Beyond weakening the administration, the seemingly incessant wave of Trump scandals seems to reinforce liberals’ narrative of the previous president. As The New Republic remarked after the resignation of Michael Flynn, “Obama went eight years without a major White House scandal. Trump lasted three weeks.” Or as Obama himself boasted in December, “we’re probably the first administration in modern history that hasn’t had a major scandal in the White House.” To the horror of conservatives, who can cite a litany of official misdeeds during the Obama years, the apparent integrity of that era will feature prominently as historians evaluate that presidency. (Spoiler: as those historians are overwhelmingly liberal, they will rate it very highly indeed.)

In a sense, though, both sides are correct. The Obama administration did a great many bad things, but it suffered very few scandals. That paradox raises critical issues about how we report and record political events and how we define a word as apparently simple as “scandal.”

Very little effort is needed to compile a daunting list of horror stories surrounding the Obama administration, including the Justice Department’s disastrous Fast and Furious weapons scheme, the IRS’s targeting of political opponents, and a stunningly lax attitude to security evidenced by Hillary Clinton’s email server and the hacking of millions of files from the Office of Personnel Management. Even on the available evidence, the IRS affair had most of the major elements of something like Watergate, and a detailed investigation might well have turned up a chain of evidence leading to the White House.

But there was no detailed investigation, and that is the central point. Without investigation, the amount of embarrassing material that emerged was limited, and most mainstream media outlets had no interest in publicizing the affair. Concern was strictly limited to partisan conservative outlets, so official malfeasance did not turn into a general public scandal.

Misdeeds themselves, however, are not the sole basis for official statistics or public concern. To understand this, look for instance at the recently publicized issue of sexual assaults on college campuses. The actual behaviors involved have been prevalent for many decades, and have probably declined in recent years as a consequence of changing gender attitudes. In public perception, though, assaults are running at epidemic levels. That change is a consequence of strict new laws, reinforced by new mechanisms for investigation and enforcement. A new legal and bureaucratic environment has caused a massive upsurge of reported criminality, which uninformed people might take as an escalation of the behavior itself.

Political scandal is rather like that. To acknowledge that an administration or a party suffers a scandal says nothing whatever about the actual degree of wrongdoing that has occurred. Rather, it is a matter of perception, which is based on several distinct components, including a body of evidence but also the reactions of the media and the public. As long ago as 1930, Walter Lippman drew the essential distinction between the fact of political wrongdoing and its public manifestation. “It would be impossible,” he wrote, “for an historian to write a history of political corruption in America. What he could write is the history of the exposure of corruption.” And that exposure can be a complex and haphazard affair.

We can identify three key components. First, there must be investigation by law enforcement or intelligence agencies, which can be very difficult when the suspects are powerful or well-connected. Facing many obstacles to a free and wide-ranging investigation, the agencies involved will commonly leak information in the time-honored Washington way. The probability of such investigations and leaks depends on many variables, including the degree of harmony and common purpose within an administration. An administration riven by internal dissent or ideological feuding will be very leaky, and the amount of information available to media will accordingly be abundant.

Second, a great deal depends on the role of media in handling the allegations that do emerge. Some lurid tidbits will be avidly seized on and pursued, while others of equal plausibility will be largely ignored. That too depends on subjective factors, including the perceived popularity of the administration. If media outlets believe they are battering away at an already hated administration, they will do things they would not dare do against a popular leader.

Finally, media outlets can publish whatever evidence they wish, but this will not necessarily become the basis of a serious and damaging scandal unless it appeals to a mass audience, and probably one already restive and disenchanted with the political or economic status quo. Scandals thus reach storm force only when they focus or symbolize existing discontents.

The Watergate scandal developed as it did because it represented a perfect storm of these different elements. The political and military establishment and the intelligence agencies were deeply divided ideologically, both amongst themselves and against the Nixon White House. Leaks abounded from highly placed sources within the FBI and other agencies. Major media outlets loathed Nixon, and they published their stories at a time of unprecedented economic disaster: the OPEC oil squeeze, looming hyper-inflation, and even widespread fears of the imminent end of capitalism. The president duly fell.

But compare that disaster with other historical moments when administrations were committing misdeeds no less heinous than those of Richard Nixon, but largely escaped a like fate. Victor Lasky’s 1977 book It Didn’t Start With Watergate makes a convincing case for viewing Lyndon Johnson’s regime as the most flagrantly corrupt in U.S. history, at least since the 1870s. Not only was the LBJ White House heavily engaged in bugging and burgling opponents, but it was often using the same individuals who later earned notoriety as Nixon-era plumbers. In this instance, though, catastrophic scandals were averted. The intelligence apparatus had yet to develop the same internal schisms that it did under Nixon, the media remained unwilling to challenge the president directly, and the war-related spending boom ensured that economic conditions remained solid. Hence, Johnson completed his term, while Nixon did not.

Nor did it end with Watergate. Some enterprising political historian should write a history of one or more of America’s non-scandals, when public wrongdoing on a major scale was widely exposed but failed to lead to a Watergate-style explosion. A classic example would be the Whitewater affair that somewhat damaged Bill Clinton’s second term but never gained the traction needed to destroy his presidency. In that instance, as with the Iran-Contra affair of 1987, the key variable was the general public sense of prosperity and wellbeing, which had a great deal to do with oil prices standing at bargain-basement levels. Both Reagan and Clinton thus remained popular and escaped the stigma of economic crisis and collapse. In sharp contrast to 1974, a contented public had no desire to see a prolonged political circus directed at removing a president.

So we can take the story up to modern times. The Obama administration did many shameful and illegal things, but the law-enforcement bureaucracy remained united and largely under control: hence the remarkably few leaks. The media never lost their uncritical adulation for the president, and were reluctant to cause him any serious embarrassment. And despite troublingly high unemployment, most Americans had a general sense of improving conditions after 2009. The conditions to generate scandal did not exist, nor was there a mass audience receptive to such claims.

So yes, Obama really did run a scandal-free administration.

What you need for an apocalyptic scandal is a set of conditions roughly as follows: a deeply divided and restive set of bureaucrats and law-enforcement officials, a mass media at war with the administration, and a horrible economic crisis. Under Trump, the first two conditions assuredly exist already. If economic disaster is added to the mix, history suggests that something like a second Watergate meltdown is close to inevitable

Philip Jenkins is the author of The Many Faces of Christ: The Thousand Year Story of the Survival and Influence of the Lost Gospels. He is distinguished professor of history at Baylor University and serves as co-director for the Program on Historical Studies of Religion in the Institute for Studies of Religion.