Incommensurability in Risk Analysis

I am reading "Should we risk it?" by Kammen & Hassenzahl. On page 150, I saw an unexpected word incommersurability, which is common in valuation literature (people use it to criticize monetary valuation--it is very challenging to convert values in different dimention, may it be environmental or socioeconomic, into $). Here is Kammen and Hassenzahl,

"By incommensurability we mean that the data available do not provide a sufficient basis to compare the risks from different chemicals or risk pathways, or that the risks themselves are so different (e.g., the potential risks of skydiving versur the riks of cancer from pesticide residues on foods) that a common metric if often arbitary."

Another reason to use MCDA :)

P.S. The currency of risk analysis seems to be, well, risk. On page 13 of the same book, e.g. they cited a table of Risks that increase chance of death by .000001 (on in one million), including: one chest x-ray taken in a good hospital (which I had to do twice in the past year unfortunately), travelling 10 miles by bicycle, and living 2 months in Denver...