rectifying the
tilt:
equality Lessons
from Religion, Disability, Sexual Orientation and
transgender
Chai R. Feldblum
I. Reflections on the
Coffin Lecture – And an Opportunity for Scholarship
It was an honor and a joy to
deliver the Tenth Annual Frank M. Coffin Lecture on Law and Public Service, and
to publish it now in the University of Maine Law Review. I thank you for this opportunity.
I have always believed that a life
worth living includes two necessary components: passion and connection. I experience those components both in my
work and in my personal life. I
love the passion I find in my work — both in my advocacy efforts to advance
justice in the world and in the teaching through which I try to pass on to
others whatever skills and wisdom I have accumulated over the years. And I love the passion I find in my
personal life, in my efforts to explore and commit to the joys and challenges of
intimacy and friendship.
The connections that I treasure
track my various passions: starting from the connections I experience in an
intimate relationship and in personal friendships, to the connections I have
with colleagues, students, and mentors.
Judge Frank Morey Coffin is a
remarkable and joyous connection in my life. It is the connection of a mentor, of a
teacher, and of a friend. And in
the example of his life, Judge Coffin has demonstrated his passion for making
the world a better place, for imparting wisdom (and jibes and practical jokes)
to his students, and for maintaining a full and happy home life.
I have been enriched by this
connection with Judge Coffin — both enriched with little pearls of wisdom and
with great peals of laughter — and I am everlastingly grateful for those
riches.
I can think of no finer way to
honor that connection, and to pay back some of the amazing gifts showered on me
by Judge Coffin, than to deliver (and publish) the Coffin lecture on law and
public service, which encompasses so many of the passions and connections of my
life. And on this occasion of the
Tenth Annual Coffin Lecture, I feel I stand here as a representative of every
former, and current, law clerk of Judge Coffin, all of whom, I know, would echo
my gratitude and joy for their connection to the judge.
I particularly appreciate the
opportunity to step back from my work and to publish a piece that draws on – but
is distinct from – the daily work I engage in. In most of my life, I operate as a
“do-er.” As Director of the Federal
Legislation Clinic at Georgetown University Law Center, I engage in anti-poverty
work, disability rights work, and anti-violence work (together with twelve
students and two full-time Teaching Fellows.) As a consultant, I engage in gay rights
work on the federal and state level.
The joy and the challenge of being
located in an academic setting is that I am also able to engage in forays
(albeit intermittent forays) into scholarly analysis. Delivering this lecture, and publishing
this piece, provides an excellent opportunity for me to engage in such a
foray. This piece, then, is a
scholarly reflection on my advocacy experiences. My goal is to use my experiences in
advocacy as fertile soil from which to create, I hope, a lovely garden of theory
and conceptual thought.
Before setting out on this
endeavor, however, I would like to offer two postulates. There are two essential qualities I
believe distinguish advocacy (the fertile soil) from scholarly analysis (the
garden). First, in advocacy, one always knows what one’s end point will
be—and that end point is decided by the client. Once an end point is decided upon
(through the choice of client), the lawyer crafts every legal argument and
analysis in the manner that will best achieve that goal. If there is a sticky, annoying problem
that stands in the way of the goal
— either an annoying problem with the law or an annoying factual
problem — the lawyer’s job is to
try to explain away, or to manage, that problem. An advocate does not respond to a
difficult legal or factual problem by deciding the opposition has the better of
the argument and dropping his or her client’s case. Rather, the lawyer crafts the best
possible arguments to convince other relevant parties (a judge, a legislator, or
an agency official) of the continued merits of his or her client’s position.
In litigation, lawyers for clients
on either side argue the merits of their position in front of a neutral third
party – a judge, a jury, or a panel of judges. The impartial third party is expected to
navigate through the respective partisan arguments and to arrive at a decision
that bests fits both the law and reality.
Judge Coffin’s two books, The Ways of a Judge and On
Appeal, are excellent discourses on the challenges and joys of acting in
that capacity as a judge.
In the legislative and
administrative arena, where I mostly operate, Members of Congress (and their
staffs) and agency officials rarely assume the role of a neutral third
party. Rather, these individuals
are themselves part of the game, and tend to align themselves with one side of
an issue. Once these individuals
have done that, they become either the champions or the opponents of lawyers
working for clients on a particular issue.
Lawyers working for clients in the
legislative and administrative arenas rarely meet for a one-time, face-to-face
“trial-like” proceeding. Rather,
these lawyers ordinarily engage in long processes of negotiation that may
include many groups concerned with a particular issue. Again, difficult legal and factual
problems usually arise that may undermine the end goal desired by a client. Again, an effective legal advocate does
not respond to those difficult problems by abandoning the client’s desired end
goal and joining the opposition.
Rather, those problems are “managed.” The legal advocate may decide that
certain compromises need to be made, or that positions must shift slightly, in
order to address the legal or factual problem at hand. The key element, throughout this
process, is that all facts and all legal arguments are filtered through the
lens of a pre-determined, client-driven end-point.
The second key quality in
advocacy, whether one is dealing with litigation or with legislation or
administrative rules, is that by the time a certain point is reached – a
lawyer must have an answer. At
a certain point in litigation, a lawyer must decide what position to take in a
brief and how to argue on behalf of that position. At a certain point in a legislative
battle, a legislative lawyer must recommend to a federal or state legislator how
to vote on a particular question (assuming the lawyer’s client has a definitive
position on how the legislator should vote.). A lawyer who says, at that point, “Well,
this is a very difficult and interesting issue, I can see two sides to this
question, and I’m not quite sure how you should vote” – is not acting as a
particularly effective advocate for his or her client.
