Constitutional History, 1980–1989
Constitutional History, 1980–1989
CONSTITUTIONAL HISTORY, 1980–1989
The major constitutional development of the 1980s was the confirmation of divided government as a legitimate alternative to presidential party government as a model for constitutional administration. In the elections of 1980 and 1984 the people chose a Republican President while returning a Democratic majority to the house of representatives. In 1986 the Democratic party regained control of the Senate, which it retained in 1988 as the republican party again won the presidency. This type of split-ticket voting was a relatively new phenomenon in modern American politics. From 1889 to 1953, no President on his inauguration faced a Congress one house of which was controlled by the opposition party. The contemporary period of divided government began in 1969 when the Republican richard m. nixon was elected President and the Democrats controlled Congress. After the resignation of Nixon in 1974, the Democrats briefly revived presidential party government with the election of jimmy carter in 1976. Their inability to govern effectively despite having power over both political branches prepared the way for the Republican capture of the White House in 1980.
Throughout the 1980s, assessments of divided government tended to be uncertain because such government contradicted what had come to be accepted since the new deal as the constitutional norm, namely, presidential government under a dominant party after a critical or realigning election. In fact, there were indications in Nixon's two election victories of a disintegration of the New Deal liberal coalition and an expectation of a political realignment that would enable the presidential party system to continue. The problem with this analysis was that while twentieth-century presidential government was historically liberal, many students of politics believed it was also inherently or in its nature liberal. Many observers were therefore reluctant to conclude that divided government could represent the deliberate choice of the electorate or that it could be a satisfactory approach to running the Constitution, despite the fact that it brought the constitutional principle of the separation of powers more directly to bear on the conduct of government.
Part of the difficulty observers had in recognizing the legitimacy of divided government was attributable to the political popularity of ronald reagan. Despite having twice been elected governor of California, Reagan was an improbable candidate for President. This improbability owed less to his being a former Hollywood actor than to his advocacy of natural rights individualism, limited government, and middle class social values that had come to be identified in the dominant political culture as the essence of right-wing reactionism. Labeled an "ideological" candidate by the national media, he ran on a platform that proposed to reverse the tide of centralized bureaucracy, restore equal opportunity for individuals, stimulate economic growth through deregulation and market incentives, and rebuild national defense. Reagan's assertion of these policies made the election of 1980 the most significant since 1932. In effect, it was a referendum on the regulatory welfare state, broadly conceived in the light of the liberal reforms of the 1960s and 1970s. When President Reagan won reelection in 1984, and Vice-President george bush won election as President in 1988, in a significant sense the era of welfare state liberalism was over, a fact difficult to deny. Divided government was the constitutional expression of this political change.
In one sense, the source or cause of divided government was the Constitution. The fundamental law organizes government into three coordinate branches and, in the words of the federalist #51, guards against a concentration of the several powers in the same department by "giving to those who administer each department the necessary constitutional means and personal motives to resist the encroachments of the others." Moreover, the Constitution permits voters to make a free political choice, including that of ticket splitting (which some critics of divided government have proposed to restrict by a constitutional amendment). Furthermore, the Constitution does not establish political parties in the structure of government, but allows them to exist as voluntary associations regulated by state and federal law. A contributing factor to divided government was the decline of partisan loyalty in the electorate, caused in part by antiparty reform measures, such as campaign finance laws. With the decline of party, political choice was based in part on the political values, principles, and governmental duties and functions associated with the executive and legislative branches or their leading officials. The separation of powers, one of the basic concepts of limited government, was thus the organizing principle of divided government.
A second apparent cause of divided government was uncertainty or ambivalence in the electorate about the basic direction of public policy. In electing Republican chief executives, the people approved of policies aimed at economic expansion, control of inflation, tax reform, limitations on government spending, and strengthened national defense. In voting Democratic control of Congress, the people expressed a desire to maintain the social welfare and regulatory programs that constituted the achievement of modern liberalism. These included entitlement programs and legally conferred benefits for individuals and groups in every social class. Although politically contradictory, these policy alternatives might be seen as reflecting complementary dimensions of the public philosophy underlying American constitutionalism : individualism, based on natural rights principles that limit government power, and the public interest, based on community consensus that requires government regulation.
