Law Review
Northwestern Law
Northwestern University Law Review : Colloquy : 2008 : Somin

A Floor, Not a Ceiling: Federalism and Remedies for Violations of Constitutional Rights in Danforth v. Minnesota

By Ilya Somin[*]

[download pdf]

Few doubt that states can provide greater protection for individual rights under state constitutions than is available under the Supreme Court's interpretation of the Federal Constitution.  More difficult issues arise, however, when state courts seek to provide greater protection than the Court requires for federal constitutional rights.  Can state courts impose remedies for violations of federal constitutional rights that are more generous than those required by the Federal Supreme Court? That is the issue raised by the Court's recent decision in Danforth v. Minnesota.[1]  In a 7-2 decision joined by an unusual coalition of liberal and conservative justices, the Court decided that state courts could indeed provide victims of constitutional rights violations broader remedies than those mandated by federal Supreme Court decisions.  I contend that this outcome is correct, despite the seeming incongruity of allowing state courts to deviate from the Supreme Court's interpretation of the Federal Constitution.  The Supreme Court should establish a floor for remedies below which states cannot fall. But there is no reason for it to also mandate a ceiling.

Part I briefly describes the facts and background to Danforth.  In Part II, I provide a doctrinal justification for the Supreme Court's decision.  It makes sense to allow state courts to provide more generous remedies than those mandated by the federal courts in cases where restrictions on the scope of remedies are not imposed by the Constitution itself, but are instead based on policy grounds.  State courts can legitimately conclude that these policy grounds are absent or outweighed by other considerations within their state systems, even if they are compelling justifications for restricting the scope of remedies available in federal courts.  State courts are in a better position to weigh the relevant tradeoffs in a state legal system than federal courts are.

Part III explains the potential policy advantages of allowing interstate diversity in remedies, most importantly inter-jurisdictional competition and an increased ability to provide for diverse citizen preferences and local conditions across different parts of the country.  The optimal remedy for a constitutional rights violation in New York may well be different from the optimal remedy for one that occurs in Mississippi.

I.  Danforth and its Origins

In its 2004 decision in Crawford v. Washington, the Supreme Court held that the Confrontation Clause of the Sixth Amendment requires that defendants have the right to "confront" witnesses against them in person at a trial.[2]  Three years later, the Court held that states are not required to apply this rule retroactively to pre-Crawford convictions.[3]  The combination of these two rulings set the stage for Danforth.

In 1996, eight years before Crawford, a Minnesota Court convicted Stephen Danforth of criminal sexual assault against a minor. The six-year-old victim did not testify at the trial, but the jury "saw and heard a videotaped interview of the child."[4]  Danforth challenged his conviction on the grounds that the use of the videotape at his trial violated the Confrontation Clause, as interpreted in Crawford.  The Minnesota Supreme Court rejected his argument, holding that Minnesota courts were forbidden by federal Supreme Court precedent to "give a Supreme Court decision of federal constitutional criminal procedure broader retroactive application that that given by the Supreme Court."[5]  Previous Supreme Court decisions had held that newly announced rules of constitutional criminal procedure do not apply retroactively unless they fall into two narrowly defined categories: rules that forbid state authorities to criminalize the conduct in question and "watershed" rules that "implicate the fundamental fairness of the trial."[6]  The Minnesota Supreme Court ruled that Danforth's case fell outside the scope of both of these categories and concluded that state courts were therefore barred from giving him retroactive relief for this violation of the Sixth Amendment.[7]

In a 7-2 decision, the United States Supreme Court overruled the Minnesota Supreme Court's ruling that state courts are forbidden to grant retroactive relief for violations of constitutional rights in cases where the Federal Supreme Court does not require them to do so.[8]  Ironically, the Court's ruling gives state courts greater latitude then they would have been allowed under the Minnesota Supreme Court's approach.[9]  The majority, held that the case turned on a question of state law remedies, not federal constitutional law.[10]  Both the four most liberal justices, and the three most conservative—Scalia, Thomas, and Alito—voted in favor of this result.

