For the Balkinization Symposium on Richard Primus, The Oldest Constitutional Question: Enumeration and Federal Power (Harvard University Press, 2025).
William Ewald
The
most striking thing about the bundle of ideas Richard Primus calls “Enumerationism”
is how distinctively American it is. (It will be helpful to capitalize
the word to distinguish the theory from the simple enumeration of federal
powers.) All constitutions apportion governmental powers, but the
near-universal American understanding that Article I, §8 is meant to limit
Congressional power (and related ideas, like his “internal limits canon,” which
says that the enumerated powers must add up to less than the powers of a
government of general jurisdiction) is not found in any other constitutional
system with which I am familiar; not, at any rate, in the same way. Why is
that? What lies behind this American exception? Is Enumerationism the
reflection of deep constitutional principle, or is it merely the result of defective
constitutional design?
Let’s
start with a couple of examples. In the United Kingdom, the fundamental
constitutional principle is the sovereignty of Parliament. Parliament
legislates; the executive acts; and executive action is subject to “judicial
review” (in the British sense) to ensure that it conforms to settled law. But
Parliament itself is free of judicial control, legally able to pass whatever
laws it wishes.
The principal constraints on its legislative power are customary and political. Numerous constitutional conventions, nowhere codified but well understood, bind the hands of the Government. The Prime Minister must have the support of the Cabinet and a majority of the Commons; proposed bills must be submitted to the House of Lords for scrutiny and suggested amendment; major legislation must first have been presented to the nation in the manifesto of the victorious party; and so on. In Primus's helpful taxonomy (pp. 37-39), these restraints are procedural. But Parliament itself is legally unbound. Plainly, in such a system, the problem of “Enumerationism” cannot arise, for the simple reason that there exists no enumeration of parliamentary powers.
It
might be objected that this is an entirely pointless example, since Britain
lacks two things that are central to American constitutionalism. It has no
written constitution, and (despite the devolved legislatures in Scotland,
Wales, and Northern Ireland) it is not a federal state. Perhaps Enumerationism is
what you get in a system based on those two things. Certainly, both loom large
in Primus’s discussions of the American debates, where Enumerationism is
presented as a great bulwark of federalism.
But
now consider a second example. Germany, on these two matters, is at the opposite
end of the constitutional spectrum from Britain. It has a detailed written
Constitution – the “Basic Law” – and the Basic Law emphatically declares
Germany to be a federal state.
Because
of those facts, the Basic Law inevitably contains an enumeration of
governmental powers. The powers are meticulously sorted into three classes:
those that belong to the federal government, those that belong to the Land
governments, and those that are shared. The enumeration is far more specific
than Article I, §8 of the U.S. Constitution. The powers to legislate concerning
(for example) hospitals, university education, telecommunications, railways,
internal waterways, nuclear energy, hunting, the protection of plant and animal
species, land-use, and so on are all specifically enumerated and assigned to
their proper sphere; so too, of course, are major powers related to the
economy, national defense, and foreign policy. [1] And in
contrast to the UK, all the instrumentalities of government – legislative
as well as executive – are subject to vigorous constitutional review at the
hands of the Constitutional Court.
Certain
things are notably absent from the Basic Law. It does not contain any
"sweeping clause" on the model of the American Necessary and Proper
Clause. That was intentional. German law has long been familiar with the
so-called “general clauses” that appear in the Civil Code, but the Hitler
regime had shown how such clauses in the Weimar Constitution could be abused
for dictatorial ends. Clauses of that sort were therefore left out of the Basic
Law.
Indeed,
the enumeration is accompanied by a fundamental principle of Rechtsstaat
constitutionalism, known as the “Vorbehalt des Gesetzes,” that serves
some of the purposes of the Tenth Amendment. The expression is not easy to translate (and,
like “due process” or “necessary and proper,” carries with it a lot of
doctrinal complexity), but the central idea is this: the state is not permitted
to act except on the basis of a law (“Gesetz”) – meaning, nearly
enough for present purposes, a statute enacted by the relevant democratic legislature.
[2]
This
legal structure creates a revealing contrast to what Primus calls the
"internal limits canon" of the American debate. In the United States,
it is largely taken for granted that the purpose of enumeration is to limit
the powers of Congress. From that assumption (which Primus argues is mistaken),
one quickly gets into disagreement about what, exactly, is forbidden by the
enumeration of powers. And that, in turn, leads to the sort of reasoning David
Schwartz mocks as "The ThereMustBeSomething Rule": if Enumerationism
is not to be empty (it is said), then “there must be something” that the
Article I enumeration of powers places beyond the legislative competence of
Congress. [3] That, in turn, is a variant of the “broccoli argument,” familiar
from the debates over Obamacare: “if the enumeration of powers permits Congress
to enact Obamacare, then what is to prevent it from requiring Americans to eat
broccoli?” In German constitutionalism, the reasoning is quite different. The
enumerated constitutional powers of the Basic Law are precisely that: they are grants
of power. They are not themselves understood to be a limitation on
power. Limitations on power are instead given by a number of constitutional
mechanisms, including the Vorbehalt des Gesetzes (which, by the way, is
drilled into even low-level civil servants: bureaucrats are not permitted to
boss people around unless they can point to a statutory authorization). That
means that, if parliamentary legislation on the eating of broccoli is not plausibly
authorized by the enumerated powers of the Basic Law, the Bundestag cannot legislate
on the matter; and if the Bundestag cannot legislate, then the bureaucracy, lacking
a statute, cannot order you to eat your broccoli. But then (one naturally
wonders) what is to be done if legislation on the eating of broccoli seems
desirable? In such a case, the constitutionally required procedure is clear. You
must amend the Basic Law.
