For several years I served as a Technically Trained Agent — a TTA — in the FBI. TTAs provide technical investigative and surveillance support across the full range of the Bureau’s cases. Part of that work, for me, was installing GPS trackers on subjects’ vehicles to monitor their movements.
It’s worth being precise about the law here, because it’s the whole point of the story. At the time, under the controlling Supreme Court precedent, attaching a tracker to a vehicle and following its signal did not require a warrant. This was routine, lawful, sanctioned practice. I installed a fair number of them, and never once with a warrant, because none was required.
They were an extraordinary tool. A tracker let a team hang back — far enough to keep everyone safe, the public included — and still not lose a subject. Trading a nerve-wracking rolling tail for a dot on a screen made certain investigations both safer and more effective. I won’t walk through any particular case, but I’ll say the installs themselves were rarely comfortable; you don’t forget the quiet of doing that kind of work in the dark. And it was, then, entirely permissible.
Then came the morning of January 23, 2012.
I was at my desk when the Supreme Court handed down United States v. Jones. I caught it on whatever news channel was running in the office and pulled the opinion. Reading it, I muttered, “Oh, $#!+,” and then, a little louder, “F—!” I printed it, walked into the Chief Division Counsel’s office, and asked if he’d seen it — because the Court had just held that attaching a GPS device to a vehicle is a Fourth Amendment search. Overnight, something I had lawfully done many times had been reclassified.
And it created an immediate, almost comic legal puzzle: if putting a device on a car is a search, what is taking one off? I’ll spare you the deliberation that followed. The point is that a single opinion turned settled practice into a live problem in the time it took to read it.
That morning rekindled my interest in the law. I dusted off my law degree and bar license to put them back to work alongside the TTA and engineering background. In the years since I’ve followed the cases that came after Jones — Riley v. California, Carpenter, and now Chatrie. I want to fix what those cases keep exposing. But first, let me show you what they expose — because the thread running through all of them is the same thread from Tuesday’s post.
Here is the part of Jones that mattered to me at my desk that morning, and that matters to this whole blog. The Court didn’t rule against me because tracking a suspect for a month invaded his privacy. Five justices didn’t reach that question at all. Justice Scalia’s majority decided it on a property theory: attaching the device to the vehicle was a physical trespass on one of the “effects” the Fourth Amendment names, and that trespass — not the surveillance it enabled — was the search. Read that against Tuesday’s post and the connection is exact. Jones protected the effect: the car, the object, the thing you can bolt hardware onto. The constitutional trigger was the government touching a piece of tangible property. I wasn’t the surveillance. I was the trespass.
That’s a clean, administrable rule, and I want to steelman it before I bury it, because it has real virtues. A property test is predictable. An agent knows whether he physically intruded on something; he often can’t know whether a court will later find a “reasonable expectation of privacy.” The trespass theory keeps judges close to the constitutional text and out of the business of intuiting privacy norms case by case. If you value rules an officer can actually follow at 3 a.m., Jones‘s property theory is not a bad one.
It has exactly one problem: it can only reach cases where the government touches your stuff. And that describes almost none of modern surveillance.
Watch the theory fail across the next two cases. In Carpenter v. United States (2018), the government didn’t touch anything of Carpenter’s — it asked his cell carrier for months of location records the company already held. No trespass, no “effect” intruded upon. The property theory had nothing to say. So the Court reached for the other idea from Jones — the one Justices Alito and Sotomayor pressed in concurrence, that aggregating enough location data invades a reasonable expectation of privacy regardless of any physical intrusion — and held that accessing that record was a search. The protected thing was no longer the object. It was the information.
Then Chatrie v. United States, decided this past June. Police used a geofence warrant to make Google hand over the location data of every device near a crime scene. Again: nothing of the defendant’s was touched. There is no car, no bumper, no “effect” to trespass on — only a record sitting on a company’s servers. The property theory that caught my work in 2012 is simply unavailable here. And the Court, 6–3, held it a search anyway, because a person keeps a reasonable expectation of privacy in his location data even when a third party holds it. Justice Kagan’s majority made a point that could serve as this blog’s epigraph: a search is a search based on what it is, not on what it turns up — the government can’t defend an intrusion by pointing to how little, or how much, it happened to collect.
Line the three up and you can watch the Fourth Amendment migrate from one word in its own text to another. Jones protected the effect — the tangible object, the thing I physically touched. By Chatrie, there is no effect in the picture at all, and the Court is protecting what is functionally the modern papers: the content, the information itself, wherever it happens to live. That is the papers-versus-effects distinction from Tuesday, playing out in real time across fourteen years of surveillance law. The object-based reading kept breaking as the technology moved the data off the object; the content-based reading is what the Court keeps being forced back to.
I lived on the seam of that migration. I stood on the “effects” side when it was solid ground, bolting devices to cars under a rule everyone understood. Then the ground moved, and it has been moving ever since — Jones to Carpenter to Chatrie, each case redefining “search” a little further to chase a technology that will not hold still.
Here’s what fourteen years and three landmark cases have actually produced: a definition of “search” that gets rewritten every few years, one fact pattern at a time, always a step behind the tools. That is a terrible way to run a rule that armed agents and federal judges have to apply in real time. It isn’t the Court’s fault — courts can only decide the case in front of them. It’s the fault of a Congress that has left the hardest questions of digital surveillance entirely to case-by-case adjudication rather than writing down, in advance, what the government may and may not do.
I spent years on the operating end of that vacuum. Now I’d rather help fill it. That’s what this blog is for — and it’s why Tuesday’s boring little section on definitions is the least boring thing I’ll write here. Define “search,” define “papers,” define what the government may do to your data before it does it, and agents stop learning the rules from a muttered expletive at their desks.
DigitalCrimPro 
Leave a comment