More fake news: how honest photographs were used to shrink an inauguration crowd
Sometimes it is not the lie you tell but the truth you select.
For a full week after the inauguration of 20 January 2017 the major American news outlets ran, and re-ran, a pair of photographs: the crowd on the National Mall at Barack Obama’s first inauguration in 2009, packed to the Washington Monument, and the crowd at Donald Trump’s, with white ground showing between the clusters. The new press secretary used his first briefing to insist the crowd had been the largest ever, which it was not, and an adviser coined the phrase “alternative facts” defending him, which did not help. This page was not about the press secretary. It was about the photographs, and about how two honest pictures can be arranged to tell a story neither of them supports.
Same time, different meaning
Both comparison photographs were genuine and both were taken at roughly the same time of day: about an hour before the ceremony began. That sounds like the fairest possible basis. It is not, because what the pair actually showed was which crowd arrived first. A third photograph, taken from the same position during the 2017 address itself, showed the Mall filled much further back than the comparison had suggested, which is what the president said he saw from the steps.
Several ordinary reasons explained the difference in timing. It rained on the morning of 20 January 2017 and did not on 20 January 2009; people arrive early to stand in sunshine and late to stand in rain. The 2009 inauguration was a unique historic event for which position on the Mall was first-come, first-served, and its attendees had reason to arrive at dawn. The District of Columbia, whose residents find the Mall easiest to reach, had voted 91 per cent for the other candidate. Nearly a hundred protest groups had announced they were coming to disrupt the day, so the security perimeter had fewer entrances and everyone passed through screening, which slowed the crowd’s arrival. And the Metro ridership figures used as corroboration, some 300,000 to 400,000 lower than 2009, ignored that ride-hailing did not exist in 2009 and that tens of thousands of its drivers now worked the capital.
Same event, different lens
A second comparison, used by many outlets, set a photograph of the 2009 crowd taken from the Capitol balcony against one of the 2017 crowd taken from near ground level in the middle of the Mall. From a balcony a crowd fills the frame to the horizon; from the ground it thins into perspective within a few hundred feet. The two photographs were of different things.
Was the 2009 crowd larger? Probably, and by a wide margin, which was never in serious doubt: the first inauguration of a Black president drew what was then the largest crowd ever recorded on the Mall. The question here was different. It was whether the photographs proved what they were presented as proving, and they did not, because they had been chosen for the hour and the angle that made the point. Compared with Bill Clinton’s second inauguration in 1997, the 2017 crowd was large. Compared with 2009, almost everything is small.
How true pictures make fake news
The method, set out here in the language of the time, was this. Pick a time of day at which photographs happen to favour the point. Present near-simultaneity as fairness. Use an elevated shot for one crowd and a ground-level shot for the other. Repeat for a week. No individual photograph is doctored and no caption lies, and the reader ends up believing something the evidence does not show. That is fake news made from true material, and it was the kind that the outlets running the story were not, in January 2017, inclined to recognise in themselves.
Later note. In 2018 a Freedom of Information request revealed that a government photographer had, at the president’s request on the day after the inauguration, cropped the official photographs to remove the empty space, and the inspector general of the Interior Department confirmed it. Both sides of this argument, in other words, chose their frames. The 2009 crowd was larger; the comparison photographs were misleading; and the official record was edited to make the smaller crowd look bigger. All three are on the record, and this page, which made only the second point, is left as it stood.