Criminal Procedure
Why it Matters
Four of the ten amendments in the Bill of Rights govern what the government may do between a first suspicion and a verdict. Anyone headed into law enforcement, corrections, or paralegal work will apply them from the inside, at a roadside or in an interview room, where nobody stops to brief the question. Public argument about these rules runs to whether they help or hinder police. The working question is narrower: which amendment governs, what standard the government must meet, how it comes out on these facts, and what remedy follows if the government loses. This activity has you run that sequence yourself, then catch an AI tutor running it wrong.
Current Context
In June 2026 the Supreme Court held 6-3 in Chatrie v. United States (opens in a new tab) that police conduct a Fourth Amendment search when they obtain Google Location History through a geofence warrant. Justice Kagan’s opinion carried Carpenter’s reasoning to location records, and the Court sent the case back to the Fourth Circuit to decide whether this warrant satisfied probable cause and particularity. Four months earlier, on February 25, 2026, the Court decided Villarreal v. Texas (opens in a new tab), upholding an order that barred a testifying defendant from talking with counsel about his ongoing testimony overnight. No Justice dissented, and the Constitution Annotated’s essay on the case (opens in a new tab) places it in the Sixth Amendment line running from Geders through Perry. One Term redrew the lines at both ends of a criminal case: what investigators may collect before charges, and what counsel may say once the trial is running.
Key Concepts
Warrant Requirement
The Fourth Amendment’s default rule: a search or seizure needs a warrant supported by probable cause. The recognized exceptions (consent, exigent circumstances, search incident to arrest, the automobile exception, plain view) do most of the work in practice. So a suppression fight is usually about whether an exception applies, not about whether the default holds.
Reasonable Expectation of Privacy
The test from Katz v. United States for whether government conduct counts as a search at all: did the person expect privacy, and does society accept that expectation as reasonable. Carpenter v. United States carried the test into digital records a third party holds, and Chatrie v. United States applied it to Google Location History in 2026. The question comes first in any Fourth Amendment analysis, because conduct that is not a search never reaches the warrant rule.
Probable Cause and Reasonable Suspicion
The two working standards of street-level criminal procedure. Probable cause, required for arrests and most warrants, means a fair probability that a crime occurred or that evidence will be found. Reasonable suspicion, the lower Terry standard, means specific and articulable facts pointing to criminal activity, enough for a brief stop and not for an arrest.
Stop and Frisk
The Terry v. Ohio doctrine: an officer with reasonable suspicion may briefly stop a person, and may pat down outer clothing if the officer reasonably suspects that person is armed and dangerous. The frisk is a weapons check and nothing more. A pat-down that turns into a hunt for contraband has left what Terry allows.
Exclusionary Rule
The remedy that makes the other rules bite: evidence obtained in violation of the Fourth Amendment is generally inadmissible, a rule Mapp v. Ohio applied to the states. It is a judicially created deterrent rather than a command in the constitutional text, which is why the Court has carved out exceptions like good faith. The suppression motion is where it gets tested.
Privilege Against Self-Incrimination
The Fifth Amendment right not to be compelled to be a witness against oneself, applied to the states in Malloy v. Hogan and enforced through the Miranda framework. Warnings are required before custodial interrogation, and the two triggers are custody and interrogation: neither one alone sets the requirement off. Statements taken in violation are generally kept out of the prosecution’s case in chief.
Right to Counsel and to a Jury
The Sixth Amendment pair that structures the trial itself. Gideon v. Wainwright made appointed counsel for indigent felony defendants binding on the states, and Argersinger v. Hamlin carried the rule to any offense that ends in jail time. Duncan v. Louisiana incorporated the jury right, Williams v. Florida then held six jurors enough, and Kian v. Florida asks the Court to reconsider that.
Resources
- US Constitution (Constitution Annotated) (opens in a new tab) (Read the Fourth, Fifth, and Sixth Amendments before you start; every scenario below turns on their text.)
- Chatrie v. United States, No. 25-112 (U.S. June 29, 2026) (slip opinion) (opens in a new tab) (The Term’s geofence decision, and the primary text behind the digital-search scenario and the Part 3 dispute you track.)
- Villarreal v. Texas, No. 24-557 (U.S. Feb. 25, 2026), full opinions (Cornell LII) (opens in a new tab) (Justice Jackson’s majority, Justice Alito’s concurrence, and Justice Thomas’s opinion concurring in the judgment; the primary text behind scenario six.)
- Villarreal v. Texas, No. 24-557 (U.S. Feb. 25, 2026): Right to Confer with Counsel (Constitution Annotated) (opens in a new tab) (The Library of Congress essay, which sets the decision against the circuit split it resolved.)
- Oyez (opens in a new tab) (Case summaries and argument audio for the spine cases below. Use it to get your bearings, then confirm every holding, date, and vote against the opinion.)
- Google Scholar (opens in a new tab) (Full opinions for any case you need to read; Westlaw via the Daytona State College library (opens in a new tab) works too.)
The cases below are the doctrinal spine of this activity, and the opinions you will be checking the model against: Mapp v. Ohio (1961), Gideon v. Wainwright (1963), Malloy v. Hogan (1964), Miranda v. Arizona (1966), Katz v. United States (1967), Terry v. Ohio (1968), Duncan v. Louisiana (1968), Williams v. Florida (1970), Argersinger v. Hamlin (1972), Rodriguez v. United States (2015), and Carpenter v. United States (2018).
