In the
United States Court of Appeals
For the Seventh Circuit
____________________
No. 16-3848
WARREN JOHNSON, et al.,
Plaintiffs-Appellants,
v.
ADVOCATE HEALTH AND HOSPITALS CORPORATION, doing
business as ADVOCATE CHRIST MEDICAL CENTER,
Defendant-Appellee.
____________________
Appeal from the United States District Court for the
Northern District of Illinois, Eastern Division.
No. 1:14-cv-08141 — Manish S. Shah, Judge.
____________________
ARGUED SEPTEMBER 15, 2017 — DECIDED JUNE 8, 2018
____________________
Before MANION, ROVNER, and HAMILTON, Circuit Judges.
ROVNER, Circuit Judge. Employees of Advocate Health
and Hospitals Corporation (Advocate) claim that they were
treated unfairly based on their race. The district court grant-
ed Advocate’s motion for summary judgment, finding that
the plaintiffs failed to offer evidence necessary to support an
element of their claim. We agree with the district court on all
2 No. 16-3848
issues but the question of the hostile work environment, and
remand to the district court for a determination of that claim.
I.
Plaintiffs Warren Johnson, Robert Pannell, Kimberly
Scott-Murray, Annette Smith, and Sherry Young all claim
that they faced race discrimination at the hands of supervi-
sors when they worked as Environmental Service Techni-
cians (EVS techs) at Advocate. EVS techs perform work that
would traditionally be called janitorial work. They clean and
disinfect hospital rooms and common areas, make beds, and
the like. The EVS techs claim that they were treated unfairly
because of their race in violation of Title VII of the Civil
Rights Act of 1964, 42 U.S.C. § 2000e et. seq. and 42 U.S.C.
§ 1981. 1
In 2012, Advocate contracted with Aramark Healthcare
Support Services and reorganized the supervision and oper-
ation of the EVS department. Under the Service Agreement
between Advocate and Aramark, Aramark was responsible
for managing the EVS department while abiding by the poli-
cies of Advocate, including, among other policies, Advo-
cate’s non-discrimination policy. See, e.g., R. 62-6 at 6, 16,
Page ID 1941, 1951. Shortly thereafter, the plaintiffs claim
that Aramark-employed supervisors Susan Castillo, Chris-
topher Skalnik, and Mariusz Michalkowski engaged in dis-
criminatory acts against the plaintiffs. The claims of discrim-
ination include: (1) Johnson and Smith were paid less than
1 We generally have applied the same prima facie requirements to
discrimination claims brought under Title VII and section 1981. Hum-
phries v. CBOCS W., Inc., 474 F.3d 387, 403 (7th Cir. 2007), aff'd, 553 U.S.
442 (2008)
No. 16-3848 3
white EVS techs; (2) Pannell and Scott-Murray were denied
promotions and raises; (3) Plaintiffs were managed and dis-
ciplined more scrupulously than their non-African-
American co-workers, and terminated in a discriminatory
fashion; (4) African-American plaintiffs were given less de-
sirable and more strenuous assignments; (5) Aramark su-
pervisors subjected the plaintiffs to offensive and derogatory
racial comments, creating a hostile work environment.
The district court granted Advocate’s motion for sum-
mary judgment on all counts, concluding that the plaintiffs
did not experience severe or pervasive race-based harass-
ment, that there was no basis for employer liability, and that
the plaintiffs failed to demonstrate that racial animus moti-
vated the decisions to terminate Johnson, Scott-Murray and
Smith. Johnson v. Advocate Health & Hosps. Corp., No. 14 CV
8141, 2016 WL 5871489 (N.D. Ill. Oct. 7, 2016). As for the hos-
tile work environment claim, the lower court held that the
comments, although concerning, were too isolated, indirect,
and sporadic, and not so serious as to have affected the
plaintiffs’ working conditions. Id. at *8. The district court al-
so concluded that there was no basis for employer liability.
Id.
II.
The plaintiffs’ brief is awash in facts and controversies.
They claim that these numerous disputes and presentations
of conflicting evidence create genuine issues of material fact.
It is true that cases with jumbles of ostensibly disputed facts
often signal the need for a trial on the facts. See Payne v.
Pauley, 337 F.3d 767, 770 (7th Cir. 2003). Not all disputed
facts, however, are relevant and material. On summary
judgment we must view the facts and make all reasonable
4 No. 16-3848
inferences that favor them in the light most favorable to the
party opposing summary judgment. Parker v. Four Seasons
Hotels, Ltd., 845 F.3d 807, 814 (7th Cir. 2017). The following
common refrains in summary judgment cases are important
to recall in a case with so many factual recitations:
On summary judgment a court may not make
credibility determinations, weigh the evidence,
or decide which inferences to draw from the
facts; these are jobs for a factfinder. Rather, the
court has one task and one task only: to decide,
based on the evidence of record, whether there
is any material dispute of fact that requires a
trial. Summary judgment is not appropriate if
the evidence is such that a reasonable jury
could return a verdict for the nonmoving par-
ty. We must look therefore at the evidence as a
jury might, construing the record in the light
most favorable to the nonmovant and avoiding
the temptation to decide which party’s version
of the facts is more likely true. As we have said
many times, summary judgment cannot be
used to resolve swearing contests between liti-
gants.
Payne, 337 F.3d at 770 (internal citations and quotations
omitted). To defeat a motion for summary judgment, the
party opposing it must make a “showing sufficient to estab-
lish the existence of an element essential to the party’s case,
and on which that party will bear the burden of proof at tri-
al.” Celotex Corp. v. Catrett, 477 U.S. 317, 322 (1986). Summary
judgment is a critical moment for a non-moving party. It
must “respond to the moving party’s properly-supported
No. 16-3848 5
motion by identifying specific, admissible evidence showing
that there is a genuine dispute of material fact for trial. Grant
v. Trs. of Ind. Univ., 870 F.3d 562, 568 (7th Cir. 2017). Infer-
ences supported only by speculation or conjecture will not
suffice. Skiba v. Ill. Cent. R.R. Co., No. 17-2002, 2018 WL
1190856, at *7 (7th Cir. Mar. 8, 2018). Neither will the mere
scintilla of evidence. Grant, 870 F.3d at 571.
Although these common refrains are familiar, the task is
often easier to describe than to perform, and plenty of credi-
bility-weighing traps lay before a court, particularly in such
fact-intensive cases. See, e.g. Payne, 337 F.3d at 771. As our
review is de novo, we affirm the district court only when no
reasonable jury could have found for the plaintiffs. See, e.g.,
Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248 (1986); Roh v.
Starbucks Corp., 881 F.3d 969, 973 (7th Cir. 2018).
On top of the normal lattice of summary judgment de-
mands, we must also apply the constructs of employment
discrimination law. For years we have tangled with a “rat’s
nest of surplus tests” in employment discrimination cases—
struggling to pigeon hole evidence into the direct or indirect
method with various overlaying requirements of “convinc-
ing mosaics” and circumstantial or direct evidence. Ortiz v.
Werner Enters., Inc., 834 F.3d 760, 764–66 (7th Cir. 2016). Our
Circuit has now clarified the singular question that matters
in a discrimination case: “[W]hether the evidence would
permit a reasonable factfinder to conclude that the plaintiff’s
race, ethnicity, sex, religion, or other proscribed factor
caused the discharge or other adverse employment action.”
