The spy cases of recent months have titillated the country, given Washington a mild case of the shudders and, above all, reinforced a popular image of the espionage craft.

It is an image of stealth and betrayal by brave or alienated men searching out secrets -- for God and country, for 30 pieces of silver, for thrills, for revenge or for love. With foggy perceptions of good and evil, they live and die in the back alleys of a dangerous world. Novelists and film-makers play on these themes. "We are the bad opposing the worst," a fictional spy laments.

In the real world of espionage in the late 20th century, however, the spy in the Burberry coat is a bit player, overshadowed by the magic of technocrats and by the huge bureaucracies that now rule the intelligence industry. "Humint," the current acronym for human intelligence collectors -- spies -- provides but a tiny portion of the torrents of information that daily pour into the U.S. intelligence community and its international counterparts.

The CIA is a case in point. At its beginning nearly four decades ago, it was literally the central intelligence agency, dominated by Ivy Leaguers addicted to the clandestine life. In its heyday in the 1960s, nearly 9,000 covert agents roamed the world in various guises. By the mid-1970s fewer than 4,000 remained. Today the agency controls no more than 10 percent to 15 percent of the "assets" -- people and money -- of a U.S. intelligence network that directly employs perhaps 200,000 people and spends upwards of $20 billion a year, 87 percent of it on "technical" means of collection.

Those numbers may understate the magnitude of the intelligence effort because of the omission of "indirect" costs. A Senate committee investigating the intelligence community in 1975-76 referred in its final report to programs involving "the activities of hundreds of thousands of individuals." The precise numbers were not published by the committee because then, as now, they were considered state secrets.

What is indisputable is that spying in 1985 is an enormous bureaucratic enterprise. Former CIA director Stansfield Turner describes it as a "colossus." It engages more manpower in the United States than the combined work forces of the departments of State, Justice, Housing and Urban Development, Education, and Energy.

It is also an enterprise of great complexity, with its many component agencies susceptible to choking on the sheer volume of material they attempt to digest. Besieged by an "information explosion," as one study commission called it even in 1970, the intelligence community has the Herculean task of sifting sense from the thousands of satellite photos, millions of conversations and billions of written words it is capable of collecting every year.

Technology compounds the problem. The 60,000 or so civilian and military employes at the National Security Agency (NSA) feed on satellites and ground-based listening posts capable of intercepting virtually any radio or telephone conversation and any cable or telex transmission occurring anywhere in the world.

These "Big Ears" are supplemented by "Big Eyes," reconnaissance aircraft and orbiting satellites able to scan with cameras and image sensors the entire surface of the Earth, reportedly identifying from the heavens objects as small as a mailbox on a country road and discriminating enough to distinguish a Guernsey cow from a Hereford. Undersea sensors plot the comings and goings of submarines. Sensors implanted on foreign soils provide seismic data, monitor nuclear tests and provide agents with clandestine communications links. Bugs and wiretaps collect a cacophony of private conversations.

Walter F. Mondale, as vice president from 1977 until 1981, was constantly impressed with the product of these technologies. "It verged," he said, "on the spectacular from time to time." Richard V. Allen, the first national security affairs adviser to President Reagan, said the NSA "never ceases to amaze me. They have incredible capabilities."

But technology creates problems, too. The billions of words and coded messages that cascade into the great computer banks at the NSA and other agencies must be processed, stored and analyzed. The same is true of the reams of photographs and infrared images descending upon the laboratories at the Washington Navy Yard, where 1,000 or more interpreters search for meanings. NSA's computers are reported to be the best in the world, but even so, some codes remain unbreakable and information can be lost or ignored in the deluge.

That was the case with the much publicized Soviet brigade in Cuba in 1980, according to Bobby R. Inman, former deputy director of the CIA and former head of the NSA. Information on the brigade had been available for years but had been overlooked. When it was "discovered" in 1979, a political issue was created that contributed to the demise of any chance for Senate ratification of the SALT II arms-control agreement with the Soviet Union.

This is not an uncommon problem. A former deputy secretary of defense, William P. Clements Jr., told a congressional committee some years ago: "In every instance I know about where there was a horrendous failure of intelligence, the information was in fact available to have averted the problem. But the analysts and the system didn't allow the raw data to surface."

