Alternative Dispute Resolution: Panacea or Anathema?

Author(s): Harry T. Edwards

Source: Harvard Law Review, Vol. 99, No. 3 (Jan., 1986), pp. 668-684
Published by: The Harvard Law Review Association
Stable URL: http://www.jstor.org/stable/1341152
Accessed: 10-05-2016 14:34 UTC
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact support@jstor.org.

The Harvard Law Review Association is collaborating with JSTOR to digitize, preserve and extend access
to Harvard Law Review

This content downloaded from 210.212.249.225 on Tue, 10 May 2016 14:34:08 UTC
All use subject to http://about.jstor.org/terms

COMMENTARY
ALTERNATIVE DISPUTE RESOLUTION: PANACEA OR
ANATHEMA?

Harry T. Edwards*
The Alternative Dispute Resolution (ADR) movement has seen an
extraordinary transformation in the last ten years. Little more than
a decade ago, only a handful of scholars and attorneys perceived the
need for alternatives to litigation. The ADR idea was seen as nothing
more than a hobbyhorse for a few offbeat scholars. Today, with the
rise of public complaints about the inefficiencies and injustices of our
traditional court systems, the ADR movement has attracted a bandwagon following of adherents. ADR is no longer shackled with the
reputation of a cult movement.
At worst, ADR is merely a highly fashionable idea, now viewed
as worthy of serious discussions among practitioners and scholars of
widely diverse backgrounds and professional interests. At best, the
ADR movement reflects a serious new effort to design workable and
fair alternatives to our traditional judicial systems. There can be no
doubt, however, that the ADR movement has drawn wide public
attention. During the past five years, there have been literally scores
of books, articles, conferences, bulletins, newsletters, and new course
offerings on ADR. Mechanisms for alternative dispute resolution are
now being established throughout the United States, with well over
one hundred and fifty minor dispute mediation centers in almost forty

states,1 and court-annexed arbitration is now actively used in both

state and federal courts.2 These are indeed heady times for those in
the ADR movement. There is reason for concern, however, that the
bandwagon may be on a runaway course.

Popularity and public interest are not sure signs of a quality endeavor. This is certainly true of ADR, because the movement is illdefined and the motives of some ADR adherents are questionable. It
appears that some people have joined the ADR bandwagon, without
regard for its purposes or consequences, because they see it as a fast
(and sometimes interesting) way to make a buck. It has also been
suggested that some of those people who promote ADR as a means
* Circuit Judge, United States Court of Appeals for the District of Columbia Circuit. Cornell
University, B.S., I962; University of Michigan, J.D., 1965. The author wishes to acknowledge
the research assistance given him by Charles Blanchard in the preparation of this Commentary.
1 Edelman, Institutionalizing Dispute Resolution Alternatives, 9 JUST. SYs. J. 134, 136
(i 984).

2 Sixteen states and ten federal district courts have authorized court-ordered arbitration
programs. See Background and Status, DISPUTE RESOLUTION F., Aug. 1985, at 4.

668

This content downloaded from 210.212.249.225 on Tue, 10 May 2016 14:34:08 UTC
All use subject to http://about.jstor.org/terms

1986] ALTERNATIVE DISPUTE RESOLUTION 669

to serve the poor and oppressed in society are in fact principally

motivated by a desire to limit the work of the courts in areas affecting
minority interests, civil rights, and civil liberties. And it is sometimes
claimed that there are those who subscribe to the ADR movement
because they view efficient and inexpensive dispute resolution as an
important societal goal, without regard for the substantive results
reached. If the ADR movement prominently reflects such thinking
then it is unclear whether the movement is a panacea for, or is
anathema to, the perceived problems in our traditional court systems.
My principal concern is that, in our enthusiasm over the ADR
idea, we may fail to think hard about what we are trying to accomplish. It is time that we reflect on our goals and come to terms with
both the promise and the danger of alternatives to traditional litigation.3 In this essay I will offer some views on the direction this
reflection should take.

