In the

    United States Court of Appeals
                  For the Seventh Circuit
                      ____________________
No. 12-2627
UNITED STATES OF AMERICA,
                                                    Plaintiff-Appellee,

                                  v.

ERNEST CLARK,
                                                Defendant-Appellant.
                      ____________________

              Appeal from the United States District Court
                 for the Eastern District of Wisconsin.
                No. 11-CR-30 — J. Phil Gilbert, Judge.
                      ____________________

      ARGUED JANUARY 7, 2014 — DECIDED JUNE 6, 2014
                      ____________________

    Before WOOD, Chief Judge, and POSNER and KANNE, Cir-
cuit Judges.
    WOOD, Chief Judge. Ernest Clark’s string of bank robberies
in the Milwaukee area attracted the attention of both state
and federal authorities. When both sovereigns decided to
prosecute Clark around the same time, the occasionally deli-
cate practices designed to coordinate the timing and pursuit
of state and federal prosecutions came into play. Though
Clark does not contest the facts underlying his federal con-
2                                                   No. 12-2627

victions, he argues that violations of the Speedy Trial Act
and the Interstate Agreement on Detainers require us to re-
verse his convictions. We conclude, however, that nothing
was amiss, and affirm the judgment of the district court.
                                 I
    Clark committed six armed bank robberies in the greater
Milwaukee area between October 2008 and August 2010. The
details of these crimes are irrelevant to his appeal, but the
timing of his apprehension and prosecution is important. On
October 8, 2010, Milwaukee police officers who knew that
Clark was wanted in connection with at least one of the rob-
beries attempted to pull him over. He fled and was appre-
hended only after a two-mile vehicle chase. Four days later,
he was charged in Milwaukee County Circuit Court with the
state felony of eluding an officer, see Wis. Stat. § 346.04(3).
That same day, federal authorities brought armed robbery
charges against him in the Eastern District of Wisconsin for
the last in his string of six bank robberies, namely, that of the
Pyramax Bank in Milwaukee on August 18, 2010. On the ba-
sis of that complaint, a magistrate judge issued a warrant for
Clark’s arrest. The FBI used that warrant to file a detainer
against him with the Milwaukee County Sheriff, who was
holding him in state custody on the eluding charge.
     Clark was convicted of the state offense and sentenced to
seven months in jail on November 23, 2010. While he was
still incarcerated for that crime, a federal grand jury returned
a twelve-count indictment against Clark on February 8, 2011,
charging him with armed bank robbery in violation of 18
U.S.C. § 2213(a) and (d), and using a firearm in furtherance
of an armed bank robbery in violation of 18 U.S.C.
§ 924(c)(1)(A)(ii) for each of his six bank robberies. He was
No. 12-2627                                                   3

arraigned seventeen days later, on February 25. Following a
four-day jury trial he was convicted on all counts, and his
motion for a judgment of acquittal under Federal Rule of
Criminal Procedure 29(c) was denied. He was sentenced to
151 months’ imprisonment for each armed bank robbery
count, to run concurrently, and to 300 months’ imprisonment
for each § 924(c) count, to run consecutively, for an aggregate
sentence of a whopping 1,951 months in prison.
   Clark has appealed, but his arguments do not focus on
the facts underlying the convictions or the length of the sen-
tence. Instead, he asserts that his prosecution violated the
Speedy Trial Act and the Interstate Agreement on Detainers,
and that some of the evidence against him was procured in
violation of the Fourth Amendment. We consider these
points in turn.
                               II
    Clark first argues that his conviction must be reversed
because his prosecution failed to comply with the timing
provisions of the Speedy Trial Act, 18 U.S.C. § 3161(b). We
review legal questions about the application of the Speedy
Trial Act de novo, and credit the district court’s factual find-
ings on any issues related to the Act’s application unless they
are clearly erroneous. United States v. King, 

, 797
(7th Cir. 2003).
     Under the Sixth Amendment, “In all criminal prosecu-
tions, the accused shall enjoy the right to a speedy and pub-
lic trial … .” U.S. CONST. amend. VI. The purpose of the right
is to “limit the time during which criminal charges are hang-
ing over a person's head unresolved.” United States v. Janik,

, 542 (7th Cir. 1983). In order to “make effective”
4                                                   No. 12-2627

this speedy trial guarantee, Congress passed the Speedy Tri-
al Act of 1974, Pub. L. No. 93-619, 88 Stat. 2076. See S. Rep.
No. 93-1021, at 1 (1974). The part of the Act relevant to this
appeal provides:
    Any information or indictment charging an individu-
    al with the commission of an offense shall be filed
    within thirty days from the date on which such indi-
    vidual was arrested or served with a summons in
    connection with such charges.
18 U.S.C. § 3161(b).
     The thirty days do not begin to run because of just any
arrest based on the conduct that ultimately supports the fed-
eral prosecution. Two limitations are at work. First, because
the states and the federal government are distinct sovereigns
for purposes of criminal prosecutions, the speedy trial pro-
tections of the federal statute apply only to arrests made for
federal charges—an arrest by a state officer on a state charge
does not start the statutory clock. 