Critics of the current system see failure built into the intelligence budget. The Senate investigating committee headed by the late Sen. Frank Church (D-Idaho) in the mid-1970s reported that more than 70 percent of the budget is spent on the collection of information, 19 percent on "processing" chores and 9 percent on analysis, the effort to find out what it all means.

Walter Laqueur of Georgetown University, in a new book, "A World of Secrets," asserts that the Church committee overstated the budget share going to analyses. The true figure, he says, is 1 percent.

Turner, the CIA director under President Jimmy Carter, has written that the analysts are not only "inundated" by raw data but are constantly vulnerable to breakdowns in communication and coordination in the intelligence community.

The problem, he said, involves "the CIA's espionage branch, the NSA, the defense organizations responsible for overhead reconnaissance, the CIA's electronic surveillance component, the State Department's diplomatic reporting system, the FBI's foreign intelligence branch, the Defense Intelligence Agency's attaches, the intelligence organizations of the military services and the intelligence offices of the departments of Treasury and Energy and the Drug Enforcement Agency." He likened this bureaucracy to an octopus.

As the technology of intelligence is constantly improved (a new generation spy satellite is being introduced), the demands and expectations of the consumers of intelligence in the White House, the Pentagon and other agencies grow apace. The story is told in the Senate Select Committee on Intelligence of an insatiable assistant secretary of state with responsibility for Latin American affairs. When asked by the intelligence community what he would like to know, he replied: "Everything."

The CIA at one time, according to Laqueur, directed that 83 categories of information were to be compiled on each of 120 countries. The Defense Intelligence Agency (DIA) had a shopping list of 200 intelligence topics for foreign countries, assigning to each topic and each country a priority number from 1 to 8.

During his tenure in the CIA, Turner decided that a major intelligence effort should be launched on behalf of the Commerce Department to acquire information helpful to American companies doing business abroad. To improve competitiveness, he said, "we should . . . expand our efforts to collect international economic data, by espionage where necessary."

Other agencies of government have involved the intelligence community heavily in an effort to solve their problems, such as terrorism and international drug trafficking. Still other consumer demands originate in Congress, contributing further to the overload of a system already subject to serious breakdowns and misguided analysis. Five months before the downfall of the shah of Iran, according to the Senate intelligence committee, the CIA concluded that "Iran is not in a revolutionary or even a pre-revolutionary situation." A month later the DIA asserted that "the shah is expected to remain actively in power over the next 10 years." There was no dissent from the State Department, according to Laqueur, in large part because the department's Bureau of Intelligence and Research had no Iran analyst on its staff.

In the 1950s, the intelligence community failed to foresee the invasion of South Korea and the subsequent entry of China into the Korean war. Major miscalculations led to claims of nonexistent bomber and missile gaps vis a vis the Soviet Union.

At the same time, assessments of Soviet economic growth were greatly overstated. It was predicted, for example, that by 1965 the gross national product of the USSR would reach 55 percent of the U.S. GNP, a level not achieved until the early 1980s.

The 1960s produced other defective analyses, beginning with the Bay of Pigs operation against Cuba in 1961. It was undertaken on the basis of erroneous assumptions about Cuban military capabilities and discontent among the populace with the Castro regime. During the war in Vietnam, the CIA and the DIA came to quite opposite conclusions about the effectiveness of the bombing of North Vietnam.

This was followed in the late 1960s and early 1970s by misinterpretations and underestimates of the Soviet intercontinental ballistic missile program and underestimates of perhaps 50 percent in the size of the Soviet military budget. The impact of the grain embargo against the Soviet Union was miscalculated in 1980.

The problems inherent in a such huge, fragmented and uncoordinated bureaucracy as the intelligence community were pointed up 15 years ago by a blue ribbon panel on defense in a document known as the Fitzhugh report.

It said that "like the rest of the intelligence community the CIA makes up for not collecting enough of the right kind of information on the most important targets by flooding the system with secondary matter. The information explosion has already gotten out of hand, yet the CIA and the community are developing ways to intensify it. The quantity of information is degrading the quality of finished intelligence. Production resources can make use of only a fraction of the information that is being collected."

