As the Senate argued its way toward passage of the Civil Rights Act one day in 1964, one of its leading liberals, exasperated, tried one more time to reassure the opposition on a critical point.

"I will start eating the pages, one after another," Sen. Hubert H. Humphrey (D-Minn.) vowed, if they contain "any language which provides that an employer will have to hire on the basis of percentage or quota related to color."

That "bugaboo," he assured at another point, "is nonexistent."

Humphrey never ate those words. But 18 years after passage of the historic act and over a decade after the Nixon administration introduced the notion of goals and timetables for hiring blacks and others, American society is still digesting the bugaboo.

Government-enforced employment quotas based on sex, race or other group status have grown into the most explosive, the most despised and, supporters say, one of the most effective of the antidiscrimination initiatives.

They are a key tool--but not the only one--in the audacious package known loosely as "affirmative action."

In the simplest terms, the phrase "affirmative action" means doing something that takes race or sex specifically into account in order to repair the effects of injustice, something beyond simply, passively, not being unjust anymore.

It lies at the heart of the civil rights issue of the decade: access to jobs. And yet Congress has passed no law spelling out what it is.

Instead, a sort of bureaucratic virgin birth occurred in about 1969, when Nixon appointees in the Labor Department made the hiring of white women and minorities a bid specification under federal procurement law, just as bids specify a certain weight of nail or grade of marble.

In place of the brave marchers and anthem singers of the 1960s, this policy has mobilized an army of technocrats and computers against an ancient devil.

Its actual record of achievements, shrouded in righteous rhetoric from all sides, is a murky one--of small, haphazard successes, crazy ironies and a high price-tag of bitterness and confusion.

Reagan administration officials have spelled out their intentions to ease enforcement and halt the use of quotas. The only clear effect of these and other of their sorties into the minefield of civil rights, so far, is to create potentially serious political problems for themselves and rally a torpid movement back to the barricades. Still, some see the nation at a crossroads on these issues.

Today's hard times and the uneven effects of affirmative action threaten the tenuous political coalition that has fought for equal opportunity. For instance, blacks, especially men, express resentment that white women are reaping greater benefits than they, the original objects of affirmative action.

Redistributing privileges is never easy, but the civil rights movement did not invent the idea. Both federal and state governments have for decades made a major exception to the merit standard by affording massive, permanent, special preferences to military veterans at virtually every level.

How has affirmative action worked for its intended beneficiaries, those known in the parlance as "protected classes?" In general, of course, they continue to lag far behind white men in terms of earnings and employment.

But evidence shows that the policy has spurred significant advances in certain categories, for certain people, at least for the short term:

White women--the largest of all groups, including white men in total population, and over 43 percent of the workforce--are the biggest gainers. The vaunted "influx of women" into the workforce has consisted essentially of white women, who have taken a share of new opportunities out of proportion to even their large numbers.

Black women, with their "two-fer" status, also seem to have taken relatively good advantage of new openings, although some specialists believe this is changing. Black women have always had a high work rate, traditionally in lower-rung jobs. The rising generation is breaking out of the house-maid role and into better-paying fields and they even out-earn white women in three occupational categories: managerial, clerical and sales.

Black men advanced in certain categories. A study for the Urban League indicates that in the last decade black men have increased by 71 percent in higher-paying jobs (managers, professionals and skilled craft workers), or twice the rate of their overall employment growth. (Blacks and other minorities such as American Indians are 12.6 percent of the workforce, and Hispanics--most of them included in the white count--are about 5.5 percent.)

But their growth stayed focused in professional jobs at the lower end of the earnings scale, including social workers, engineering and science technicians especially, but also including lawyers, according to the Labor Department.

Protected classes have also improved markedly in police work, managerial bank jobs, underground mining and certain other work settings where the affirmative action artillery has been specifically focused.

However, while affirmative action clearly has accelerated the progress of certain young, well-educated, hard-working or otherwise outstanding members of the affected groups, it has not affected broad traditional patterns of inequality.

For instance, the proportion of black men in the labor force has actually declined since 1948, especially among teen-agers, the old and the less-educated. And all working women still earn only about 50 cents on the dollar compared with working white men.

Still, in the last half of the 1970s, white men and women were taking more than their share of new jobs. Together they made up 64 percent of labor force growth but took 74 percent of the new jobs, according to the National Commission on Employment Policy.

And more than half of these new jobs were taken by white women, it said, although they made up only one-third of the increase.

