Emory Law Journal

Lessons from California’s Recent Experience with Its Non-Unitary (Divided) Executive: Of Mayors, Governors, Controllers, and Attorneys General
Vikram David Amar Professor of Law and Associate Dean for Academic Affairs, University of California at Davis. Some of the themes raised and arguments made in this Essay were first presented, in tentative and undeveloped form, in various on-line columns the author wrote for the website findlaw.com. These on-line columns are viewable on the author’s Findlaw archive page, which is available at http://writ.news.findlaw.com/amar.

It is often said that one of the great advantages of a federalist system is that states can operate as laboratories of democracy, experimenting with common law and statutory frameworks in ways that provide useful policy information to other states as well as the federal government. 1However fairly, the “laboratory” metaphor for policy experimentation is often attributed to Justice Louis Brandeis, who wrote that “[i]t is one of the happy incidents of the federal system that a single courageous State may, if its citizens choose, serve as a laboratory; and try novel social and economic experiments without risk to the rest of the country.” New State Ice Co. v. Leibmann, 285 U.S. 262, 311 (1932) (Brandeis, J., dissenting). The utility of this framework is not limited to the common law or experiments by legislatures; it applies with equal, albeit underappreciated, force to matters of constitutional law.

Thus, in a symposium dedicated to examining the meaning and future of the federal “unitary executive,” the experience of states—almost all of which reject a unitary executive model—warrants some inquiry. Recent episodes in the most populous state, California, involving struggles over two of the most prominent and consuming controversies of our day—recognition of gay marriage and how best to deal with unprecedented public budget shortfalls—serve as the focus of my Essay. These episodes highlight both the pitfalls and the possibilities of a divided (that is, plural) executive model.

Lesson # 1: Dispersal of Core Executive Power Can Generate Dubious and Ultimately Harmful Claims of Power by Executive Officials in Ways That Increase Disorder and Cynicism during Critical Times

A. San Francisco Mayor Gavin Newsom and the Origins of California’s Gay Marriage Saga

“California, like virtually every other state, employs a divided executive in that [the] State Constitution divides the executive powers of state government among several” actors, both horizontally at the state level and vertically within the geographical subdivisions that comprise the state. 2Gilb v. Chiang, No. 34-2008-80000026, at 8 (Cal. App. Dep’t 29 Super. Ct. Mar. 18, 2009). The tensions and disputes created by this allocation of authority arise most frequently between various statewide constitutional officers, but the battles over gay marriage—one of the biggest storylines in the state and the nation over the last decade—demonstrate that the division of executive power between state officials on the one hand and regional or county officials on the other generates its own potential for mischief.

In the spring of 2004, in direct and overt defiance of state statutes limiting state-sanctioned marriage to heterosexual couples, 3 See, e.g., Cal. Fam. Code § 308.5 (West 2009) (providing that “[o]nly marriage between a man and a woman is valid or recognized in California”), invalidated by In re Marriage Cases, 183 P.3d 384 (Cal. 2008). Gavin Newsom, the mayor of San Francisco (both a county and a city) began directing the Clerk of San Francisco County to make whatever changes were necessary “in order to provide marriage licenses on a non-discriminatory basis, without regard to gender or sexual orientation.” 4Lockyer v. City and County of S.F., 95 P.3d 459, 464 (Cal. 2004). Newsom took this position based on his view that state and federal equal protection and due process principles required it. 5 See id. at 464–65, 465 n.4. Under Newsom’s direction, the county issued licenses to thousands of same-sex couples whose weddings were performed at city hall by county bureaucrats and elected officials. 6 Id. at 465.

The issuance of these licenses quickly prompted a number of lawsuits brought by private groups seeking to enforce California state statutes forbidding same-sex marriage. 7 E.g., id. at 465–66. California trial courts declined to grant the immediate injunctive relief sought by the private plaintiffs, 8 E.g., id. at 466. largely on the ground that even if Mayor Newsom had been improperly violating state law, the plaintiffs would not suffer substantial “irreparable injury” because of it. 9 See, e.g., Thomasson v. Newsom, No.CGC-04-428794, at 1 (Cal. Super. Ct. County of S.F. Feb. 25, 2004) (order denying temporary restraining order).

A few days later, California Attorney General Bill Lockyer filed papers directly with the California Supreme Court, making three requests: (1) that the court issue an immediate cease-and-desist order directing the county clerk to stop issuing new licenses; (2) that the court declare the marriages that had already been performed invalid based on allegedly illegal actions of the mayor; and (3) that the court take up and resolve the question of whether, as advocates of same-sex marriage like Newsom had argued, California’s constitution prohibits discrimination against same-sex couples in the context of marriage. 10 Lockyer v. City and County of S.F., 95 P.3d 459, 466 (Cal. 2004). The attorney general noted that the court could choose to grant the cease-and-desist order and invalidate the marriages already performed, but then allow lower courts to decide the state constitutional equal protection and due process issues even before they were addressed by the state supreme court. But the attorney general indicated that it was his belief that the better course was for the court to examine the merits right away, without waiting for the lower courts to rule, insofar as the due process and equal protection questions were adequately framed and presented. Id.

Mayor Newsom’s only imaginable defense for his actions was his stated belief that the state statutes violated the California and federal constitutions and that the oaths he swore when he took office permitted—perhaps even compelled—him to disregard such statutes that conflict with these higher laws. 11 See supra notes 3–5 and accompanying text (discussing Mayor Newsom’s decision to issue thousands of marriage licenses to same-sex couples).

One specific provision of the California constitution posed a problem for Mayor Newsom. Article III, Section 3.5 of the California constitution admonishes that an “administrative agency . . . has no power . . . [to] refuse to enforce a statute, on the basis of it being unconstitutional unless an appellate court has made a determination that such statute is unconstitutional.” 12 Cal. Const. art. III, § 3.5.

Section 3.5 appears to have been added to the state constitution to provide an orderly judicial process to resolve questions about the constitutionality of state laws. It essentially directs state agency officials who believe a statute is unconstitutional to obey the statute until appellate courts have decided the constitutional question, rather than disregard the will of the legislature in the name of constitutional conscience. 13 See id.

Attorney General Lockyer relied heavily on Section 3.5 in his request for an immediate cease-and-desist order, 14 See Lockyer, 95 P.3d at 466. and his argument seemed to have the force of the law behind it. Even if state statutes banning same-sex marriage did violate state due process or equal protection principles, Section 3.5 on its face appeared to preclude the mayor from disregarding these statutes in the absence of invalidation by a state appellate court.

Indeed, the only real question concerning the applicability of Section 3.5 in the minds of many observers was whether San Francisco is an “administrative agency” within the meaning of this section. The better view is that San Francisco is an “agency” for these purposes. It was, after all, the San Francisco county clerk who issued the licenses, 15 Id. at 465. and the state constitution elsewhere defines “counties” as “legal subdivisions of the state.” 16 Cal. Const. art XI, § 1. A prominent treatise on the California constitution summarizes the prevailing view that counties “serve as regional agencies and instrumentalities of some state-level functions and are thus treated as legal and operational subdivisions of the state government itself.” 17 Joseph R. Grodin, Calvin R. Massey & Richard B. Cunningham, The California State Constitution: A Reference Guide 182 (11th ed. 1993).

If one moves beyond the text of Section 3.5 to its evident purposes, the actions of San Francisco officials in the same-sex marriage context are hard to justify. The licenses each county grants are valid not just in that county, but rather throughout the state and perhaps in other states as well. 18Whether the marriage licenses are valid in other states depends on the states’ interpretations of the federal Constitution’s Full Faith and Credit Clause. See U.S. Const. art. IV, § 1. Thus, the desire to avoid disorder and confusion that underpins Section 3.5 is implicated in a significant way by the issuance of marriage licenses.