Nevertheless, in a relative historical sense, by contrast with the period of presidential party government which it succeeded, the public philosophy of divided government represented a partial attempt to restore an older conception of limited government. This was the idea that federal power was limited to specific ends or objects in accordance with the federal principle of divided sovereignty. Reagan administration proposals for a new federalism that would return certain policy matters to the states expressed this outlook. Underlying these proposals was the more basic idea of restoring a sense of discipline and limitation in the conduct of government, seen in attempts to limit government spending, reduce the federal deficit, and revive the concept of a balanced budget.
The political expression of this conservative idea was a state-based movement for a constitutional convention to limit federal spending and achieve a balanced budget. By the mid-1980s this movement had begun to produce political effects in Washington. The Republican-controlled Senate and executive branch supported a balanced-budget constitutional amendment. The Democratic party switched from supporting deficit spending to arguing for deficit reduction as a reaction against Reagan administration defense spending in a year (1985) when the deficit reached $200 billion. With the executive and legislature each blaming the other for the deficits, the situation was ripe for a bipartisan solution. The result was the Balanced Budget and Emergency Deficit Control Act of 1985.
Known as the grammrudman-hollings act, the law required the federal budget deficit to be reduced by stages until it was eliminated in 1991. It contained a triggering mechanism by which automatic across-the-board spending cuts were mandated if deficits exceeded specified levels at certain target dates. Finding it difficult to meet the reduction requirements, Congress put some spending items "off budget" so that they would not be counted in the reckoning of the deficit problem. In 1987 it revised the law to postpone the balanced budget date to 1993. Although the Washington political establishment generally disliked the law, it had the effect of reducing the deficit and slowing the rate of increase in government spending. Despite sharp differences over spending priorities, limitation of spending had become a bipartisan objective or requirement, replacing the presumption of indefinite government expansion based on taxing and spending that marked the 1960s and 1970s. In this sense, divided government signaled the kind of change in the policy agenda associated with a political realignment.
Although compromise could be said to be the logic of government under the separation of powers, as seen in the disposition of such major issues as the budget deficit, tax reform, and control of immigration, ideological conflict was the predominant political effect of divided government. The struggle between the executive and legislature for control of the administrative state, a continuing theme in twentieth-century constitutional history, was exacerbated by the ideological polarization of the 1980s.
Powerful governing instruments were available for carrying on the struggle for policymaking and administrative control. Having forced President Nixon to resign through application of the impeachment power, Congress curbed executive power by passing laws respecting presidential actions in regard to war powers, intelligence activities, and budgetary matters. In 1978, Congress took the major step of creating the office of special prosecutor outside the executive branch to investigate wrongdoing in high-level executive offices. By vigorous use of the legislative veto and the appropriations and oversight powers, Congress in the 1980s challenged the president not only in domestic policy but also in foreign affairs.
Although the executive was weaker than in the era of presidential party government, this branch also experienced a restoration of authority under divided government. Despite legislative measures aimed at limiting the executive, the principal elements of presidential power from the pre-watergate period remained intact. The White House staff of 600 and the Executive Office of the President staff of 5,000 employees were powerful institutions that functioned as a policymaking structure parallel to the regular executive departments. Perhaps the main element in the power of the chief executive was the fact that political responsibility for government continued to fall primarily upon the president. On the whole, despite severe second-term problems in the iran-contra affair, President Reagan was reasonably successful in meeting that responsibility. Possessing aptitude suitable for a plebiscitary type of presidency, he used the media effectively to communicate directly with the electorate and shape public opinion in support of administration policies.
Although President Reagan's political appeal rested in part on his opposition to big government, he responded to the constitutional imperative of the modern administrative state by expanding executive authority to achieve the policy ends of his administration. In 1981 he issued Executive Order 12291, giving the office of management and budget (OMB) authority to require executive agencies to submit cost-benefit analyses. Focused on budgetary impact, the order was intended as a check on the regulatory process, aimed at eliminating waste and inefficiency. In 1985, President Reagan issued a more far-reaching directive, Executive Order 12498, intended to coordinate and establish White House control over bureaucratic policymaking. This order required executive agencies to submit proposed regulations to OMB for substantive approval to ensure that they were consistent with overall White House policy.
