A Completely Non-Partisan Policy Conundrum

In an effort to provide an occasion for political discussion that engages our political and moral views but not our partisan passions, I offer the following policy question, which I and some colleagues have written about (registration required). (A longer relative of this paper with more biology and a less developed moral argument is here.) It's about embryonic stem cells, but nothing about the moral question we raise turns on whether or not it's OK to use human embryos to derive stem cells. It also assumes that we have figured out how to use stem cells to cure diseases.

As I said in the last discussion of stem cells, any treatment with stem cells will involve transplanting stem cells into a person's body. Like any other transplant, therefore, there will need to be a good genetic match between the stem cells and their recipient; otherwise, the recipient's immune system will reject the stem cells. In popular discussions of embryonic stem cells, a lot of attention has been given to the possibility of creating stem cell lines with their recipients' own DNA, thereby creating a perfect match and avoiding all problems with immune rejection. But for the foreseeable future, this will be too expensive to be generally useful; moreover, this approach will never work in treating any disease that must be treated as quickly as possible, since it will always be quicker to use an existing line of stem cells than to create one from scratch. For this reason, we're likely to need a bank of stem cell lines that we can use to treat diseases.

While stem cells are like other transplants in requiring a good genetic match between the transplanted material and its recipient, however, they are unlike other transplants in two important respects. First, in order to for your organs to become available for transplant, you typically have to either die or undergo a horrible operation. By contrast, to donate a stem cell line you need only either donate your excess embryos from IVF treatment or (if we're using therapeutic cloning) let someone swab a few cells from you. Donation is thus much less arduous. Second, solid organs can be used only once: put a heart in someone and you no longer have that heart available for transplant. But stem cell lines can be used over and over: take some cells from the line and transplant them, and you still have the original line and can grow more. For these reasons, while the demand for solid organs nearly always outstrips supply, and thus people who run organ transplant networks have to work with whatever they can get, the people who design stem cell banks will probably have the ability to decide which stem cell lines their bank will include.

So how should they answer this question? The most obvious answer is: they should include the lines that will give the greatest number of people the ability to find a match, and thus get stem cell therapies. (What sense would it make to include only lines that have an incredibly rare genetic profile, and would be usable by only a few people?) But there is one problem with this approach. The particular genes that are relevant to matching come in a (large) number of different versions. Suppose you made a chart showing which versions are most common in a given group of people. The chart for men would look similar to the chart for women; the chart for right-handed people would look very similar to the chart for left-handed people; but unfortunately, the charts for different ethnic groups would look very different. Specifically, if you drew charts for people whose ancestors were East Asian, Sub-Saharan African, Caucasian, and Native American (including indigenous South Americans), their charts would look very, very different. (The likely reason for this is that the relevant genes are important to the immune system, and they have evolved to meet the diseases people tend to encounter. Since people who live in very different parts of the world tend to encounter different diseases, they also tend to have different versions of these genes.)

What this means, in practice, is that a cell line that is useful to a high proportion of whites is likely to be useful to only a small number of people of East Asian, Sub-Saharan African, or Native American descent, and vice versa. The fact that some people are of mixed descent helps a bit, but not enough to alter the basic problem, which is the following: suppose, for the sake of argument, that within each ethnic group, the most common versions of the relevant genes give the same percentage of people in that group a match. Obviously, a given percentage of a large number is bigger than the same percentage of a smaller number. In this country, there are many more Caucasians than members of any other ethnic group. For this reason, a stem cell line with the version of the relevant genes that is most common among Caucasians will cover a much larger number of people than a line with the version most common among any other ethnic group.

