He wasn’t sure what he was looking for. He was still at the stage when the important thing was to look at everything, absorb as much data as he could. The decisions about what was significant, the search for patterns—that would come later.

He realized he’d been in too much of a rush the first time. He needed to slow down. He’d discovered over the years that one of the most destructive errors a detective can make is to leap to a possible pattern with too little data. Because once you think you see a pattern, there’s an inclination to dismiss data that doesn’t fit into it. The brain’s natural affinity for pattern formation devalues dots that don’t contribute to the picture. Add to that a detective’s professional need to grasp the outline of a situation quickly, and the result is a tendency to jump to premature conclusions.

The period of simple looking, listening, absorbing had tremendous value. Giving that period its full due was always the best way to begin an investigation.

Begin an investigation?

Begin an investigation of what, exactly? At whose request? With what legal authorization? In potential collision with Schiff and who else?

He decided to simplify the matter—or at least detoxify the terminology—by thinking of it as nothing more than a private fact-finding initiative, a modest effort to answer a few questions. Questions such as:

Who was behind the original “pranks” that had disturbed Kim?

Which was closer to the truth—Kim’s characterization of Meese or his characterization of her?

Who set the vicious little trap that had thrown him to the basement floor? Was he the intended victim or was Kim?

If the whisper had been real, who was the whisperer? Why was he lurking in the basement? How and when had he gotten into the house, and how had he gotten away?

What was the meaning of the warning “Let the devil sleep”?

And what, if anything, did these present events have to do with a ten-year-old series of roadway murders?

Gurney envisioned his fact-finding initiative beginning with a review of everything in the incident reports, the report annexes, the ViCAP reports, the FBI profile, the status reports in Kim’s project folder, and the notes he’d taken while listening to Hardwick’s acerbic summaries of the victims’ personalities.

All of that he could address by himself. But he also felt a growing urge to sit down with Rebecca Holdenfield and delve more deeply into the Good Shepherd profile and the case hypothesis—how the primary data was gathered, analyzed, prioritized; how theoretical alternatives had been tested; how consensus had emerged; and whether any beliefs she had about the case had changed over the years. He was also curious to know whether she’d ever spoken to Max Clinter.