One of the great strengths of federalism as a structure of constitutional governance is its flexibility. Federalism offers this flexibility at two distinct points in the constitutional life cycle. During the initial design phase, federal institutions can be calibrated to accommodate a wide variety of political considerations, from demands for autonomy by cultural or linguistic minorities, to concerns over efficiency and innovation in the generation and implementation of public policy, to civic fears concerning the protection of liberty. Perhaps even more important is the oft-remarked capacity of federal structures to adjust themselves to changing conditions once a constitutional system is up and running.1 On this view, changes in social and political facts on the ground lead inevitably to changes in the goals and interests of national and subnational governments. These changes lead in turn to evolution in the grounds upon which such governments contest for power, producing corresponding adjustments to the allocation of power between national and subnational governments and sometimes even to the nature and goals of the federal system itself.
In Partisan Federalism,2 Professor Jessica Bulman-Pozen argues that American federalism has recently undergone just such a change, evolving from its initial design as a system of contestation between state and national polities to one of contestation between and among geographically concentrated partisans of the two major national political parties. She welcomes this change, praising constitutional doctrines that enhance the porosity of state borders and advocating further legal change to facilitate the formation by citizens of multiple, transborder subnational political identities. I have no serious quarrel with Bulman-Pozen’s descriptive claim, but I do have reservations about the normative desirability of the changes she documents. These changes, I fear, have the potential to undermine the most significant design goal of American federalism: the protection of liberty.
Unlike the federalism practiced in many other states, American federalism has from the beginning opted for a relatively decentralized model not to accommodate ethnonational claims of geographically concentrated minorities, but to protect liberty by institutionalizing a system of permanent intergovernmental contestation.3 At a minimum, federalism is said to achieve this goal by assigning to the states the function of checking or resisting dangerous uses of national power and by giving them power to do so effectively. Somewhat more ambitiously, the federal system could be understood as authorizing states to improve the lives of Americans by resisting exercises of national power that do not directly threaten liberty but are merely bad on the merits.
At the structural level, then, the important question is: What kind of entities must American states be to perform these functions consistent with the constitutional plan? In my judgment, effective execution of the constitutional plan requires states to be political communities — the federalism constructed by the U.S. Constitution is meant to pit political communities, rather than other kinds of communities, against one another. This conclusion follows from three features possessed uniquely by formal political communities: their power, their completeness, and their autonomy.
Power. The United States government is powerful, and its exercise of power is democratically legitimate. Successful resistance to this kind of power generally can be mounted only by powerful and equally legitimate actors. In our system, political communities are by definition the principal bearers of this kind of power. To be sure, the function of checking national power can be and is exercised by other kinds of organizations — political parties, private associations, the organized press, civil society groups — but they do not possess formal, official power, and such power is, in some circumstances, the only tool capable of effectively resisting national misbehavior. These other manifestations of the checking function are pluralist, but they are not federal.
Completeness. Ideally, the kind of power wielded by governments should not be entrusted to an incomplete community. To do so is, in Madisonian terms, to hand over the levers of power to a “faction,”4 an outcome that much of the institutional architecture of the U.S. Constitution is meant to prevent. Entrusting rule to political communities lessens the risk of factional control because, unlike other kinds of communities, a political community is capable, at least in principle, of comprehending all the varieties of interests that a human community can contain. Its decisions are decisions of a genuine polity, conferring on them an important form of presumptive legitimacy. This helps explain why we often worry about the wisdom and legitimacy of conferring the full weight of governmental authority on narrow and incomplete political entities such as capital cities (Buenos Aires, Mexico City) or, in the international setting, microstates (Monaco, Liechtenstein) — the same intuition that makes many Americans balk at the idea of elevating the District of Columbia to the status of a state. In all these communities, the range of interests and human experience may be too narrow to impart the kind of legitimacy a complete political community enjoys.
Autonomy. Finally, a political community is, in principle, the kind of institution most capable of acting with true autonomy. Indeed, the idea of an independent, free-standing political community is, in our Lockean political tradition, the very definition of a genuinely autonomous actor, fully in charge, as much as fortune permits, of its own destiny. The autonomy enjoyed by political communities allows them in turn to make independent and self-generated judgments about the two most important considerations informing the contestatory dynamic of federalism: the welfare of their citizens and the performance of the federal government. The autonomy that political communities, and only political communities, enjoy thus offers a stronger possibility than other forms of communal organization for (a) improving the welfare of subnational citizens and (b) effectively checking national power.
Seen in this light, Bulman-Pozen’s story is a story of the diminishment of what states are. She claims that the identity of political communities defined by state boundaries has collapsed into a kind of checkerboard of national partisan identity, at least for purposes relevant to federalism. This is plausible, and it may very well be descriptively accurate. I have worried about this phenomenon myself: in a recent article, I argued that national political parties have served as conduits through which national politics “colonizes” state politics, thereby establishing the agenda and range of positions available to state political actors.5
Where I think we part company is over whether these developments pose any kind of problem. Bulman-Pozen argues, optimistically, that enhancing the porosity of subnational borders and communities to facilitate the formation of partisan state and individual political identities would mark a beneficial accommodation to the current reality. I am less sanguine. The pertinent question, it seems to me, is not whether “partisan federalism” is on its own merits an adequate, or sufficient, or even a superior form of federalism. The question, rather, is whether states that function politically as outposts of the national political parties are capable of performing the checking and resistance functions that the existing constitutional plan of federalism contemplates for them. Considered in terms of the three characteristics of political community reviewed earlier, there are grounds for doubt.
