I'm a bit afraid to wade into this, but couldn't resist making two points  
about the assumption that "human desires tend to be unlimited" and the  
discussion of that assumption by Fred and John.  
  
A) To avoid a general claim about human desires, it is easier to simply  
assume that the range of demands on resources exceed the resources  
available. That is all that scarcity need imply. Also, the simplier  
assumption allows us to ask questions about how economic organization in  
different cultural settings create different incentive structures for  
adjudicating the relation between human desires and the resources available  
(to me, a culture is primarily an institutional configuration, rather than  
primarily a set of beliefs and values). On this, an excellent book is Barry  
Gordon's The Economic Problem in Biblical and Patristic Thought. He shows  
how the Judaic law provides several different solutions to the economic  
problem, and also ponders how Jesus and several patristic thinkers might be  
interpreted as responding to the problem of scarcity. The response to  
scarcity always occurs within a particular institutional context: markets,  
law, moral codes, etc.  
  
This raises the interesting question regarding axioms: are there axiomatic  
structures which are "institution free"? I doubt it. For example, do they  
assume particular kinds of contracts? Particular kinds of markets? Armen  
Alchain is great on this, of course, as are others.  
  
I don't have a problem suggesting that scientific progress is made by  
generalizing the axioms as we recognize their institutional limitations. But  
I expect that the less institution context included, the less the science  
has to offer to policy.  
  
Ross Emmett