The Constitution gives control over matters of war and peace to the president and Congress, in the words of Alexander Hamilton, as a “joint possession.” This pretty much guarantees continuing strife between the two branches. Yet after two centuries of a seesawing contest for primacy, lines have been crossed in the last year that take the relationship between the two branches into new territory. What has happened raises the question of whether the structure of political power in the United States today dangerously restricts what the country can aim to achieve abroad.

Never before has Congress openly tried to undermine a president in the midst of negotiation with a foreign adversary. In 2015 it did so twice: by inviting Israeli Prime Minister Benjamin Netanyahu to attack President Obama’s policy on Iran’s nuclear program before a joint session of Congress, and through an “open letter” to Iran’s leaders, drafted by Senator Tom Cotton of Arkansas and signed by forty-six fellow Republican senators. The letter asserted (incorrectly) that any agreement not approved by Congress would have no legal validity and threatened to overturn it at the legislature’s will. Knowing, as the senators did, that Iran’s Supreme Leader and political elite were already deeply skeptical of Washington’s trustworthiness and of the administration’s ability to fulfill a deal, the letter appeared designed to strengthen the deal’s opponents in Tehran.

Later in the year, Senate Majority Leader Mitch McConnell (who also signed the Cotton letter) tried to derail another major negotiation. He dispatched an aide to meet with foreign emissaries to describe the GOP’s plans to overturn any deal reached at the upcoming Paris summit on climate change. “The president’s international negotiating partners,” McConnell warned, “should proceed with caution before entering into an unattainable deal…because commitments the president makes there would rest on a house of cards of his own making.” Both interventions—on Iran and climate—were made while the outcome of these negotiations was very much in doubt.

It is true that something similar may have happened before—in Richard Nixon’s efforts to derail a cease-fire negotiation with the North Vietnamese during the 1968 presidential campaign, and in an alleged deal between Ronald Reagan and Ayatollah Khomeini to delay release of the American hostages in Iran until after the 1980 election. The intense secrecy that surrounded both episodes is the point: the interventions were recognized as far beyond the pale. In a taped phone conversation between President Johnson and Senate Minority Leader Everett Dirksen, Johnson describes the Nixon team’s actions, saying, “They oughtn’t to be doing this. This is treason.” Dirksen’s reply, in full: “I know.” Neither Nixon nor Reagan acted on behalf of the legislature, but each was, at the time, his party’s leader, as McConnell was last year.

Some commentators called Senator Cotton’s letter treasonous, but largely it was dismissed as embarrassingly silly. Both it and the invitation to Netanyahu seemed to be simply further instances of the ferociously politicized reflexive taking of positions along party lines that has become all too familiar. McConnell’s intervention in the negotiations before Paris went almost unnoticed. It was hardly news that the GOP was determined to block action on climate change.

To judge whether discord between the executive and the legislature on foreign policy amounts to more than the partisan polarization that afflicts all issues today requires a wider perspective than the past year or even the last seven. It has been a quarter-century since the end of the cold war in 1991, the last major global milestone. Is there a coherent record of behavior by both branches of government through these twenty-five years? If so, how does it fit with the global scope of Washington’s interests, the world’s deepening interdependence, and the American people’s expectations of their country’s position abroad?

In its allocation of overlapping powers to the executive and the legislature, and its many silences on foreign policy, the Constitution bequeathed, in the words of Edward S. Corwin, one of the great constitutional scholars, a permanent “invitation to struggle.” The president may be commander in chief, but Congress is given the power to declare war and to raise armies. The president has the authority to “take care that the Laws be faithfully executed,” but Congress has the responsibility to write them. The president’s power to receive foreign envoys was extended by President Washington to include the right to decide on diplomatic recognition, but Congress must confirm ambassadors and appropriate funds for embassies. Just two foreign policy decisions are explicitly reserved to the legislature: to declare war and to approve treaties signed by the president by a two-thirds vote of the Senate.

Through congressional default, the war power has withered to almost nothing. Congress has formally declared war five times in the country’s history—most recently after Pearl Harbor. But presidents have begun military action more than one hundred times without prior congressional approval. The 1973 War Powers Act attempted to clarify the executive and congressional authority, but it satisfied no one and has been ignored.

