EIGHT MONTHS after Britain and Argentina ended their 1982 Falklands war, the U.S. Navy published a military analysis of tactics, equipment and lessons to be learned from the war in the south Atlantic.

Curiously, one year after the American operation in Grenada, the Pentagon has yet to release a comparable critique.

This silence is not because our military knows more about other countries' wars than about its own; numerous analyses of the Grenada invasion have circulated within the Pentagon. Rather, the Defense Department has publicly declared Grenada an unqualified success, and it does not want that judgment sullied by tales of Army radios that could not talk to Air Force radios, or of U.S. soldiers mistakenly killed and injured by U.S. Navy bombers. So the Grenada reports, unlike the Falklands analysis, have been stamped secret -- even though an airing of the problems might lead to improvements that could, in turn, prove crucial if the United States runs into something more than 700 troops and armed construction workers in its next campaign.

The decision to withhold the Grenada reports reflects a growing tendency within the Pentagon to release only information that will make the Pentagon look good. Embarrassing information is kept secret while self-serving facts are aggressively put forward, blurring the line between public information and public relations. Secrecy allows the Pentagon to dodge accountability for everything from cost overruns to war in Central America.

Obviously, the desire to show the best possible face is nothing new, and many secrets must be protected in the interest of national security. Nor is there anything new in a beat reporter's frustration with the age- old, and in many ways healthy, war between bureaucracy and press.

But the current attempts at information management by top civilian officials seem to go beyond traditional practices, stemming from a view that perception of military strength is as important as strength itself. Purchasing weapons that they have little expectation of using, for example, many Pentagon officials believe that the performance of the weapons they buy is less important than how the "other side" thinks they will perform. In such a situation, almost any obfuscation can be justified in the interest of national security. To admit failure is to create a perception of weakness -- which is the same as weakness itself.

The danger with this policy is that failures are hidden from the American public as well as from any undeclared enemy. And the absence of public debate and questioning can encourage the Pentagon to believe its own propaganda -- which can allow real weaknesses, as opposed to perceived ones, to persist.

In recent years, the Pentagon has been hiding more and more of its budget items. Only a few officials in the administration and in Congress can say how many billions of dollars are now tucked away in "black programs," classified projects that don't appear on any budget line. Several officials acknowledge that the amounts are growing fast.

Richard Barnard recently wrote in the trade publication, Defense Week, that the amount of hidden money in the budget has grown more than sevenfold in the past five years, from $742 million in 1980 to at least $5.7 billion in 1985. Pentagon officials called his estimate conservative.

The best-known "black" program is the Stealth bomber, a "secret" weapon first discussed publicly by Defense Secretary Harold Brown in 1980. Everyone knows that the Air Force is developing the Stealth, an advanced-technology bomber designed to be nearly invisible to enemy radars.

Only a few officials, however, know how much has been spent or will be spent on the program. The Wall Street Journal recently estimated, based on interviews with industry and government officials, that the program will be worth as much as $45 billion to Northrop Corp. and other firms, but no one in the public knows for sure.

One result of the secrecy, ostensibly, is to complicate military planning for the Soviets, who do not know how many Stealth planes we will build. But a more evident result is that if Stealth ends up costing more than Northrop has promised, the public will never know. There can be no embarrassing cost overruns without public cost estimates.

"Black" accounting similarly allows the Pentagon to dodge questions about military involvement in Central America. As with the Stealth bomber, everyone knows the United States is supporting an army trying to overthrow the government of Nicaragua.

Because that effort is officially secret, however, the Pentagon can comment when it chooses and then act as though no war exists when silence suits the Pentagon's purpose. The Pentagon has built airstrips now used by the CIA-backed rebels; it has provided them with military equipment; it almost certainly gathers intelligence that they use. Nonetheless, Michael I. Burch, chief Pentagon spokesman, won't answer questions about those activities.

When The Washington Post reported that the Air Force had donated two small planes through the CIA to the contras -- the forces opposing the Sandinista government in Nicaragua -- Burch would say only that "another government agency" had taken possession of the planes. Beyond that, the Pentagon -- the government -- did not feel obliged to account.

