section, that is: cases of GGE that cannot be properly classified as either treatment or enhancement.
2.1 The Borderline Cases Argument
the words “treatment” and “enhancement” are both vague and, as we shall see in section 3-4, largely a matter of definition. Since the words and concepts that carry the distinction are both vague the problem with this distinction is partially semantic. “A vague word admits of borderline cases – cases in which we do not know whether to apply the word or not, even though we have all the kinds of information that we would normally regard as sufficient to settle the matter.”6 A vague word such as “treatment” and “enhancement” will therefore inevitably lead to borderline cases: interventions that cannot be classified as one or the other, where we don’t know whether to apply the word or not, despite having all the information usually needed for a correct verdict. What vagueness is, in of itself, a bit of a philosophical rabbit hole; it could be a property of language, a property of the world or a property that we, ourselves possess. Vagueness could also be the absence of fact or of a definite truth.7 For my discussion, I will consider vagueness as the absence of a sharp boundary, since I believe this speaks most clearly of the problem I´m trying to highlight with the treatment/enhancement distinction.
Since vagueness leads to borderline cases this means that there will always be CRISPR edits and cases of GGE that will not fall either under the description of “treatment” or
“enhancement”. These borderline cases will be something in between treatment and
enhancement and, despite having all the information needed about the intervention at hand, we will fail to classify them as either one or the other. However, some ethicists, like Nick Bostrom and Julian Savulescu, argue that vagueness would not present a problem for an ethical demarcation line. They argue that a line between what we consider problematic enhancements and non-problematic enhancements must not be sharp. In extension, they also note that many important and useful philosophical terms are, in fact, vague. Nonetheless they argue that the line we draw must meet two challenges, essentially the features I have outlined in my introduction: first, it must make an intelligible and non-arbitrary account of what counts as an enhancement and, second, this account must track a morally relevant distinction.8
6 Ibid. 41
7 Ibid. 41–2
Bostrom and Savulescu might be right that vagueness, in of itself, does not present an issue for the treatment/enhancement distinction and that many useful philosophical terms are, in fact, vague. Torbjörn Tännsjö for example, distinguishes between negative medical
intervention, positive interventions and enhancement and then go on to clarify that the biggest shortcoming of the distinction is that it is vague.9 He believes the vagueness arises mainly because we struggle to draw a sharp line between what is considered healthy and not, but more of this in section 3-4. Despite this, he goes on to use the distinction anyway. Plenty other ethicist also note the vagueness of the distinction and then apply it nonetheless. But arguably, neither Bostrom nor Savulesco or Tännsjö are concerned with drawing any type of ethical demarcation lines between acceptable and unacceptable interventions since they are exclusively concerned with the consequences of the edits made. To them, the only thing that matters is whether the edits create more or less happiness, utility and human well-being rather than whether the edits were made with treatment or enhancement purposes. But only counting the consequences is a roll of the dice, as we do not yet know the actual consequences of GGE. It therefore seems reasonable to inforce a distinction between ethically
acceptable/unacceptable applications of CRISPR that can function as an ethical demarcation line. The risk of rolling the dice with our own genetic makeup is simply too high.
However, to create a demarcation between problematic and non-problematic uses of CRISPR, vagueness will present an immense and possibly unavoidable problem: vagueness will always lead to borderline cases. These borderline cases will then make it impossible for the first challenge from Bostrom and Savulescu to be met: making the line “non-arbitrary”. Since some interventions will classify as “borderline” and not properly be classified as either treatment or enhancement this line will remain arbitrary, intelligible but not useful. This will also, in extension, make it impossible to rise to the second challenge presented by Bostrom and Savulescu: making the distinction track a “morally relevant distinction”. Because our distinction will be cluttered and covered in borderline cases, it will never be morally relevant since any opposition to this line will claim that there are cases that don’t fall on either side of the line we have drawn.
If so, then one could object that for all the borderline cases that fall outside (or in between) the treatment/enhancement distinction we simply need a third category: borderline cases in need of case-to-case evaluation. This would make the distinction into a three-category distinction: clear cases of treatment (considered ethically acceptable), clear cases of enhancement (considered ethically problematic) and clear borderline cases (considered ethically unclear and vague). This would be the best solution if our main concern is the prospect of slippery slopes towards eugenic practices since it creates a clear demarcation line towards enhancement and radical enhancement. It would also be reasonable since it would allow for edits made with the purpose of treating severe and chronic illnesses, if we assume that GGE is made sufficiently medically safe. The three-category distinction would also solve the problem of borderline cases since these would be considered ethically questionable, neither clearly problematic or unproblematic.
But on the other hand, the three-category distinction doesn’t solve any issues of vagueness, it rather procrastinates the problem because it presumes that we can draw a clear demarcation line between treatment and borderline cases and between borderline cases and enhancement. This would create the same concerns as the two-category distinction since the fuzziness of the words and concepts would remain. Moreover, there is nothing to support the assumption that GGE made with the purpose of treatment would, by definition, be ethically unproblematic and that GGE made with the purpose of enhancement would, by definition, be ethically
2.2 The Slippery Slope
As I have mentioned in the introduction, the distinction between ethically acceptable and ethically unacceptable applications of CRISPR that the treatment/enhancement distinction represents, must enable a reasonable demarcation against slippery slopes. This is the third feature needed for the distinction not to be vague. The problem of vagueness is that it leads to borderline cases, and these borderline cases then enable a slippery slope towards eugenic practices, radical enhancement and dangerous applications of CRISPR that we will no longer be able to control. I have already argued that the vagueness of this distinction is partially a semantic issue since the words “treatment” and “enhancement” are both vague and largely a matter of definition. But there is also an epistemic problem with the treatment/enhancement distinction, since the issue is not simply knowing where to draw the line but also about being able to know when we crossed the line.