If we try to design a stem cell bank to cover as many people as possible, therefore, we will pick a line with the version of the relevant genes that's most common among Caucasians. Moreover, our second choice will not be a line with the version most common among some other ethnic group; it will be a line with the version of the relevant genes that's the second most common among Caucasians. We will have to go pretty far down the list of 'best for Caucasians' before our best choice, by the numbers, is a line with the version of the relevant genes that's most common among any other ethnic group. For this reason, if we try to design a stem cell bank that ill help the greatest number of people, our bank is likely to be overwhelmingly more favorable to Caucasians than to any other ethnic group. We ran the numbers for one version (technical details available in the articles linked), and found that it would provide matches for 40% of whites, 7.8% of blacks, and 3.6% of Asians. That is a huge difference, and it is due to the fact that the 25 most common versions of the relevant genes are the 25 versions most common to whites. (We haven't yet gotten to the point where the most common version for even one other ethnic group is included.)

So here's the question: should we design such a bank to provide the greatest number of people with a match, even though in practice this bank will be much, much more favorable to whites than to members of any other ethnic groups? Or should we try to provide members of all ethnic groups with more or less equal probabilities of finding a match, even though that will mean covering fewer people overall?

Bear in mind a few things. First, the inequality of the first option isn't the result of racism; it's the result of population genetics and the demographics of the U.S. Second, no one is entitled to stem cell therapies: it's not as though, if we choose the second option, we're depriving people of something they have a right to. (Or if we are, this is not obvious.) What do you think we should do?

Comments

[Bernard Yomtov]

It is not clear to me what you mean to imply by saying "no one is entitled to stem cell therapies."

Does that mean people have no claim to these therapies beyond their ability to pay - that there are no moral claims, just economic ones? Then isn't the answer that you bank the lines with the greatest profit potential?

Now I don't agree that there would be no non-economic claims, for a number of reasons. So the question you pose translates, I think, into the question of either:

a. maximizing the probability that a randomly chosen person needing a transplant will be able to get one.

or

b. equalizing the probability across the population.

I don't know the answer, but is it fair to say, in line with the question of the basis of claims, that this may depend on the decision-maker - that the proper decision for the government might well be different than the proper decision for a private actor?

[hilzoy]

Bernard -- I didn't mean to imply that there are no moral claims. What I meant was something more like this: Suppose my two nephews (adorable, aged 5 and 7) have some cookies, and one of them eats all the cookies. This is wrong. But the way in which it's wrong depends on whose cookies they were. If they were all his cookies, then it would have been nice of him to share, but still, they are his cookies. If each of my nephews was given half the cookies and one ate all of them (without the other's permission), then he stole his brother's cookies. In the second case, the brother was entitled to the cookies: he owned them, and he should have gotten to eat them. In the first case, the brother was not entitled to the cookies: it would have been nice for the one who ate the cookies to share, and we can criticize him for not sharing, but he didn't take anything that anyone else was entitled to.

Likewise, in the stem cell case: if you paid for your insurance, and just when you needed to use it I came and (somehow) took it away and said, "I am going to use it to help several other people; you would just use it for yourself", I'm taking something you're entitled to. But if there's a government program and, because it was designed one way rather than another, it benefits a group of people that doesn't include you, I have not taken anything you're entitled to. I might be criticized on any number of other (moral) grounds, but not that one. This cuts both ways in the original problem.

You're right that it matters who designs the bank and pays for it. Suppose it's the government: what should it do?

[sidereal]

In this case, I believe competition is probably the best solution.

One of the fundamental techniques in market competition is identifying a disenfranchised or unserved demographic and marketing to them. I guarantee that in the model you've described above (40% whites, 7.8% blacks, 3.6% asian), if a private entity established a service exclusively targeting blacks, it would be highly competitive and therefore viable.

Another issue is regional variation. I might have missed it, but it's not clear to me what the incentives are for maintaining a smaller number of lines (as opposed to a huge bank of every possible line). If it's logistical, than providing different lines in different areas seems to be reasonable. A bank in Atlanta, for example, would have substantially different from a bank in Seattle, which in turn would be radically different from Orange County.

If the incentive is the moral one. . that we should only develop a few lines because each new one costs us a life, then I don't think there's any good solution. You have to do the most good per available slot, and I think the only rational measure for most good is the most potential people served.

