The dignitaries in the Treaty Room strategy session are conducting themselves with all the direction of Shriners in convention.

The President, the Vice President, the First Lady, and their highest echelon of advisers are celebrating the Illinois primary blowout of the night before -- it will be Carter 163 delegates, Sen. Edward M. Kennedy just 16 -- with backslaps and belly laughs.

Jimmy Carter is cheerily demanding to know why his crack team allowed Kennedy to win even those few delegates. Robert S. Strauss is talking about how they are going to get Kennedy out of the race now once and for all. Patrick Caddell is passing out sheets of paper with the latest poll results for next week's primary state.

The room falls silent as those assembled read the first sheet, and then the second. They are looking at an unforseen development that will plague the President for the rest of the campaign: the birth of the anti-Carter protest vote.

All of the public polls have been showing Carter headed for a huge win in New York, the next primary, that would surely mean the end of the Kennedy candidacy. Caddell's figures also show Carter 51 percent, Kennedy just 32.

But Caddell's technique for probing beyond the initial responses with a unique "second vote" question which has produced the most reliable of all political polling figures this year -- shows that Kennedy has actually taken the lead: 43 percent to Carter's 39.

There follows a third figure that is the most amazing of all.

Kennedy holds this lead despite the fact that he is viewed "unfavorably," as opposed to "favorably," by a majority of people. And Carter is trailing even though he is viewed "favorably" by a substantial majority of the people. (The figure is 60 percent "favorable," 38 percent "unfavorable.")

Everyone's eyes just popped out," recalls one adviser who was at the meeting. Caddell explains to the president and his advisers that apparently people are now suddenly willing to vote for Kennedy even though they do not care for him. He says that now that people believe that Carter will be the party's nominee, they are focusing on Carter almost exclusively, and they are deciding they do not like what they see.

The "protest vote" proves to be at the heart of what is happening throughout the rest of the Democratic primary campaign of 1980.

For a while it will confound the Carter strategists. They will not find a way of coping with it for several weeks, until they are well into the Pennsylvania primary. Finally, they will neutralize it significantly with a new batch of negative ads created by media adviser Gerald Rafshoon that will focus attention -- and doubts -- on Kennedy once again. (In the process they will also focus controversy upon Rafshoon.)

But it will come back to afflict Carter once more, at the end of the primary campaign. And in fact, it is probably a harbinger of things to come for Carter in a general election contest against Republican Ronald Reagan.

Understanding the "protest vote" means understanding much about what is going on in this country this year; it means understanding ourselves and the way we work our politics.

Just how the President and his campaign came to view this anti-Carter vote is seen in a study of the internal statistics of Caddell's campaign surveys, some of which have been made available to The Washington Post. They basically show numerically what journalists had been gleaning and writing during their own lengthy interviews during the primaries: that many people just did not like Carter's performance as president. Even though they did not like Kennedy, the more they thought about Carter, the more willing they were to vote for his opponent. Kennedy was most successful when he became brand X.

The Carter campaign of 1980 could not be like the Carter campaign of 1976. This year, Carter had to make his opponent the issue, which is far different from 1976, when he won the nomination mainly by convincing people to like him. He made them view him as trustworthy and decent and so on; he made them vote for him, not against his opponents.

A glimpse at how the Carter campaign came to cope with this "protest vote" with new negative ads shows how polling and advertising mix in the politics of 1980, and how campaign spending decisions are made. It also offers a marked contrast between the Carter and Kennedy campaign organizations.

For by New York, the Kennedy campaign had mostly run out of money, which greatly limited its polling. And it had been running its advertising at the outset through a consortium of media experts. They produced media results about as successful as the storied committee that tried to assemble a horse -- and produced a camel.

New Yorkers had more than just Carter's campaign successes to give them cause to protest his leadership. There was that U.S. vote in the United Nations Security Council for an anti-Israel resolution -- quickly followed by that U.S. renunciation of its own vote.