In a highly influential article by G. A. Cohen, he writes about a man that substitutes his “fleshy parts” for artificial ones participating in, what I would call, a process of radical enhancement.10 Cohen then goes on to conclude that the problem with gene editing lies not only in the difficulty of drawing the line between what we now find ethically acceptable or unacceptable, but rather in the risk of a slippery slope. Cohen’s main concern is not simply that we don´t know where to draw the line, but rather that for every line we draw we will later draw it somewhere else.11 This is a particular kind of slippery slope, Cohen argues, because as our tolerance for CRISPR increases, our desire to edit will increase with it and as a result we will keep moving the line we have drawn into a more and more accepting place. The basic idea of the slippery slope argument is that, for every edit we make, we will inevitably edit even more. Even if we knew where to draw the line between treatment/enhancement, we would draw it at a different place later in time as acceptance and tolerance for the technology increases. The slope is slippery because we do not know where to stop editing or how much we should edit before we are satisfied. We might suggest today, that there is nothing ethically wrong in using GGE to prevent diseases such as cystic fibrosis or Huntington’s disease but it
10 G. A. Cohen, “Rescuing Conservatism: A Defense of Existing Value” in Reasons and Recognition: Essays on the Philosophy of T.M. Scanlon, Eds. R. Jay Wallace, Rahul Kumar and Samuel Freeman (Oxford: Oxford University Press, 2011), 7
will then, inevitably, lead to GGE being used for other genetic traits that are considered “undesirable”, such as obesity or shortness.12
There are different types of slippery slope arguments, however. Some that might be called “conceptual” or “logical” slippery slopes where it is argued that once the first important step is taken, we are logically committed to all subsequent steps and consequently sliding to the bottom of the slope. The solution would be to draw a non-arbitrary line that no one is allowed to cross, but as we shall see in the next sections this is a difficult task since the
treatment/enhancement distinction remains too vague to fill this function. Another type of “slippery slope” is a psychological one, where the argument is based on a prediction that one practice will lead to another. If we accept edits against cystic fibrosis, it will then be difficult for us not to accept edits against diabetes for example. Accepting one type of edit will
psychologically prepare us to accept other edits as we will fail to recognise any meaningful or morally significant difference between them.13 The biggest shortcoming with this line of argument is that it is simply a prediction of what might happen, it rests on the idea that “once we start using CRISPR we will probably escalate our use as time passes”. But this is not a necessary rule or logical conclusion, it is rather a fear of how the process might unfold. In opposition to this, one could argue that in using current selective reproduction tools like in vitro fertilization (IVF) followed by pre-implantory genetic diagnosis (PGD = embryo
selection) we can avoid over 250 genetic diseases such as cystic fibrosis, Huntington’s disease and haemophilia.14 But the number of genetic traits that we can choose against keep
increasing and the list of genetic diseases we can avoid keeps growing. It does seem like the more genetic traits we can choose from, the more the demand for this type of product increases and in the end, there is nothing stopping us from also using IVF + PGD to choose genetic traits like gender or eye colour.
Even if the slippery slope arguments warn against the danger of eugenics, enhancement, designer babies or discrimination against the sick and the disabled they do not deem germline therapies themselves as ethically unacceptable. Some argue that this is the greatest weakness
12 Françoise Baylis, Altered Inheritance: CRISPR and the Ethics of Human Genome Editing (Cambridge, Massachusetts: Harvard University Press, 2019), 175–76
13 Ibid. 176
of these arguments because they criticise the ethically unacceptable supposed consequences of the applications of technologies such as CRISPR, but their objections remain
suppositions.15 “Ultimately, slippery slope arguments cannot convince of their implicit assumption, that legal or other barriers are not sufficient to control the development of technologies.”16 The critique being that the slippery slope arguments cannot convince us that CRISPR could not be sufficiently regulated through legislation and strict policy for its applications. This counter argument is hardly convincing because as we shall see in section 3.3, today’s selective reproduction like PGD and prenatal screenings does lead to as many as 90% of foetuses effected by Down’s syndrome being aborted and the CRISPR-twins born in 2018 came into the world during a supposed “moratorium on germline gene editing.”. In a global world and under a capitalist market economy, it will become increasingly harder to regulate the use of CRISPR since “the cat is already out of the box”. The technology exists, there are clearly people willing to use it and there is a market for this product.
This concludes section 2 of my argument, in the next section I present the current and most prominent arguments in favour of a treatment/enhancement distinction arguing that this demarcation line remains vague.
3. Arguments in Favour of a Distinction
There are many different reasons as to why some ethicists defend and enforce the treatment/enhancement distinction. In this section I present some of the more prominent arguments and reasons given in favour for this distinction to illustrate that it fails on the first and second feature needed for this ethical demarcation line not to be vague. Not only can the distinction not draw a sharp line between treatment and enhancement, and therefore creates borderline cases, but it also fails to properly frame the issue and correspond with what is genuinely morally problematic and not – it misses the point. The distinction remains vague and as such, fails to be action-guiding.
15 Matthias Braun, Hannah Schickl and Peter Dabrock, “Between Moral Hazard and Legal Uncertainty: An Introduction” in Between Moral Hazard and Legal Uncertainty: Ethical, Legal and Societal Challenges of Human Genome Editing, eds. Matthias Braun, Hannah Schickl and Peter Dabrock (Wiesbaden: Springer Fachmedien Weisbaden GmbH, part of Springer Nature, 2018), 7
3.1 The Difference Between Treatment and Enhancement
The treatment/enhancement distinction relies on two separate concepts, that of what can properly be defined as “treatment” and that of what can properly be defined as “enhancement” and the difference between them. I will first bring forward a defence for what can correctly be defined as “treatment” presented by Eric T. Juengst and then a defence for what can correctly be defined as “enhancement” presented by Andrew McGee.