The Reagan executive orders resisted an inherent tendency in Congress toward micromanagement of the executive branch, a tendency encouraged by the politics of divided government. Reaction to the directives revealed the ambiguous constitutional status of administrative agencies subject to the control of both the President and Congress. Functionally the executive departments and independent regulatory agencies are in the executive branch, for they are concerned with enforcing and administering laws. The President can order administrative officers to carry out policies within their statutory discretion, and if they fail to follow his direction, he should be able to remove them, on the theory that the Constitution intends law enforcement to be managed by the chief executive. Adminstrative departments and agencies owe their existence, however, to Congress. By its lawmaking power, it creates the units of the administrative state, defining the purpose and powers of each department or agency and the terms and conditions of holding office. From this point of view, the executive departments and regulatory agencies are accountable to Congress and may exercise rule-making discretion only within their statutory mandates.
Claiming discretionary rulemaking authority exercised by previous administrations, Reagan officials frequently acted to withdraw, revoke, or alter agency rules. When they did so, they were sometimes charged, by congressional committees and by private interest groups that opposed the changes, with violating their statutory authority. In the period of divided government, rulemaking under the delegation of legislative power to the executive, usually considered a basic feature of the modern administrative state, was thus subject to attack as lawless executive conduct.
Despite serious deregulatory efforts, the structure of the regulatory welfare state changed very little in the 1980s. Few agencies were eliminated, and given the balance of political forces and congressional defense of the status quo, it was difficult to effect major policy changes. On the government-expansion side, the Veterans Adminstration was elevated to cabinet rank. The main difference between this period and the 1970s was that the regulatory state functioned in a more accommodating and less antagonistic spirit in relation to regulated groups and associations. Corporations opposed the wholesale dismantling of regulatory structures that the "Reagan revolution" at first seemed to threaten, because it would reopen costly political and legal battles. What corporations wanted was greater flexibility in government—business relations and greater reliance on economic analysis in regulatory policy. After initially strong deregulatory efforts met stiff resistance in the subgovernments of the administrative constitution, the Reagan administration moderated its regulatory policy accordingly.
In seeking to preserve the regulatory welfare state, the Democratic congressional establishment evinced tendencies toward legislative supremacy inherent in the doctrine of the separation of powers and the theory of popular sovereignty. As in the past, the appropriations power, regarded as the quintessential legislative power, was the most effective instrument for asserting the congressional will.
The congressional budget and impoundment control act of 1974 made Congress an equal participant in budget planning. The act created the Congressional Budget Office to compete with the OMB, and it solved the problem of presidential impoundment of funds by placing tight restrictions on presidential nonspending (called deferrals and rescissions). At the same time, the act succeeded in one of its implicit purposes, to facilitate congressional spending, which increased significantly in the 1970s and 1980s. Under divided government, the cooperation envisioned in the budget act between the executive and legislature was elusive, and by 1985 it was widely believed that the budget process was not working satisfactorily, for reasons that implicated both branches. The Gramm-Rudman-Hollings Act did not improve the process.
Failing to reach agreement on the series of executive department appropriations bills required by the law, Congress from 1986 to 1988 enacted each year a single measure, the omnibus continuing resolution, to fund the entire federal government. The continuing resolution previously had been a technical expedient used to keep an agency in operation when action on an appropriations bill was not complete by the end of the congressional session. It now was transformed into a comprehensive budget act. Concealed within the continuing resolutions were many substantive policy decisions, unknown to anyone but their subcommittee sponsors, that were enacted into law without public scrutiny and debate. The continuing resolution for 1988, for example, appropriated $605 billion, was 1,057 pages long, and was accompanied by a 1,194-page conference report. President Reagan, with only one day to consider the bill, was virtually forced to sign it if he did not want to shut down the government for lack of funds. In effect, Congress deprived the President of his veto power.
Another constitutional innovation of Congress was the creation of commissions outside the government to decide public policy matters. For example, Congress established the National Economic Commission; the Commission on Executive, Legislative, and Judicial Salaries; and the Commission on Sentencing Guidelines to deal with politically controversial subjects. These were but a few of 596 federal commissions created in 1988 by Congress, 117 of which were determined by the General Accounting Office to be concerned with substantive policy questions. A contemporary manifestation of the progressive belief that government could be purified by separating politics from administration, government by commission added a new wrinkle by proposing to separate politics from legislation. It was an acknowledgment that the legislative process itself, where interests are properly expressed, was so immobilized by faction and ideology that Congress was prepared to abdicate its constitutional responsibility for lawmaking.
