This is the old version of the H2O platform and is now read-only. This means you can view content but cannot create content. You can access the new platform at https://opencasebook.org. Thank you.
William P. Marshall, 115 Yale L.J. 2446 (2006)
The issue of attorney general litigation control remains in dispute. Attorneys general are often “directed” by other branches of state government to take positions in litigation or, in some cases, to withdraw from litigation.
The majority rule is that the attorney general retains control over litigation and that “client” agencies do not have the final word. This means that although a part of the state executive branch, state attorneys general are not required to follow instructions from the governor or other executive agencies.
A 2006 law review article by William Marshall is one of the few that explores the historical and policy reasons for the majority rule that states attorneys general control litigation.EDIT ANNOTATED ITEM INFORMATION DELETE ANNOTATED ITEM
Proponents of the federal unitary executive have contended that its adoption by the Framers “swept . . . plural executive forms into the ash bin of history.” The federal model, however, has not been embraced by the states. The states, rather, employ a divided executive that apportions executive power among different executive officers not subject to gubernatorial control. In forty-eight states, for example, the Attorney General does not serve at the will of the Governor; and in many states, other executive branch officers such as the Secretary of State, Treasurer, and Auditor are also independent. The divided executive holds the theoretical advantages of dispersing power and serving as a check against any particular officer’s overreaching, virtues that might be seen as particularly appealing given concerns about executive branch excesses at the federal level. But the structure also potentially undermines the virtues of energy and efficiency, political accountability, and separation of powers that the Framers of the Federal Constitution associated with the unitary executive model. The question then arises as to whether the divided executive provides a viable and workable model for executive power implementation.
Focusing on the Office of the Attorney General, this Essay examines the divided executive. Part I examines the state experience. It provides a brief discussion of the history and evolution of the Office of the Attorney General, explores how the divided executive works in practice, and canvasses the cases that address how conflicts between governors and state attorneys general are resolved. Part I concludes that the divided executive model can foster an intra- branch system of checks and balances without undercutting the ability of the executive branch to function effectively. Part II then probes the question of whether the federal government should borrow from the state experience and make the Federal Attorney General an independent officer. We live in an era of increasing (and, some would say, increasingly unchecked) presidential power. Part II accordingly considers whether the federal government should construct an intra-branch system of checks and balances, consistent with the state experience, in order to guard against executive branch excess.
I. THE STATE EXPERIENCE WITH THE DIVIDED EXECUTIVE: GOVERNORS AND STATE ATTORNEYS GENERAL10
A. Common Law Origins of the Office of the Attorney General12
The roots of the Office of the Attorney General date back to the thirteenth century, when English kings appointed attorneys to represent regal interests in each major court or geographical area. Initially, the attorneys had limited powers, based either on the courts in which they appeared or the business that they were assigned to conduct. During the Middle Ages, however, this practice was superseded by the appointment of a single attorney with broad authority, including the power to appoint subordinates to carry out his responsibilities. The Attorney General emerged as chief legal adviser to the Crown and was often appointed for life tenure—a practice that continued until the reign of Henry VIII when it was changed to service at the pleasure of the Crown.
Throughout the sixteenth and seventeenth centuries, the duties of the Attorney General continued to evolve and expand; with eminent tenants such as Edward Coke and Francis Bacon, the Office also continued to gain in prestige. The Attorney General was often summoned by writ of attendance to the House of Lords where he was consulted on bills and points of law. In 1673, he began to sit in the House of Commons, advising that body and assisting in the drafting of legislation. He also gave legal advice to the various departments of state and appeared for them in court.
Importantly, during this period, the Attorney General established that his duty of representation extended to the public interest and not just to the ministries of government. In fact, by 1757, the Attorney General was able to refuse “to prosecute or to stop a prosecution on the orders of a department of the government, if he disapproved of this course of action.” Accordingly, the Attorney General became less the government’s lawyer and more an independent public official “responsible for justice.”
B. The State Attorneys General17
The Office of the Attorney General was brought over to the colonies, where it was modeled after its English counterpart; and at the time of the founding, it existed in all thirteen of the original states. The terms of tenure varied considerably. North Carolina, for example, provided for a lifetime appointment by the legislature. In New York, the Attorney General was appointed by the Governor with the advice and consent of an Executive Council but he could be impeached and removed from office for “mal and corrupt conduct” only by a two-thirds vote of those present in the Assembly. Delaware allowed the Governor to appoint the Attorney General, upon confirmation by the Privy Council, for a term of five years. Rhode Island, alone among the original states, provided that the Attorney General would be popularly elected.
The Framers of the Federal Constitution apparently placed the Attorney General under the control of the President, thereby adopting the model of the unitary executive, at least insofar as they did not directly create separate federal officers independent of the President. But the federal model proved to have very little influence over the development of state government. In fact, in the years following the ratification of the Federal Constitution, the states tended to reject the federal model because they were concerned with the concentration of too much power in one executive officer. Ohio, for example, in reaction to a territorial Governor who was perceived to be too autocratic, drafted its first state constitution in 1802 specifically to minimize the authority of the Governor by dispersing executive power over a range of independent executive branch officers.
As the nation matured, many states created independent attorneys general and afforded the Office even greater autonomy by making it a popularly elected position. Again, the states’ purpose was to weaken the power of a central chief executive and further an intra-branch system of checks and balances. Thus, the Minnesota Supreme Court observed, in reference to the state’s 1851 constitution, that:
Rather than conferring all executive authority upon a governor, the drafters of our constitution divided the executive powers of state government among six elected officers. This was a conscious effort onthe part of the drafters, who were well aware of the colonial aversion to royal governors who possessed unified executive powers.
C. Governors and State Attorneys General25
Not surprisingly, a divided executive creates substantial opportunities and incentives for conflict. First, there are matters of simple politics. In states where the Governor and the Attorney General are independently elected, the two officers may come from different political parties with diametrically opposed partisan agendas. If so, they can be expected to be in constant political opposition to each other. Moreover, even when from the same party, the two officers can, and often are, divided by personal rivalries or ideological differences. And even when the two officers agree on a particular issue, they may compete with each other to be the most aggressive in addressing the issue to curry favor with a particular constituency. Add to this the political reality that the Office of the Attorney General has long been seen by many of its occupants as a stepping stone to the Governor’s office and the blueprint for confrontation and conflict is manifest. Finally, disputes may occur because of the differing visions the officers may have concerning each other’s roles. Governors tend to view attorneys general as subservient officers. But most attorneys general, while acknowledging some obligation to represent the Governor and the other parts of state government, tend to perceive their overriding obligation to be to the broader concerns of representing the state, the law, and the public interest.
What is remarkable, then, in reviewing the state experience, is that debilitating conflict has not materialized. This is not to say that serious disputes have never occurred or that governors have never complained about having to deal with independent attorneys general (or vice versa). Certainly they have. And it is also true that the divided executive has occasionally been the target of reforms that would make the Attorney General subject to gubernatorial appointment and removal. But history suggests that both governors and attorneys general have generally learned to cooperate effectively within a divided executive framework.
The reasons why cooperation, rather than conflict, has been the rule are not complex. On one side, the Governor, even if he believes he is unduly constrained by an attorney general’s position, has the general incentive to comply because he may not want to be seen as defying the Attorney General on matters for which the public expects that the Attorney General, as chief legal officer, will have greater expertise. A Governor who rejects the Attorney General’s position therefore risks expending political capital by appearing reckless, if not lawless. Moreover, he risks even greater vulnerability on that point if his legal position eventually fails in court.
On the other side, the Attorney General may also be restrained from overreaching because she is aware that her role is, in large part, defined by public expectations and that her primary obligation is to defend, not contradict, the policies of state officers or agencies, except when those policies violate the law. Indeed, this understanding is so prevalent that virtually all of the state attorneys general have institutionalized it in in-house memoranda.
Many of the more powerful incentives for cooperation, moreover, are mutual. To begin with, as repeat and interdependent players, both sides have the incentive to maintain a functioning relationship to ensure they can fulfill the duties of their respective offices. They may also feel significant political pressure to work together because it will be harmful to both if they are seen as unwilling or unable to work across political divides. The electorate, after all, does not tend to reward those who bring government to a standstill. Further, both sides may be motivated to come together because reaching internal consensus may fortify their actions against third parties. When both the Governor and the Attorney General agree that a course of action is permissible, the authority behind that position is greater than when either party reaches that conclusion alone. Finally, and perhaps unduly idealistically, the Governor and Attorney General may be united by a common sense of duty. As one court has noted, a divided executive requires the executive officers to “combine and cooperate (even if they have differing policy views and perspectives) to provide an efficient and effective executive branch of government.” It may be that state governments traditionally have taken that duty seriously.
