Before Donald J. Trump came on the electoral scene, not a lot of people had heard of the Goldwater Rule. By now many surely have. The eponymous rule was established by the American Psychiatric Association, my own guild organization. It stipulated that no psychiatrist should make a diagnosis of a person he or she has not examined face-to-face and who has not given consent to discuss his mental health publicly.
While do-it-yourself diagnosing of a certain presidential candidate has become a cottage industry in this election—the consensus about Donald Trump is that he might have a severe case of Narcissistic Personality Disorder—psychiatrists have been understandably reluctant to weigh in, even as people have clamored to say that they deserve to know everything possible about a potential president’s mental fitness.
How did the rule come to be? In the lead up to the 1964 election, Fact magazine surveyed over 12,000 psychiatrists about the personality traits of Sen. Barry Goldwater, the Republican nominee. Only a modest percentage of the psychiatrists responded but of those that did, almost half felt the candidate was psychologically unfit to be president.
They savaged Goldwater, judging him “warped,” “narcissistic,” “impulsive” and a “paranoid schizophrenic” who harbored unconscious hatred of his Jewish father and was scarred by his rigid toilet training. One respondent saw him as “a frightened person who sees himself as weak and threatened by strong virile power around him—and that his call for aggressiveness and the need for individual strength and prerogatives is an attempt to defend himself against and to deny his feelings of weakness and danger.” Even his supporters came in for unfavorable psychoanalysis.
In the aftermath, two things happened. First, Goldwater, who lost the presidency in a landslide, sued Fact for libel and was awarded $75,000 in punitive damages. He deserved to win the suit. The psychiatrists’ intra-psychic assessments were deeply intrusive, disrespectful, and basically meaningless. Warmed-over Freudianism is no way to foretell one’s actions in office. Second, the APA issued the Goldwater Rule, which meant to prohibit such distant assessment.
The American Psychiatric Association still stands by the Goldwater Rule and no psychiatrist can go wrong by following it. Of course, many can easily skirt around it—it’s thanks to the rule that modern-day assessments of celebrities’ mental health are couched with the “I have not personally examined this person, but … ”
It goes without saying that any clinical encounter with a nonpsychotic or imminently dangerous person is strictly confidential unless the subject gives permission to reveal the findings to anyone else. But I think there are conditions under which psychiatrists should have more latitude in making determinations from a distance.
To me, the reason why it can be responsible to profile a patient without examining him or her (in other words, to circumvent the Goldwater Rule), has to do with the way diagnoses are made today compared to the way they were made then. I’m referring to the Diagnostic and Statistical Manual, Fifth Edition, which many consider the gold standard for evaluating mental disorders.
The diagnostic entities contained in the current DSM are based on signs (what clinicians observe) and symptoms (what patients report). But in Goldwater’s time, Freudian theory dominated American psychiatry and pathology: Even severe depression and schizophrenia were tied to psychoanalytic formulations. Symptoms such as mood or hallucinations or obsessions were almost irrelevant—instead, diagnosis relied almost entirely on what internal conflicts or stalled developmental stages a patient had experienced in his or her lifetime. This required a deep understanding of the patient’s life. None of the psychiatrists quoted by Fact had access to that full picture of information. Thus, not only were their opinions ill-informed, they were based on certain theories of mental health that are now considered highly questionable by many in the field.
But the assessment of mental disorders changed to a more objective system of taxonomy in 1980 with the publication of the DSM-III. A number of diagnoses are now made largely on a person’s observable behavior or what can reasonably be inferred from it.
So it is now possible to make a psychological assessment from afar. The question remains of whether it is appropriate. One of the biggest concerns about armchair diagnosing is that it’s a drive-by affair: Quick pronouncements are made based on a single transgression or a glimpse of erratic behavior. But even when a clinician conducts a formal interview it may not pick up a lot. In an hourlong interview, a savvy politician could easily present as better adjusted than he or she actually is—in some cases, such impression management might go undetected, even by experienced examiners.
What’s more, as we have seen, some politicians, like Trump in particular, can run circles around most interviewers, and it’s doubtful that even trained psychiatrists could get much more information from them in a direct examination. Thus, a formal clinical interview with him might not shed much light.