The issues concerning the treatment/enhancement distinction leads ethicists such as Juengst to the conclusion that a line can be drawn between prevention and enhancement uses of gene therapy, and therefore a line can be drawn between treatment and enhancement, but only if we employ a correct understanding of health and disease; that is, a proper understanding as to what can correctly be classified as “treatment”.17 The problem with the distinction is that oftentimes we enhance the body in order to prevent disease, for example we enhance our immune systems with vaccines in order to prevent diseases such as polio. This makes the distinction blurry since enhancements are used to treat and prevent disease.18 Juengst argue that the problem with drawing a line between treatment and enhancement lies in the fact that the distinction springs from concepts such as “Normal Function”, as formulated by Norman Daniels, to define the limit of what interventions can count as legitimate health care needs. “On these accounts, to be healthy is to function, under typical circumstances, with the typical efficiency of members of one's age, gender and species.”19 What is healthy and not is
therefore defined according to the general health of other people of the same age and gender. The basic idea here is that interventions that aims to restore “normal function” are treatments, and interventions that go beyond “normal function” are enhancements. A similar account is that of “positive health”, that is interventions that allow for more than “normal function” but less than enhancement. The idea is that “negative health” interventions would aim to avoid or cure diseases such as cancer but “positive health” interventions would allow for interventions that for example gives us better sight or that increase athletic stamina.
17 Eric T. Juengst, “Can Enhancement be Distinguished from Prevention in Genetic
Medicine?”, The Journal of Medicine and Philosophy: A Forum for Bioethics and Philosophy of Medicine, Volume 22, Issue 2 (April 1997): 127–28
The shortcoming of both the “positive health” account and the “normal function” account, Juengst argue, is an epistemic one because advocates for these views cannot use the lines they draw since they can never tell when the line is being crossed.20 They might be able to draw a line between treatment and enhancement but this line cannot indicate, on its own, when it is being crossed.21 The solution is to adapt the old notion of a “disease entity” and accept these entities as metaphysically real – if our understanding of disease is made robust enough and objectively knowable then we can correctly argue that the aim of legitimate preventive genetic healthcare should be to protect us from these entities. Only this way can we say that “…what distinguishes prevention from enhancement, in other words, is that medical
prevention seeks to improve the body's defences against disease.”22 Treatment is therefore defined as interventions that improve our defences against diseases.
Instead of defending the concept of “treatment” as Juengst does, Andrew McGee criticises the understanding of “enhancement” that some ethicists, like John Harris and Allen Buchanan, employ when denying the distinction between treatment and enhancement.23 Notably McGee uses the term “therapy” instead of “treatment” as I do, for my purpose the terms can be used interchangeably since they aim to describe interventions of the same kind and I will therefore continue to use the term “treatment”. McGee argues that doubts about the utility of the line drawn between treatment and enhancement is often overrated and based on a bias in favour for enhancement.24 This is a highly probable claim since those who deny the line between treatment and enhancement usually go on to argue that since there is no difference between the two, CRISPR should be used for enhancement purposes. One way of attacking the treatment/enhancement distinction is therefore to claim that all treatments are a type of enhancement, essentially the same issue that Juengst discusses. McGee claims that this problem arises because definitions of enhancements are often too broad, which allows philosophers such as Harris to fuse together the concept of treatment and enhancement.25 Harris is known for statements such as “If it wasn’t good for you it wouldn’t be enhancement. In terms of human functioning an enhancement is by definition an improvement on what went
20 Ibid. 130 21 Ibid. 133–34 22 Ibid. 135–39
23 McGee, “Using the therapy and enhancement distinction, 70–71 24 Ibid. 71
before.”26 According to Harris, everything that aims to improve is therefore an enhancement, and this argument will be discussed further in section 4.1. In extension, this basically means that every type of intervention, even medical, is an enhancement, which McGee argues is a faulty way of looking at these concepts: “To put the point another way: only by already first defining enhancement broadly (so as to include therapy) can Harris then go on to deny the difference between therapy and enhancement. He thereby succeeds in questioning the distinction only by stipulative fiat.”27 Only in defining enhancement as “everything that improves” can Harris deny the difference between treatment and enhancement.
This line of argumentation from McGee is complicated by the fact that descriptively speaking, all treatments are enhancements since they aim to improve a sick individual in respect to “normal function” so that in theory, all GGE could be considered a type of enhancement.28 The difference, I would argue, is that not all enhancements will be made for health-related purposes. In cases of diseases related to old age, however, the treatment/enhancement distinction could be harder to apply, McGee admits, but this doesn’t imply that its useless, rather that in this case it is harder to apply and might need to be revised.29 But as we have seen in previous sections, edits made to prevent diseases of old age to extend our life
expectancy will probably remain a borderline case that simply falls in the middle of the vague treatment/enhancement distinction. Even if we decided to use CRISPR for these types of diseases and claimed that the edits made are clearly treatments, the distinction would not help us to decide which diagnoses to edit out and not. In opposition to this, one could argue that we should simply edit out all diseases related to old age, since every edit made to prevent a disease is a type of treatment. But then it would become harder to claim that this is strictly “treatment” purposes since most of us would consider an ageing process free of diseases an enhancement, since diseases of old age are very much a part of “normal function”. As such it would once again blur the line between treatment and enhancement, the distinction therefore fails to meet the first criteria needed for it not to be vague: to draw a sharp line between treatment and enhancement so we can know which is which. More importantly – the distinction still fails to be action guiding.