. Sec-
ond, even an arrest by the federal authorities is insufficient if
the person is immediately released without any federal
charges being filed. 

Amendment speedy trial
right, from which the Speedy Trial Act draws its substance,
applies only to persons who are formally accused of a crime.
United States v. MacDonald, 

, 16 (1982). To qualify as
a federal arrest and trigger the 30-day period under the Act,
therefore, an arrest must be for the purpose of bringing fed-
eral charges, and charges must be “pending” when the per-
son is arrested. Someone who is only the target of a criminal
investigation has no right to have the government wrap up
its investigation quickly and bring charges, even if the target
is aware of the investigation. See id.; see also United States v.
No. 12-2627                                                    5

Samples, 

, 301 (7th Cir. 1983)(explaining that the
Sixth Amendment right and the statute apply “only to an ac-
cused”).
    Clark suggests two alternative dates as triggers for his
speedy trial clock, either of which would provide him the
relief he seeks. The earlier one is the day he was arrested by
Milwaukee police: October 8, 2010. Although he was arrest-
ed by state police, he maintains (vaguely) that the arrest
“was involved with” federal charges. His point seems to be
that the state officers arrested him with the ulterior motive of
bringing federal charges, either because they were acting as
the federal government’s agents at the time or because the
arrest was always meant to lead to federal prosecution.
    If an agency relationship were present this might be a
closer point, but Clark has no proof of agency. He bases his
argument entirely on testimony by the arresting officer that
his armed bank robbery was the reason for his arrest. He
then reasons that the fact that his ultimate trial in federal
court was on armed robbery charges amounts to proof that
his original arrest was on this charge. Yet the record is de-
void of any indication that the federal government was in-
volved at all in his prosecution at the time of the October 8
arrest, let alone that it instructed the state authorities to ar-
rest Clark or that it wanted him arrested so that it could ini-
tiate proceedings against him. Wisconsin unsurprisingly has
its own laws against armed robbery. See Wis. Stat.
§ 942.32(2). The decision of state officials to arrest someone
because he is wanted for conduct in violation of state law
does not force the federal government to initiate whatever
proceedings it might bring for the same underlying conduct
6                                                  No. 12-2627

at the same moment. The federal arrest necessary to trigger
the Speedy Trial Act did not occur on October 8.
    Clark’s second theory is that his speedy trial clock started
with the filing of the federal complaint and detainer with the
Milwaukee County Sheriff on October 12. He maintains that,
because he would have been brought into federal custody
that day if his state charges had vanished, these were the
functional equivalent of a federal “arrest or summons” to
initiate the 30-day period under the Speedy Trial Act.
    Several of our sister circuits have considered the argu-
ment that some form of accusatory document short of an in-
dictment—such as a complaint or detainer—is sufficient to
trigger the 30-day rule even if the accused has not been
brought into federal custody. They have almost uniformly
rejected it. The key to their reasoning is that there must be a
“federal deprivation of liberty” in connection with the de-
tainer or complaint before the accused is entitled to the pro-
tections of the Speedy Trial Act. United States v. Bloom, 

, 491 (2d Cir. 1989).
    A good example is the Eleventh Circuit’s decision in
United States v. Shahryar, 

(11th Cir. 1983) (per
curiam). Shahryar was first arrested on state arson charges,
and four days later a federal complaint was filed against him
for violations of wire fraud and firearms statutes. A federal
magistrate judge issued an arrest warrant and set a tempo-
rary bond; the warrant was lodged with state authorities as a
detainer. Shahryar posted bond on his state charges five
months after that, and was immediately taken into federal
custody. He was indicted by a federal grand jury two weeks
later. 

The court reasoned that the Act was not vio-
lated because the 30-day period did not begin until Shahryar
No. 12-2627                                                      7

posted bond on his state charge and entered federal custody.
Until that point, “there was no federal arrest of the appellant
and no taking of him into federal custody.” 