Power, of course, is not an issue under partisan federalism, or any other kind: states will have at their disposal a considerable amount of official, legitimate, governmental power so long as the system remains one that is recognizably federal. The nub of Bulman-Pozen’s argument, however, is that the levers of state power are being operated by something other than autonomous political communities — that, as a matter of contingent fact, states have become essentially branch offices of the national political parties. If true, this development seems to me worrisome from the point of view of the goals American federalism is meant to achieve.
One important reason why this is the case is because a state community whose ideas and whose very identity are furnished by national political parties is an incomplete community, one that by hypothesis lacks the breadth and texture of human interests that a true political community contains. Like any faction, such a community is unlikely to generate a suitably broad and complete range of policy positions for potential state adoption; on the contrary, it will be limited by definition to those that appear on the agendas of national political parties. These positions are especially likely to reflect a relatively narrow range of views, since national parties by definition are compelled to adopt positions and commitments that appeal to national majorities — a very significant limitation. Certainly a partisan state community will be less capable than a complete political community of generating positions and preferences different from those made available on the menu of national party commitments. Bulman-Pozen correctly observes that state affiliates of national political parties can and sometimes do adopt partisan positions that differ from those of the national parties,6 but no evidence of which I am aware suggests that state affiliates are capable of taking positions that do not lie comfortably within a range of opinion defined by parameters established at the national level and oriented toward the possibility of appealing to the median national voter.
Perhaps even more worrisome is that a partisan state community lacks the autonomy of a true political community. Such a community is less likely than a complete political community to be able to generate robust and independent conceptions of the common good, both of the local populace itself and of the national polity. Bulman-Pozen notes, again correctly, that the generation of independent conceptions of the common good does not depend on some Romantic conception of states as organic, Hegelian communities. It is more than sufficient that states contain differing proportions of nationally distributed subpopulations, or that differing and contingent historical experiences produce path-dependent divergences of opinion among state populations.7 But a state community that takes its identity from national political parties is not, even in this more limited sense, a fully independent political community, and thus lacks the degree of autonomy that American federalism seems to demand. Its close relation to and overlapping identity with national political parties make it dependent on those parties to supply both a policy agenda and a set of positions to be taken up concerning items on that agenda. At the end of the day, what is missing from such a community is a demos; it has no resources from which to draw a conception of a distinct local or national common good — precisely the kinds of conceptions that federalism contemplates that states will, in appropriate circumstances, offer in opposition to positions generated at the national level.
Even this situation might not undermine the contemplated operation of federalism if state polities could, whenever necessary, extract themselves from alignment with national partisan positions to assert a wider range of more distinctive, homegrown positions when necessary to recognize, check, or resist bad national policy. But can they? As thinkers from Aristotle to John Stuart Mill have observed, citizenship is a faculty that requires regular exercise. It is not at all clear that if state residents stop thinking of themselves as, and acting like, citizens of a subnational political community, they can simply dust off and reacquire the habit when circumstances demand. If not, then the only available grounds for state resistance to national policy choices and actions may be partisan, yet partisan grounds may be too limited in scope to fulfill the role that state power is meant to serve in the American system of federalism.
In sum, I agree with Bulman-Pozen that it is pointless to cling to idealized notions of a federalism that no longer exists, if indeed it ever did. The rise in the United States and around the globe during the late twentieth century of regimes of robust, judicially enforced human rights may well be testament to the demise of structural mechanisms as meaningful constitutional constraints on government misbehavior. Still, I think it is too early to abandon the American constitutional ideal of redundant systems of liberty protection, and we might therefore consider how a meaningful state check on national power might be preserved, even in a diminished form, before we agree to see it ushered off the constitutional stage.
* Bridget and Thomas Black SUNY Distinguished Professor, SUNY Buffalo Law School, The State University of New York.
1. See William S. Livingston, Federalism And Constitutional Change 11—12 (1956); Nathalie Behnke & Arthur Benz, The Politics of Constitutional Change Between Reform and Evolution, 39 Publius 213, 216 (2009).
2. Jessica Bulman-Pozen, Partisan Federalism, 127 Harv. L. Rev. 1077 (2014).
3. The idea is Madison’s. I describe the Madisonian account more fully in James A. Gardner, Interpreting State Constitutions 81—83 (2005).
4. The Federalist No. 10 (James Madison).
5. James A. Gardner, The Myth of State Autonomy: Federalism, Political Parties, and the National Colonization of State Politics, 29 J.L. & Pol. 1, 17 (2013).
6. The canonical work is Robert S. Erikson, Gerald C. Wright, & John P. Mciver, Statehouse Democracy (1993).
7. Gardner, Interpreting State Constitutions, supra note 3, ch. 2.