This indictment, coupled a few years later with revelations by the Church committee of inept and scandalous CIA covert operations, might have led in time to a major upheaval in the intelligence bureaucracy: purges, reorganization and reallocation of resources. The opposite has occurred.

In the aftermath of the Church committee investigation, a bipartisan consensus emerged that what the intelligence agencies needed was more money and more people. This reasoning was based in part on findings by the Church committee that in the years between 1962 and 1975, the budget for intelligence had declined by 30 percent, when adjusted for inflation.

The intelligence bureaucracy, civilian and military, had been reduced from 153,800 people to 101,500. But these cuts, the committee said, had not crippled the intelligence effort. James R. Schlesinger, a former secretary of defense and a former CIA director, took the view at that time that intelligence budgets had been bloated in the late 1960s by excessive investments in technology that had led to "gross redundancies" in intelligence operations.

Nonetheless, sharp increases in the intelligence budget commenced in 1978 and have continued. The increases, according to the Senate intelligence committee, have averaged 5 to 6 percent "real growth" (over and above inflation). They have been "fat years" for the community, the committee said. Billions of new dollars have been made available. New technologies have been commissioned. Thousands of recruits have been brought in. Covert operations, in very bad odor just a few years ago, have been back in style since the last two years of the Carter administration.

But the problems of bigness remain and contribute to the community's vulnerability to "penetration" and theft by foreign adversaries and their agents. Roughly 4 million Americans have access to "classified" materials: 2.4 million in the Defense Department, including nearly 150,000 intelligence personnel; 1.2 million employes of defense contractors and 400,000 other government employes and researchers. A half-million of these people, according to a study by Editorial Research Reports, have access to "top secret" material and 125,000 have access to material in an even higher classification, "Sensitive Compartmented Information."

Each year, according to Laqueur, government and nongovernment offices produce 80 billion documents (in addition to the 300 billion on file). Twenty million federal documents each year are "classified," 350,000 as "top secret." So the amount of material potentially vulnerable to espionage and theft is very large. In a perverse way, however, the sheer mass of the material is protective. Low-level "moles" can only nibble at the fringes.

On the other hand, more mass means more management problems. Despite the recommendations of study commissions, congressional committees and even presidents, the vast American intelligence community still has no "czar," no director or directorate to establish priorities, eliminate redundancies, enforce quality standards and allocate money and people to the areas of greatest need.

In theory, the CIA director presides over the community. But in practice neither the current director, William J. Casey, nor his predecessors have filled that role. One former director, Richard M. Helms, observed that he was not in a position to act as a czar when the Pentagon controlled 85 percent of the intelligence budget.

Lack of coordination and centralized control likewise afflict the government's counterintelligence efforts. Each agency goes its own way, although the FBI's deputy assistant director, Phillip D. Parker, insists that it cooperates closely with the CIA and other counterintelligence units.

There is some dissatisfaction with the current state of affairs in the intelligence community. Sen. David F. Durenberger (R-Minn.), chairman of the intelligence committee, is impressed like many of his colleagues with the technology of spying and assumes that one day the intelligence community may "achieve the ultimate," such as imagery techniques to penetrate clouds, darkness and even concrete.

But he is concerned about converting the incoming masses of data into useful intelligence. Intelligence failures in Iran, Lebanon and Central America, he said, have come about because the wrong information was sought.

"The Philippines," Durenberger said, "is now similar to Iran. We know that Marcos is in trouble, and we know why. But what happens when and if he goes down? That's the kind of intelligence we need, and that kind of intelligence is our principal weakness."

Other senators are critical of the counterintelligence effort. They complain about Casey's personality. Mondale has been critical of the DIA -- "a dumping ground," as he described it -- and for years has urged that more academic scholars be utilized as consultants and advisers. (Former national security affairs adviser Allen, on the other hand, developed a "high respect" for the DIA). There is, finally, a general concern about upgrading quality and getting better analysis.

But the remarkable political fact is that the intelligence agencies, despite past failures, despite the scandals of the 1960s and 1970s, despite obvious inefficiencies, have a broad and deep base of support today in both Congress and the administration. They may not, Durenberger warns, continue to receive the lavish budgetary treatment of the past seven years. But their Bay of Pigs is far, far behind them