I. THE PROBLEM IN PERSPECTIVE

If alternative dispute resolution mechanisms are most significant

as substitutes for traditional litigation, then it is important to assess
the specific problems facing our judicial system that ADR seeks to
address.4 Fortunately, the literature on this subject is so extensive
that it is unnecessary here to rehash the issues or to resolve the ongoing
debate as to whether we are truly an overly litigious society.5 It is
enough to note that, in recent years, the cost of litigation has substantially increased and the number of cases filed in state and federal
courts has mushroomed. For example, between I960 and I980 the
number of filings per capita in federal district courts nearly doubled.6
Although our judicial systems recently have been adjusted to meet

3 By "alternative" dispute resolution I mean to focus on any ADR system that resolves
disputes pursuant to methods other than traditional litigation or government regulatory action.
Some alternatives - such as court-annexed arbitration - act as adjuncts to courts. Others,
however, use private fora not connected in any way to government institutions.
4 This Commentary will focus on the caseload problems of our judicial systems. However,
I recognize that ADR is also responding to other problems with the legal system. See Abel,
The Contradictions in Informal Justice, in THE POLITICS OF INFORMAL JUSTICE 31O (R. Abel
ed. I982); Edwards, Hopes and Fears for Alternative Dispute Resolution, 21 WILLAMETTE L.
REV. 425 (1985).

5 See, e.g., Edwards, The Rising Workload and Perceived "Bureaucracy" of the Federal
Courts: A Causation-Based Approach to the Search for Appropriate Remedies, 68 IowA L. REv.
871 (1983); Galanter, Reading the Landscape of Disputes: What We Know and Don't Know
(And Think We Know) About Our Allegedly Contentious and Litigious Society, 3I UCLA L.
REV. 4 (I983); Miller & Sarat, Grievances, Claims, and Disputes: Assessing the Adversary

Culture, I5 LAW & Soc'Y REV. 525 (I98I); Trubeck, Sarat, Felstiner, Kritzer & Grossman, The
Costs of Ordinary Litigation, 31 UCLA L. REv. 72 (1983).

6 See Galanter, supra note 5, at 37.

This content downloaded from 210.212.249.225 on Tue, 10 May 2016 14:34:08 UTC
All use subject to http://about.jstor.org/terms

670 HARVARD LAW REVIEW [Vol. 99:668

this massive increase in caseload,7 it is somewhat pollyanish to view

the addition of still more judges as an acceptable solution to our
society's ever increasing demand for judicial resources.
Of course, it is misleading to look at statistics on court congestion
as conclusive evidence of the faults of judicial process because, in
state and federal courts, about ninety percent of all cases are settled
without adjudication.8 Although (or maybe because) case filings are
high, we already have an "alternative dispute resolution" system that
emphasizes negotiation rather than adjudication. Unfortunately, our
experiences with litigation-settlement negotiations have been far short
of satisfactory. Recent research reveals widespread dissatisfaction
among trial attorneys, with a "staggering" eighty-five percent agreeing
that the ad hoc processes now employed in connection with litigationsettlement negotiations could be significantly improved. The parties
involved complain that compromise comes too late, is too expensive,
and is too stressful.9
While there is obvious room for improvement in the way we settle
cases - perhaps by encouraging a more active judicial role in settlement negotiations - it is probably naive to think that this alone will
fully solve the problems with our burgeoning caseloads. Many judges
simply lack the mediation skills necessary for the successful resolution
of cases through compromise. There is, unfortunately, no obvious
match between the characteristics that make for excellent judging and
the skills required for successful mediation. Additionally, we cannot
depend on private litigants to settle cases satisfactorily on their own;
too many lawyers view the suggestion of compromise as an admission
of weakness and therefore delay the initiation of negotiations with the

hope that the onus of suggesting settlement will fall on opposing

counsel. 10 Also, lawyers often become so convinced of the merits of
their clients' positions that they may have wholly unrealistic expectations regarding the outcome of a case, thereby lessening the possibilities of successful early negotiation. For these reasons, we should
be highly skeptical of existing trial settlement processes as we search
for viable mechanisms for alternative dispute resolution. Parties will