Congress authorized the use of force before each of the major wars since the cold war (the 1991 Gulf War, Afghanistan, and Iraq). In the Gulf War and Iraq, however, very large forces had been deployed to the region long before the vote, making it both likely that a decision had already been made to use them and very difficult for Congress to oppose. But today’s expanding operations against the Islamic State (ISIS) in Syria and Iraq are sanctioned only by the Iraq War Resolution of 2002 (with Saddam Hussein and his supposed WMD threat gone in 2003) and by the 2001 Authorization for the Use of Military Force (AUMF) against those responsible for the September 11 attacks. ISIS did not exist at the time.

A year ago, six months after launching operations against ISIS, the president asked Congress for a new AUMF, though he asserted that as a legal matter he did not need it. His proposed language was seen as too open-ended or as too restricted and weak, and virtually everyone thought it too vague. But the real reason that Congress has not acted is the one that has held it back so often before: members see this as a vote that will not please their constituents no matter how they vote. The administration’s ineffectual push for the AUMF has made it easy to do nothing. Both branches, it seems, prefer to fight this war without having to explicitly draw Americans’ attention to the fact that we are engaged in yet another war in the Middle East or to confront the unpleasant task of defining its goals.

Congress’s record on treaties, on the other hand, is decisive. In the past twenty-five years it has found almost no multilateral treaties—other than on trade—that it can pass. Trade agreements are a unique, hybrid type in which the president negotiates under a prior grant of congressional approval. Congress approved the final North American Free Trade Agreement (NAFTA) in 1993 and the highly significant and even more controversial Uruguay Round trade deal creating the World Trade Organization, which squeaked through a year later.

In these same hopeful years after the end of the cold war, Congress approved three multinational treaties: the International Covenant on Civil and Political Rights (which had been awaiting action since 1966); the Framework Convention on Climate Change; and the Chemical Weapons Convention, banning the production and use of these weapons of mass destruction. Because the Framework Convention only set a broad goal of avoiding dangerous climate change and contained no limits on emissions or means of enforcement, it was an easy vote for President George H.W. Bush to recommend. By contrast, the Civil and Political Rights agreement—which guarantees the right to life, freedom of religion, and freedom of speech, among other things—presented heavy political obstacles. When it was finally ratified,1 the Senate did so with five reservations, five understandings, four declarations, and one proviso, the net effect of which was to remove, in great part, the treaty’s substance.

The rest of this quarter-century has been a wasteland for multilateral commitments. There are many ways to defeat an accord. The Senate can vote down a treaty. More often it simply leaves an agreement on the shelf, sometimes for decades (it took thirty-nine years—from Truman to Reagan—to approve the Genocide Convention2). Increasingly, administrations choose to avoid a fight by not sending forward a signed treaty that the Senate is known to dislike. When opposition is strongest a president may choose not to sign a treaty, or to sidestep the Senate by negotiating an executive agreement instead (more on this below).

By one of these means or another, the Senate (sometimes together with President George W. Bush) has turned away more than a dozen major international accords in this period on matters of the environment, arms control, the rule of law, and public health. They include one—the Biodiversity Convention on species conservation—to which every other member of the United Nations (195 of them) is a party. Most of the rest have upward of 160 state parties. Other major agreements the US has chosen not to join include the Law of the Sea Treaty, the Comprehensive Test Ban Treaty (CTBT), the Kyoto Protocol on climate, the Protocol on Torture, and a convention protecting people with disabilities. The US has also refused to approve treaties to protect genetic resources, ban persistent organic pollutants, ban cluster bombs, restrict trade in conventional arms, and control international trade in tobacco. President Bush withdrew President Clinton’s signature on the International Criminal Court statute (though there is no doubt that the Senate would have voted against it), and three presidents have left the anti-personnel land mine ban unsigned.

In a few cases, there are real differences between US policy and the corresponding international agreement, regrettably, on torture, on climate, and on anti-personnel land mines (which the government believes it still needs in Korea). But in most cases there is little or no substantive distance between the US position and international agreements that cannot command the support of a post-Reagan Republican president or two thirds of the Senate.

The opposition of the senators is based more on symbolism, ideology, or mythically exaggerated fears of loss of sovereignty. The Law of the Sea Treaty has been strongly supported by administrations of both parties, by the military, and by environmental advocates. It remains unratified after more than twenty years, although Washington recognizes its provisions as general international law. There is no foreseeable prospect of ratifying the Comprehensive Test Ban Treaty either, though both of the original arguments against it have fallen apart. The US has observed a unilateral moratorium on nuclear testing for twenty-four years now—long enough to thoroughly negate the argument that it must test to assure the safety and reliability of its nuclear weapons. The technology to assure that the world would know if a country cheats has also been fully demonstrated.

