Fact and Value

 In showing the complicated nature of the distinction between fact and value, Amartya Sen introduces the proposition, "a government should not increase the money supply more rapidly than the increase in national production."

That statement has a "should" in it so a simple application of Hume's guillotine suggests that this is a value statement. Thus it cannot be demonstrated to be so by facts. 

Yet of course people advocating it will have facts to discuss, as will those who oppose it.  I'll expand upon Sen's point here a bit. 

I believe that increasing the money supply beyond productivity in the underlying economy produces inflation and I hold inflation to be a bad thing because I believe the smooth functioning of a price system is a good thing. That good thing has various good consequences for buyers and sellers alike, and for their dependents. 

In principle IF you could persuade me that the smooth functioning of a price system is NOT a good thing, then I would have to rethink any propositions built on that one, including an anti-inflation lemma, and the theorem italicized above. 

Could I define "smooth functioning" in value-free terms?

Instead of attempting that, we could pursue a separate rabbit hole. My belief in the italicized value judgment is generally dependent on a theory about the causes of inflation. If in my view, inflation occurs because a government -- or more specifically a central bank -- increases some currency's supply more quickly than the underlying nation's productivity is increasing, and if inflation is a bad thing, then it plausibly follows that central banks should not do that. 

But the theory as to causation is more complicated than it appears, because actions can have more than one consequence. Suppose the increase in money supply creates or worsens inflation on the one hand, but allows for the provision of relief to the poor on the other hand. 

That situation raises a nest of further questions, both of fact and of value. 

I have gone further with the thought experiment than Sen did, and will go no further for now. But I do want to get back to his text. He has distinguished between "basic value propositions" and "non-basic value propositions." Obviously, the italicized one above is non-basic. But then Sen also raises in a footnote the possibility that they are all non-basic. 

He does this in an understated way, in a single sentence, "It is not asserted here that both the categories must be non-empty." Nice.

Translating that double-negative into a positive assertion: it may be that the category of "basic" unmixed value judgment is empty. The rabbit holes could go on forever.