In December 2014, in the must-pass legislation known as the “CRomnibus,” Congress raised the limit on contributions that political parties could accept for their conventions and their headquarters. Good-government organizations such as Common Cause, and their allies at every newspaper editorial board, issued familiar denunciations, decrying yet another opening to big money and corruption, and noting the sticker-shock sums that a single wealthy donor would now be able to give to influence the process: $777,600 a year, or $1.5 million over a two-year election cycle. Others pointed out that another controversial provision in the bill—a rollback of a key Wall Street regulation, reportedly drafted by lobbyists for Citigroup—showed precisely the influence of the massive political spending that the first provision would invite even more of.

To an increasingly influential group of scholars and journalists, though, the higher contribution limits were a welcome development: Letting political parties raise more money should not be seen as driven solely by “the petty motives of the party hacks,” as one political scientist, Ray La Raja of the University of Massachusetts-Amherst, put it, but as a positive social good. La Raja and others were not embracing Senator Mitch McConnell’s argument that all limits on political money trample the First Amendment, but making a modest and more pragmatic claim: There’s going to be money in politics, and for government to function, it’s better for that money to go through parties—even if, or perhaps because, they are “hacks”—than through other channels such as the ideological super PACs that have proliferated since 2010.

La Raja’s argument fits into a more comprehensive challenge to many of the assumptions and unexamined verities of those who aspire to reform the American political process, whether from the left, center, or even the Tea Party right. This school of skeptics is less interested in an idealized democracy—“democratic romanticism,” as one of them, NYU law professor Richard H. Pildes, calls it—than in effective governance. They worry less that government is pulled away from the public interest by money and concentrated power than that it can’t address problems in even the messy, transactional, and imperfect way that it once did.

The idea that American democracy should be more transparent and more inclusive, that it should put the broad public interest ahead of partisanship or local or private interests, is so benign that it’s hard to find a coherent argument against those aspirations. Who speaks for partisanship, patronage, corruption, or secrecy? The main obstacles to reform would seem to be the inertia of the political system itself, and the power of partisans, lobbyists, and donors who benefit from the existing campaign finance system, from patronage, and from “crony capitalism,” rather than any cogent political theory. (One exception is the First Amendment argument against campaign finance limits, which is a serious claim, but the Supreme Court’s recent decisions presume that corruption doesn’t really exist outside of quid pro quo deals, not that corruption is healthy.)

But the new skeptics of reform are not hacks, and they raise valuable critiques that deserve a hearing, even if they are sometimes vulnerable to romanticism, naïveté, and nostalgia of their own. Besides La Raja and Pildes, whose recent article, “Romanticizing Democracy,” in the Yale Law Journal is the strongest summary of the skeptical case, doubters include Bruce Cain of Stanford, author of the 2014 book Democracy More or Less; Jonathan Rauch of the Brookings Institution, who was brave enough to make an explicit “Case for Corruption”; in The Atlantic last year; and Jason Grumet of the Bipartisan Policy Center, author of 2014’s City of Rivals: Restoring the Glorious Mess of American Democracy. These books and articles vary greatly in tone and depth, ranging from Cain’s cool-eyed analysis of paradoxes in dozens of aspects of political reform at the state and federal level, to Grumet’s nostalgia for the era when handshake agreements were made in adjoining chairs in the Senate barbershop. But these authors share a belief that fixing politics would be a lot easier if we just learned to love the messy and selfish motives of transactional politics—though even the mess may be harder to achieve than they recognize.

The Skeptics’ Arguments

The skeptics generally share at least four specific arguments, although they approach each differently. First, they worry that transparency, an objective of many reformers, actually precludes the quiet backroom deals that allow politicians to back down from hard ideological pledges and find common ground. Second, they argue that eliminating earmarked appropriations took away from congressional leaders a kind of currency that had allowed them to buy off members and induce them to compromise. Third, they claim that restrictions on campaign money, particularly the “soft money” run through party committees that dominated campaigns in the late 1990s, have weakened parties and led to the dominance of ideological extremists. Finally, they express concern that the alternative solution favored by campaign reformers—systems that encourage small donors—will further increase polarization because small donors tend to hold more extreme positions.

