We perform a survey experiment on the issue of immigration. People are presented with a situation where public opinion is at odds with the election promise. In our control group, no information is given about public opinion. In the treatment groups, respondents are told that 55 per cent or 80 per cent of the people are against the project. When respondents are informed about the election promise but are not told about public opinion, 64 per cent say that the party should fulfill its promise. That percentage drops to 51 per cent when people are informed that a slight majority (55 per cent) are opposed to the project and to 42 per cent when they are told that a strong majority (80 per cent) are opposed. Citizens thus believe that politicians should pay attention not only to the majority view but also to the size of that majority.

I outline four conditions on permissible promise-making: the promise must be for a morally permissible end, must not be deceptive, must be in good faith, and must involve a realistic assessment of oneself. I then address whether promises that you are uncertain you can keep can meet these four criteria, with a focus on campaign promises as an illustrative example. I argue that uncertain promises can meet the first two criteria, but that whether they can meet the second two depends on the source of the promisor's uncertainty. External uncertainty stemming from outside factors is unproblematic, but internal uncertainty stemming from the promisor's doubts about her own strength leads to promises that are in bad faith or unrealistic. I conclude that campaign promises are often subject to internal uncertainty and are therefore morally impermissible to make, all else being equal.