By contrast, the hallmarks of
scholarly reflection are the exact opposite of these two characteristics. First, in scholarship, there is no
client and, hence, no pre-determined end point. Every scholar presumably begins his or
her scholarly endeavor with a set of values. Moreover, each scholar also arrives at
the endeavor with a set of ideas about how best to achieve those values in a
legal setting. But because there is
no client, a scholar can also be truly open to thinking about, and
exploring, alternative points of view, alternative legal conclusions, and
alternative means for achieving what the scholar presumes to be a normatively
good result.
Rarely have I found that my
scholarly endeavors have changed my values. That is, rarely have I found that my
beliefs about what is normatively good have shifted significantly because
of my research. I have found,
however, that scholarly endeavors have allowed me (indeed, have forced me) to
critique a “party line” espoused by lawyers for a client (including myself,
when I have acted in such a role) on a particular issue. Scholarship both allows for and demands
such a result because a scholar does not filter every legal and factual problem
through the lens of a client who has already determined a desired end point.
The second defining characteristic
of scholarship, which sets it apart from advocacy, is that a legal scholar need
not have all the answers at any particular point in time. Because there is no client, and hence no
defined end game that will play out at a certain point, the answers to difficult
legal and theoretical problems can develop in a slower, more organic
fashion. Thus, for example, I
explore in this piece a model for dealing with one type of equality conflict —
when providing anti-discrimination protection for gay people threatens the
beliefs of certain religious people who believe homosexuality is morally
wrong. When Congress considered
this question in the Religious Liberty Protection Act, there was little
opportunity for lawyers representing either gay rights groups or religious
groups to consider the complexity of the issue removed from the needs of their
clients. By contrast, I hope this
piece can help scholars with different values move the conversation forward in a
deliberative fashion, without having the pressure of “a client” on either side
of the issue.
II. The Fertile Soil of
Advocacy
The bottom-line of this piece is
that a legal mandate to offer reasonable accommodations should be viewed as a
form of legislating equality, not a form of legislating “equality-plus.” This is not a new idea. Whether one deals with the
“sameness-difference” debate in feminism,
affirmative action,
or the debate concerning English as a second language,
the idea that substantive equality requires more than formal equality is nothing
new. Nevertheless, what I hope to
offer in this piece is a model (and a visual) that can further explicate and
apply this principle by drawing on the development of the reasonable
accommodation mandate in the areas of religion and disability, and on the
absence of the development of such a mandate in the areas of sexual orientation
and transgender status.
As I note above, my conclusions in
this piece will not be pre-determined by any client’s established end
point. But the fertile soil from
which I hope to grow a theoretical garden is filled with practical experiences
on behalf of clients. Indeed, it is
precisely those experiences that have demonstrated to me that what should be an
“old” idea is still not an accepted idea. Moreover, those practical experiences
have also lead me to believe that a new model and visual might help opposing
parties reach different conclusions as to what may be right or fair when
equality claims conflict.
A. Congress’ Section 5
Power under the Fourteenth Amendment
In 2000, I wrote an amicus
brief on behalf of seven Members of Congress who had been legislative leaders in
the passage of the Americans with Disabilities Act (ADA). The brief was filed in the case of
Board of Trustees of the University of Alabama v. Garrett,
in which the Supreme Court was faced with the question whether Congress had
acted pursuant to section 5 of the fourteenth amendment when it subjected states
to the requirements of the ADA.
In City of Boerne v.
Flores
the Supreme Court laid out the analysis to be followed for questions such as
these. The Supreme Court is the
sole arbiter of what it means for a state to deny its citizens the equal
protection of the laws. Moreover, although three levels of
judicial “standards of review” under the equal protection clause have been
developed by the Supreme Court as a means of constraining courts in their review
of legislative initiatives,
the Supreme Court has recently applied those standards of review as a means of
constraining Congressional initiatives designed to implement the equal
protection clause. Thus, in order for Congress to enact a
law that legitimately waives a state’s sovereign immunity to money damages in a
lawsuit, Congress must first identify a pattern of likely unconstitutional
conduct on the part of the states. This is conduct that would be ruled
unconstitutional under the level of scrutiny the Supreme Court has established
for actions based on the characteristic at issue.
Congress also has constitutional
authority to prohibit certain additional activities, as long as such
prohibitions are “proportionate and
congruent” to the likely unconstitutional conduct. Thus, any action that Congress wants to
establish as illegal under a statute (in which a state will be subjected to
money damages in a lawsuit) must be tailored, in a proportionate manner, either
to remedy past unconstitutional conduct on the part of states or to
prevent possibly future unconstitutional conduct.
Thus, in Garrett, the
Supreme Court had to decide whether Congress had identified a pattern of likely
unconstitutional conduct on the part of the states prior to passing the ADA, and
whether the requirements of the ADA had been a proportionate and congruent
response to such likely unconstitutional conduct. The ADA is a law that requires employers
both to ignore a person’s disability in determining whether to hire or
promote the individual, and to affirmatively take the disability into
account when determining whether the person is qualified for a job. This latter circumstance arises when an
individual with a disability needs an “accommodation” – the modification of a
policy, the addition of a device or an auxiliary aid (such as a reader or a sign
language interpreter), or a modification in physical space -- in order to enable
the person to compete and perform equally with others. The ADA’s mandate of non-discrimination
thus includes affirmative requirements on covered entities to make modifications
and accommodations, as long as such actions do not reach a statutorily
established limit.