More than one-quarter of New York's Democratic primary voters are Jewish, and Caddell's figures showed that in a 10-day period after the United Nations vote, New York's Jewish voters had shifted from favoring Carter by a 9-point to favoring a Kennedy by a 17-point margin.

This contributed greatly to Carter's problem in New York, Caddell was saying a week before the vote there, but it could not account for all of it. "It wasn't just the Jews," Caddell says. "It was upstate. It was across-the-board. It was something else."

Caddell was more concerned about what he saw as the larger problem for Carter: that Kennedy could have the lead despite the fact that a majority of the people gave him an "unfavorable" rating. Additional speeches by Carter surrogates were scheduled, and new Carter media spots were aired.

Still, publicly, all signs were pointing to a Carter landslide. By the Friday preceding the Tuesday, March 25 primary, the Louis Harris poll in the New York Daily News was showing Carter with a huge 27-point lead.

Caddell, using his own "second vote" adjustment technique, was coming up with something else. He and his associate, John Gorman, had devised a two-step polling procedure which had proven very accurate in this tumultuous year.

First the poll-takers would ask who he or she would vote for. Then they would ask a short series of questions designed to get the persons being interviewed to think as intensely about the choice as he would on election day when walking into the voting booth. Then, they would ask the person again, in a "second vote," to choose the candidate he or she would vote for. Invariably, much of Carter's support dropped away in that second vote.

On Friday, Caddell's date were showing that Carter had climbed back to a very narrow lead. But that same day, Secretary of State Cyrus R. Vance was testifying in Congress that the controversial U.N. resolution vote really had coincided with Carter administration policies after all, an admission that made for stunning front page news in New York on the weekend before the primary election.

Carter's strategists were apoplectic.

Vance's testimony was politically undiplomatic -- and potentially disastrous. But also damaging, Caddell felt, was that Friday's Harris poll, which the New York Daily News headlined, "Carter's the One -- in a Pollslide." It could fuel the anti-Carter vote. "

Sunday, 2 p.m. Caddell is in the kitchen of his home on R Street NW when his associate Gorman calls with the latest results.

"I have some good news and some bad news," Gorman says. "The good news is that we have now seen the most dramatic shift ever in our regression analysis [factors that people view as important when they made their voting decisions].

"The bad news is that . . . Ted Kennedy has disappeared as a factor in this election. It's Carter versus Carter. And Carter is going to get murdered."

The protest vote has hit with full force. Publicly, the Harris poll is still showing a 20-point Carter lead, but the bottom has fallen out of the president's prospects for a victory in New York that would have meant the end of Kennedy's challenge. On Friday, Caddell had shown Carter with a 4-point lead. On Saturday it was Kennedy who had a 4-point lead. Now, on Sunday, Kennedy's lead has swelled to nine points.

Caddell calls the President at Camp David. He goes over the figures with Carter. He explains how his associates have done this analysis of internal questions to determine what factors are leading people to vote they way they say they will.

Usually, trustworthiness of a candidate is the number one factor -- and that has always been a strong plus for Carter. Now, Caddell explains, trustworthiness has "fallen off the charts"; it is not even among the top 12 factors that New Yorkers consider important in the making of their presidential decision.

What comes first is a "protest question" that Caddell has added to his survey. It is a statement read to people by his interviewers: "Carter can't handle the presidency; we would be better off trying a new president." Suddenly, a majority of the New Yorkers are agreeing with this statement; two weeks earlier, a majority were disagreeing with it.

The President listens to all the bad news without comment. When Caddell finishes his hurried presentation there is still silence.

"We'll try to do everything we can," Caddell says.

"Thank you," says the President.

New York and Connecticut turn out to be dual debacles for the President; most pollsters and even the Kennedy camp are stunned by the outcome, which is Kennedy by 18 points in New York and by 5 points in Connecticut.

No longer are the President's advisers thinking about forcing Kennedy out of the race early. Now they are looking thankfully ahead to a respite in the Wisconsin primary. They call it their "safety net."