Ortiz, 834 F.3d at 765. “Evidence must be considered as a
whole, rather than asking whether any particular piece of
6 No. 16-3848
evidence proves the case by itself … . Relevant evidence
must be considered and irrelevant evidence disregarded.” Id.
Plaintiffs allege that the district court’s analysis failed to
comport with this new standard in Ortiz, but we disagree.
The district court did exactly as Ortiz demands and ignored
the old “convincing mosaic” language and cut straight to the
“bottom line and determine[d] whether there [was] evidence
from which a reasonable trier of fact could infer discrimina-
tion from Advocate’s actions as to each particular plaintiff.”
Johnson, 2016 WL 5871489, at *5 (N.D. Ill. Oct. 7, 2016), (cit-
ing Liu v. Cook Cty., 817 F.3d 307, 315 (7th Cir. 2016)).
Despite doing away with the need for separate tests and
“mosaics,” the well-known and oft-used McDonnell Douglas
framework for evaluating discrimination remains an effi-
cient way to organize, present, and assess evidence in dis-
crimination cases. David v. Bd. of Trs. of Cmty. Coll. Dist. No.
508, 846 F.3d 216, 224 (7th Cir. 2017) (noting that “Ortiz,
however, did not alter [t]he burden-shifting framework cre-
ated by McDonnell Douglas Corp v. Green, 411 U.S. 792
(1973)”). There is no magic to this test; it is merely one way
of culling the relevant evidence needed to demonstrate
whether a reasonable factfinder could conclude that an em-
ployer engaged in an adverse employment action based on
the plaintiff’s race or other proscribed factor. But because the
framework is helpful we use it to evaluate each of the plain-
tiffs’ claims by looking to see whether the plaintiffs (1) are
members of a protected class; (2) performed reasonably on
the job in accord with their employer[’s] legitimate expecta-
tions; (3) were subjected to an adverse employment action
despite their reasonable performance; and (4) similarly situ-
ated employees outside of the protected class were treated
No. 16-3848 7
more favorably by the employer. David, 846 F.3d at 225. All
of the plaintiffs are African-American and thus fall within a
protected class. Using this framework as an organizational
guide, we address each of the plaintiffs’ claims individually,
beginning with the plaintiffs’ claims that they received lower
pay than non-African-American EVS workers.
A. Pay disparity
Plaintiffs Johnson and Smith claim that they were paid
less than white associates because of their race. Plaintiff
Pannell also asserts that he did not receive “charge pay”—
extra pay for performing work outside of his job description.
To make these claims successfully, the plaintiffs needed to
produce evidence that similarly situated non-African-
American employees were treated more favorably. Reed v.
Freedom Mortg., 869 F.3d 543, 549 (7th Cir. 2017). Similarly
situated means “directly comparable” in all material re-
spects. Id. “The proposed comparator need not be identical
in every conceivable way, however, and courts must con-
duct a common-sense examination.” Id. (quoting Perez v.
Thorntons, Inc., 731 F.3d 699, 704 (7th Cir. 2013)). “The simi-
larly-situated inquiry is flexible, common-sense, and factual.
It asks ‘essentially, are there enough common features be-
tween the individuals to allow a meaningful comparison?’”
Coleman v. Donahoe, 667 F.3d 835, 841 (7th Cir. 2012). In the
Coleman opinion, the court warned against using a mechani-
cal “magic formula” for the similarly-situated inquiry but
yet set forth some examples of evidence that would be re-
quired in the usual case. These included whether the em-
ployees being compared (1) were supervised by the same
person, (2) were subject to the same standards, and (3) “en-
gaged in similar conduct without such differentiating or mit-
8 No. 16-3848
igating circumstances as would distinguish their conduct or
the employer’s treatment of them.” Id. (citing Gates v. Cater-
pillar, Inc., 513 F.3d 680, 690 (7th Cir. 2008)).
Once again, this is not a hard and fast test, and there is no
magic to these considerations. In the employment discrimi-
nation context, the requirement to find a similarly situated
comparator is really just the same requirement that any case
demands—the requirement to submit relevant evidence.
Relevant evidence means evidence having “any tendency to
make a fact more or less probable than it would be without
the evidence.” Fed. R. Evid. 401. Evidence of what has hap-
pened to other employees is only relevant if that employee is
in the same boat as the plaintiff. For example, one would ex-
pect an employee with ten years’ experience to be paid more
than an employee with ten months’ experience. Whether a
comparator is similarly situated is typically a question for
the fact finder, unless, of course, the plaintiff has no evi-
dence from which a reasonable fact finder could conclude
that the plaintiff met his burden on this issue. Coleman, 667
F.3d at 846–47; Reed, 869 F.3d at 549.
The plaintiffs argue that Johnson identified two white
EVS technicians named “Kelly” and “Diane” (last names
unknown) who disclosed their salary information to plaintiff
Johnson and, according to Johnson, told him that they were
paid at a higher rate than he was. The plaintiffs argue that
Advocate did not “adequately justify or reconcile this pay
discrepancy” and that even if it did, Advocate mistakenly
thought that Johnson was referring to Diana Esparaza when
actually Johnson was referring to a different person named
“Diane.” (Appellants’ Brief at 10). The defendants, however,
did not have an obligation “to justify or reconcile this pay
No. 16-3848 9
discrepancy.” It is the plaintiffs’ responsibility, on summary
judgment, to make a “showing sufficient to establish the ex-
istence of an element essential to the party’s case, and on
which that party will bear the burden of proof at trial.” Ce-
lotex, 477 U.S. at 322. It is the plaintiffs’ responsibility to go
beyond the pleadings and designate specific facts showing
that there is a genuine issue for trial. Id. at 324. There is no
requirement that the moving party support its motion with
any evidence negating the opponent’s claim. Id. at 323.
Other than Johnson’s statement that “Kelly” and “Diane”
told him they had higher salaries, and that Diane told John-
son that she had no prior experience in a hospital, the plain-
tiffs make no showing to establish an essential element on
which they would bear the burden at trial—that is, whether
Kelly and Diane were similarly situated to the plaintiffs.
They submitted no pay records, nothing about their qualifi-
cations or experience (other than that Diane was an aqua
teacher at a park and had not worked at a hospital before),
who supervised Kelly and Diane, how long they had worked
for the hospital, what types of reviews they received, and if
they had been subject to any discipline. In fact, the plaintiffs
never revealed the last names of “Kelly” and “Diane” such
that the defendant could, on its own, look for the answers to
any of these questions. Even if Kelly and Diane did, in fact,
receive more money than Johnson, without knowing wheth-
er they were similarly situated, a court has no way of dis-
cerning whether this information is relevant to a claim of
race discrimination.
Advocate also argues that Johnson’s report of what
“Kelly” and “Diane” told him is hearsay. Johnson retorts
that this is a statement by the defendant’s employee on a
10 No. 16-3848
matter within the scope of that employee’s relationship with
Advocate and thus not hearsay pursuant to the exclusions
enumerated in Federal Rule of Evidence 801(d)(2)(D). It is
unlikely, however, that this statement falls within the exclu-
sion of Rule 801(d)(2)(D). In order to be excluded, the de-
clarant must be involved in the “decisionmaking process af-
fecting the employment action.” Simple v. Walgreen Co., 511
F.3d 668, 672 (7th Cir. 2007). These declarants were not.
But we need not belabor this hearsay point too much. As
we have just reasoned, even if the statement does not consti-
tute hearsay, it does not provide any relevant information
that would help defeat summary judgment. For purposes of
Title VII, plaintiffs need to produce evidence that similarly
situated non-African-American employees were treated
more favorably. Montgomery v. Am. Airlines, Inc., 626 F.3d
382, 394 (7th Cir. 2010). And the plaintiffs have not done so.
Smith makes the same claim—that two white associates,
who did not have previous housekeeping experience, in-
formed her that their hourly rates were higher than hers.
R. 46-3 at 63, Page ID 676. Smith’s claims fail for the same
reasons that Johnson’s do—lack of evidence that the other
employees were similarly situated.
Likewise, Pannell claims that he was often directed to
perform work outside of his job description thus warranting
extra “charge pay,” but yet never received such pay. He
claims that there was no evidence that white employees in
his job category were asked to perform these tasks, nor was
there evidence that they were deprived of the charge pay.
Once again, however, the burden was on Parnell to produce
evidence of a disparity and he failed to do so in any manner.
He did not submit affidavits from white EVS technicians
No. 16-3848 11
who were paid for their work, records of assignments, pay,
or any other scintilla of evidence of how similarly situated
white employees were treated. In sum, the plaintiffs failed at
their burden of putting forth relevant evidence on these
facts, and thus cannot survive a motion for summary judg-
ment on the claim that African-American plaintiffs were
paid less than non-African-Americans, either in salary or
charge pay.
B. Failure to promote
The plaintiffs also contend that there are disputed ques-
tions of material fact regarding whether the African-
American EVS tech plaintiffs were passed over for promo-
tions in favor of white EVS workers with less experience and
seniority. Here again, plaintiffs have failed to make a show-
ing sufficient to establish the existence of the same element
discussed above which is essential to their case and on
which they will bear the burden of proof at trial—that is,
that there are similarly situated non-African-American
workers who were treated more favorably. See Celotex, 477
U.S. at 322.
The plaintiffs allege that Johnson, Pannell, Smith and
Scott-Murray applied for promotions that were instead giv-
en to white EVS techs with less experience. We can quickly
dispense with Scott-Murray’s claim as he does not dispute
that he did not apply for any promotions. R. 64 at ¶54, Page
ID 2040. Advocate offered Pannell a Tech 2 position but he
declined it because it was not on his preferred shift. R. 64 at
¶36, Page ID 2029–30; R. 62-10 at ¶13, Page ID 1986. Thus,
according to the undisputed facts, neither Pannell nor Scott-
Murray were denied a Tech 2 position. No reasonable juror
could infer that Pannell and Scott-Murray were passed up
12 No. 16-3848
for promotions because of their race. They were not passed
up at all.
Johnson and Smith, on the other hand, did apply unsuc-
cessfully for other positions. They both claim that Advo-
cate’s hiring personnel passed over their applications and
hired non-African-American people with less seniority or
experience in their stead. Smith’s evidence consisted of her
deposition testimony that she applied for positions, did not
get the jobs, and that white applicants did. Save for one ex-
ception, she did not know the names of any of the appli-
cants, nor anything about their experience, qualifications,
resumes, or interviews. R. 46-3 at 9, 11–12, 65–67, Page ID
622, 624–25, 678–80. Smith did know that a white woman
named Mary Harris, received a dispatcher position for
which she applied. R. 46-3 at 9, Page ID 622. Other than
knowing that Harris had worked in an office before, howev-
er, Smith did not know anything else about Harris’ resume,
background, qualifications or why she was selected for the
position. Id. Just as before, it was the plaintiffs’ burden to set
forth factual information to demonstrate that these other ap-
plicants were similarly situated, and they failed to do so.
Johnson’s claim contains the same defect. Although he
does identify four white employees who received positions
for which he unsuccessfully applied—Gina Lika, John
Mueller, Lance White, and Fred (Last Name Unknown)—he
offers no evidence of their background, qualifications, re-
sume, or any other information that would satisfy the plain-
tiff’s burden to demonstrate that the comparators were simi-
larly-situated applicants. Although the brief points to Plain-
tiff’s Response to Defendant’s Statement of Material Facts as
to Which There is No Genuine Issue (Appellant’s Brief at 12–
No. 16-3848 13
13), and that document, in turn, points to depositions and
declaration pages, we could not find, despite our digging,
any relevant material evidence that the employees who re-
ceived the promotions for which Johnson and Smith applied
were similarly situated to the latter. It was, of course, the
plaintiffs’ responsibility to point us to any evidence that
would have supported this essential element of their claim.
It has become axiomatic in this Circuit to remind parties in
colorful terms that “[j]udges are not like pigs, hunting
for truffles buried in” the record. Albrechtsen v. Bd. of Regents
of Univ. of Wisc. Sys., 309 F.3d 433, 436 (7th Cir. 2002) (citing
United States v. Dunkel, 927 F.2d 955, 956 (7th Cir. 1991)).
The most we could find through our charitable truffle
hunting was a sprinkling of speculative statements that the
white employees who received the positions seemed like
they may have been less qualified than African-American
candidates. For example, Pannell stated in his deposition
that a white candidate who was hired must have had less
experience than Scott-Murray because “[h]e [the white can-
didate] was hired, and he had to be trained, so I mean if
[Scott-Murray’s] already been trained, he has got less experi-
ence.” R. 46-1 at 9, Page ID 460. And in regard to a different
position, Pannell stated that he spoke with the hired candi-
date and, “[h]er background, as far as being a tech two, I
talked with her and she said she had never done that type of
[tech two] work before.” R. 46-1 at 10, Page ID 461.
Finally, Johnson stated in his deposition that two white
women, whose names he did not know, received jobs in po-
sitions for which he had applied. According to Johnson, one
had the state approved certification for the job (which John-
son did not have), but Johnson did not know anything about
14 No. 16-3848
her other qualifications. The other was, at that time, going
through the state certification program and had no other
work experience other than “she said she worked on and off
at Popeyes.” R. 46 at 62–63, Page ID 422–23. The brief makes
no claim that any of these employees who were hired were
similarly situated to Johnson or Smith and the citations in
the brief do not point to any relevant support that they were.
The plaintiff has failed to make a showing sufficient to estab-
lish the existence of the element that is essential to their
claim—that similarly situated non-African-American em-
ployees were treated more favorably by the employer vis-à-
vis promotions. Celotex, 477 U.S. at 322; David v. Bd. of Trs. of
Cmty. Coll. Dist. No. 508, 846 F.3d 216, 225 (7th Cir. 2017).
C. Disparate terminations
The lack of a relevant comparator similarly dooms sever-
al other of the plaintiffs’ claims. Smith, Johnson, and Scott-
Murray each claim that they were terminated because of
their race. But once again, none of the plaintiffs sets forth
any admissible evidence that similarly situated white em-
ployees were treated more favorably. For example, Smith
alleges she was terminated for refusing to work in her as-
signed area, while another white employee, Jolanta, who al-
so refused to work in her assigned area, was not terminated,
reprimanded, or sent home. The brief is silent as to any evi-
dence that Jolanta was similarly situated. The plaintiffs offer
no evidence about who Jolanta was, what her position was,
who supervised her, why she refused to work in her as-
signed area, and whether she had a similar disciplinary rec-
ord and similar performance reviews. Arguments that con-
sist of conclusory allegations that are not properly devel-
oped are waived. Anderson v. Catholic Bishop of Chicago, 759
No. 16-3848 15
F.3d 645, 649 (7th Cir. 2014). But even if we dig through the
statement of facts (which, again, we should not need to do),
the only evidence that the plaintiff and Jolanta were similar-
ly situated is that they shared the same supervisor R. 64 at
¶ 78, Page ID 2052–53. We have no idea whether Jolanta was
otherwise similarly situated, or whether there were any un-
usual circumstances, mitigating factors, difference in seniori-
ty, that would have made the comparison inapt.
The same deficiencies plague Scott-Murray’s claims of a
racially-based termination. Her non-specific claim that she
was disciplined for false reasons is unsupported by any de-
tails or any comparison to similarly situated non-African-
American employees. She does raise one example of a non-
African-American employee being treated differently for the
same conduct. Advocate ultimately terminated Scott-Murray
for taking an unauthorized break. Scott-Murray alleges that
she was actually performing her job by cleaning the confer-
ence room at the time of the alleged infraction. She also as-
serts that Diana Esparaza, who is not African American, was
shown more favor by the supervisors and was not terminat-
ed even when she sat in the lobby using her phone and
watching television. The brief provides no information as to
whether Scott-Murray and Esparaza had the same supervi-
sor, or whether they were subject to the same standards.
Johnson also disputes the motivations behind his termi-
nation claiming that management unfairly criticized him,
held him to unfair standards, nitpicked and micromanaged.
This evidence cannot support a claim for race discrimination
unless he can set forth some evidence that similarly situated
non-African-American employees were not treated in the
same manner. A supervisor who nitpicks, micromanages,
16 No. 16-3848
and holds employees to unreasonable standards is simply a
bad boss. It is only if she applies this poor management une-
qually based on race that a plaintiff has a claim for race dis-
crimination.
As evidence of this claim, Johnson makes only the same
type of similarly broad and conclusory allegation that
“white EVS were not subject to such excessive scrutiny, mi-
cromanagement and monitoring.” (Appellant’s Brief at 15).
But even if such an undeveloped, generalized argument that
supervisors treated white employees better than black em-
ployees is not waived (and we believe it is, see Anderson, 759
F.3d at 649), it certainly is not sufficiently supported by evi-
dence so as to survive a motion for summary judgment.
Johnson’s only evidence on this matter was his testimony
that “they walk up behind me, the fact that they nitpick.
They won’t follow white people or Polish people. They will
just follow black people. It’s a fixation upon black people.”
R. 46 at 51, Page ID 411. Johnson can certainly set forth evi-
dence of discrimination in his own declaration or deposition,
but speculation as to an employer’s state of mind is not suf-
ficient to create an issue of material fact. Payne, 337 F. 3d at
772. Nor is a vague claim based on the employee’s subjective
belief that supervisors “nit-picked black” but not white em-
ployees. See Id.
Johnson tries to create a material dispute of fact by point-
ing to inconsistencies in the defendant’s claims about how
many beds each employee was expected to clean on each
shift and whether he was adequately meeting the cleaning
standards and expectations of the employer. There may be
disputed facts about how many beds employees were ex-
pected to clean, but unless there was some evidence that Af-
No. 16-3848 17
rican-American employees were held to a different standard
than white employees, the dispute is not material to the
question of race discrimination.
D. Discriminatory assignments
The related claims—that supervisors assigned African-
American EVS technicians to more strenuous and less desir-
able assignments—also lack evidentiary support. The plain-
tiffs testified that some assignments required more effort
than others. Even accepting the bare allegations that some
floors were less desirable and that more African-American
employees were assigned to these floors (both allegations
without factual support), the plaintiffs have not set forth suf-
ficient evidence to demonstrate that the non-African-
American employees who cleaned the allegedly easier floors
were similarly situated. We do not know, for example,
whether those non-African-American employees had more
seniority or different qualifications or were assigned to those
floors by request or for other reasons. Moreover, the evi-
dence that higher floors were more strenuous came either
from the plaintiffs subjective opinion or speculation as to co-
workers states of mind. See, e.g. R. 46-1 at 8, Page ID 459 (“Q:
Is it possible that some people would find that [a lower
floor] to be more difficult? A: I’ve worked there. I wouldn’t
say so, no.”) In short, this claim fails for the same reason as
the others—the plaintiffs have failed make a showing suffi-
cient to establish the existence of an element essential to their
case. Celotex, 477 U.S. at 322.
E. Hostile work environment/racially derogatory speech
Although we find that the plaintiffs have failed to make a
showing sufficient to defeat summary judgment on their
18 No. 16-3848
claims of race discrimination in pay discrepancy, failure to
promote, termination, and discriminatory work assignments
claims, the claims regarding racially derogatory speech and
a hostile work environment fall on the other side of the line
and survive a motion for summary judgment.
To state a claim for discrimination based on a hostile
work environment, the plaintiffs must show that (1) they
were subject to unwelcome harassment; (2) the harassment
was based on their race; (3) the harassment was so severe or
pervasive as to alter the conditions of employment and cre-
ate a hostile or abusive working environment; and (4) there
is a basis for employer liability. Alamo v. Bliss, 864 F.3d 541,
549 (7th Cir. 2017). The conduct alleged must be “sufficiently
severe or pervasive to alter the conditions of employment.”
Scruggs v. Garst Seed Co., 587 F.3d 832, 840 (7th Cir. 2009).
Whether conduct meets this bar depends on “the severity of
the allegedly discriminatory conduct, its frequency, whether
it is physically threatening or humiliating or merely offen-
sive, and whether it unreasonably interferes with an em-
ployee’s work performance.” Id.; see also Milligan-Grimstad
v. Stanley, 877 F.3d 705, 714 (7th Cir. 2017). Sometimes our
cases phrase the test differently, looking instead for evidence
that the workplace was both subjectively and objectively of-
fensive—either in lieu of the first prong—that the employee
was subject to unwelcome harassment—or the third prong
—whether the harassment was severe or pervasive enough
to rise to the level of a hostile work environment. See Cole v.
Bd. of Trs. of N. Ill. Univ., 838 F.3d 888, 896 n.6 (7th Cir.
2016), cert. denied, 137 S. Ct. 1614 (2017). In the end, we have
concluded that the inquiry is the same. Id.
No. 16-3848 19
We expect a certain level of maturity and thick skin from
employees. “Offhand comments, isolated incidents, and
simple teasing do not rise to the level of conduct that alters
the terms and conditions of employment.” Passananti v. Cook
Cty., 689 F.3d 655, 667 (7th Cir. 2012). “Discrimination laws
do not mandate admirable behavior from employers,
through their supervisors or other employees. Instead, the
law forbids an employer from creating an actionably hostile
work environment for members of protected classes.” Russell
v. Bd. of Trs. of Univ. of Ill. at Chicago, 243 F.3d 336, 343 (7th
Cir. 2001). As the dissent rightfully points out, the environ-
ment need not reach the point of “hellishness,” as some cas-
es once argued. The Supreme Court standard dictates that
the discrimination must be only so severe or pervasive so as
to affect the terms and conditions of employment. Harris v.
Forklift Systems, Inc., 510 U.S. 17, 21–22 (1993). This is a far
cry from hellish.
The district court determined, as a matter of law, that the
conduct complained of was not sufficiently severe or perva-
sive. We disagree. Whether harassment was so severe or
pervasive as to constitute a hostile work environment is gen-
erally a question of fact for the jury. Passananti, 689 F.3d at
669. See also, Smith v. Rock-Tenn Servs., Inc., 813 F.3d 298, 310
(6th Cir. 2016); Mosby–Grant v. City of Hagerstown, 630 F.3d
326, 335 (4th Cir. 2010); E.E.O.C. v. PVNF, L.L.C., 487 F.3d
790, 798 (10th Cir. 2007). In order to remove such a question
of fact from the jury on summary judgment, the court would
have to determine that no reasonable jury could find the
conduct at issue severe or pervasive. In this case, it is certain-
ly possible that a reasonable jury could find that the conduct
was pervasive or severe based on the claims of racially de-
rogatory speech used by Aramark supervisors. Once again,
20 No. 16-3848
we remind ourselves that as we review the facts on sum-
mary judgment we do so in the light most favorable to the
plaintiffs and refrain from resolving swearing contests. We
have divided those facts—the incidents of racially derogato-
ry speech—into three categories: comments made directly to
plaintiffs; comments made directly to non-plaintiff co-
workers and comments made to non-plaintiffs about which
others were aware.
1. Racially charged language
a. Comments made directly to plaintiffs
The defendants argue that the plaintiffs have little admis-
sible evidence to support assertions that supervisors used
racially derogatory language (Appellees’ Brief at 22). This is
not so. The plaintiffs cited to deposition testimony and
sworn declarations in which the plaintiffs testified that they
heard racially derogatory language. Nothing more is re-
quired on summary judgment. “Provided that the evidence
meets the usual requirements for evidence presented on
summary judgment—including the requirements that it be
based on personal knowledge and that it set forth specific
facts showing that there is a genuine issue for trial—self-
serving” testimony is an acceptable method for a non-
moving party to present evidence of disputed material facts.
Payne, 337 F.3d at 773.
Supervisor Castillo told plaintiff Johnson that he
“cleaned like a monkey.” R. 64 at ¶ 25, Page ID 2002; R. 46 at
68, 87, Page ID 428, 447. Supervisor Michalkowski would use
the N-word around African-American employees and would
“mock Johnson as if to say African Americans only speak
slang,” and answer “yo” when Johnson would say “hello.”
No. 16-3848 21
R. 64 at ¶ 26, Page ID 2023; R. 46 at 54, Page ID 414. He
would try to rap, or say “I’m from the hood,” or speak using
stereotypical African-American slang when African-
American EVS techs were around. R. 46-1 at 56–57, 83, Page
ID 507–08, 534; Supervisor Skalnik told plaintiff Scott-
Murray not to give him the “black girl ghetto attitude.” R. 64
at ¶ 45, Page ID 2035; R. 46-2 at 16, Page ID 553. Pannell once
heard Supervisor Hudson call plaintiff Young “a black B. Bi-
polar. Crazy.” R. 64 at ¶ 42, Page ID 2033–34; R. 46-1 at 54,
Page ID 505. Supervisor Hudson called Young “black wig-
wearing witch[ ].” R. 46-4 at 5, Page ID 690. Hudson also
called Young “bipolar” and “crazy.” R. 64, ¶ 42, Page ID
2033–34. Plaintiff Young testified that she often heard
Michalkowski mocking African Americans and on several
occasions (“about three,” she testified), she heard him use
the N-word:
Mike always had the black jokes, you know.
The—he always the black, the slums, the slang
talk. And make fun about how, you know, we
talk, Black people talk. And I found it to be
very offensive … you know, yo this and we be
saying and, you know how black people—
black slang talk. Oh man, we be doing this, but
he talk, try to talk the exact same way the black
guys be talking, yo ni**er this, yo ni**er that
because that’s the words they used to each oth-
er … He did it quite a bit. Had his black slang
talk, you know, yo ni**er this, yo ni**er.
R. 46-4 at 70, Page ID 755 (edits ours).
22 No. 16-3848
b. Direct comments made to non-plaintiffs
Comments made to non-plaintiff co-workers carry less
weight in the evaluation of a hostile environment claim, but
they are not irrelevant. Russell, 243 F.3d at 343 (noting that
“[w]hen harassing statements are directed at someone other
than the plaintiff, the impact of such second hand harass-
ment is obviously not as great as the impact of harassment
directed at the plaintiff.”) After all, the plaintiffs’ burden in a
hostile work environment claim is to demonstrate that ra-
cially charged comments were severe or pervasive. Evidence
that the same supervisors made racially derogatory com-
ments to other employees in the same positions is certainly
relevant evidence tending to make it more likely that the
discrimination was pervasive. Eventually a jury will be able
to weigh such evidence and, with instruction, give it the ap-
propriate amount of weight.
EVS tech Fernando Carpintero gave testimony through a
declaration that Susan Castillo told him directly, while in-
quiring if he knew anyone who would be interested in work-
ing at Advocate, “I don’t want any blacks. They’re lazy.”
R. 62-12 at 2, Page ID 1993. A dispatcher for the EVS de-
partment, Mary Harris, testified through a declaration that,
“[o]n one occasion, I told Susan Castillo that I think I could
handle being a supervisor in the EVS department and Susan
responded, ‘Honey, you’re the wrong color.’” R. 62-11 at 2,
Page ID 1989.
c. Comments made to others about which the plaintiffs
heard
The weakest evidence the plaintiffs present is evidence
about comments that they were told supervisors made—in
No. 16-3848 23
other words, hearsay. It appeared to be well-known
throughout the hospital that Castillo had referred to an em-
ployee as a “porch monkey.” For example, Connie Lockridge
reported to her supervisor that it had happened. R. 62-2 at
14, Page ID 1740; R. 62-3 at 95, Page ID 1853. One dispatcher,
Mary Harris, declared that she heard several African-
American employees, including Lockridge, complain that
Castillo used racially offensive terms such as “porch mon-
key.” R. 62-11 at 2, page ID 1989. Another stated in his decla-
ration that he was “personally aware that Castillo referred to
African Americans as ‘porch monkeys.’” R. 62-12 at 2, Page
ID at 1993. Other employees reported hearing that Susan
Castillo preferred to have Polish people cleaning rather than
African-American people. See e.g. R. 46-3 at 39, Page ID 652.
One of the supervisors who was himself accused of racism,
Michalkowski, testified that an employee came to him to
complain that supervisor Susan Castillo had called another
employee a porch monkey. R. 62-2 at 13–14, Page ID 1739–
40. Finally, an Advocate vice president said to another em-
ployee, within earshot of Johnson, that it was hard to get Af-
rican-American employees to leave. R. 46 at 89, Page ID 449.
Although these comments are largely hearsay, and in some
cases, double hearsay, some of it could be admitted under
hearsay exceptions. For example, it is evidence, not of the
veracity of the remark of course, or even of the truth as to
whether Castillo uttered it or not, but of the fact that em-
ployees understood their environment to be one in which
derogatory statements were pervasive. This latter group of
evidence might have some bearing in a trial on the merits,
but it would be for the district court to determine whether
these comments were more prejudicial than probative. We
would caution against elevating workplace rumors to evi-
24 No. 16-3848
dence of a hostile work environment, although coupled with
other evidence this testimony might have relevance in a hos-
tile work environment claim.
2. This evidence is sufficient to create a triable issue of
fact
The district court found that, although these comments
were “cause for concern … they were not so serious (on their
own or in combination) or so numerous that they materially
influenced plaintiffs’ working conditions.” Johnson, 2016 WL
5871489 at *8. The district court asserted that there was no
evidence that racially derogatory comments made to one
plaintiff had any bearing on other plaintiffs’ work environ-
ments, and thus combining the plaintiffs’ experiences does
not advance any one plaintiff’s claim. Id. And in conclusion
the district court stated that, “[t]he things each plaintiff
heard were too isolated, indirect and sporadic to be actiona-
ble. Although plaintiffs should not encounter racism in their
workplace, a hostile work environment claim under TitleVII
and § 1981 does not provide relief from the comments made
to plaintiffs.” Id. at 19–20.
The district court, evaluating these claims on summary
judgment—where relevant, material, admissible facts must
be assumed to be true—erred by concluding that no reason-
able jury could find these comments to be severe or perva-
sive. We have noted that “Given American history, we rec-
ognize that the word ‘ni**er’ can have a highly disturbing
impact on the listener. Thus, a plaintiff’s repeated subjection
to hearing that word could lead a reasonable factfinder to
conclude that a working environment was objectively hos-
tile.” Hrobowski v. Worthington Steel Co., 358 F.3d 473, 477 (7th
Cir. 2004) (citing Virginia v. Black, 538 U.S. 343 (2003)) (edit
No. 16-3848 25
ours); see also Passanati, 689 F.3d at 668 (“in claims of racial
harassment, racially-charged words certainly can suffice.”);
Rodgers v. Western–Southern Life Ins. Co., 12 F.3d 668, 675–76
(7th Cir. 1993) (finding an actionable hostile work environ-
ment when supervisors and employees referred to plaintiff
by the term “ni**er” between five and ten times during his
employment).
The plaintiffs put forth evidence that one supervisor rou-
tinely harassed them by trying to mock what he thought was
the speech of the African-American EVS techs, directly call-
ing the African-American plaintiffs “ni**er.” They also put
forth evidence of supervisors calling plaintiffs “black wig-
wearing witches,” “a black B. Bipolar. Crazy,” and telling a
plaintiff she had a “black girl ghetto attitude.” It may be, as
the dissent points out, that some individual plaintiffs heard
relatively few uses of racist language aimed directly at them
as individuals. But, as we made clear through our catalogue
of the evidence of racist statements made not only to plain-
tiffs, but to others (see supra pp. 22-23), the collective evi-
dence is sufficient to infer that all of these co-workers knew
that their supervisors were repeatedly using racist language
towards many employees to define the working environ-
ment for all of them. And, it is also sufficient to infer that
these supervisors were using racist language in a pervasive
way to establish racial and hierarchical dominance in the
workplace. That evidence should allow a reasonable jury to
find that each of the plaintiffs experienced a racially hostile
working environment.
The plaintiffs’ evidence that the harassment altered the
terms of their employment is thin perhaps, but it is enough
to survive summary judgment. Particularly because we have
26 No. 16-3848
noted that “a plaintiff's repeated subjection to hearing that
word [ni**er] could lead a reasonable factfinder to conclude
that a working environment was objectively hostile.” Hrob-
owski, 358 F.3d at 477. Several of the plaintiffs complained
about the harassment through the official channels and testi-
fied about the stress they felt because of it, or the increase in
the difficulty of their assignments. See, e.g. R. 64 at ¶¶ 14, 17,
23, 28, 30, 42, 46, 65, 69, Page ID 2015, 2017, 2021, 2024, 2026,
2034, 2036, 2045, 2047. As to the subjective component of the
inquiry, all that the plaintiffs had to establish was that they
perceived the environment to be hostile or abusive. Hrob-
owski, 358 F.3d at 477. A reasonable jury could find that these
words, among them one of the most racially derogatory
word in the English language, that the plaintiffs heard were
unwelcome, and therefore there is an issue of material fact
regarding subjective hostility. Id.
3. Employer liability
A showing of severe or pervasive harassment is neces-
sary but not sufficient to survive summary judgment. The
plaintiffs must also demonstrate a basis for employer liabil-
ity. The question of employer liability has become a bit
muddled in this case. The defendant in these proceedings is
Advocate. The alleged discrimination came at the hands of
Aramark’s supervisors 2. In Title VII cases, liability is direct
2For purposes of this appeal, the claims are basically limited to ac-
tions taken by Aramark supervisors. One Advocate employee, Margaret
DeYoung, was alleged to have made one comment within earshot of
plaintiff Johnson that “it’s a hassle to get [blacks] to leave.” R. 46 at 89,
Page ID 449. Otherwise all the claims of the direct discrimination were
directed at words and actions of Aramark employees, and then at Advo-
cate for failure to stop or remedy the discrimination.
No. 16-3848 27
rather than derivative. Dunn v. Washington Cty. Hosp., 429
F.3d 689, 691 (7th Cir. 2005). The employer is actionable only
for its own deeds. Id. Thus employers are strictly liable for
the discriminatory acts perpetrated by supervisors and they
are liable for the discriminatory acts of others—coworkers,
independent contractors, customers, inmates etc.—only if
they are negligent either in discovering or remedying the
harassment. Nischan v. Stratosphere Quality, LLC, 865 F.3d
922, 930 (7th Cir. 2017). The plaintiff bears the burden of
showing that the employer knew of the problem and that the
employer did not act reasonably to remedy the issue once it
had knowledge. See Dunn, 429 F.3d at 691.
But which entity was the plaintiffs’ employer for purpos-
es of the Title VII claim, and must there be but one? It is not
uncommon for employers to hire subcontractors or other-
wise outsource supervision, just as Advocate did here. Un-
der Title VII law, an employee is considered the supervisor
of the alleged victim of discrimination when “the employer
has empowered that employee to take tangible employment
actions against the victim.” Vance v. Ball State Univ., 570 U.S.
421, 431 (2013). In other words, a supervisor is the one with
the “power to directly affect the terms and conditions of em-
ployment. This power includes the authority to hire, fire,
promote, demote, discipline or transfer a plaintiff.” Nischan,
865 F.3d at 930. When analyzing which entity—Aramark,
Advocate or both—served as the plaintiffs’ de facto employ-
er for purposes of Title VII liability, we use a five-factor test
which requires us to consider:
(1) the extent of the employer’s control and su-
pervision over the worker, including directions
on scheduling and performance of work, (2)
28 No. 16-3848
the kind of occupation and nature of skill re-
quired, including whether skills are obtained
in the workplace, (3) responsibility for the costs
of operation, such as equipment, supplies, fees,
licenses, workplace, and maintenance of opera-
tions, (4) method and form of payment and
benefits, and (5) length of job commitment
and/or expectations.
Knight v. United Farm Bureau Mut. Ins. Co., 950 F.2d 377, 378–
79 (7th Cir. 1991). See also, Nischan, 865 F.3d at 929. The most
important of these factors is the ability to supervise and con-
trol the employees. Love v. JP Cullen & Sons, Inc., 779 F.3d
697, 702 (7th Cir. 2015). And of those control factors, the abil-
ity to hire and fire ranks as most significant. Id.
For purposes of Title VII an employee can have more
than one employer. Harris v. Allen Cty. Bd. of Comm’rs, No.
17-2577, 2018 WL 2275752, at *3 (7th Cir. May 18, 2018). An
entity can be an indirect employer or a joint employer or
have some other complex combined relationship with an
employee. Whitaker v. Milwaukee Cty., Wisc., 772 F.3d 802, 810
(7th Cir. 2014). And when more than one entity is potentially
involved in the employment relationship, a court should ap-
ply the Knight factors to determine which entity or entities
qualify as an employer for purposes of applying Title VII.
Love, 779 F.3d at 701–02. 3
According to the plaintiffs, the Aramark supervisors ap-
proved overtime, created and changed schedules, approved
3
The same Knight factors are used for differentiating between em-
ployees and independent contractors and also for determining who is an
employer for purposes of Title VII liability. See Love, 779 F.3d at 702–05.
No. 16-3848 29
and denied vacation requests, assigned work, disciplined
employees including issuing “Corrective Action Notices,”
did quality checks, and completed performance reviews.
R. 63 at ¶ 30, 32, Page ID 2003. The parties agree that by con-
tract, Advocate maintained ultimate control over discipli-
nary, hiring, and firing actions and that Aramark supervi-
sors could not hire, fire or discipline any EVS tech without
prior approval from Advocate. See R. 63 at ¶ 33, Page ID
2002–03; (Appellants’ Brief at 20). In other words, the Ara-
mark supervisors did have some supervisory and control
powers in the sense that they supervised work, created
schedules and assignments, recommended discipline, etc.,
but, by contract, a significant amount of control remained
with Advocate. We therefore conclude that, because signifi-
cant control remained in the hands of Advocate it was (at
least one of) the plaintiffs’ employers for purposes of Title
VII liability.
The district court did not address the remaining Knight
factors, nor delve further into the role that Aramark supervi-
sors played in “tangible employment actions” directed at the
plaintiffs. See Vance, 570 U.S. at 431. Such a discussion was
not critical to its decision at the time, as it had found no evi-
dence of discrimination. The court concluded, therefore,
with a short-shrift explanation in a footnote, that the Ara-
mark supervisors were not agents of Advocate. Johnson, 2016
WL 5871489 at *9 n.1. Now however, because we conclude
that there is a factual question regarding the racially deroga-
tory speech, we think it worth a full airing by the district
court as to the relationship between Advocate, Aramark and
the plaintiffs.
30 No. 16-3848
It is true that by the terms of the contract between the
parties, Advocate retained final control of most of the tangi-
ble employment actions like hiring and firing. But it is also
possible that, in reality, Advocate acted as a mere rubber
stamp for the recommendations of Aramark supervisors
about these tangible employment actions. Once the District
Court ferrets out the question of supervisor status and the
relationship between the entities, including whether “the
employer may be held to have effectively delegated the
power to take tangible employment actions” to the Aramark
supervisors (Vance, 570 U.S. at 447), it can determine under
which standard Advocate might be held liable for actions by
Aramark supervisors—strict liability or negligence. And de-
pending on which standard applies, the district court may
need to consider whether Advocate took “prompt and ap-
propriate corrective action reasonable likely to prevent the
harassment from recurring.” Cole v. Bd. of Trustees of N. Illi-
nois Univ., 838 F.3d 888, 898 (7th Cir. 2016), cert. denied, 137 S.
Ct. 1614 (2017).
The district court will also have an opportunity to con-
sider the remaining four Knight factors. These were not ad-
dressed by the parties or the district court because no dis-
crimination had been found. But based on the record before
this court now, it appears that many, but not all, of those fac-
tors confirm that Advocate was an employer.
For example, considering the second factor, it appears
that, according to the contract between the parties, Aramark
was responsible for some training of EVS techs. See R. 62-6
at 7, Page ID 1942 (Aramark leaders “will coordinate train-
ing and management of the hourly service employees on
Advocate’s payroll in those departments managed by Ara-
No. 16-3848 31
mark … .”). There has been no factual development on this
point and the district court did not consider it.
The remaining factors all point to Advocate as the em-
ployer. Advocate was responsible for the costs of operation
of the hospital including the equipment, fees, licensing, facil-
ities and maintenance. Advocate paid the plaintiffs’ salaries
and the EVS tech employees expected to stay in their posi-
tions indefinitely.
The answer to the question of whether Advocate took
“prompt and appropriate corrective action reasonable likely
to prevent the harassment from recurring.” Cole, 838 F.3d at
898 also will require a hard look at the particular facts and
circumstances of the case at hand. But we conclude that
there is sufficient evidence of notice to the employer to pro-
ceed past a motion for summary judgment. Johnson did not
report the “porch monkey” comment to Advocate human
resources, nor did he report Michalkowski’s mocking slang,
but other employees did. Johnson also reported to the presi-
dent of the hospital that a supervisor had stated that she pre-
ferred to hire Polish workers rather than African-American
workers because “they clean better.” And Johnson submitted
a list of African-American associates who he believed had
been discriminated against and gave this list to Advocate’s
human resources department. Smith made several com-
plaints to Advocate’s human resources department about
what she believed to be racially motivated behavior. R. 46-3
at 61, Page ID 674; R. 62-1 at 104–112, 144–145, Page ID 1649–
57, 1689–90. Scott-Murray reported to Advocate human re-
sources department that her Aramark supervisor told her
“not to give him that black girl ghetto attitude.” R. 64 at ¶ 45,
32 No. 16-3848
Page ID 2035. Based on these reports, a jury could conclude
that Advocate had adequate notice of the problem.
As for adequate steps to remedy the discrimination, Ad-
vocate investigated many of the complaints, held some
meetings, and met with some supervisors. It is true that an
employer’s course of action need not have been perfect, but
it cannot be negligent. Williams v. Waste Mgmt. of Ill., 361
F.3d 1021, 1030 (7th Cir. 2004). Advocate argues that it took
sufficient corrective action by investigating “the allegations
brought by Plaintiffs and [taking] remedial action in the
form of training and rounding with associates.” (Appellee’s
Brief at 41). Advocate also notes that it reassigned Castillo
when certain allegations were made in fall 2014, and moved
Skalnik away from Plaintiff Scott-Murray when she com-
plained that he placed his hands on her. The district court
pointed out that Advocate human resources employees in-
vestigated Scott-Murray’s complaint against Skalnik; inves-
tigated Smith’s complaints of improper treatment; provided
discrimination and harassment training; instructed leader-
ship to round with associates in EVS; and investigated John-
son’s complaint about payroll errors and resolved some of
them. Johnson, 2016 WL 5871489 at *3–4.
Plaintiffs argue that Advocate did not reprimand or ter-
minate any of the harassers, including Castillo, Skalnik, or
Michalkowski. Advocate did not separate Castillo from the
employees who raised concerns about her. Although Advo-
cate’s human resources employee, Abigail Oman, stated that
she investigated Johnson’s claims that he and other African-
American EVS techs were being treated unfairly, the plain-
tiffs argued that the investigation was wholly inadequate.
Johnson had supplied Oman with a list of other African-
No. 16-3848 33
American EVS techs who he claimed were also being treated
unfairly. Oman went back through her notes of conversa-
tions she had in the past with some of those African-
American EVS techs to see if they had any past reports that
corroborated Johnson’s, but she did not re-interview any of
the EVS techs after Johnson’s complaint, and she did not in-
terview workers on the list with whom she had never spo-
ken. Nor did she contact any white EVS techs to compare
their experiences. Certainly a jury could conclude that an in-
vestigation such as this one—a review of notes of previous
conversations with some of the employees on Johnson’s
list—was negligent.
Finally, although Advocate held discrimination training
sessions, it is undisputed that Michalkowski, one of the main
alleged perpetrators of the discrimination, did not attend.