Common to all of these arguments is the goal of strengthening parties, and in particular the power of party leaders to discipline members and manage the political and legislative process. In theory, this will weaken the hold of ideological extremists—or “intense policy demanders,” in the language of political science. The desire for strong and disciplined political parties is an old one among political scientists; in 1950, a task force of the American Political Science Association published a call for “responsible parties,” by which they meant parties with enough of a coherent policy agenda that they would present voters with a distinct ideological choice. (It was really about civil rights: They envisioned a Democratic Party that could stand forthrightly for justice, rather than being constrained by the Southern conservative Democrats who held disproportionate power in Congress.) But the reform skeptics see something different: They see parties as essential to mediating citizens’ engagement in democracy. And they see strong parties as an alternative to ideology that can center and stabilize the political system, because parties’ job is to win elections, which pulls them toward the views of the average voter.

Here is Stanford’s Cain, with the mildest statement of the skeptics’ theory of parties, which may sound familiar from your Poli Sci 101 textbook:

[Parties] provide cognitive assistance to voters looking for electoral shortcuts.… There is at least a reputational incentive for these information brokers to be reliable, which puts limits on their incentive to mislead. Politicians heed public opinion fairly closely, and despite party polarization they have so far managed to act when they have had to. In the end…the American politics system has produced an imperfect but stable and sufficiently effective government.

What Pildes calls “democratic romanticism,” Cain calls “populism,” by which he means not the redistributive economic policies advocated by, say, Senator Elizabeth Warren, but rather a naïve vision implicit in many reform strategies—including campaign finance reforms and visions of using the Internet to improve civic participation—of citizens who are deeply and directly engaged in all the issues of the day. Citizens have extraordinary capacity, and experiments in deliberative democracy have shown that they can master and reach consensus on the most complex issues. But most don’t have the time or interest, and reformers often hold an idealized view of democracy in which everyone is as absorbed and informed as they are. Whether it’s old urban political machines, parties, or interest groups, reformers often seem to disparage the very institutions that make meaningful participation possible and have—at times—produced an “imperfect but stable” system. Cain’s “reform pluralism” is a useful corrective.

But the second half of Cain’s description of political parties doesn’t sound much like the actual parties in our world, particularly the current Republican Party, which Thomas E. Mann and Norman Ornstein have described as “an insurgent outlier in American politics.” If politicians and parties “heed public opinion fairly closely,” it’s hard to explain how a party whose congressional wing had a 71 percent disapproval rating shortly before the 2014 election could nonetheless gain seats and expand its majority. (Congressional Democrats had a 65 percent disapproval rating.) There seem to be few “limits on their incentive to mislead” on issues such as the Affordable Care Act. The rule of thumb that parties and politicians gravitate toward the views of the median voter—which Jacob Hacker and Paul Pierson have skeptically called the “master theory” of American politics—has been put to the test and proven obsolete. A combination of identity politics, partisan redistricting, ideological media, money, and the use of aggressive tactics to control who votes and who doesn’t makes it possible for an extremist party or an ideological faction to gain control and hold it. The median voter has become little more than an angry spectator. The mechanisms that should stabilize American politics and pull it back toward the center don’t seem to be working, and the inequalities of power in the economy and politics only seem to reinforce one another.

Parties and Politics

At best, then, the parties as imagined by the reform skeptics should be considered aspirational—what we want parties to look like—rather than descriptive. Some of the skeptics, notably Pildes, acknowledge this, which is why he minimizes “polarization” as the issue and focuses instead on “fragmentation”: The parties are okay, but they have too weak a core to control their members, and thus extremists such as Senators Ted Cruz and Elizabeth Warren (yes, Pildes sees them as parallel; they’re not) dominate the agenda because they have their own base of donors. For Pildes, as for Rauch, the purpose of letting parties raise more money is to give party leaders the power, by withholding or providing campaign funds, to pull their members back toward the center, where they can negotiate and compromise with the other party.