The required two-thirds vote makes treaty approval hostage to small, but passionate, minorities. The homeschooling lobby played a large part in defeating the Convention on the Rights of the Child, for example, for it objected to children having the right to publicly supported education. As interest groups—especially well-funded ones—have acquired more and more influence over US politics in recent decades, it becomes easier to block anything that requires a supermajority.

But what is most notable, and most ironic, about the list of rejected treaties is that the United States actively supports nearly all of them—indeed in several cases is the most active international supporter. At regular meetings of the parties to each treaty the American delegation is frequently the biggest, but as a nonparty it is a guest, sitting on the sidelines. Its delegates are backed by money, science, commitment, strong support from American nongovernmental organizations—everything but the right to vote as the treaty regime is developed.

Even more striking is the fact that most of these agreements were American initiatives to begin with or are concords that put American domestic law into international form. The test ban, rejected in 1999, follows a vision first offered by President Kennedy and sought by the US for decades. The Convention on the Rights of Persons with Disabilities is modeled on the Americans with Disabilities Act. The Biodiversity Convention follows American conservation laws. The Framework Convention on Tobacco copies American anti-smoking legislation. Still the international versions have been steadily rejected.

A president’s ultimate solution to legislative opposition is to turn from treaties to executive agreements. Neither the Constitution nor The Federalist Papers makes any mention of such unilateral exercises of executive authority. But notwithstanding the howls of outrage on Capitol Hill last year challenging the president’s right to reach a binding agreement with Iran without submitting it to Congress, there was nothing at all unusual in President Obama’s choice of this method. Indeed, executive agreements have been the principal means of doing the government’s international business since the beginning of World War II—and by no means only for minor matters.

Momentous accords of that era include the Atlantic Charter and the Yalta and Potsdam agreements. In the year 1952, the president made 291 executive agreements, more than the total reached from 1789 to 1889. The trend continued over subsequent decades as the executive’s authority in foreign affairs expanded and as the sheer amount of international engagement made the slow and cumbersome process of treaty approval unfeasible.

Presidents of both parties have used executive agreements for the most weighty and controversial commitments, including Richard Nixon’s 1973 peace agreement with Vietnam and Gerald Ford’s 1975 Helsinki Final Act, which legitimized the boundaries of the USSR. In the five presidential terms in office from Jimmy Carter through Clinton’s first term, American presidents reached an average of eight hundred executive agreements in each four-year term.

If anything, there have been times when it seemed to members of Congress as though major decisions were reached through executive agreements while the treaties on which they had a say were relegated to the small stuff. “We get many treaties dealing with postal affairs and so on,” complained Senate Foreign Relations Committee Chairman William Fulbright. “Recently, we had an extraordinary treaty dealing with the protection of stolen art objects. These are treaties. But when we put troops and take on commitments in Spain, it is an executive agreement.”

Many members of Congress and countless commentators have insisted, as senators did in the Cotton letter, that an agreement with Iran or on climate change would be meaningless, valueless, illegitimate, and, above all, not binding without congressional approval. They are wrong. There are legal disputes about executive agreements, as there are about everything constitutional, but under international law the domestic process by which a country makes an international commitment is irrelevant. Thus an executive agreement is no less binding than a treaty. A country that chooses to repudiate a commitment it has made, whether by treaty or executive agreement, accepts the consequences.

To the limited extent that the Supreme Court has ruled on these matters, its 1936 decision in US v. Curtiss-Wright Export Corp., reached by a 7–1 vote (and ironically drafted by a former member of the Senate Foreign Relations Committee), is unequivocal. The president, the Court held, is “the sole organ of the federal government in the field of international relations.” He (or she) “alone has the power to speak or listen as a representative of the nation…. Into the field of negotiation the Senate cannot intrude; and Congress itself is powerless to invade it.” Based on law and custom, of the past six decades especially, what makes the Iran deal unusual is not that it was an executive agreement, but that President Obama was forced to compromise and submit it to Congress for approval.

Much, then, is not new. Discord between the president and Congress over their respective responsibilities in foreign affairs is as old as the Republic. Periods in which the Senate constantly disagrees with the president have occurred before. John Hay, secretary of state to both William McKinley and Theodore Roosevelt, wrote that he doubted that “another important treaty would ever pass the Senate.” Woodrow Wilson dubbed the Senate’s treaty-making power its “treaty-marring power.” And avoiding Congress altogether, even on the most critical decisions of war and peace, is a well-established practice of both Republican and Democratic presidents. What is new is a dramatically altered political environment, the intensely polarized character of the current struggle, the instant and worldwide exposure of American differences in today’s media, and most fundamentally the consequences for American foreign policy.