7 See id.

8 A study of cases in five federal district courts and at least one state court in each federal
district found that less than 8% of the cases went to trial. In 22.5% of the cases, the judge
either dismissed the case or rendered judgment on the merits summarily. The remainder were
resolved by settlement. See Trubeck, Sarat, Felstiner, Kritzer & Grossman, supra note 5, at
89.

9 See W. BRAZIL, SETTLING CIVIL SUITS i (i985). According to Brazil, "the process through
which the parties eventually reach agreement often is difficult to launch, then can be awkward,
expensive, time-consuming, and stressful. The route to resolution can be tortuously indirect
and travel over it can be obstructed by emotion, posturing, and interpersonal friction." Id. at
44.

10 See id. at 45.

This content downloaded from 210.212.249.225 on Tue, 10 May 2016 14:34:08 UTC
All use subject to http://about.jstor.org/terms

i986] ALTERNATIVE DISPUTE RESOLUTION 67I

continue to settle cases, but it is unlikely that the settlement process

will improve if we rely solely on the ad hoc negotiation processes
currently in use.

Given the inadequacy of traditional responses to the manifold

problems with our court systems, it is not surprising that many commentators believe that we must develop new approaches for dispute
resolution in lieu of litigation. Generally, I concur, but I think that
there are two critical threshold inquiries that we must make before

we leap to embrace any system of ADR. First, we should consider

whether an ADR mechanism is being proposed to facilitate existing
court procedures, or as an alternative wholly separate from the established system. Second, we must consider whether the disputes that
will be resolved pursuant to an ADR system will involve significant
public rights and duties. In other words, we must determine whether
ADR will result in an abandonment of our constitutional system in
which the "rule of law" is created and principally enforced by legitimate branches of government and whether rights and duties will be
delimited by those the law seeks to regulate. Perhaps the best way
to conceptualize these critical issues is by reference to a simple matrix:
Private Disputes Public Disputes

ADR in Private Disputes Resolved Public Law Issues

Court by Adjuncts to Courts Resolved by Adjuncts
to Courts

ADR Outside Private Disputes Resolved Public Law Issues

Court by Independent Resolved by Independent
Mechanisms Mechanisms

Obviously, many disputes cannot be easily classified as solely private disputes that implicate no constitutional or public law. Many
commentators have tried to distinguish "public" and "private" disputes;
but, in my view, no one has been fully successful in this effort. The
problem is that hidden in many seemingly private disputes are often
difficult issues of public law. In this Commentary, I offer no easy
solution to the definitional problem of public/private disputes. I do
suggest, however, that there are a number of public law cases that
are easily identifiable as such. These include constitutional issues,
issues surrounding existing government regulation, and issues of great
public concern. The latter category might include, for example, the
development of a legal standard of strict liability in products liability

cases.11 Although less easily identifiable than constitutional and regulatory issues, such issues of great public concern can be accommo11 See, e.g., Escola v. Coca Cola Bottling Co., 24 Cal. 2d 453, 150 P.2d 436 (I944) (Traynor,
J., concurring).

This content downloaded from 210.212.249.225 on Tue, 10 May 2016 14:34:08 UTC
All use subject to http://about.jstor.org/terms

672 HARVARD LAW REVIEW [Vol. 99:668

dated so long as ADR mechanisms are created as adjuncts to existing

judicial or regulatory systems, or if these issues can be relitigated in
court after initial resolution pursuant to ADR.12

My purpose in creating a public/private law matrix is not to give

court administrators a fool-proof method of assigning cases to appropriate dispute resolution systems.13 Instead, the matrix helps to illuminate those aspects of ADR that should give rise to the greatest
concern. In particular, we must focus on the quadrant of the matrix
that would allow for the resolution of public law disputes in ADR
systems that are totally divorced from courts. ADR mechanisms falling within this quadrant, I believe, are wholly inappropriate.

