Why I Should Still Offset Rather Than Do More Good

ABSTRACT

Stefansson (forthcoming) argues that by emitting and offsetting, we fail to fulfil our justice-based duty to avoid harm owed to specific individuals. In this paper, I explore a case where offsetting fails to prevent some but not all risks of harms that our emissions impose on them. By drawing on a distinction between general and specific duties not to (risk) harm, I argue that if by emitting and offsetting, we satisfy some (if not all) of our specific duties we owe others, then this gives us stronger moral reasons to offset than give to charities that do good more effectively.

KEYWORDS:

• offsetting
• climate change
• duties
• risk
• John Broome

Acknowledgments

I’d like to thank Orri Stefansson, Andreas T. Schmidt, and Sven Nyholm for their helpful written feedback on initial draft of this paper. I’m also thankful to Crystel Hajjar and Conrad Bakka for discussion on the topic.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

1. It is perhaps important to clarify that Stefansson’s view is not that your act of emitting-and-offsetting causes or risks causing harmful event E; rather, it is that when you go for your Sunday drive, you cause or risk causing the harm even though you emit and offset.

2. In criticizing Broome’s pro-offsetting view this way, Stefansson joins a range of philosophers who have similarly questioned, albeit in different ways, whether one really satisfies their justice-based duty by emitting and offsetting (C.f. Cripps, 2016; Hyams & Fawcett, 2013).

3. One might wonder about the relationship between the general and the specific duties. While I lack the space to develop this further, I am inclined to say, for now, that the relationship is simply one of derivation, whereby the latter duties derive from the former.

4. This is because the probability of a disjunction of independent events (such as satisfying duty1, or duty2, or duty3 … and so on) is higher than zero (that is, the probability that at least one of the disjuncts obtain is higher than zero) even if one of these events has close to zero probability of obtaining.