D. The Cases Addressing the Relative Powers of Governors and Attorneys General30
Not all disputes between governors and attorneys general regarding their respective powers are resolved internally and some, not surprisingly, proceed to litigation. The relatively few cases addressing intra-executive branch disputes, however, are significant for our purposes in that they provide useful insight into the types of legal conflicts that can be triggered by a divided executive, how courts might approach these conflicts, and, by implication, whether a divided executive is a viable and sustainable structure. These cases can be broken into three categories: (1) cases in which the Attorney General chooses to exercise independent legal judgment and either refuses to represent the Governor (or other executive officers or agencies) or takes an opposed position in litigation; (2) independent actions brought by the Attorney General directly against the Governor or other members of the executive; and (3) cases raising the issue of whether the Attorney General has the right to initiate enforcement actions against private parties without the Governor’s approval or in direct contravention of the Governor’s wishes. This Section first canvasses the cases within each category and then evaluates whether the approaches utilized by the courts are effective in furthering the purposes the divided executive is designed to achieve.33
1. The Power of the Attorney General To Exercise Independent Legal Judgment in Litigation35
The first and most common category of cases addresses the right of the Attorney General to refuse to take the Governor’s (or other executive officer’s or agency’s) position in court. Must the Attorney General represent the position of the Governor on a disputed legal issue, or is she free to substitute her own independent legal judgment as to the best interests of the state? The majority rule favors attorney general independence. Her primary duty, as the state’s chief law officer, is to represent the public interest and not simply “the machinery of government.”
In Secretary of Administration & Finance v. Attorney General, for example, the Massachusetts Supreme Court held that the Attorney General can refuse to appeal an adverse decision despite the contrary wishes of his executive agency client: “[W]hen an agency head recommends a course of action, the Attorney General must consider the ramifications of that action on the interests of the Commonwealth and the public generally, as well as on the official himself and his agency.” An Alabama case, Ex parte Weaver, states this principle even more broadly:
The most far-reaching of the attorney general’s common-law powers is the authority to control litigation involving state and public interests. It is generally accepted that the attorney general is authorized to bring actions on the state’s behalf. As the state’s chief legal officer, the attorney-general has power, both under common law and by statute, to make any disposition of the state’s litigation that he deems for its best interest.
2. The Power of the Attorney General To Sue the Governor or Other Executive Officers45
The second category of cases comprises those in which the Attorney General sues the Governor or other executive officers. For example, an issue occasionally arises regarding the power of the Attorney General to challenge the constitutionality of a state enactment by suing the state executive charged with its enforcement, including the Governor when appropriate. In such cases, the majority rule vests power in the Attorney General to bring the action. Thus, in People ex rel. Salazar v. Davidson, a Democratic Attorney General contended that a redistricting plan signed by the Republican Governor violated the state constitution and sued the Secretary of State to invalidate the plan. The Colorado Supreme Court affirmed the Attorney General’s prerogative, holding that “the Attorney General must consider the broader institutional concerns of the state even though [those] concerns [are] not shared by” other executive officers.
Case law also supports the power of the Attorney General to sue the Governor over matters involving the Governor’s own actions. In State ex rel. Condon v. Hodges, the South Carolina Supreme Court allowed the Attorney General to sue the Governor for attempting to circumvent the provisions of an appropriations bill. Rejecting the argument that a lawyer cannot sue his own client, the court held that the Attorney General has a dual role as the Governor’s attorney and as the executive official charged with vindicating wrongs against the citizens of the state, with the power to seek legal redress for separation-of-powers violations by other state executive officers.
Although there are few cases in which the Attorney General directly sues the Governor, Hodges is not the only example. The Mississippi Supreme Court has allowed the Attorney General to intervene on behalf of plaintiff legislators seeking to declare that a Governor’s partial vetoes of certain bills were unconstitutional. The Kentucky Supreme Court, although holding that the Attorney General had not justified his claim for injunctive relief on the merits, allowed him to bring an action to enjoin the Governor from being sworn in and acting as a member of the state university board of trustees pursuant to the Governor’s own self-appointment. And the Florida Supreme Court allowed the Attorney General to bring a quo warranto action against the Lieutenant Governor seeking his removal because he lacked necessary qualifications.
Nevertheless, the right of the Attorney General to sue executive branch officers or agencies has not been universally approved. In Arizona State Land Department v. McFate, for example, the Arizona Supreme Court held that the Attorney General could not bring suit against a state agency to enjoin its sale of public lands. The court explained that “the Governor alone, and not the Attorney General, is responsible for the supervision of the executive department and is obligated and empowered to protect the interests of the people and the State.” Similarly, in Hill v. Texas Water Quality Board, the Texas Court of Civil Appeals held that the Attorney General lacked the authority to bring suit to set aside an agency rule, finding no independent authority for the Attorney General to represent the public interest against the specific interests of his agency client.
3. The Power of the Attorney General To Initiate Enforcement Actions Against Private Parties50
The final category of cases concerns the power of the Attorney General to proactively initiate civil or criminal actions against private parties. This power, needless to say, may have a profound effect on a state’s policy agenda. For example, a governor who promises to create a pro-business climate could be hampered in achieving this result if the state’s attorney general is aggressive in maintaining consumer protection or antitrust actions against the state’s industries. Similarly, a governor who runs for office as an anti-pornography crusader will be seriously limited in his ability to deliver on this issue if the state’s attorney general refuses to bring pornography prosecutions.
Whether the State Attorney General has the power to initiate criminal or civil actions independent of the Governor is largely a function of statutory authority and, particularly in civil matters, whether the Attorney General is deemed to enjoy common law powers. Thus, in Ohio v. United Transportation, Inc., the court held that, because he had common law authority, the Attorney General of Ohio could bring an antitrust action under state and federal law against local taxicab companies without the approval of either the Governor or the General Assembly. The court stated that “the broad inherent common law powers of the attorney general in . . . contesting infringements of the rights of the general public” had been long recognized. This common law power, moreover, is quite broad. As the court held in Florida ex rel. Shevin v. Exxon Corp., the Attorney General is entrusted, under the common law, with “wide discretion” and a “significant degree of autonomy” in determining what is in the public interest. Indeed, the Attorney General’s common law authority is so unfettered that it may allow her to bring suits in the public interest even when other executive officers or agencies oppose such actions.
In other states, however, the courts have held that the Attorney General’s powers are far more circumscribed. In State ex rel. Haskell v. Huston, for example, the Oklahoma Supreme Court held that the Attorney General must have the Governor’s permission to maintain a civil nuisance action against an oil company because it is within the Governor’s responsibility to see that the laws are “faithfully administered.” Moreover, in a few states, not only is the Attorney General prohibited from initiating actions without the Governor’s approval, but the Governor can also compel the Attorney General to prosecute an action even when the Attorney General does not want to proceed.
4. The Cases in Theoretical Perspective55
Some of the results in the cases reviewed in the previous Subsections can be explained simply as the product of statutory interpretation by the courts. The McFate decision, for example, was based on the relatively broad powers accorded to the Governor under the Arizona Constitution compared to the narrow grant of authority vested in the Attorney General. In other cases, such as Shevin, when the constitutional and statutory principles were less explicit, the courts had to rely on more general principles. But whether derived from constitutional provision, statutory text, or judicial gloss, two general approaches have emerged in deciding how the powers of the Governor and the Attorney General are to be allocated in a divided executive. The first, based on ethics, suggests that the conflicts should be resolved in accord with the principles of the attorney-client relationship. The second, based on the structure of the divided executive, looks to the policies and understandings underlying that model as the basis for resolution. Each will be discussed in turn.58
a. The Argument from Ethics60
The leading case in support of the position that an attorney general is bound by the principles of the attorney-client relationship to represent the interests of his state officer or agency client is People ex rel. Deukmejian v. Brown. As the California Supreme Court stated in that case, there is nothing unique to the duties of the Attorney General that “justif[ies] relaxation of the prevailing rules governing an attorney’s right to assume a position adverse to his clients or former clients.” The approach taken in Deukmejian has an initial, intuitive attraction. After all, if the Attorney General is the lawyer and the Governor the client, the normal expectation would be that the former should advance the latter’s legal positions. In fact, however, the attorney- client relationship approach is easily dismissed.
To begin with, this approach ignores the fact that the Attorney General’s role is significantly more complex than that of a private attorney. Since seventeenth-century England, the Attorney General has generally been deemed to represent the “state” or public interest and not only the machineries of government. Moreover, in the modern era of expansive government, the Attorney General is also often charged with representing a wide range of state officers and agencies, many of whom have positions diametrically opposed to each other. Accordingly, and in recognition of this reality, most courts have held that an attorney general does not violate ethical rules when she engages in the dual representation of competing state entities. It is therefore not a giant step to conclude that dual representation of a state entity and the state or public interest is also not an ethical violation and, indeed, a majority of jurisdictions have so held.