[liberal japonicus]

Isn't there another dimension here, which would be the internationalization of research? A stem cell bank set up to maximize the usefulness to the population of the US would possibly stymie foreign researchers. This kind of thinking seems to make this research akin to military research, which can only be shared with allies and then only if they are absolutely trustworthy. One could take that line, but it seems contradictory to the way we want human history to go.

There are good and bad arguments to support this. A bad argument, (though one that I think is true) is that other countries with different social/moral climates will be able to do research in different ways. We see that already with cloning. A good argument is that the interconnectedness of researchers is an small but significant way that we can promote international exchange. A second good argument is that it attempts to recenter medical research on questions of global application. As we can see with drug company research or the harvesting of genetic biomaterial from third world countries, first world priorities will inevitably win out, so a stem cell bank should be composed as a reflection of the world's genetic makeup, not Americas, or we will continue to develop treatments that largely help us and not others. This may mean raising in prominence a line that would be much further down the list, but this happens all the time in constructing sampled models.

Furthermore, one would imagine that the stem cell bank would first be constructed for research, so in order to take advantage of serendipity, we would want as diverse a stem cell bank as possible, so we would already have the foundation (I don't know how easy or hard it is to keep stem cell lines pure) for a diverse bank, after which we would then supplement it with ones that are more statistically appropriate. At least this is my assumption on how things work.

[hilzoy]

Sidereal: It's unclear to me that competition will solve this, in the short run at least. The line that would make its creators the most money would (other things equal) be the one that could be used for the most people, and that has not already been created. Competition would therefore seem to lead to individual companies moving down the list of versions of the genes that cover the most people, which would, for the reasons given above, lead to a much greater proportion of whites finding a match.

LJ: I think there are reasons why, after setting up whatever sort of research bank would be best, we might enlarge it to serve our own citizens best. This wouldn't hold anyone else back, or prevent them from developing their own lines, or licensing ours if ours were useful to them. But I would think that if the government were to set something like this up, it would make sense for us to set it up for the benefit of US citizens.

It's also true, though, that an ancillary benefit of setting it up in such a way as to benefit people from different ethnic groups, we would create lines that would be of use to people in other countries, some of which (e.g., in Sub-Saharan Africa) would not be able to afford to do so themselves. How much, if any, weight you give to this ancillary benefit is a different question.

[Bernard Yomtov]

I think there is no simple rule but that it depends on the trade-off involved in a specific case.

I think my characterization of the choices is correct, so let me work with that. Suppose the US has only two ethnic groups which, purely for convenience, we label black (10%) and white (90%). Now we have a choice of banking lines (set x) which will benefit half of all whites and no blacks - 45% of the population, or lines (set y) which will benefit 40% of each group, hence obviously 40% of the population. Assume further that, if we choose set x, there is available no compensating benefit we could provide blacks.

Given this specific choice I would choose the second alternative. But if set y helped only 5% of each group I would favor the first choice. And of course we can lower the 5% to any arbitrary number, so unless we are adamant about equal probabilities there is some point at which we must favor set x.

Why not always? Because equal treatment counts for a lot. Even from a purely utilitarian point of view, it seems to me that there are significant externalities from equal treatment. Certainly there are negative externalities from unequal treatment. So I think that unless there is a huge disparity, as in my 45% - 5% case, we should generally prefer equal probabilities.

So I have no clear answer, just a leaning. And I don't know where to make the break either.

[liberal japonicus]

Thanks for the reply, Hilzoy. I'm not sure (even after reading the links) how the research operates. The second paper discusses that clinical application necessitates avoiding problems of HVA rejection and getting good matches. I don't know if matches are done on a simple percentage basis or if there are specific points that have to match, which I think would be an important consideration.

