The Side-Effect Effect and Conversational Pragmatics
Fred Adams and Annie Steadman were among the first to attempt to explain the side-effect effect. Adams and Steadman (2004a) suggest that we should be careful about using facts about people's intuitive judgments to reach conclusions about the nature of their underlying concepts.
In particular, they argue that the side-effect effect does not tell us anything about the nature of our folk concept of intentional action but is, instead, best explained entirely in terms of pragmatic considerations - in particular, in terms of conversational implicature.5 According to Adams and Steadman, people believe that the chairman is blameworthy in the Environmental Harm case. This belief introduces a problem. The only way that people can express this belief is by saying that the chairman harmed the environment intentionally. If they said that the chairman did not harm the environment intentionally, then this would imply that the chairman isn't to blame for harming the environment. Recognizing this, when people are asked whether or not the chairman harmed the environment intentionally, they tend to say that he did. This tendency to say that the chairman harmed the environment intentionally doesn't reflect the nature of people's folk concept of intentional action, it merely reflects their desire to blame their chairman for harming the environment.6Knobe (2004b) argues that the best way to evaluate the merits of this explanation would be to find a different method for determining whether or not people regard the chairman's actions as intentional - a method that doesn't depend on pragmatic considerations involved in people's use of the word ‘intentional'. We could compare the results obtained using this second method against the original results.
If the results obtained using the second method differ significantly from the original results, then this would give credence to Adams and Steadman's suggestion that the original results were due to pragmatic considerations and don't tell us anything about the nature of our folk concept of intentional action. On the other hand, if the results obtained using the second method agree with the original results, then this would give credence to the suggestion that the original results were not due to pragmatic considerations and do tell us something about the nature of our folk concept of intentional action.Knobe (2004b) asked people whether or not the chairman helped or harmed the environment in order to increase profits. According to Knobe, people are unwilling to say that someone performed an action in order to achieve a particular goal unless they believe that she performed the action intentionally. As such, asking people whether or not the chairman helped or harmed the environment in order to increase profits provides us with an indirect method of determining whether or not people regard the chairman's action as
intentional. Knobe found that most people who were asked to consider the Environmental Harm case were willing to say that the chairman harmed the environment in order to increase profits, while most people who were asked to consider the Environmental Help case were not willing to say that the chairman helped the environment in order to increase profits.7 Since the results obtained using this second method agreed with the original results, Knobe concluded that the original results were not due to pragmatic considerations and did tell us something about the nature of our folk concept of intentional action.
Knobe's (2004b) study did little to convince Adams and Steadman. In a paper responding to this study, Adams and Steadman (2004b) argue that the same pragmatic considerations that played a role in people's willingness to say that the chairman harmed the environment intentionally are playing a role in people's willingness to say that the chairman harmed the environment in order to increase profits.
They agree with Knobe that people are unwilling to say that someone performed an action in order to achieve a particular goal unless they believe that she performed the action intentionally. But, they argue, this is precisely why we should suspect that whatever pragmatic considerations played a role in people's willingness to say that the chairman harmed the environment intentionally are playing a role in people's willingness to say that the chairman harmed the environment in order to increase profits.Adams and Steadman are right. If pragmatic considerations are playing a role in people's willingness to say that the chairman harmed the environment intentionally, then they are also likely playing a role in people's willingness to say that the chairman harmed the environment in order to increase profits. It is not clear, however, that pragmatic considerations are actually playing the role that Adams and Steadman suggest. Malle (2006) points out that people who were asked to consider the No Skill/Immoral Outcome case were invited to express blame directly. It is odd to think that people would feel the need to conversationally imply something that they were allowed to express directly.8 Still, it might be the case that people's attributions of intentionality are influenced by their attributions of blame in a different way: people's attributions of intentionality aren't made in order to conversationally imply blame but are made for the sake of consistency. That is, maybe people feel that it would be inconsistent to say both that someone is to blame for her actions and that she committed those actions unintentionally. But, this also doesn't seem to be the case. Instead, it seems as if people are quite willing to blame others for unintentional actions. Consider the following vignette:
Drunk Driver:
Bob got rip-roaring drunk at a party after work.
When the party ended, he stumbled to his car and started driving home. He was very drunk at the time - so drunk that he eventually lost control of his car, swerved into oncoming traffic, and killed a family of five.9Knobe (2003b) found that most people asked to consider this vignette judge that Bob was to blame for killing the family of five even though they also judge that Bob did so unintentionally. So, it seems that people's attributions of intentionality are neither made in order to conversationally imply blame (since the same attributions are made in contexts in which it is possible to directly attribute blame), nor made for sake of consistency (since there are contexts in which people are willing to attribute blame without attributing intentionality). Putting these two considerations together, it seems hard to sustain the suggestion that the side-effect effect is best explained in terms of pragmatic considerations.
4.