The point of departure for my essay on "The Trouble with Limited Government" in the Claremont Review of Books was a National Review cover story that Senior Editor Ramesh Ponnuru wrote last November. In that article, he said that:

[T]he real crisis of conservatism…can be boiled down to two propositions. The first is that, at least as the American electorate is presently con­stituted, there is no imaginable political coali­tion…capable of sustaining a majority that takes a reduction of the scope of the federal govern­ment as one of its central tasks. The second [proposition] is that modern American conserva­tism is incapable of organizing itself without tak­ing that as a central mission. 

Thus, we are presented with "Ramesh's dilemma": There can be no political coalition built around the principle of limited government, yet limited govern­ment is one of the central principles of modern Amer­ican conservatism.

After thinking about the crisis Ramesh described, I went through the Office of Management and Budget historical tables that accompany the annual budget to compile some statistics for my own article. I used them to belabor the obvious, which is that conserva­tives have very little to boast about in their fight against Big Government. The Ronald Reagan-Newt Gingrich-George H. W. Bush era saw a more emphat­ically conservative Republican Party attain parity-or something a little better than parity-with a more hes­itantly liberal Democratic Party. Nevertheless, in 1981, the year President Reagan took office, the federal government spent $678 billion; in 2006, it spent $2,655 billion. Adjust that 292 percent increase for inflation, and the federal government is still spending 84 percent more than it did when Reagan became President-in a country whose population has grown by only 30 percent.

A conservative who wanted to put the best face on the past quarter-century would point out that spending by all levels of government in America- federal, state, and local-was 31.6 percent of gross domestic product in 1981 and 31.8 percent in 2006. Or he might say that conservatives haven't reduced the size of government but have caused it to grow more slowly than it used to. Per capita fed­eral spending, adjusted for inflation, was 41 percent higher in 2006 than in 1981; by comparison, it was 94 percent higher in 1981 than in 1956.

Conservatives who defend such "accomplish­ments," however, are clearly defining efficacy down. Conservatives who cheered when the newly in­augurated President Reagan declared his "intention to curb the size and influence of the federal establishment" did not anticipate that they would be asked to celebrate a "victory" that looks so much like a long retreat.

Even the stable size of the public sector, relative to GDP, says more about the dynamic economy of the past quarter-century-the Dow Jones Indus­trial Average closed at 777 in August 1982-than about restraining Big Government. A more pros­perous society will need more of some government functions than a less prosperous one. More people will travel by plane and drive cars, for example, requiring more roads, airports, and air traffic controllers. More children will spend more years attending school, requiring more teachers and classrooms. Generally, though, a growing economy should see more people able to spend more money on their own health, education, and welfare, allow­ing the government to spend less. It hasn't worked out that way.

The reason why conservatives have failed to rein in Big Government is that the task is very hard. If it were easy, it would have been done by now. The rea­son it's hard is that liberalism's sales pitch-we want the government to give things to you and do things for you-has never gone out of style. (The conser­vative counterargument, about how such a govern­ment is going to have to do things to you and take things from you, makes bigger, more problematic claims on voters' time and attention.)

Worse still from conservatives' perspective, liber­alism lends itself to narrowcasting. Liberals don't have to sell Big Government or the welfare state all at once. Every program creates its own constituency, and every constituency is much more interested in sustaining and expanding that program than the general public will ever be in ending or curtailing it. The correlation of political forces, then, is always pushing the scope of governmental activity outward.

What Is to Be Done? Can the Dilemma Be Solved?

So what is to be done? This is Lenin's question applied to Ramesh's dilemma. I have no tidy answer to offer today. I do wish to bring your attention to several considerations in the hope that conserva­tives can think through the trouble with limited government.

The responses to Ramesh's dilemma fall into two broad categories: We can either reject it or accept it. Rejecting it means challenging its two propositions. Accepting the dilemma, on the other hand, means accepting that both propositions are true, and that they're incompatible, and then trying to figure out a way to either live with or resolve Ramesh's dilemma.

Can We Dispute Its Propositions? Can a Majority Be Built Around the Idea of Limited Government?

Let's talk first about rejecting the dilemma by disputing its propositions. The argument against Ramesh's first proposition is that a majority coali­tion for reducing the scope of the federal govern­ment is not unimaginable. That proposition rests on an interpretation of recent political history. One of the most important things graduate schools try to teach young historians is to be on guard against the fallacy of "retrospective determinism"-the belief that because something did happen, it had to have happened. Once an event has occurred, it acquires the weight of its own factuality. It becomes seduc­tively clear that every relevant prior event was pav­ing the road that led, inevitably, to the occurrence under consideration.

Retrospective determinism is a problem not just for historians, but for anyone trying to make sense of some slice of the human record. You see it on the sports pages every day. The Boston Red Sox won the World Series last month, giving rise to dozens of articles about the superiority of their management, scouting, and personnel decisions. Boston, how­ever, trailed the Cleveland Indians three games to one in the American League Championship Series. If Cleveland had managed just one more victory, Boston would not have gone to the World Series, and none of those articles about how the Red Sox are so well run would have been published-even though the objective facts about the organization wouldn't have been any different.

The corollary of the proposition that because something did happen, it had to have happened is that because something didn't happen, it couldn't have happened. The fact that conservatives didn't rein in Big Government after 1980 more or less proves they couldn't have. Just because this is our brains' default option for making sense of things doesn't prove that it's wrong, of course. When the history of the Reagan-Gingrich era is written, the impossibility of downsizing government may turn out to be the best explanation for what did and didn't happen.

But it's not the only possible explanation. There's a case to be made that the Reagan-Gingrich attempt to curtail Big Government failed for reasons that were accidentally rather than essentially related to the enterprise. If different calculations had been made, if different people had been in power, then conservatives might have made a lot more headway than they did. The depressing account of the 1995 government shutdown in Major Garrett's book on the class of '94 Republicans, for example, shows that despite all the miscalculations that led to that juncture, the GOP was still very close to achieving real reductions-until panic set in and Republicans snatched defeat from the jaws of victory.

We don't have time today to scrutinize fiscal pol­itics over the past quarter-century. I raise the point only to suggest that if we accept or even entertain the idea that the past might have turned out differ­ently, then it's possible to be less fatalistic about the prospects for conservatives who want to limit gov­ernment in the future. Even if I had more time and wanted to talk about nothing else today, I don't believe that I would try to strike through Ramesh's first proposition. But I believe it's possible to at least put a question mark after it.

Can Modern American Conservatism Abandon the Cause of Limited Government?

The argument against Ramesh's second proposi­tion would be that conservatism is capable of orga­nizing itself without taking the reduction of the scope of government as a central task. Conserva­tives needn't become enthusiasts for Big Govern­ment or even stop opposing it. The terms of the proposition lend themselves to making the reduc­tion of government a second-order goal rather than a central one. Since, if Ramesh's first proposition is true, negligible gains against Big Government are what conservatives are going to have to settle for anyway, quietly reordering our priorities may be only a concession to reality.

The question is about the nature of such a post- limited government conservatism. Conservatism will have to be about something else. Conservatives will conserve or restore other things, not limited government.

Generically, conservatives want to conserve things that are both valuable and vulnerable. If they're valuable, they deserve conservation, and if they're vulnerable, they need it. In America, the republic- the experiment in self-government-is both valu­able and vulnerable. Conservatives who make its stewardship a high priority might concentrate on defending the nation against those who threaten it. Opposing the Soviet Union was a central task for conservatives for nearly half a century. Opposing jihadism is such a task now, and wary vigilance against China may be part of it in the future.

The domestic agenda for post-limited govern­ment conservatives could be concerned with pro­moting the dispositions and habits in the citizenry that are essential to the success of the experiment in self-government. Whatever you think of the group that went by the name, America does need a moral majority, because enlightened self-interest can do only so much to support the cause of republican government.

There's no guarantee that a post-limited govern­ment conservatism preoccupied with the politics of national security and the sociology of virtue will either (a) command the assent and enthusiasm of enough conservatives to form a coherent political force or (b) attract enough votes to be more elector­ally competitive than limited-government conserva­tism. Indeed, the enthusiasm, coherence, and votes lost by downgrading the importance of limiting government may outweigh the gains. But conserva­tives are going to look for alternatives to watching their movement succumb to the crisis Ramesh has described, and this attempt, or something like it, will be one of them.

Dealing With, or Resolving, the Dilemma

The Libertarian Response

The alternative to rejecting Ramesh's dilemma is accepting it, which means finding a way to either live with it or resolve it. Libertarians have their way of living with the dilemma. Reducing the scope of government in America is, for them, a non-negotia­ble principle. Thus, it is better for conservatism to lose elections, even lots of elections, than lose its soul. If push comes to shove, libertarians would rather see conservatism become, again, a scorned fringe movement than one that compromises its rai­son d'etre in order to be electorally competitive.

Libertarians, however, hope push won't come to shove and believe they have a way to resolve Ramesh's dilemma. Their hope is that the principle of limited government will pervade the entire con­servative agenda and attract enough votes to make it electorally competitive. They want to apply that principle to not only the welfare state, but social policy and national security as well.

The result is an ideology that is a photographic negative of the post-limited government conserva­tism we've sketched out. Welfare state programs would be cut back, lifestyle diversity would be tol­erated serenely, and threats to national security would be construed narrowly enough to justify a diminished military.

Libertarianism asks conservatives to say "Wel­come aboard" to a lot of voters who don't think of themselves as particularly conservative and puts a lot of other conservatives who've been on board a long time into the life rafts. The question that must be posed about the libertarian resolution is the same one that must be posed about post-limited govern­ment conservatism: Is the number of voters it gath­ers in greater or lesser than the number it drives out?

Post-limited government conservatism may be problematic, but it's hard to see how the conserva­tive coalition can expand by doing all the con­tracting libertarians have in mind. Once you've shown social conservatives the door and then ushered out conservatives who stress the vigorous prosecution of national security, you're going to have to find lots of new voters to come out ahead. Perhaps the marginalization of conservatism is not only a result libertarians are willing to accept, but one they actually welcome.

The Neoconservative Alternative

There's a second way to accept Ramesh's dilem­ma. I'm afraid I'm going to have to call it the neo­conservative alternative. I hesitate to do so, since that term has been the source of so much confusion and controversy. It was, however, the godfather of neoconservatism, Irving Kristol, who called for a "conservative welfare state." For my purposes today, I'm using "neoconservatism" strictly to refer to the efforts to grapple with Kristol's term, to figure out what a conservative welfare state can and should mean.

The libertarian position, of course, is that a con­servative welfare state is a contradiction in terms, and the effort to achieve one is going to be a waste of time, at best, and probably counterproductive in the sense that accommodating the welfare state will change conservatism for the worse much more than conservatism will change the welfare state for the better. The question, then, is whether conservatives can conservatize the welfare state without abolish­ing it and what such an endeavor would mean.

There are two possible answers to this question, distinct but not mutually exclusive. The first is that the central task of a conservative welfare state is to mitigate the economic and social damage inflicted by the liberal rendering of it. The second is that the conservative welfare state will work to undo the political damage caused by the liberal one.

Mitigating the Economic and Social Consequences of Modern Liberalism

Conservatives who want a welfare state that mit­igates liberalism's economic and social consequenc­es look at Western Europe's social democracies- France in particular-and say, "That's what we don't want to happen here." What they mean is that we don't want to have policies and attitudes that dis­courage the multiplication of wealth but encourage the division of it. We don't want dependency, we don't want endless clamoring for ever-larger entitle­ments, we don't want endless hectoring of entrepre­neurs, and we don't want regulation to reach the point that private enterprise becomes a contradic­tion in terms, since businessmen worry much more about placating the bureaucracies that govern their destiny than about satisfying their customers.

George Will has recently described the welfare state tempered by these fears. It will, he said, "presuppose economic dynamism sufficient to generate investments, job-creation, corporate profits and individuals' incomes from which come tax revenues needed to fund entitlements." Thus, it will worry much more than American liberals and European social democrats are inclined to about excessive taxes, regulations, and government spending. Moreover, says Will, a conservative welfare state will work to encourage rather than discourage "attributes and attitudes-a future orientation, self-reliance, individual responsibility for healthy living-that are essential for dignified living in an economically vibrant society."

The conservative welfare state, like the liberal one, rests on the belief that a decent society is obli­gated to prevent the small minority of citizens who are chronically unable to fend for themselves, and the larger minority occasionally and transitionally unable to do so, from leading miserable lives and that government programs will be a necessary part of how a decent society acts on that concern.
The conservative welfare state, unlike the liberal one, emphasizes that a nation wealthy enough even to have a welfare state is wealthy enough to have lots of people who don't need most of what the welfare state provides. It will express that belief by limiting its programs to poor people through means-testing and by devising, for people who aren't poor, incentives like Health Savings Accounts and 401(k) plans to keep them out of poverty and, thus, ineligible for the means-tested programs.

Mitigating the Political Damage of Modern Liberalism: Rebuilding the "Legitimacy Barrier"

The second understanding of the conservative welfare state-determined to mitigate the political damage inflicted by liberalism-takes its bearings from the political scientist Theodore Lowi, who's not even a conservative. Lowi argued 40 years ago in his book The End of Liberalism that modern liberalism has left us with "a government that is unlimited in scope but formless in action." Such a government "can neither plan nor achieve justice" because, Lowi says, "liberalism replaces planning with bargaining" and creates a regime of "policy without law."

For conservatives who view our political history this way, the decisive break occurred not in 1933, when the New Deal started creating new programs and spending more money, but in 1937, when the Supreme Court basically threw in the towel. In a series of decisions upholding exactly the sort of programs it had been striking down during and before FDR's first term, the Court replaced the Con­stitution of enumerated powers and limited gov­ernment with a new, anything-goes constitutional framework.