26 John Harris, “Enhancements Are a Moral Obligation” in “Human Enhancement” Eds. Julian Savulescu, & Nick Bostrom, (Oxford: Oxford University Press 2009), 131
27 McGee, “Using the therapy and enhancement distinction,” 72 28 Baylis, Altered Inheritance, 61–4
McGee also criticises the argument from Harris and Buchanan that we already accept enhancements in our daily life by, for example, wearing glasses for better vision or using smartphones for communication. Buchanan even claims that literacy is a ‘fantastic cognitive enhancement’. This view, McGee argues, fails to distinguish between us and our tools as well as neglecting the fact that an “enhancement” such as learning how to read does not change or human nature nor alter our genetic makeup in the way that genetic enhancement would.30 This is a point well made since we can always take off the glasses we use for reading but we
cannot reverse or remove genetically enhanced vision. The idea that “we already use enhancements” fail to recognise the fact that genetic enhancements are permanent,
irreversible and most importantly: heritable. We might have full morphological freedom over our own bodies but it is less clear if we have full morphological freedom in respect to our future children and the generations to come. Some would argue that this raises concerns of consent, that the moral issue would circle around the lack of consent from the (future) people that would be directly affected by the edits made. But I think this argument lacks the power needed to be convincing, because arguably, none of us consent to anything concerning the genetic makeup we are born with: with or without GGE we cannot choose our parents. More importantly, we would lack this consent for all edits made, whether for treatment or
enhancement purposes. This supports the argument of Missing the Point because once again, the treatment/enhancement distinction fails to correspond to what is genuinely morally problematic and not.
The accounts presented here by Juengst and McGee are defences for the concepts and definitions of “treatment” and “enhancement” which imply that these arguments rely on the assumption that there is a natural or conceptual difference between treatment and
enhancement that, if interpreted correctly, could be given moral significance. Juengst and McGee would probably argue against my thesis, that the line between treatment and enhancement is too vague to be ethically or normatively guiding, that if the concepts and definitions were correctly defined the problem of vagueness would go away. The problem of drawing the line, for McGee and Juengst, would be an epistemic one or possibly a semantic issue. I agree that the problem is probably a combination of a semantic and an epistemic issue but none the less the vagueness of the distinction will always cause borderline cases such as
diseases of old age – and prevention against these types of diseases will neither be correctly classified as treatment or enhancement.
An obvious shortcoming with the distinctions outlined here are that they inevitably rely on an idea of what is a disease and not, but what counts as a “disease entity” or as an illness might partially be a social construction. As such they are changed and reshaped over time, as societal norms and views grow and adapt. For example, the World Health Organization (WHO) defined homosexuality as a disease up until 1990 and even today their list of diseases keeps being reviewed and updated.31 Even our definitions of “normal” are socially
constructed and not everyone views a genetic condition such as deafness or blindness as a deficit or deficiency worth editing out.32 Another example is intelligence, today an IQ bellow 70 is considered as a handicap or disability. But why is that? Probably because an IQ over 70 is what is required for us to work and function autonomously in today’s society. As Tännsjö puts it: “Depending on what society requires of people, we draw a line between what is normal and healthy and what is abnormal and dysfunctional.”33 What is normal is therefore decided in respect to what society requires of us. If we applied these examples to the
treatment/enhancement distinction, we could imagine that it would be less problematic to edit out a genetic predisposition for homosexuality (if we imagine that sexuality is genetically predetermined) if we still viewed homosexuality as a disease rather than a naturally occurring variation of what it means to be a human. This shows that simply being classified as a disease does not automatically imply that we should genetically edit this trait out, or that these edits would automatically be considered “ethically unproblematic”. This is highly problematic for the treatment/enhancement distinction because it shows how the distinction fails to meet the second feature needed for it not to be considered vague: the distinction must correspond with what is genuinely morally problematic and not. But if simply employing a slightly more tolerant view of what it means to be “human” and “normal” we shift what is a disease and not, this complicates the view that GGE made for treatment purposes would automatically be considered ethically unproblematic. This, again, supports The Argument of Missing the Point and seen this way, the treatment/enhancement distinction remains vague and arbitrary.
31 World Health Organisation, News and Events: International Day against Homophobia, Transphobia and Biophobia: https://www.who.int/life-course/news/events/intl-day-against-homophobia/en/
32 Baylis, Altered Inheritance, 69
3.2 Unwanted Consequences and Human Nature
The arguments presented in this section might be called “consequentialist arguments” since they operate on the assumption that a line should be drawn between treatment and
enhancement with the moral motivation of avoiding certain worst-case-scenarios. The line is being drawn to avoid certain unwanted consequences. A highly influential argument in favour for a distinction between acceptable and unacceptable genetic intervention is from Jürgen Habermas, as he argues that GGE would represent an intrumentalization of human nature which could change our ethical self-understanding as a species.34 He argues that a vast amount of our self-understanding and ethics rely on our identification as human beings, grounded in a belief in human dignity. The manipulation of the human genome risks not only to violate these values but to blur the distinctions between the grown and the made. In
participating in this process, we risk to obliterate the boundary between persons and things, opening for unexpected and unwanted consequences for ourselves, our interpersonal
relationships and the society we live in.35 The idea is that enhancements violate human nature and therefore these edits can be distinguished from treatments, interventions that do not risk to alter or violate our human nature. This assumption will be discussed further in section 4.2.