What to Do
You will work six police-encounter scenarios with an AI tutor, check every conclusion it reaches against the sources, and follow one live fight over investigative technology. The skill is applying doctrine to facts, and noticing when the model doing it for you has gone wrong.
Part 1: Work the Scenarios
Copy the Encounter Analyzer into a fresh chat and work the six scenarios below one at a time. For each, record the governing amendment, the standard the government must meet, the likely outcome on these facts, and the remedy if the government loses. Then check each conclusion against the Constitution Annotated or the opinion itself, and note every place the model hedged, missed a controlling case, or stated a rule too broadly.
Scenario 1. Acting on an anonymous call reporting a man with a gun, an officer stops a man matching the description on a public sidewalk and pats down his outer clothing.
Scenario 2. During a routine stop for a broken taillight, the officer finishes writing the ticket, then holds the driver eight more minutes so a drug dog can arrive and circle the car.
Scenario 3. Investigating a robbery, detectives ask Google for a list of every device present within a block of the scene during a one-hour window, without seeking a warrant.
Scenario 4. Detectives question a suspect for two hours in a closed interview room at the station without Miranda warnings, having told him once at the start that he is free to leave.
Scenario 5. A defendant charged with a misdemeanor that carries possible jail time tells the judge he cannot afford a lawyer and is told to proceed on his own.
Scenario 6. Midway through a defendant’s testimony, the judge recesses trial overnight and orders him not to discuss his ongoing testimony with his attorney until he retakes the stand.
Part 2: Build a Doctrine Timeline
Pick one right to trace in depth: search and seizure, self-incrimination, right to counsel, or trial by jury. Run the Doctrine Timeline Builder, then verify every entry it produces, starting with the two the model itself flags as most likely wrong. Keep notes on what it got wrong and on why the error was believable, because a believable error is the one that survives a careless check.
Part 3: Track a Live Dispute
Run the Tech Dispute Tracker and pick one current fight over investigative technology. Confirm through your own searching that the dispute is real and fairly described, and read at least one of the sources the model cites. What you want by the end is a working picture of where Fourth Amendment doctrine is heading after Carpenter and Chatrie.
Part 4: Write Your Post
Write a 400–600 word post with three components. A scenario table covering all six scenarios: the right at stake, the standard, your verified conclusion, and the source you checked it against. A full prose analysis of the one scenario you found hardest, walking through the doctrine step by step. A closing paragraph covering the errors you caught the model making across all three parts and a 3–4 sentence account of the live dispute you tracked.
Part 5: Share Chat Link
Include one AI chat link with a 1–2 sentence explanation of what the conversation shows and why you chose to share it.
Suggested AI Prompts
Use these prompts as a starting point, then adjust them to fit your goal. Strong prompting develops through trial, revision, and testing. It’s a foundational skill that grows into more advanced AI work such as context engineering and agent-based workflows.
The Encounter Analyzer
You are a criminal procedure tutor for a college constitutional law student. I will paste one police-encounter scenario at a time. For each scenario, work step by step rather than announcing a conclusion: identify the governing amendment and the specific right at stake, name the controlling doctrine and the leading Supreme Court case that established it, state the standard the government must meet, apply that standard to the facts I gave you, and say what remedy a court would consider if the government’s action falls short. Where the outcome depends on a fact the scenario does not state, name the missing fact explicitly and explain how each version of it would change the result. End every analysis by naming the single best source for me to confirm the rule you applied: the Oyez page for the controlling case, the relevant Constitution Annotated essay, or the opinion itself. Ask me for my first scenario.
Decomposition and missing-fact elicitation: the model reasons in ordered steps and names what it does not know, which gives you analysis you can actually check. The closing instruction builds verification into every exchange instead of leaving it to your good intentions.
The Doctrine Timeline Builder
You are a constitutional criminal procedure researcher. Begin by asking me which right I want to trace: search and seizure under the Fourth Amendment, self-incrimination under the Fifth, right to counsel under the Sixth, or trial by jury under the Sixth. Then build a chronological timeline of the Supreme Court cases that shaped that right, from the earliest significant decision to the present, giving each entry the case name, the year, a two-sentence statement of the holding, and one sentence on how it changed what came before. Flag any entry where you are uncertain about the year, the holding, or the sequence instead of presenting it with false confidence, because I will verify every entry against authoritative sources before relying on it. End by naming the two entries in your own timeline you consider most likely to contain an error and why.
Announced verification changes how the model answers: told that every claim will be checked, and made to nominate its own likeliest errors, it surfaces the doubt models usually hide. Those self-nominated errors are your triage list.
The Tech Dispute Tracker
You are a legal research assistant tracking Fourth Amendment disputes over investigative technology. Begin by asking me which technology I want to explore: geofence warrants after Chatrie v. United States, facial recognition, automated license plate readers, cell-site simulators, or another I name. Then use your web search tool to find one live dispute from the past two years involving that technology: a pending case, a new statute, or an agency policy fight. Explain who the parties are, what the Fourth Amendment question is, and how Carpenter v. United States and Chatrie bear on it. Cite the sources you relied on so I can confirm the dispute is real and fairly described, and keep what your sources establish clearly separated from your own analysis of where the doctrine is heading.
Source grounding: live search with required citations keeps the model out of its training data, where recent legal developments are least reliable. Making it separate sourced fact from its own analysis exposes the seam you most need to check.