Furthermore, the nature of an independent attorney general belies the conclusion that an attorney general should be ethically bound to represent her officer client. Ethical rules do not provide an attorney with much room to reject the position of her client and, if they in fact limited her authority, there would be little reason for an attorney general to have independent status. Certainly, an attorney general, ethically bound to represent a governor, would not serve as a check on a governor who was intent on exceeding his constitutional or statutory authority. At best, she would be able only to refuse to facilitate the governor’s actions.
Finally, ethical concerns also weigh against binding an attorney general by the attorney-client relationship. As the Colorado Supreme Court noted in People ex rel. Salazar v. Davidson, imposing a rigid obligation on the Attorney General to advance the executive’s positions can undermine the Attorney General’s ethical obligations to uphold the law and constitution when the Governor seeks to defend a measure that the Attorney General believes is unlawful.
b. The Argument from Structure65
The structural approach to disputes between the Governor and the Attorney General focuses on the respective roles of the two officers in the divided executive and questions which role deserves particular deference in a specific context. In certain circumstances, specifically with respect to policy judgments, a structural analysis supports the authority of the Governor (or other executive officer or agency) over that of the Attorney General. Consider Motor Club of Iowa v. Department of Transportation of Iowa, in which a motor club challenged the validity of a state agency rule establishing a sixty-five foot length limitation for trucks. After losing in the trial court, the agency decided against an appeal because a majority of agency commissioners no longer supported the length limit. The Attorney General, however, attempted to pursue the appeal without agency approval. The court held that the Attorney General did not have the authority to proceed without agency authorization.
From a structural perspective the decision makes sense. After all, if the agency no longer supports its own rule, why should the Attorney General, the chief legal officer, be able to substitute her policy judgment for that of the entity empowered to make the policy decisions? Similarly, if the Governor is the officer charged with setting state policy, it makes sense that the Attorney General should defer to the Governor’s (non-legal) policy judgments.
The structural argument, however, favors the Attorney General in matters involving legal, as opposed to policy, judgments. Presumably, a primary reason for having an independent attorney general is to allow for independent legal judgment. Empowering the Governor to be the final authority on legal decisions would make this independence a nullity (as well as, nonsensically enough, vesting in a non-legal officer the power to have the final say on legal meaning).
To be sure, the line between legal judgment and policy decision is sometimes blurred. (Some might even suggest that all law is policy-based.) But even if all legal decisions have some policy overtones, as Motor Club of Iowa suggests, not all policy decisions involve law. The truly difficult cases, in this respect, are those in the third category discussed in this Section, dealing with the Attorney General’s power to institute lawsuits against private parties on behalf of the state. No doubt the decision to bring cases such as the antitrust action in United Transportation96 or the civil nuisance action in Haskell involves the exercise of legal judgment. But it also involves non-legal considerations that can be integral to a state’s overall policy agenda. Accordingly, whether final authority for such decisions should be deemed to be in the province of the Governor, the Attorney General, or both, may depend on the particular context, or, as is often the case with statutory enforcement matters, legislative intent.
The structural argument more consistently favors the Attorney General in the first category of cases previously discussed, those concerning the power of the Office to refuse to take the position of executive branch officers or agencies in ongoing litigation. First, assuming the Attorney General’s actions are based upon legal, rather than policy, judgments, her authority to refuse to take the executive branch client’s position reflects her structural role as the state’s chief legal officer. Second, recognizing her prerogatives in this respect also furthers the policy of having an executive officer whose fealty extends primarily to the rule of law rather than to the litigation needs of any particular administration. Third, allowing the Attorney General to oppose the Governor or other executive branch officer in court reflects another benefit of the divided executive—it promotes a fuller and more thorough examination of intra- executive disputes, both in court and in pre-litigation consultation, than would occur if the Governor were empowered to impose his position unilaterally. Indeed, the values of intra-branch litigation have been implicitly recognized even within the federal executive in cases like United States v. Nixon and Tennessee Valley Authority v. United States EPA, where courts have refused to dismiss intra-branch litigation as non-justiciable on grounds that the requisite adversarial component was missing when the U.S. government was effectively suing itself. Rather, the courts heard both sides of the issues involved, presumably reaching a more considered judgment than might have occurred if the matters had been decided entirely within the executive branch. The results in state cases involving intra-branch disputes, one would suspect, would be similarly informed.
Finally, the structural argument plays its clearest role in supporting the Attorney General’s power in the second category of cases, those in which she sues another part of the executive branch for exceeding its authority. Indeed, if the purpose of the divided executive is to create an intra-branch system of checks and balances,104 there is no better mechanism to achieve this result than dividing executive power between a chief executive and a chief legal officer. After all, who other than the state’s chief legal officer is better poised to make the judgment of whether a state officer has exceeded his legal and constitutional authority? (Moreover, because the Attorney General is further removed than the Governor from the political pressures and demands that face state government,105 she may be able, other things being equal, to approach the issues regarding the bounds of authority more dispassionately.106) The most compelling structural argument supporting the Attorney General’s authority to police the boundaries of executive power, however, rests in the inherent weaknesses of the alternative solution—specifically with the lack of inherent checks that would occur in a system in which the Governor had the final say. For example, in State ex rel. Condon v. Hodges,107 the South Carolina Supreme Court permitted the Attorney General to sue the Governor for circumventing the provisions of an appropriations bill. Had the court allowed the Governor to quash the action, the advantages of the divided executive would have been eviscerated because the Governor would effectively have become the judge of his own authority. There would be neither check nor balance in such a structure.108
E. Lessons from the Divided Executive70
The preceding Sections suggest that the state experience with the divided executive and the independent Attorney General hold a number of lessons. First, as its architects intended, the divided executive model disperses power109 and checks executive branch excess.110 Second, under the divided executive, the Office of the Attorney General is, or can be, appropriately independent of gubernatorial control. Neither ethical constraints nor structural concerns, properly understood, demand that the Attorney General exclusively represent the Governor’s interests. Third, by insulating the Attorney General’s legal authority from gubernatorial control, the divided executive protects against executive branch overreaching by dedicating an executive officer to uphold the rule of law. Additionally, as the example of intra-branch litigation suggests, attorney general independence promotes fuller decision-making before governmental action by assuring consideration of a wider range of concerns than if the Governor acted alone.111 Fourth, the divided executive can be constructed to accommodate a variety of interests. A state, for example, may protect the right of an attorney general to exercise independent legal judgment against the Governor’s position in a particular matter while still requiring the Attorney General to advance the interests of the Governor when her disagreement is based on pure policy112 or upon any other factor deemed to fit best within the final authority of the Governor. In this way, the Governor’s prerogatives can be accommodated as well.
This then leads to a final lesson. The proponents of the federal unitary executive have argued that other structures are destined to fail because they would lead to weakened executives fraught with internal conflict and lack of accountability. The state experience has shown, however, that this has not occurred. After all, the divided executive has been the rule, rather than the exception, in virtually every state for most of the nation’s history, yet there is little to suggest that it has created endemic dysfunction. The primary lesson of the state experience with the divided executive, in short, is that it works. The next Part, accordingly, will ask whether the model may also be appropriate for the federal government.
ii. AN INDEPENDENT FEDERAL ATTORNEY GENERAL?76
A. The Increasingly Powerful (and Unchecked) Presidency78
More than fifty years ago, Justice (and former Attorney General) Robert Jackson observed that the “real powers” of the presidency had expanded far beyond the authority afforded the Office of the Attorney General under the Constitution.113 Since Jackson’s era, as many of the participants in this Symposium attest, presidential power has only continued to increase, particularly in the areas of foreign policy and national security.114 The reasons for this expansion extend beyond the ambitions and personalities of those who have held the Office.115 Rather, the exigencies of decision-making in these areas inevitably vest power in the entity that can react most swiftly; in virtually every case, this entity is the executive. Congress, for example, cannot decide quickly enough after hostilities break out whether those hostilities are a sufficient basis on which to declare war; the courts cannot adjudicate the question of whether the President should have first consulted with Congress before taking military action.116
The contemporary inter-branch imbalance, moreover, is further compounded by the fact that the President has at his command resources unimaginable at the time of the Founding. In addition to the enormous military power that the President is able to unleash without any significant ex ante check, the President has at his disposal agencies such as the CIA and the FBI, which provide the President with ample opportunity to use their enormous capabilities for mischief, including the invasion of individual rights through investigation, surveillance, and detention. At the same time, because their activities are inherently secretive, these agencies are not meaningfully subject to effective oversight by the other branches or by the media.