In the remainder of this Commentary I will explore the hazards

and possibilities presented by each quadrant in the matrix, beginning
with two quadrants that involve the use of ADR as an adjunct to our
traditional court system.

II. THE ROLE OF ADR WITHIN THE TRADITIONAL COURT SYSTEM

One way to deal with the caseload problem is simply to divert

cases from litigation by limiting the jurisdiction of the courts. There
are two difficulties with such a "demand-side" approach. First, limiting the jurisdiction of courts may result in diminished rights for
minorities and other groups, whose cases in areas like civil rights,
prisoner suits, and equal employment are likely to be the first removed
from the courts. Second, the jurisdiction-limiting solution fails to
recognize the potential role of ADR within the traditional court systems. If we rush to limit the substantive jurisdiction of our courts,
we may lose our best opportunity to experiment with the promise of
ADR.

Implicitly recognizing these two difficulties, many ADR advocates

have suggested the use of ADR as an adjunct to federal and state
12 Cf. Alexander v. Gardner-Denver Co., 4I5 U.S. 36 (I974) (holding that title VII claims
should be heard de novo in federal court even after the claims are heard in grievance arbitration).

13 The recent Supreme Court decision in Thomas v. Union Carbide Agric. Prods. Co., 05
S. Ct. 3325 (I985), is not inconsistent with my central thesis that public law should not be
resolved by private ADR mechanisms. Thomas held that article III does not prohibit Congress
from selecting binding arbitration as the mechanism for resolving compensation disputes among

participants in the Federal Insecticide, Fungicide, and Rodenticide Act's pesticide registration
scheme. In Thomas, arbitration was chosen by Congress pursuant to standards that it set.
There was no danger that private parties would decide issues of public law.

Thomas is also noteworthy because it employed a public/private distinction. It was, however,

using the distinction as developed in article III jurisprudence. This Commentary employs the
public/private distinction for an entirely different purpose. While article III is concerned about
the exercise of judicial power by the political branches of government, my concern is that public
law issues may be resolved by nongovernmental bodies. Nevertheless, the complexity of the

public/private distinction, as exemplified by Thomas, reinforces my belief that decisions to use

ADR should be made on a case-by-case basis.

This content downloaded from 210.212.249.225 on Tue, 10 May 2016 14:34:08 UTC
All use subject to http://about.jstor.org/terms

I986] ALTERNATIVE DISPUTE RESOLUTION 673

court systems. ADR would not replace litigation, but instead would
be used to make our traditional court systems work more efficiently
and effectively. Because the vast majority of all court cases are settled
rather than adjudicated, many commentators believe that ADR has
an enormous potential for reducing caseloads by enhancing the effectiveness of settlement; at the same time, because ADR would be under
the careful supervision of courts, there is far less danger that ADR
would become a nefarious scheme for diminishing the rights of the
underprivileged in our society.
There are several ways in which the enormous settlement-enhancing potential of ADR can be tapped. Many lawyers insist that a
neutral, penetrating, and analytical assessment of a case greatly enhances the prospects of a successful negotiation by offering a realistic
view of what could transpire if a case goes to full-blown adjudica-