In a forceful dissent, joined by Justice Anthony Kennedy, Chief Justice John Roberts argued that remedies for violations of federal constitutional rights are indeed a matter of federal law, and that the Constitution requires nationwide "uniformity of decisions throughout the whole United States" on all federal constitutional issues.[11]

At least at first glance, it seems as if Roberts has a point. After all, remedies for violations of constitutional rights are elements of the rights themselves.  For example, the Fifth Amendment right to "just compensation" for a taking of private property necessarily includes the right to sue the government for compensation if it takes a citizen's property without paying for it.[12]  There is, therefore, some intuitive appeal to the claim that they must be uniform "throughout the whole United States."  As Roberts put it, the majority's approach allows "the Federal Constitution . . . to be applied differently in every one of the several States," thus creating the kind of "disuniformity" that the Constitution was in part established to prevent.[13]  However, there are good reasons to permit such "disuniformity" that are largely ignored by both the dissenters and the majority.

Roberts's logic is correct insofar as it requires states to provide a minimal level of remedies for violations of federal constitutional rights—a floor. But his logic does not apply with equal force to allowing the Supreme Court to impose a ceiling.

II.  Federalism and Policy-Based Limits on Remedies for Rights Violations

Neither the majority nor the dissent in Danforth ever seriously considered the fact that limits on the retroactivity of remedies for rights violations do not rest on constitutional mandates but on policy concerns.[14]  When the courts refuse to remedy an admitted constitutional rights violation because of policy considerations that weigh against retroactivity, they are, in effect, subordinating a constitutional concern to a policy preference.

Yet this is precisely what happened in Whorton v. Brockling, the 2007 Supreme Court case that held that states are not required to apply the Crawford Confrontation Clause decision retroactively.[15]  In Whorton, the Court noted that the procedure followed by the state court in convicting the defendant had violated his Confrontation Clause rights by interpreting the Clause in a way inconsistent with the intent of "the Framers" of the Bill of Rights.[16]  However, the Court refused to apply this ruling to pre-Crawford cases because the old rule—although based on a flawed interpretation of the Sixth Amendment—did not significantly increase the chances of an inaccurate conviction and therefore did not outweigh the policy considerations weighing against retroactive application of new Supreme Court decisions under Teague.[17]  Justice O'Connor, the author of Teague, has characterized its presumption against retroactivity as an example of how "federal courts exercising their habeas powers may refuse to grant relief on certain claims because of 'prudential concerns' such as equity and federalism."[18]  Other relevant "prudential concerns" weighing against retroactivity include the need to ensure finality in criminal proceedings[19] and the danger of recidivism by offenders released prematurely if their convictions are invalidated.

One might legitimately question whether it is ever permissible for the Court to allow "prudential concerns" to trump constitutional rights.  After all, a crucial purpose of enshrining any interest as a constitutional right is precisely to ensure that it overrides ordinary policy considerations, "prudential" or otherwise.  The tradeoff between a constitutional right and other objectives that might conflict with it is not for the courts to decide.  That decision has already been made by the framers and ratifiers of the Constitution. There are good reasons to believe that the policy judgments of the supermajorities that produce constitutional amendments are likely to be better than those of the Supreme Court.[20]  Significantly, the framers and ratifiers did not include a non-retroactivity exception in the Bill of Rights.  Perhaps the Court should respect that "prudential" decision rather than subordinating the enforcement of constitutional rights to its own interpretation of prudence.

Nonetheless, I do not pursue this more radical criticism of non-retroactivity here.  Justifiably or not, the Court often weakens remedies for constitutional rights when it perceives weighty prudential considerations on the other side.  This is particularly true when a new precedent overrides a long-established decision that government officials have relied on in good faith.  Most famously, the Court adopted this approach when it ruled in Brown v. Board of Education II that southern states need only desegregate their education systems with "all deliberate speed," rather than immediately[21] —despite the fact that continued segregation in what turned out to be a lengthy interim period would lead to an ongoing violation of constitutional rights.

At the same time, the Supreme Court should not have the same kind of power to impose its "prudential" policy preferences on the states as it does when it enforces actual constitutional rights.  Federal courts may indeed be in the best position to weigh conflicting policy priorities in federal legal proceedings (assuming that such weighing a legitimate judicial function at all).  State courts, however, are better placed to weigh these issues in the context of state proceedings, as in Danforth.