For the moment, there is no urgency about figuring out how to solve the protest vote problem -- for, after New York and Connecticut, people are no longer looking at Carter as a sure winner. Several weeks down the trail, Pennsylvania looms as a potential New York. But for now, Caddell's polls are showing Carter with a comfortable lead in Wisconsin.

This is somewhat of a surprise to Caddell.

Back in the planning stages, he had advocated skipping Wisconsin, contending that the state was simply too liberal, that it would be a Kennedy state. Vice President Mondale had argued that Caddell was wrong, that the state was solid, midwestern and a good Carter state.

Caddell took a poll and came back with a report: "Mondale was absolutely right."

With New York behind them and Pennsylvania ahead, the Carter advisers needed a decisive win in Wisconsin on April 1, as one adviser noted, "to show that we are not in a complete state of collapse."

Public opinion on Carter's handling of the crisis in Iran had begun to sour and several days before the primary Caddell's surveys showed that Carter's lead was beginning to narrow. "The Iranian thing was clearly a problem," Caddell says. "The president's ratings on his handling of it were declining."

Privately, there were signs that an agreement was near on transferring control of the American hostages out of the hands of the militants. But publicly, things took a turn for the worse when: Iran went public with what it said was a message from Carter that was apologetic in tone; and next, press secretary Jody Powell issued what seemed to be a denial that any message had been sent to Iran's leaders; then Swiss officials said they had indeed delivered some type of message from Carter, Powell tried to clarify that he had been saying was that no apologetic message had been sent, but the effort seemed mostly to be making the worst of a bad thing.

The Carter officials were desperate to turn things around. The President personally met with a group of newspaper representatives, and later talked with television network anchor-persons as well, to let it be known that agreement seemed near.

And on election day in Wisconsin, the President summoned reporters to his office just after 7 a.m. -- just in time for the network morning news shows -- to proclaim that the latest statement from Iran that morning was a "positive step."

The resulting good news did the political trick, short term.

"When it was made clear over the weekend that there was going to be some progress, that bumped the race up for us," Caddell says. "It went from a lead of 15 to 18 points to a win of almost 30 points.

Wisconsin was the only state where we ever got the undecided to go for us in the end."

But there are some members of Carter's highest council who feel that there were also long-term setbacks to that last-minute attempt at election day dramatics.

"The president decided that he wanted to do it -- and now Jody is kicking himself for going along with it instead of thinking about it and stopping it," says one senior adviser. "We paid a price for that -- a real price, especially as far as our relations with reporters are concerned. That 7 a.m. thing crossed the line. Carter no longer seemed decent and honorable, but manipulative. Ironically, it probably had no impact -- usually it takes a 24-hour gestation period for big events to have an impact on the public."

The President's strategists are locked in a major strategy dispute.

It is Saturday, and with the overkill of Wisconsin behind them, the Carter campaign's board of directors has just heard Caddell's latest analysis, which is that the April 22 Pennsylvania primary is looking "very New Yorky."

All of the public polls are showing a huge Carter lead in Pennsylvania, and Caddell's "first response" figures are also showing that: Carter 53 percent Kennedy 30. But his "second vote" response in those same interviews shows that there already has been a dramatic reversal, with Kennedy leading Carter, 43 to 40 percent.

Once again, "trustworthiness" is not being listed as a major factor in people's decisions. And people are agreeing in large numbers with the "protest" statement of Caddell's questionaire that it it time to try a new president.

Tim Kraft argues that the Carter campaign should write off Pennsylvania. His argument -- at this meeting attended by Jordan, Rafshoon, Caddell, and campaign officials Tim Smith and Tim Finchem -- is that the campaign will soon be hard pressed for money and hard pressed to stay within the legal spending limits of the primary campaign.

Pennsylvania is a lost cause; don't waste money on it, Kraft says. Rafshoon vehemently disagrees.

