Trump’s War on Fact-Finding

Transparency about errors is a strength of journalism and science, not a weakness.

On February 20, 2019, President Donald J. Trump tweeted “The New York Times reporting is false. They are a true ENEMY OF THE PEOPLE.” He did not indicate what reporting he had in mind, but the timing suggested that he was angered by the newspaper’s February 19 story about his use of intimidation against the various investigations into his campaign, presidency, and businesses. It is one thing to claim a press report is false, but quite another to label an entire publication an “enemy of the people.”

At this point we have gotten used to the president labeling whatever press reports displease him as “fake news” and scientific claims that contradict his policy priorities as “hoaxes.” White House officials even go so far as to promulgate “alternative facts” instead of claims reported by traditionally reliable sources. These attacks undermine the trust we should have in the best justified claims we have.

There is a nefarious logic at work in such denouncements. The norm that is cynically alleged to be violated is accuracy. Accuracy is, certainly, the goal of good journalism and of good science, and failures of accuracy are worthy of criticism. But Trump’s denunciations implicitly appeal to a misleading view of what counts as a failure to uphold the norms of accuracy.

Trump exploits an absolutist perspective that accuracy must be devoid of all error. Thus failure occurs if a journalist or scientist is not perfectly correct. Any mistake, slip, or error, however large or small, whatever its cause, warrants condemnation of the offender, if not his employer and whole profession. There are three problems with this view.

First, to err is human, and hence inevitable. Reports by journalists and scientists are not always going to be 100-percent accurate. All claims to knowledge are fallible. Second, errors come in degrees of seriousness. A misprint is not equivalent to fraud. And third, there are different explanations for error, from individual failures (moral or mental) to structural failures (absence of oversight or equipment failure). The absolutist “devoid of all error” view does not reflect the real world, yet Trump uses that perspective to dismiss the entire practice of legitimate, justified claims that meet more realistic epistemic norms. When the claims do not support what he finds ideologically desirable, he swings the hammer of “absolute accuracy” to smash the very practice of fact-finding.

In fact-finding practices such as journalism and science, failures are not only expected, but their exposure is used to develop more accurate and trustworthy reporting. Minor mistakes fall within the noise of human fallibility. Typos are simply reported and fixed. Reporting of verifiably false claims are the basis for rethinking and revising methods for gathering and using evidence. If the false claim was the result of deception by a reporter’s source, or by an unanticipated failure of an experimental instrument, the reporter or scientist might be responsible for lack of due diligence or for sloppiness or carelessness, but not of lying or fraud. The tolerance for acceptable error is set by the standards of practice.

For journalists there are explicit rules of ethical behavior, including using reliable sources, minimizing harm, avoiding conflicts of interest, and being accountable and transparent. Notice that it does not include the impossible demand to publish only what is 100-percent accurate. No matter how assiduously a journalist follows the rules, errors will still occur. Error isn’t eliminated (and this is the point that Trump obscures to achieve his own ends); it is managed by professional norms and structures based on understanding the sources of past failures of accuracy. Accountability warrants epistemic and professional sanctions if the error is generated by an individual’s inexcusable lapse of judgment or care. Error propagation is limited by corrections or by retracting the claims or entire publications once an error is exposed.

Transparency allows a window into the circumstances and contexts that might compromise any individual’s ability to follow due diligence. For example, the current pressure to publish almost immediately thwarts efforts to follow more thorough fact and source checking. Sometimes urgency cannot be avoided, but our judgments and sanctions should be moderated accordingly. Where there is lying, deception, and dishonesty, then condemnation and more are appropriate. However, failure of accuracy includes much more than that extreme and requires more nuanced responses. If errors are never exposed, then we cannot learn how to redress them, or determine whom to blame, or what circumstances contributed. When errors occur and are investigated and understood, then we learn how to restructure and reform legitimizing practices to produce more accurate reporting in the future.

Science shares the goal of accuracy with journalism and scientific claims are similarly fallible, though the sources of error are different. The best practices of science also include transparent pursuit of truth and accountability, explicit in the 2018 “Code of Ethics for Researchers.“ Rather than fact-checking by multiple sources or official sources, scientists rely on peer review and replicability to limit idiosyncratic bias. Transparency of data and statistical algorithms increase trustworthiness in the results. This is because revealing the details on which claims are based permits identification of sources of error. Knowing the sources can prompt modifications to scientific practices to prevent similar errors in the future.

For science, there have been studies of retractions of scientific papers akin to the National Transportation Safety Board’s studies of plane crashes. Why did they happen, and how can we do better? What new regulations would prevent the publication of erroneous results or the frequency of plane crashes? Retraction Watch, which has produced a database of more than 18,000 scientific papers that have been retracted since the 1970s, has the motto “Tracking retractions as a window into the scientific process.” While roughly 60 percent of all retractions are due to misconduct (by a disproportionately few scientists), 40 percent are ascribed other sources of error from mistakes in printing to errors in data, methods, or analysis.

In a political climate where any claim if it is not 100-percent accurate is deceivingly labeled utterly false and its author condemned for fraud, defending the less-than-perfect, fallible, yet justified claims of reporters and scientists becomes increasingly difficult. Will we become less willing to seek the many sources of varying degrees of error and admit when something is inaccurate? If so, we will have embraced a cynical and sophistic view of “accuracy” that allows almost any justified statement of fact to be impugned and replaced with an “alternative fact” that better fits the political or economic goals of those in power. We need to confront the misrepresentations of what knowledge is in order to preserve the role Justified facts as the best resources humans have to understand the world, navigate our day-to-day decisions, and ground our social policies.

Sandra D. Mitchell

Sandra D. Mitchell is distinguished professor in the Department of History and Philosophy of Science at the University of Pittsburgh, and author of Unsimple Truths: Science, Complexity and Policy (University of Chicago Press, 2009).