This is not to say that Section 3.5’s approach to the problem of what executive branch agencies should do when they think a statute is unconstitutional—namely, enforce the statute until an appellate court acts—is the only reasonable answer. To be sure, disagreements that arise within an executive branch as to whether an action is constitutional must be settled through some process. In the federal system, the President usually has the last word, and sometimes the President’s word need not even agree with that of the courts. But all others within the Executive Branch must fall in line with the President—an agency head (despite his or her oath to uphold the Constitution) cannot contravene the President’s position. For example, a regional EPA official or a regional federal prosecutor could not, relying on his oath, defy the President’s directive to enforce a federal statute that the official believes is unconstitutional. Such localized authority would be intolerably chaotic and inefficient, with regional agency representatives frustrating the public’s expectations by refusing to follow decrees from Washington.

If the federal system can opt in this way for a “unitary executive,” states certainly can as well. One way of understanding Section 3.5 is that California has embraced at least some aspects of the federal model in this respect; the California constitution and cases recognize the Governor and attorney general as the state’s chief law enforcement officers, and Section 3.5 does not allow lesser state agencies to diverge from the positions taken by these state executive leaders, at least until an appellate court has directed them to do so.

Thus, just as at the federal level regional representatives and agency heads must fall in line with the President, Section 3.5 suggests that at the state level mayors must fall in line with the Governor when it comes to law enforcement. 19 See supra note 12 and accompanying text. Whether a Governor or an attorney general could have a decent claim to be free from the constraints of Section 3.5 because they are not “agencies” (just as the President might claim a right to disregard congressional enactments that he believes are unconstitutional regardless of what an appellate court has said), 20 See infra notes 51–68 and accompanying text. a lower-level state executive official, like a mayor, simply cannot make such a claim.

Cities in California (and other states) do enjoy some state constitutional protection to decide for themselves certain local matters free from state control. 21 See Cal. Const. art. XI, § 5. But the California constitution understandably limits these local autonomy realms to “municipal affairs.” 22 Id. See also Grodin, Massey & Cunningham, supra note 17, at 187–90. “Municipal affairs” are understood to include things such as setting salaries for city employees, contracting for city construction projects, taxing local residents, and the like. 23 See Cal. Const. art. XI, § 5. But because of the significant intra- and interstate implications of marriage, the issuance of marriage licenses does not fall within the “municipal affairs” category. Instead it is part of a uniform statewide administrative scheme.

This becomes clear if we consider how the system would have reacted if Mayor Newsom’s objections to state statutes concerning marriage had been based not on a perceived conflict with the California constitution, but rather only on his policy sense of what types of marriage ought to be recognized. Suppose, for example, that a California mayor sincerely believed that persons who are fourteen years old should be allowed to marry (in violation of state law) but grounded his argument on policy views rather than on an interpretation of state constitutional principles? No one would ever think that a mayor and a city could go with their own policy views based on local powers over “municipal affairs.”

Possibly because they sensed some weakness in their position for the reasons just discussed, Mayor Newsom and his lawyers eventually asserted that state statutes prohibiting same-sex marriage might violate the federal, as well as the state, Constitution, and that this changed things. 24 See Lockyer v. City and County of S.F., 95 P.3d 459, 467 (Cal. 2004). In essence, Newsom’s lawyers argued that Section 3.5 would violate federal constitutional supremacy 25 See U.S. Const. art. VI (“This Constitution . . . shall be the supreme Law of the Land. . . any Thing in the Constitution or Laws of any State to the Contrary notwithstanding.”). if it were invoked to require a state official to do something he believed violated federal law. 26 See Lockyer, 95 P.3d at 467.

Although it was an interesting argument, upon reflection I reject the position taken by Newsom’s lawyers. Whenever a genuine disagreement arises between the institutional actors within the state as to the legality of a state statute and there exists a good-faith belief that the statute might, in the end, be upheld, nothing in the federal Constitution prohibits a state from setting forth an orderly process to administer a state law that ultimately might be invalidated as violating the federal Constitution. If a situation arose in which a state constitution directed state officials to enforce laws that all reasonable persons would agree, based on clearly settled law, run afoul of the federal Constitution, a federal due process problem might very well arise. But that is not what Section 3.5 asks of mayors and other state agencies. Instead, it directs that, when reasonable people could disagree about how a state or federal constitutional challenge will be resolved by the judiciary, agencies should wait for a court of appeals decision on the relevant issue before declining to implement the will of the legislature as expressed in the statute. 27 See Cal. Const. art. III, § 3.5.

For example, under the argument I make here, the situation would have been different if the U.S. Supreme Court had already held, in a case from another state, that the federal Equal Protection Clause requires that same-sex couples have equal access to marriage. If, in that situation, Governor Arnold Schwarzenegger had issued a directive telling local officials to continue honoring heterosexual but not homosexual marriages, that directive could be ignored by state agencies, notwithstanding Section 3.5 of the state constitution. Under such circumstances, the state would not have been able to defend its attempts to compel mayors to comply with the directive on the ground of promoting intra-executive stability and good order. It would have been plain that instead of seeking intra-executive stability, the state was simply trying to frustrate federal rights and nothing more. Such obvious intent to disregard federal rights might directly violate the federal Supremacy and Due Process Clauses.

But so long as measures like Section 3.5 are understood to apply only to situations in which a reasonable person could argue that the state statute in question does not violate the federal Constitution, there is no federal bar to requiring state agencies to implement a state statute unless and until an appellate court invalidates it. 28 Id. For these reasons, San Francisco needed an appellate court decision in its favor before granting licenses for same-sex marriages. And no appellate court had ruled on California’s heterosexual-only marriage statutes at the time of Mayor Newsom’s actions. Thus, Section 3.5 suggests San Francisco should have enforced the marriage laws, as they then existed, until an appellate court had ruled on them.

The California Supreme Court promptly saw problems with the mayor’s actions. Within days of receiving the attorney general’s request, the court enjoined San Francisco officials from issuing any more same-sex marriage licenses—but only after thousands of same-sex couples had been granted licenses. 29 Lockyer, 95 P.3d at 467. Halting further issuance of the licenses, the court asked for briefing on the narrow question of whether a mayor has the power to decline to follow state marriage statutes when he believes the statutes to be unconstitutional. 30 Id.

The overwhelming majority of informed analysts correctly predicted that the mayor’s actions would be unanimously rebuffed by the California high court. 31 See, e.g., Bob Egelko, Court Halts Gay Vows; SURPRISE RULING: Newsom Says the City Will Make a Strong Case. State Justices Take S.F. Case on Narrow Constitutional Issue, Promise Quick Decision, S.F. Chron., Mar. 12, 2004, at A1. And yet, the mayor’s actions caused tremendous confusion and instability as to the administration of the state marriage regime and the validity of marriage licenses already issued. It is possible that many years from now, people might look back on the episode and argue that, as a political matter, Mayor Newsom’s actions kicked off a sequence of events that, either through federal litigation or a ballot measure, ultimately ended up guaranteeing marriage rights for same-sex couples in California. But without the benefit of a crystal ball predicting the future, I cannot help but conclude that the mayor’s actions generated much undesirable disorder and disarray on a matter of statewide concern. Further, the disorder he created was not confined to the jurisdiction that elected him. The validity of the marriage licenses issued by the County of San Francisco posed a quandary for officials and others throughout the whole state.

This kind of disruption is a product of a very divided state executive branch. Within the federal government, disagreements over the administration of nationwide programs—for example, federal criminal laws or EPA or FAA licensing schemes—are, or at least can easily be, resolved at a centralized location within the Executive Branch. This means that various regions of the country are subject to different enforcement regimes only to the extent that the highest officials in the administration decide that having different rules in different places is for some reason beneficial in the short run. In other words, under the federal system, any temporary uncertainty or disorder would be a product of deliberate choice in the moment—for which the President and his advisors would be fully accountable—rather than a function of institutional design. 32 See Printz v. United States, 521 U.S. 898, 922 (1997) (noting the importance of unity in the Executive Branch and invalidating Congress’s attempt, through the Brady Act, to transfer federal law enforcement “responsibility to thousands of CLEOs in the 50 States, who are left to implement the program without meaningful Presidential control”). In short, the Gavin Newsom episode seems like “Exhibit A” for those who argue that a unitary executive at the federal level promotes coherence, stability, and political accountability for important executive decisions.

