> Do you the agree the question of whether or not the second derivative
> of an utility function exists is only meaningful if utility attains
> at least an interval measurement-scale level? That is, the canonical
> neoclassical model does indeed assume that utility is cardinal?


I'm no Markus, but the existence of some derivate does in nobloodyway
imply that utility is cardinal. You should know better than making
this kind of stupid claim; the measurement of utility is arbitrary
other than the ranking of the alternatives.
 
>> And I have no reason to believe 
>> that this is not a good approximation.
 
> The empirical reasons for doubting the canonical neoclassical model
> on these grounds lie in Amos Tversky's experiments and the empirical


Tversky's experiment only cast doubt on some of the simplest forms of
expected utility, not on neoclassical economics. You need another
angle to reject neoclassical economics.

[snip]

> D. Wade Hands shows that the Slutsky conditions do indeed imply that
> neoclassical theory is based on the existence of a conservative vector field,
> as I indicated in a previous post. But it's never been clear to me
> exactly what's conserved. As I stated, the conservation of the sum of
> income and utility is implied only if income effects are negligible.


Although adding income and utility might be logically possible
operation, the obtained result contains nothing interesting from econ
point of view, from phil of science, or on any other reasonable point
of view, for that matter.