In 1929, after learning about a secret code-breaking intelligence operation jointly run by the Army and State Department, Secretary of State Henry Stimson promptly de-funded it, declaring, “Gentlemen do not read each other’s mail.”

Actually, they do, and they always have – even the gentlemen who founded the United States. In fact, in 1775, the Second Continental Congress created the Committee of Correspondence, which funded propaganda, performed covert operations, developed codes and – gasp – intercepted mail. While we’re on the subject of Revolutionary War history, even Gen George Washington sometimes got bad intel – or got good intel too late: Gen. Washington sent a task force to Bermuda with orders to seize gunpowder stored at the British arsenal. But when his men arrived, the ammunition was gone. It had already been secretly acquired by agents of the Continental Congress. Ever since, the high-stakes nature of intelligence operations and the self-critical nature of our representative system of government have conspired to expose intelligence failures and to impugn this “ungentlemanly” line of work.

It’s important to keep this in mind – and keep a proper perspective on the necessary, albeit unpleasant and imperfect work of the Intelligence Community – as our government strains to fend off a withering onslaught of Russian hacking and weaponized leaks; Chinese cyber-espionage and traditional espionage; fifth-column anarchists; and asymmetrical attacks emanating from rogue states like Iran and North Korea as well as non-state actors like ISIS and al Qaeda.

Surprises

Of course, the Continental Congress didn’t invent the imperfect science of intelligence-gathering. There’s been a need for intelligence – and spies to gather it – for as long as there have been tribes, nations and kingdoms.

Some 1,500 years before the birth of Christ, the Book of Numbers describes how Moses ordered a select group of men into the land of Canaan to gather intelligence: “See what the land is like and whether the people who live there are strong or weak, few or many,” he said. “What kind of land do they live in? Is it good or bad? What kind of towns do they live in? Are they unwalled or fortified? How is the soil? Is it fertile or poor?”

They carried out their mission into hostile territory and returned to report that the land was teeming with fertile crops and healthy livestock. But they also cautioned, “The people who live there are powerful, and the cities are fortified and very large.”

Upon receiving the intel report, Moses and his advisors – the policymakers of their day – faced a dilemma. Some argued for an operation to “go up and take possession of the land, for we can certainly do it.” Others countered, “We can’t attack those people; they are stronger than we are.” This latter group’s report frightened and divided the people.

The more things change, the more they stay the same. Then, as now, intelligence is just data and information. It’s left to policymakers to make sense of it and ultimately act on it – or not act on it. Then, as now, employing intelligence is a mix of guesswork, gut instinct, analysis and calculation.

In other words, the Intelligence Community doesn’t cause policymakers to choose a particular course of action. Policymakers choose the course. And while it’s fair to hold the IC accountable when the intel is wrong, it’s not fair to scapegoat the IC when the course of action turns out to be wrong.

Consider America’s entry into both world wars, which is often blamed on intelligence failures. Much of the blame is unfair.

Intelligence services actually played a crucial role in uncovering the Zimmerman Telegram, which detailed German plans to leverage Mexico as a proxy for war against the United States. Of course, it was British intelligence that cracked the German code, underscoring how inadequate U.S. intelligence was – an enormous policymaking failure.

Playing catchup, the Army created Military Intelligence Section 8, which scored perhaps its most significant successes after the war. Cryptologists from MI-8 and the State Department broke the code Japanese negotiators used to cable back and forth to Tokyo during the Washington Naval Conference in 1921-22, enabling U.S. diplomats to outmaneuver their counterparts. Before the decade was out, however, the codebreaking operation was shut down, thanks to Stimson.

Predictably, that led to limited code-breaking capabilities at the onset of World War II. Even so, America’s hamstrung intelligence assets sounded the alarm over Japan in time to avert disaster. In January 1941, almost a full year before Pearl Harbor, the Secretary of the Navy warned that hostilities on the part of Imperial Japan “would be initiated by a surprise attack upon the fleet or the naval base at Pearl Harbor.” Two weeks before the attacks, the Navy Department warned of “a surprise aggressive movement in any direction by the Japanese.”