B. The 2008–2009 (and Beyond?) Budget Battles in California Between Arnold Schwarzenegger and John Chiang

Perhaps the only state news item that has rivaled same-sex marriage in prominence in California headlines over the last few years is the massive, unprecedented state budget fiasco that began in summer 2008. The causes of the budget crisis are contested and manifold, but the magnitude of the predicament was undeniable. Beginning in the summer of 2008, the state was facing a yearly deficit of approximately $40 billion, 33Jim Christie, UPDATE-2 California Delaying Tax Refunds Amid Cash Crisis, Reuters, Jan. 16, 2009. a staggering amount for an entity that is not authorized to borrow for yearly operating costs and that had annual general fund revenues in the ballpark of $100 billion before the recession. 34 See, e.g., State of Cal. Dep’t of Fin., 2008–09 Final Budget Summary, Sept. 23, 2008, available at http://www.documents.dgs.ca.gov/osp/GovernorsBudget/pdf/fbudsum_0809.pdf.

In response to this crisis, Governor Arnold Schwarzenegger undertook many emergency measures designed to save the state multiple billions of dollars a year. These measures included a plan to pay most state employees only the federal minimum wage until after a state budget had been passed 35 See, e.g., Michael Rothfield, The State; Gov. Lays Off 10,000 Workers; Part-Timers Are Fired. Most Others Would Get Minimum Wage. The Controller Vows to Block the Pay Cut., L.A. Times, Aug. 1, 2008, at A1. and to “furlough” hundreds of thousands of state workers a few days each month. 36 Id.

Managing an unprecedented fiscal crisis is difficult even for the most functional of governments, but Republican Governor Schwarzenegger’s efforts were hindered by many other state executive officials—most notably California State Controller John Chiang, a Democrat. The Governor had to sue the elected controller twice (and had to defend himself in a third action in which the controller joined union plaintiffs suing the Governor) 37 See Gilb v. Chiang, No. 34-2008-80000026 (Cal. App. Dep’t 29 Super. Ct. Apr. 13, 2009) (final judgment); Prof’l Eng’rs in Cal. Gov’t v. Schwarzenegger, No. 34-2008-800000126-CU-WM-GDS (Cal. Super. Ct. County of Sacramento Feb. 11, 2009) (judgment regarding verified petition); Schwarzenegger v. Chiang, No. 34-2009-900000158-CU-WM-GDS (Cal. Super. Ct. County of Sacramento Feb. 10, 2009) (minute order regarding notice of related cases). to get Chiang to implement the directives that the Governor and his Department of Personnel Administration had issued regarding temporarily reduced pay.

The arguments Controller Chiang asserted for refusing to abide by the Governor’s edicts fell into three main categories. First, he argued that the Governor’s fiscal plan was unwise, unnecessary (given state cash flow projections), and practically unfeasible. 38 See, e.g., Gilb, No. 34-2008-80000026, at 13 (Cal. App. Dep’t 29 Super. Ct. Mar. 18, 2009). Second, echoing the tack taken by Mayor Newsom in San Francisco, he argued that the Governor’s proposed course of action would violate federal employee wage laws and thus run afoul of the Supremacy Clause. 39 Id. at 10. Finally, he argued that the Governor’s orders trammeled his state constitutional independence to administer the payment of state monies and to oversee his office, which the state constitution set up as distinct from the Governor within the executive branch. 40 See Schwarzenegger, No. 34-2009-900000158, at 2–3 (minute order).

To date, each of these arguments has been rejected by the courts in thoughtful trial court opinions, as most observers had expected. However, the cases are still winding their way up the appellate ladder, all the while continuing to magnify the confusion and uncertainty of the budget “fixes” currently in place. 41 See, e.g., Kevin Yamamura, State Controller: I Won’t Carry Out Governor’s Furlough Order, Sacramento Bee, Jan. 22, 2009, at 4A, available at http://www.sacbee.com/capitolandcalifornia/story/1563773.html. These cases also have the potential to shift significant power during this episode to the members of the state judiciary, who might have felt (and still feel) emboldened to second-guess the Governor’s plans since the challenges to those plans come not just from employee labor unions but also from statewide constitutional officeholders. As discussed below, 42 See infra notes 74–83 and accompanying text. there are particular reasons to think the controller’s legal assertions of independence and discretion were (and still are) weak, but they were plausible enough to have interjected additional tension and unpredictability into an already dysfunctional budget process. 43 See Yamamura, supra note 41. As was true with the episode involving Mayor Newsom, the chaos created by the Schwarzenegger–Chiang conflict has no federal analogue. President Obama will likely never be in the position of needing to sue his Treasury Secretary over policy or legal disagreements. 44 See, e.g., Paul Brest, Sanford Levinson, J.M. Balkin & Akhil Reed Amar, Processes of Constitutional Decisionmaking: Cases and Materials 674 (4th ed. 2000) (noting that the independence of the Federal Reserve may raise its own problems and solutions). Whatever the problems may be with the way the federal government has responded to the economic crises of the last year, internecine battles in the Executive Branch do not appear to have been the monkey wrench in D.C. that they seem to have been in Sacramento. Chalk up another point for the unitary executive, at least in this fiscal management setting.

Lesson #2: Divided Is Conquered? Ambitious Assertion of Plural Executive Power Can Undermine the Entirety of Executive Authority and Independence

A second interesting facet of the Mayor Newsom episode is the particular way in which the California Supreme Court repudiated his actions. 45Lockyer v. City and County of S.F., 95 P.3d 459, 459 (Cal. 2004). Importantly, it did so by rejecting not just local executive power, but rather all executive power to decline to enforce state statutes. 46 Id. at 486. Thus, the state regime’s dispersal of executive power in a way that allowed for Mayor Newsom’s independent actions may have diminished the sum total of state executive power relative to the amount of power under a more unitary system. Logically, nothing dictates that divided executive powers should lead to diminished executive powers, but it is interesting to note that many of the most ardent proponents of a unitary executive regime at the federal level (like Justices Scalia and Thomas) also read executive power quite broadly. It may therefore be no coincidence that divided executive schemes incline some jurists (like those on the California Supreme Court) to construe the totality of executive power somewhat narrowly. Apart from all of that, as I explain below, the aggressive maneuver undertaken by a participant in the divided executive arena (Mayor Newsom) may have, in the short run, hurt the very cause (same-sex marriage rights) he was ostensibly championing. 47 See infra notes 69–72 and accompanying text.

To be sure, the outcome of the California Supreme Court’s appraisal of the mayor’s decision to authorize same-sex marriage licenses in violation of state statutes 48 Lockyer, 95 P.3d at 487. came as no surprise. Almost all legal commentators who were not involved in the litigation predicted that the court would reject Newsom’s position. 49 E.g., Egelko, supra note 31.

What was less foreseeable than the high court’s decision itself, however, was its reasoning. Exactly how would the California Supreme Court explain its result? And exactly what—if anything—would the court say and do about the 4,000 or so marriage licenses that San Francisco had already issued to gay couples?

In some respects, the court’s opinion exemplified judicial restraint and care. The court explicitly steered clear of expressing any views on the “ultimate” question in the California gay marriage controversy—whether the statutes that define marriage as only between a man and a woman are consistent with California’s constitution 50 See, e.g., Lockyer, 95 P.3d at 486–87. (a question later decided by the court and discussed later in this Essay).

But in contrast to its careful avoidance of that issue, the court was quite expansive in its discussion about the limits on executive power that made Mayor Newsom’s actions directing the issuance of same-sex marriage licenses untenable. In rejecting Newsom’s claim, the court could simply have said that local executive officials who are part of a statewide hierarchical system, such as mayors, lack such power. But in its opinion, the court did not explicitly home in on local executive officials, and it went further to suggest that no executive official—local or statewide—could ever have such power. 51 See, e.g., id. at 485–86, 487 (asserting that because statutes are always presumed constitutional and the statutes at hand were not patently unconstitutional, their directives must be followed by all public officials).