In light of the controversy over affirmative action in the Reagan era, this series will focus on the achievements and the costs of affirmative action in practice, rather than on the complex and persistent problems of discrimination itself. It will focus primarily on the employment arena (rather than, say, education), based on dozens of interviews around the country with personnel administrators, top executives, hard-hats and secretaries and others on the work-a-day frontlines.

Beyond the numbers, this survey suggests the following conclusions:

Affirmative action measures can achieve long-term progress only in combination with improvements in the pipeline of education and training; an economic climate that creates new opportunities, and the personal commitment of top managers who, after all, still retain considerable control of their workforces.

Despite political and visceral reactions against it, affirmative action has sufficient momentum to survive, at least in the flowchart and paperwork sense. It has withstood a decade of constant legal challenge, and some of its most powerful lobbyists are not civil rights activists but corporate executives, lawyers and consultants who make their living in the field.

"Anybody who says he can't find a qualified woman or minority doesn't know what he's talking about," says former Nixon official Bill Marumoto, a Japanese-American "minority" and a professional headhunter who uses a computerized data bank to help employers fill high-level job vacancies. "It's a science for us now."

A significant peripheral effect of affirmative action (along with other fair employment laws) is that it has helped expose to public scrutiny the rituals by which all personnel decisions are made. Some employers report the changes can actually be good for business.

Under pressure to achieve demonstrable change, some employers have hired unqualified white women and minorities and applied a lower standard in judging their performance on the job. Because of persistent inequities in education and culture, some say this is the only way to induct significant numbers of them into certain occupations any time soon.

But this two-track approach also sets some up for failure, reinforces prejudices, cheapens the success of the truly qualified and aggravates hostility and resentment on all sides. Supporters of affirmative action insist that forcing employers to hire the unqualified was never the policy's intent.

Affirmative action is such a prickly pear politically that Congress has given it wide berth, generally acquiescing in whatever was done by a succession of presidents, courts and bureaucrats.

Black economist Thomas Sowell, conservative nemesis of civil rights leaders, describes the process as one in which "nonelected judges, the media and the intellectual establishment" have allowed themselves to be swept along uncritically by emotional rhetoric and "a horror of being classed with bigots."

The system has been molded to a great degree by Republicans--notably the Nixon Labor Department and the Nixon Supreme Court. Arthur Fletcher, a black Nixon appointee, rather than a Great Society partisan, signed the key paper. As he puts it, "We started the counting."

A succession of presidential orders as far back as Roosevelt had called vaguely for positive action by government contractors to end discrimination.

Lyndon Johnson, in 1965, signed a crucial order creating a new bureaucracy to enforce the mandate. And he gave poetic voice to the philosophy of it: "You do not take a person who, for years, has been hobbled by chains and liberate him, bring him up to the starting line of a race and then say, 'You are free to compete with all the others,' and still justly believe that you have been completely fair . . ."

The problem, Fletcher observed recently, was that employers simply ignored the flowery but toothless presidential exhortations. Their attitude was, as a company lawyer told him one day in the mid-'60s, "When the government tells us what affirmative action is, then we'll comply."

"A year later," he said with an extravagant chuckle and a they-asked-for-it tone, "I was an assistant secretary of labor."

The idea was to emulate the cleverest manipulators of government rules--the defense contractors, he said. But in order to link job equity to procurement law, he added gleefully, "You got to count."

In their opening salvo in 1969, known as "the Philadelphia Plan," Fletcher's staff specified that certain Pennsylvania construction contractors would have to have minorities and women working a certain number of man-hours.

"We made sure there were enough hours required that one minority couldn't work 'em all," he said. "That was called 'motorcycle compliance' in those days, where they'd put one black on a motorcycle and run him from one job to another all day long.

"And we got around calling them quotas. If we'd just said 18 percent, that would be inflexible, and it would be a quota. So we set a range, of say 18 to 20 percent."

Some civil rights groups have tried to draw elaborate semantical distinctions between this notion of a "range" (known as "goals and timetables") and quotas. But others on both sides argue that this is usually a specious distinction.

To reach that range, Fletcher went on, "We called for a 'good faith effort.' And how do you show that? Simple: you count the man-hours, the cost, the number of personnel it took to make the effort. That's procurement law."

Other top Nixon labor officials have since called the idea a horrid mistake. But Fletcher, now a consultant in the field, says he would do it again. "No court has ever declared our concept unconstitutional."

The far-reaching presidential order is only one of the policy's legal legs. The other is in the Civil Rights Act, which provides vague authority.

The significant difference is that, on this second footing, the courts can order affirmative action only as a remedy after a finding of discrimination. Enforcement of the presidential order requires no such finding.

