[Confidence 50%, this is speculative and good counterarguments will improve my view of these matters.]

I.

When I do moral philosophy, I am looking for a usable repeatable framework for analyzing moral decision-making in a very broad range of cases. I accept that moral philosophy is an inexact science and cannot be made into a machine that mechanically outputs the right answer. There are simply too many parameters that go into real decision-making to make a moral philosophy complete in the way that an account of electro-magnetism is. 

But I would like for something that doesn’t fall flat on its face at the first objection. The standard Thomistic framework is a great starting point but is locked in intractable controversies. Perhaps chief among them is how to identify what an act is.

You see, ‘action’ is one of the three key metaphysical categories by which we judge praiseworthiness or acceptability of an action in the Thomistic system.

You know the rhyme:

With Intentions Good, the mind is right.

An act acceptable gives hand its might.

And should an evil too befall

Let pain be healing, evil forestall.

Or if lists are your preference:

1. Your intention must be toward the good ends, not the bad ones which may result from the action. Cause a stinging pain in your child’s knee to clean the wound, not see him bleed.
2. The act must be acceptable in itself. That is, it cannot be one of things that “we just don’t do” no matter what.
3. There must be proportionally good outcome that is not caused by the bad outcomes. You can kill to protect your home, but you cannot kill to collect better data on death rattles. Generally, we group all the considerations that include causal chains, uncertainty, consequences, and proportionality into one bucket called “Circumstances.”

I have always been troubled by step 2. For many years, I denied the existence of the “act itself” for moral analysis. What is the act, I thought, but the marriage of intentions with the mechanical actions to make them happen. The twitching of fingers, the vibrating of vocal cords, the speaking of vows, the motions of homework these are not meaningful at all in themselves. The circumstances and intentions give them meaning. A philosopher might say abortion is unacceptable in itself, because “the whole life is sacred from conception to natural death” belief commitment. 

I would respond in two ways: One abortion is a noun not a verb. The acts that make it up vary a lot and so there is not one act called “an abortion.” And then, since this is a bit obtuse to most, I would point out that there are actions that have the same effect which are acceptable in some circumstances, i.e. removing an ectopic pregnancy. What matters in the moral analysis are not the actions but the intentions and circumstances, even for the person categorically against abortion.

Another thing that I have trouble with is how to identify an act. We might say that playing baseball is an act. But is it? Isn’t playing baseball made of swinging, catching, throwing, running, shouting. Those are the acts, no? Or maybe swinging isn’t an action. Swinging is composed all sorts of subparts that make up the reified action of swinging?

Is vowing an act distinct from speaking? Isn’t it the circumstances of the words that make a speech act a vow as opposed to the words themselves? Aren’t words themselves social conventions anyway? And thus, identifying an act is just abiding by a specific social convention. A vow is an act. Marriage is an act. Life itself is one great deed.

I am revising this view. My motivation is that I think we should have a category of actions that we do categorically rule out. Yes, the category will be fuzzy, but a good life requires being the type of person who can draw a hardline at times to say, “This I simply will not do, and no argument or circumstance will make me do it.” I am talking about exceptionless moral norms. We should live by rules which over many iterations will lead to a better society and protect us from our worst habits of casuistry to justify viciousness. Exceptionless moral norms do this by saying that there are some powers of judgment we cannot trust ourselves to exercise. None of us can wield unfiltered moral reasoning, and if we coordinate around some universalized principles, we will save ourselves from self-destruction.

Are there circumstances where torture may be justified? Perhaps there are. But whom do we trust to correctly identify those circumstances when they have power? No, we want the dark powers chained, buried in the sea, or set beyond our grasp.

I don’t know what the exceptionless moral norms are. But I do want a way to preserve them. They are important Schelling points for behavior, guide action, and decrease the frequency of the worst offenses.

II.

Before continuing, I would like to revisit our terms.

1. Intention: A person’s intention is to complete all the acts and cause all the effects which lead to their desired end. Some of the effects might be necessary (or probabilistic) consequences of their actions but are not necessary to the end. These effects are foreseen, but not intended.
   