That
then raises the question, how can the Basic Law be amended? The process is carefully
crafted and has two principal requirements. First, a constitutional amendment
may not violate any of the fundamental human or civil rights guaranteed in the
opening articles of the Basic Law. Those guarantees (roughly equivalent to an
expanded version of our Bill of Rights) are placed beyond amendment altogether.
(In Primus’s taxonomy, these are external controls; that is, they
explicitly forbid the state to do certain things.) Secondly, the amendment must
secure a two-thirds supermajority both in the Bundestag and the Bundesrat (the
second legislative chamber, representing the Land governments). In Primus’s
taxonomy, this is a procedural limitation. And as a procedural matter,
amending the Basic Law is roughly as difficult as overriding an American
president's veto – difficult, but by no means impossible. That means that the
ordinary functions of government and the legislative capacities of the federal
and Land legislatures can be adjusted and clarified and redistributed as
circumstances require. So long as there is a solid, democratic, national
consensus in favor of the change, and so long as none of the fundamental rights
is violated, it can be done. The Basic Law has been frequently amended in this
way, roughly every year or so, to bring its powers into alignment with current constitutional
practice and to make precise exactly what is entailed by the language of the
Constitution; most of the time, the amendments are technical adjustments and
uncontroversial.
A
couple of observations are now in order. First, despite the constitutional differences
between Britain and Germany, neither possesses anything like the constitutional
theory Primus calls Enumerationism. Both systems take for granted that
legislative powers must be adaptable, capable of change in response to changed
circumstances; both provide mechanisms for adapting to change while, at the
same time, securing fundamental constitutional restraints on government;
neither views the granting of legislative power as itself imposing limits. (Of
course, any time a power is granted, the question can arise whether a
particular exercise is outside the scope of the power, and in that sense, there
is always an internal limit; but that is trivial.) Secondly, the restraining mechanisms they
deploy are (in Primus's taxonomy) a mixture of the procedural and the external.
The precise mix varies, but neither system emphasizes Primus's third category of
internal restrictions; still less do they discuss anything resembling
his "internal limits canon." So far as I am aware, Enumerationism is
a distinctively American peculiarity, and other constitutional regimes do not
rely upon it – most likely because, as Primus says, internal restrictions are
neither precise about the boundaries they set nor especially effective in
maintaining them.
But
that leaves us with a puzzle. If Enumerationism is an American constitutional oddity
and if, as I have argued, it is not the consequence either of a written
Constitution or of federalism, what explains its prominence in American
constitutional theory? I suggest it is the product of three interlocking
things, each of which is a further American peculiarity.
First,
American constitutionalism is tethered to a text that was drafted nearly a
quarter of a millennium ago and that, even at the time, contained ambiguities
and evasions and signs of hasty drafting. No other constitutional system is
like that. (I set aside religious legal systems that base themselves on the
interpretation of a sacred text.) Secondly,
the constitutional text is exceptionally difficult to amend, largely because of
the hurdles of the amendment process in Article V.
Those two
facts open a gap between the enumerated powers of 1787 and the tangible
realities of government today, and constitutional theory is forced into awkward
and undignified contortions to paper things over.
This
has happened repeatedly. In the 1930s, there was a national consensus in favor
of the New Deal, but how was it to be reconciled with the 1787 text?
Essentially, by squeezing the regulatory powers into the commerce clause and by
taking a relaxed interpretation of the separation of powers. In 1964, there was
a national consensus in favor of civil rights, but what clause from 1787
allowed Congress to pass the Civil Rights Act? The answer was essentially the
same, and the power to prohibit racial discrimination became an exercise of the
power to regulate interstate commerce. What about Obamacare and national health
insurance? That is justified by the power to tax.