The paper also talks about the instrumental creation of stem cell lines, which some would scream is a euphemisim for harvesting embryos. I'm not sure if you are hinting at this in your post, but it is ironic that because fertility treatment is the domain of the relatively well-off, and, as the 2nd paper points out, African-Americans have a distrust of the medical establishment, so a status quo solution that avoids harvesting would end up reinforcing inequities in the delivery of health care (I could be wrong about how the clinical applications would work and thus, be wrong about this. Corrections gratefully accepted)

As one of those international type liberals, I think that the US government has, in some ways, a requirement/burden/calling to help the rest of the world (as do all first world countries and for-profit corporations that gain their money from the first world) so my 'leaning' is towards an approach that would aim to benefit as wide a range of people as possible, even if it has to do so by giving a short shrift to American citizens. I rationalize that out by suggesting that this kind of approach yields more benefits in the long run than an approach that seeks to maximize benefits for ourselves (bearing in mind that no one here is placing these points in such stark contrast) I don't think we can predict the course of stem cell research and taking a statistical model that, say, mirrors the US population rather than the world population may be closing off avenues that could make a huge difference. For example, given that the majority of people who lose limbs are in third world countries that have experienced war and have large numbers of mines, widespread clinical applications of would depend on having the correct stem cell lines for those countries. Similarly, the Vietnam conflict has produced a huge number of birth defects, as depleted uranium ordinance seems to have the potential to do. Perhaps stem cell treatment does not need immensely large numbers of patients to provide such clinical trials, but if it does, then these locations provide the most number of patients.

It would also depend on whether stem cell treatments are going to require a huge investment in other drugs and treatments to have them work. Again, I have no idea, but if we look at the disparity in AIDS treatment (and understanding that it is the inability to precisely identify the particulars of the AIDS virus and patient demands that drives much of the progress), and we imagine that stem cell treatment has a greater impact, I would be very uncomfortable seeing the gap in application that would arise in the future.

[Bernard Yomtov]

LJ,

Good point about international effects. But doesn't that just shift the debate to a different arena? Should we bank stem cell lines that potentially benefit a high percentage of Chinese, say, in preference to those that benefit a smaller number of people overall, but people who are widely distributed around the globe?

[hilzoy]

LJ: The short answer is: you could create a bank using either excess IVF embryos or embryos created to order by somatic cell nuclear transfer. The second would allow for a much better way of creating a stem cell bank, but raises moral problems of its own; thus my not having mentioned it.

Matching: your immune system could, in principle, check the entire genome of any cell introduced into the body. But that would be, let's say, cumbersome. So it chooses a quicker approach: different bits of your immune system check genes at certain sites, and if they recognizes those genes as 'yours', they don't attack. It's not quite a matter of percentages, both because not all your genes are involved and because the different attacks that can be triggered differ in ferocity.

The reason that creating embryos to order works best is this: you have two copies of each of the genes your immune system looks for. If we had stem cell lines whose two copies of each of these genes was the same, then if we transplanted this line into you and you had one copy of the relevant version of each gene, then this line would be a match for you even though your other copy of each gene is different. So this one cell line would match a lot more people than a line with different versions of each gene. And therefore it would be a lot more efficient to create a bank with these sorts of lines.

The problem is finding them. They are fairly rare. Using IVF embryos, one could screen an awful lot of embryos and hope to get lucky. But if we can create embryos to order, we can use DNA from adults who have already, independently, been identified as having this genetic profile.

Did that make sense?

[meaux]

How much more variety in the political persuasion of commenters would hilzoy have had if she had limited her discussion to adult and umbilical stem cells. the linked article notwithstanding? That way she could bring in those who disagree on the embryonic issue, and avoid the embryonic discussion in facor of this interesting redistributive justice problem.

For my part, assuming we have wide-open embryonic stem cell research, it is not clear to me that the lines themselves are that precious, and assuming we don't get into a big patent mess (which is a real possibility) letting the market - in the form of academia, private industry, and probably some NIH grants - decide seems the most efficient solution, and the most likely one.

[hilzoy]

Bernard: I basically agree with you about the sliding-scale nature of this. (You can also get similar results by moving other numbers around: in your example, I would not be profoundly troubled if the result of plain maximizing helped 50% of group A and 49% of group B.) But I'm not sure the problem is exactly that we're not granting different people equal treatment.