If the policy environment was not conducive to legislation, Congress could influence administrative policy-making through the legislative veto. In the era of presidential party government the legislative veto was used mainly to effect executive reorganization plans and strengthen presidential policymaking. In the aftermath of Watergate, Congress used the veto extensively to supervise regulatory policymaking, including it in more than 200 statutes. The Supreme Court declared the legislative veto unconstitutional in the case of immigration and naturalization service v. chadha. Nevertheless, Congress continued to employ devices that were the functional equivalent of the legislative veto. For example, it required agencies to get the approval of appropriations committees before taking an action and used committee reports and notification requirement to supervise bureaucratic policy-making.
The most significant congressional constitutional innovation in the period of divided government was the creation of a major executive office, the independent counsel, outside the executive branch. The independent counsel is a special prosecutor whose duty is to investigate allegations of criminality by high-level officials in the executive branch. After the Watergate affair, bills were introduced into Congress to provide for the prosecution of executive branch wrongdoing by an officer not subject to the political influence or legal control of the President. The premise on which these proposals rested was that an inherent conflict of interest prevented the President and the Justice Department from conducting an unbiased investigation of malfeasance in the executive branch. Publicly the proponents of the special prosecutor defended it as an "auxiliary precaution" under the theory of separation of powers and checks and balances for controlling the government, especially the executive power. The Carter administration supported this legislation, and the office of special prosecutor—renamed independent counsel in 1983—was included in the Ethics in Government Act of 1978.
The independent counsel act assumes that the executive branch is peculiarly prone to illegal activity and conflict of interest. Its provisions apply to the President, vice-president, cabinet officers, senior White House staff, and sundry directors of agencies constituting a class of about seventy officials. The law provides that upon receiving information about a possible criminal violation by a covered official, the attorney general shall conduct an inquiry to decide whether an investigation by an independent counsel is needed. Restrictions on the attorney general make further investigation almost necessary for dealing with allegations of wrongdoing. The judiciary committee of either house is also authorized to request the appointment of a special prosecutor. At the request of the attorney general, a panel of judges from the District of Columbia Circuit Court (called the Special Division) appoints an independent counsel, whose jurisdiction is defined by the Special Division and who can be removed by the attorney general only for cause.
Although the act purports to separate the independent counsel from the legislative branch, the effect of the law is to create a major executive office outside the executive branch. It provides a means by which Congress can do things that for political reasons it might not wish to do through the use of constitutional powers otherwise available to it, such as impeachment. Investigation and indictment through the independent counsel process can be seen as a substitute for impeachment. Despite the appointment of the counsel by the attorney general and the appearance of administrative independence, the independent counsel is in effect an agent of Congress, for congressional committee investigations are the primary source of information on which the appointment process is based. Furthermore, when an executive official is targeted for investigation, the independent counsel is under political pressure to find a violation. The counsel searches for a crime to pin on the executive official, rather than looking for the person to fit the crime, as traditional law enforcement does.
In Morrison v. Olson (1988) the Supreme Court upheld the constitutionality of the independent counsel act. In earlier decisions, such as Chadha,northern pipeline co. v. marathon pipe line co. (1981), and bowsher v. synar (1986), the Court had struck down acts of Congress as violations of the separation of powers. Its acceptance of the independent counsel was therefore a major victory for Congress. The independent counsel's functions appeared to be a "purely executive power" that under the Court's previous separation of powers decisions required placing the officer under the full removal power of the President. The Court dropped this line of analysis, however, concluding simply that the removal provisions of the act and other limits on the President's ability to control the discretion of the independent counsel did not interfere with the President's exercise of his constitutional duty to execute the laws. Morrison v. Olson in effect invited Congress to create additional special prosecutors to enforce laws that are judged to be too important to leave to the discretionary litigation policy of politically appointed officers in the executive branch.
Although the office of independent counsel was created ostensibly to assure impartial investigation of executive branch improprieties, it encouraged lawmakers to pursue policy disagreements with the executive branch in a partisan manner, to the point of criminalizing them. The Iran-Contra affair illustrated this tendency. It also epitomized the conflicts of divided government carried into the sphere of foreign policy. In this pivotal event, the President relied on executive discretion in foreign affairs, congressional committees investigated executive officials in the accusatorial manner established in the McCarthy era, and independent counsel obtained indictments leading to criminal trials that obviated impeachment proceedings.