Most disability rights activists
and writers have long viewed the ADA’s negative and affirmative requirements as
equally core applications of “non-discrimination” and “equality.” Thus, when I was drafting the
Garrett amicus brief, I wanted to argue that reasonable accommodation is
a form of core equal protection – once equal protection and equality are
correctly understood. But, as an
advocate, I was compelled to consider whether such an argument would best
advance the interests of my clients.
In this case, my clients were several Members of Congress who wanted the
Supreme Court to conclude that the ADA was an appropriate application of
Congress’ section 5 power. My
colleague, Professor Michael Gottesman, convinced me that my sweeping argument
regarding reasonable accommodation would not advance my clients’ interests. Professor Gottesman, who was
representing the respondent in Garrett at the Supreme Court level,
believed the Court would view my argument that a failure to provide reasonable
accommodations to people with disabilities is itself a denial of equal
protection as far-fetched.
Professor Gottesman offered,
instead, a creative argument that presumed only irrational discrimination based
on prejudice was a violation of equal protection, but then explained the
reasonable accommodation mandate as protecting against such irrational
discrimination.
Upon reflection, I chose to follow
that same tack in the Congressional amicus brief. But the experience of writing the brief
(and making that particular decision) made me want to explore – in a scholarly
setting – why the reasonable accommodation requirement seemed so
naturally to be viewed as “special rights,” rather than equal rights, to those
outside the disability world.
If writing the Garrett
amicus brief brought home to me the infirmities in the public’s understanding of
reasonable accommodation in the disability context, representing a gay rights
group during Congressional consideration of the Religious Liberty Protection Act
brought home a different lesson: that it is difficult for civil rights groups to
grapple with providing a reasonable accommodation to religious people when
providing such an accommodation creates an equality conflict with another
group. That experience made me want
to explore whether there was a model we could develop that would help civil
rights groups and religious groups better grapple with this dilemma.
B. The Saga of the Religious Liberty Protection
Act
In 1993, Congress passed the
Religious Freedom Restoration Act (RFRA) in response to the Supreme Court’s
decision in Employment Division v. Smith. In Smith, the Court ruled that a
neutral government law that results in a burden on religious practices does not
violate the First Amendment guarantee of free exercise of religion. In response, Congress passed RFRA, which
required governments to prove that any neutral law that burdened religious
practices was narrowly tailored to a compelling government interest.
Passage of RFRA was premised on
something of a myth –that the Supreme Court had previously adhered to a
consistently “strict scrutiny” standard for any neutral law that burdened
religious practices. In reality,
the Court had announced a strict standard in an unemployment compensation case,
but had applied the standard in a less than strict manner in a range of other
cases.
RFRA generated the law suit that
ultimately resulted in the City of Boerne case, which brought us the
Supreme Court’s restrictive view of Congress’ Section 5 power. But there was another problem with
RFRA. The law began to be used by
landlords to deny rental housing to cohabitating, unmarried couples, on the
grounds that providing housing to such couples burdened the landlords’ religious
beliefs that cohabitation prior to marriage was sinful. Some courts ruled that civil rights laws
prohibiting marital discrimination were not a burden on the landlords’ religious
beliefs. Others ruled that such civil rights laws
did burden the landlords’ religious beliefs, but the case law was then mixed on
whether such laws were narrowly tailored to a compelling government purpose.
Following the Supreme Court’s
decision in City of Boerne, advocates for religious and civil liberties
groups prevailed on Members of Congress to introduce the Religious Liberty
Protection Act (RLPA). This bill
created the same strict standard for governments to meet whenever a neutral law
burdened the religious beliefs of those covered under the law—that is, the
government had to prove the law was narrowly tailored to a compelling government
interest. The only difference was
that Congress’ asserted constitutional authority to enact the legislation
stemmed primarily from its Commerce Clause Power and its Spending Power.
Unlike the uniform religious and
civil rights advocacy stand that characterized the effort to pass RFRA, RLPA ran
into greater difficulties. Gay
rights groups and some civil rights groups (such as the American Civil Liberties
Union) were concerned that landlords and employers would use the religious
defense created by RLPA to avoid compliance with state and local civil rights
laws prohibiting discrimination based on sexual orientation and marital status,
and perhaps even race.
In order to avoid this result, the
gay rights groups and their allies asked the religious coalition supporting RLPA
to exempt civil rights laws from the scope of that law. The problem with that request, of
course, is that it was difficult to state a rationale for why civil
rights laws should be exempted from RLPA’s religious defense, but not child
abuse laws, or environmental laws, or presumably many other laws passed to
advance the public good. If
protecting the free exercise of religion was really so important that any
time a neutral law burdened religious belief or practice the government had
to justify its law as narrowly tailored to a compelling government interest,
then why should civil rights laws be treated any differently from other neutral
laws? One answer that civil rights
advocates offered was that civil rights laws should be presumed to always pass
the strict scrutiny test, and thus, exempting such laws from RLPA’s scope would
obviate the need for unnecessary litigation. But would it always be true that every
civil rights law would meet this strict standard? And even if all such laws would meet
that standard, were there really no other laws that should be presumed to meet
the strict standard as well?
Precisely because it is so hard to
draw a coherent line once any exemption from the RLPA defense is accepted, the
coalition supporting RLPA had decided (several years before) to resist
any exemptions at all. This
was a hard line position that was politically impossible to deviate from.
A more coherent position for gay
rights groups and civil rights groups to have taken would have been to question
whether a strict scrutiny standard made sense at all as a means of accommodating
religious beliefs and practices in this country. But that position, although advanced by
various academics,
was a political non-starter. I
advanced a soft version of that position in testimony before the House
Subcommittee on the Constitution, but I knew that approach was unlikely to serve
as the basis for any resolution of the conflict between gay rights and religious
rights. But the most compelling fact of that
time period, for purposes of this piece, is that there was no real
conversation in the advocacy world on the underlying, tough question: i.e., what
should government’s responsibility be to religious people who live in a
world in which their religious beliefs and practices may be burdened, including
burdened by civil rights laws?
