The Scope of Miranda's Public Safety Exception

By Mike Dorf

By now it has been widely reported that federal investigators and prosecutors plan to interrogate Dzhokhar Tsarnaev without first reading him his Miranda warnings, pursuant to the "public safety" exception to Miranda v. Arizona.  That exception allows the government to interrogate a suspect without first issuing the Miranda warnings, and then introduce any evidence obtained as a result, when the initial interrogation is undertaken for the purpose of protecting the public from an immediate threat.  A leaked 2010 internal FBI memo contends that in certain terrorism cases, the exception can be broader.  [NB: I've since rethought that reading of the memo, as described in a short follow-up post.]  Here is the money quote:

There may be exceptional cases in which, although all relevant public safety questions have been asked, agents nonetheless conclude that continued unwarned interrogation is necessary to collect valuable and timely intelligence not related to any immediate threat, and that the government's interest in obtaining this intelligence outweighs the disadvantages of proceeding with unwarned interrogation.

Whether the courts find this more expansive view persuasive remains to be seen, but here I want to suggest that it's not even 100% clear that the regular public safety exception applies in a case like Tsarnaev's, much less an expanded exception.

To be sure, the exception only comes into play if the government attempts to introduce a statement obtained from Tsarnaev at his trial.  As Orin Kerr explains on the Volokh Conspiracy, there is no freestanding right of arrestees (or anyone else) to be read warnings.  The warnings only become relevant when the government offers a statement obtained as a result of custodial interrogation.  So, let's assume that happens.  That is, let's suppose that: Tsarnaev recovers; the FBI interrogates him without first giving Miranda warnings, asking him questions designed primarily to detect information about an immediate threat; Tsarnaev makes a statement in response; in addition to using the statement for any information it contains about ongoing threats, the government attempts to introduce it as evidence of Tsarnaev's guilt at his criminal trial.  Is it admissible under the public safety exception?

The public safety exception was first announced in New York v. Quarles and the Court rested it on a number of considerations.  In some respects, interrogation of a terrorism suspect makes out a stronger case for the application of the public safety exception than an ordinary criminal case like Quarles, but in other respects it's a weaker case.  Let's look at the relevant considerations in Quarles and compare them with the Tsarnaev case.

The Quarles opinion is short and I encourage readers interested in this subject to read it.  The main points are these:

1) The Miranda rule balances the needs of law enforcement for convictions of guilty persons against the protection of suspects from the risks to their Fifth Amendment right against compelled self-incrimination--risks that inherently arise from custodial interrogation.  But, as the Court said in Quarles, "the need for answers to questions in a situation posing a threat to the public safety outweighs the need for the prophylactic rule protecting the Fifth Amendment's privilege against self-incrimination."

There are in that statement two points, and they face in opposite directions.  In Quarles, the threat to public safety was a gun that Quarles had discarded.  In the Tsarnaev case, the potential threat to public safety is much greater.  Tsarnaev or his brother or as-yet-unknown co-conspirators may have planted or plotted additional mayhem.  Thus, the "public safety" consideration in the public safety exception is heightened here.

2) However, note that the language I have just quoted from the late CJ Rehnquist's opinion tacitly makes another point, by referring to Miranda as a "prophylactic rule."  He was pointing to the fact that the Court in Miranda did not say that the Fifth Amendment requires the Miranda warnings.  That's the point Kerr makes.  But there's something else too.  CJ Rehnquist was also saying that the Constitution does not even require that Miranda warnings be given if the government later attempts to introduce the suspect's statement as evidence.  All that Miranda does is to say that the pressure inherent in custodial interrogation requires some safeguards to mitigate that pressure, and that the warnings are one such effective set of safeguards.  But, the Court implies in Quarles, because the Miranda rule is judge-made--i.e., "prophylactic"--the Court may choose not to apply it where its costs outweigh its benefits.

That looks pretty good for the government in the Tsarnaev case but for one major problem.  In 2000, in Dickerson v. United States, the Court, in another Rehnquist opinion, rejected the idea that Miranda is a mere prophylactic rule in favor of the view that it was a "constitutional decision."  In dissent, Justice Scalia argued that, in light of Quarles and other cases seemingly holding that Miranda was prophylactic, the Dickerson majority made little sense.   In an article in the Supreme Court Review shortly after the case came down, Barry Friedman and I agreed that the majority opinion did seem somewhat inconsistent with the language of various prior Miranda cases, but that this did not mean those earlier cases had been overruled.  What it did mean was that the rationales of those prior cases needed to be reconceptualized.  Quarles, in particular, could no longer be based on an ad hoc weighing of costs and benefits, but must be reconceptualized in some way.

3) If we look to the other factors at play in Quarles, they appear inapplicable to a case like Tsarnaev.  One point the opinion emphasizes is that time is of the essence: If the gun is not found, perhaps the suspect himself or a hidden co-felon will use it, or a child will discharge it accidentally.  The Court is envisioning a very immediate emergency, referring to the decisions police must make "in a matter of seconds."  That sort of exception would certainly allow the FBI, upon arresting Tsarnaev, or upon his awakening and gaining the power of speech, to ask him a few urgent questions.  But Quarles itself does not envision an extended interrogation.

4) The Quarles Court also thought it noteworthy that Quarles was asked one question--"where's the gun?"--upon his arrest, rather than in the inherently coercive environment of the station house for which the Miranda warnings were originally designed.  It's possible that Tsarnaev will be questioned at his hospital bedside.  That seems somewhat less coercive than the station house, but depending on the length of the questioning, would seem more like it than different.

How does all of the foregoing cash out?  I think it's a fairly close case, although I strongly suspect that, if it came to it, most judges would find the sort of interrogation I'm envisioning falling within the public safety exception.  My point here is simply that it's not an obvious slam dunk.  The seriousness of the threat is doing a lot of the work.

In any event, I also would note a certain unreality to the whole discussion.  The Quarles opinion cites as one important factor the risk that the giving of Miranda warnings will lead suspects not to talk.  But the fact is that Miranda has only a marginal impact on a suspect's willingness to talk.  Occasionally a warned suspect will clam up and/or lawyer up because of the warnings.  But usually they make no difference.  Even innocent people confess after receiving the warnings.