The precautionary principle, and the spread of AIDS

Jay Manifold has a piece in which he quotes an article on the precautionary principle, which essentially states that if you have invented some new technology, you should be absolutely sure it is benign before using it at all. This sounds good at first, but of course the consequence is that if you take a very long time to prove something is benign, then you lose the benefits of having had it during that period. And of course the best way to prove something is benign is to actually use it, so the precautionary principle lengthens such periods.

Jay (and the people he quotes) make the point that if someone is sick, and you have a drug that might cure them, and you don't use it because of such a principle, then you bear some culpibility if that person dies. If you have better crops that could grow more food, and you don't use them out of concerns about the ethics of using genetically modified crops, then you are implicated if people die of starvation due to low crop yields. (There is also a discussion of stem cells). At times, inaction can be as big a crime as the wrong action can be. Therefore, clearly, we need to look at the costs of the precautionary principle, and look at how serious the consequences of things going wrong could be, and look at the possible benefits of things going right, and find some middle ground. Simply saying the precautionary principle should be sancrosanct isn't a terribly useful position.

The article makes a good case that many of the medical breakthroughs we now take for granted may have either never taken place are may have taken place much later if this principle was always in case. However, it then gives an example that makes me nervous. Lots of scientists are asked to give examples of progress that would have been halted (or is being halted) by overuse of the precautionary principle. Stuart Derbyshire (assistant professor of anaesthesiology and critical care medicine at the University of Pittsburgh) gives an example of "Xenotransplantation could benefit thousands, but is being held back by ill-founded concerns about porcine retroviruses".

Okay, in this instance we are talking about giving people drugs to ensure that rejection does not occur, and transplanting the organs of pigs into people for who there are not human organs available for transplants. People are worried about the spread of pig diseases to people, particularly diseases that are benign (possibily due to millennia of evolution) in pigs but dangerous to people.

Sometime before 1960, two retroviruses managed somehow to cross the species barrier from primates into human beings. A virus managed to get transferred from chimpanzees to humans, and in humans it is known as HIV-1. Another virus managed to get transferred from a species of monkey called the sooty mangabey, and this is known as HIV-2. These are different viruses (although of the same type) and cause very similar symptoms, have long incubation periods and are usually fatal. HIV-1 is responsible for the bulk of the AIDS crisis, although HIV-2 has also killed a substantial number of people, primarily in Africa. We know that HIV-1 crossed into humans on at least two occasions, as one form of the virus, HIV-1 Group O, is genetically sufficiently different from the other variants that it clearly came from a different population of chimpanzees. Thus we know that AIDS crossed over into the human population on at least three separate occasions, all probably between 1930 and 1960.

Nobody knows how the viruses crossed over into humans. Given that the viruses were in chimpanzee and sooty mangabey populations for thousands of years, and the cross over occurred three times in a very short period, it seems certain that changed human activity in central Africa was in some way responsible.

In particular, a theory that the disease was transferred from primates to humans during trials of polio vaccines that had been partly developed using infected primate tissues in Africa in the 1950s was argued by Edward Hooper in the book The River. Hooper's evidence was circumstantial, and since the book was published the circumstantial evidence has been demonstrated to be less convincing than Hooper argued, and most people do not believe it. Tests of samples of the vaccine have found no evidence of the virus. (Of course, this does not disprove it, as only a small number of samples needed to be contaminated. However, it does make it less likely). The most common argument put forward for the transfer of the viruses to people is that people hunting primates somehow managed to infect themselves via the spill of Chimpanzee blood. I find this argument unconvincing, as this could have happened at any time in the last several thousand years. That it should not happen (or at least not cause an epidemic if it did happen) for this length of time, and that it should then happen at least three times in a few years strikes me as implausible. I find the fact that it occurred at least three times in a short period deeply troubling, to say the least. I think that something that changed due to human civilization caused the species jump, but I do not know what. (It could be something very simple like the widespread use of dirty syringes, some of which came into contact with animal tissue).

I personally doubt the polio vaccine theory is true, but I cannot categorically say that it is false. Even if it is false, there is a chance that the spread was in some way an unexpected consequence of medical efforts or medical treatment of some kind. If this is true, then the people arguing for the precautionary principle have a stronger case, at least in situations where the spread of a new disease is an (even remote) possibility. The trouble with situations that involve the spread of a new disease is that the worst case scenarios are so dreadful that they must be taken into consideration even if they are very unlikely. It is not unreasonable to talk in terms of propabilities. Figure out the number of lives that will be saved by a new technology if things go well, and multiply by the probability of this happening. Figure out the number of lives that will be lost if things go badly, and multiply by the probability of that happening. subtract one from the other, and you have an expected benefit (or not) of using new technology. If the expected benefit is large enough, then go forward.

However, in cases where the spread of a disease is a possibility, and the worst case scenario is "a billion people will die", then the expected benefit can be negative even if the probability of things going wrong is very small. These types of cases create situations where the ethical dilemmas get even more complicated, and where I find arguments for something like the precautionary principle to be more convincing.

This is why the idea of xenotransplants still makes me very nervous. The possibility of spreading a harmful retrovirus from pigs to people is very small. The most convincing argument that this is so is that people and pigs live very close to one another throughout the world, and people come in contact with pigs blood regularly. Therefore, if there were any such viruses in pigs, they would have crossed the species barrier long ago. I find this argument quite convincing. However, the number of potential beneficiaries from xenotransplants is low compared to the number of potential beneficiaries from genetically modified wheat. And while the risk is low, it is a risk of something truly terrible. So forgive me if the idea makes me nervous.

I stress that this is a special case. In a situation where you are (say) dealing with genetically modified wheat, the balance is different. The number of potential beneficiaries is much larger, and the consequences of something going wrong are much smaller. (Say, we create a strain of wheat that turns out to be unhealthy or inedible. We stop growing it, and we stop eating it, and we plant some other strain instead. Perhaps we end up creating a particularly nasty weed. This is bad, but not catastrophic).

What am I arguing. In the end you have to muddle through on a case by case basis. Being ethical doesn't involve taking simplistic (but simple to arrive at) absolute stances. It involves accepting that compromises sometimes have to be made, and that the morally best outcome is probably somewhere in the middle of the minefield, rather than somewhere on either side of it.