While this logic is appealing in the context of some specific situations, such as Cruz’s role in nearly forcing a government default in 2013, it’s less persuasive as a general theory. For one thing, even in times of more productive government, there have been freewheeling legislative entrepreneurs, some destructive and others vital. Before there were Cruz and Warren, there were Jesse Helms, whose National Congressional Club, a direct-mail operation, could out-raise the Republican Party itself, and liberal Ohio Senator Howard Metzenbaum, who, like many legislators, was wealthy enough that he didn’t need any help from the party. (And, no, Helms and Metzenbaum weren’t parallel either, except in that both sometimes annoyed their colleagues.) Other legislative entrepreneurs have been centrists, defying their party’s leadership to cut deals across the aisle. If the Senate has any value, it is precisely its openness to political entrepreneurs of all stripes, who can demand votes on amendments and force new ideas and perspectives onto the agenda, often building new coalitions, things that are harder to do under tight party control.

Further, the parties aren’t nearly as fragmented as Pildes suggests. As Seth Masket of the University of Denver has argued, they are better described as “networked,” made up of organizations, especially fundraising vehicles, well beyond the official party committees. A recent profile of Republican National Committee chair Reince Priebus, for example, reported that rather than fighting with the funders and organizations outside the party, “Priebus has found a role in their ecosystem,” encouraging donors to give to his committees first and only then to deliver their millions to the supposedly independent groups. Parties are not nearly as marginalized in the fundraising business as the skeptics suggest, and even the Tea Party is a fully integrated component of the Republican network.

The skeptics’ case in favor of earmarks, despite its appealing contrarianism, is a good example of losing sight of the deeper changes in American politics. Earmarks—specific decisions about where and how to spend money written into legislation rather than left to administrative agencies—were not widely used except by the most powerful members of Congress until 1996, when Newt Gingrich made them available to his newly elected Republican members, so they could use pork-barrel spending to strengthen their re-election chances, much as Democrats elected to competitive districts in the 1970s had mastered the art of intensive constituent services. The number of earmarks quintupled between 1996 and 2005, but it was hardly the most productive decade in modern politics, and the practice was renounced after the 2010 elections. The swing of earmarks in and out of fashion, and now back in, illustrates one of the many dilemmas of political reform that Cain identifies: On the one hand, we want technical decisions about spending to be made by dispassionate professional agencies focused on the public interest; on the other hand, we want members of Congress to be responsive to their constituents.

But as a potential solution to the problem of governance, the key fact about earmarks is not that they aren’t available. It’s that junior members of Congress don’t want them anymore. House Speaker John Boehner and his deputies would surely like to have that currency to lure back reluctant members, but a currency has no value if members don’t want it. And they don’t want it because they don’t need what it buys—they can win re-election without bringing home pork-barrel spending projects. Governors have turned down multibillion-dollar federal infrastructure projects in the depths of the recession—and won re-election. Some members of Congress even renounced standard constituent services, declaring that they would not help anyone navigate the Affordable Care Act—and they, too, won re-election. The rules in this political world are unfamiliar, and simply restoring earmarks won’t bring a return to the customs of 1975 or 1995.

The skeptics bring a similar contrarian instinct to transparency, arguing that well-intentioned laws such as the federal Open Meetings Act or the Freedom of Information Act (FOIA), along with C-SPAN coverage of congressional proceedings, have the perverse effect of preventing negotiation, as legislators play to the cameras or take hard-line positions that they can’t back down from behind closed doors. But while not all disclosure is useful and some is harmful, the case that transparency is to blame for government dysfunction is unpersuasive. Transparency can be an invasion of privacy: Disclosure of small campaign contributions, for example, tells us less about politicians and influence than about the ordinary citizens who give modest amounts, and La Raja has found evidence that such disclosure might deter participation. Other forms of transparency increase public cynicism without giving citizens or journalists much useful information, and dump so much information that more is buried than is revealed. And, yes, some transparency may indeed make it hard to reach quiet bargains. But “transparency backfires” is as naïve a position as the familiar Louis Brandeis quote, “Sunlight is said to be the best of disinfectants,” which it isn’t always. There is plenty of room for middle ground. For example, in their 2007 book, Full Disclosure: The Perils and Promise of Transparency, Archon Fung and two colleagues from the Harvard Kennedy School suggest principles of “targeted transparency” that are based in how citizens, consumers, and journalists actually use information for constructive purposes.