   For example, hitting a home intruder with my home defense flashlight causes his unconsciousness and my safe home, but his unconscious state, while foreseen, is not strictly speaking intended. The intended state is the safety of my home. You should think of this as a standard Pearlian causal inference diagram. One action leads to two states. Those states are causally conjoined by the action, and the logic of language says the unconscious person causes the safety of the home. But ignore the logic of language: states do not cause states. Actions cause states. The action causes both consequences.
   
   An intention is known by external actions chosen. If a person claims that they intend to learn, but instead cheats on their assignments, then they haven’t learned, nor was the cheating part of that intention. How we judge a person’s intention is by what acts they chose to do. This is both how legal system works and how common sense operates. If you plan to assassinate someone, but do so in a way that causes a lot of unnecessary collateral damage as well, then your intention should rightly be inferred from the specific actions you took. This is essential for understanding the act itself discussion.
2. An act is acceptable in itself, if it upholds exceptionless moral norms. That is, it preserves the essential character of the good intention.
   
   My go to example is taken from Steve Coll’s Ghost Wars. In 1998, Osama bin-Laden was being tracked. The US was 50% sure he was in a compound in Kandahar and special ballistics and mapping experts were busy calculating how the compound could be bombed in such a way as to collapse as few as possible of the surrounding domiciles. In the end the president, decided against bombing on such a probability of success, given the number of civilian casualties that would likely result, some around 60. What catches my eye about this story is that the ballistics experts are not only supposed to measure how to destroy the target, but also how to minimize collateral damage and unnecessary loss of life. If you want to act morally your actions need to show that you don’t intend the evil effects of a bombing.
   
   Everything that occurs between two end states in a causal diagram is the act.
3. Circumstances. These are the relative weights of the good and bad effects of the action, the uncertainty around these, relationships between people and institutions, the norms and obligations that are affected by the action. These all have some bearing that allow for judgements of proportionality, acceptability, and blameworthiness/ praiseworthiness of an action.

III.

With these redefinitions, we have a reason to think about exceptionless moral norms for actions, but then we suddenly have an issue about how to identify whether an action is within the bounds or not? A standard retort is to make the indirect/direct distinction and either argue, prohibition breaking must be either indirectly voluntary or indirectly caused by the agent.

What does it mean to indirectly vs directly to cause a harm? It is this distinction that I primarily object to and think needs reframing.

Is it how many mechanistic events intervene between your action and the effect? If I cut someone’s hand with an axe, that is more direct than if I cut their hand with a Rube Goldberg machine.

But would you call killing with gun which has a series of intervening events(trigger, hammer, gunpowder, combustion, propulsion) less direct than killing with a sword? All else held equal we wouldn’t consider these different moral acts at all. In a causal diagram necessary/inexorable effects do not need to be modelled.

Maybe the difference between indirect and direct is probabilistic?

If your action contributed to an effect, but was not sufficient to cause it then it is indirect.

But would all doctor’s prescriptions and most medical interventions then be considered indirect? That’s crazy talk, I think, for we would hold doctors responsible for unnecessary risk-taking and foreseeable harms even if they are only probabilistically going to occur. The probabilities matter for the intended effect, but the direct vs indirect distinction does not add anything.

Possible scenarios in this causal diagram.

The claim of the indirect/direct distinction is that if Action D leads to both a good effect and the bad effect. Action A leads to the good effect and bad effect just as well, but makes our voluntary actions more distant the bad effect it is, all else being equal, it is the more praiseworthy action. I am not exactly sure how this could be.

In the final column, I posited an commission-omission distinction. I have a problem with that distinction, but perhaps in light of the above argument...

Maybe it works like this? Action D is an exceptionless moral norm, but Action A, which causes state D is not. So we should prefer A, even if it seems like a fiction since D is caused anyway? Maybe avoiding dirtying our hands is an aspect of morality worth upholding. 

This is all a response to SEP 4.4. The whole article is very good! https://plato.stanford.edu/entries/double-effect/

I look forward to your thoughts on how you would think through the idea of direct/indirect distinction mattering.