Braham and Holler: Distributing causal responsibility in collectives

Braham and Holler. 2006. Distributing causal responsibility in collectives.

In Brief

The puzzle to solve: How to solve the "many hands problem" (suggested by Thompson 1980, and many others)? Determining the causal contribution of an individual in a collective activity or joint-decision: "Is it true that the person's actions contributed to the occurrence of that state of the world?" (Note: this article explicitly doesn't address the question of whether the person knew/should have known the consequence of action, nor if they acted freely. They only purport to establish a "factual connection.")

Main points: First, it is possible to overcome many hands problem. Second, the authors show how to rank agents by causal responsibility, and how to distribute and measure responsibility.

(The main point of the article is to explain the NESS test. See notes on Wright 1985 for more details on the NESS test and on ranking a power index. The logic here is much different, just a formalization and extension.)

How to overcome many hands problem and rank responsibility?

Use the NESS test "necessary element of a sufficient set" (Wright 1985; 1988). (NESS test is the same as INUS (Mackie 1965): "causal condition is an insufficient but necessary part of a condition which is itself unnecessary but sufficient for the result.")

The problem of the many hands problem is that it assumes strong necessity is required to show causality. B&H argue that this is not the case (see table for types of necessity and sufficient conditions on p. 2).

NESS Test

Weak necessity/strong sufficiency requirement. This circumvents the problem of duplicative causation--e.g. 2 marksmen kill same person simultaneously. Who is responsible?

NESS doesn't require that C (cause) is the causal mechanism for E (event) in ALL occurrences: "weak necessity/strong sufficiency means it need not be the case that a single condition is sufficient for E but that a set of conditions be sufficient (strict sufficiency is then the special case of a singleton set) on this occasion and combines this with a necessity requirement in that the members of this set are necessary for the set to be sufficient" (7).

Comments and Criticisms

When it comes to assigning political responsibility, does it depend on the system? Is it easier to ascribe responsibility in presidential system?

Why do we care? Here is my answer: If we can't attribute positive or negative states of the world to specific politicians or groups of politicians, voting is largely irrelevant because you can (or should) only hold someone accountable if you know that they were causally responsible.

If all individuals in a collective are causally responsible, and person A attempts to prevent event E from occurring but is unsuccessful, is person A still responsible?

What if person A tries to prevent event E but doesn't try as hard as he could. How does that alter his responsibility (if at all)?