“The problem, of course, is not genetic engineering, but the mode and scope of its use.”36 Again, the tool itself isn’t the problem but rather how we use it and for what purpose. This, Habermas argues, leads to the problem of drawing a line between gene edits we consider justified and those we consider problematic. But once again this presents us with a paradoxical challenge, because the line between the two is vague both on conceptual and practical grounds. So, in a dimension where boundaries are fluid and vague we are supposed to draw and inforce particularly clear-cut lines. Therefore, the problem with drawing the line between treatment and enhancement, to Habermas, is twofold and paradoxical. Because on the one hand the line will remain vague and fluid, but on the other hand, if we refuse to accept a distinction between treatment and enhancement, we will leave the choice of what
interventions can be accepted and practiced to the liberal capitalist marketplace.37
34 Jürgen Habermas, The Future of Human Nature (Cambridge: Polity Press & Blackwell Publishing Ltd, 2003), 37–47
Another highly influential argument is brought forward by Michael J. Sandel who believes the main problem with enhancements and genetic engineering is not that they undermine our humanity or that it erodes human agency, as Habermas believes, but rather: “The deeper danger is that they represent a type of hyperagency – a Promethean aspiration to remake nature, including human nature, to serve our purposes and satisfy our desires. The problem is not the drift to mechanism but the drive to mastery.”38 Sandel believes this drive to mastery represents a failure to appreciate the giftedness of life, to take some things as given rather that to shape everything into our liking. In extension, this means that bioengineering and genetic enhancement threaten the relationship between parents and children because: “To appreciate children as gifts is to accept them as they come, not as objects of our design or products of our will or instruments of our ambition.”39 This, however does not imply that parents cannot do anything to prevent illness or disease in their future children, because medical intervention with the purpose of preventing illness or restoring the injured to health does not violate nature or limit the future child’s possibility to flourish. The problem, essentially, is not genetic intervention with the purpose of treating disease but rather the hubris that comes with
“designing parents” will to enhance their future children. This, then, is how Sandel draws the line between enhancement and treatment, as he argues that medical intervention with the purpose of curing or preventing illness does not desecrate nature but rather honours it as it does not override the child’s natural capacities but rather allow them to flourish.40 While enhancement is morally impermissible because it represents an unhealthy strive towards hyperagency, a drive to mastery and a failure to appreciate life, and our children, as a gift.
On the discussion of whether GGE violates human nature or represents a drive to mastery Mark Sagoff brings forward an interesting point of view in examining the concept of “nature” and “human nature” in the debate on germline intervention and gene editing.41 Similarly to Juengst, he argues that the conclusions drawn from these types of arguments rely on our view of nature and what is “natural” as well as on our view of medicine.42 But the conclusion he
38 Michael J. Sandel, “The Case Against Perfection: What’s Wrong With Designer Children, Bionic Athletes, and Genetic Engineering” in “Human Enhancement” Eds. Julian Savulescu and Nick Bostrom, (Oxford: Oxford University Press, 2009) 78
39 Ibid. 79 40 Ibid. 80
41 Mark Sagoff, “Nature and Human Nature” in Is Human Nature Obsolete? Genetics,
Bioengineering, and the Future of the Human Condition, Eds. Harold W. Ballie and Timothy K. Casey, (Cambridge, Massachusetts: The MIT Press, 2004), 69
draws is slightly different from Juengst, Habermas and Sandel since he argues that the
problem with GGE is not that we are becoming alienated from our human nature or from what is naturally given, but rather that we are becoming increasingly responsible for it. The good thing about the genetic makeup we inherit, even if full of defects and diseases, is that it is no one’s responsibility. But with GGE, our genetic makeup is becoming a matter of choice rather than chance and in extension we become responsible for the choices we make.43 Habermas make a similar reflection on the connection between moving from chance to choice and control when it comes to our genetic makeup and how consequentially this makes us, as humans and parents in particular, responsible for our children’s genetic conditions. He believes, similarly to Sandel, that in giving the parents the possibility to choose between desirable and undesirable genetic traits as if this was a product of their choosing, the parents are exercising a type of control that, inevitably, they can later be held responsible for by their future children.44 Genetic modifications would in this way shift the line between chance and choice.45 Besides from the possibility that enhancements violate values such as “human nature” and “human dignity” I would argue that there is this obvious move from what might be called the “genetic lottery of chance” to parents choosing the genetic traits of their children giving them the control over their genetic makeup and as a consequence, making them
responsible for it. The unfolding of this process being something in the line of:
Chance Choice Control Responsibility
I call this the Peter Parker principle of CRISPR and GGE, because “With great power comes great responsibility” – with the immense power of editing DNA we are increasingly
becoming responsible for the edits we make and for the edits we choose not to make. Making us, in extension, responsible for the genetic makeup of our species. This, if we follow the line of argumentation of Habermas, Sandel and Sagoff, is a consequence of GGE that we might not be ready for. More importantly, I would argue that this type of responsibility applies independently of whether the edits were made with treatment or enhancement purposes. So, once again the treatment/enhancement distinction fails to be action-guiding and to correspond with what is genuinely morally problematic and not. Because if the moral issue concerning GGE boils down to the fact that we become responsible for the genetic makeup of our
43 Ibid. 90
offspring then this responsibility applies independently of whether things fall one the treatment or the enhancement side of the distinction. In opposition to this, one could argue that we are already responsible for at least half the genetic makeup of our children (counting that they are 50% genes from each parent). But I would argue that the active choice that CRISPR would give us in respect to the genetic makeup of our children does present a new type of responsibility: both for the edits we choose to make and the edits we choose not to make. And that, once again, this supports The Argument of Missing the Point since the responsibility applies independently of the treatment/enhancement distinction.
3.3 Eugenics and Disability Rights
Besides from the objection that enhancements might violate values such as “human nature” and “human dignity” in a way that treatment doesn’t, some claim that enhancements are morally impermissible because it’s eugenics. These arguments stem from the idea that it might be possible to draw a line between what is morally permissible and impermissible in an exact place, together with the assumption that morals and ethics should concern given and exact limits against impermissible actions. There might be practices and actions, that we as moral agents, are never allowed to do. The choosing between what we consider “lives worth living” against “lives not worth living” that some believe CRISPR represents, could be such a line that we are not allowed to cross no matter what. This is another way of drawing the treatment/enhancement distinction, since treatments would aim to cure diseases (not
considered problematic) but enhancements would be considered eugenics: aimed at improving the human genepool by choosing the genes we consider desirable and thereby choosing which lives are worth living and not. However, it is worth noticing that some argue for “less good lives” rather than simply “lives not worth living”.