Minnesota courts presumably have greater knowledge about the impact of retroactivity on their own future proceedings than the justices of the Federal Supreme Court.  They also have greater incentives to use their knowledge effectively. Should they make a ruling that imposes undue costs on the Minnesota legal system, Minnesota political authorities could curb the state courts' powers by choosing new judges with different views or by passing jurisdiction-limiting legislation.  In the twenty-two states with elected judiciaries, including Minnesota, judges are subject to electoral checks.[22]  In other states, judges are appointed by the governor or the legislature, sometimes with participation by "merit commissions."[23]  Both methods give judges at least some incentive to consider policy considerations important to their states' judicial systems.  By contrast, Minnesota officials and voters have much less influence over the selection of federal judges.

Undoubtedly, both electoral and non-electoral constraints on state judges have significant flaws. For example, widespread political ignorance may greatly reduce the ability of voters to monitor state judges' performance and deny reelection to those who have reached poor decisions.[24]  Even so, state judges are clearly more accountable to their states' voters and government officials than federal judges, and therefore have stronger incentives to give due consideration to state-level policy concerns.

The superior knowledge and incentives of state judges relative to federal judges may have little significance in cases where state discretion is limited in order to enforce federal constitutional rights.  There, the Constitution does indeed seek to impose "uniformity" of the kind emphasized by Chief Justice Roberts in his dissent.  However, the superior position of state judges is very relevant to situations where the supposed justification for federal imposition is simply a matter of "prudential" policy considerations.  Here, superior knowledge and incentives counsel in favor of letting state courts set their own rules.

This is especially true with respect to policy arguments against retroactivity that do not apply to state courts invalidating their own state's convictions.  For example, Justice O'Connor listed "federalism and comity" among the "prudential concerns" justifying the Teague rule.[25]  Obviously, these considerations simply do not apply to a state court reviewing the validity of state convictions within its own jurisdiction.  By definition, there is no issue of comity in cases like Danforth since comity problems only arise in a situation where one sovereign refuses to respect the decision of another.  Nor can there be any "federalism" problem when one state court overrules the decision of another court from its own state.

The Danforth majority did recognize that "federalism and comity considerations" do not apply to state courts reviewing their own state's convictions, and also noted that "finality of state convictions is a state interest, not a federal one."[26]  It even emphasized that there is a "fundamental interest in federalism that allows individual States to define crimes, punishments, rules of evidence, and rules of criminal and civil procedure in a variety of different ways—so long as they do not violate the Federal Constitution . . . ."[27]  This bedrock principle of federalism, the Court concluded, cannot be constrained by "any general, undefined federal interest in uniformity."[28]  However, it failed to draw the more general conclusion that state courts, not federal courts, are in the better position to decide policy issues arising from state judicial rules.  Thus, there is a fundamental difference between Supreme Court decisions that enforce federal constitutional rights and those that limit such enforcement on the basis of "prudential" policy considerations.

III.  Interstate Variation in Remedies and the Benefits of Federalism

Although I have argued that state courts are generally better placed to evaluate policy concerns about state court remedies than federal courts, it is theoretically possible that Chief Justice Roberts is right to argue that federally imposed "uniformity" in remedies is desirable.[29]  Perhaps this is an exception to the general rule that state courts are better judges of state legal rules than federal courts.  However, there is good reason to believe that allowing interstate variations in remedies captures some of the standard benefits of federalism.  It allows us to reap more of the benefits of interstate diversity, mobility, and competition.

A.  Diversity

The ability to satisfy the diverse preferences of populations in different parts of the country is a classic rationale for federalism.  Both objective local conditions and citizen preferences may differ from one state to another.  It makes sense to allow states to adopt divergent policies in order to take account of such differences.[30]

This point applies to diversity in remedies as much as to other types of policy diversity among states.  There are many reasons why the optimal remedy for a constitutional rights violation in one state might be different from the optimal remedy in another.  For example, rights violations might be a more common problem in some states than others, which might justify stronger remedies in order to increase deterrence in the state where government officials are more prone to violate the right in question.  Similarly, public opinion in State A might value a particular right more than that in State B. A divergence in remedies (with a more generous remedy in State A) could help satisfy the preferences of voters in both states.  A uniform federal rule, by contrast, would leave at least one state's voters relatively dissatisfied.

There is also a strong case for interstate variation with respect to the specific question of retroactivity at issue in Danforth.  If a state has a long, egregious history of violating a particular constitutional right, retroactive application of remedies might be needed in order to root out the systemic consequences of past rights violations.  By contrast, this need is likely to be less pressing in a case where the state has rarely violated the right in question.  To take one of the most notorious examples in American history, many states—particularly in the South—systematically violated the rights of African-American criminal defendants for decades.[31]  In states with this kind of record, retroactive remedies might be more defensible than in states with less history of abuse.  Additionally, the costs, as well as the benefits, of retroactivity are also likely to vary between states.  In some states, for example, there may be less danger of recidivism and less need to insist on finality of convictions than in others.

