Title: Three Conceptions of the Measurement of Value
Abstract: There are many different ways in which things can be good - or in other words, many different values. But goodness comes in degrees: some things are better than others. Is it possible, at least in principle, to measure how good something is? If so, values would have to possess certain features that would make such measurement possible. There are three conceptions of these features that philosophers have at least implicitly entertained: the model of extensive measurement; the decision-theoretic model; and the model of difference measurement. According to what is argued here, the first model is not a plausible conception of the measurement of value; the second model is plausible for certain sorts of extrinsic value (such as "instrumental value" or "news value"); and the third model is plausible for intrinsic values. Additive models of aggregation should not be interpreted on the model of extensive measurement, but as accounts of how certain values are related to other values. For example, an additively aggregative model may provide the best account of how the values grounding each of the various reasons for and against the available courses of action are related to the value of the overall choiceworthiness of these courses of action.