The result of this is that Congress and the courts seem increasingly unable to check and balance presidential power in particularly critical areas.117 Consequently, we have seen the President’s escalating ability to unilaterally lead the nation into armed conflict, avoid public oversight in the war on terror and other matters, detain and suspend the civil liberties of individuals (including American citizens), and advance an expansive understanding of inherent constitutional powers that flies in the face of congressional and international restrictions.118 Accordingly, if Justice Scalia was correct in writing that the “purpose of the separation and equilibrium of powers in general, and of the unitary Executive in particular, was not merely to assure effective government but to preserve individual freedom,”119 there are serious questions as to whether the existing structure can still effectively promote this goal. Too much presidential power now lies unchecked.
On paper at least, there is a watchdog guarding against executive branch excess. The Federal Attorney General reviews the legality of executive branch action, either in preparation for litigation or in her capacity as legal adviser to the President.120 And consistent with the Office’s common law heritage,121 some of the tenants have claimed that their primary duty is to the law rather than to the administration that they were appointed to serve.122 But under the unitary executive framework, it is the President’s, and not the Attorney General’s, position on the duties and obligations of the Office that controls. And by his power of appointment or otherwise,123 the President can assure that the Attorney General’s primary fealty is to his administration and not to some abstract view of the law. Without any structural assurance of independence, in short, the Office of the Attorney General is only as independent as the President wants it to be.
B. An Independent Federal Attorney General?83
The question, then, is should the Office of the Attorney General become independent? The suggestion is not novel. Congressional hearings124 on the subject were held in the wake of the Watergate scandal, and President Carter was sufficiently intrigued that he asked the Justice Department to opine formally on whether a proposal to make the Office an independent agency would be constitutional. (The Department concluded that it would not.125) The fact that forty-eight states employ such a structure also suggests that the idea is not all that radical, particularly when one remembers that it is not at all clear that the Office was intended to be controlled by the President in the first place. Early versions of the Judiciary Act of 1789, which established the Office, contemplated that the Supreme Court would appoint the Attorney General. Even the enacted provision did not clearly provide for presidential appointment.126 Moreover, the Judiciary Act did not expressly provide that the Attorney General would serve at the President’s will, as it had provided for other executive branch positions.127 The creation of an independent Office of the Attorney General, in short, may not have been all that far outside the Framers’ design.
That said, there are some reasons why the state experience with divided government in the form of an independent attorney general may not easily translate to the federal government. First, most state governments do not conform to a three-branch separation of powers model as rigidly as the federal government, and the inclusion of a separate independent executive officer may upset the balance and design of the federal structure in a more fundamental way than would occur in the states. Second, the need for an independent attorney general to check against executive branch overreaching may be greater at the state level because state legislatures are often part-time and therefore unable to effectively police the actions of the full-time officers of the executive branch.128 (To be sure, there is a strong counterpoint to this argument in that there may be a greater need for an additional check at the federal level because, while the federal government may be available to check against any excesses by the state executives, there is no comparable external authority that can check the federal government.) Third, the powers of the Federal Attorney General are far greater, particularly in her centralized authority over criminal matters, than in any of the State Attorney General offices because, in most states, prosecutorial authority is localized and not under attorney general control. Creating an independent attorney general at the federal level would, accordingly, carve out a far broader swath of executive power than at the state level.
The most important distinction suggesting that the structures of the state and federal governments are not analogous, however, is that the federal government’s role in national security and foreign policy is unlike any responsibility within the province of the states. The President’s need to act with dispatch and expedience in these areas may create a greater need for decision-making to be concentrated in one individual than exists in the states. Moreover, separating the Attorney General’s powers from the President may infringe upon the President’s ability to execute foreign policy and promote national security because questions of legal authority are so critical in this area. The argument thus comes full circle. The President’s national security and foreign affairs duties arguably call for concentrating power in the President, but the dangers of excess in those areas also raise the greatest need for an intra- branch system of checks and balances. Accordingly, in appraising this tension, it may be worthwhile to revisit the classic arguments of energy and efficiency, political accountability, and separation of powers that have been advanced in support of the unitary executive.
1. Energy and Efficiency88
The first classic objection to dividing the executive, stemming from Alexander Hamilton in Federalist No. 70, is that unitariness is needed to foster energy and efficiency.129 Undoubtedly some energy and efficiency concerns would arise if the Attorney General were independent130 because the President would need to consult another executive officer and work out any disagreements prior to taking action. In part, however, this concern may be overstated. Every President already confers with legal advisers when his legal authority to take a specific action is ambiguous. The only difference is that there would now be an independent voice at the table.
Still, as the state experience shows, inefficiencies exist. A governor who does not need to worry about negative legal advice from an independent officer is less likely to be chilled in taking particular actions falling within the gray areas of her authority. Inefficiencies are also created, as Perdue v. Baker131 demonstrated, in litigation when the State is a party. It is anything but efficient when both the Governor and the Attorney General separately represent the State and take opposite positions. And even when the Attorney General is deemed the State’s official representative in litigation, the power of the Governor to intervene separately still fosters inefficiency in the allocation of resources—not to mention presenting a decidedly mixed message to the courts.
But the issue, in any event, is not simple inefficiency or lack of energy. As the Framers’ three-branch design already recognizes, inefficiencies and inhibitions on government actions are not always negatives and can affirmatively foster other important goals, such as dispersing power and maintaining a system of checks and balances. The actual issue, then, in choosing between a unitary and a divided executive is optimal inefficiency: Are the benefits offered by the divided executive worth the inefficiency costs? Certainly a President who must work through an independent attorney general, for example, to initiate an extensive program of warrantless electronic surveillance or detention of American citizens may be stilled in his efforts. But having presidents less energetic in testing the boundaries of their powers would also presumably serve the goal of protecting individual liberty.132
The second classic argument, also from Federalist No. 70, is that a divided executive undermines political accountability.133 As Hamilton argued, a plural executive “tends to conceal faults, and destroy responsibility”134 by either increasing the chances that various officers may blame others for any miscalculations or errors or by colluding in the first instance to deliberately cloud the lines of responsibility and avoid subsequent blame. Additionally, as the experience with the independent counsel may have shown, if the Attorney General is truly independent, there will be few checks on her when she engages in questionable behavior. The possibility for abuse then, as Justice Scalia foresaw in Morrison v. Olsen, is considerable.135
However, it is once again unclear how well these arguments actually contradict those in favor of establishing an independent Office of the Attorney General. To be sure, lines of accountability between the President and the Attorney General could become blurred in certain circumstances. But although there may be some problems with blurred accountability, they will not be as extensive as in the types of plural executives of concern to Hamilton if the scope of the Attorney General’s authority does not extend to all executive decisions and pertains only to matters of legal judgment.136
In fact, an independent attorney general would arguably foster greater accountability than the unitary structure. To begin with, there is often no political accountability in the current unitary executive because accountability requires transparency and, particularly in the areas of national security and foreign affairs, so much executive action is done in secret. The ability (and predilections) of the unitary executive to take action removed from all oversight, in short, undercuts the accountability claim. To the extent that requiring the President to consult with an independent officer leads to greater transparency, the interests of accountability are served. Moreover, a divided executive has the potential to foster greater accountability than the unitary model in another respect. As Peter Shane has argued, the persuasiveness of the accountability argument as support for the unitary executive may be overstated because the electoral process requires the voter to combine a series of political choices into a vote for a single personality who is unlikely to reflect her views on all those issues.137 A voter who is pro-life or anti-tax might vote to re-elect a President who reflects these positions even if she disagrees with the latter’s legal stance on the limits of presidential power. Allowing her to vote separately for the officer charged with formulating legal positions may promote greater realization of her policy choices. To be sure, this argument may prove too much, as it would suggest in its extreme that the executive should be divided into an elective office for every galvanizing political issue.138 But again, if the role of the Attorney General were defined in relatively narrow terms, the overall political accountability of the executive branch could be increased. Finally, if the Attorney General is independently elected, as in most states, the problems associated with an independent counsel would not exist in the first place. Unlike an independent counsel, the Attorney General would not be an officer with only one charge and no accountability to any electorate. Rather, she would have authority over a wider range of legal matters and responsibility to the electorate for deficiencies or errors in judgment.
In any event, the question of whether a divided executive truly undermines political accountability may have been answered by the state experience. There is currently little to suggest that, in the overwhelming number of states where the Attorney General is independent, the division of authority between the Governor and Attorney General has made either politically unaccountable.