These benefits of diversity do not undercut the case for establishing a federally mandated "floor" for constitutional remedies.  In the absence of such a floor, states could deny remedies for rights violations entirely, thereby negating the main purpose of creating enforceable constitutional rights in the first place.  However, there is no comparable justification for a federally imposed ceiling.  If state courts, for their own reasons, decide that they want to provide broader remedies for constitutional rights violations than the Supreme Court requires, they may well have good diversity-based reasons for doing so.

B.  Interstate Mobility and Competition

A second crucial rationale for decentralized federalism is the ability of citizens to "vote with their feet" for the state government whose policies they prefer.[32]  People dissatisfied with the policies of their state can vote with their feet against them by migrating to a different jurisdiction whose policies they find more congenial.  If states are free to adopt diverging policies, there will be more options for potential foot voters.  Moreover, competition for taxpaying residents and firms gives states incentives to adopt policies that will attract migrants and convince current residents to stay.[33]

However, foot-voting and competition may not apply as readily to interstate differences in remedies as to other policies.  Given the costs of moving, few people or firms are likely to migrate merely because one state has better remedies than another for violations of constitutional rights.  This, in turn, reduces the likelihood that states will try to compete with each other on this dimension.  Nonetheless, there might be exceptions to this generalization.  Residents who are particularly concerned about the danger of a given rights violation may take remedies into account in their moving decisions.  In the Jim Crow era, when federal courts were extremely lax in enforcing constitutional protections for African-American criminal defendants, black migrants did indeed take into account the fact that northern criminal justice systems treated them more favorably than southern ones.[34]  On the other hand, excessive remedies that overdeter law enforcement might be curtailed by migration on the part of residents seeking to move to areas with lower crime rates.

On balance, foot-voting and interstate competition are less compelling rationales for allowing variations in remedies than diversity. But they have some force, nonetheless.


There is good reason for the Supreme Court to establish a floor for remedies for federal constitutional rights violations.  On the other hand, there is no comparable justification for it to also establish a ceiling that state courts are not allowed to exceed.  To the extent that the Supreme Court's Danforth decision tracks this distinction, it should be welcomed.

At this time, the extent that the Court's ruling applies outside the Sixth Amendment context remains unclear.  Presumably, the Court's reasoning applies to all cases where state courts provide more generous remedies for violations of federal constitutional rights than the Supreme Court mandates.  However, the Danforth decision fails to provide a comprehensive explanation of the right-remedy distinction and also fails to explicitly consider the question of how broadly its ruling will apply.  But the Court did hold that "the remedy a state court chooses to provide its citizens for violations of the Federal Constitution is primarily a question of state law" and is therefore not subject to a federal court-imposed ceiling.[35]  This suggests that its logic applies to all such remedies, not just those involving criminal proceedings.  Certainly, the justification offered here for the floor-ceiling distinction in Danforth applies with equal force to similar cases involving other constitutional rights.



1.  128 S. Ct. 1029 (2008) (link).

2.  Crawford v. Washington, 541 U.S. 36, 68–69 (2004) (link).

3.  Whorton v. Bockting, 127 S. Ct. 1173 (2007) (link).

4Danforth, 128 S. Ct. at 1033.

5.  Danforth v. State, 718 N.W.2d 451, 456 (Minn. 2006), rev'd, 128 S. Ct. 1029 (2008).

6.  Teague v. Lane, 489 U.S. 288, 310–12 (1989) (plurality opinion) (link). Although the so-called "Teague rule" was first outlined in a plurality opinion joined by only four justices, it was accepted by the majority of the Court in a later decision that followed soon afterwards.  See Penry v. Lynaugh, 492 U.S. 302, 313–19 (1989) (endorsing and applying the Teague rule).

7Danforth, 718 N.W.2d at 460–61.

8Danforth, 128 S. Ct. at 1038–47.