Towards the end of the period that I was working on
RLPA, I heard the beginnings of one promising conversation on the topic. At the annual conference of the American
Association for Law Schools in 1999, the gay rights section and the religion
section of AALS jointly cosponsored a presentation on RLPA. I was among the four panelists, as was
Professor Mike McConnell.
In his talk, Professor McConnell
first sketched out various forms of non-discrimination based on religion,
including the affirmative obligations imposed by a reasonable accommodation
mandate. He then addressed the civil rights conflict regarding RLPA in the
context of the importance and meaning of reasonable accommodation for religious
people.
I completely enjoyed Professor
McConnell’s talk, even though I arrived at a different policy position than he
did. Indeed, when I stood up to
speak, I started my remarks by saying: “I love it. You totally understand how the mandate
to provide a reasonable accommodation is a form of legislating basic, simple
equality. We should write an
article together about the importance and the meaning of reasonable
accommodation.”
The fact that I appreciated
Professor McConnell’s approach did not, however, answer for me the particular
question on the table – that is, should there be an exemption for civil rights
laws in RLPA? Nor did it answer for
me the harder, underlying question – was the strict scrutiny standard the right
standard to apply to governmental actions in order to ensure the
appropriate level of accommodation for religious people? But at least I felt we were in the right
territory by conceptualizing the question as one of reasonable
accommodation for religious people.
Now I just felt we needed a better model with which to think through the
question of what to do when providing such an accommodation undermines equality
for other groups.
These two experiences in my
personal advocacy – one in disability rights and one in gay rights – gave rise
to the challenge that the Coffin Lecture has now allowed me to turn my attention
to. That is, I want to see if there
is a way to explain the reasonable accommodation mandate in a manner that makes
it easier for individuals (those in the lay public, as well as in the judiciary)
to perceive the mandate as a form of ensuring equality – not as a provision of
special rights or equality-plus. I
also wonder whether a richer understanding of reasonable accommodation might not
help steer individuals of good will to a better conversation when providing an
accommodation for one group necessarily undermines equality rights for a second
group.
I call the model I have developed:
“rectifying the tilt.” But before I
set forth the elements of that model, I want to tell two stories: the
development of the reasonable accommodation mandate for religious people and for
people with disabilities, and the absence of such a mandate yet for gay people
and transgender people. These two
stories help establish the foundation on which I build the components of the
“rectifying the tilt” model.
III.
The Story of Reasonable Accommodation
A.
Reasonable Accommodation for Religious People
Our story begins during the
intense civil rights struggle of the late 1950’s and early 1960's. In 1964, Congress finally breaks the
stranglehold of conservative Southern Democrats and passes a major civil rights
law. Title VII of Civil Rights Act of 1964,
which applies to private employers with a certain number of employees, prohibits
discrimination in employment on the basis of race, color, national origin,
religion, and sex. There has been a
fair amount written about the addition of “sex,” offered in a floor amendment in
the House of Representatives, to the list of protected characteristics. But there has been very little written
about what Members of the 1964 Congress thought about the inclusion of religion
among the list of protected categories. Indeed, I have not found any extended
discussion in the legislative record to the 1964 Act concerning either the need
for anti-discrimination protection on the basis of religion, or what the
components of such an anti-discrimination mandate would be.
The job of explicating the
non-discrimination mandate of Title VII fell to lawyers working at the newly
created Equal Employment Opportunity Commission (EEOC). The analysis for race, color, and
national origin seemed pretty straightforward. It meant that employers would have
to ignore the fact of person’s race, color, or national origin whenever
the employer made any type of employment decision.
But how should the EEOC deal with
an anti-discrimination mandate with regard to religion? Would it be sufficient to require
employers to ignore the religious beliefs of applicants and
employees? That type of rule would
be sufficient to invalidate as discriminatory a policy that said: “No Jews or
Muslims need apply for this position.”
But consider an employment policy that requires all employees to work a
Saturday and Sunday shift every six weeks, and requires all employees to accept
mandatory overtime any weekend the employer needs their services. Would it be discriminatory to require an
Orthodox Jew or a Seventh Day Adventist, whose religious beliefs prohibit them
from working from sundown on Friday night to sundown on Saturday night, to
comply with this even-handed, broad employment policy? Such individuals are not helped, in any
practical way, by a law that requires employers to ignore their religious
beliefs. Indeed, it is the
very fact that the employer is ignoring these individuals’ religious practices
and beliefs that such individuals experience as an act of employment
discrimination.
The EEOC lawyers took two stabs at
this question. First, in 1966, the
EEOC issued guidelines that required an employers to “reasonably accommodate” a
religious person who did not wish to work on his or her Sabbath, as long as the
employer could make that accommodation without “serious inconvenience to the
conduct of the business.” The guidelines also provided, however,
that if an employee had established an expected work week, with mandated
overtime, and all employees knew beforehand that such a schedule existed, the
employer did not have to accommodate an employee who wished to avoid work
on his or her Sabbath.
A year later, in June 1967, the
agency modified its position. In
the new guidelines, the agency required that employers accommodate the religious
beliefs of all employees (regardless of the establishment of any pre-existing
work schedule), but limited accommodations to those that would not impose an
“undue hardship” on the employer’s business.
In the meantime, in 1969, Mr.