tion.14 Furthermore, because too many lawyers view the suggestion

of compromise as an admission of weakness, mechanisms that place
the onus of suggesting settlement negotiations on neither party have
tremendous potential for initiating settlement at much earlier stages
in the litigation. 15
Indeed, many private litigants and courts already use ADR because
it offers such a neutral assessment and requires parties to think about
compromise at earlier stages in the litigation. For example, several
corporations have pioneered the resolution of large and complicated
business disputes by mini-trials. 16 Although the result is nonbinding,
mini-trials have been tremendously successful in settling cases quickly.
Business litigants frequently find the opinion of a third party invaluable in deciding how best to settle many quite complicated cases. The
mini-trials also have the virtue of forcing corporate litigants to confront the weaknesses in their cases. 17 Unfortunately, however, minitrials are a realistic option only for the wealthy, and the success of
mini-trials may result from the fact that they are initiated by the
parties, who thereby show their predisposition to settle.
Court-annexed arbitration - quickly being adopted in many state
and federal courts - may offer a "poor man's mini-trial." Many

14 See W. BRAZIL, supra note 9, at 44-46.

15 See id. at 45.

16 The first use of the mini-trial was a patent infringement action brought by Telecredit
against TRW. After three years of litigation, the two parties held a nonbinding arbitration
before executives of both corporations and former Judge James Davis of the Court of Claims.
Thirty minutes after the hearing, the parties settled. See Green, Recent Developments in
Alternative Forms of Dispute Resolutions, ioo F.R.D. 512, 5I4-I6 (I983). Judge Lambros has
introduced a "summary jury trial" modeled after the mini-trial. Lambros, The Summary Jury
Trial and Other Alternative Methods of Dispute Resolution, I03 F.R.D. 46I (I984).
17 The mini-trial has been successful in settling disputes of several major corporations including Control Data Corp., Burroughs Corp., Gillette Corp., and Texaco. See Green, supra
note i6, at 5I7.

This content downloaded from 210.212.249.225 on Tue, 10 May 2016 14:34:08 UTC
All use subject to http://about.jstor.org/terms

674 HARVARD LAW REVIEW [Vol. 99:668

jurisdictions have compulsory arbitration for particular classes of cases

- primarily tort and contract disputes with potential damage awards
below an established dollar ceiling. Critically, therefore, court-annexed arbitration is most often used to resolve private disputes rather
than difficult public law issues.18 Indeed, by diverting private disputes to arbitration, federal and state courts may be able to expend
more time and energy resolving difficult public law problems.
The experience in most state court-annexed arbitration programs
is very encouraging. A large percentage of the disputants accept the
arbitrated settlements and express satisfaction with the arbitration
process. In Pittsburgh, for example, court-annexed arbitration ends
three-quarters of all cases without appeal, and the median time to a
hearing is three months, in marked contrast to an eighteen-month
wait for trial. 19 In Michigan, although disputants accept the arbitration award in less than half of the cases, only seven percent of all
cases in which the arbitration award is rejected actually go to trial.20
Of course, in the excitement over the docket-clearing potential of
court-annexed arbitration, we must not make the mistake of ignoring
the quality of arbitration outcomes. The evidence on this is sparse,

but a study of the Pittsburgh program did find that most participants
viewed arbitration outcomes as fair.21 Additionally, court-annexed
arbitration has many of the characteristics of adjudication - most
notably the application of rules of law by neutral decisionmakers.
Unfortunately, the success of arbitration programs has been less
than uniform. The seventh amendment right to a jury trial requires
that arbitrated settlements be nonbinding, unless the parties agree
otherwise.22 In cases seen to be very important to the litigants
whether for monetary reasons or otherwise - losing parties are rarely
willing to accept the result of arbitration as long as a trial de novo
remains available and they have so little to lose by resorting to fullblown litigation. Only if parties agree beforehand to waive their jury
rights can arbitration be fully effective.23
18 Obviously, some small tort and contract disputes can often present novel and important
public law issues. The danger of this occurring, however, is much less than that which would
occur if constitutional and regulatory cases were arbitrated. Furthermore, court-annexed arbitration ensures the parties eventual access to the courts, where novel public law issues can be
resolved.