The
court explained that “[f]or the time limit of the Act to com-
mence a person must be held for the purpose of answering
to a federal charge.” 

It grounded its rule in the
doctrine of dual sovereignty, “deeply rooted concepts of fed-
eralism,” and “common sense.” 

    Other circuits similarly have held that there must be both
a federal charge and federal custody for the Speedy Trial Act
to apply; neither one alone suffices. See United States v. Bag-
ster, 

, 611 (10th Cir. 1990); United States v. Johnson,

, 312 (5th Cir. 1987); see also 

(relying on Shahryar to hold that when a person is arrest-
ed and released on the same day without formal charges,
and a complaint is filed after that release, the Speedy Trial
Act does not apply until that person is brought back into
custody). The only apparent holdout is the Eighth Circuit. It
has not directly addressed whether the Speedy Trial Act’s
pre-indictment timing provision is triggered by a complaint
alone, but it seemed to assume so in United States v. Solomon,

, 1251–52 (8th Cir. 1982). We note, however, that
this issue was not squarely presented in Solomon, and that
the court there was deciding if an arrest alone triggered the
Speedy Trial Act, or if the 30-day period began only when a
complaint issued. 

have not had the occasion to address the re-
lation between a federal detainer and the Speedy Trial Act,
we have held that recapture of an escaped prisoner does not
begin the speedy trial clock for new charges stemming from
the escape. See United States v. Zukowski, 

, 177
8                                                  No. 12-2627

(7th Cir. 1988). The defendant there argued that bringing
him back into federal custody was necessarily an “arrest” for
the purposes of the Act, but we reasoned that there was no
“arrest” because “the restraints imposed on the prisoner
[were] not new but [were] based on his original conviction.”

rights, we said, “do not protect an individual
against unreasonable delay before accusation by arrest or the
filing of charges.” 

It is true that we were constru-
ing the constitutional rather than the statutory speedy trial
right in that portion of the opinion. Nonetheless, because the
Speedy Trial Act uses an arrest or summons as the trigger for
the 30-day requirement, the principle is the same.
   Returning to Clark’s case, it does not take a significant ex-
tension of Zukowski to find that he was not “arrested” for
purposes of the Speedy Trial Act when the detainer was
lodged with Milwaukee police on October 12. The sine qua
non of an arrest for purposes of the Act is the act of unam-
biguously bringing the accused into federal custody. Clark’s
detainer argument falls short of the mark for two reasons: he
was not brought into federal custody at all on that date, and
he was not subject to any restraints beyond those that Wis-
consin already was imposing legitimately. We accordingly
agree with our sister circuits that have held a “federal depri-
vation of liberty” is required to start the 30-day period under
the Speedy Trial Act. Since neither of the dates Clark pro-
poses initiated federal custody, we find no violation of the
Act in Clark’s prosecution.
                                III
   Clark’s second challenge to his conviction requires us to
consider another aspect of federal-state coordination on
prosecuting criminals: the Interstate Agreement on Detainers
No. 12-2627                                                   9

(IAD), 18 U.S.C. app. 2 § 2. The district court’s application of
the IAD presents a question of law that we review de novo.
United States v. Jones, 

, 646 (7th Cir. 2006).
    The IAD is designed to ensure cooperation both between
member states and between states and the federal govern-
ment. The federal government and every state except Louisi-
ana and Mississippi have signed the compact. The portion of
the IAD at issue here is Article V, which governs when and
how a state (or the federal government) wishing to prosecute
a person already imprisoned in another state may take tem-
porary custody of the prisoner for the purpose of pursuing
its charges. Article V(d) provides in relevant part:
   The temporary custody referred to in this agreement
   shall be only for the purpose of permitting prosecu-
   tion on the charge or charges contained in one or
   more untried indictments, informations, or com-
   plaints which form the basis of the detainer or detain-
   ers or for prosecution on any other charge or charges
   arising out of the same transaction.
18 U.S.C. app. 2 § 2 art. V(d).
    Clark maintains that his prosecution violated Article V(d)
and that the appropriate remedy is dismissal of his convic-
tions. To evaluate his claim, we first must examine the de-
tainer the FBI filed in his case. The complaint issued by the
federal magistrate judge on October 12, 2010, contained two
charges: armed bank robbery and brandishing a firearm in
furtherance of such a robbery, both based on the last of
Clark’s six bank robberies. (The record does not reveal
whether state or federal authorities were aware of Clark’s
connection with the other five robberies at that time.) The
10                                                 No. 12-2627