Each year, the defense secretary's annual report to Congress contains less and less useful information about the U.S. military, according to those who have read or worked on the reports for more than a decade. At the same time, successive editions of Soviet Military Power, first published by the Pentagon in 1981, contain more and more useful data.

Millions of documents are classified secret, as always, but now the Pentagon is preparing new rules to keep even unclassified documents secret in many cases.

The Catch-22 of secret documents, of course, is that only the classifier knows whether publication would really threaten national security. Once in a while, however, a declassified document can give us some indication.

Recently, the Pentagon refused to release its inspector general's report on the Army's troubled Divad, a new antiaircraft gun system that costs about $6 million apiece and that hasn't worked very well in early tests. The report was classified, officials said; even to explain why would violate the law.

A Senate subcommittee got hold of the report and released it, however, after deleting a few paragraphs that discussed the gun's range and other information of potential use to adversaries. The rest of the report, it turned out, was merely embarrassing: it described how the Army withheld negative test results from top Pentagon officials whose job was to decide whether to purchase the Divad system.

Can there be any justification for concealing information about Grenada snafus, or cost overruns or improprieties in the weapons testing process? Some officials think so.

To understand why, one must understand the importance of perception in defense thinking. Nicholas Lemann pointed out in a recent Atlantic Monthly article that this administration views increased military spending as a symbol of national resolve likely to improve security, regardless of how the money is spent.

That notion is not entirely new. It was the Carter administration, after all, that made 3 percent real growth for NATO military spending a symbol of western steadfastness. In both cases what came to be emphasized was symbolism more than substance. By this line of reasoning, anything that detracts from the administration's ability to rally Congress and the public to support the full request detracts from national security.

Similar arguments have become virtually the sole support for the beleaguered MX missile. Few now make a military case for the MX, but many argue that the MX must be built to show the Soviet Union that we have the national will to see a new missile system through to deployment.

A House subcommittee recently issued a report charging that, despite increased spending on new weapons, the U.S. military lacked the ammunition, spare parts and cargo planes needed to sustain a war. Defense Secretary Caspar W. Weinberger labeled the report "dangerous" as well as wrong. Why? Anything that gives the impression of weakness is as harmful as, or more harmful than, weakness itself, according to this view.

The Pentagon looks on many of its critics as people "who were never for strong defense anyway," as Weinberger said recently, and who seize on spare parts procurement or weapons testing problems as the most convenient pretext to attack defense.

In that context, it becomes easier to understand the ongoing Pentagon campaign that, according to internal memos, is designed to "enhance media and public understanding of and support for" the administration's management of the military. The memos call for an orchestrated series of speeches, press conferences and other media events.

"It can be expected that attacks on Pentagon management procedures will continue . . . ." one memo said. "These attacks should be contested in the press by an active letter-to-the-editor program."

Pentagon spokesman Burch explained the campaign -- after Common Cause got hold of and released the memos -- by saying that "we have a responsibility to the public to keep them informed of the good and the bad." But there was nothing about the bad in these memos; this was sheer public relations, not public information.

Not long ago, the independently-published weekly Navy Times reported that one U.S. soldier died and 15 were wounded when a Marine-Navy spotter team called in the wrong coordinates for a Navy light attack plane's strafing run during the Grenada operation.

A U.S. Army unit had requested air support against a Cuban position about 100 yards away, Navy officials confirm. Instead, the plane attacked the ramshackle hut where the U.S. soldiers had taken refuge, tearing the hut in half with 20mm cannon fire and severely wounding 16. One 23-year-old sergeant died last June of injuries received in that incident, which the Pentagon did not acknowledge at the time.

Does one tragic incident of "friendly fire" suggest that the Grenada campaign was poorly executed? Of course not.

Does it offer lessons to decrease the chances of similar incidents in the future? Quite possibly.

Has our military absorbed those lessons? That we may never learn. Adm. Wesley L. McDonald, who had overall command of the Grenada operation, told Congress simply that the campaign was "a complete success."

That, the Pentagon believes, is all we need to know.