Transparency rules are an exceptionally weak explanation of congressional dysfunction, as Gary D. Bass, Danielle Brian, and Norman Eisen pointed out in a paper published by the Brookings Institution, since there are ample opportunities for members of Congress to negotiate behind closed doors, and laws such as FOIA don’t apply to Congress. (Complaints about congressional negotiations breaking down because of transparency often cite leaks intended to stir up opposition, but leaks are not transparency. They’re a breach of an agreement to negotiate quietly.)

A Changed Washington

Members of Congress don’t sit down together to find common ground because they don’t want to, not because they can’t. Why don’t they want to? For two reasons: One, as noted above, they don’t need to. They can accomplish nothing and still win re-election, and nothing is often exactly what their financial backers prefer them to accomplish. Two, it’s issues, not just institutions, that can form a ground for compromise—but there are now almost no issues on which there is broad agreement on the need to do something, with only different views about how to do it.

Health care used to be one of those issues; it no longer is. The No Child Left Behind Act of 2001 is one of the last great cross-partisan compromises; a decade and a half later, education is now much more divisive, marked by the backlash against Common Core standards. Immigration reform and legislation to address climate change once seemed likely candidates for cross-partisan bargaining; both have fallen from the agenda. President Obama’s eagerness for a “grand bargain” on taxes and spending can be seen as a bid to revive what had traditionally been a locus for negotiation that often went well beyond fiscal policy, as in the 1990 budget talks at Andrews Air Force Base or in Gingrich’s pact with President Bill Clinton. But that avenue, too, has been closed. If the goal is a government that can get things done, then it’s useful to look at some of the particular things that should be getting done and consider why they’re not.

Often, the reform skeptics fall into the same trap as reformers, seeing a literal cause and effect between a procedural reform and a governmental result: If transparency or campaign finance limits coincide with an undesirable outcome, the reform must have had an unintended consequence. But all reforms fit into a bigger context of a changing political culture, and have different effects as that culture changes. Neither reformers nor skeptics pay enough attention to the role of race, region, and identity in shaping partisanship and politics. In particular, the reemergence of a one-party, conservative South, after four decades of more fluid and competitive politics in the region, has had a bigger impact on our politics than any procedural reform.

There’s a great deal of value in the skeptics’ overlapping arguments. In particular, it’s good to put the goal of having a government that gets things done at the forefront, rather than a utopian vision of a wholly equal, participatory process of democratic deliberation. Many campaign finance reform laws and proposals, especially those based primarily on limits, merely shift money around and, being poorly designed, may invest more power in those who pull the system away from consensus. Transactional politics has its uses. Transparency is not an end in itself and needs to be focused on the information that’s genuinely useful to democracy.

But is this a set of correctives, or an alternative way of looking at how American politics could work, a new set of reforms-in-reverse that would unleash the “glorious mess” of democracy? Most don’t go that far, but Grumet seems to, arguing that all that’s needed is to devise more ways for legislators to spend time together—in those closed rooms and that iconic Senate barbershop.

This nostalgia is romanticized in its own way. History suggests that a political system that is closed to outside scrutiny and tightly managed by party and committee leaders, one that is also awash in unregulated and unknown money and driven by quiet deals for votes that benefit parochial interests, not only falls short of a democratic ideal, but also fails by any measure of government effectiveness. That describes American politics in the Gilded Age of the 1890s or the Congresses of the 1950s, which were paralyzed in dealing with the urgent cause of civil rights. Meanwhile, the most productive period in American government, from about the mid-1970s through the bipartisan achievements of the early George W. Bush years, followed the post-Watergate reforms that opened the process to scrutiny, gave individual members more freedom to operate on their own, and put restrictions on money in politics. (The first part of this era was chronicled in a book by Ira Shapiro titled, accurately, The Last Great Senate.)

The question that books like Cain’s or articles like Pildes’ should lead us to ask is, how do we reform American politics so that their pluralistic vision—“imperfect but stable,” messy, unromantic—might actually describe reality, rather than the conditions of another era? How do we make voters matter again, so that politicians and parties feel the obligation to respond to the average citizen, whether of their own districts or the whole nation? How do we ensure that the inequalities of political and economic power, which are unavoidable, are not continually reinforcing each other in a cycle that will lead to societal stagnation? These questions, informed and inspired by these thinkers, could lead to a new and far more productive era of political reform.