Moreover, the courts have declared it legal for employers (and unions) to adopt voluntary, private affirmative action measures--including quotas--as a kind of preventive medicine, in order to head off lawsuits.

In 1967, women were added as a protected class, followed by the physically handicapped and Vietnam veterans.

Nobody knows for sure exactly how many workers are covered directly by the federal affirmative action machinery, but Labor Department officials estimate it could be as high as 300,000 employers of perhaps 40 million American workers. Most of the workforce is affected, one way or another.

And every white woman, black, Cuban, Mexican American, Puerto Rican, American Indian, Alaskan Native, Asian or Pacific Islander, Vietnam veteran or handicapped person who works for pay in this country, indeed virtually everyone except able-bodied white men--whether underclass or uppercrust, whether or not hired specifically under the affirmative action banner--may be counted as an affirmative action statistic.

On one side, then, are around 48 million able-bodied white Anglo males in the civilian workforce not covered. On the other, roughly 36 million white women, 10.7 million blacks, 4.8 million Hispanics, 2.4 million American Indians, Asian and Pacific Islanders and Alaskans, 3 million or 4 million white handicapped males and around 7,000 white Vietnam veterans--or a total of 57 million or more are in the protected groups estimates are based on the latest federal figures, in 1979.

Even though statistical formulas are the bedrock upon which affirmative action has been built, the government has never refined the statistics in a way that measures the program's impact very clearly. The picture clicks and changes like a kaleidoscope.

In selected occupational categories or industries that have been specifically targeted by affirmative action enforcers, change is evident.

Before the advent of Fletcher and his "goals and timetables," for example, his first target--the highly paid, hidebound iron workers, electricians and other construction workers in the Philadelphia area--had a minority membership of under 1 percent. Today that proportion has risen to 12.1 percent. A dramatic increase, though far short of the government-imposed goal of 19 to 26 percent.

In 1970, there were about 24,000 black police officers. The latest figures show that has risen to 43,500, or 9 percent of the total.

"You can say there has been a sea change in the workforce," said Barry Goldstein of the NAACP Legal Defense Fund. "What you cannot say is that exactly 10,000 are there because of affirmative action, another 15,000 because of other fair employment laws and other factors" such as a suddenly perceived need, in the wake of the 1960s' urban riots, to recruit more black policemen to patrol the inner cities.

Under pressure from women's groups, the government had targeted the banking industry for special attention. And the proportions of female officials and managers zoomed.

But at the same time civil rights groups hailed these changes as proof that affirmative action works, they also branded such figures as potentially hollow. Huge increases in the category of officials and managers "may indicate that banks are changing women's job titles, not their pay, duties or responsibilites," a report by Working Women warns.

In another targeted industry, the number of female underground coal miners rose from zero in 1973 to 3,295, or 8.7 percent, by late 1980.

Women's groups offer this not only as an example of an effective program but also as a clear illustration of how affirmative action can outstrip the kind of simple "passive nondiscrimination" which requires only that the employer select fairly from among all applicants.

In a more general sampling: among physicians, women are 13.4 percent and minorities are 10.8 percent; among lawyers and judges, women are 12.8 percent and minorities are 4.2 percent, and among editors and reporters, women are 50 percent and minorities are 5.7 percent.

Still, nationwide more than 60 percent of daily newspapers employ no black reporters or editors. And at the rank of city editor and above, only 1 percent are black and other minorities.

Opposing sides disagree about exactly what the workforce statistics prove about the impact of affirmative action.

Whatever progress has been made, experts on all sides agree it is impossible to measure how much of it would have occurred as a result of economic and other pressures, without any affirmative action.

Stunningly little specific data or case studies are available. Personnel information is protected as "proprietary" and employers (including this newspaper) generally refuse to make it public unless forced by a court.

The government has demanded vast statistical recordkeeping by employers but, in 1980, the U.S. Commission for Civil Rights informed top Carter administration equal employment opportunity officials that its researchers had found the data virtually useless in assessing the effects of affirmative action in given organizations since 1973, according to interoffice correspondence.

Some records have been lost over the years as they were shuttled from one government office to another during successive reorganizations; some data tapes were of poor quality, and perhaps three-fourths of employers (mostly small ones) had simply failed to file the most basic form.

Officials responded that they hoped a reorganization of enforcement and a new computerized system at the budget office would improve the situation. But "up-to-the-minute, 100 percent accurate" data might be impossible.

Meanwhile, out in the real world, affirmative action had taken on a life of its own.