A physicocausal account of double effect
In response to Zippy Catholic's invitation to come up with a new term for the account he describes in his post, I would submit "physicocausal account of double effects" as an alternative term. The reason for the first change that "physicalist" seems blatantly inappropriate for an analysis that views the physicocausal aspects in terms of the intent (meaning proximate end in this case) for one's behavior. The reason for the second change is that my physicocausal test is only intended to classify effects as directly or indirectly intended, irrespective of whether that causes them to be classified as intrinsically evil.
Zippy's description appears to be correct in the main:
The approach we must reject goes something like the following: We take the decision a person makes to act, figure out the intended end for which he makes it, and construct a physical account from what he does to the achievement of that end. Everything which is a physical cause leading up to his desired result, then, is considered to be intended; anything which is not causally prerequisite to achieving his end, on a physicalist account, is considered to be unintended[* Other language is sometimes used to label what I have labeled intended and unintended. One traditional way is to refer to the intended and the indirect voluntary; another is to say directly intended and indirectly intended. But these are merely semantic choices about how to label things, and do not as far as I can tell change the substance of what we are discussing.].
In order to resolve ambiguity, I would replace "intended end" with "proximate end." In other words, it is the immediate physicocausal result of one's behavior that one desires to achieve and nothing more remote than that. I would also add "permitted" as a synonym for "unintended," "indirect voluntary," or "indirectly intended." Most Catholics (and indeed, most non-Calvinist Protestants) can intuitively grasp that God, even though he sustains in existence the actions of evildoers, merely permits evil rather than intending it. The mundane concept being used for the theological analogous term in this case is identical to the one I have in mind.
So far, so good. Then the picture starts getting murkier:
His act is intrinsically evil if and only if any of the things he intends (on this account of intention) is evil.
First, this account is wrong. On the physicocausal account of double effect, if there is a moral species of act in which the act is intrinsically evil merely by indirectly intending some effect, then the physicocausal account of double effect would simply say that it was indirectly intended, being entirely agnostic about whether indirectly intending that effect is evil or not. It makes no judgments about whether the action is evil or not; perhaps there are species of conduct in which indirectly intended effects make the action intrinsically evil. All the physicocausal account of double effect says is that if the evil effect is intended as proximate end or means, then the action is definitely evil and therefore cannot be weighed proportionally against any good. Note that it doesn't matter whether the action is intrinsically or extrinsically evil, and really, all of this discussion connecting the concept of intrinsic evil to double effect is simply irrelevant.
Second (and this is the reason I say this is murkier), I am not aware of a single species of intrinsically evil conduct that is evil solely on account of its effect. About the only pathological case I could conceive offhand was one with a masseuse who knows a married man is attracted to her, who is repulsed by the idea, but who is in desperate need of payment and so agrees to perform a therapeutic massage in exactly the same professional manner as she would for any client. I would argue that this could be intrinsically evil as adulterous conduct even though the effect could arguably be considered a side effect of what would otherwise be a therapeutic massage, a neutral or good act in itself, because she knows that this ordinary action has a side effect of inflaming the married man's passions. So it could be that knowingly inflaming someone's passions is evil no matter what the direct intent of one's actions are. That would seem consistent with immodesty more generally, since there are probably numerous people who mean nothing by wearing light clothing in warm weather but ought to know with certainty that their potentially innocuous choice will be an occasion of sin for others.
The reason I bring this up is as follows:
Basically, this account of intrinsic evil takes the principle of double-effect to apply to all acts, and elevates the double-effect requirement "the bad effect must not cause the good effect" to the status of a rule which determines whether or not an act is intrinsically immoral.
Even without doing further work we can see that this approach is fundamentally question-begging. Rather than applying the principle of double-effect to an act which is not intrinsically immoral, this approach applies the "bad effect must not cause the good effect" rule - which in reality only applies to acts which are not intrinsically immoral - in order to conclude that the act is not intrinsically immoral.
Furthermore, this account of intrinsic evil renders the requirement "the act must not be evil in its object" nonsensical. If the rule "the bad effect must not cause the good effect" is the very thing which tells us whether the act is evil in its object, then the inclusion of the additional requirement that the act must not be evil in its object is superfluous nonsense.
In the first place, this is simply false, because the physicocausal account actually includes both intended ends and means, not merely means. But the bigger trouble is that it seems to presume "the act must not be evil in its object" is NOT superfluous. That is debatable for exactly the reason I stated above: namely, I can't think of any clear case in which a species of intrinsically evil conduct is defined so that merely indirectly intending an effect makes the conduct intrinsically evil.
To put the point more plainly, "double effect" analysis simply codifies what St. Paul says regarding not doing evil so that good may come of it, so that if evil is being done, then no good effect can offset it (and that applies with regard not only to intrinsic evils but actions that are evil by intention or circumstance as well). In other words, it specifies conditions in which proportionality cannot be applied, thus restricting what can legitimately be called a "double effect," and it then requires proportionality even in those cases. That's precisely why there is a well-known 3-point formulation of double effect analysis:
1. Intentionality. The good effect and not the bad effect must be intended.
2. Causality. The good effect must not be caused by the means of the bad effect.
3. Proportionality. If (1) and (2) are met, the bad effect must not outweigh the good effect.
There can be additional details to these analyses. In particular, the causal prong is often evaluated by the "if by a miracle" test, more aptly described as the "test of failure" to determine whether a particular effect in the causal chain, if removed, would terminate the chain of causality. But those three requirements are the basic conditions for "double effect."
Sometimes, an alternative formulation is given with another step:
0. The act must not be good or morally neutral, nor intrinsically evil, in its object.
Now, it isn't clear to me that there is even one case where an act is classified as intrinsically evil for intending the evil effect as neither ends nor means but only as indirectly intended effect. But as a strict matter of logic, it could be the case that such a class exists, meaning that (0) might not be entirely superfluous. Or perhaps it is simply intended as a clarification in cases (like the masseuse above) where an argument might be made that the evil effect is indirectly intended, in order to forestall rationalizations. In any case, (0) is certainly a step that is so rare in catching any scenario not already caught by (1) or (2) that one could argue that it is very close to superfluous, to the point that some moral theologians don't even consider it necessary and rely on the 3-point test instead.
What is important to realize is that the physicocausal account of double effect is directed only to steps (1) and (2), and all it says is that if (1) or (2) is met, then the action clearly cannot be evaluated under double effect. It is agnostic as to step (0), and generally, it is agnostic as to whether the act in intrinsically evil or not, because that doesn't matter for double effect purposes anyway. If it's an evil act, whether extrinsically or intrinsically, it can't be done for the sake of any other good. What is absolutely clear, however, is that (0) is not some sort of separate test for moral object or intrinsically evil conduct with (1) and (2) pertaining solely to ulterior intention or circumstances. If anything, that's exactly reversed. (0) is only directed at catching those few pathological cases that are not already caught under (1) or (2).
More or less, double effect, which applies to any action that is evil intrinsically or extrinsically, should not be confused with the determination of whether something is intrinsically evil. You might need to look at remote intentions to determine if something is intrinsically evil (as in the case of theft), and you might need to look at remote intentions to determine if an effect is proportional under double effect. But just looking at remote intentions in both cases does not make the analysis identical.