Robert Dewey sued his employer, Reynolds Metal Company, charging employment
discrimination based on religion. Reynolds had negotiated a collective
bargaining agreement that required all employees to be available for mandatory
overtime, including on Sunday. Mr. Dewey, a member of the Faith
Reformed Church, believed it was a sin to work on Sunday or to ask others to
work in his stead and asked to be exempted from the required overtime. The company refused, and after Mr. Dewey
failed to show up on several Sundays without finding a replacement, he was
fired.
The district ruled that the company policy was discriminatory in its effect
and in violation of Civil Rights Act of 1964. The appeals court for the Sixth Circuit
reversed, concluding that Title VII prohibited only non-discrimination
based on religion, not the special treatment based on religion requested by Mr.
Dewey. As the court explained:
The fundamental error of Dewey and
the Amici Curae is that they equate religious discrimination with failure to
accommodate. We submit these two
concepts are entirely different.
The employer ought not to be forced to accommodate each of the varying
religious beliefs and practices of his employees.
The appeals court’s view of equality was also quite simple
and direct:
The simple answer … to all of Dewey’s claims is that the collective
bargaining agreement
was equal in its application to all employees and was uniformly applied,
discriminating against no
one.
In June 1971, in a per curiam opinion. the Supreme Court
affirmed the appeals court decision by an equally divided Court.
Clearly, for religious people who
wanted a different meaning for the anti-discrimination mandate of Title VII, a
return to Congress was in order. As
luck would have it, the Equal Employment Opportunity Enforcement Act of 1971 was
proceeding through Congress at the time.
The bill primarily included additional enforcement power for the EEOC. In
January 1972, the bill reached the Senate floor and Senator Randolph offered an
amendment which he suggested Senator Williams, the floor manager for the bill,
would find non-objectionable.
Senator Randolph’s amendment modified the definition of “religion”
for purposes of Title VII. With the amendment, the definition of “religion” read
as follows: “The term ‘religion’ includes all aspects of religious observance
and practice, as well as belief, unless an employer demonstrates that he is
unable to reasonably accommodate to an employee’s or prospective employee’s
religious observance or practice without undue hardship on the conduct of his
business.”
The little discussion of the
amendment that occurred was primarily personal and anecdotal. Senator Randolph explained that he was a
member of a small denomination called Seventh-Day Baptists. Such individuals observe the Sabbath
from Friday night to Saturday night, as do Orthodox Jews and Seventh-Day
Adventists, and like members of those other denominations, Seventh-Day Baptists
believe it is a sin to work on the Sabbath. Senator Randolph bemoaned the fact that
employers often refused to accommodate this religious practice, and that indeed,
members of the younger generation of Seventh-Day Baptists were abandoning the
religion because it was too hard to mesh the religious requirements with work
schedules. Senator Randolph explained his amendment
would require employers to accommodate religious practices, as a part of the
non-discrimination law, unless engaging in such an accommodation would impose an
undue hardship on the employer.
Senator Williams posed very few
questions to Senator Randolph regarding the amendment. He asked whether it would be an undue
hardship for an employer to accommodate an employee who needed one day off on
the weekend – if the job was located at a resort and took place only on
Saturdays and Sundays. It was no
surprise that Senator Randolph was able to answer that such an accommodation
would, indeed, pose an undue hardship.
Senator Williams then pronounced himself satisfied with the amendment,
and the amendment passed without opposition. There was no discussion as to whether
this new provision in the law ensured equality for religious people, or created
a form of equality-plus for religious people. Nor was there any discussion of how the
“undue hardship” limitation would work when the interests of other employees
might be at stake.
Congress’ reticence in this area
ultimately back-fired for supporters of the amendment. In 1977, in the case of TWA v.
Hardison, the Supreme Court ruled that any change that would create more
than a “de minimis” cost for the employer would constitute an “undue
hardship.” Following the Court’s ruling, many
plaintiffs bringing religious accommodation cases found themselves without
practical recourse, as courts almost consistently ruled that the requested
accommodation would impose an “undue hardship” on the employer. In 1997, the Workplace Religious Freedom
Act (WRFA) was introduced, providing the term “undue hardship” in Title VII with
the same definition the term had in the ADA. While hearings have been held on the
bill,
and the bill has been reintroduced each Congress, there has been no real
legislative activity on the bill.
B. Reasonable Accommodation for People with Disabilities
If the 1990’s found the religious
community looking to disability laws as a model, the roles were reversed in
1973. Just two years after passage
of the Equal Employment Opportunity Enforcement Act, Congress passed the
Rehabilitation Act of 1973, reauthorizing a series of grant programs for
vocational rehabilitation programs.
Staff people for a few Senators decided to add a section to the law that
would be modeled on Title IX of the Education Amendments of 1972 (on which
several of the staff people had worked as well) and Title VI of the Civil Rights
Act of 1964. This section, which became section 504,
provided that any entity receiving federal financial assistance could not
discriminate against an otherwise qualified person with a handicap.
There was little discussion or
analysis in Congress regarding section 504 of the bill. Thus, there was little discussion as to
what “non-discrimination” on the basis of handicap would entail. This time the agency directed with the
drafting of regulation was the Department of Health, Education and Welfare
(HEW), since that agency provided significant federal funding to a range of
entities. The attorneys in HEW’s
Office of Civil Rights were charged with drafting regulations that would
implement the new non-discrimination mandate of Section 504.
These lawyers borrowed extensively
from the EEOC’s guidance regarding reasonable accommodation for religious
practices, and from Congress’ codification of the accommodation concept in
1971. Thus, the Section 504 regulations
defined “discrimination” as a failure to make a reasonable accommodation for a
person with a handicap, as long as making that accommodation would not place an
undue hardship on the recipient of federal funds. The regulations did not provide a
definition of “reasonable accommodation.”
Instead, it provided examples of accommodations — building a ramp for
someone who uses a wheelchair; modifying a work schedule so someone could go to
the doctor; having a sign language interpreter for a deaf person, or a reader
for a blind person. The common theme was that all these
modifications gave the individual with a handicap the opportunity to participate
equally in a job or in the receipt of services.
The agency also adopted the
limitation of “undue hardship,” but did not define the term. Instead, the regulations listed a series
of factors that were to be considered in determining whether a particular
accommodation would impose an undue hardship on the recipient of federal
funds.
When those of us who were involved
in the drafting of the ADA addressed the reasonable accommodation provisions of
the bill, we made several decisions.
First, we decided to use the same term -- “reasonable accommodation” –
that had been used in the Section 504 regulations. We made that decision in spite of the
fact that none of us intended the word “reasonable” to mean
“reasonable.” Instead, we intended
the term to mean “effective.” That is, a reasonable accommodation was
one that would effectively allow a person with a disability to perform a
job or benefit from a service. Despite our best efforts, a number of
courts (not surprisingly) have given the term “reasonable” its plain meaning,
and have imported into the ADA a separate defense to the provision of an
accommodation, beyond the “undue hardship” defense established by the statute. In May 2002, the Supreme Court rejected
the argument that the term “reasonable,” in “reasonable accommodation,” grants
employers an open-ended additional defense under the ADA, although it allowed
the term to be used as a defense in the unique circumstances in which provision
of an accommodation would frustrate settled and expected seniority rights of
others employees.
Although those
of us drafting the ADA did not provide a definition of “reasonable
accommodation,” we did provide a definition of “undue hardship.” That definition stated that an “undue
hardship” was something that required “significant difficulty or expense.” Moreover, the committee reports
explicitly observed that the standard of “undue hardship” in the ADA was
different from, and more stringent than, the weak standard enunciated by the
Supreme Court for purposes of religious accommodation under Title VII.
The committee reports never
explain why, as a conceptual matter, the provision of reasonable accommodation
for people with disabilities should benefit from a stricter standard than that
accorded religious people who need accommodations. Indeed, it would have been hard to
articulate such a justification, had one been demanded from the drafters of the
ADA. At bottom, the decision to
strictly demarcate “undue hardship” in the ADA from its statutory sibling in
Title VII was purely pragmatic.
None of us believed the Supreme Court had correctly discerned
Congressional intent when it had interpreted “undue hardship” to mean anything
more than a “de minimis” cost for purposes of Title VII. But it is often hard to get Congress’
attention and have it revisit statutory language once a court has mangled a
term’s interpretation. Thus, advocates advancing the ADA decided to establish
the correct standard for reasonable accommodations in the disability context,
and presumably, hoped the standard would ultimately change for religious people
at some point in the future.
C. Reasonable Accommodation
for Religious People -- Redux
If the Workplace Religious Freedom Act (WRFA) were to pass, the standard
for “undue hardship” that governs reasonable accommodations for religion in the
workplace would be identical to the high standard that exists under the
ADA. But even without passage of
WRFA, issues of religious
accommodation continue to arise under Title VII. Some of these cases pose the
precise conflict between religious beliefs and gay rights that Congress averted
when it transformed RLPA into “baby RLPA.”
Consider the case of Bruff v.
North Mississippi Health Services, decided in 2001, by the Fifth Circuit. Sandra Bruff was hired as a mental
health counselor by the North Mississippi Health Services. The company had contracts with a number
of employers in Mississippi to run their employee assistance plans (EAP), which
included counseling. Three
counselors divided up the work; at different times, one counselor would go out
on location to provide the counseling.
In 1996, Sandra Bruff counseled a woman through the EAP program. A few months later, the woman returned
for another session, explaining she was having trouble in her relationship. This would have been a pretty run-of-the
mill case for an EAP counseling session, except that the woman also explained
that she was in a lesbian relationship.
Bruff then refused to counsel the woman, explaining it was against her
religious beliefs to assist someone in a lesbian relationship.
The employee complained to her
employer who subsequently complained to the company. Upon hearing of the complaint, Bruff
asked the company to accommodate her religious beliefs by letting her decline to
counsel anyone in a gay relationship or anyone in an unmarried, sexual
relationship. The company
considered whether Bruff could be accommodated by shifting responsibilities
among the three EAP counselors.
Ultimately, the company decided such an accommodation would not feasible,
given the small number of counselors and the fact that counselors traveled alone
to provide counseling across the state.
A jury concluded that Sandra Bruff had
been discriminated against on the basis of religion and awarded her $32,000 in
back pay, $320,000 in compensatory damages, and $1.7 million in punitive
damages. The Fifth Circuit overturned the
judgment, ruling that accommodating Bruff would impose an undue hardship on the
company.
So Sandra Bruff has to seek
another job because of her religious beliefs. Is that fair? Well, would it have been appropriate to
accommodate her as she wanted? How
would her co-workers feel, having to travel with her in case there was a
counseling session she couldn’t do?
Moreover, should the employer be required to bear the costs of greater
expenditure of resources because of Bruff’s personal beliefs? And what about the gay employee? Her employer has contracted with a
company to provide employee assistance, and now she walks into a counseling
session only to be told she can’t be counseled.
There is clearly a conflict of rights here. At the moment, the artificially low
standard for “undue hardship” under Title VII may ensure that most religious
employees are not accommodated in the fashion requested by Sandra Bruff. But, as the remainder of this paper will
make clear, I believe the standard for “undue hardship” under Title VII should
be the same high standard that applies under the ADA. But if such a change is to be made, we
need a better way to think through the conflict of rights that will arise under
an invigorated “undue hardship” standard for religious accommodation under Title
VII.