Suppose there were no differences between the proportions of different ethnic groups that were covered by the bank. In this case, we'd presumably say that even though some people, just in virtue of their genes, are not going to be covered, everyone was treated equally, since in deciding which lines to include we used a fair rule that doesn't discriminate against anyone. (I mean, if we can't cover everyone, at least we gave everyone a fair shot at it.)

One might ask: why doesn't the same principle hold in the case I described? We can't over everyone; we used a fair, nondiscriminatory rule, and some people, unfortunately, didn't luck out. Why does it make a difference that those unlucky people include almost all blacks? -- One might also ask: suppose that the relevant genes didn't sort along ethnic lines, but a way that meant that some group that has never been discriminated against was left out. (Say that we substitute 'right- and left-handed people' for 'whites', and 'the ambidextrous' for 'blacks', in your example.) Do you feel the same way in this case?

Just asking ;)

[Bernard Yomtov]

Hilzoy,

Good questions. Having had a nice dinner with several glasses of wine, I think the best thing is to sleep on it. My brain is not in turbo mode at the moment. I'll get back to you.

[hilzoy]

meaux: I wasn't trying to put anyone off, though I did worry that the topic might do that. (That's why I put in the note about the problem being independent of the morality of using human embryos.) The main reason I didn't use cord blood or adult stem cells as an example is that there are already cord blood banks and bone marrow donor registries (bone marrow being at present the main source of adult stem cells, along with cord blood), and they take the place of stem cell banks in these cases. Since banking cord blood, for instance, is a lot cheaper than establishing and maintaining stem cell lines, I don't think a stem cell bank will be needed here. -- Embryonic stem cells are different: the only alternative way of banking them would be to set up an embryo bank, which no one is suggesting, thank God.

[liberal japonicus]

Thanks hilzoy, yes it did make sense. It sounds like the research is much more advanced than I was aware, which is no surprise given the state of my knowledge about medicine. I still have to work out what my feelings are about it.

Bernard, yes, it does simply move it to a different arena but that arena opens up a lot of possiblities, many of which we might not be prepared to face. Hilzoy says that no one is suggesting setting up an embryo bank, but I would not be surprised if such a bank is not being contemplated somewhere in Asia, given the social norms concerning abortion and such, especially if the research is as advanced as it seems.

As for just ending up with a bank that represents more Chinese than Caucasians, I am thinking that a cell bank like this would not simply be for providing therapeutic treatment, but for also adding to the store of human knowledge. As such, a wider range is more appropriate, but again, it depends on if it is for learning more about how to use stem cells or if it for immediate therapeutic interventions.

[hilzoy]

LJ: to be clear about the state of the research: much of what I wrote in my last comment is just basic immunology applied to the (now hypothetical) case at hand. The actual research on using embryonic stem cells to develop therapies for disease is still in its early stages. This is partly because human embryonic stem cells were only isolated in 1998, and partly because the funding was not immediately available. (Note: I say this as a factual claim, not as a complaint. Some of the reasons for the delay -- e.g. the need to determine whether embryonic stem cell research was prohibited by the Congressional ban on destroying embryos for research purposes, the need to set up guidelines for funding, and the need to solicit grants and evaluate them, were clearly justifiable. Those who think Bush's policy is the right one will of course think that those delays that are due to it are also justifiable. Whether they are right or wrong is a separate claim from the one I am making, which is that, right or wrong, it created further delays.)

There have been very interesting results in experiments on animals, but the consensus seems to be that human therapies are a ways off. One needs, basically, to figure out how to do the following: (a) get the stem cells where they need to go, (b) get them to turn into the right sorts of cells, (c) get them to do whatever it is you want them to do, and (d) get them to do it in a well-regulated way. (E.g., if you're trying to cure diabetes, it's not enough to arrange things so that stem cells get to the right place, turn into the right cells, and start producing insulin. You also need to make sure that they produce the right amount of insulin at the right times; otherwise your patients might go into insulin shock, which is obviously not the desired result.) This is very difficult, and researchers are just getting started on figuring it out.

