The Burden of Proof
<>On the impossibility of technology assessment for the Human Genome Project
<>Michael Drieschner, Bochum

The editors asked me to elaborate a bit on a point I made during the conference at a panel discussion, namely on the complexity of natural processes and the warning it gives us, not to tinker about with it: If it were not for other reasons then for the one reason that we cannot assess the consequences.

Actually my point of view is no different from what any reader of newspapers could think of. I feel encouraged to publish this view in the present conference volume because of my special experience with this kind of questions from my time in the "Max-Planck-Institut zur Erforschung der Lebensbedingungen der wissenschaftlich-technischen Welt", where in the course of interdisciplinary research and philosophical considerations (and, correspondingly, much disappointment about its possibilities) many problems of this kind were discussed.—So, what is at issue?

We are enjoying progress. On average we reach twice the age of our ancestors. Nobody has to go hungry. Practically all of the 80 million dense population of Germany live in affluence, most of them rather struggle with their overweight. We can reach every location on earth physically within 24 hours. Informationally we can even be anywhere within fractions of a second, if somebody is there with a cellular phone.

We can determine over ourselves to an extent unthinkable still a few hundred years ago, from the election of the president to worker participation, form free choice of partners and trade to free choice of the number of children. We are even promised that in the near future we shall be able to determine the quality of our children, or at least that only those will be born who are free from handicaps. It should be like paradise. Only people do not seem to be happier or more satisfied now. Is it possible that happiness does not depend on this kind of changes?

In the late 60s we could have learnt a historical lesson from a coincidence: The Hudson Institute, Herman Kahn's "think tank", published a book "The year 2000" (Kahn and Wiener 1967), an enthusiastic picture of the land of milk and honey we stand on the threshold of, where all problems that harass us will be solved.—In the very same year there came up the discussion about problems of the environment and of the Third World. At that time we still mainly spoke of traffic noise and starvation in the colonies (Heinrichs 1968). In the meantime the discussion has broadened, and we see that we are well on our way of making the world uninhabitable for our descendants if we continue to live as we do now. Beyond that, our way of life is such that only a small fraction of humanity can participate in it (Meadows 1972).

Thus instead of Cockaigne we had suddenly the Apocalyptic Riders. What had happened? Don't we have everything under control, as we thought?

Let us have a look at a case that is far from genetic engineering, that we already overlook, though, historically for some time: The automobile! When Mr. Benz built his first motor carriage the technology assessment of that time could have emphasized the saving of space for horse stables, the advantage that the motor carriage would not cost anything when it does not ride, and the higher velocity of travel. Traffic victims, the problems of parking, cities adapted to the automobile, splashing into the open countryside, monoculture of department stores in the inner cities, the dependence of Germany's entire economy on the automobile: The very best technology assessment could not have foreseen all that!

Individual traffic is still a limited, relatively small field, open before our eyes, that has got its heavy impact on life only by the development of the automobile. But I think I can show in this field what my concern is: That for the questions of technology assessment it is all important to look up from what is immediately before one's eyes and to gain an entirely new perspective.

It is interesting to see that we seem to have so much less time today than in stagecoach times (when people of our status had to walk, anyway). Did it not seem self evident that we would save time when we rode faster, taking less time for the same distance?—I know two quite different approaches to explaining this discrepancy. Both are rather schematic. But both have a quality that seems necessary: they help look up from the workpiece right before my eyes.

The first one is Ivan Illich's quite serious calculation that by car one is really even slower than walking or on bicycle. This comes out because he does not only calculate the time immediately needed for the ride, but also the time necessary in advance for being able to ride: Time for car administration and repair; and mainly time to earn the money that driving a car costs.

Let us take a rather realistic example: Somebody goes 1000 km by his car per month, at an average speed of 50 km/h. This costs about 600 DM per month, according to the table of the German automobile association ADAC. Let us take the net wage of an average worker being 15 DM per hour. It takes him 20 hours for plainly driving 1000 km, but for earning the necessary money it takes this average worker another 40 hours. So actually it takes him 60 hours for 1000 km, which results in an average speed of 16 km/h. You go that fast on bicycle as well! In reality nobody would go 1000 km every month if he had not that easy and seemingly swift method to do so by car. On foot he would maybe go 100 km per month and take for that distance, very roughly, 20 hours. Now he prefers a car: If he wants to have a car ready at hand, even if he does not use it, he would have to spend the equivalent of 20 hours of work. In order to cover 100 km "quickly" he would use another 2 hours. So on the whole he would spend 22 hours to cover 100 km: A human being living the way of life of stagecoach times could not save time by using a car, but it would take him extra time!

An entirely different model is used by Carl Friedrich von Weizsäcker in order to make understandable the shortage of time of our contemporaries. Maybe this model is as valid as Ivan Illich's. Weizsäcker gives the following calculation: Let us assume that I go to see friends or visit events that are within an hour of my home (or whatever time you choose; for our calculation this does not matter). I assume for simplicity that such opportunities are distributed uniformly around my home. Then the number of opportunities I can reach easily is proportional to the surface I could cover within an hour. When I walk, that surface is a circle with radius 5 km. When I drive it is a circle with radius 50 km, that is with a surface one hundred times as large. Thus a driver has about one hundred times as many opportunities he could or might want to visit as a pedestrian.—No wonder he runs short of time!

Those calculations may serve as examples for an unconventional widening of our view. At the same time they may illustrate that technology assessment depends largely on our imagination: If there is a possibility we have not thought of, but which is crucial, our whole technology assessment is useless.