Was the court’s seemingly broad rejection of all executive power to decline to enforce statutes sound? Dissenting Justice Kathryn Werdegar believed that it was not and criticized the majority’s expansive tone and language:

Under these circumstances, I see no justification for asserting a broad claim of power over the executive branch. Make no mistake, the majority does assert such a claim by holding that executive officers must follow statutory rather than constitutional law until a court gives them permission in advance to do otherwise. 52 Id. at 510 (Werdegar, J., dissenting).

Consider, in this regard, the court’s citation to a mid-nineteenth century U.S. Supreme Court case for the proposition that “to contend that the obligation imposed on the president to see the laws faithfully executed implies a power to forbid their execution is a novel construction of the constitution, and entirely inadmissible.” 53 Id. at 464 (quoting Kendall v. United States, 37 U.S. 524, 613 (1838)). But history is more complicated than the California court made out. Let us look at the founding era. While Marbury v. Madison 545 U.S. 137 (1 Cranch) (1803). does say that “it is emphatically the province and duty of the judicial department to say what the law is,” 55 Id. at 177. “emphatically” is not “exclusively.” And many a president has historically, and I think properly, asserted a power to decline to enforce a congressional statute that he believed to be unconstitutional—a power deriving from both the oath he took to support the Constitution and his duty to see that the law, especially the Constitution, is “faithfully executed.” 56 U.S. Const. art. II, § 3.

Take, for example, the 1798 Alien and Sedition Acts, through which Congress effectively sought to outlaw criticism of the incumbent Federalists—an obvious First Amendment violation. 57 See John C. Miller, Crisis in Freedom: the Alien and Sedition Acts 75–76 (1951) (describing the politically-motivated Federalist congress as passing the Sedition Act to “ensure the perpetuation of their party’s control”). Even though the federal courts had upheld the Act, 58 See id. at 129–30 (describing a federal circuit court’s jury charge that resulted in upholding the constitutionality of the Sedition Act). President Thomas Jefferson pardoned everyone who had been convicted under the statute. 59 James Magee, Freedom of Expression 22 (2002). To Jefferson, the question was not simply what decisions courts had reached or might reach regarding the constitutionality of the Act, but what his own independent constitutional conscience dictated.

In 1832, six years before the Court decided Kendall v. United States, the case cited by the California justices, 60See supra note 53 and accompanying text. President Andrew Jackson vetoed a bill on constitutional grounds—again using independent judgment despite a prior ruling indicating that the bill would have been judicially approved. 61See Jonathan D. Varat, William Cohen & Vikram D. Amar, Constitutional Law: Cases and Materials 35 (13th ed. 2009). There, the Supreme Court had already upheld a similar bill against constitutional challenge. 62 Id. But Jackson wrote:

The Congress, the Executive, and the Court must each for itself be guided by its own opinion of the Constitution. Each public officer who takes an oath to support the Constitution swears that he will support it as he understands it, and not as it is understood by others. It is as much the duty of the House of Representatives, of the Senate, and of the President to decide upon the constitutionality of any bill or resolution which may be presented to them for passage or approval as it is of the supreme judges when it may be brought before them for judicial decision. 63 Id.

More recent authority from the U.S. Supreme Court also indicates that the President may sometimes have the power to reject or refuse to enforce a law that he believes is unconstitutional. For instance, as then-Assistant Attorney General (and later acting Solicitor General) Walter Dellinger observed in a 1994 memo, “[o]pinions dating to at least 1860 assert the President’s authority to decline to effectuate enactments that the President views as unconstitutional.” 64 Brest et al., supra note 44, at 56. Dellinger noted that “[m]ore recently, in Freytag v. Commissioner, all four of the Justices who addressed the issue agreed that the President has ‘the power to veto encroaching laws . . . or even to disregard them when they are unconstitutional.’” 65 Id. (citation omitted).

Nor are executive branch officials institutionally or practically incompetent to perform the kind of sophisticated constitutional interpretation that would justify executive independence and refusal to enforce statutes. The California court observed:

Certainly attorneys have no monopoly on wisdom, but a person trained for three years or more in a college of law and then tempered with at least a decade of experience within the judicial system is likely to be far better equipped to make difficult constitutional judgments than a lay administrator with no background in the law. 66Lockyer v. City and County of S.F., 95 P.3d 459, 490 (Cal. 2004) (quoting S. Pac. Transp. Co. v. Pub. Utils. Comm’n, 556 P.2d 289, 321 (Cal. 1976)).

  • Yet many executive officers throughout American history have been among the most talented and insightful of constitutional lawyers. Perhaps the best example of this is President Abraham Lincoln, who many believe was a more sophisticated constitutional lawyer and thinker than anyone on the Supreme Court during his day.

Moreover, the comparison between the aptitude of courts and the aptitude of executive officials that the California court suggests we undertake is a false one. The real comparison should be between the aptitude of the legislature that passed the statute in question and the executive official who does not wish to enforce it because he believes it to be unconstitutional. Legislators are no more likely than executive officials to be learned lawyers. Yet the California court’s broad suggestion is that all executive officials must obey statutes that legislators think are constitutional, even though no court has yet weighed in to support either the legislature or the executive branch positions.

For this reason, the “presumption of constitutionality” 67 Id. at 493 (quoting Davies Warehouse Co. v. Bowles, 321 U.S. 144, 153 (1944)). that attaches to each legislative statute—and which does much work in the court’s analysis 68 See, e.g., id. at 485–86. —is itself undermined by the court’s functional analysis of the legal sophistication of the relevant actors. There is no reason to believe that a legislator’s instincts about what is constitutional are likely to be any better than an executive official’s.

The California Supreme Court also ruled more broadly than necessary on another question: Do—or did—the licenses already issued by San Francisco to same-sex couples have any legal force and effect? The majority bluntly responded no, holding that because the licenses were issued unlawfully, they never had, and never will have, any effect at all. 69 Id. at 498.

To some extent, the majority’s instinct was right—until and unless California statutes defining marriage as between only a man and a woman have been judicially invalidated, the San Francisco same-sex licenses cannot have any legal force, and the persons who hold them cannot enjoy any of the legal benefits distinctive to the institution of marriage. Thus, couples possessing these licenses should know—and the California Supreme Court was right to remind them—that these licenses should not be relied upon in the absence of a court opinion invalidating the marriage statute.

But the court went a step further, saying that even in the event that California marriage statutes were later invalidated by appellate courts, same-sex couples would have to go through the marriage process (again) to obtain marital benefits: “[S]hould the current California statutes limiting marriage to a man and a woman ultimately be repealed or be held unconstitutional, the affected couples then would be free to obtain lawfully authorized marriage licenses.” 70 Id.

Justice Werdegar’s dissent invokes a persuasive analogy to demonstrate that the court’s conclusion here is far from obvious. 71 Id. at 508 (Werdegar, J., dissenting). She observes that “interracial marriages that were void under antimiscegeny statutes at the time they were solemnized were nonetheless recognized as valid after the high court rejected those laws in Loving v. Virginia.” 72 Id. At the time the interracial marriages were voided, no court had held the statutes banning interracial marriage to be unconstitutional. Yet, as Justice Werdegar suggests, after the Supreme Court so held, interracial marriages that had already been solemnized were recognized automatically. 73 Id. Why should same-sex marriages be treated differently? The majority gives no adequate answer.