The term “eugenics” was coined in 1883 and it means “well-born”. Eugenic thought is
grounded in an idea of improving the human gene pool and the shadow that eugenic practices cast over GGE and enhancement makes some argue that genetic selection through PGD and gene editing with the help of CRISPR would be nothing more than privatized or free-market eugenics.46 Those who defend enhancement practices argue that if the state coercion is removed and if the genetic choice is freely chosen then eugenics would no longer be morally impermissible. Some political philosophers argue that a moral distinction could be drawn
between the old eugenics practiced through state coercion and a “liberal eugenics” allowing for genetic enhancements that does not restrict the autonomy of the future child.47 In a well cited passage from Nicholas Agar he states that: “While old-fashioned authoritarian eugenics sought to produce citizens out of a single centrally designed mould, the distinguishing mark of the new liberal eugenics is state neutrality.”48 What sets “liberal eugenics” apart and what makes it morally permissible is the fact that it is not state imposed and the choice if freely made.
The idea that state coercion could be removed from eugenic practices is examined by Susumu Shimazono who analyses the Japanese resistance against genetic selection through prenatal diagnosis and selective abortions. In today’s Japan pre-implantation genetic diagnosis (PGD) of test-tube fertilized eggs (IVF) is not publicly permitted, nor is selective abortions due to prenatal genetic diagnosis. As a result, in 1996 less than 10% of the foetuses with Down’s syndrome were aborted, while 50% of foetuses with Down’s syndrome were aborted in France and Britain the same year.49 The selective abortions of foetuses with Down’s
syndrome in western countries, including the Unities States, were as high as 90% in the mid-nineties even if the exact number might vary slightly between countries.50 It is important to note that abortions are fully legal in Japan but they are more hesitant in regards to selective abortions than western countries. Shimazono argues that if society and doctors compel pregnant women to undergo screenings, examinations and diagnosis of the foetuses, this represents a will to intentionally manipulate women’s self-determination and freedom of choice when it comes to selective abortions.51 State coercion can be practiced in many ways and the fact that women are pushed to undergo screenings and testing might have a negative effect on the autonomy of their choice to proceed with the pregnancy or not. In extension, this type of societal manipulation represents a discriminatory attitude against disorders and
handicaps since the desire for a “healthy child” is being shifted into a belief that “a child must be born healthy”.52 Shimazono also argues that in legislating that a foetus can be aborted
47 Ibid. 84
48 Habermas, The Future of Human Nature, 48–9 & Sandel, “The Case Against Perfection,” 85
49 Susumu Shimazono, “Reasons Against the Selection of Life: From Japan’s Experience of Prenatal Genetic Diagnosis” in Human Enhancement, eds. Julian Savulescu and Nick Bostrom (Oxford: Oxford University Press, 2009), 292–3
50 Bostrom and Savulescu, “Introduction”, 12
because of a genetic disorder we are stipulating in law that the value of the life of disabled people is less than that of non-disabled people, thus ensuring a discriminatory judgement by law. From this Shimazono concludes that it is becoming “natural” to prevent disabled people from being born simply because society has created the idea that they should be excluded: and this idea is grounded in eugenic thought.53
Philosophers such as Shimazono obviously draw the line between treatment/enhancement and acceptable/unacceptable practices of genetic selection in a much more conservative and harsher way than many other ethicists but it is important and interesting because it highlights the fact that state coercion might not be the only thing making the “old eugenics” morally impermissible and that the “state neutrality” that Agar believes is the “distinguishing mark of the new liberal eugenics” might not be so neutral after all. Coercion and state pressure can be practiced in many ways. I am not arguing against PGD and prenatal diagnosis but rather arguing against the claim that PGD does not lead to eugenic practices. It´s often claimed that “The experience of PGD shows that it is possible to limit a method to its medically indicated use through weighed policy-making and legal regulations”54, but this is not fully true. If some 90% of foetuses effected by Down’s syndrome are aborted, then we are in fact choosing between lives considered worth living and not – which should be considered beyond simply “medical use”.
In Denmark for example, all women regardless of age and risk factors are offered prenatal screenings since 2004. Almost all expecting mothers choose to take the test and as a result, more than 95% of foetuses with Down’s syndrome are aborted. In 2019 only 18 children were born with Down’s syndrome in all of Denmark. This is important because Down’s syndrome is frequently referred to as the “canary in the coal mine” of selective reproduction since it is among the least severe genetic conditions to be routinely screened for and more importantly, it is highly compatible with a long and happy life.55 Even Tännsjö talks about Down’s
53 Ibid. 298
54 Giovanni Rubeis, “Human Germline Genome Editing in the Clinical Context: The Case of Disease Prevention” in Between Moral Hazard and Legal Uncertainty: Ethical, Legal and Societal Challenges of Human Genome Editing eds. Matthias Braun, Hannah Schickl and Peter Dabrock (Wiesbaden: Springer Fachmedien Weisbaden GmbH, part of Springer Nature 2018), 154
55 Sarah Zhang, “The Last Children of Downs Syndrome” The Atlantic, December 2020 Issue,
syndrome as a “dying culture” as he believes we are moving towards a society where Down’s will be either extremely unusual or completely eradicated. However, he doesn’t think there is anything wrong about this group ceasing to be among us, we might regret that some diversity is lost but he rather believes this is the natural course of society: “We will have other
minorities in their place. It´s sad for those who are the last of a group, but there is basically nothing tragic in one minority replacing another.”56
It is important to clarify that I´m not saying that Down’s syndrome isn’t a genetic disease but it might be worth to keep Hume’s law in mind: simply stating that something is a genetic disease doesn’t imply that we ought to edit it out. The point here is that plenty of people with Down’s syndrome live long, fulfilling and happy lives and as such this is vastly different from someone born with, say, Huntington’s disease or cystic fibrosis which usually entails a short life in perpetual suffering. Editing with the purpose to eliminate Down’s syndrome might be another borderline case because only by first classifying Down’s as an undesirable genetic disease can we then argue that these edits are made for treatment purposes. But these edits could just as easily be considered enhancements because there is little or no evidence to support the idea that people with Down’s live less happy or fulfilling lives than any other person. The wellbeing of children affected by Down’s syndrome might depend more on how tolerant of a society they live in rather than whether we edit against this diagnosis or not. As in the case of homosexuality, mentioned in section 3.1 it is possible that we would see edits against Down’s as enhancement rather than treatment if we simply employed a more tolerant view of what it means to be human.