detainer was lodged with the Milwaukee County Sheriff on
the basis of those two charges. It remained in place, without
amendment, throughout his state prosecution, until he was
indicted on February 8, 2011, on the full complement of fed-
eral charges for which he was ultimately convicted. Clark
argues that because only the two counts in the complaint
“form[ed] the basis of the detainer” that led to his federal
custody, under Article V(d) the federal government could
prosecute him only on those counts. The IAD forbade, he as-
serts, the addition of other counts that were not part of the
complaint on which the detainer was based.
    The magistrate judge who first heard Clark’s argument
rejected it on the ground that Clark’s presence in federal
court was procured via a writ of habeas corpus ad prosequen-
dum rather than a written request for a prisoner transfer un-
der the IAD. See 18 U.S.C. app. 2 § 2 art. IV. But that distinc-
tion was rejected in the Supreme Court’s decision in United
States v. Mauro, 

(1978). There, while construing
Article IV of the IAD, the Court held that once the federal
government has filed a detainer, a writ of habeas corpus ad
prosequendum is itself a “written request for temporary cus-
tody” for the purposes of the IAD. 

Though that
case involved Article IV’s provisions governing the timing of
interstate prosecutions, there is no material difference in this
case between Article IV and Article V. As the Supreme Court
explained, proceeding by way of the writ “in no way reduces
the need for … prompt disposition of the charges underlying
the detainer.” 

(There is one difference between the
IAD process and the writ, however. As United States v. Pleau,

(1st Cir. 2012) (en banc), notes, the writ of habeas
corpus ad prosequendum is a stronger tool for the federal gov-
ernment, because a state governor has no power to refuse to
No. 12-2627                                                  11

honor the writ, while the governor does have power to dis-
approve a request for transfer under the IAD. 

6.)
    The district court spotted the Mauro error and instead
reasoned that the detainer was constructively amended by
the February 8 indictment; at that point, the court ruled, the
extra 10 counts of the indictment became additional bases
for the detainer. The government admits on appeal that this
reasoning is unprecedented and urges us in the alternative
to hold that any violation of the IAD in this case was harm-
less. Reversing Clark’s convictions based on such a small vio-
lation, it contends, would not serve the purposes of the IAD.
    Our take is slightly different from the district court’s. We
note at the outset that Article V(d) says temporary custody is
“only for the purpose of permitting prosecution” on an untried
charging document that forms the basis of the detainer. This
is different from temporary custody being “for the purpose
of prosecution only on charges that form the basis of the de-
tainer,” which the IAD does not say. Clark’s detainer was
based on the armed robbery and firearm-brandishing charg-
es arising out of the Pyramax Bank robbery. Under the IAD,
he could be transferred so that the receiving sovereign could
try him on those charges. The only wrinkle is that before the
transfer occurred, a superseding indictment was returned
that added more charges. Once he found himself in federal
custody, he had to stand trial for both the original charges
and the new ones, since they had been joined as Federal Rule
of Criminal Procedure 8(a) permits. Thus, upon his transfer,
Clark was tried both for the charges that formed the basis of
his original detainer and for the additional charges raised in
the superseding indictment.
12                                                No. 12-2627

    As long as the original charges are not lodged in bad
faith or fraudulently as a basis for obtaining a detainer, or
the later-added charges are not improperly delayed to avoid
the strictures of the IAD, we see nothing to prevent the addi-
tion of new charges. Suppose the government had taken
Clark’s view of the matter and had tried him only on the
charges that supported the detainer. Surely Clark does not
mean that the government was forever barred from bringing
the additional charges related to the other five bank rob-
beries. Once the first trial was over, the government could
have sought a new indictment (assuming no statute-of-
limitations problem) and proceeded with a new trial, maybe
even under a new detainer. But what sense would there be in
such a rule? The purpose of the IAD is to “prevent compet-
ing jurisdictions from transferring prisoners back and forth,
because such transfers can undermine the prisoner’s right to
a speedy trial and the rehabilitative process of the system in
which the prisoner is currently serving a sentence.” United
States v. Jones, 

, 448 (3d Cir. 1991).
    Article V(d) in particular was designed to “safeguard the
prisoner’s right to contest the transfer by appealing to the
Governor of the sending state pursuant to Article IV(a) of the
IAD,” which allows governors to disapprove requests for
temporary custody within 30 days. Cooney v. Fulcomer, 