The Iran-Contra affair initially involved the secret sale of arms to Iran, with which the United States had been in a state of undeclared war since 1979. The sale was a foreign policy maneuver intended to secure the release of American hostages held by Arab terrorists. This covert action arguably violated the requirement of the national security act that the President, in conducting such an operation, make a factual "finding" and notify Congress. The Reagan administration used the profits from the Iranian arms sale to aid the Contras in Nicaragua seeking to overthrow the left-wing Sandinista government. The foreign policy question here was whether the United States should support the rebels. Congress cast the dispute in legal terms, which was made easier by a series of riders it had attached to defense appropriations acts from 1982 to 1986. These riders, known as the boland amendments, prohibited any funds from being spent by any agency of the United States involved in "intelligence activities," for the purpose of supporting the military overthrow of the Nicaraguan government.
Fundamental constitutional issues concerning the powers of President and Congress in foreign policy were hung on the peg of narrow statutory questions concerning the meaning of the Boland amendments. Yet as products of legislative compromise and executive-legislative accommodation in the (unacknowledged) spirit of divided government, the riders were deliberately vague and ambiguous. A key question was whether the National Security Council (NSC) was an "intelligence activity" in the sense intended by the legislation. The riders did not expressly identify it as such, although they did so identify other agencies. A further question, assuming that NSC was covered, was whether its activities conformed to the permissible scope of intelligence activities assisting the Contras that the Boland amendments approved. Neither the Iran-Contra hearings nor subsequent judicial trials of NSC officials satisfactorily resolved these questions.
If basic constitutional questions could not be reached, answers to narrow legal questions were also elusive because of the basic ambiguity in congressional foreign policy. Sometimes Congress voted for military and humanitarian aid to the Contras; at other times it barred using military equipment to overthrow the Nicaraguan government. Congress desired to challenge the President's control of foreign policy, but was afraid of critical public reaction if it did so too clearly or extremely. Therefore, it resorted to imprecise statutory language that did not unequivocally block the executive branch from carrying out its pro-Contra policy. For his part, President Reagan did not veto the Boland amendments, either because he viewed them as a compromise that permitted the administration to pursue its policy or because he feared impeachment.
The latter was a reasonable fear, for when information about the Iranian arms sale was revealed in 1986 a Watergate-type impeachment mood gripped many law-makers and critics of the executive branch. The office of independent counsel, however, in conjunction with the traditional legislative power of investigation, was available as a less politically risky alternative. In November 1986, President Reagan promptly agreed to a congressional request to seek the appointment of an independent counsel to investigate the Iran-Contra affair. NSC officers Lt. Col. Oliver North, Admiral John Poindexter, and Robert McFarlane were the most prominent targets of the investigation, which resulted in at least six criminal convictions.
The trial of Oliver North illustrated the process by which constitutional controversy over the conduct of foreign policy was reduced to minor criminal convictions. North was a principal figure in the administration's policy who was called to testify in 1987 before the House and Senate select committees investigating the Iran-Contra matter. He was given a grant of criminal-use immunity, which meant that his testimony—televised to the nation in dramatically staged hearings—could not be used against him in any criminal prosecution that might result from the independent counsel's investigation. North's testimony was subsequently used against him, however, in a supposedly nonevidentiary way that arguably violated North's Fifth Amendment right against self-incrimination as interpreted by the Supreme Court in kastigar v. united states (1972). North was indicted on a dozen counts. He was not charged with violating the Boland amendments, and he was acquitted on the most serious charges, such as defrauding the government by diverting funds to the Contras. North was convicted on three charges: accepting an illegal gratuity, falsifying and destroying government documents, and obstructing Congress.
The precise purpose of Congress in the Iran-Contra affair was a matter of some dispute. In policy terms, the Democratic leadership opposed aiding the rebels, yet it pursued this policy ambiguously and inconsistently. More clear was the constitutional purpose of Congress—to assert the power of the legislative branch in foreign affairs. Historical practice and constitutional law recognized a broad sphere of executive power and discretion in this area. Congress therefore required cogent arguments against the administration's policy to overcome the presumption that tended to favor executive authority in the making of foreign policy, especially in the twentieth century. To bring foreign policy in a detailed operational sense under the rule of law, as the Boland amendments purported to do, asserted a congressional claim of authority to rival that of the President. Criminalizing the foreign policy disagreement through the use of independent counsel prosecutions was a suitable means of discrediting executive authority. An administration that conducted foreign policy by illegal or lawless means could not be considered constitutionally legitimate. This was the point that criminal convictions in the Iran-Contra affair were intended to make.