I just wanted to say this because it is, as Laura Bush said, important not to create false hopes that a cure is around the corner. On the other hand, this seems to me like a very bad reason not to continue the research. Which is, of course, not to say that there aren't other good reasons.

[Bernard Yomtov]

Hilzoy,

Good questions, but trappy.

(By the way, I just read your article. Could have saved myself some effort by doing the assigned reading ahead of time, I suppose. Old habits die hard).

Still, when I used the term "black" I was not thinking in terms of historical disadvantage. Rather, I was thinking in terms of an ex ante easily identifiable group. This could easily be left-handers, or ethnic Hungarians, or anybody. So no, that doesn't change my thinking by a lot.

It seems to me that there are two arguments for equalization:

1. A principle that says government should treat all citizens equally even at some cost in efficiency.

2. A utilitarian idea that says failure to treat all citizens equally will have negative consequences that outweigh the benefits of a pure maximization approach.

To the extent we rely on the second, the damage from excluding left-handers is much less than from excluding African-Americans, so we might adjust the sliding scale somewhat. But I wonder to what degree we are able to identify the probabilities of a line helping members of these sorts of groups - left-handers, bald men, blue-eyed people - with the same accuracy that we can identify the probabilities with respect to ethnic groups. Is it even likely that these things would usually matter?

In your paper you note that it is unwise to create a situation where minorities have reason to wonder whether their interests are being taken into account. I would add that it is also unwise to create a situation where some members of the majority group might conclude that it is acceptable not to take minority interests into account.

[Bernard Yomtov]

One further point occurs to me about these decisions. I suppose, though I don't know, that stem cell therapies will be more effective in treating some ailments than others. If so, then perhaps we should be talking in terms of the frequency of those ailments in various groups, rather than the portion of the population they represent.

Could this lead to an argument for the equalization strategy based on the relative poor health of disadvantaged groups? We have taken maximization to mean maximize matches. But another version of "maximization" approach is to maximize the benefit from a line, taking into account that some conditions are more severe than others, and that some conditions afflict more members of some ethnic groups than others. So if blacks suffer more often from treatable diseases than whites, then surely it makes sense to take this into account.

Take my previous example of matching 50% of whites or 40% of the whole population. Now add the possibility that 25% of blacks have treatable diseases while only 10% of whites do. Now the equalization strategy actually maximizes the number of people being treated. Bank lines for half the whites and you treat 4.5% of the people. Bank lines for 40% of everyone and you treat 4.6% (.4 X .1 X .9 + .4 X .25 X .1).

This seems to me to be an important consideration.

Sorry to double post, but this one is on a different idea, so I thought it OK to separate them.

[Sebastian Holsclaw]

"Embryonic stem cells are different: the only alternative way of banking them would be to set up an embryo bank, which no one is suggesting, thank God."

There is an interesting question about why the following three cases should be considered morally different as in the above sentence:

1. Banking IVF excess embryos

2. Creating stem cell lines from excess IVF embryos.

3. Engaging in IVF practices which you know will created embryos to be discarded.

4. Outright creation of embryos purely for the purpose of adding them to tranplant bank.

But that is almost certainly a thread-jacking.

As far as the direct topic of the thread, the competition model should do fine for creating stem cell lines once the standardized methodology comes about so long as the standardized methodology is not ridiculously expensive. Since it is effectively a cloning technique which will make it possible, I don't see the technique as likely to be very expensive in the medium-run. Furthermore the question of the bank seems unlikely to become a serious issue since the same advances in technology would make cloning for the purposes of embryonic transplant equally easy, and no worry about rejection needed whatsoever.

Another interesting sidenote is that this ought to provide an excellent test case for those who suggest that therapuetic research is likely under European research models without US support. If that claim is true, the delay in stem cell research funding from the US ought not significantly delay the actual research, it will only keep the discoverers from being US scientists.