, 44 (3d Cir. 1989); but see 

. That purpose
is fully served in situations such as this one, where the ulti-
mate transfer is for the purpose of trying the original charges
that formed the basis of the detainer, and the additions to the
indictment are imposed before the prisoner is transferred.
The system Clark seems to be proposing would actually
work at cross-purposes with the IAD, insofar as it would re-
quire the federal government to send Clark back to Wiscon-
No. 12-2627                                                    13

sin so that it could re-indict him and lodge a new detainer
before bringing him back to federal custody for a trial.
    Clark has not alleged any prejudice that flowed from the
fact that the government did not file a new detainer after the
indictment. Relying on Alabama v. Bozeman, 

(2001), he argues instead that the doctrine of harmless error
is categorically unavailable for violations of the IAD. Bo-
zeman held that there is no de minimis exception to the anti-
shuttling provision of Article IV(e) of the IAD, which re-
quires that a transferred prisoner be tried in the receiving
state before being returned to the sending state. The wording
of Article IV(e) was central to the Court’s reasoning: it says
that if a prisoner is returned to the sending state before trial,
the indictment “shall not be of any further force or effect, and
the court shall enter an order dismissing the same with prej-
udice.” 

(quoting 18 U.S.C. app. 2 § 2
art. IV(e)). This language, the Court said, left no room for an
exception for de minimis violations.
    Article V(d), in contrast, does not mandate any conse-
quence. This distinguishes it not only from Article IV(e), but
also from other parts of Article V, such as Article V(c) (direct-
ing courts in the jurisdiction where an indictment was issued
to dismiss it with prejudice if the jurisdiction “refuses or fails
to accept temporary custody” of a prisoner pursuant to a de-
tainer). The drafters knew how to require dismissal of an ac-
tion for failure to comply with provisions of the IAD, and
their decision not to impose that draconian consequence for
violations of Article V(d) should not be disregarded.
    Automatic reversal of a criminal conviction is generally
imposed only when the type of error at issue is “so intrinsi-
cally harmful as to require automatic reversal (i.e., “affect
14                                                   No. 12-2627

substantial rights”) without regard to their effect on the out-
come.” Neder v. United States, 

, 7 (1999). While
Congress can modify that rule by statute for particular types
of error—as it did in Article IV(e) of the IAD—in the absence
of such a legislative command, we subject most alleged er-
rors to harmless error analysis. See 

Harmless error
analysis seems particularly appropriate here, where even if
there was a violation, Clark has not identified a single reason
why he was harmed by the supposed violation of the IAD or
why the anti-shuttling purpose of the agreement was un-
dermined. Accordingly, even though we do not think the
IAD was violated, we are satisfied that if it was, harmless
error analysis applies to the type of claim Clark is raising,
and any error here was harmless.
                                 IV
    Clark’s final claim is that a search warrant issued for his
DNA while he was in the Milwaukee County jail failed to
satisfy the Fourth Amendment’s particularity requirement.
Shortly after Clark’s initial arrest, a state judicial officer is-
sued a search warrant to conduct a buccal swab for DNA
from “Ernest F. Clark, B/M [black male], 11-01-1968 at the
City of Milwaukee Prisoner Processing Center.” The DNA
obtained from this search was later linked to a mask and pair
of gloves found during the investigation of Clark’s last bank
robbery. Clark argues that this evidence should have been
suppressed because his birthdate (at least using the month-
day-year ordering common in the United States) is 01-11-
1968 (i.e. January 11, not November 1).
   The particularity requirement of the Fourth Amendment
protects against open-ended warrants that leave the scope of
the search to the discretion of the officer executing the war-
No. 12-2627                                                    15

rant, or permit seizure of items other than what is described.
E.g., Stanford v. Texas, 

, 485–86 (1965). A warrant
satisfies this requirement if it leaves nothing about its scope
to the discretion of the officer serving it. Jones v. Wilhelm, 

, 462 (7th Cir. 2005).
    The warrant issued for Clark’s DNA was not unconstitu-
tionally vague. It correctly identified his name, race, sex, and
the location where he was housed, and contained what is at
most a transposition of the date and month of his birthday.
This small error (if it was one, and not just a rendering of his
birthday in the day-month-year order that is common
throughout much of the world and used by the U.S. military)
casts no doubt on the fact that Clark was the intended target.
It did not give the officer discretion to search anyone but
him, or to collect anything but Clark’s DNA by means of a
buccal swab. There was no confusion about who was to be
searched and what was to be collected; the Fourth Amend-
ment is not a bulwark against typos.
                              *   *   *
    We are satisfied that there were no statutory or constitu-
tional errors requiring reversal of Clark’s convictions. The
judgment of the district court is AFFIRMED.