The advisers have already concluded that they will mostly concentrate their field operations money on the caucus states and the media money on the primary states. Rafshoon argues that with a good media plan, Pennsylvania can be saved. There are too many delegates at stake to surrender them to Kennedy at this late stage.

Jordan quietly his own counsel. He will spend a couple of days asking if Carter aides think it would be smart to write off Pennsylvania; he seems to be siding with Kraft. But his decision will be to spend for the media plan and trust Rafshoon to deliver.

Rafshoon takes a camera crew to Pittsburgh and begins interviewing people on the subject of what they think of Kennedy. The idea, he concedes, is far from original.

"I remembered what Ford was able to do to us at the end of the 1976 campaign with those man-in-the-street ads," Rafshoon says, recalling the devasting ads that featured Georgians saying why they were going to vote against Carter.

It is early morning in Pittsburgh. As Rafshoon is working with his camera crew, he notices a middle-aged woman loitering to the side, studying him suspiciously. Finally she approaches, a sparkle in her eye.

"I know you," she says. "You're . . . you're. . . ."

"Dan Rather!" says the short, mop-haired Rafshoon, who is to Dan Rather what Woody Allen is to Sir Laurence Olivier.

"I knew it?" the woman beams, shaking his head before walking away.

The ads Rafshoon produces are negative -- but then again, so is Kennedy's daily stump rhetoric,

Rafshoon's ads show different people saying that Kennedy is "too liberal . . . a big spender . . . wrong on welfare . . ." The closest thing to a personal attack is one commercial which deals only subtlely with character:

MAN: I don't think Kennedy's qualified to be president.

WOMAN: I don't think he has any credibility.

WOMAN: I don't believe him.

Woman: I don't think he's the man for the job.

MAN: I don't trust him.

WOMAN: You're taking a chance with Kennedy.

WOMAN: I'm going to vote for Carter because I think he's the best qualified.

MAN: Carter's got his hands full with Kennedy. He's got his hands full with the country and the economic situation. He's got his hands full with Iran and Afghanistan. He's not doing a bad job.

MAN: Between Kennedy and Carter, I would defintely go with Carter myself. I trust him.

"The ads spoke to the perceptions of Kennedy that people already had in their minds," Rafshoon says, taking care to always refer to them as the man-in the street ads. "We needed to remind people, that's all."

The ads begin running during the last week of the Pennsylvania campaign. Kennedy has been steadily lengthening his lead in the Caddell polls; Friday's "second vote" figures give him a 9-point lead over Carter.

But significantly, they also show a sharp increase in the number of people who now believe that Kennedy can win the nomination. In the next couple of days, voters begin expressing concerns about Kennedy's character.

Like those unpleasant television commercials where the Rolaids neutralize stomach acid by the pitcherful the Carter people believe they are watching Rafshoon's ads neutralizing the anti-Carter vote right before their eyes.

Caddell's polls are showing that people are once again listing "trustworthiness" as the main factor in their election day decision-making -- just as it used to be in Carter's good old days when he was winning big. A majority of the people are now agreeing with the statement that Kennedy is a big spender. Caddell's last poll, finished the Sunday before Tuesday's election, shows Carter and Kennedy dead even, at 40 percent.

That is just the way it is on Tuesday night, when Kennedy wins by a whisker. The reality of it is that Carter -- whose aides almost had him abandoning the state -- comes away with half of Pennsylvania's 185 delegates.

And added to the landslide victory in the Missouri caucuses the same day, Carter comes away from a day that could have been a disaster by making up all of the ground he lost in that New York-Connecticut debacle.

"Jerry's negative ads made it possible," Caddell will later say. "They made Kennedy the issue again."

The month of May will give Carter a delegate cushion that will provide crucial comfort as he goes through what has become his traditional, quadrennial June 3 tail-first crash landing. In 1980, as in 1976, he wins in Ohio but loses in New Jersey and California.

"Luckily," Caddell says, "the month of May made June 3 irrelevant."