[Sebastian Holsclaw]

Neat, I think 4=3.

[hilzoy]

Sebastian: it was a quick 'Thank God', more along the lines of: thank God we don't have to worry about that! -- In fact, I think that the most significant difference between 1 and 2 (assuming that no one plans to implant the embryos) is just that they will be stem cell lines ready for use. This is practically significant, since (as I think I said somewhere, but who can keep track?) cloning will not, in fact render banks superfluous. For one thing, there are a lot of diseases that might be treatable with stem cells and that need to be treated ASAP (stroke leaps to mind). Using an existing line will always be faster than growing a new one to order, and so for these diseases banks will be key.

I also agree with you about 3 and 4, with one qualification: the reason why one would clone an embryo and then harvest its stem cells, thereby destroying it, is to try to save people's lives. The reason why one would create more embryos than one knows one needs for IVF is convenience. This is why it's a mystery to me why the pro-life people aren't making every bit as big a deal about IVF as they do about therapeutic cloning.

[Sebastian Holsclaw]

"This is why it's a mystery to me why the pro-life people aren't making every bit as big a deal about IVF as they do about therapeutic cloning."

That is an artifact of the reporting. Especially when IVF was first being researched, the excess embryo issue was extremely important in the pro-life community. It has never broken out into public conciousness beyond the pro-life community so we have decided to fight more realistic battles. That historical fact is one of the reasons I get incredibly angry when I have IVF thrown in my face in the cloning or stem cell debates. One of the key pro-life arguments against creating excess embryos for IVF was that we should not create human life and discard it for convenienc--with the idea that such cheapening of life leads to commodification and the like. We have always been worried about the danger of looking at such embryos as 'excess'. It saddens me greatly to see that we now discuss the very things which pro-life advocates were dismissed as fear-mongers for raising when the particulars of IVF were being worked out. That is why I am extremely skeptical about certain cloning and transplant techniques--we've already fallen down the slippery slope once on this issue.

[hilzoy]

I agree about its being wrong to create excess IVF embryos, though not about therapeutic cloning. This being, of course, because I do not see a week-old embryo as a person, but I do see it as something not to be treated lightly. Creating excess embryos for IVf seems to be to count as 'treating it lightly'; creating it for research into a cure for, say, multiple sclerosis does not.

I'm not sure which slippery slope it is that therapeutic cloning leads us down, though. I mean, of course if you think that the fetus is a person from the moment of conception, then there's no slope at all; there's a moral cliff, and therapeutic cloning has fallen over it. But if a slippery slope were to be relevant, it would presumably be to someone who thought therapeutic cloning itself might be OK, but it would lead to something else.

Leon Kass seems to think this, for reasons that have always seemed to me bizarre -- he thinks therapeutic cloning would lead to reproductive cloning. But I don't know why he thinks this -- and why, in particular, he thinks that the fact that these embryos might be implanted in someone justifies banning therapeutic cloning, and why the argument for such a ban is stronger than, say, banning hammers because you can use them to kill people. I don't know what the other slippery slopes are.

[Sebastian Holsclaw]

"But if a slippery slope were to be relevant, it would presumably be to someone who thought therapeutic cloning itself might be OK, but it would lead to something else."

I think the slippery slope is likely to be fetal organ or fetal tissue transplantation from older fetuses.

For example fetal transplantation of brain tissue which is being researched now. Fetal brain tissue does not exist in embryos, but once there is a precedent of intentionally growing embryos for research and treatment you get to the line drawing problem and an intersection with abortion reasoning. 5 days, ok-embryo. 2 weeks? 8 weeks? 11 weeks? All in the first trimester, right? 20 weeks? Still in the second trimester, right? Where do you draw the line, and on what basis do you draw it?

This is especially concerning considering that fetal brain transplantation research is not a slippery slope, but something morally disturbing which is already going on.