D. Reasonable Accommodation For Gay People
and Transgender People-- A Very Short Story
There are still two
characteristics that may generally cause an individual to be fired from a job,
subjected to harassment, or passed over for a promotion and for which there is
no explicit federal anti-discrimination protection: being gay or being
transgendered. As a matter of legal
doctrine, it is possible to argue that existing sex discrimination laws,
properly interpreted, already prohibit such discrimination. Nevertheless, given that most
courts have not yet interpreted sex discrimination laws in such a manner, it is
(probably) necessary to enact new laws that explicitly prohibit discrimination
based on being transgender or gay.
Since 1974, a bill has been
introduced in every Congress to prohibit discrimination based on sexual
orientation in employment, housing, public accommodations, and government
services. No serious action has
ever been taken on this legislation.
Since 1994, a more targeted bill has been introduced in every Congress,
prohibiting discrimination based on sexual orientation by private employers (the
Employment Non-Discrimination Act (ENDA)).
This bill has received serious attention, and failed to pass the Senate
by merely one vote in September 1996.
ENDA represents a classic example
of the failure to include reasonable accommodation as a basic component
of equality. If one asked a random
sample of gay people in America today whether they face discrimination in the
workplace, one would probably receive a range of answers – from people who
experience severe harassment and discrimination in their workplaces to those who
experience no adverse reactions at all in their workplaces. But if you ask that same random sample
if they are treated fairly and equally with their heterosexual counterparts with
regard to the benefits of employment – most particularly, health
insurance benefits – not one person will be able to claim that he or she is
treated equally with non-gay workers.
Most employers who offer health insurance coverage to the spouses and
children of non-gay workers do not offer equivalent coverage to the domestic
partners (and children of those partners) of gay employees. Even those employers who do offer such
coverage cannot ensure complete equality for their gay employees. Federal tax law requires those employees
to pay a tax on the income equivalent of the health insurance benefit,
something non-gay workers are not required to do.
One would expect that such a
glaring example of non-equality would be addressed by a bill titled the
Employment Non-Discrimination Act.
In fact, the bill does the exact opposite. In two separate sections, the bill
ensures that any discriminatory acts that might flow from the denial of marriage
rights to gay people are left undisturbed in the employment sector. In a section titled “Benefits,” the bill
explicitly provides that “this Act does not apply to the provision of employee
benefits to an individual for the benefit of the partner of such individual.” And in the following section, titled “No
Disparate Impact,” the bill establishes that “the fact that an employment
practice has a disparate impact, as the term ‘disparate impact’ is used in
section 703(k) of the Civil Rights Act of 1964, on the basis of sexual
orientation does not establish a prima facie violation of this Act.” Hence, ENDA will not redress any
inequalities with regard to workplace benefits, nor will it contest any neutral
employment rule based on marriage that results in a disparate impact on gay
people.
I do not highlight this point to
cast aspersions on either the motivations or legal drafting competencies of the
drafters of ENDA. (Indeed, as one
of the primary drafters of the bill, that would be quite odd.) Rather, I highlight this point to
demonstrate how the need to make ENDA politically palatable and viable required
a move back from full equality.
I hope the section below on “morality and the tilt” will shed some light
on why such a compromise was necessary to include in ENDA – at least, in the
ENDA as originally introduced in 1994.
While ENDA does not offer full
equality to gay people, it does at least prohibit certain explicit forms of
discrimination based on sexual orientation. Transgender people are not as
lucky. Unless ENDA is amended prior
to passage to prohibit discrimination based on transgender status (as it should
be), transgender people will still be subject to blatant forms of
discrimination, with their only recourse being existing sex discrimination
laws. Thus, neither the equality of having
one’s characteristic ignored, nor the equality of having one’s
characteristic taken into account, will be available to transgender
people, absent the passage of a law that explicitly provides for both types of
equality.
IV.
“Rectifying the
Tilt” – A Proposed Model
A.
Equality Means Treating Others “As Equals”
Being on a tilt is not easy. It is also not “equal,” if being equal
means being treated “as an equal” in one’s society. There is, admittedly, a side benefit:
one is able to share stories and comradeship with others who are also on the
tilt. But that fact of community is
often used primarily as a means of dealing with the hardship of being on the
tilt in the first place. Moreover,
one should still expect to see communities developing among people who are all
standing upright on level ground.
Creating a level ground does not mean erasing all differences between
people in society. It simply
requires that each different person (and each different community) be treated as
an equal within that society
It should be noted that the
differential norm itself does not inherently create the tilt. In a different world, if that person’s
norm were also the societal norm, that person would be standing upright
and everyone else would be on a tilt.
For example, in a society in which the Sabbath is celebrated from Friday
night sundown to Saturday night sundown, and all businesses and transportation
are shut down during that period, an Orthodox Jew would be standing upright and
a secular Jew would be on a tilt.
It is the result of the particular society’s pre-existing choices
that determine whether a person with a different norm will be on a tilt.
If we believe that equality means
treating every person in our society “as an equal,” and if we believe
that the different norm that creates a tilt for a person in society is not
morally problematic (see more on this below), then any non-discrimination law
that purports to establish equality should be required to do something about the
tilt.