In contrast, a political campaign for president offers an abundance of unfiltered observable behavior. Indeed, voters have been inundated—they have watched the candidates on a near daily basis for many months. And for these two candidates in particular, the volumes of documentation go back decades: we have books, articles, and interviews with people who have known them. In this case, the public actually has more “data” than many psychiatric evaluators have to go on when they see certain patients in certain settings.
Research supports the importance of extensive exposure to a subject’s behavior. According to psychologist Scott Lilienfeld, a personality researcher and professor at Emory University, research has demonstrated that informal, unstructured interviews of the kinds typically conducted in direct examinations of psychiatric patients are often not especially reliable–that is, consistent across different mental health professionals. “What’s more,” Lilienfeld says, “informant reports from those who know the person well are often more predictive of people’s behavior than are self-reports, which are much of what one obtains from a direct interview.”* More broadly, he says, assessments of personality are most likely to be valid when multiple sources of information are integrated.
There is still one huge technical catch to diagnosing at arms’ length. The DSM requires that a diagnosis must include the presence of "significant impairments in self (identity or self-direction) and interpersonal (empathy or intimacy) functioning." The first part of that diagnosis is hard, if not impossible, to make without a private interview, especially in someone who has succeeded at a high level based in part on these traits. So, while I believe it is occasionally appropriate to speculate, any assessment should be considered just that—an assessment, not a diagnosis.
This brings us to Narcissistic Personality Disorder, which is most often invoked by his detractors to describe Donald Trump.
Here are the nine diagnostic criteria for NPD in DSM-V. (These criteria go far beyond what we usually mean when we casually refer to someone as a narcissist.) Five are needed to be eligible for the diagnosis:
- A grandiose logic of self-importance
- A fixation with fantasies of infinite success, control, brilliance, beauty, or idyllic love
- A credence that he or she is extraordinary and exceptional and can only be understood by, or should connect with, other extraordinary or important people or institutions
- A desire for unwarranted admiration
- A sense of entitlement
- Interpersonally oppressive behavior
- No form of empathy
- Resentment of others or a conviction that others are resentful of him or her
- A display of egotistical and conceited behaviors or attitudes
Sure, it seems obvious that Trump could qualify—perhaps he is even a severe case. But keep in mind that some would easily attribute these traits to Hillary Clinton, too. And herein lies the real virtue of the Goldwater Rule: Allowing psychiatrists to diagnose or assess at a distance would open a floodgate of efforts, many of which will be inaccurate, slanted, or politically motivated. The Goldwater Rule attempts to keep these gates closed. The DSM cannot become a political instrument.
Of course, people don’t like this answer. As voters, we feel entitled to absolute transparency about our candidates, even about their health.
But we shouldn’t. So what if a candidate appears to meet criteria for NPD? It is no secret that many of the DSM criteria listed above are manifest in Trump. But even a clinically authentic diagnosis (which has not been made, or at least surfaced publicly) would not mean, prima facie, that he should not run for president, or that he would necessarily be a bad one if elected. Surely, there have been people with NPD who have held positions of power and achieved good and great things. There are also plenty of people who don’t have the disorder who have held positions of power and failed.
We are too caught up in diagnosis. Even if Trump were to be officially diagnosed with this disorder, it would do little to sway voters from either supporting or reviling him.
After all, the enduring traits that many find so problematic in Mr. Trump–and that his supporters fail to see, dismiss, or downplay because other aspects of his candidacy appeal so strongly to them—are not ones that necessarily indicate a formal mental condition. I’m referring, of course, to various dispositions, such as a lack of interest in learning about the governing process or world affairs, a tendency to ignore advisers, impulsive and crude modes of retaliation, a reflexive instinct to scapegoat others, and so on. One does not need to be a specialist of the mind to understand why many would not want to vote for such a person, or why others might fear being governed by such a man.
So, I am a minority view within my profession because I do not believe it is necessarily an ethical lapse for a psychiatrist to venture an assessment of a public figure—especially because the “evidence” in this election has been so widely accessible and formal assessment is often made on the basis of that very evidence.
But in this case, I don’t think a diagnosis would actually provide any new insight into the choice we have to make about who we should elect president. It does not matter whether Trump might have NPD.
That’s because diagnoses do not allow us to predict future actions with great certainty. Past behavior and enduring character traits, on the other hand, do—and we have seen more than enough to understand what Trump’s (and Clinton’s) are.
*Update, Oct. 13, 2016: An earlier version of this story omitted the word “one” from a quote by psychologist Scott Lillienfeld. It has been updated.