Well, no. Utilitarian systems are based on a utility function (although I’m not aware of any requirement that it be immutable… actually, what do you mean by “immutable”, exactly?). Consequentialist systems don’t have to be utilitarian.
Even so, the origin of a utility function is not that mysterious. If your preferences adhere to the von Neumann-Morgenstern axioms, then you can construct a utility function (up to positive affine transformation, as I understand it) from your preferences. In general, the idea is that we have some existing values or preferences, and we somehow assign utility values to things (“things”: events? world states? outcomes? something) by deriving them from our existing preferences/values. It’s not a trivial process, by any means, but ultimately the source here is the contents of our own brains.
Well, no. Utilitarian systems are based on a utility function (although I’m not aware of any requirement that it be immutable… actually, what do you mean by “immutable”, exactly?). Consequentialist systems don’t have to be utilitarian.
Even so, the origin of a utility function is not that mysterious. If your preferences adhere to the von Neumann-Morgenstern axioms, then you can construct a utility function (up to positive affine transformation, as I understand it) from your preferences. In general, the idea is that we have some existing values or preferences, and we somehow assign utility values to things (“things”: events? world states? outcomes? something) by deriving them from our existing preferences/values. It’s not a trivial process, by any means, but ultimately the source here is the contents of our own brains.
The problem is that most (all?) people’s preferences don’t.