3. Separation of Powers98
The third classic concern, raised by Madison in Federalist No. 51, is that a divided executive undermines separation of powers by weakening the executive in its battles with the other two branches of government. Madison theorized that because those in power would inevitably attempt to expand their authority, fortifying each branch was necessary to prevent the encroachments of another.139 To Madison, the legislature had the greatest ability to invade the prerogatives of the others.140 He thus concluded that, in order to assure that the branches were protected “commensurate to the danger of attack,” the legislature needed to be divided into two. The executive, however, was to be unitary not because it was intended to be powerful for its own sake, but because it was needed to constrain the power of the legislature.141 Certainly, dividing the executive could weaken it in its struggles with Congress. But if the bases of Madison’s initial calculations have changed, and the executive, and not the legislature, is now the most dangerous branch, then restructuring the government to reflect the new reality would be consistent with Madison’s vision and design.142 The separation-of-powers argument, in short, defends the unitary executive only if the original calculations of the defenses needed to counter “the danger of attack” are still accurate. If the balance among the branches has shifted in favor of the executive, however, this same argument militates in favor of the divided executive.101
4. Designing the Office of the Attorney General103
There are undoubtedly other objections to creating an independent Office of the Attorney General beyond the concerns discussed in the last Subsection. For example, practically speaking, a President may choose not to consult with an attorney general if the latter is independent.143 Thus, in creating the Office, it is important to establish the President’s duty to consult before taking certain types of actions. Another concern is that even if the presidency is not inordinately weakened in relation to Congress, an independent attorney general might be weakened, suggesting adoption of measures to protect the Office from over-retaliation.144 Finally, whether the position is elected or appointed, steps should be taken to assure that the Office’s ability to function effectively is not undermined by politicization.145
No solution is likely to be free of difficulty, and designing the optimum approach will take some development and empirical study that are beyond the bounds of this Essay. The critical question, however, is not whether the creation of an independent Federal Attorney General would be a perfect solution but whether it would preferable to the current model in which the Attorney General is politically dependent on and subservient to the President. The workability of the state experience with independent attorneys general provides a starting point for assessing the viability and desirability of this option as a method for restraining presidential power. The increasing inability of the current federal system to check presidential excesses provides reason to consider this approach seriously.
The debate over the unitary executive has tended to disregard the state experience, although virtually every state government has a divided executive structure. As the state experience demonstrates, a divided executive presents its share of concerns. Proponents of the unitary executive correctly point out that the structure can impose inefficiency and coordination costs. But the structure offers benefits as well. State attorneys general who are not under the control of governors are freer to offer objective advice and better able to act in accordance with the rule of law rather than in the pursuit of a particular political agenda. An independent attorney general’s ability to do so without imposing substantial burdens on the efficacy of state government makes the model an attractive candidate for adoption at the federal level.The current presidency has the potential of becoming a law unto itself as the expediency and demands of modern government have, in some critical areas, freed the President from the effective oversight of the other two branches. At the same time, the President’s ability to control the Office of the Attorney General makes him effectively the only arbiter of the legality of his actions. An independent attorney general, in the form of the state divided executive, may therefore be an appropriate model from which to reconstruct a workable system of intra-branch checks and balances.110
1 Steven G. Calabresi, Some Normative Arguments for the Unitary Executive, 48 ARK. L. REV. 23, 25 (1995).
2 Patrick C. McGinley, Separation of Powers, State Constitutions & the Attorney General: Who Represents the State?, 99 W. VA. L. REV. 721, 722 (1997).
3 The Attorney General is independently elected in forty-three states and is appointed by the legislature in Maine and the Supreme Court in Tennessee. CHI S. KEON ET AL., THE BOOK OF THE STATES 268 (2005). In New Jersey, New Hampshire, and Hawaii, the Attorney General is appointed by the Governor but is not removable at will. See HAW. CONST. art V, § 6; N.H. CONST. pt. 2, arts. 46, 47, 73; N.J. CONST. art. V, § IV, paras. 3, 5. Only in Alaska and Wyoming does the Attorney General serve entirely at the Governor’s behest. See ALASKA CONST. art. III, § 25; WYO. STAT. ANN. § 9-1-601 (2005).
4 See DANIEL R. GRANT & LLOYD B. OMDAHL, STATE AND LOCAL GOVERNMENT IN AMERICA (5th ed. 1986).
5 This Essay assumes, for purposes of discussion, that making the Office of the Attorney General independent, either by election or appointment, would require a constitutional amendment. See Proposals Regarding an Independent Attorney General, 1 Op. Off. Legal Counsel 75, 77-78 (1977).
6 6 WILLIAM HOLDSWORTH, A HISTORY OF ENGLISH LAW 459 (2d ed. 1937).
8 Id. at 460-61.
10 Rita W. Cooley, Predecessors of the Federal Attorney General: The Attorney General in England and the American Colonies, 2 AM. J. LEGAL HIST. 304, 307 (1958).
11 6 HOLDSWORTH, supra note 6, at 463.
12 Id. at 465.
13 Cooley, supra note 10, at 307.
14 12 WILLIAM HOLDSWORTH, A HISTORY OF ENGLISH LAW 305 (1st ed. 1938).
16 NAT’L ASS’N OF STATE ATTORNEYS GEN., STATE ATTORNEYS GENERAL: POWERS AND RESPONSIBILITIES 6 (Lynne M. Ross ed., 1990) [hereinafter STATE ATTORNEYS GENERAL].
17 DANIEL J. MEADOR, THE PRESIDENT, THE ATTORNEY GENERAL, AND THE DEPARTMENT OF JUSTICE 5 (1980). Notably, the Crown granted colonial attorneys general the same powers and duties as the attorneys general had at home. The effectiveness of the colonial attorneys general, however, was far more limited than their English counterparts owing to their significant lack of resources. STATE ATTORNEYS GENERAL, supra note 16, at 6.
18 See generally Oliver W. Hammonds, The Attorney General in American Colonies, in 2 ANGLO- AMERICAN LEGAL HISTORY SERIES, ser. 1, 3 (Paul M. Hamlin ed., New York Univ. Sch. of Law 1939).
19 N.C. CONST. of 1776, art. XIII.
20 N.Y. CONST. of 1777, arts. XXIII, XXXIII.
21 DEL. CONST. of 1776.
22 This practice dated back to 1650. See R.I. Sec’y of State, Office of the Attorney General, http://www.state.ri.us/govtracker/index.php?page=DetailDeptAgency&eid=3877 (last visited Aug. 5, 2006). The Office of the Attorney General was formally established by constitutional provision in 1842. R.I. CONST. of 1842, art. VIII, § 1.
23 As will be discussed subsequently, it is somewhat ambiguous whether the Office was originally intended to be subject to presidential control. See infra notes 126-127 and accompanying text.
24 The question of whether Congress could create officers or agencies not subject to presidential control has been, of course, the dominant issue in the unitary executive debate. See, e.g., Steven G. Calabresi & Saikrishna B. Prakash, The President’s Power To Execute the Laws, 104 YALE L.J. 541 (1994); Lawrence Lessig & Cass R. Sunstein, The President and the Administration, 94 COLUM. L. REV. 1, 5 (1994).
25 STEVEN H. STEINGLASS & GINO J. SCARSELLI, THE OHIO STATE CONSTITUTION: A REFERENCE GUIDE 163 (2004). Interestingly, the Attorney General was not one of the executive officers established in Ohio’s first constitution and was created first by statute in 1848 and then by constitutional provision in 1851. Id. at 163-64.
26 State ex rel. Mattson v. Kiedrowski, 391 N.W.2d 777, 782 (Minn. 1986).
27 See supra note 3.
28 Scott M. Matheson, Jr., Constitutional Status and Role of the State Attorney General, 6 U. FLA. J.L. & PUB. POL’Y 1, 28 (1993).
29 The authority of attorneys general in specific subject areas is catalogued in STATE ATTORNEYS GENERAL, supra note 16.
30 Id. at 278-79.
31 See, e.g., In re Cardizem CD Antitrust Litig., 218 F.R.D. 508, 521 (E.D. Mich. 2003) (describing variations in the common law powers of attorneys general across states). Not every state, however, invests the Attorney General with such authority. See, e.g., Blumenthal v. Barnes, 804 A.2d 152, 165 (Conn. 2002) (holding that the Connecticut Attorney General does not have common law powers).
32 See, e.g., ARIZ. REV. STAT. ANN. § 41-192(A) (2006); COLO. REV. STAT. § 38-13-102(2.5) (2005); GA. CODE ANN. § 45-15-10 (2006); MISS. CODE ANN. § 7-5-1 (2006); see also STATE ATTORNEYS GENERAL, supra note 16, at 40.