[hilzoy]

But Sebastian: fetal tissue research, as I said, comes from fetuses that have already died by means that are already legal, whether or not either of us thinks they should be. It is illegal to do any research in which you offer any inducement for an abortion to occur, or affect the timing of it in any way, or anything like that. It is, in other words, treated as exactly on a par with any other research using cadaveric tissue. Is there any good reason to think that fetal tissue research is more morally problematic than research using tissue from a dead infant or adult, so long as lots of care is taken to prevent anyone from causing deaths to occur to get the tissue? (Something which would be unlikely in any case, since there is no shortage of either sort of tissue obtained without trying to bring about death for research.)

If you say (I mention this response only because iirc you said something like it earlier) 'what happens when we don't have enough fetuses?', why isn't this an equally good objection to research done on dead adults who donated their bodies to science? I mean, we don't just do research on them; we use their body parts for all kinds of purposes (during my last surgery, I only barely avoided having cadaveric tissue implanted into me.) So if anything the need might be greater. But somehow I am confident that we will know how to draw the line that separates (a) letting research continue on cadavers obtained the way they are now, namely w/o killing people for the sake of their tissue and with their consent, from (b) deliberate killing for the sake of valuable tissue, and allow the former but forbid the latter.

[Sebastian Holsclaw]

If you say (I mention this response only because iirc you said something like it earlier) 'what happens when we don't have enough fetuses?', why isn't this an equally good objection to research done on dead adults who donated their bodies to science?

It isn't an equally good objection because (to use the example above) if fetal brain tissue transplantation ends up being a ueseful therapy for any of the more common brain diseases, already-aborted fetuses will not be enough--especially given tissue matching issues. This is not a leap of logic on my part, because the scientists doing the research are complaining that they don't get enough tissue even to do research. This strongly suggests that a transplantation therapy would have far more trouble getting enough tissue.

"But somehow I am confident that we will know how to draw the line that separates (a) letting research continue on cadavers obtained the way they are now, namely w/o killing people for the sake of their tissue and with their consent, from (b) deliberate killing for the sake of valuable tissue, and allow the former but forbid the latter."

Examine the number of adults the state allows people to kill now. Compare that with the number of fetuses the state allows people to kill. This suggests that there is a difference in how the two are treated. Why do you then posit that there will then be no difference in harvesting treatment? Further posit a successful Alzheimer's treatment using such tissue. Considering that we allow abortions anyway, why are you so confident? Equally interesting--why would you care?

[hilzoy]

Sebastian: what makes you say that researchers say they don't have enough tissue to do research? The ones I talk to (and I was talking to a Parkinson's guy today) say they can't imagine running short.

About your last paragraph: you're talking about allowing people to pay other people to have abortions, and/or to conceive a child in order to abort it. I do not think that will ever be legal, for one thing because a lot of pro-choice people would oppose it. Me, for instance. I do not believe abortion should be illegal, for a number of reasons, some of which have to do with ethics and some of which have to do with the law. But neither I nor anyone I know who's pro-choice thinks that abortion is something to be done lightly, nor is there any reason why or pro-choice views imply that we should think so.

Besides which, there are other interesting treatments for Parkinsons around. I spent the early evening talking about one of them, which involves deep-brain stimulation via (removable, turn-on-and-offaable) electrodes. Serious improvement in the vast majority of the (carefully selected) people they do the procedure in, without the worst of the side effects of l-dopa. Testing similar electrodes for use in psychiatric illnesses (OCD and severe depression) has just started. Extremely interesting, both scientifically and morally.

[Sebastian Holsclaw]

I'm not against other types of treatment. Go for electrodes. I find it disturbing that so many people act as if Parkinson's and Alzheimers cannot possibly be treated without chopping up human or slightly pre-human people.

"I do not think that will ever be legal, for one thing because a lot of pro-choice people would oppose it. Me, for instance."

I fully expect to have the opportunity to hold you to that statement.

[liberal japonicus]

Well, folks, as George Allen used to say the future is now Comments?