19 See INSTITUTE FOR CIVIL JUSTICE, AN OVERVIEW OF THE FIRST FIVE PROGRAM YEARS
36 (I983).

20 See Shuart, Smith & Planet, Settling Cases in Detroit: An Examination of Wayne County's
"Mediation" Program, 8 JUST. SYs. J. 307, 315 (I983).

21 INSTITUTE FOR CIVIL JUSTICE, supra note I9, at 36.

22 The nonbinding nature of court-annexed arbitration has been the key factor permitting
federal court-annexed arbitration to survive seventh amendment challenge. See New Eng.
Merchants Nat'l Bank v. Hughes, 556 F. Supp. 7I2, 714 (E.D. Pa. I983); Kimbrough v.
Holiday Inn, 478 F. Supp. 566 (E.D. Pa. 1979).

23 Over half of all arbitrated settlements in the Maryland Health Claims Arbitration System,

This content downloaded from 210.212.249.225 on Tue, 10 May 2016 14:34:08 UTC
All use subject to http://about.jstor.org/terms

i986] ALTERNATIVE DISPUTE RESOLUTION 675

Even with this problem of finality, court-annexed arbitration has

increased the ease with which cases are settled. Most parties that
reject arbitration decisions eventually settle - often earlier than they
would have in the absence of arbitration. Even if parties do not
accept the outcome of arbitration, the arbitrator's decision forces both
parties to focus on a neutral third-party's realistic assessment of the
case.24

As our experience with court-annexed arbitration demonstrates,

federal and state courts are striving mightily to accommodate and
encourage the development of demonstrably effective dispute resolution mechanisms, especially in cases involving private disputes. At
the same time, because these alternatives allow for careful supervision
by the judiciary, there is less danger that the poor will find no room
on the docket. And, most importantly, under these ADR mechanisms,
which function as adjuncts to existing court systems, there is little
likelihood that we will see the creation or development of public law
by private parties. By focusing on that quadrant of the matrix offering
the least concern - the resolution of mostly private disputes by ADR
systems that act as adjuncts to courts - programs such as courtannexed arbitration may diminish the pressure on courts to reduce
substantive rights in response to perceived or actual excessive caseloads.

III. THE ROLE OF ADR AS AN "ALTERNATIVE" SYSTEM

It is clear, however, that a number of ADR proponents have a far

more ambitious vision of ADR than that set forth so far. Some, such
as Jerold Auerbach, seem to favor community resolution of disputes

for example, are refused by parties who demand a trial de novo. See MacAlister & Scanlan,
Health Claims Arbitration in Maryland: The Experiment Has Failed, I4 U. BALT. L. REv.
48I, 50I (I985). Similarly, about 6o% of the litigants studied in three federal district courts
refused arbitrated settlements in the court-annexed arbitration programs adopted by those courts.
See E. LIND & J. SHAPARD, EVALUATION OF COURT-ANNEXED ARBITRATION IN THREE FEDERAL DISTRICT COURTS 76 (I983).

24 A study of three federal district court programs found that, in two of the three districts
studied, the time from filing to disposition decreased as a result of arbitration. In most cases

this was because arbitration encouraged earlier settlements. See E. LIND & J. SHAPARD, supra
note 23, at 76-77.

In addition, some private groups have begun using agreements designed to prevent litigation
altogether. IBM and Hitachi, for example, have agreed, as part of a consent decree in a major
trade secret case, to resolve all future trade secrets disputes by negotiation and arbitration. See
S. GOLDBERG, E. GREEN & F. SANDER, DISPUTE RESOLUTION 545 (I985). Although not court-

annexed, this type of agreement between two private parties is not troubling because it is
unlikely to implicate public law issues.