33 See, e.g., ILL. CONST., art. 5, § 8 (“The Governor shall have the supreme executive power, and shall be responsible for the faithful execution of the laws.”); MONT. CONST. art. 6, § 4 (same); PA. CONST. art. 4, § 2 (same).
34 Thad L. Beyle, Governors, in POLITICS IN THE AMERICAN STATES 180, 192 (Virginia Gray et al. eds., 4th ed. 1983) (“These two offices [the Governor and the Attorney General] . . . have the potential for built-in conflict at several levels, from politics to policy to administration.”).
35 See, e.g., Al Baker, Pataki, Environmentalist? Little and Late, Critics Say, N.Y. TIMES, Feb. 18, 2003, at B2.
36 See William N. Thompson, Should We Elect or Appoint State Government Executives? Some New Data Concerning State Attorneys General, 8 MIDWEST REV. PUB. ADMIN. 17, 29-31 (1974).
37 See Matheson, supra note 28, at 12 & n.57 (1993) (citing the articles of two state attorneys general, William A. Saxbe, Functions of the Office of Attorney General of Ohio, 6 CLEV.- MARSHALL L. REV. 331, 334 (1957), and Lacy H. Thornburg, Changes in the State’s Law Firm: The Powers, Duties and Operations of the Office of the Attorney General, 12 CAMPBELL L. REV. 343, 359 (1990)).
38 See, e.g., id. at 28 n.148.
39 James E. Tierney, The State Attorney General: Who Is the Client? (Sept. 1, 1995), http://c-128.port5.com/articles/art2.html.
41 State ex rel. McGraw v. Burton, 569 S.E.2d 99, 109 (W. Va. 2002) (emphasis added).
42 The cases may also have implicit significance in that the very fact that courts have been able to entertain intra-branch disputes reinforces the viability of the divided executive by suggesting that an effective judicial backstop may be available to resolve any potentially debilitating conflicts.
43 Manchin v. Browning, 296 S.E.2d 909, 923 (W. Va. 1982) (Neely, J., dissenting) (urging that the rule in the majority of jurisdictions be adopted by the court).
44 Commonwealth ex rel. Hancock v. Paxton, 516 S.W.2d 865, 867 (Ky. 1974); see also id. at 868 (“[I]n case of a conflict of duties the Attorney General’s primary obligation is to the Commonwealth, the body politic, rather than to its officers, departments, commissions, or agencies.”). The Hancock court noted that at common law the Attorney General represented the king, “he being the embodiment of the state. But under the democratic form of government now prevailing the people are the king . . . .” Id. at 867 (internal citation omitted); see also Sandersen v. Blue Cross & Blue Shield of Ala. (Ex parte Weaver), 570 So. 2d 675, 684 (Ala. 1990) (holding that the Attorney General had the authority to dismiss legal proceedings over the objection of an executive agency).
45 326 N.E.2d 334 (Mass. 1975).
46 Id. at 338. Two years later, in Feeney v. Commonwealth, 366 N.E.2d 1262, 1266-67 (Mass. 1977), the Massachusetts Supreme Court came to the same result when the parties’ intentions were reversed, holding that the Attorney General could prosecute an appeal even when his executive agency client objected. See also State ex rel. Morrison v. Thomas, 297 P.2d 624, 628 (Ariz. 1956) (holding that even though the Attorney General did not retain common law powers, he had the statutory authority to appeal an adverse decision without the approval of the agency client because the Attorney General “may, like the Governor, go to the courts for protection of the rights of the people”).
47 570 So. 2d 675.
48 Id. at 677 (internal citations and quotations omitted). Ex parte Weaver also suggests that the Attorney General should allow the state agency to employ counsel to represent its position if the Attorney General refuses to do so. Id. at 678-79.
49 296 S.E.2d 909, 921 (W. Va. 1982). The Manchin court did acknowledge, however, that its decision did not follow the majority rule. Id. at 921 n.6.
50 Id. at 919-21; see also Chun v. Bd. of Trs., 952 P.2d 1215, 1234 (Haw. 1998) (holding that when the Attorney General’s views differ from those of her agency client, the Attorney General cannot control the litigation “as to advance her view of the ‘public welfare’”).
51 Manchin, 296 S.E.2d at 921.
52 530 P.2d 360 (Ariz. 1975).
53 Id. at 362 (citing Ariz. State Land Dep’t v. McFate, 348 P.2d 912 (Ariz. 1960)).
54 586 S.E.2d 606 (Ga. 2003).
55 Id. at 610.
56 See, e.g., Commonwealth ex rel. Hancock v. Paxton, 516 S.W.2d 865, 867-68 (Ky. 1974) (“[T]he duty of the Attorney General to uphold the Constitution . . . surely embraces the power to protect it from attacks in the form of legislation as well as from attacks by way of lawsuits by other persons against state officers or agencies.”).
57 Cf. State ex rel. Douglas v. Thone, 286 N.W.2d 249 (Neb. 1979) (allowing, without discussion, the Attorney General to bring an action against the Governor to enjoin the implementation of a statute).
58 Hansen v. Barlow, 456 P.2d 177, 177-78 (Utah 1969). But cf. State v. Burning Tree Club, 481 A.2d 785 (Md. 1984) (holding that the Maryland Attorney General does not have common law, statutory, or state constitutional authority to initiate a declaratory judgment action challenging the constitutionality of a state statute).
59 79 P.3d 1221 (Colo. 2003).
60 Id. at 1231.
61 562 S.E.2d 623 (S.C. 2002).
62 Id. at 627-28.
63 Fordice v. Bryan, 651 So. 2d 998 (Miss. 1995). Even more recently, the Mississippi Attorney General sued to block the Governor’s cut-back on Medicaid. See James Dao, In Mississippi, Setting the Pace for a New Generation of Republican Governors, N.Y. TIMES, Feb. 8, 2005, at A18.
64 Commonwealth ex rel. Cowan v. Wilkinson, 828 S.W.2d 610 (Ky. 1992).
65 State ex rel. Attorney-General v. Gleason, 12 Fla. 190 (1868); cf. United States v. Troutman, 814 F.2d 1428, 1438 (10th Cir. 1987) (holding that it was proper for the Attorney General to assist federal officials in the prosecution of an executive officer because “a state attorney general has a primary responsibility to protect the interests of the people of the state and must be free to prosecute violations of those interests by a state officer regardless of his representation of the state officer in past or pending litigation”).
66 348 P.2d 912 (Ariz. 1960).
67 Id. at 918. See also ARIZ. CONST. art. V, §§ 1, 4 (charging the Governor with the faithful execution of the laws and stating that the duties of the Attorney General shall be as prescribed by law).
68 568 S.W.2d 738 (Tex. Civ. App. 1978).
69 506 F. Supp. 1278 (S.D. Ohio 1981).
70 Id.; see also Florida ex rel. Shevin v. Exxon Corp., 526 F.2d 266 (5th Cir. 1976) (affirming the power of the Attorney General to maintain an antitrust suit against various oil companies).
71 United Transp., 506 F. Supp. at 1281-82; see also In re Cardizem CD Antitrust Litig., 218 F.R.D. 508, 520-21 (E.D. Mich. 2003) (“Plaintiff States, by their Attorneys General, had the authority to settle and release indirect purchaser claims in a parens patriae or other representative capacity.”).
72 526 F.2d at 266.
73 Id. at 268-69, 271.
74 See id. at 272; see also State v. Tex. Co., 7 So. 2d 161, 162 (La. 1942) (holding that the Attorney General “is not required to obtain the permission of the Governor or any other executive or administrative officer or board in order to exercise” his right to sue on behalf of the State); State ex rel. Bd. of Transp. v. Fremont, E. & M.V.R. Co., 35 N.W. 118, 120 (Neb. 1887) (holding that the Attorney General could proceed with the prosecution of a case over the objections of the executive agency involved in the suit).
75 97 P. 982 (Okla. 1908).
76 Id. at 985-87 (concluding that the Governor has the sole and exclusive right to exercise executive discretion to determine if a suit should be brought on behalf of the State, and that the Attorney General cannot interfere with the Governor’s discretion); see also State ex rel. Cartwright v. Ga.-Pac. Corp., 663 P.2d 718 (Okla. 1982) (noting that the Attorney General must seek the Governor’s permission to initiate a suit).
77 Ariz. State Land Dep’t v. McFate, 348 P.2d 912, 912 (Ariz. 1960).
78 526 F.2d at 266.
79 624 P.2d 1206 (Cal. 1981). Deukmejian, although the leading case in support of this position, is actually somewhat unusual in that the Attorney General had previously counseled the state agency about how to implement the law at issue.