One postwar inquiry concluded that policymakers in Washington and Hawaii “were fully conscious of the danger from air attack…and they were adequately informed of the imminence of war.” Congress found that the Navy and War departments had “failed to give careful and thoughtful consideration to the intercepted messages from Tokyo to Honolulu.” In other words, U.S. intelligence was doing its job. But the policymakers weren’t listening.

This foreshadows the leadup to 9/11. IC memos throughout the late 1990s and early 2000s warned of Osama bin Laden’s determination to attack the U.S. homeland. A 1999 report commissioned by the National Intelligence Council predicted that “al-Qaeda's Martyrdom Battalion could crash-land an aircraft…into the Pentagon, the headquarters of the Central Intelligence Agency or the White House.” But the policymakers weren’t listening – and thus weren’t willing to act.

Indeed, in the late 1990s, intelligence assets tracked bin Laden and had him in their sights on several occasions, but disagreement among policymakers about whether to kill or capture him allowed him to escape. In addition, before 9/11, the FBI and CIA were barred from sharing information about terrorist threats. And on 9/11, just 1 percent of the CIA’s 18,000 employees were tasked to counter-terrorism.

These were policymaking failures on the part of Congress and the Executive – not intelligence failures.

Reminders

In 500 BC, Chinese warrior Sun Tzu concluded, “What enables the wise sovereign and the good general to strike and conquer, and achieve things beyond the reach of ordinary men, is foreknowledge.” That’s another word for intelligence. “If you know the enemy and know yourself,” Sun Tzu went on, “you need not fear the result of a hundred battles. If you know yourself but not the enemy, for every victory gained you will also suffer a defeat. If you know neither the enemy nor yourself, you will succumb in every battle.”

When the intelligence is bad – when information about the enemy is wrong – the result is fog rather than foreknowledge, as we learned in Iraq.

Given that intelligence agencies underestimated Saddam Hussein’s nuclear program in 1991, it’s no surprise that they, in a sense, overcorrected and extrapolated the worst in 2002-03. And given that Saddam apparently pretended to maintain his WMD arsenal as an internal deterrent, it seems unfair for policymakers to blame the IC for reporting what its sources inside Iraq were saying.

However, those sources were a problem. The IC relied too heavily on defectors (who were motivated not by the pursuit of knowledge and truth, but by personal and tribal grievances) and on the assessments of foreign intelligence agencies (which were based on assumptions from other foreign intelligence agencies). This created a cycle that fed on itself and led to what Sen. Pat Roberts called “a global intelligence failure.”

As a consequence, even as the American military succeeded in ousting a persistent threat to U.S. interests and in opening the way for the Iraqi people to build a representative government, the United States sustained a number of self-inflicted wounds – diplomatic, economic, geopolitical. And the U.S. military suffered through its bloodiest conflict since Vietnam.

Iraq serves as a reminder that uncertainty is a given in the world of intelligence, that the work of intelligence presupposes some amount of interpretation, that when humans interpret the motives and behavior of other humans we are bound to get it wrong sometimes. Pearl Harbor and 9/11 serve as reminders that intelligence is worthless if not acted upon. The Zimmerman Telegram serves as a reminder of the absolute need for a robust intelligence apparatus. And all of these underscore the importance of prudent policymaking to avoid as many of the pitfalls as possible – and statesmen-like politics to pull the nation out of those pitfalls it fails to avoid.

Safety

From Gen. Washington’s day to our own, Americans have engaged in intelligence operations, cryptography, reconnaissance and espionage to defend our freedom and way of life, to foil the plans of hostile regimes, and to maintain some semblance of global order. It may be considered dirty work, but it’s necessary.

Ian Fleming, who is best known for creating the James Bond character, also served as a high-ranking official in British Naval Intelligence. After Moscow shot down an American U-2 spy plane in 1962, he lamented how so many in the West flatly misunderstood espionage; explained that “Americans can lie more safely in their beds today, and Englishmen too, because of the intelligence brought back by planes of the U-2 class”; chided those who blamed America and the West for conducting such operations; and heaped scorn on “men who think espionage is a dirty word. It isn’t. It has got to be done well—that’s all.”