A recent article in the Texas Review of Law and Politics, by my Independence Institute colleague Rob Natelson (who is also a constitutional law professor at the University of Montana) applies some cultural context to the original meaning of the spending clause.
At the time of the Founding, the legal principles of "fiduciaries," "servants," or "agents" were well-known. A fiduciary/servant/agent was expected to act impartially, not for his own self-interest, on behalf of the principals. For example, if a person were the trustee of an estate for three children, the trustee would be required to give each of the children equal earnings from the trust. (Unless there were a good reason not to. i.e., the one child was independently wealthy from other income; one child had made an improvident marriage).
Natelson argues that the fiduciary principles are incorporated into the Constitution, and provide the basis for "rational basis with bite" review of congressional spending. (That is, until the 1936 Butler case is over-ruled, and Congress is again required only to spend in furtherance of enumerated powers, rather being allowed to spend for anything it chooses.)
For example, Congress spends money to establish a National Institute of Obesity Research in Mississippi. The spending will provide a much greater financial benefit to Mississippi than to any other state, but the purpose of the NIOR is clearly to benefit the entire nation, by improving everyone's health. This spending passes the rational basis with bite test.
In contrast, consider the earmark which former Senator Conrad Burns created, in order to provide funds for capital construction at the University of Montana Law School. This would appear to be special-interest spending for the benefit of a single state, not for the nation as a whole. Such spending would fail the rational basis with bite test. If, perhaps, Congress made a finding that some states were chronically underlawyered, and provided capital construction grants for expanded law school facilities in all such states, then the spending might pass the Natelson test. (The above examples are my own, not Natelson's, although he does cite the Montana earmark as an example of pork.)
Rob Natelson, the Spending Power, and Fiduciary Duty: