United States of America v. Sagarsen Haldar

In the
United States Court of Appeals
For the Seventh Circuit
No. 13-1238
UNITED STATES OF AMERICA,
Plaintiff-Appellee,
v.
SAGARSEN HALDAR,
Defendant-Appellant.
Appeal from the United States District Court for the
Eastern District of Wisconsin.
No. 10-CR-268 — Rudolph T. Randa, Judge.
ARGUED DECEMBER 3, 2013 — DECIDED APRIL 30, 2014
Before POSNER, MANION, and HAMILTON, Circuit Judges.
HAMILTON, Circuit Judge. A jury found Sagarsen Haldar
guilty of conspiring to defraud the United States by using his
position as leader of a Hare Krishna temple in Milwaukee to
obtain religious-worker visas for people who were not actually
religious workers. Haldar appeals his conviction and seeks a
new trial by raising three issues, none of which he raised in the
district court. He argues: (1) certain statements from the
prosecutor and a government witness improperly called into

-- 1 of 17 --

2 No. 13-1238
question the validity of his temple and were unfairly prejudi-
cial under Federal Rule of Evidence 403; (2) the prosecutor
misrepresented testimony during his closing argument and
relied on facts outside the record; and (3) the district court on
its own initiative should have instructed the jury not to
scrutinize the religious qualifications of the visa recipients.
Haldar cannot satisfy the stringent “plain error” standard that
he must meet to win a new trial when raising such issues for
the first time on appeal. We affirm his conviction.
I. Factual and Procedural Background
Haldar is an Indian citizen who came to the United States
in 1999 and has been a permanent resident since 2006. A few
years after his arrival he founded the Gaudiya Vaisnava
Society, Inc., known as GVS-Milwaukee. Gaudiya Vaisnava (or
Vaishnava) is another name for the Hare Krishna religion.
From 2004 to 2007, GVS-Milwaukee sponsored twenty-five
applicants for religious-worker visas known as “R-1 visas,” see
8 C.F.R. § 214.2(r)(1), seventeen of which were approved.
For reasons not fully explained in the record, by mid-2007
the State Department had advised the Department of Home-
land Security (DHS) that GVS-Milwaukee might be involved
in visa fraud. DHS had also received an anonymous tip along
the same lines. As a result, the agency began an investigation
that included temple visits, surveillance, searches of Haldar’s
luggage on international trips, and interviews with GVS-
sponsored visa recipients. Haldar was charged in 2010 with
conspiracy to defraud the United States under 18 U.S.C. § 371.
At trial, DHS agents testified about their investigation. One
agent, Ali Lubbad from the fraud detection arm of U.S.

-- 2 of 17 --

No. 13-1238 3
Citizenship and Immigration Services (USCIS), described the
investigation’s inception and initial stages. He mentioned the
State Department’s suspicions of visa fraud and that DHS had
received a tip about GVS-Milwaukee. As he was about to
describe the content of the tip, defense counsel objected to the
testimony as hearsay. Before the judge could rule, the prosecu-
tor interjected, “Actually I think I can take care of it this way.
Without going into the substance of the letter, is it fair to say
you received a tip letter that indicated that you should look
into GVS?” Lubbad responded, “Yes.” Haldar’s lawyer did1
not renew his hearsay objection or object to this testimony on
another basis.
Agent Lubbad also told the jury that, in any event, DHS
routinely investigates organizations sponsoring R-1 visa
applicants because up to 33 percent of all applications are
sponsored by “[c]ompletely fraudulent” organizations,
meaning there was actually “no organization” at all. Defense
counsel did not object to Lubbad’s testimony on this point.
Agent Lubbad went on to describe a surprise visit he made
to the main GVS temple location. During that visit, Haldar
seemed anxious for him to leave because no religious activity
Although Haldar does not raise the hearsay problem with Lubbad’s
1
testimony on appeal, such “course of investigation” evidence is only rarely
relevant. When the prosecution offers this type of testimony, it invites
witnesses to give inadmissible hearsay and invites appellate courts to
reverse on evidentiary or Confrontation Clause grounds. See Jones v.
Basinger, 635 F.3d 1030, 1044–48 (7th Cir. 2011) (ordering habeas corpus
relief where “course of investigation” evidence violated defendant’s
Confrontation Clause rights).

-- 3 of 17 --

4 No. 13-1238
was apparent: “He was encouraging me to leave that location.
Kept saying I have another Temple at Ramsey. You know, it’s
better to see. And it seemed—my feeling was that he wanted
to leave that location as soon as possible.” Following Haldar to
the Ramsey Avenue temple location in a separate car, Lubbad
saw him talking on his cell phone during the drive. Lubbad
testified that he thought the people he encountered at the
Ramsey location, who were loudly performing music, looked
as if they were expecting a visit. Haldar’s lawyer objected to
this testimony as speculative, to which the prosecutor re-
sponded that Lubbad could testify about his own perception of
the situation. The court allowed the testimony to stand, and
Haldar’s lawyer offered no other objection.
Near the end of Agent Lubbad’s direct examination, the
prosecutor asked how DHS keeps track of organizations that
sponsor religious workers. Lubbad answered in part by once
more volunteering his statistic about the rate of fraud in R-1
visa applications, this time saying that “a very significant 33
percent [of applications] were – had some elements of fraud in
them.” Again Haldar’s lawyer made no objection.
Another agent from DHS, Scott Engelhardt, who became
involved with the investigation after it was transferred from
USCIS to U.S. Immigration and Customs Enforcement (ICE),
testified that he twice supervised the inspection of Haldar’s
luggage at O’Hare Airport when Haldar returned from trips to
India. The luggage contained completed R-1 visa applications
and supporting documents, including numerous passports,
letters to U.S. consulates, and training certificates that Haldar
had signed. Engelhardt also testified that he interviewed three
R-1 visa recipients sponsored by GVS-Milwaukee who were

-- 4 of 17 --

No. 13-1238 5
working in secular jobs rather than for Haldar or GVS-Milwau-
kee or any other religious organization. These men, Engelhardt
explained, agreed to cooperate with the investigation.
The testimony of the three visa recipients, along with covert
audio recordings of two meetings one of them had with
Haldar, formed the heart of the government’s case. All three
men testified that Haldar had agreed to help them obtain visas
in exchange for payments ranging from $20,000 to $30,000. He
had then fulfilled his end of the bargain by making GVS-
Milwaukee their sponsoring organization and arranging for
them to be provided with training certificates and other
supporting documents by a colleague in India. The visa
recipients also testified that—although they had received
enough training before leaving India to enable them to pass for
Hare Krishna priests if interviewed by U.S. officials—they had
always intended, as Haldar knew full well, to work in secular
jobs in the United States.2
The recordings were made when one of the visa recipients,
Gurpreet Singh, wore a recording device to two meetings with
Haldar. According to the translated transcript of the first
meeting, Singh explained to Haldar that his visa would soon
expire and he needed help getting an extension. Haldar
responded by scolding Singh for having failed to pay all he
had promised for the visa: “You people did not pay my
money … it has been such a long time. … You should just think
A person need not be a priest or minister in his or her religion to receive2
an R-1 visa, but the visa applicants here pretended to be priests. “Religious
worker” is a broader category, though the requirements for qualifying are
quite demanding. See 8 C.F.R. §§ 214.2(r)(1)(iii), (r)(3).

-- 5 of 17 --

6 No. 13-1238
about … what a difficult job for me it was to get your job done.
…You should have paid my money … I have asked so many
times for it.” Singh promised to pay Haldar what he already
owed and also to pay more for the visa extension. He then
gave Haldar $500. Haldar laughed and said, “You owe me
$10,000 and you are paying me $500 only,” but he agreed to
accept installment payments and to assist with the visa
extension. At the second meeting, Haldar accepted another
$500 payment.
On cross-examination, Singh and another visa holder,
Kashish Chopra, admitted that their removal from the United
States had been deferred so that they could testify against
Haldar. Chopra also admitted that he had gotten married for
the sole purpose of staying in the United States. The third man,
Sanjiv Kumar, admitted that he initially denied to investigators
that he had ever paid Haldar any money. And all three
answered in the affirmative when asked if they would “do
anything” to stay in the United States. (None of the three had
been charged with a crime when they testified, but none had
been promised immunity either.)
In his closing argument, the prosecutor summed up the
evidence against Haldar, stressing the recordings and the visa
recipients’ testimony that Haldar was a party to the fraud.
Haldar’s lawyer countered that Haldar was himself a victim,
deceived by the visa recipients. He argued that the visa
recipients’ testimony about Haldar’s knowledge was not
credible. The three could not be believed on any subject, the
lawyer contended, because they had admitted on cross-
examination that they “would do whatever it takes” to stay in
the United States. Yes, he conceded, “This case does involve

-- 6 of 17 --

No. 13-1238 7
fraud. It just doesn’t involve fraud on the part of Mr. Haldar.
It involves fraud on the part of three guys that got caught and
are doing whatever it takes to stay here.”
II. Analysis
The jury found Haldar guilty, and the district court
sentenced him to 37 months in prison. On appeal Haldar
argues that three aspects of his trial call for reversal and a new
trial. Because Haldar never raised these issues in the district
court, he has forfeited any appellate claim based upon them
and our review is only for “plain error.” A plain error is an
error that is obvious, affected the defendant’s substantial
rights—meaning that the defendant likely would have been
acquitted otherwise—and seriously affected the fairness,
integrity, or public reputation of the judicial proceedings.
United States v. Jones, 739 F.3d 364, 368 (7th Cir. 2014). In effect,
a finding of plain error says that an error was so obvious and
prejudicial that the district judge should have intervened
without being prompted by an objection from defense counsel.
See United States v. Alexander, 741 F.3d 866, 870 & n.1
(7th Cir. 2014).
A. Questioning the Religious Legitimacy of GVS-Milwaukee
During his opening statement, the prosecutor said that
Haldar “identified himself” as GVS-Milwaukee’s founder and
president and “held himself out” as holding various other
titles. Haldar argues that this unfairly insinuated that he was
a charlatan and that GVS-Milwaukee was a sham religious
organization. This insinuation was reinforced, Haldar says, by
the testimony of Agent Lubbad, who not only said that the
people he saw at the second temple location were putting on

-- 7 of 17 --

8 No. 13-1238
a show for him but also suggested a high likelihood of fraud by
citing the statistic that one-third of all R-1 visa applications
were sponsored by “completely fraudulent” organizations or
had “some elements of fraud” in them. Haldar argues that the
prosecutor’s statements constituted prosecutorial misconduct
and that those statements and Agent Lubbad’s testimony were
substantially more prejudicial than probative, violating Federal
Rule of Evidence 403.
Haldar’s claim of prosecutorial misconduct is based on his
belief that any questioning of GVS-Milwaukee’s legitimacy was
barred by a promise the prosecutor made before the jury was
empaneled. The purported promise came in the context of a
request from Haldar’s lawyer that potential jurors be asked
during voir dire about their religious beliefs and views on
religious freedom, with the goal of identifying anyone who
might be intolerant of Hare Krishnas. The lawyer explained:
Hare Krishnas historically in this country have had
negative connotation. From movies where they’re
seen—for example, the movie Airplane. A recurring
theme. They’ve been made fun of. They’ve been the
subject of controversy throughout the 60’s, the 70’s.
And depending upon the age of the jurors they
could have various opinions on the Krishnas them-
selves from prior experience. And I think we at least
need to address their understanding of whether or
not they have those biases. And a big question at the
end of the day is, is this a bona fide religious organi-
zation? That’s part of the require-ment [for obtaining
a religious-worker visa], and I think we’re going to
get into it.

-- 8 of 17 --

No. 13-1238 9
The judge responded that he did not think the government
was disputing whether “the GVS … was an established
religious organization,” and the prosecutor confirmed, “No,
we’re not disputing that.” The judge then explained that he
would ask prospective jurors about any prejudice against the
Hare Krishna religion but would not inquire into their religious
beliefs.
We accept the premise that a prosecutor who promises not
to raise certain issues during trial but nonetheless does so—at
least without warning and an opportunity for the defense to
object—may engage in misconduct. Cf. United States v. Melvin,
730 F.3d 29, 32 (1st Cir. 2013) (prosecutor’s use of proffered
information at trial after assuring defendant that he would not
so use it violated due process). But we cannot conclude that
such a promise was broken in this case, let alone that there was
a plain error.
First, an assurance regarding GVS-Milwaukee would not
necessarily extend to Haldar personally. He was being charged
with conspiracy to defraud the United States, so attacks on his
sincerity or integrity could not have been a surprise. Second,
the prosecutor’s statements about Haldar’s use of religious
titles were not followed by an assertion that he did not hold
those titles; the implication was, at worst, that the prosecutor
did not know whether he actually held them. Third, but most
important, whether the prosecutor was making an assurance
regarding GVS-Milwaukee specifically or about Hare Krishnas
in general is not clear from the transcript, but the absence of
any objection from Haldar at trial weighs heavily in favor of
concluding that the prosecutor broke no promise.

-- 9 of 17 --

10 No. 13-1238
Haldar also argues that the prosecutor’s opening statement
and Agent Lubbad’s testimony were substantially more
prejudicial than probative and so should have been excluded
under Federal Rule of Evidence 403. Calling into question
Haldar’s religious sincerity, he contends, inflamed the jurors
against him.
Accusations that a person is irreligious or has acted
contrary to a religion’s teachings can be inflammatory. See,
e.g., Slagle v. Bagley, 457 F.3d 501, 518 (6th Cir. 2006) (baseless
implication from prosecutor that defendant did not pray was
designed “to inflame the passions of the jury” but affirming
denial of habeas corpus relief); United States v. Giry, 818 F.2d
120, 133–34 (1st Cir. 1987) (prosecutor’s comparison of defen-
dant’s actions “to Peter’s denial of Christ” was “an irrelevant
and inflammatory appeal to the jurors’ private, religious
beliefs” but not a plain error). And sometimes such accusations
are highly prejudicial, as in Nichols v. American Nat’l Ins. Co.,
154 F.3d 875, 885, 890 (8th Cir. 1998), which reversed a defense
verdict where the jury learned that the plaintiff had an abor-
tion “even though it was against her religion,” a piece of
information that “increased the likelihood that the jury would
view her as immoral and not worthy of trust and reach its
verdict on such basis,” requiring a new trial.
But the prosecutor in Haldar’s case did not question his
faith, nor did the prosecutor criticize the Hare Krishna religion.
While the general theme of the prosecution undeniably was
that Haldar used his position as the leader of GVS-Milwaukee
to carry out the fraud, this theme was not improper. Haldar’s
misuse of his position was the essence of the crime he was

-- 10 of 17 --

No. 13-1238 11
charged with committing. Nothing the prosecutor said during
his opening statement was unfairly prejudicial.
Aspects of Agent Lubbad’s testimony are more troubling.
Haldar focuses on Lubbad’s claim that one-third of R-1 visa
applications are fraudulent, his reference to a tip about GVS-
Milwaukee, his description of the first temple location he
visited as devoid of religious activity, and his stated belief that
the people at the second temple were putting on a show for his
benefit.
In addition to stressing religion’s special inflammatory
potential, Haldar argues that Lubbad’s testimony was equiva-
lent to the testimony that we found unduly prejudicial in
United States v. Klebig, 600 F.3d 700, 710 (7th Cir. 2010), where
the defendant was charged with firearms offenses and the jury
was told that the defendant had a sign on his door warning
that nothing inside was “worth dying for.” This testimony in
Klebig, combined with photographs of the defendant’s excep-
tionally cluttered home, had no probative value and instead
painted the defendant as a “perhaps unbalanced man” and a
“gun nut,” inviting the jury to convict him based on fear or
dislike rather than the evidence of a crime. Id. at 712–13.
Haldar also draws a comparison to United States v. Wolf,
787 F.2d 1094, 1098 (7th Cir. 1986), where irrelevant innuendo
implied that the defendant in a Mann Act case once had a
venereal disease, and to United States v. DeGeratto, 876 F.2d 576,
583–86 (7th Cir. 1989), where the defendant, charged with
crimes related to the interstate transport of stolen property,
was painted as a dangerous loan shark who was also running
a prostitution ring. In each case, we concluded that the

