What McCulloch v. Maryland Got Wrong: The Original Meaning of 'Necessary' Is Not 'Useful,' 'Convenient,' or 'Rational'

Posted: 2 Sep 2022 Last revised: 16 Dec 2022

Date Written: August 31, 2022


McCulloch v. Maryland, echoing Alexander Hamilton nearly thirty years earlier, claimed of the word 'necessary' in the Necessary and Proper Clause: 'If reference be had to its use, in the common affairs of the world, or in approved authors, we find that it frequently imports that one thing is convenient, or useful . . . to another.'. Modern case law has translated that understanding into a rational-basis test that treats the issue of necessity as all but nonjusticiable; The Supreme Court has never found a congressional law unconstitutional on the ground that it was not 'necessary . . . for carrying into Execution' a federal power. Marshall, and Hamilton before him, were simply wrong in their empirical claim about the meaning of 'necessary'. We show, using founding-era dictionaries, an extensive corpus linguistic study of founding-era sources, and intertextual and intratextual analysis, that the original meaning of 'necessary' cannot plausibly be equated with 'convenient', 'useful', 'conducive to', or 'rational'. The case against Marshall and Hamilton’s linguistic claim is simply overwhelming. That does not mean that executory laws are 'necessary' only if 'indispensable', as the State of Maryland, echoing Thomas Jefferson, argued in McCulloch. While that strict meaning finds support in many of the sources that we examine, it does not constitute the best meaning in the specific context in which the term 'necessary' appears in the Constitution: A clause defining the incidental powers of agents. In that setting, familiar from the law of agency, a better fit is James Madison’s view that executory laws are necessary if they exhibit “a definite connection between means and ends, ”'showing some obvious and precise affinity' between the laws and the powers which they implement. In modern parlance drawn from another context, one might say that executory laws are necessary if they are congruent and proportional to the task to which they are put. Our principal goal in this article is not to defend this Madisonian view of necessity but simply to show that Marshall and Hamilton’s linguistic claim about the meaning of 'necessary' is false. We do not offer a comprehensive account of the original meaning of the Necessary and Proper Clause beyond this simple observation. But because McCulloch’s dictum has become canonical, we examine some of the leading cases involving federal power to see whether substituting a congruence-and-proportionality test for the test of usefulness, convenience, or rationality would make a large difference in outcomes. Holding all other elements and applications of doctrine equal, we find only a few cases in which getting right the original meaning of 'necessary' might make a difference – and those cases are already widely seen as anomalous under current doctrine. Nonetheless, there is value in getting such things right, including focusing attention on the extent to which the Necessary and Proper Clause rather than the Commerce Clause is the key to understanding the scope of federal power.

Suggested Citation

Calabresi, Steven G. and Lawson, Gary and Kostial, Elise, What McCulloch v. Maryland Got Wrong: The Original Meaning of 'Necessary' Is Not 'Useful,' 'Convenient,' or 'Rational' (August 31, 2022). Baylor Law Review, Forthcoming, Boston Univ. School of Law Research Paper No. 24, Northwestern Public Law Research Paper No. 22-34, Available at SSRN: https://ssrn.com/abstract=4207043 or http://dx.doi.org/10.2139/ssrn.4207043

Steven G. Calabresi (Contact Author)

Northwestern University - Pritzker School of Law ( email )

375 E. Chicago Ave
Chicago, IL 60611
United States

Gary Lawson

Boston University School of Law ( email )

765 Commonwealth Avenue
Boston, MA 02215
United States
617-353-3812 (Phone)

Elise Kostial

Yale University - Law School

Do you have negative results from your research you’d like to share?

Paper statistics

Abstract Views
PlumX Metrics