Let us return to the consequences of technical interventions into our lives: They are much more complicated than transportation! Already the climate on earth is so complex a structure that we have striven in vain for many decades to predict the consequences of the increase of CO2 in the air. How much more this must be valid for the extremely complex interrelations within a human organism or human society! An intervention as simple as the introduction of "the pill"—how much has it revolutionized not only the relations between sexes, but our whole way of life. Or take the phone: materially entirely harmless, barely noticeable; but it has turned our entire social structure upside down!

So now we go ahead and start assembling our offspring from blueprint. That unfortunate debate about "the certificate" ("Schein") of the last months had one merit, though: it brought into public awareness once again that abortion could be a problem. Whoever thinks about it at all in Germany, as, for example, the colleagues from the medical ethics department, mostly sees the issue—that seems self-evident!—only under the aspect of self determination of the parents: To the same extent as my profession or my consumption is at my disposal, I should be able, as far as technically possible, to determine the quality of my offspring. The next step of technical progress seems to be within our scope: Just now TV news report that the first human chromosome has been sequenced completely; the rest of the human genome will probably follow shortly. Nobody knows up to now what we can do with this knowledge. But certainly a technical application will be found, just as in the cases of all scientific innovations until now: There was always a (profitable!) application at hand! Think only of the technology of in vitro fertilization—it was the one invention that made the misfortune of a childless couple a real tragedy in that it promised a technically realizable remedy.

In the meantime we have become prudent; we think: "We have learnt that every new technology was charged with side effects we had not imagined when it was introduced. Thus we are doing technology assessment to gain control over this problem. Unfortunately this does not work yet one hundred percent satisfactorily, we are still at the beginning. So we shall have to do massive technology assessment research!"—Has anybody ever considered that this beats everything in human hubris, over-estimating human abilities? As if it were ever possible to "gain control over" the consequences of human action, beyond the very immediate ones!

If we have learnt anything at all from genome research of the last decades, it is this: The structure of the genome is extremely more complex than we had ever imagined beforehand. Any progress in this research brings to light new complications, every glance around a new corner opens our view at a whole new scenery of streets and squares of a complexity inconceivable before. Structurally one could have imagined something like that from the beginning. But what kind of complexity this really would be, nobody could have guessed beforehand. Still much less anyone is capable of guessing the consequences of technical interventions in systems that are so complex.

Certainly it is conceivable that we might find interventions that have a very narrow positive effect, for example cure a certain hereditary disease. But even there we should ask if such intervention makes sense. Take the example of a hereditary disease: There are even today good ways to prevent the birth of handicapped children, in that parents who are in danger to have offspring with hereditary diseases would not have offspring at all. There are many, in their majority worse reasons for which couples decide not to have children, so why not for the reason that those children might be handicapped?—We have learnt from this well-known method of preventing hereditary diseases, which abyss can open behind "pure technique", in this case the whole complex of eugenics. This turns out true even for methods the technique of which is simple and the consequences foreseeable. How much more this must be true of a so far unknown technology, even if our first impression is simple and harmless as well. Thus the best requirement for the introduction of a new technology would be the proof, according to the strongest criteria, that its application is harmless.

This requirement is recognized even today as valid, but, alas!, rather according to the guideline: "If nobody can think of anything that can be harmful with that technology, we can regard it as harmless."—The criterion should rather be the other way around, that the one who proposes a new technology is obliged to show that he has checked all (really all!) possibilities, and has found that none is harmful, including "social" consequences, as in the case of the eugenics debate.

One can easily see that nobody will be able to do that, regardless of the type of technology. That means that we should not start at all tinkering with the genome!

We started with examples from transportation and other "simple" technologies. For those, in principle, the same considerations apply. But would not the whole of society come to a stand-still if we stuck to the advice not to tinker with complex systems?

What do those stories with cars and blueprints aim at?—They are supposed to open up a view that is not given automatically with considering technological problems, that is not included in "technology assessment". The examples should verify my conclusion:

There might be cases where the harvest for humanity from a technology is obvious. Take, for instance, the necessity to check cholera: Sewage had to be constructed, and nobody gave long considerations about the question whether sewage could be harmful. But sewage was a technology well tried and tested from Roman times on.—When it came to carrying through smallpox vaccination one had to weigh up the consequences of an epidemic against the consequences of vaccination. It was a forced decision under conditions of ignorance—as it is necessarily in almost all practical cases.


In the case of genetic engineering the problem is that we know, on the one hand, so little about the very complex systems involved, such that technology assessment is practically impossible; and that, on the other hand, the consequences of not applying genetic engineering are obviously harmless.—This is probably an unorthodox view; because it is rather easy to draw a line from any scientific method or scientific result whatsoever to consequences that are, if not salvation of mankind, at least a decisive improvement of its fate. To prove that it suffices to read at random some last paragraphs from papers in, say, Scientific American!—Such claims are rather easily unmasked as public relations talk. If one gives it a closer look one can see, in general, that there is no harm in keeping away from the proposal. And one can see as easily that nobody can in the least imagine what the consequences of realizing the proposal would be for the very complex system at stake.

It is so easy—with one false move within the almost absolute power of our technical means—to destroy what cannot be reconstructed within centuries!

To weigh soberly in such discussions, not to let oneself seduce by arguments of "technical sweetness": Such are the needs of the moment.



Heinrichs J (1968) Welternährungskrise oder: Ist eine Hungerkatastrophe unausweichlich? Rowohlt, Reinbek

Kahn H, Wiener A (1967) The year 2000; a framework for speculation on the next thirty-three years. Macmillan New York

Meadows D (1972) The M.I.T.–Club of Rome Project on the Predicament of Mankind. M.I.T Press, Cambridge, Mass.