I would argue that this shows how the treatment/enhancement distinction remains vague because Down’s clearly has a genetic cause but if we viewed it as a normal variation of what it means to be human and not as a “undesirable defect” edits of this kind would count as enhancement and not as treatment. Once again, the distinction struggles in drawing a sharp line between treatment and enhancement because simply in employing a wider view of what it means to be human, with all our defects, variations and diversity, we shift the line between the two. In extension, there is still nothing to indicate that edits made for treatment purposes
56 Eric Hilmersson, “Downs syndrome på väg att försvinna” Göteborgs Posten, 28 December 2020. *Quote translated by author Ellen M. Svensson
would automatically be considered ethically unproblematic. More importantly, drawing the line at eugenics would also have implications on practices we already accept and allow such as embryo selection through PGD and prenatal screenings, this will be discussed further in section 4.3. This supports The Argument of Missing the Point as the distinction also fails on the second feature needed for it not to be vague: to correspond to what is genuinely morally problematic and not.
This concludes section 3, in the next section I bring forward some of the arguments usually invoked to undermine or deny the treatment/enhancement distinction.
4. Arguments Against a Distinction
Just as there are many different reasons given in favour for a distinction between treatment and enhancement there are different reasons against such a distinction. In this section, I bring forward arguments from ethicists that deny a demarcation line between
treatment/enhancement as well as highlighting some of the arguments that speak directly against the arguments brought forward in section 3. It is worth noticing however, that not all accounts in this section are arguments in favour for enhancement but rather, they illustrate the difficulty of supporting the line illustrated in section 3 as well as the unwanted implications that such a line might have.
4.1 Limited Utility
In this section I bring forwards the arguments against the treatment/enhancement on the grounds that the distinction is of little or limited utility. John Harris, mentioned in section 3.1 as McGee criticises his broad interpretation of enhancement, argue that enhancements are a moral obligation since: “Enhancements are so obviously good for us (if they weren’t they wouldn’t be enhancements) that it is odd that the idea of enhancements has caused and still occasions so much suspicion, fear and outright hostility.”57 Again, enhancements are defined as everything that is good for us and that improves our condition so they cannot, by definition, be bad or invoke controversy. Harris believes the distinction between treatment and
enhancement is of little utility since the boundaries between treatment/enhancement and therapy/enhancement are not precise nor mutually exclusive. As such, they cannot be coherently or consistently maintained.58 The core of this argument relies on a critique of
Daniels definition of “normal species functioning” as Harris argue that this definition is of little use since both therapy and enhancement serve the same purpose and strive towards the same goal. “The overwhelming moral imperative for both therapy and enhancement is to prevent harm and confer benefit. Bathed in that moral light, it is unimportant whether the protection or benefit conferred is classified as enhancement or improvement, protection or therapy.”59 It doesn’t matter if CRISPR is used with the intention of treatment or
enhancement because both strive towards the same goal and moral imperative: to prevent suffering and create an improved condition compared to what we would have had without the intervention.
As we have seen in previous sections, the idea of “normal function” is a continuous problem for the treatment/enhancement distinction and I would argue that Harris is correct in that both edits made for treatment and enhancement, seen in this way, strive towards the same purpose: to improve our life quality and prevent harm. On the other hand, Harris might be right that descriptively treatment and enhancement strive towards the same moral imperative but intuitively there does seem to be a conceptual difference between extending our life
expectancy by one year compared to extending it to 600 years. It also seems incorrect to argue that both GGE with the intent to prevent someone from being born with cystic fibrosis and GGE with the purpose of choosing, for example, skin or eye colour could be bathed in “the same moral light”. There is another complicating factor to this argument, that Harris seems to neglect, which is the question of “for the benefit of whom the edits are made?”. For the benefit of the parents? For the benefit of the future child? For the benefit of society? As we have seen in section 3.3 edits made to avoid Down’s syndrome, for example, seems to be a type of edit made with the purpose of conferring benefits to society or possibly the parents rather than conferring benefits to the future child, since a diagnosis like Down’s syndrome is fully compatible with a long and happy life. But edits made to avoid a future child being born with cystic fibrosis rather seems to be with the intention of benefiting the future child by preventing them from a life with a chronic, incurable and painful disease. Can we then say that both edits to avoid Down’s syndrome and edits made to avoid cystic fibrosis can be bathed in “the same moral light?”. I would argue that we cannot.