In the words of the legal historian Bernard Schwartz, "To the post-1937 Supreme Court, the Congress, charged with all of the legislative powers granted by the Constitution, is entitled to its own choice among all rationally permissible opinions as to what the Constitution allows." The spirit of this new understanding moved one law professor to argue recently that the "proper constitutional response" to concerns about Congress overstepping its authority and regulating any activity it chooses is, "So what?"

The public's understanding of the Constitution has followed the Court's, and vigilant antipathy to government expansion, which was once a decisive factor in American political life, is now a faint mem­ory. The political scientist James Q. Wilson calls this development the fall of the "legitimacy barrier." In the aftermath of that fall, "New programs need not await the advent of a crisis or an extraordinary majority, because no program is any longer 'new'- it is seen, rather, as an extension, a modification, or an enlargement of something the government is already doing. Since there is virtually nothing the government has not tried to do," Wilson says, "there is little it cannot be asked to do."

The crucial task for this second understanding of the conservative welfare state is to rebuild a legiti­macy barrier that clearly distinguishes which goals the government may pursue, and which means it may employ, from those it may not. Such an endeav­or will be different from the libertarian goal of zero­ing out the welfare state.

Michael Greve, the director of the American Enterprise Institute's Federalism Project, admires the jurisprudence of the pre-1937 Supreme Court. He notes, however, that much of the New Deal, much of the welfare state generally, would be per­mitted by the kind of legitimacy barrier that existed then and which he would like to see again. "New Deal programs that served a discernible public pur­pose were never [in danger of being ruled unconsti­tutional]," he writes. "Nothing in the Constitution precluded the New Deal from paying unemployed artists to adorn U.S. post offices with Soviet-Realist murals. And contrary to progressive myth, a prop­erly structured Social Security system was never likely to be invalidated" by the Court.

The question is whether conservatives can re­build the legitimacy barriers demolished in the 1930s and, if so, where? The pre-1937 Constitu­tion constrained government because jurists em­phasized the due process and interstate commerce clauses, the enumeration of congressional powers, and the impermissibility of Congress delegating those powers to administrative agencies. The re­construction of a legitimacy barrier formed by those particular restraints is a difficult and doubt­ful prospect. The alternative to pursuing it is to erect the legitimacy barrier elsewhere. What these new restraints on government will be, and how they can be enacted and enforced, are difficult questions some conservatives have wrestled with but to which there are, as yet, no clear and com­pelling answers.

Conclusion: Limited Government vs. Small Government

I wish I myself had some clear and compelling answers. Then, instead of just stopping my talk, I could conclude it with wise and persuasive words about the present state of conservatism and where we should go from here. I'm afraid, however, that the best I can do is to leave you with a distinction that arises from the second variant on the idea of a conservative welfare state. That distinction is between small-government conservatism and limit­ed-government conservatism.

Small-government conservatism is opposed to Big Government. But "big" and "small" are relative terms, and liberals are always working to move the goal posts. "We're in favor of a lot of things and we're against mighty few," Lyndon Johnson said in 1964, and that's about as rigorous as the liberal theory of the welfare state ever gets.

American liberals wake up every morning think­ing about all the suffering and injustice they could alleviate if only the U.S. public sector weren't forced to scrape by with 32 percent of our GDP. The trou­ble is that Sweden's social democrats wake up every morning thinking about all the suffering and injus­tice they could alleviate if only their public sector weren't forced to scrape by with 57 percent of Swe­den's GDP. American liberals speak frequently and forcefully about all the ways the U.S. would be a better place if we ran this country more like Swe­den. They speak rarely and vaguely about which features of the European social democracies, if any, they consider too expensive or too intrusive.

Limited-government conservatism confronts this problem more directly, because it is opposed not to Big Government, but to Unlimited Govern­ment. The evil it works against is not that liberalism gives us Big Government but that liberalism lacks a limiting principle. The absence of such a principle is not just an oversight, not something liberals have been meaning to get around to for 75 years but which somehow got lost in their in-basket. The inability-the refusal, really-to say what it would mean for the welfare state to grow too big and spend too much is a defining feature of American liberalism.

Conservatives who want to fight that battle on behalf of limited government are going to need an answer to the question, "Limited by what?" Histori­cally, American government was limited by the principle that it existed only to secure our natural rights. The future of the conservative movement will depend heavily on three questions:

• Can the natural rights principle be reinvigorated as a constraint on government?
• If not, can some other principle effectively restrain the growth of government?
• And, finally, if the natural rights principle is dead and nothing new can take its place, how do conservatives arrange terms of surrender to the advocates of Unlimited Government and move on to fight other battles?

William Voegeli, Ph.D., is a Visiting Scholar at the Henry Salvatori Center for the Study of Individual Freedom in the Modern World at Claremont McKenna College in Claremont, California.