-- 11 of 17 --

12 No. 13-1238
insinuations constituted unfairly prejudicial attacks on the
defendant’s character.
Agent Lubbad’s testimony was not as likely to instill fear in
the jurors as the testimony in Klebig or as likely to inflame them
as the testimony about abortion could have in Nichols. Nor do
we think Haldar’s character was maligned unfairly as in Wolf
and DeGeratto. Still, Lubbad’s provision of the statistic about
R-1 visa fraud was both irrelevant and unfair. “Evidence is
unfairly prejudicial if it induces the jury to decide the case on
an improper basis rather than on the evidence presented.”
United States v. Conner, 583 F.3d 1011, 1025 (7th Cir. 2009).
Telling the jurors that a large proportion of R-1 applications are
fraudulent encouraged them to conclude, without any evi-
dence specific to this case, that the applications Haldar spon-
sored had a strong chance of being fraudulent. The rest of
Agent Lubbad’s testimony was not especially prejudicial on its
own, but book-ended as it was by the irrelevant and prejudicial
statistic he provided twice, the testimony did exacerbate the
implication that the GVS-sponsored visas were likely to fall
into the supposedly massive pool of fraudulent applications.
A jury should no more have been told of that pool than it
should have been given overall conviction rates for the
category of crime that was charged.
The government tries to justify Agent Lubbad’s testimony
as a hedge against any accusation that the government had
unfairly singled out GVS-Milwaukee. No such accusation had
been made, though, and the government was playing with fire
when it chose to anticipate one with Lubbad’s flimsy assertion
of the one-third statistic.

-- 12 of 17 --

No. 13-1238 13
Haldar, however, did not object at trial to Agent Lubbad’s
testimony about the rate of R-1 visa fraud. In fact, he made no
relevant objection to any part of Lubbad’s testimony. (His
objection that Lubbad’s description of people at the second
temple was speculative and his abandoned hearsay objection
to the tip testimony did not preserve any objection based on
Rule 403. See, e.g., United States v. Blount, 502 F.3d 674, 677–78
(7th Cir. 2007) (an objection on one ground does not preserve
a challenge based on another).) As a result, our review is only
for plain error, Jones, 739 F.3d at 368, and Haldar cannot satisfy
that standard.
A finding of plain error requires not only that the error be
obvious but that the defendant probably would not have been
found guilty if not for the error. Id. Considering the testimony
of the three visa recipients and the recordings of Haldar
seeking payment for R-1 visa sponsorship from a person he
knew was not a religious worker, Haldar was highly likely to
have been convicted regardless of the statistic Agent Lubbad
offered. A defendant is guaranteed a fair trial but not a perfect
trial. Brown v. United States, 411 U.S. 223, 231–32 (1973); United
States v. Harris, 271 F.3d 690, 704 (7th Cir. 2001). Agent
Lubbad’s testimony was not so prejudicial as to amount to a
plain error that denied Haldar a fair trial.
B. Prosecutor’s Closing Argument
Haldar next argues that the prosecutor in his closing
argument improperly bolstered the testimony of the visa
recipients with evidence not in the record and misrepresented
some evidence that was in the record. The prosecutor’s
purported misconduct is contained in this passage, which

-- 13 of 17 --

14 No. 13-1238
followed the prosecutor’s acknowledgment that the visa
recipients’ testimony was inconsistent on some smaller points:
In fact, with respect to all of these different types of
information, that is the payment of a substantial fee
at Haldar’s direction, the payment in cash, the
payment in India in Rupees, the installments, the no
training as a Priest, all of those things, they remain
consistent all the way through. They were—that’s
what the witnesses told the Agents the first time
they met. That’s what they testified to here, with the
exception of Sanjiv Kumar, who as you heard from
him, held back the fact that he had paid for his
religious worker Visas. But he never held back on
the fact that this was fraud. On that, all these wit-
nesses have been definite from the beginning, and
continued to be here at the trial.
Haldar did not object to the prosecutor’s closing at the time.
A basic foundation of the law governing federal trials is
that lawyers, and especially prosecutors, should not refer to a
witness’s prior consistent statement if the earlier statement is
not in evidence, even though doing so does not automatically
call for reversal. See United States v. Tucker, 714 F.3d 1006, 1013
(7th Cir. 2013). The extra-record statements that Haldar
contends the prosecutor referred to are from the witnesses’
grand jury testimony, very little of which was introduced at
trial. But Haldar’s reading of the prosecutor’s words is unnatu-
ral. The prosecutor’s explicit reference to the visa recipients’
interviews with DHS agents made clear that the consistency he
was stressing was between those interviews, which the jury