This discussion is further complicated by the possibility to, for example, edit with the purpose of changing skin colour. Imagine a scenario where two black parents are worried about their future child being a victim of racism. They know that black people, especially in America, have a higher statistical risk of being convicted of a crime and of being victims of police brutality so they decide to genetically edit the skin colour of their future child. Can we argue that this is the same as editing out the gene that causes Huntington’s disease? I would argue that they are edits made with widely different intentions. Because the parents who genetically change skin colour of their future child do so to avoid a structural and societal problem: something that could be solved with social reform and progress. Whereas the parents who chose to edit out the gene that causes Huntington’s disease do so to avoid a disease that there is currently no cure for: the suffering caused by Huntington’s disease cannot be avoided by changing the structures of society. To argue that “everything is an enhancement because enhancements make everything better” is a faulty shortcut through the moral concerns that GGE raise but at the same time, it is a shortcut that the treatment/enhancement distinction enables. Because not only does the distinction struggle to divide between the two concepts but the distinction also fails to correspond to what is genuinely morally problematic and not. Once again the treatment/enhancement distinction fails to indicate anything about how these different intentions, benefits and reasons should be evaluated or weighted against each other.
A similar critique of the utility of the treatment/enhancement distinction is brought forward by Bryan Cwik who argue that: “Even if we had a clear therapy/enhancement distinction, this would do little to shed light on some very difficult problems regarding therapeutic uses of GGE.”60 His point is rather that as of current applications and developments of GGE, the prospect of enhancement is not only technically far off but it would require a different type of use and application of the technique than is currently being developed. Cwik also argues, more importantly, that even the applications of GGE technologies such as CRISPR for what is clearly therapeutic or treatment purposes present significant ethical questions for their clinical use. Simply being classified as “treatment” might not be enough to give a moral justification for their application. An argument I fully agree with since the treatment/enhancement
distinction fails to correspond to what is genuinely morally problematic and not. Because of this, Cwik believes that “…issues about enhancement and eugenics are, for the foreseeable
future, likely a red herring.”61 That is, discussions of enhancement and eugenics lead astray from the much more relevant and important subject of where to actually draw the line between acceptable and unacceptable uses of CRISPR. Cwik suggest that possibly the line could be drawn at introducing novel genes into an otherwise healthy individual, something that might be impermissible no matter what. This might be a more helpful way of defining radical enhancement: to genetically modify an otherwise healthy individual by possibly introducing “novel genes” – even genes from another species. If so, this definition could enable a reasonable demarcation line against the slippery slope towards radical enhancement and eugenic practices. But on the other hand, we would still be presented with the problem of defining what it means to be an “otherwise healthy individual”.
Another suggestion could be to draw the line at edits to one gene, which has many potential clinical uses but presents less of a medical risk than multiple gene edits.62 As we stand today, GGE still represents a significant risk for off-target effects (edits made in the wrong place) as well as unwanted on-target effects (edits made in the right place that cause unwanted and harmful consequences) and mosaicism. Mosaicism is a form of incomplete editing resulting in a scrambled sequence of DNA which leads to an embryo with both edited and unedited cells, this reportedly happened to one of the CRISPR-twins created by Jiankui He in 2018.63 These unwanted side-effects open for the possibility of what Françoise Baylis call
“multigenerational harm” since the default edits are heritable and will be passed down to any future offspring and generations.64 Because of these risks, Cwik concludes that: “Given the current state of gene editing technology, it is hard to see multiple gene edits in human
embryos being a safe and effective clinical possibility in the foreseeable future, so the de facto line of permissible use of gene editing will be drawn at single gene edits anyway.”65 Because of technical difficulties and medical risks, the actual line between acceptable/unacceptable uses of CRISPR might be drawn at single gene edits in any case.
61 Ibid. 696 62 Ibid. 703–4
63 Baylis, Altered Inheritance, 90–91 64 Ibid. 91
4.2 What “Human Nature”?!
As we have seen in section 3.2, ethicists such as Habermas, Sandel, Sagoff and many others discuss and invoke the notion of “human nature” in the debate of GGE through CRISPR as a possible way to draw a line between treatment and enhancement. The idea being that
enhancements violate human nature in a way that treatment doesn’t and therefore one can be distinguished from the other. However, there is no consensus whatsoever on whether human nature has inherent value or solely derivative normative status. There is not even consensus on how the term should be defined or whether there even exists such an entity.66 In extension, the term “human nature” can be used in both a strictly empirical sense; that is to describe the physical and psychological traits of human beings and in a non-empirical sense; that is in a metaphysical sense. In the metaphysical sense “human nature” is usually used to describe some “universal human essence” to distinguish human beings from other living beings and animals. To do this, philosophers have to identify some unique character that human beings possess: “The most promising candidates are probably the capacities for reason, autonomy and morality... or as beings with a special moral status and moral capacity.”67 The difficulty with these accounts lies in somehow proving not only that something like “human nature” exists, but also to prove that it is unique and has some type of value (inherent or derivative). Because of this, some argue that we should refrain from any use of the term “human nature” in the debate concerning GGE since it offers no ethical guidance; “There’s no such thing as a unicorn, and, some have said, there’s no such thing as human nature, either.”68 The idea of “human nature” is simply a myth. The problem, one could argue, lies not only in the
ontological difficulty of proving that something like “human nature” exists but also in proving that there is something morally impermissible about changing it. History and evolution rather proves the point that our “human nature” is in constant change. In opposition to this some argue that the ethical issues with GGE lies in altering our genetic makeup to the extent that we cease to be human.69 The idea being that if enough of our DNA is changed, repaired and
66 Nadia Primc, “Germline Modification as a Severe Intervention into Human Nature” in “Between Moral Hazard and Legal Uncertainty: Ethical, Legal and Societal Challenges of Human Genome Editing” eds. Matthias Braun, Hannah Schickl and Peter Dabrock
(Wiesbaden: Springer Fachmedien Weisbaden GmbH, part of Springer Nature, 2018) 100–1 67 Ibid. 101–2
68 Tim Lewens, “Human Nature: The Very Idea,” Philosophy and Technology, vol. 25 (Springer-Verlag, 2012): 459