I try
to explain these things to the foreign LLM class every year and always get a
room full of puzzled expressions. These readings of the constitutional text are
flagrantly implausible ways to understand the language of 1787; they seem
natural only after several years of expensive postgraduate education. Naturally
enough, this way of reading the 1787 Constitution prompts questions, and it is
those questions that lead into the morass of Enumerationism and the debate
about broccoli. From there, it is a short step to the doctrinal contortions
visible in cases like Lopez and Morrison. The American tradition
is one of adaptation by subterfuge: change the practice and leave the text
unaltered. The result is a lack of analytical precision about the foundations
of constitutional law. In other countries, such problems are handled with
greater transparency. In Britain, national health insurance was proposed as
part of the Labour Party's 1945 electoral manifesto, resoundingly approved by
the voters, and duly enacted by Parliament. In Germany, the language of the
Basic Law is routinely updated to bring the constitutional text into alignment
with constitutional reality, with the fortunate consequence that a foreigner
can obtain a reasonably accurate understanding of the current powers of the
German government by carefully reading the text of the Basic Law. That is
obviously not possible with Article I, §8.
But
there is a further point. Lurking behind these two peculiarities and giving
them much of their force is a third: the sacralization of the events of 1787.
No other nation possesses quite such an extensive literature devoted to
celebrating the Wisdom of the Framers or to defending the rightness of
virtually every clause of the Constitution, from the Preamble to the electoral
college to the Connecticut Compromise. The mythology is that the entire
document was carefully thought through and purposefully designed: if you do not
like the enumeration of powers, well, take it up with Mr. Madison.
The
problem with this adulatory literature is not, of course, that it praises the
Framers, but that it overlooks the main thing they did during the summer of
1787, which was to disagree. The common understanding holds that Enumerationism
was present from the beginning, was a matter of deliberate design, and that a
desire to restrain the national government was the reason for enumerating
congressional powers. That story of the Framing is at the heart of
things and underpins the rest of the Enumerationist construction; the other peculiarities
depend upon it. That means it is
necessary to look closely at what happened behind the closed doors of the
Philadelphia Convention: not, notice, because today’s constitutional law must
follow the original intent of the Framers – that is a separate issue – but
because one wishes to assess the truth of a widespread and influential
historical assumption.
Primus’s
discussion of the Revolutionary-era history in his opening chapters (and
especially in Chapter 2) is surefooted, and I have little to add. He has read the
documentary evidence carefully and points out that the list of powers that
emerged from the Committee of Detail was at least as likely to have been intended
to confer power on Congress as to restrict it. That reading of
the Committee of Detail is plainly correct, certainly for Wilson. (I would
supplement his discussion with another recent piece of meticulous close
reading, John Mikhail’s article on “The Progress of the Prerogatives,” which is
directly on point. [4])
Primus
is careful not to overstate his case. He points out that the evidence is
incomplete, that some delegates (Madison, Wilson) saw enumeration primarily as
a conferring of power, but that others (Randolph, Mason) saw it primarily as a
limitation and would likely have endorsed some version of Enumerationism; he
also emphasizes that the polished final text did not unambiguously resolve the
matter. [5] For the purposes of his argument, that is enough.
That
last point can perhaps be put more emphatically. Primus's discussion brings
together essentially everything that is known to have been said about
enumeration of powers by the 1787 Framers, before, during, and after the
Convention; and the crucial point to observe is just how little there is. A
handful of delegates expressed carefully considered views; a few others made
occasional brief observations; the great majority said nothing at all. Even the musings of Madison, who gave the
matter the most sustained attention, would scarcely fill more than a few pages,
and his remarks display numerous doubts and considerable hesitation.
That
should not be surprising. The writing of a constitution was a novel enterprise,
and it was not clear – it could not have been clear – to anyone in the
room exactly what to expect or how the system would function in practice. The
majority of the delegates probably had no settled view about what we now call
Enumerationism, and could not have told you, even at the time, exactly what had
been agreed to. The problem is not that we do not know what the Framers
intended, but that we know that they achieved no settled, collective intent, and
that even the deepest thinkers among them regularly changed their minds.
NOTES
[1] The
principal enumeration occurs in GG Articles 70-74, though other powers are scattered
throughout the document. My listing here is only a selection; dozens of powers are
explicitly mentioned.
[2]
The principle is doctrinally located in Article 20, §3 of the Basic Law, though
the historical roots go back much further. The doctrine is entwined with the
separation of powers, the principle of democracy, and the rule of law; as one
might guess, the details are intricate. I am unaware of a comprehensive
treatment in English and have relied on the discussions in the Handbuch des Verfassungsrechts (Matthias Herdegen, Johannes Masing, et al., eds., 2021).
[3] Richard Primus, The Oldest Constitutional Question:
Enumeration and Federal Power 245 (2025).
[4]
John Mikhail, The Progress of the Prerogatives,
63 Am. J. Leg. Hist. 196 (2023).
[5] See, for instance, his remarks at pp. 73-76 and 114-115.
William Ewald is the O’Brien Professor of Comparative Law
and Professor of Philosophy, University of Pennsylvania Carey Law School. You can
reach him by e-mail at wewald@law.upenn.edu.