There is, moreover, a clear factual dispute as to whether Cas-
tillo—the other Aramark supervisor most accused of dis-
crimination—attended the human resources training ses-
sions. The district court found it “unfortunate” that the al-
leged discriminators did not attend the anti-discrimination
training. We think it is more than unfortunate; it creates a
material fact as to whether Advocate took reasonable actions
to remedy the alleged discrimination. A jury might well find
that Advocate’s actions were minimally sufficient or even
completely so, but that is a factual determination based on
all of the facts and circumstances of this particular case. And
those “facts and circumstances” include the “gravity of the
harassment alleged.” May v. Chrysler Grp., LLC, 716 F.3d 963,
971 (7th Cir. 2013); see also Berry v. Delta Airlines, Inc., 260
F.3d 803, 811 (7th Cir. 2001) (“An employer’s response to al-
leged instances of employee harassment must be reasonably
calculated to prevent further harassment under the particu-
34 No. 16-3848
lar facts and circumstances of the case at the time the allega-
tions are made.”). In short, the question as to whether an
employer’s response was reasonably likely to end the har-
assment is fact specific and must be analyzed according to a
totality of the circumstances review. By dismissing on sum-
mary judgment Advocate’s liability for the hostile work en-
vironment, the district court declared that no reasonable jury
could have found that Advocate’s response to the discrimi-
nation was inadequate. We disagree.
Perhaps Advocate’s response was adequate given the
fairly low bar this court has set. Williams, 361 F.3d at 1030.
But that is a factual determination to be made based on the
individual facts and circumstances of this case. Erickson v.
Wisc. Dep’t of Corr., 469 F.3d 600, 606 (the determination of
employer liability is “fact-specific and must be analyzed ac-
cording to the totality of the circumstances.”); Guess v. Beth-
lehem Steel Corp., 913 F.2d 463, 465 (7th Cir. 1990) (noting that
the effectiveness of an employer’s corrective action is a ques-
tion of fact). In this case, we conclude that it could not be de-
termined as a matter of law on a motion for summary judg-
ment.
On a final note, we reject entirely the defendant’s notion
that it “strains credulity” to think that African-American
employees might be subject to a hostile work environment
where many of the managers in the workplace are also Afri-
can-Americans. There are many reasons why women and
minorities in management might tolerate discrimination
against members of their own class, or why they might par-
ticipate in discriminatory acts themselves. See, e.g., Ramit
Mizrahi, "Hostility to the Presence of Women": Why Women
Undermine Each Other in the Workplace and the Consequences for
No. 16-3848 35
Title VII, 113 Yale L.J. 1579 (2004). We need not delve into
this extensive psycho-social research for purposes of this
case.
III.
In sum, we reverse the district court’s grant of summary
judgment on the hostile work environment claim in regards
to the racially derogatory language, and remand to the dis-
trict court for a determination on the merits of that claim.
The decision of the district court is affirmed in all other re-
spects.
36 No. 16-3848
MANION, Circuit Judge, concurring in part and dissenting
in part. I agree with the court that the plaintiffs failed to ad-
duce enough evidence to avoid summary judgment on their
various disparate treatment claims, so I join its opinion in
those respects. However, I disagree with the court’s conclu-
sion that the plaintiffs’ hostile work environment claims
should go to trial. For the reasons stated by the district court,
I would affirm the judgment in favor of Advocate in its en-
tirety.
It is of course true that even “[o]ne overtly racial comment
is too many in terms of basic civility and expected workplace
norms.” Johnson v. Advocate Health & Hosps. Corp., No. 14-CV-
8141, 2016 WL 5871489, at *7 (N.D. Ill. Oct. 7, 2016) (opinion
below). Nevertheless, “Title VII is not a general code of work-
place civility, nor does it mandate admirable behavior from
employers.” McKenzie v. Milwaukee Cty., 381 F.3d 619, 624 (7th
Cir. 2004). “We will not find a hostile work environment for
mere offensive conduct that is isolated, does not interfere with
the plaintiff’s work performance, and is not physically threat-
ening or humiliating.” Yankick v. Hanna Steel Corp., 653 F.3d
532, 544 (7th Cir. 2011). Under this standard, we have held that
“[a] handful of comments spread over months” is insufficient
to demonstrate “severe or pervasive” harassment as a matter
of law. Baskerville v. Culligan Int’l Co., 50 F.3d 428, 431 (7th Cir.
1995); see also Patt v. Family Health Sys., Inc., 280 F.3d 749, 754
(7th Cir. 2002) (eight comments over the course of employ-
ment not enough). In sum, although we have disclaimed our
former use of the descriptor “hellish” to describe an actiona-
ble work environment, compare Wyninger v. New Venture Gear,
Inc., 361 F.3d 965, 977 (7th Cir. 2004) (quoting Rogers v. City of
Chicago, 320 F.3d 748, 752 (7th Cir. 2003)), with Jackson v. Cty.
of Racine, 474 F.3d 493, 500 (7th Cir. 2007), plaintiffs still must
No. 16-3848 37
clear a high bar to avoid summary judgment, see Boumehdi v.
Plastag Holdings, LLC, 489 F.3d 781, 789 (7th Cir. 2007) (com-
paring the “high standard for hostile work environment
claims” with the even higher one for constructive discharge
claims).
This is not a class action; each individual plaintiff must ad-
duce enough evidence for a reasonable juror to conclude that
he or she was subject to a hostile work environment. When
viewed in this manner, the district court was clearly correct
that “[t]he things each plaintiff heard were too isolated, indi-
rect, and sporadic to be actionable.” Johnson, 2016 WL
5871489, at *8. Indeed, once one discounts the plaintiffs’ dis-
parate treatment claims (as the court does), some of the plain-
tiffs’ hostile work environment claims are extraordinarily
weak. Plaintiff Kimberly Scott-Murray, for example, com-
plains of just one racially charged comment the entire time
she was employed at Advocate. She had no interaction with
Susan Castillo and just one minor interaction with Mike
Michalowski, so the conduct of these two major offenders had
little to no impact on her work environment. Plaintiff Robert
Pannell’s claim is on even shakier ground; the only racially
charged comments which he testified to hearing were not ad-
dressed to him. And neither the court today nor the district
court pointed to any race-based comment made in the pres-
ence of Plaintiff Annette Smith. At the very least, we should
affirm summary judgment for Advocate against these three
plaintiffs.
Plaintiffs Warren Johnson and Sherry Young have some-
what stronger claims, but even their allegations fall short of
demonstrating the type of work environment necessary to
survive summary judgment. I agree with the district court
38 No. 16-3848
that “[e]ach of these comments should be cause for concern,
but they were not so serious (on their own or in combination)
or so numerous that they materially influenced plaintiffs’
working conditions.” Id. These “sporadic, offensive com-
ments” are simply not enough to clear the high bar this court
has set in order to bring hostile work environment claims be-
fore a jury. Id. Therefore, I would also affirm the district court
with respect to Johnson’s and Young’s hostile work environ-
ment claims. 1
Because I conclude that none of the plaintiffs proffered
sufficient evidence to avoid summary judgment on their hos-
tile work environment claims, I respectfully dissent from that
portion of the court’s opinion and judgment. I would affirm
the judgment below in full.
1Because I conclude that the plaintiffs’ hostile work environment claims
should fail, I express no opinion on whether employer liability was appro-
priate in this case.