Key Takeaways
- Use immunity bars the government from using a witness’s statements, and evidence derived from them, but still allows prosecution based on independent evidence.
- Transactional immunity is a broader promise not to prosecute for the covered crimes at all, and in the federal system it usually comes through a negotiated agreement rather than the immunity statute.
- Federal prosecutors, with approval from senior Justice Department officials, decide whether to seek immunity, and a judge cannot order them to grant it.
Is criminal immunity a “get out of jail free” card? When a witness or defendant is granted immunity, what exactly is promised, who decides, and what does the government get in return?
In this TalksOnLaw Brief, Bennett Capers, a former Assistant U.S. Attorney in the Southern District of New York and now the Stanley D. and Nikki Waxberg Chair and Professor of Law at Fordham Law School, explains how immunity works in the federal system. Speaking from his years as a federal prosecutor, where the issue came up often, he walks through the two forms of immunity, how a defendant goes about seeking it, and why prosecutors rather than judges control the process.
Use Immunity and Transactional Immunity
Professor Capers describes use immunity as a promise that the government will not use what the person says against that person in a later prosecution. If an immunized witness admits to a crime, the statement itself cannot be used to convict him. But if prosecutors develop independent evidence of guilt, they can still bring charges. Transactional immunity, by contrast, is a blanket promise not to prosecute for the covered crimes at all, even if independent evidence later emerges, which is why defendants prefer it.
Under the federal immunity statute, the protection a court orders is use and derivative use immunity: neither the compelled testimony nor any evidence derived from it can be used. In Kastigar v. United States, the Supreme Court held that this protection is enough to overcome a witness’s Fifth Amendment privilege, and that the government bears a heavy burden to prove that its evidence came from a wholly independent source. Broader protection resembling transactional immunity is typically provided through a negotiated non-prosecution agreement, and some states continue to grant transactional immunity by statute.
Making the Pitch
Professor Capers stresses that federal prosecutors grant immunity rarely, in part because they have other ways to obtain information. A defendant who wants to cooperate usually works through counsel to arrange a proffer session, often called a “queen for a day” meeting, with the defendant, defense lawyer, and prosecutor present. The defendant lays everything out and explains why the information is valuable, in the hope of earning immunity or another benefit in exchange.
Prosecutors Hold the Cards
As Professor Capers explains, this is one of the few areas where a federal judge has little power. A judge cannot order a prosecutor to immunize a witness. The decision belongs to the prosecutor, and in the federal system it also requires approval from senior Justice Department officials in Washington.
What to Know Now
The framework Professor Capers describes remains in place. The federal immunity statute and Kastigar still govern, and Justice Department policy continues to require approval from Main Justice before a court-ordered grant. Federal courts have generally rejected requests by defendants to force immunity for their own witnesses, though defense lawyers continue to argue that fairness sometimes requires it. Readers may also encounter the word immunity in very different settings: presidential immunity, addressed in Trump v. United States, and qualified immunity for police officers are separate doctrines about who can be sued or prosecuted, not deals that prosecutors offer witnesses.