-- 14 of 17 --

No. 13-1238 15
heard about, and the trial testimony. There was no misconduct
here, let alone misconduct that could support a finding of plain
error.
Like any other lawyer, a prosecutor is also obliged to avoid
misrepresenting the evidence that was presented to the jury.
United States v. Wolfe, 701 F.3d 1206, 1214 (7th Cir. 2012).
Haldar contends that the prosecutor misstated the evidence in
his case by telling the jury in his closing argument that the visa
recipients agreed that “this was fraud” and that they had
received “no training” to be priests. The witnesses, Haldar
points out, never used the word “fraud” themselves, and they
explained that they were at least trained how to behave like
priests.
Considering the prosecutor’s statements in context, see
United States v. Roe, 210 F.3d 741, 747 (7th Cir. 2000), we
conclude that he did not misrepresent the visa recipients’
testimony. Although the witnesses did not say “fraud,” even
Haldar’s lawyer candidly acknowledged in his own closing
argument that what they described was fraud. The prosecu-
tor’s argument was certainly fair comment on the evidence.
And the prosecutor’s “no training” statement, while not
precise, was a permissible shorthand reminder of the wit-
nesses’ testimony that their training was meant to enable them
only to pass for priests rather than to work as priests. Again, we
find no prosecutorial misconduct.
C. Instruction on Religious Qualifications
Haldar never asked the court to instruct the jury that the
visa recipients were qualified religious workers, but he argues
now that the lack of such an instruction is reversible error. He

-- 15 of 17 --

16 No. 13-1238
relies in part on our decision in McCarthy v. Fuller, 714 F.3d 971
(7th Cir. 2013), where we reversed a district court’s ruling that
a jury could decide whether the defendant was a member of a
Roman Catholic religious order. The church denied that the
defendant was a member of a religious order, and we held that
“once the court has satisfied itself that the authorized religious
body has resolved the religious issue, the court may not
question the resolution.” Id. at 976. Haldar also cites Hosanna-
Tabor Evangelical Lutheran Church & Sch. v. EEOC, 132 S.Ct. 694
(2012), which recognized the implied “ministerial exception”
in federal employment discrimination statutes, for the general
proposition that the government cannot choose a church’s
ministers. Notably, Haldar cites no case requiring a court to
give an instruction about religious qualifications when no such
instruction was requested, and we have found none.
Some of our cases treat a failure to request a jury instruction
as a waiver of any argument that such an instruction should
have been given, see, e.g., MMG Financial Corp. v. Midwest
Amusements Park, LLC, 630 F.3d 651, 658 (7th Cir. 2011), while
others consider the argument merely to have been forfeited,
see, e.g., United States v. Johnson, 655 F.3d 594, 605 (7th Cir.
2011). But even if we treat Haldar’s argument as only forfeited
and subject to plain-error review rather than waived, we find
no error, plain or otherwise. The jury was told to decide
whether Haldar sponsored the visa applicants with the
expectation that they would be religious workers. Performing
that task would have been impossible if an instruction had
required the jurors to accept that the applicants were religious
workers simply because Haldar had managed to obtain
certificates for them. Our analysis would be different if the

-- 16 of 17 --

No. 13-1238 17
jurors had been asked to determine how much training is
required to become a Hare Krishna priest or whether the visa
recipients had been given enough, but no such questions were
posed to them. Haldar has not shown there was any error in
the jury instructions.
D. Cumulative Error
Haldar’s final argument is that even if none of the individ-
ual issues he has raised constituted plain error, they satisfy that
standard in the aggregate. To win reversal based on a
cumulative-error theory, Haldar would need to demonstrate
at least two errors and show that in combination they caused
his trial to be fundamentally unfair. United States v. Moore,
641 F.3d 812, 830 (7th Cir. 2011). Because we have found only
one minor error in Haldar’s trial (the admission of Agent
Lubbad’s statistic), we necessarily reject his cumulative-error
argument.
The judgment of the district court is AFFIRMED.

-- 17 of 17 --