The conventional wisdom about Sen. Edward M. Kennedy's withdrawal from the Democratic presidential race is that it marks the demise of liberalism, now consigned to the Smithsonian's attic. The way is supposedly clear for a new generation of rightward-leaning "centrists," typified by Rep. Richard Gephardt of Missouri, Gov. Bruce Babbitt of Arizona and former Virginia governor Charles Robb, anxious to give the party a no-nonsense new image.

This theory is perversely misconceived. It is latest version of the wishful thinking that imagined Sen. John Glenn sweeping to victory on a powerful "centrist" tide in 1984 -- a tide that never rolled in.

The new "centrists" present themselves as unrelentingly realistic, but they are actually in the grip of a raging fantasy. They think Democrats can win the nomination and the presidency by talking like Republicans. The Republicans they sound like are the prudent, moderate Republicans of pre-Reagan days, stalwarts of the balanced budget and business-like government, who were regularly trounced in presidential elections.

Far from marking the demise of liberalism, Kennedy's withdrawal may set the stage for its revival; for his departure sets liberalism free from the long Kennedy captivity and, potentially, ushers in a period of healthy debate and redefinition. The "centrists" started that debate. Now, Kennedy's withdrawal frees the liberals to join it.

The practical problems of nomination politics also push the Democratic field leftward. Anyone who wins the party's nomination must capture many of the forces that Kennedy has just released -- thousands of activists, millions of dollars and tens of millions of voters -- which have not vanished. To acquire these assets, the candidates must be able to claim legitimacy as heirs to the tradition that, until last December, was indistinguishable from Kennedy himself.

The premises upon which the "centrists" have built their hopes do not stand up well under analysis.What they ignore are some basic realities about the Democratic party:

The party doesn't have a clearly defined center, located somewhere between a clearly defined left and a clearly defined right. It has activists, interest groups and power brokers who care passionately about certain issues. Gephardt, Babbitt and Robb -- like Glenn before them -- are chasing a middle ground that exists only in abstraction, not in the muddy arena where the nomination battle is waged.

There is no southern trend in the party, distinct from the national party. Polls show that southern Democrats over the past generation have come to share the same general beliefs as other Democrats. This may confound the centrist scenario, which assumes that a southern super-primary early in the race will aid right-leaning candidates. In fact, the real winner may not be a centrist, but Jesse Jackson, who ran strongly in the South in 1984.

A Democrat stands the greatest chance of winning in 1988 by presenting a choice, rather than a faint echo of Reagan conservatism. The centrists present themselves as an alternative for Democrats, but not as an alternative to Reagan. They view the deficit as conservatism's tragic flaw, failing to grasp that budget-balancing is the principal instrument by which Democratic programs and politicians are undone. They believe that their fearless talk about fiscal discipline will allow them to escape the fate of Mondale. By seizing upon domestic budget-cutting as the cure for the Democratic malaise, they cast their own party's political base as the enemy.

The centrists, for their part, contend that the Democratic nominating process has become a suicidal trap. Any candidate who captures the hearts, minds and checkbooks of Democratic activists and wins the nomination, they warn, is unelectable. They argue that the Democrats are suffering from a syndrome that often afflicts a party out of power -- control by fringe activists whose views are out of touch with the national electorate. Cautionary comparisons are made to the Goldwater Republicans of 1964 and the McGovern Democrats of 1972.

The centrists believe that they are shaking off the burden of the past. Yet they have shackled themselves to the immediate circumstances of the post-debacle year of 1985, when the commanding figures were Kennedy and Reagan, the references points by which the centrists positioned themselves. But the removal from the 1988 campaign scene of both of these landmarks suggests that a centrism carefully calibrated between them may have little relevance in the future.

In Reagan's Washington, liberalism is distinctly unfashionable. Within Democratic convention politics, however, many versions of liberalism still flourish; and the vast majority of those who participate in this particular forum are not subject to the capital's compromising pressures. If Walter Mondale, Gary Hart and Jesse Jackson can be broadly classified under the liberal rubric, regardless of their significant differences in 1984, then it is undeniable that liberalism dominates the national Democratic discourse.

The resources pent up by Kennedy's prospective candidacy make up much of the Democratic Party.

"Kennedy built a good financial base since 1980," said Tim Finchem, the national finance director for Jimmy Carter in 1980 and for Walter Mondale in 1984, and now a partner in National Strategies, a lobbying firm. At stake, he said, is "no less than $6 million and up to $10 million for 1987 alone, the critical period before the primaries."

Finchem explains that the Democratic Party in presidential politics "is left of center. The direct mail donor is likely to go with a candidate who is a liberal and conveys a liberal message in the mail." About 150,000 donors, inscribed on the data base of Kennedy's political action committee, are now looking for a new presidential candidate.

Just as Kennedy has freed money he has freed activists. "The number of field organizers, with two or more presidential races under their belts, who are deployable from a standing start into a state that is not their residence, is a universe of less than a thousand," said Paul Tully, the director of Kennedy's political action committee, Fund for a Democratic Majority. "Kennedy was in a position to deploy at least 250, about a third."

The next universe of activists are citizens who may have some experience with nomination politics and are ardent, well-informed participants, whose numbers may range from 25,000 to 40,000. Then there is the quadrennial flood of new participants, few of whom are known to the politicians and their staffs, attracted by the intensity of the campaign. During the 1984 Democratic contest, about 100,000 of these people emerged, according to Tully's estimate, almost all of them motivated by an excited idealism.

Kennedy's political fortune cannot be inherited by someone who claims to have no use for it. To take advantage of all these material resources -- activists, money and voters -- candidates must first define themselves within the terms of the liberal philosophy. "There are enormous numbers of people out there who are active for reasons. It's not a free energy pool," said Tully.

Recent Democratic history is partly a chronicle of Kennedy withdrawals. This history supports the theory that Kennedy's departure may animate liberalism, rather than hobble it. For when he withdrew from contention in the past, his absence opened space for others. "If Kennedy had run in 1972, it would have been a very difficult situation for me," said George McGovern. "It's true that (Edmund) Muskie was the overwhelming favorite, but Ed never had the emotional hold over people that Ted did. That opportunity was now open to someone else. I made the appeal to the grassroots, to the most committed activists, rather than going through the Democratic-labor establishment. The anti-establishment approach made it different."

Jimmy Carter made much of being an outsider, but in his anti-establishment campaign for the nomination he did not place himself outside the liberal tradition. He ran, rather, as the civil-rights candidate, the favored candidate of blacks. He also asserted a rhetorical economic populism, labeling the tax system "a disgrace to the human race." It was the failure of Carter's larger liberal promise that impelled Kennedy into a quixotic venture against a sitting president of his own party. Paradoxically, the only people who remember Carter as a liberal are conservatives.

Walter Mondale's disastrous defeat was the failure of a particular brand of liberalism. Mondale offered himself as the master middle-man of interest-group demands, after the fashion of the Minnesota Democratic Farmer-Labor Party, in compensation for his lack of charisma. He recast liberalism into a shape that is now generally assumed to have been its eternal condition. Mondale concluded his campaign with a television ad depicting himself as standing for "ideas that will not die." Liberalism, he seemed to be insisting, was geriatric.

The centrists of 1984, in the meantime, hardly demonstrated their potency. Sen. John Glenn of Ohio had won a mere 12.7 percent of the primary votes when he withdrew.

This time the self-conscious centrists believe that history will absolve them, that Kennedy's withdrawal will vindicate their strategy. The opening in the nomination process, as they see it, will be to the right, an opening guaranteed by the Super-Tuesday primary embracing the entire South.

"The political effect will be the enhancement of the chances of a moderate, middle-of-the-road candidate winning our nomination for president," said Dick Lodge, chairman of the Tennessee state Democratic Party, one of the southern primary's principal promoters.

The centrist scenario for a southern primary, however, rests on a series of unproven assumptions: that southern Democratic primary voters are more conservative than other Democrats; that the South will vote as one and not fragment; and that the more conservative Democratic voters won't be drawn into the Republican presidential primary.

Yet, according to all public opinion polls, southern Democrats are virtually indistinct from Democrats elsewhere, except New York. In 1984, the South split its votes among the candidates, giving the fewest to the only self-proclaimed centrist, John Glenn. With a heated Republican primary going on in 1988, the Democratic centrists may look less like their party's vital wing and more like the weak end of the Republican field. For a centrist adjustment of Reaganism, why not vote for George Bush, who makes a more authentic case?

Until he departed, Kennedy's political strategists privately viewed the South as a wonderful opportunity for his campaign. They considered the centrist assumptions to be precisely wrong, but didn't speak up for fear of discouraging the effort to construct a southern primary. Once in place, they believed, it would prove to be a bulwark for the frontrunner Kennedy and an insuperable barrier to challengers. "We looked upon the southern primary the way B'rer Rabbit looked upon the briar patch," said a source close to Kennedy. Kennedy's projected strategy, as it turns out, required the opposition of centrists, the legatees of Glenn.

The centrist dream of a southern primary may not materialize as they wish. As they position themselves to take advantage of the distant mega-event and enhance their Washington reputations, their standing around the country declines. Last fall, for example, Babbitt and Robb journeyed to California to make the case for their new centrism. In Los Angeles they attracted an attentive crowd of more than 250 leading Democrats, "the heart of the party here," said one of those present. Babbitt attacked the Medicare program, while Robb defended Reagan's policies in Central America. The reception was uniformly and strongly negative.

In a subsequent speech, Babbitt charged that pre-Reagan Democrats "looked out upon the vast, sprawling landscape of government spending and stood for a one-word platform: 'More.' Theirs was the Brezhnev Doctrine of domestic politics . . . " He added: "The enemy is us . . . it's our entitlements."

"There's no reason they should do better than Glenn," said one of the most prominent Democrats at the Los Angeles gathering. "They're younger, but their message is no better."

So long as Kennedy appeared to be the frontrunner, all the other candidates could position themselves as not Kennedy. The terms were easily understood: They were the insurgency, he was the orthodoxy. Understandably, when Kennedy withdrew, his potential rivals experienced vertigo. Now they must explain who they are, rather than who they aren't.

The first stage of the new generational politics, which burst out in the Hart primary victories of 1984, pitting a younger against an older candidate, concluded with Kennedy's withdrawal. Almost every possible candidate is a peer in age, with the obvious exception of New York Gov. Mario Cuomo, a volcanic personality, who reacts most sharply to the most ancient grievances and has suggested that he might be provoked into running because of perceived ethnic "slurs" -- the first Italian-American protest candidate.

The second stage of generational politics has begun. Vaguely appealing to the post-World War II generation of voters simply as a representative of their age-group will not set one candidate off from another.Now, the question is appealing to the new generation with a program both material and spiritual.

The Kennedy legacy that falls to the Democrats is an unfulfilled promise. He presented the hope of holding together the diverse party coalition through his leadership, which would transcend all particular constituency demands and reflect the national idea. By this, liberalism would be seen as a philosophy for the public good.

The task of realizing the Kennedy promise, which eluded Kennedy himself, may now be assumed by others. Liberalism awaits the candidate who can redefine it against the only fixed reference point in the 1988 campaign, the established conservative dogma and its works, the Reagan bequest.