Abstract: Two objects of valuation are said to be incommensurable when neither is better than the other, nor are they equally good. Hitherto, incommensurability has always been taken to be an ‘on-off’ matter. We argue instead that this relation comes in many degrees, which approximate different forms of commensurability, such as the ‘better than’ relation, to greater or lesser extents. In particular, there is an ‘almost better than’ relation, which we may easily conflate with the ‘better than’ relation itself.

We apply our account to the notorious ‘Continuum Argument’, made famous by Parfit (though he does not endorse it; indeed, he wants to resist it). The argument has us imagine a fairly large population of people with excellent lives that initiates a sequence of populations, each with a slightly lower life quality than its predecessor, but much larger and therefore putatively better. Eventually we reach a huge population of people with lives barely worth living. By the transitivity of ‘better’, the conclusion is that this population is better than the one we started with. Like many authors, we agree with Parfit that this conclusion is “repugnant”, and that the Continuum Argument must somehow be resisted.

Degrees of (in)commensurability to the rescue! At least some of the populations in the argument’s population sequence are merely almost better than their immediate predecessors. This makes it possible for the last population to be worse than the first population. We give a formal model of degrees of (in)commensurability. Among other things, it shows that ‘almost better’ is not transitive and thus that the repugnant conclusion can be avoided.