Externally, the world could hardly be more different from 1991. As powerful as it remains, the US can no longer take dominance as a given, except in the military sphere. Its share of the global economy has shrunk as a result of the success of its efforts to foster others’ economies. Threats that metastasize across the borders of collapsing states, whether as refugees or terrorists, are less amenable to the application of military force than are those from a strong opposing state. And because countries no longer have to choose between a Soviet or American security umbrella, the US can no longer count on automatic followers. Today, the support of foreign countries must be earned, and earned anew, by policies that meet their interests. Washington can still twist the arms of many nations, but in recent years it has been surprised over and over again by the world’s willingness to make international policy in the face of American disapproval—for example in banning land mines that devastate economies and civilians, especially children, and in creating an international criminal court.

But while America’s power is relatively diminished, the scope over which it must act is dramatically larger. What the globalization of the last two decades means in practice is that matters that were once domestic are now international. Whether it is energy or agriculture or air pollution, the capitalization of banks, the control of viruses, climate change, or the capture and punishment of pornographers, the concerns of foreign affairs have grown exponentially. And on most issues the relevant solutions are not bilateral or even regional accords, but fractious global ones.

For many Americans, and for today’s Republican Party in particular, this not-so-new state of affairs ranges from discomfiting to unacceptable. Negotiations, because they require compromise, seem acts of weakness. Settling for the best outcome in a bad situation seems like less than a hegemon should require. Calls on the campaign trail to recapture American greatness, and the enthusiastic response they receive, are not so much considered demands for more toughness or greater military strength as they are a vague wish for a world that is past.

Washington is under no obligation to approve every multilateral treaty. Some treaties are more hortatory than they are wise. But a pattern of rejecting virtually all of them is unmistakably not in the US interest. Life as a treaty nonparty deprives Washington of a vote as new policies and institutions take shape, and weakens its leverage over others’ choices. It drains moral authority. If they were parties to the Comprehensive Nuclear Test-Ban Treaty, the US and China would be in a stronger position, for example, to deal with North Korea’s nuclear tests than they are today.3 And when US participation is essential to international success as it is, for example, on climate change, the US’s inability to negotiate and join a multilateral commitment lowers the bar for the world.

If the US can bind itself through an executive agreement, does it matter that the Senate is passing through an era when it does not approve multilateral treaties? It does matter, because a treaty, carrying the weight of the Constitution, requiring a public debate, and bearing the approval of both branches of government shows a more serious national commitment. US v. Curtiss-Wright stands as legal precedent, but political considerations now matter more, and the polarization of American politics is so deep and politicization so all-encompassing that the traditional deference to presidential authority on foreign affairs is gone.

Future presidents can certainly do better than President Obama has in trying to work with Congress, but when opposition to a president from the other political party is nearly automatic, it is likely that the bitter divisions of our politics are going to prove increasingly costly to the international leadership that most Americans expect and to a world that badly needs American leadership. One way to estimate what that might mean is to ask what current politics would have meant to successes of the past.

In the 1960s authorities expected thirty or more nuclear-armed states by the turn of the century. That there are nine today is due to the highly imperfect Non-Proliferation Treaty. Like the Iran deal, its commitments were only temporary. It allowed nonnuclear states access to fissile material, reprocessing, and enrichment because that was all that could be negotiated. And it committed the nuclear states to the extreme step of nuclear disarmament. It was flawed, but it was enough to build on. It is not pleasant to contemplate how many nuclear powers there would be today if the US had rejected the NPT in 1968.

More recently, the discovery of a large hole in the stratospheric ozone layer triggered an international response in the 1980s. The Montreal Protocol called for a 50 percent cut in the chemicals that caused the problem, but gave treaty parties leeway to change the number based on evolving science, without renegotiation. The cut was soon tightened to a ban. The ozone hole has since healed. Would today’s Senate politics and special interest influence allow approval of such a treaty?

The globalized world in which every country’s well-being depends on a thickening web of international rules and cooperation in nearly every sphere is not going to reverse course anytime soon. An America that can be counted on to say no as this process continues won’t look like a leader, even one leading from behind, but more like it is expecting to be able to lead from outside the room. Nor is it obvious how or when our disabling partisan dissension might be eased. Unless Congress and the executive branch can find ways for Washington to speak more often with a unified voice abroard, and in a voice that recognizes growing global interdependence, the United States, and the world, will pay.

This article originally appeared in the New York Review of Books.