80 Id. at 1209; see also Tice v. Dep’t of Transp., 312 S.E.2d 241, 246 (N.C. Ct. App. 1984) (same); Manchin v. Browning, 296 S.E.2d 909, 920 (W. Va. 1982) (holding that the Attorney General is bound by rules governing the attorney-client relationship).
81 See Bill Aleshire, Note, The Texas Attorney General: Attorney or General?, 20 REV. LITIG. 187 (2000).
82 For a thoughtful discussion of the ethical issues involved, see Justin G. Davids, State Attorneys General and the Client-Attorney Relationship: Establishing the Power To Sue State Officers, 38 COLUM. J.L. & SOC. PROBS. 365 (2005).
83 See supra notes 14-16 and accompanying text.
84 E.g., Conn. Comm’n on Special Revenue v. Conn. Freedom of Info. Comm’n, 387 A.2d 533 (Conn. 1978); People ex rel. Sklodowski v. State, 642 N.E.2d 1180 (Ill. 1994); Pub. Util. Comm’n v. Cofer, 754 S.W.2d 121 (Tex. 1988).
85 E.g., People ex rel. Salazar v. Davidson, 79 P.3d 1221 (Colo. 2003); EPA v. Pollution Control Bd., 372 N.E.2d 50 (Ill. 1977); Commonwealth ex rel. Hancock v. Paxton, 516 S.W.2d 865 (Ky. Ct. App. 1974); Humphrey ex rel. State v. McLaren, 402 N.W.2d 535 (Minn. 1987); State ex rel. Allain v. Miss. Pub. Serv. Comm’n, 418 So. 2d 779 (Miss. 1982). But see Deukmejian, 624 P.2d at 1206; City of York v. Pa. Pub. Util. Comm’n, 295 A.2d 825 (Pa. 1972).
86 See, e.g., OHIO CODE OF PROF’L RESPONSIBILITY EC 5-1 (2004) (“The professional judgment of a lawyer should be exercised, within the bounds of the law, solely for the benefit of his client and free of compromising influences and loyalties.”); see also MODEL RULES OF PROF’L CONDUCT R. 1.2 (2004).
87 Manchin v. Browning, 296 S.E.2d 909, 923 (W. Va. 1982) (Neely, J., dissenting) (arguing that defining the Attorney General’s role with reference to the attorney-client relationship renders the Attorney General “analogous to a legal aid attorney for State employees sued in their official capacity . . . [who is] bound to advocate zealously the personal opinions of the officer whom he represents”).
88 79 P.3d 1221, 1231 (Colo. 2003).
89 For a discussion of the Attorney General’s obligations to refuse to defend unconstitutional laws, see Dawn E. Johnsen, Presidential Non-Enforcement of Constitutionally Objectionable Statutes, LAW & CONTEMP. PROBS., Winter/Spring 2000, at 7; and Seth P. Waxman, Defending Congress, 79 N.C. L. REV. 1073, 1088 (2001).
90 251 N.W.2d 510 (Iowa 1977).
91 Id. at 512.
92 Id. at 516.
93 Affording the Attorney General the power to exercise independent legal judgment (e.g., to provide the Governor with an interpretation of the meaning of a law) is not necessarily inconsistent with the Governor’s duty to assure that the laws are faithfully executed.
94 See Manchin v. Browning, 296 S.E.2d 909, 924 (W. Va. 1982) (Neely, J., dissenting) (“To take the control of the State’s case away from the ‘chief “law-trained” officer of the State’ and inject the opinions of [an executive] officer who has no legal training is nonsensical.”).
95 Cf. LAWRENCE M. FRIEDMAN, AMERICAN LAW IN THE 20TH CENTURY 589 (2002) (observing that all lawyers and judges are at times legal realists).
96 Ohio v. United Transp., Inc., 506 F. Supp. 1278 (S.D. Ohio 1981); see also supra notes 69-71 and accompanying text.
97 State ex rel. Haskell v. Huston, 97 P. 982 (Okla. 1908); see also supra notes 75-76 and accompanying text.
98 See generally LINCOLN CAPLAN, THE TENTH JUSTICE: THE SOLICITOR GENERAL AND THE RULE OF LAW 277 (1987) (describing the Solicitor General’s Office as independently committed to the rule of law).
99 For this reason, the common rule that the Governor may retain separate counsel when the Attorney General refuses to take his position also makes sense. See, e.g., Ex parte Weaver, 570 So. 2d 675 (Ala. 1990) (allowing the Governor to intervene and take a position in opposition to the Attorney General).
101 278 F.3d 1184 (11th Cir. 2002), opinion withdrawn in part sub nom. Tenn. Valley Auth. v. Whitman, 336 F.3d 1236 (11th Cir. 2003).
103 As Neal Devins reports, the Supreme Court, in furtherance of its interest in fully hearing an issue, has occasionally chided the Solicitor General for not reporting intra-branch disputes. See Neal Devins, Unitariness and Independence: Solicitor General Control over Independent Agency Litigation, 82 CAL. L. REV. 255, 315-16 (1994).
104 Seesupranotes25-27and accompanying text.
105 She may also, because of the traditions of her office, have greater insulation from political pressure because of her perceived role in upholding the rule of law, although one would think that this perception might vary widely among specific personalities.
106 This is not to say that politics will never play a role in an attorney general’s decisions. It is undoubtedly no accident that the legal positions of Attorneys General Salazar and Baker in their respective redistricting and reapportionment cases reflected the positions of their political party. See People ex rel. Salazar v. Davidson, 79 P.3d 1221 (Colo. 2003); Perdue v. Baker, 586 S.E.2d 606 (Ga. 2003).
107 562 S.E.2d 623 (S.C. 2002).
108 See People ex rel. Deukmejian v. Brown, 624 P.2d 1206, 1212 (Cal. 1981) (Richardson, J., dissenting) (noting that allowing the Governor to prohibit the Attorney General from seeking a judicial pronouncement on the legality of legislation that the Governor would implement would cause the “system of checks and balances envisioned by the Constitution [to] fail”).
109 See, e.g., State ex rel. Mattson v. Kiedrowski, 391 N.W.2d 777, 782 (Minn. 1986) (holding that the legislature may not strip a constitutionally established, independent, executive officer of her independent core functions because to do so would “thwart” the Framers’ intent to divide executive powers).
110 See Condon, 562 S.E.2d at 623 (holding that the South Carolina Attorney General can sue the Governor for appropriations violations).
111 See supra notes 101-103 and accompanying text; see also Abner S. Greene, Checks and Balances in an Era of Presidential Lawmaking, 61 U. CHI. L. REV. 123, 134 (1994) (“Diversifying the voices heard in government not only helps to prevent one point of view from becoming too strong, but also promotes the affirmative goal of democratizing governmental decision- making.”). Involving more than one actor in the decision-making process, as the divided executive requires, also can improve transparency which, in turn, can help improve the democratic process by informing the electorate as to the bases of executive branch actions. See Erik Luna, Transparent Policing, 85 IOWA L. REV. 1107 (2000).
112 Iowa appears to be one state that has adopted this approach. Compare Motor Club of Iowa v. Dep’t of Transp., 251 N.W.2d 510 (Iowa 1977) (holding that the Attorney General does not have the power to supersede the policy decision of a state agency in pursuing an appeal), with Fisher v. Iowa Bd. of Optometry Exam’rs, 476 N.W.2d 48 (Iowa 1991) (holding that the Attorney General has the authority to guide state litigation consistent with what he believes are the interests of justice).
113 Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 653-54 (Jackson, J., concurring).
114 See Neal Kumar Katyal, Separating from Within: Internal Checks on Today’s Most Dangerous Branch, 115 YALE L.J. _ (2006) TK (proposing the creation of checks and balances within the executive branch); Harold Hongju Koh, Turning the World Right-Side Up, 115 YALE L.J. _ (2006); Jonathan Macey, Executive Branch Usurpation of Power: Corporate Control and Capital Markets, 115 YALE L.J. _ (2006) (discussing how the federal executive has taken on a disproportionate role in what was previously a central domain of state law regulation). But see Steven G. Calabresi & James Lindgren, Commentary, The President: Lightning Rod or King?, 115 YALE L.J. _ (2006); Todd D. Peterson, The Law and Politics of Shared National Security Power, 59 GEO. WASH. L. REV. 747, 761 (1991) (reviewing HAROLD HONGJU KOH, THE NATIONAL SECURITY CONSTITUTION: SHARING POWER AFTER THE IRAN-CONTRA AFFAIR (1990)) (arguing that Congress has substantial power to control the President’s national security powers). According to Peterson, the problem in this area is not that the President has assumed too much power; it is that Congress has exercised too little. See Peterson, supra, at 767.
