I. General information on slippery slope arguments

The slippery slope argument is a figure of reasoning by which an action that seems morally acceptable is conclusively shown to be morally impermissible because it has morally unacceptable consequences. In general, two variants of slippery slope arguments can be distinguished based on two possible consequences: the causal variant and the conceptual variant.

The causal variant postulates that the introduction of a practice that is morally acceptable will, with a certain probability, also lead to the establishment of a morally unacceptable practice in the future. Therefore, contrary to first appearances, the practice that is morally acceptable had better not be introduced. A recurring weakness of causal slippery slope arguments is that in most cases it is difficult or impossible to state seriously accurate probabilities for the occurrence of the undesirable consequence. Even if plausible probabilities can be identified in individual cases, a causal slippery slope argument is only valid if the negative consequence is sufficiently weighty and its probability of occurrence is high enough to outweigh the foreseeable benefits associated with the introduction of the morally acceptable practice.

The conceptual variant of the slippery slope argument often occurs in a certain variety, the so-called vagueness variant: If action (1) is permissible, then an action (2) that deviates minimally from it is also permissible, which in turn justifies an action (3) that goes beyond it, and so on. Thus, one finally arrives at an act (n) that is morally unacceptable. But since act (n) is logically related to act (1) via a chain of minimally deviant acts, the argument goes, act (1) is already impermissible. Conceptual slippery slope arguments may fail because the minimal differences between the acts in question are ethically relevant, so that only the introduction of act (1) is justified, but not the introduction of acts (2) to (n).  

Whether or not a slippery slope argument has the theoretical shortcomings outlined must be examined on a case-by-case basis.

II. Slippery Slope Argumentation in the Case of PGD

Against the permissibility of PGD for the purpose of early detection of diseases with subsequent possibility to kill the embryo in case of a genetic defect, it is argued that this would also make the use of PGD for other purposes (such as the selection of an embryo on the basis of certain characteristics like eye color or intelligence) more probable (causal) or would have to be consistently permissible in other cases as well (conceptual). However, since PGD should not be established beyond the early detection of disease, PGD should not be used for the purpose of early detection of disease. The causal variant of this slippery slope argument, as outlined above, is only successful if plausible probabilities for the threatened expansion of PGD can be asserted and the negative consequences (weighted by their probability of occurrence) outweigh the positive consequences of PGD (weighted by the probability of their occurrence). Moreover, if the causal or conceptual dam-break argument is to work, the expansion of PGD must actually be a negative consequence. This, in turn, is asserted by reference to another slippery slope argument:

If some expectant parents were to use PGD for the purpose of selecting an embryo to grow into a preferably efficient human, then, according to a causal slippery slope argument, societal pressure would arise on other expectant parents to also inspect the genome of their embryos and to select them according to performance-indicating criteria. This is because in a society with genetically selected people (depending on the predictive accuracy of PGD), overall performance is likely to be higher than in a society without such selection, which tends to put all those who have not been selected at a competitive disadvantage. As a result, existing class differences in a society would be exacerbated and perpetuated, since socioeconomically better-off groups would use PGD on their embryos, in addition to the advantages already existing. Against the background of these foreseeable consequences, a liberal legal regulation that would allow the possibility of PGD for purposes other than the early detection of diseases would be ethically impermissible. Although it is difficult to answer the question of a specific probability of occurrence (or even a range of probability) of the competitive dynamic, such a development does not seem completely remote. Highly speculative, however, is the follow-up question whether a more unequal society, partly based on performance-oriented PGD, would be beneficial (higher GDP, higher social expenditures, etc.) or detrimental (competition, low esteem, etc.) for the socioeconomically worse-off groups. However, this very information could influence the ethical evaluation of the consequences of PGD for purposes other than early detection of disease, leaving the slippery slope argument against PGD for the purpose of higher performance in limbo.