Donagan, Alan - Consistency

09/24/2010

The Theory of Morality ch 5, University of Chicago Press, 1979

This chapter attempts to deal with possible inconsistency in the moral system author has laid out in the previous three chapters. Author is less concerned about consistency between all the different precepts, since undoubtedly he mis-formulated some of them in a way that would generate moral conflict. The larger question is whether a system such as this in general is prone to inconsistency. Inconsistency is equated to moral perplexity-- there being a circumstance where the moral precepts give conflicting answers on what to do. Viewed in this light, author looks at Aquinas' cases of moral perplexity: perplexity on account of some prior misdeed, and essential perplexity. Perplexity due to some prior misdeed is the circumstance where a prior action or intention was impermissible or culpable and now the agent is faced with conflicting choices about how to get aright. Author back others like Kant and Aquinas in claiming that this conditional perplexity is not a threat to the system's consistency. (pg145) This still leaves moral perplexity 'simpliciter'.

The possibility of perplexity simpliciter is first given by Bishop Kirk, claiming that only a moral system with one principle could be said to contain no contradictions a priori. Any system with more than one inviolable principle may results in a conflict between the principles. Author denies this, claiming that one can easily imagine two inviolable principles ('absolute prohibitions'), such as not lying and not killing, which would never come in conflict. (pg146-7) Author also considers Peter Geach's conception of Divine law, which never conflicts due to divine providence but can appear to do so because of limits on human understanding. Author does not help himself to this because of the theological commitments it entails, and Donagan has taken a secular route. Author also does not use an ordering principle or weighting, since it is 'incompatible with the very nature of deductive systems'.(pg148) Instead, author claims that there is a structural principle that is derivable from the fundamental principle of respect for rational agents: 'It is impermissible to do evil that good may come of it'. (pg149) [his argument is on pg155] Author labels this the Pauline principle after St. Paul.

Author reviews the previous work, specifically second-order and first-order precepts. The two levels can't conflict since all manner of combinatorial schemes are possible without moral perplexity. Furthermore, author claims that second-order precepts are not independent enough from first-order ones. Thus if there is any inconsistency, it is among the first-order precepts (permissible and impermissible actions). (pg149-150) Author then reviews the different types of duties given in the first-order precepts: duties to self and duties to others within and without institutions. During this time author claims that it isn't possible that one needs to hurt oneself in order to respect others. In cases where there must be a choice between two impermissible acts (not because of perplexity simpliciter, but perplexity after a prior violation), author does not admit that there are degrees of moral impermissibility. Instead, impermissible wrongs can be more or less 'grave'. (pg152) Hence the precept: 'When you must choose between evils, choose the least.' During the review, the principle of beneficence arises as a potential source of inconsistency (pg153). Here the Pauline principle does some work, as does Kant's 'perfect' and 'imperfect' duties. Author argues that some acts are impermissible, but among the permissible ways to promote human well-being, there are many different plans of action. One must be beneficent, but it is left open exactly how. (pg154) Author also discusses Jonathan Bennett's objection to absolute systems because they do not take into account bad consequences. (pg156-7)

The next section (5.3) deals with a 'double effect', an action that is both good and bad. Author talks about the exact formulation of this idea, notably a conflict in formulation between JP Gury and Germain Grisez. (pg158) A case of double effect is killing in self-defense (if killing is wrong). Here it is permissible to defend oneself, even to the death, and yet killing is also impermissible and presumably a grave evil. (pg159) Author also takes the case of abortion: under Gury's formulation it is impermissible to kill the baby to save the mother's life. However, it would probably be permissible to extract a 'cancerous womb' to save a woman's life, even if that cancerous wound happened to contain a baby. For Gury, the principle turns on whether the evil was the means of the good effect or not.(pg159) Author rejects Gury's formulation because of author's previous work in the theory of action; both cases are agent actions voluntarily done.

Author contrasts Gury with the thinking of Germain Grisez, for whom the principle of double effect was about analyzing two factors: the unity of performance (acts) and the unity of intention. If an act cannot be divided into smaller acts which might avoid double effect, then an examination of the intention with which the act is performed is the important factor for double effect. For Grisez, an intention is licit if it is performed for the good effect and not for the bad, and the bad is out-proportioned by the good. (pg161) For instance for Grisez, an abortion may be performed in order to save the mother's life, and yet it is equally permissible to kill the mother in order to save the fetus. (pg162) The problem here is the absolute abstention from violating or coercing an aggressor. The Judeo-Christian tradition sees a fetus that happens to threaten the mother's life as an aggressor, somewhat ungrateful for its life. (pg162) For author however, the performance of an impermissible action leaves one open to coercion (or death, if necessary) because she has forfeit the right to respect. For author then, the principle of double effect is useless since the cases it was meant to cover are not problematic. (pg163) This is lucky for author, since author also objects to Grisez's treatment of intentions as the only morally relevant components of actions (see ch 4).

The final section (5.4) discusses a unique feature of the principle of beneficence, pregnancy. There is a possibility for moral perplexity here because of the possibility of the human race outstripping the available resources. If birth control is impermissible and Malthusian (overpopulation) problems are possible, it seems there could be conflict. However, author argues that contraception is permissible. (pg167) However, abortion is impermissible. (pg168-170) Author considers two arguments for the permissibility of abortion. The first is that a fetus is not a human being. Author: 'that is forced' (pg168). The second is that humans do not have the right to be respected as rational agents, only full-fledged members of a moral community have that right. Author: even if I grant that, isn't it true that a fetus will become a moral community-member if you nurture and let it? 'If respect is owed to beings because they are in a certain state, it is owed to whatever, by its very nature, develops into that state' (pg171). Author does allow for an abortion in the case of rape, since the mother cannot be subjected to labor and child-rearing without a prior voluntary action (pg169). For author, any case where the pregnancy is not the result of a voluntary action would excuse the mother removing the fetus, viable or not.