How sources are vetted.

The source hierarchy

In descending order of weight:

1. Primary clinical and research literature. The DSM-5-TR itself; peer-reviewed papers in journals like the Journal of Personality Disorders, Psychological Medicine, the Journal of Abnormal Psychology, the Annual Review of Clinical Psychology; foundational books by clinicians on the underlying disorder (Kernberg, Kohut, Pincus).
2. Practicing clinicians writing for non-specialist audiences. Books and longer essays by practicing clinicians who treat the population in question — Ramani Durvasula, Wendy Behary, Karyl McBride, Bill Eddy, Lundy Bancroft. Used for clinical observation that hasn't been reduced to controlled-trial form but is consistent across multiple respected practitioners.
3. Survivor-experience literature, used carefully. First-person accounts and survivor-facing books, used to describe the lived shape of the pattern but not as evidence for clinical claims about its cause or prevalence.
4. Established secondary sources. Reputable encyclopedias, professional-association explainers (American Psychiatric Association, APA), credentialed journalism. Used for orientation and dating, rarely as the primary evidence for any specific claim.

What does not get treated as evidence

• Anonymous online posts, Reddit threads, TikTok explainers — even those that quote clinicians correctly. We go to the clinicians.
• Self-published books without independent corroboration. Some are excellent; the verification cost is too high to use them as primary support.
• Single-source clinical claims where no other clinician has corroborated. These may be true; they're flagged as provisional in the prose.
• Anything coming through litigation contexts where the source has a clear motivation to characterize a specific person or pattern. The patterns we describe are general; the evidence for them is general.

Where the evidence is firm and where it isn't

Some claims on this site are well-supported across multiple lines of evidence:

• The DSM diagnostic criteria for NPD.
• The grandiose-vs-vulnerable distinction in personality research.
• The existence of malignant narcissism as a clinically meaningful pattern (Kernberg's framework).
• Complex PTSD as a recognizable trauma syndrome distinct from single-incident PTSD (Herman, ICD-11).
• DARVO as a measurable, replicable phenomenon in perpetrator response (Freyd's work).

Other claims rest more on clinical observation than on large-scale studies, and the prose says so:

• The specific gender-distribution claims about covert vs. grandiose presentation.
• The specific mechanisms by which family courts and intake systems under-detect female covert perpetrators.
• The exact prevalence of covert NPD in the general population (the diagnostic instruments under-detect this; the truth is that we don't have a confident number).

What is editorially excluded

• Naming individuals. Real names of private individuals will not appear on this site. Public figures may be mentioned where directly relevant, with sourcing.
• Diagnostic claims about anyone. No page on this site asserts that a specific named person meets criteria for NPD, malignant narcissism, or any other disorder.
• Reader stories presented as evidence. The site does not solicit, collect, or publish first-person reader narratives.
• Case vignettes that combine material from multiple real people. The risk of identification is too high and the explanatory benefit is small.
• Lurid amplification. Where two equally accurate framings are available, we use the more clinically restrained one.

Corrections

Errors of fact are corrected on identification. Substantive factual changes are noted at the bottom of the affected page. Stylistic edits are not. If you spot a misstatement of the literature or a misquoted source, please flag it through the contact channel listed on the about page.

What this methodology cannot do

It cannot make a topic this politically charged uncontroversial. It cannot guarantee the prose is right in every detail; it can only commit to the line above and to revising when the line is shown to have been crossed. It cannot prevent the material from being misused — by people quoting individual sentences as personal weapons, by litigants trying to repurpose general descriptions as case-specific evidence, by readers who would prefer a tidier story than the literature supports. The site goes ahead anyway because the under-recognition described on the female covert page has its own cost, and it is the larger one.