Lesson #3: The Potential Drawbacks of a Divided Executive Depend Very Much on the Particular Offices Involved in Any Clash

Let us return to the Schwarzenegger–Chiang battles over the state budget. For many reasons, the general instinct among lawyers and the population is that Chiang’s various gambits in court should, did, and will continue to lose. First, it is odd that the controller would believe he is legally authorized to take positions in court contrary to the Governor’s positions. In a seminal ruling in 1981, 74People ex rel. Deukmejian v. Brown, 624 P.2d 1206 (Cal. 1981). the California Supreme Court held that even the attorney general could not take a position adverse to the Governor in court because under the state constitution, the Governor retains the “Supreme Executive power” to determine the public interest, and under state statutes he is to “supervise the official conduct of all executive and ministerial officers.” 75 Id. at 1209. While the controller is often designated as the state’s chief financial officer, and while he is assigned by the constitution to sit on some meaningful boards, 76 See Larry Berman, Bruce Allen Murphy & Milton Clarke, Approaching Democracy: California Edition 712–13 (5th ed. 2005). California courts have held that the office of the controller is largely “ministerial.” 77 See, e.g., Tirapelle v. Davis, 26 Cal. Rptr. 2d 666, 676 (Cal. Ct. App. 1993) (“[T]he greater part of the duties devolved upon [the Controller] are of a . . . purely ministerial character.” (quoting People ex rel McCauley v. Brooks, 16 Cal. 11, 55 (1860))). The job of the controller consists of being the state’s chief auditor and accountant, but not its fiscal steward. The Governor plays that role, and, in the case of state employment, fiscal stewardship lies in the hands of the legislatively established Department of Personnel Administration, which the Governor oversees. Given that even the attorney general—an officer whose job is to litigate—is not permitted to formally sue the Governor or take legal positions adverse to the Governor’s positions, the controller’s assertion that he may disregard the Governor’s directives and challenge them in court seems quite far-fetched.

It is true that some of the controller’s arguments—arguments asserted on behalf of himself and a variety of other independently elected statewide officers such as the treasurer and Lieutenant Governor—were ostensibly asserted to protect the independence of the office of the controller and the employees housed within it rather than simply based on disagreements with the Governor about general policy. 78 See, e.g., Schwarzenegger v. Chiang, No. 34-2009-80000158-CU-WM-GDS, at 2–3 (Cal. Super. Ct. County of Sacramento Mar. 12, 2009) (minute order). But even if the controller had standing to raise those arguments about the independence of his office—in the way that the President sometimes declines to defend Congress’s laws that impinge upon the Executive Branch 79 Cf. INS v. Chadha, 462 U.S. 919 (1983). —that rationale would not justify the bulk of the controller’s legal arguments. These arguments focused on his disagreement with the Governor over which resolution to the budget crisis was in the state’s best interest, disagreements he had with the Governor over the meaning of federal law—questions on which even the attorney general is supposed to defer to the Governor under the ruling in People ex rel. Deukmejian v. Brown 80People ex rel. Deukmejian v. Brown, 624 P.2d 1206, 1209 (Cal. 1981) (“The constitutional pattern is crystal clear: if a conflict between the Governor and the Attorney General develops over the faithful execution of the laws of this state, the Governor retains the ‘supreme executive power’ to determine the public interest; the Attorney General may act only ‘subject to the powers’ of the Governor.” (quoting Cal. Const. art. V, § 13)). —and questions about the feasibility of various payment methods, 81Gilb v. Chiang, No. 34-2008-80000026, at 10–13 (Cal. App. Dep’t 29 Super. Ct. Feb. 27, 2009) (ruling after hearing). See also, e.g., Yamamura, supra note 41. which again are matters of pure policy and executive implementation squarely within gubernatorial competence.

Related to and perhaps as important as this “structural” 82I borrow this term from Bill Marshall, who uses it to describe an approach that “focuses on the respective roles of . . . officers in the divided executive and questions which role deserves particular deference in a specific context.” See William P. Marshall, Break Up the Presidency? Governors, State Attorneys General, and Lessons from the Divided Executive, 115 Yale L.J. 2446, 2464 (2006). critique of the need for controller independence is the political reality that, as compared to the Governor, the controller hardly even counts as a prominent statewide elected official. This is evidenced by the fact that voter interest in the unknown candidates’ statewide race for controller is virtually nil, 83 See, e.g., Mark DiCamillo & Mervin Field, The Field Inst., The Field Poll: Awareness, Image Ratings and Preference Measurements in Seven Down Ballot Statewide Election Contests 5 (2002) (showing that considerably more than half of voters surveyed had “no opinion” about either major party candidate for the controller office); Allison Hoffman, Candidates for Controller Struggling for Attention, Torrance Daily Breeze, Oct. 29, 2006, at A6; Matthew Yi, CAMPAIGN 2006: State Controller; Unknowns Clash in Battle to Claim Powerful State Financial Position, S.F. Chron., Oct. 14, 2006, at B2 (observing that more than eighty percent of voters had no opinion of either major party candidate for controller in the weeks leading up to election). and the amount of money spent on the controller race pales in comparison with the amount spent on the gubernatorial race. 84 See Yi, supra note 83 (documenting that both major party candidates had, only two months before the election, raised only around $1 million). A campaign for the Governor’s office in California can easily require more than $50 million. Moreover, the fiscal and budgetary management policies of gubernatorial candidates are prominent during the election, whereas the controller’s platforms tend to be more technical, and citizen awareness of the identity of any statewide executive office other than the Governor and the attorney general seems to be exceedingly low. While a substantial majority of people know who the Governor is, and over half the voters typically have an opinion regarding attorney general candidates in election polls, generally no other statewide official is recognized by more than twenty percent of the populace. 85 See, e.g., Mark Baldassare et al., Pub. Policy Inst. of Cal, Californians and Their Government 19 (1999). The controller’s office is so under-the-radar that it has not even been included in the most prominent surveys of voter knowledge conducted over the years.

In short, the idea that in a time of true fiscal calamity (and the last year has certainly been calamitous) 86 See supra notes 33–36 and accompanying text. a clash of policy between the Governor and a controller in California should warrant judicial resolution, which would result in delaying certainty about the course of fiscal action to be pursued, is laughable. And yet this is an entailment of California’s distinctive scheme of divided executive powers (one of the many areas of fiscal management that cries out for reform in the state). There seems to be very little modern justification for a separately elected state controller, let alone one who can claim any legitimate authority to challenge the Governor’s sense of what is in the best fiscal interests of the public. I am not suggesting, of course, that the Governor should be treated as an emperor in times of financial emergency; he must obey existing laws, contracts, and federal limitations. What I am saying is that those laws, contracts, and federal limitations can be enforced by courts in lawsuits brought by aggrieved employees and citizens. The office of the controller need not confuse and complicate things more by asserting its own (often poorly reasoned) views in court.

Contrast the Governor–controller conflict with the tension that sometimes exists between the attorney general (the state’s chief legal officer) and the Governor. As the thoughtful constitutional scholar William Marshall suggests in one of the few significant essays looking at divided state executive branches and the lessons they may offer for the federal model, 87 See Marshall, supra note 82. many states have a strong tradition of attorney general independence. Even in California, where the state supreme court held that the attorney general may not sue or “take a position adverse to” 88People ex rel. Deukmejian v. Brown, 624 P.2d 1206, 1209 (Cal. 1981). the Governor, the California Supreme Court has read state statutes broadly (perhaps overly so) to recognize the attorney general’s power to decline to assert a position desired by the Governor (or other executive branch clients) and bow out of the case entirely, leaving it to the Governor’s office or other executive agency to assert its legal views on its own. 89 Id. at 1209. The court read California Government Code Section 11040 as recognizing the attorney general’s prerogative to withdraw from representation of an executive agency, even though the statute’s terms do not explicitly confer that power. Id. Thus, while the attorney general may not contradict the Governor, he need not be the Governor’s mouthpiece if he feels required by his legal conscience, and his desire to keep his legal hands clean, to disagree with the chief executive.