9.  Several other state supreme courts have held that the Teague rule does not constrain state courts acting to remedy constitutional rights violations that occurred during state postconviction legal proceedings.  See, e.g., State v. Whitfield, 107 S.W.3d 253, 266–68 (Mo. 2003); Colwell v. State, 59 P.3d 463, 470–71 (2002) (per curiam); State ex rel. Taylor v. Whitley, 606 So. 2d 1292, 1296–97 (La. 1992); Cowell v. Leapley, 458 N.W.2d 514, 517–18 (S.D. 1990); Daniels v. State, 561 N.E.2d 487, 489 (Ind. 1990); cf. State ex rel. Schmelzer v. Murphy, 548 N.W.2d 45, 49 (Wisc. 1996) (holding that state courts are not required to follow the Teague rule, but adopting it of its own volition as a matter of state law).

10.  Danforth, 128 S. Ct. at 1045–47.

11 1053 (Roberts, C.J., dissenting) (quoting Martin v. Hunter's Lessee, 14 U.S. (1 Wheat.) 304, 347–48 (1816)) (internal quotation marks omitted).

12Cf., Armstrong v. United States, 364 U.S. 40 (1960) (upholding such a suit as part of the right to "just compensation") (link).

13Danforth, 128 S. Ct. at 1053–54.

14.  Professor Michael Dorf's recent critique of Danforth—the only previous published analysis of Danforth by a legal scholar that I know of—also fails to consider the importance of this distinction to the outcome of Danforth. See Michael Dorf, Did Justice Stevens Pull a Fast One?  The Hidden Logic of a Recent Retroactivity Case in the Supreme Court, Findlaw's Writ, Feb. 25, 2008, (link).

15.  127 S. Ct. 1173 (2007). 1182. 1181–84.

18.  Withrow v. Williams, 507 U.S. 680, 699 (1993) (O'Connor, J., concurring in part and dissenting in part) (emphasis added) (link).

19See Danforth v. Minnesota, 128 S. Ct. 1029, 1040–41 (2008) (noting that the Teague rule is partly based on the need to respect the "finality of state convictions").

20.  For a more detailed discussion of this point, citing relevant literature, see Ilya Somin, "Active Liberty" and Judicial Power: What Should Courts Do to Promote Democracy?, 100 Nw. U. L. Rev. 1827, 1855–59 (2006) (link).

21.  349 U.S. 294, 757 (1955) (link).

22.  Stephen J. Choi, G. Mitu Gulati & Eric A. Posner, Professionals or Politicians: The Uncertain Empirical Case for an Elected Rather than Appointed Judiciary, 3 (U. of Chi. John M. Olin Law & Econ. Working Paper Series, Paper No. 357, (2d Series) 2007), available at (link).


24See, e.g., Ilya Somin, Voter Ignorance and the Democratic Ideal, 12 Critical Rev. 413, 415–38 (1998) (discussing the impact of political ignorance).

25.  Withrow v. Williams, 507 U.S. 680, 699 (1993) (O'Connor, J., concurring in part and dissenting in part) (link).

26.  Danforth v. Minnesota, 128 S. Ct. 1029, 1041 (2008) (emphasis in original).



29See supra text accompanying notes 10–12.

30.  For summaries of the diversity rationale for decentralization, see, for example, Michael W. McConnell, Federalism: Evaluating the Founders' Design, 54 U. Chi. L. Rev. 1484, 1493–95 (1987); Ilya Somin, Closing the Pandora's Box of Federalism: The Case for Judicial Restriction of Federal Subsidies to State Governments, 90 Geo. L.J. 461, 464–68 (2002).

31See, e.g., Randall Kennedy, Race, Crime, and the Law 76135 (1997) (detailing this history).

32See Ilya Somin, Political Ignorance and the Countermajoritarian Difficulty: A New Perspective on the Central Obsession of Constitutional Theory, 89 Iowa L. Rev. 1287, 1344–50 (2004) (discussing "foot voting" and contrasting it with conventional ballot box voting).

33.  For the benefits of interstate competition, see Somin, supra note 30 at 468–71; see also Thomas R. Dye, American Federalism: Competition Among Governments 1–33 (1990).

34See, e.g., Daniel M. Johnson & Rex R. Campbell, Black Migration in America: A Social and Demographic History 84–85 (1981) (discussing this motive for black migration to the North in the early twentieth century).

35.  Danforth v. Minnesota, 128 S. Ct. 1029, 1045 (2008).


Copyright 2008 Northwestern University      

Cite as: 102 Nw. U. L. Rev. Colloquy 365 (2008).

Persistent URL:


(back to top)