B. What Should We Do About the Tilt?
There are different ways to
approach the phenomenon of various members of society being forced to live on a
tilt. Obviously, one choice is to
change the societal norm so that the person with the differential norm is able
to stand upright. This will be a
phenomenally poor policy choice (which would never be adopted) if the subsequent
effect were to place everyone else on a tilt. But, in certain circumstances,
once a society becomes aware of a particular tilt, it is able to change
the societal norm so that everyone is able to stand upright. For example, a society can mandate that
all new buildings be built with ramps, rather than steps. People who are able to use either ramps
or steps will still experience themselves as standing upright in such a
society. And people who use
wheelchairs will similarly experience themselves as standing upright.
When it is possible to rectify the
tilt by changing the societal norm itself (without putting anyone else on a
tilt), that usually represents the best means for rectifying the tilt. Such a change not only results in
everyone standing upright, it also integrates those with the differential norm
into society in a way that best reaches the goal of treating everyone “as an
equal.”
In many circumstances, however, it
will not make practical sense (or may not seem to make practical sense)
to change the societal norm for everyone.
For example, if we changed our societal norm so that every person was
taught both English and American Sign Language, at home and in school, there
would no longer be a significant tilt for deaf people in our society. Such individuals might presumably still
feel a certain community with each other because of experiencing the world of
(no) sound similarly, but they would not experience themselves as un-equals in
society.
It is difficult to imagine,
however, that our society would choose to engage in such a radical change of the
social norm of our national language.
Hence, an alternative way to rectify the tilt is to change the incline
under that person, and that person only. This could be achieved, for example, by
providing a sign-language interpreter at a conference that a deaf person
attends, or providing a sign-language interpreter who will accompany a deaf
individual on a job. At first
blush, such a requirement might appear to be “special treatment” for the deaf
person – some form of “equality-plus.”
But it clearly is not. If
one accepts that equality means treating all people in society “as equals,” and
if one accepts that the reason the deaf person is on a societal tilt is because
of an affirmative decision by society not to teach all its members American Sign
Language, then providing the sign-language interpreter to the deaf person should
be viewed as the minimal step society is required to take to rectify the
tilt.
In order to fully accept this
proposition, I believe it is important for majority members of society to
acknowledge that the reason they are standing up-right is because society has
already established a level ground underneath them. A critical, concomitant acknowledgment
is that this particular incline of the land has not “just happened.” Rather, the incline has resulted from a
series of affirmative actions and decisions, taken collectively (albeit, often
unconsciously) by society over time.
These actions and decisions may seem natural, and even unassailable, but
they have still been created, collectively, by society.
C.
Should There Be Limits to Rectifying the Tilt?
Under the ADA, the obligation to provide a reasonable accommodation –
what I call, under this model, the obligation to rectify the tilt – has a
statutory limitation built into it.
For example, an employer is not required to make a reasonable
accommodation if doing so would impose an “undue hardship” on the business. While “undue hardship” is defined in the
statute as an action requiring “significant difficulty or expense,” there are no
clear, bright-line rules in the law as to what rises to the level of a
“significant difficulty or expense.”
Indeed, the law is explicitly set up as a flexible, “deep-pocket”
bill. Whether an action rises to
the level of a significant expense depends on a list of factors, including the
size of the business, the number of employees, and the cost of the
accommodation. Thus, the larger the
business, the more the employer is expected to pay for a reasonable
accommodation.
Several commentators who favor disability rights have challenged the
ADA’s bona fides as a civil rights law in light of the statutory
limits built into the law. And,
from an opposite perspective, several economists (and other commentators) have
criticized the ADA as faulty public policy because it places an inappropriate
burden on individual employers and businesses. If rectifying the tilt is an
appropriate societal obligation to ensure that all members of society are
treated as equals, should there be any financial or programmatic limits
to such an obligation? And
regardless of whether one believes there should be such limits, is the approach
taken by the ADA the best one for rectifying the tilt?
I have a strong instinct with regard to the latter question, and an open
mind still on the first question.
Viewed within the “rectifying the tilt” model, the ADA’s approach is
clearly not the optimal one. The
assumption of the model is that all of us, as members of society, have
contributed to the establishment of norms that have created the tilts for our
fellow society members. Hence, it
should be all of us that pay to rectify the various tilts. Why should one particular employer have
to pay for a ramp, because that employer was “unlucky” enough to have one of its
employees involved in an accident that required the employee to use a
wheelchair? Why should one company that offers training sessions be required to
pay for a sign language interpreter, because that company is “unlucky” enough to
have one deaf person sign up for its training? Why should one company have to reorder
its personnel line-up because it is “unlucky” enough to have hired six Orthodox
Jews? Why should one company have to pay extra
premiums to a health insurance carrier because it has a significant number of
gay employees with domestic partners or pay to install a one-person unisex
bathroom to accommodate a transgender employee?
In light of the fact that it was society, collectively, that engaged in
the affirmative actions that created the norms and the tilts, I believe we
should establish a fund that we all contribute to equally. This fund, which I call the RTF
(Rectifying the Tilt Fund), should be available for covering the costs of
rectifying the tilt for any characteristic for which the law has created a
reasonable accommodation mandate.
The political fallibility of such a fund should be readily apparent to
anyone who can spell “taxes.” In
recent decades, the idea that “our money belongs to us, and not the government,”
has held powerful sway over the electorate. The converse idea – that we the people
are the government – and that, we the people, have collective
responsibilities to each other, seems to have had much weaker political
pull. Nevertheless, as a
theoretical matter, I do not believe the reasonable accommodation mandate
should fall haphazardly on those businesses that happen to be the ones faced
with an employee or customer with a disability (or other characteristic
requiring an accommodation).
In light of the fact that a RTF does not exist (and is not likely to be
established in the near future), the “undue hardship” limitation, together with
its “deep pocket” approach, seems a pragmatically necessary limit for the
moment. It is appropriate to
require an entity operating within society to take resp