In California, the unfolding gay marriage episode showcased this independence of the attorney general. 90Ironically, the attorney general is now Jerry Brown, who was the Governor in 1981 when he was challenged by then-Attorney General George Duekmejian in the dispute that led to the 1981 Brown ruling in favor of the Governor. Id. After the California Supreme Court repudiated Mayor Newsom’s actions, 91Lockyer v. City and County of S.F., 95 P.3d 459, 459 (Cal. 2004). it then heard the state constitutional equal protection–due process challenge to California’s statutes limiting marriage to opposite-sex couples on the merits. 92 In re Marriage Cases, 183 P.3d 384 (Cal. 2008). In these California same-sex marriage cases, 93 Id. Attorney General Jerry Brown filed a brief that differed from the Governor’s position on the merits, asserting different nuanced views of state equal protection. 94Answer Brief in Response to Petition for Extraordinary Relief, In re Marriage Cases, 183 P.3d 384 (Cal. 2008) (No. S147999). Technically, this independent view might have run afoul of the Brown decision—the attorney general’s position was different from and arguably adverse to the Governor’s stance. 95Although the differences in the positions asserted are minor, the Governor’s lawyer (who argued separately from the Attorney General’s Office at oral argument) did advert to differences in legal views. Transcript of Oral Argument at 54–56, In re Marriage Cases, 183 P.3d 384 (Cal. 2008) (No. S147199). The “clean hands” rationale allowing an attorney general to decline to represent a client he thinks is acting illegally would not seem to justify the attorney general’s decision to file separately.

It is possible, of course, that the Governor authorized or encouraged the attorney general to make a distinct filing. Perhaps the Governor thought the court was best served by having multiple approaches to equal protection briefed before it. But there does not seem to be a public record of such approval, creating at least the possibility of attorney general independence that is even stronger than that recognized as legitimate by the California courts.

Professor Marshall argues that the tradition of attorney general independence from governors is beneficial and is explained in large part by a structural analysis of the job of a top lawyer. 96Marshall, supra note 82, at 2464–67. Governors are not typically elected for their legal—as opposed to policy, managerial, or fiscal—prowess, so recognizing latitude by state attorneys general makes good democratic sense. Such a structural account seems to explain the instinct in California that the attorney general’s independence is more understandable than that of the controller, although both are elected officials. 97It does bear noting that at election time the attorney general is a much more salient and visible office and one about which voters have many more views and opinions. And history shows that formal, legal independence does not confer real world credibility and authority as much as political or structural running room does. Prosecutors appointed by, and subject to removal by, the President often have more real world credibility and clout than so-called “independent” prosecutors, who by virtue of their selection process may be legally protected from termination but are vulnerable to the charge of political gamesmanship.

Even if Professor Marshall is correct that state attorneys general do and should enjoy independence from their governors, the recent California experience again counsels caution in some respects. In the most recent phase of the California gay marriage saga, after the voters of the state reinstated a ban on same-sex marriage, 98For a discussion of the federal challenge, see Dan Levine, Prop 8 Backers Don’t Want to Give Up E-mail, Recorder (S.F., Cal.), Sept. 28, 2009. a federal lawsuit was filed challenging the state ban on federal grounds. So far, the attorney general has declined to defend the state law and again has gone beyond a “clean hands” approach to formally attack the law 99 See Answer of Attorney General Edmund G. Brown Jr. at 3, 8–10, Perry v. Schwarzenegger, No. 09-CV-02292 VRW (N.D. Cal. June 12, 2009) (asserting that Proposition 8 violates the federal Constitution). (as he had done in the most recent battles over state law grounds). The Governor, who has made a number of separate filings, has declined to weigh in on the merits of the case, saying that he lacks a basis for responding to the plaintiff’s legal arguments. 100 See Administration’s Answer to Complaint for Declaratory, Injunctive, or Other Relief, Perry v. Schwarzenegger, No. 09-CV-02292 VRW (N.D. Cal. June 16, 2009). While politically understandable, the Governor’s assertion that he is not in a position to weigh in makes little legal sense. Does he agree with the attorney general’s refusal to defend (and decision to attack) the state law? If so, why doesn’t he make that clear? If not, what is his position? As the district court judge handling the case has observed, it is odd—and frustrating—that the Governor refuses to state his views on this most important of modern legal issues. 101Lisa Leff, Judge Sets January Trial Date for Prop. 8 Case, Associated Press, Aug. 19, 2009 (noting Judge Vaughn Walker’s “pointed advice for the lawyer who was in court representing Gov. Arnold Schwarzenegger,” and that the judge “was surprised to find Schwarzenegger standing on the sidelines ‘on an issue of this magnitude and importance,’” and that the judge observed that “[t]he Governor’s thoughts and views would be very much welcome and appreciated”). Even if a divided executive were to accomplish some laudable goal of checks and balances, surely clarity about how divided the executive is—and how much we can infer about the Governor’s views from what the attorney general files—would promote the undeniably important accountability 102 Cf. New York v. United States, 505 U.S. 144 (1992); see also Perdue v. Baker, 586 S.E.2d 606, 609 (Ga. 2003) (suggesting Georgia may favor a “concurrent power” approach to legal representation). the unitary executive theory tries to vindicate.

Lesson (or Question) #4: Can the Federal Government Learn from the California Attorney General’s Disinclination to Defend the Ban on Same-Sex Marriages?

I close this Essay not with a lesson, but with a question. Might scholars of the federal model learn from the example set by the California attorney general in the gay marriage context, where he failed to defend a state law against a constitutional challenge? As many (including Marty Lederman, whom I quote here) have pointed out:

As a general matter, the [federal Justice] Department has traditionally adhered to a policy of defending the constitutionality of federal enactments whenever “reasonable” arguments can be made in support of such statutes—i.e., whenever the constitutionality of the law is not fairly precluded by clear constitutional language or governing Supreme Court case law. This practice has been predicated on the notion that because the political branches—the Congress that voted for the law and the President who signed it—have already concluded that the statute was constitutional, it would be inappropriate for DOJ lawyers to take it upon themselves to reject the constitutional judgment shared by the President and the legislature. 103Marty Lederman, John Roberts and the SG’s Refusal to Defend Federal Statutes in Metro Broadcasting v. FCC, Balkinization (Sept. 8, 2005, 00:11), http://balkin.blogspot.com/2005/09/john-roberts-and-sgs-refusal-to-defend.html.

Although there are exceptions to this rule (involving, as noted above, situations where a law allegedly impinges on the Executive Branch or where the President has already publicly committed to not enforcing a law), there is a strong federal tradition of the Executive Branch defending the constitutionality of statutes. Although he has not made clear his reasoning, California’s Attorney General Brown must not embrace this approach. Like the federal Defense of Marriage Act (DOMA), which the Obama Administration is defending in court, the California ban on same-sex marriages can be legally defended by non-frivolous arguments. Indeed, they are, at least under current case law, likely to be winning arguments.

Some might urge Attorney General Brown to more thoroughly explain his disinclination to defend the statute. But I would urge the federal Department of Justice (DOJ) to explain more thoroughly its decision to defend in virtually all cases. What exactly makes it more important to safeguard against separation of powers violations than, say, equal protection violations? And why is the President’s stance so different in litigation than in legislation? After all, the President need not give much interpretive deference to Congress when deciding whether to veto a law. Why should the Administration be more solicitous of Congress’s handiwork in court? Indeed, as noted above, presidents sometimes decline to enforce federal statutes, a position that would seem more extreme than a refusal to legally defend a statute. But should DOJ necessarily be more reluctant to refuse to defend than to refuse to enforce a statute? It is not clear.

There are a number of possible distinctions between Attorney General Brown’s decision to attack Proposition 8 and the Obama Administration’s decision to defend DOMA. First, Proposition 8 is a voter initiative, not a statute. But since Attorney General Brown’s objection to Proposition 8 is substantive, and does not relate to whether the Proposition went through the requisite procedural hoops to become a part of the state constitution, that distinction may not be relevant. Second, Attorney General Brown is elected, and the federal Attorney General Eric Holder is not. But President Obama is elected, and he could instruct his Attorney General not to defend the DOMA if he chose. Also, as noted earlier, even if a state attorney general were not elected, his independence might not be drastically reduced. Third, state standing rules permit more parties and intervenors, such that a state attorney general’s attack, or refusal to defend, might not impair the ability of the court to hear the strongest defense of the law in state court as much as in federal courts, which are more stringent about standing and other justiciability doctrines. Perhaps this is true, but these federal doctrines might not be as rigid as they sometimes appear. Further, Congress is able to assert its own defense of federal statutes if it chooses. 104 See Note, Executive Discretion and the Congressional Defense of Statutes, 92 Yale L.J. 970 (1983).