Criminal conviction was an effective policymaking tool because it could be justified by the traditional republican goal of imposing the rule of law on executive power, which is always the potential source of tyranny, according to the antiexecutive strain in American political thought. Although the Iran-Contra defendants were not tried for violating the Boland amendments, members of Congress did not hesitate to make accusations and to conclude that officials of the executive branch broke the laws enacted by Congress to prevent the United States from getting involved in a war in Latin America. Senators and representatives could make these accusations, confident that under the separation of powers doctrine their possession of the lawmaking power makes them superior to the executive branch and establishes a presumption that the will of Congress is tantamount to the rule of law.
Even if the Boland amendments are considered constitutional, it is questionable whether the lawmaking power can be effectively employed in making foreign policy. The reason is that in foreign affairs U.S. officials are required to deal with the representatives of other countries, who are not subject to the authority and rules of action that constitute U.S. law. The rule of law as propounded by legislative enactments under the separation of powers is further questionable because the purpose of foreign policy is to protect the public safety and national security. Prudence, wisdom, and discretion in the exercise of power are required to achieve this end, rather than the general and prospective rules of action that characterize the rule of law. The statesmanship on which the successful conduct of foreign policy depends has usually been thought more likely to result from the actions of the chief executive, who can be held politically accountable, than from the deliberations of hundreds of lawmakers in Congress.
In the 1980s, tendencies toward legislative assertiveness clashed with an executive authority that had significantly recovered from the power deflation and loss of respect suffered during the eras of the vietnam war and Watergate. President Reagan was in many respects a nonpolitical chief executive, uninterested in the details of partisan maneuvering and administrative management, who in spite of himself refurbished the presidential office. That he successfully served two terms and employed military force in foreign affairs without interference from Congress under the War Powers Resolution was evidence of executive branch revitalization.
Divided government contradicted the theory of presidential party government. Yet it was not, as some observers argued, a historical and procedural accident. On the contrary, divided government resulted from a reconsideration of the public philosophy of interest-group liberalism and the constitutional theory of the regulatory welfare state. It was a partial repudiation of the twentieth-century tendency toward governmental activism and centralized sovereignty. Divided government was a confirmation of the relevance and utility of the separation of powers principle, one of the basic concepts of limited government.
(see also: Bork Nomination; Civil Liberties; Civil Rights; Congress and Foreign Policy; Congressional War Powers; Conservatism; Constitutional Reform; Criminal Justice System; Economy; Establishment Clause; First Amendment; Freedom of Speech; Freedom of the Press; Line-Item Veto; Procedural Due Process of Law, Civil; Procedural Due Process of Law, Criminal; Racketeer Influenced and Corrupt Organizations Act; Religious Fundamentalism; Religious Liberty.)
Carter, Stephen L. 1988 The Independent Counsel Mess. Harvard Law Review 102:105–141.
Eastland, Terry 1989 Ethics, Politics, and the Independent Counsel: Executive Power, Executive Vice, 1789–1989. Washington, D.C.: National Legal Center for Public Interest.
Harris, Richard A. and Milkis, Sidney M., eds. 1989 Remaking American Politics. Boulder, Colo.: Westview Press.
Helco, Hugh 1986 General Welfare and Two American Political Traditions. Political Science Quarterly 101:179–196.
Lowi, Theodore J. 1985 The Personal President: Power Invested, Promise Unfulfilled. Ithaca, N.Y.: Cornell University Press.
Mansfield, Harvey C., Jr. 1985 Pride Versus Interest in American Conservatism Today. Government and Opposition 20: 194–205.
Shapiro, Martin 1985 The Constitution and the Bureaucracy. This Constitution 9:11–16.
Sundquist, James L. 1989 Needed: A Political Theory for the New Era of Coalition Government in the United States. Political Science Quarterly 103:613–636.
Wildavsky, Aaron 1980 How to Limit Government Spending. Berkeley: University of California Press.