115 This is not to say that personality has not played a part. The efforts of Presidents Reagan and Clinton, for example, to give the President greater control over federal agency action have helped to consolidate presidential authority over the administrative state. See Elena Kagan, Presidential Administration, 114 HARV. L. REV. 2245 (2001).
116 The power that comes with being the first to act, moreover, does not substantially abate even after the initial crisis is over. Crisis decisions are not easily undone. When the executive decides to commit the military to armed conflict, the inevitable result is a “rally round the flag” reaction that reinforces the initial decision. Mark Tushnet, Controlling Executive Power in the War on Terrorism, 118 HARV. L. REV. 2673, 2678 (2005); see also Korematsu v. United States, 323 U.S. 214, 244-45 (1944) (Jackson, J., dissenting) (discussing the practical difficulties involved in having courts second-guess military decisions).
117 See, e.g., Martin S. Flaherty, The Most Dangerous Branch, 105 YALE L.J. 1725, 1727 (1996).
118 E.g., Memorandum from Jay S. Bybee, Assistant Attorney Gen., to Alberto R. Gonzales, Counsel to the President (Aug. 1, 2002), in THE TORTURE PAPERS: THE ROAD TO ABU GHRAIB 172, 172-73 (Karen J. Greenberg & Joshua L. Dratel eds., 2005) (arguing that the President’s power to act under his authority as Commander-in-Chief is absolute and cannot be circumscribed by domestic or international prohibitions on torture).
119 Morrison v. Olson, 487 U.S. 654, 727 (1988) (Scalia, J., dissenting); see also Steven G. Calabresi & Kevin H. Rhodes, The Structural Constitution: Unitary Executive, Plural Judiciary, 105 HARV. L. REV. 1153, 1155 (1992) (“The genius of the American Constitution lies in its use of structural devices to preserve individual liberty.”).
120 The key divisions of the Justice Department in this respect are the Office of the Solicitor General, charged with litigating cases before the Supreme Court, and the Office of Legal Counsel, charged with providing legal advice. See Cornelia T.L. Pillard, The Unfulfilled Promise of the Constitution in Executive Hands, 103 MICH. L. REV. 676, 682 (2005) (characterizing these offices as the “principal constitutional interpreters for the executive branch”).
121 See supra notes 14-16 and accompanying text.
122 For example, Attorney General Edward Bates, who served under Lincoln, reportedly stated that it was his duty “to uphold the Law and to resist all encroachments, from whatever quarter, of mere will and power.” LUTHER A. HUSTON ET AL., ROLES OF THE ATTORNEY GENERAL OF THE UNITED STATES 51 (1968). Robert Jackson, on the other hand, apparently viewed his obligations differently. Looking back at his role as Attorney General from the perch of a Supreme Court Justice, he described an opinion he offered as Attorney General as “partisan advocacy.” Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 648 n.17 (1952) (Jackson, J., concurring).
123 See, e.g., Dan Eggen, Staff Opinions Banned in Voting Rights Cases, WASH. POST, Dec. 10, 2005, at A3 (discussing how the President’s political appointees can remove or redeploy staff attorneys if they find them too independent).
124 See Removing Politics from the Administration of Justice: Hearings Before the Subcomm. on Separation of Powers of the S. Comm. on the Judiciary, 93d Cong. (1974).
125 See Proposals Regarding an Independent Attorney General, 1 Op. Off. Legal Counsel 75, 77- 78 (1977).
126 See Susan Low Bloch, The Early Role of the Attorney General in Our Constitutional Scheme: In the Beginning There Was Pragmatism, 1989 DUKE L.J. 561, 567 & n.24. According to Bloch, “The President nevertheless immediately assumed that responsibility, went to the Senate for advice and consent, presumably reading article II, section 2 to support and perhaps require this approach.” Id. at 567 n.24. Moreover, as Bloch notes, unlike the language found in the organic acts establishing the Departments of Foreign Affairs and War, the text of the Judiciary Act did not label the Office of the Attorney General as executive. Id. at 578.
127 Id. The Office is also not, in any event, purely executive. As a functional matter, the position is at least quasi-judicial, both in its role in issuing formal opinions and in its capacity as an officer of the court. See Henry J. Abraham & Robert R. Benedetti, The State Attorney General: A Friend of the Court?, 117 U. PA. L. REV. 795, 797-98 (1969). The Office may also be considered quasi-legislative in the states in which it is also charged with the duty of providing advice to the legislature. See STATE ATTORNEYS GENERAL, supra note 16, at 55-56.
128 See Brief for Georgia Legal Foundation as Amicus Curiae at 4, Perdue v. Baker, 586 S.E.2d 606 (Ga. 2003) (No. S03A1154).
129 THE FEDERALIST NO. 70, at 423-24 (Alexander Hamilton) (Clinton Rossiter ed., 1961). Hamilton’s argument actually involves four separate points. First, a unitary executive is necessary to allow the executive to act with dispatch; second, a plural executive could lead to internal disagreements that would weaken the executive’s ability to carry out its operations; third, a unitary executive prevents divisive internal executive branches from developing as they would if there were numerous executives competing for power; fourth, a unitary executive, by having a national constituency would be more energetic on behalf of the entire national community and not distracted by local geographic pressures.
The last two arguments can be immediately dismissed in the context of the divided executive. A Federal Attorney General would not be subject to the pressures of local faction as would a member of Congress because, like the President, she would be a nationally selected officer (whether elected or appointed) and therefore responsive to the needs of the national constituency. Hamilton’s concern with intra-executive factions, in turn, would not be implicated because the models of a plural executive that he addressed (the Roman model, in which two magistrates shared expansive power, and an executive council model, which required the approval of an independent council before the executive could undertake significant action) involved broadly shared powers. Hamilton did not consider a model in which the chief executive held most of the power while a secondary officer had relatively limited authority. When the vast majority of executive power rests with one person, the incentive for intra-executive cabals to develop to support the officer with significantly less authority seems less likely.
130 There is, of course, an initial question as to whether the Framers’ concern with energy still applies in the modern era. “Modern government at its most lethargic is energetic beyond the Founders’ most reckless speculations.” Flaherty, supra note 117, at 1826.
131 586 S.E.2d 606 (Ga. 2003).
132 Moreover, to the extent that certain law enforcement functions of the Attorney General are truly necessary for the President’s ability to conduct foreign policy and foster national security, the provision granting independence to the Attorney General could be qualified, consistent with the division of executive powers in some states, to grant presidential authority to direct the Attorney General to take particular actions. See, e.g., State ex rel. Jackson v. Coffey, 118 N.W.2d 939 (Wis. 1963). Such an approach could both protect some presidential prerogative while providing a political guard against overreaching.
133 THE FEDERALIST NO. 70, at 427-28 (Alexander Hamilton) (Clinton Rossiter ed., 1961).
134 Id. at 427; see supra note 129.
135 487 U.S. 654, 728 (1988) (Scalia, J., dissenting).
136 Additionally, to the extent that the value of political accountability is less to foster majoritarian results and more to allow the people to protect themselves from government tyranny, a divided executive may complement, rather than undermine, this purpose. See Rebecca L. Brown, Accountability, Liberty, and the Constitution, 98 COLUM. L. REV. 531 (1998).
137 Peter M. Shane, Political Accountability in a System of Checks and Balances: The Case of Presidential Review of Rulemaking, 48 ARK. L. REV. 161, 197-99 (1994).
138 Id. at 199 (arguing that if true representation had been the Framers’ goal, they would have created a multiple presidency).
139 THE FEDERALIST NO. 51, at 321-22 (James Madison) (Clinton Rossiter ed., 1961) (“But the great security against a gradual concentration of the several powers in the same department consists in giving to those who administer each department the necessary constitutional means and personal motives to resist encroachments of the others. The provision for defense must in this, as in all other cases, be made commensurate to the danger of attack. Ambition must be made to counteract ambition.”).
141 See Greene, supra note 111, at 141-48.
142 Flaherty, supra note 117, at 1727.
143 Indeed, in this respect, it is notable that the trend in state government has been that governors have increasingly employed their own counsel. Matheson, supra note 28, at 19; Tierney, supra note 39.
144 The Constitution protects, for example, judicial independence by providing Article III judges with life tenure and guaranteed compensation. See U.S. CONST. art. III, § 1.
145 Such steps might include making the election non-partisan, holding the election in a different year from the presidential election, and making former attorneys general ineligible to run for President or Vice President.
December 04, 2014
Break Up the Presidency?: Governors, State Attorneys General, and Lessons from the Divided Executive
This is the old version of the H2O platform and is now read-only. This means you can view content but cannot create content. If you would like access to the new version of the H2O platform and have not already been contacted by a member of our team, please contact us at firstname.lastname@example.org. Thank you.