Lastly, I should note one significant similarity between the Proposition 8 and DOMA litigations—they both involve core equal protection challenges to statutes. Attorney General Brown points out in his federal court Proposition 8 brief that his attack on, rather than defense of, the measure finds precedent in another California equal protection case that was filed in the federal courts—the discriminatory housing case of Reitman v. Mulkey 105387 U.S. 369 (1967). in which state officials attacked the state law under challenge. Moving to the federal, as opposed to state, DOJ, one of the early cases in which the federal DOJ declined to defend a federal law also involved a core “separate but equal” Fourteenth Amendment issue. 106 See Simkins v. Moses H. Cone Mem’l Hosp., 211 F. Supp. 628 (M.D.N.C. 1962), rev’d 323 F.2d 959 (4th Cir. 1963), cert. denied, 376 U.S. 938 (1964). Perhaps if additions are to be made to the list of situations in which the federal DOJ should decline to defend federal statutes, they should build on an equal protection base, like the one used in California.

Footnotes

Professor of Law and Associate Dean for Academic Affairs, University of California at Davis. Some of the themes raised and arguments made in this Essay were first presented, in tentative and undeveloped form, in various on-line columns the author wrote for the website findlaw.com. These on-line columns are viewable on the author’s Findlaw archive page, which is available at http://writ.news.findlaw.com/amar.

1However fairly, the “laboratory” metaphor for policy experimentation is often attributed to Justice Louis Brandeis, who wrote that “[i]t is one of the happy incidents of the federal system that a single courageous State may, if its citizens choose, serve as a laboratory; and try novel social and economic experiments without risk to the rest of the country.” New State Ice Co. v. Leibmann, 285 U.S. 262, 311 (1932) (Brandeis, J., dissenting).

2Gilb v. Chiang, No. 34-2008-80000026, at 8 (Cal. App. Dep’t 29 Super. Ct. Mar. 18, 2009).

3 See, e.g., Cal. Fam. Code § 308.5 (West 2009) (providing that “[o]nly marriage between a man and a woman is valid or recognized in California”), invalidated by In re Marriage Cases, 183 P.3d 384 (Cal. 2008).

4Lockyer v. City and County of S.F., 95 P.3d 459, 464 (Cal. 2004).

5 See id. at 464–65, 465 n.4.

6 Id. at 465.

7 E.g., id. at 465–66.

8 E.g., id. at 466.

9 See, e.g., Thomasson v. Newsom, No.CGC-04-428794, at 1 (Cal. Super. Ct. County of S.F. Feb. 25, 2004) (order denying temporary restraining order).

10 Lockyer v. City and County of S.F., 95 P.3d 459, 466 (Cal. 2004). The attorney general noted that the court could choose to grant the cease-and-desist order and invalidate the marriages already performed, but then allow lower courts to decide the state constitutional equal protection and due process issues even before they were addressed by the state supreme court. But the attorney general indicated that it was his belief that the better course was for the court to examine the merits right away, without waiting for the lower courts to rule, insofar as the due process and equal protection questions were adequately framed and presented. Id.

11 See supra notes 3–5 and accompanying text (discussing Mayor Newsom’s decision to issue thousands of marriage licenses to same-sex couples).

12 Cal. Const. art. III, § 3.5.

13 See id.

14 See Lockyer, 95 P.3d at 466.

15 Id. at 465.

16 Cal. Const. art XI, § 1.

17 Joseph R. Grodin, Calvin R. Massey & Richard B. Cunningham, The California State Constitution: A Reference Guide 182 (11th ed. 1993).

18Whether the marriage licenses are valid in other states depends on the states’ interpretations of the federal Constitution’s Full Faith and Credit Clause. See U.S. Const. art. IV, § 1.

19 See supra note 12 and accompanying text.

20 See infra notes 51–68 and accompanying text.

21 See Cal. Const. art. XI, § 5.

22 Id. See also Grodin, Massey & Cunningham, supra note 17, at 187–90.

23 See Cal. Const. art. XI, § 5.

24 See Lockyer v. City and County of S.F., 95 P.3d 459, 467 (Cal. 2004).

25 See U.S. Const. art. VI (“This Constitution . . . shall be the supreme Law of the Land. . . any Thing in the Constitution or Laws of any State to the Contrary notwithstanding.”).

26 See Lockyer, 95 P.3d at 467.

27 See Cal. Const. art. III, § 3.5.

28 Id.

29 Lockyer, 95 P.3d at 467.

30 Id.

31 See, e.g., Bob Egelko, Court Halts Gay Vows; SURPRISE RULING: Newsom Says the City Will Make a Strong Case. State Justices Take S.F. Case on Narrow Constitutional Issue, Promise Quick Decision, S.F. Chron., Mar. 12, 2004, at A1.

32 See Printz v. United States, 521 U.S. 898, 922 (1997) (noting the importance of unity in the Executive Branch and invalidating Congress’s attempt, through the Brady Act, to transfer federal law enforcement “responsibility to thousands of CLEOs in the 50 States, who are left to implement the program without meaningful Presidential control”).

33Jim Christie, UPDATE-2 California Delaying Tax Refunds Amid Cash Crisis, Reuters, Jan. 16, 2009.

34 See, e.g., State of Cal. Dep’t of Fin., 2008–09 Final Budget Summary, Sept. 23, 2008, available at http://www.documents.dgs.ca.gov/osp/GovernorsBudget/pdf/fbudsum_0809.pdf.

35 See, e.g., Michael Rothfield, The State; Gov. Lays Off 10,000 Workers; Part-Timers Are Fired. Most Others Would Get Minimum Wage. The Controller Vows to Block the Pay Cut., L.A. Times, Aug. 1, 2008, at A1.

36 Id.

37 See Gilb v. Chiang, No. 34-2008-80000026 (Cal. App. Dep’t 29 Super. Ct. Apr. 13, 2009) (final judgment); Prof’l Eng’rs in Cal. Gov’t v. Schwarzenegger, No. 34-2008-800000126-CU-WM-GDS (Cal. Super. Ct. County of Sacramento Feb. 11, 2009) (judgment regarding verified petition); Schwarzenegger v. Chiang, No. 34-2009-900000158-CU-WM-GDS (Cal. Super. Ct. County of Sacramento Feb. 10, 2009) (minute order regarding notice of related cases).

38 See, e.g., Gilb, No. 34-2008-80000026, at 13 (Cal. App. Dep’t 29 Super. Ct. Mar. 18, 2009).

39 Id. at 10.

40 See Schwarzenegger, No. 34-2009-900000158, at 2–3 (minute order).

41 See, e.g., Kevin Yamamura, State Controller: I Won’t Carry Out Governor’s Furlough Order, Sacramento Bee, Jan. 22, 2009, at 4A, available at http://www.sacbee.com/capitolandcalifornia/story/1563773.html.

42 See infra notes 74–83 and accompanying text.

43 See Yamamura, supra note 41.

44 See, e.g., Paul Brest, Sanford Levinson, J.M. Balkin & Akhil Reed Amar, Processes of Constitutional Decisionmaking: Cases and Materials 674 (4th ed. 2000) (noting that the independence of the Federal Reserve may raise its own problems and solutions).

45Lockyer v. City and County of S.F., 95 P.3d 459, 459 (Cal. 2004).

46 Id. at 486.

47 See infra notes 69–72 and accompanying text.

48 Lockyer, 95 P.3d at 487.

49 E.g., Egelko, supra note 31.

50 See, e.g., Lockyer, 95 P.3d at 486–87.

51 See, e.g., id. at 485–86, 487 (asserting that because statutes are always presumed constitutional and the statutes at hand were not patently unconstitutional, their directives must be followed by all public officials).

52 Id. at 510 (Werdegar, J., dissenting).

53 Id. at 464 (quoting Kendall v. United States, 37 U.S. 524, 613 (1838)).

545 U.S. 137 (1 Cranch) (1803).

55 Id. at 177.

56 U.S. Const. art. II, § 3.

57 See John C. Miller, Crisis in Freedom: the Alien and Sedition Acts 75–76 (1951) (describing the politically-motivated Federalist congress as passing the Sedition Act to “ensure the perpetuation of their party’s control”).

58 See id. at 129–30 (describing a federal circuit court’s jury charge that resulted in upholding the constitutionality of the Sedition Act).

59 James Magee, Freedom of Expression 22 (2002).

60See supra note 53 and accompanying text.

61See Jonathan D. Varat, William Cohen & Vikram D. Amar, Constitutional Law: Cases and Materials 35 (13th ed. 2009).

62 Id.

63 Id.

64 Brest et al., supra note 44, at 56.

65 Id. (citation omitted).

66Lockyer v. City and County of S.F., 95 P.3d 459, 490 (Cal. 2004) (quoting S. Pac. Transp. Co. v. Pub. Utils. Comm’n, 556 P.2d 289, 321 (Cal. 1976)).

67 Id. at 493 (quoting Davies Warehouse Co. v. Bowles, 321 U.S. 144, 153 (1944)).

68 See, e.g., id. at 485–86.

69 Id. at 498.

70 Id.

71 Id. at 508 (Werdegar, J., dissenting).

72 Id.

73 Id.

74People ex rel. Deukmejian v. Brown, 624 P.2d 1206 (Cal. 1981).

75 Id. at 1209.

76 See Larry Berman, Bruce Allen Murphy & Milton Clarke, Approaching Democracy: California Edition 712–13 (5th ed. 2005).

77 See, e.g., Tirapelle v. Davis, 26 Cal. Rptr. 2d 666, 676 (Cal. Ct. App. 1993) (“[T]he greater part of the duties devolved upon [the Controller] are of a . . . purely ministerial character.” (quoting People ex rel McCauley v. Brooks, 16 Cal. 11, 55 (1860))).

78 See, e.g., Schwarzenegger v. Chiang, No. 34-2009-80000158-CU-WM-GDS, at 2–3 (Cal. Super. Ct. County of Sacramento Mar. 12, 2009) (minute order).

79 Cf. INS v. Chadha, 462 U.S. 919 (1983).

80People ex rel. Deukmejian v. Brown, 624 P.2d 1206, 1209 (Cal. 1981) (“The constitutional pattern is crystal clear: if a conflict between the Governor and the Attorney General develops over the faithful execution of the laws of this state, the Governor retains the ‘supreme executive power’ to determine the public interest; the Attorney General may act only ‘subject to the powers’ of the Governor.” (quoting Cal. Const. art. V, § 13)).

81Gilb v. Chiang, No. 34-2008-80000026, at 10–13 (Cal. App. Dep’t 29 Super. Ct. Feb. 27, 2009) (ruling after hearing). See also, e.g., Yamamura, supra note 41.

82I borrow this term from Bill Marshall, who uses it to describe an approach that “focuses on the respective roles of . . . officers in the divided executive and questions which role deserves particular deference in a specific context.” See William P. Marshall, Break Up the Presidency? Governors, State Attorneys General, and Lessons from the Divided Executive, 115 Yale L.J. 2446, 2464 (2006).

83 See, e.g., Mark DiCamillo & Mervin Field, The Field Inst., The Field Poll: Awareness, Image Ratings and Preference Measurements in Seven Down Ballot Statewide Election Contests 5 (2002) (showing that considerably more than half of voters surveyed had “no opinion” about either major party candidate for the controller office); Allison Hoffman, Candidates for Controller Struggling for Attention, Torrance Daily Breeze, Oct. 29, 2006, at A6; Matthew Yi, CAMPAIGN 2006: State Controller; Unknowns Clash in Battle to Claim Powerful State Financial Position, S.F. Chron., Oct. 14, 2006, at B2 (observing that more than eighty percent of voters had no opinion of either major party candidate for controller in the weeks leading up to election).

84 See Yi, supra note 83 (documenting that both major party candidates had, only two months before the election, raised only around $1 million). A campaign for the Governor’s office in California can easily require more than $50 million.

85 See, e.g., Mark Baldassare et al., Pub. Policy Inst. of Cal, Californians and Their Government 19 (1999).

86 See supra notes 33–36 and accompanying text.

87 See Marshall, supra note 82.

88People ex rel. Deukmejian v. Brown, 624 P.2d 1206, 1209 (Cal. 1981).

89 Id. at 1209. The court read California Government Code Section 11040 as recognizing the attorney general’s prerogative to withdraw from representation of an executive agency, even though the statute’s terms do not explicitly confer that power. Id.

90Ironically, the attorney general is now Jerry Brown, who was the Governor in 1981 when he was challenged by then-Attorney General George Duekmejian in the dispute that led to the 1981 Brown ruling in favor of the Governor. Id.

91Lockyer v. City and County of S.F., 95 P.3d 459, 459 (Cal. 2004).

92 In re Marriage Cases, 183 P.3d 384 (Cal. 2008).

93 Id.

94Answer Brief in Response to Petition for Extraordinary Relief, In re Marriage Cases, 183 P.3d 384 (Cal. 2008) (No. S147999).

95Although the differences in the positions asserted are minor, the Governor’s lawyer (who argued separately from the Attorney General’s Office at oral argument) did advert to differences in legal views. Transcript of Oral Argument at 54–56, In re Marriage Cases, 183 P.3d 384 (Cal. 2008) (No. S147199).

96Marshall, supra note 82, at 2464–67.

97It does bear noting that at election time the attorney general is a much more salient and visible office and one about which voters have many more views and opinions.

98For a discussion of the federal challenge, see Dan Levine, Prop 8 Backers Don’t Want to Give Up E-mail, Recorder (S.F., Cal.), Sept. 28, 2009.

99 See Answer of Attorney General Edmund G. Brown Jr. at 3, 8–10, Perry v. Schwarzenegger, No. 09-CV-02292 VRW (N.D. Cal. June 12, 2009) (asserting that Proposition 8 violates the federal Constitution).

100 See Administration’s Answer to Complaint for Declaratory, Injunctive, or Other Relief, Perry v. Schwarzenegger, No. 09-CV-02292 VRW (N.D. Cal. June 16, 2009).

101Lisa Leff, Judge Sets January Trial Date for Prop. 8 Case, Associated Press, Aug. 19, 2009 (noting Judge Vaughn Walker’s “pointed advice for the lawyer who was in court representing Gov. Arnold Schwarzenegger,” and that the judge “was surprised to find Schwarzenegger standing on the sidelines ‘on an issue of this magnitude and importance,’” and that the judge observed that “[t]he Governor’s thoughts and views would be very much welcome and appreciated”).

102 Cf. New York v. United States, 505 U.S. 144 (1992); see also Perdue v. Baker, 586 S.E.2d 606, 609 (Ga. 2003) (suggesting Georgia may favor a “concurrent power” approach to legal representation).

103Marty Lederman, John Roberts and the SG’s Refusal to Defend Federal Statutes in Metro Broadcasting v. FCC, Balkinization (Sept. 8, 2005, 00:11), http://balkin.blogspot.com/2005/09/john-roberts-and-sgs-refusal-to-defend.html.

104 See Note, Executive Discretion and the Congressional Defense of Statutes, 92 Yale L.J. 970 (1983).

105387 U.S. 369 (1967).

106 See Simkins v. Moses H. Cone Mem’l Hosp., 211 F. Supp. 628 (M.D.N.C. 1962), rev’d 323 F.2d 959 (4th Cir. 1963), cert. denied, 376 U.S. 938 (1964).