This content downloaded from 210.212.249.225 on Tue, 10 May 2016 14:34:08 UTC
All use subject to http://about.jstor.org/terms

676 HARVARD LAW REVIEW [Vol. 99:668

using community values instead of the rule of law.25 Others, such as

the Chief Justice, complain that "there is some form of mass neurosis
that leads many people to think courts were created to solve all the
problems of mankind," and believe that ADR must be used to curb
the "flood" of "new kinds of conflicts" (such as "welfare . . . claims
under the Equal Protection Clause") that have purportedly overwhelmed the judicial system.26 In either case, these ADR advocates
propose a truly revolutionary step - the resolution of cases through
ADR mechanisms free from any judicial monitoring or control.
If we can assume that it is possible to finance and administer truly
efficient systems of dispute resolution, then there would appear to be
no significant objections to the use of even wholly independent ADR

mechanisms to resolve private disputes that do not implicate important

public values. For instance, settling minor grievances between neighbors according to local mores or resolving simple contract disputes by
commercial norms may lead to the disposition of more disputes and
the greater satisfaction of the participants. In strictly private disputes,
ADR mechanisms such as arbitration often are superior to adjudication. Disputes can be resolved by neutrals with substantive expertise,
preferably chosen by the parties, and the substance of disputes can

be examined without issue-obscuring procedural rules.27 Tens of thousands of cases are resolved this way each year by labor and commercial arbitration,28 and even more private disputes undoubtedly could
be better resolved through ADR than by adjudication.
However, if ADR is extended to resolve difficult issues of constitutional or public law - making use of nonlegal values to resolve
important social issues or allowing those the law seeks to regulate to

delimit public rights and duties - there is real reason for concern.
An oft-forgotten virtue of adjudication is that it ensures the proper
resolution and application of public values. In our rush to embrace

alternatives to litigation, we must be careful not to endanger what

law has accomplished or to destroy this important function of formal
adjudication. As Professor Fiss notes:
Adjudication uses public resources, and employs not strangers chosen
by the parties but public officials chosen by a process in which the
public participates. These officials, like members of the legislative
and executive branches, possess a power that has been defined and

25 See J. AUERBACH, JUSTICE WITHOUT LAW? I38-47 (I983). However, even Auerbach
recognizes serious limitations on the use of local values in modern society, see, e.g., id. at I44
("[A]lternatives prevent the use of courts for redistributive purposes in the interest of equality,
by consigning the rights of disadvantaged citizens to institutions with minimal power to enforce
or protect them.").

26 Address by Chief Justice Warren E. Burger, in Minneapolis, Minnesota 4 (Aug. 2I, I985).

27 See S. GOLDBERG, E. GREEN & F. SANDER, supra note 24, at I89.

28 See Meyerowitz, The Arbitration Alternative, A.B.A. J., Feb. I985, at 78-79.

This content downloaded from 210.212.249.225 on Tue, 10 May 2016 14:34:08 UTC
All use subject to http://about.jstor.org/terms

1986] ALTERNATIVE DISPUTE RESOLUTION 677

conferred by public law, not by private agreement. Their job is not

to maximize the ends of private parties, not simply to secure the
peace, but to explicate and give force to the values embodied in
authoritative texts such as the Constitution and statutes: to interpret
those values and to bring reality in accord with them.29

The concern here is that ADR will replace the rule of law with
nonlegal values. J. Anthony Lucas' masterful study of Boston during

the busing crisis highlights the critical point that often our nation's
most basic values - such as equal justice under the law - conflict
with local nonlegal mores.30 This was true in Boston during the
school desegregation battle, and it was true in the South during the
civil rights battles of the sixties. This conflict, however, between
national public values reflected in rules of law and nonlegal values
that might be embraced in alternative dispute resolution, exists in
even more mundane public issues.
For example, many environmental disputes are now settled by
negotiation and mediation instead of adjudication. Indeed, as my
colleague Judge Wald recently observed, there is little hope that Superfund legislation can solve our nation's toxic waste problem unless
the vast bulk of toxic waste disputes are resolved through negotiation,

rather than litigation.31 Yet, as necessary as environmental negotiation may be, it is still troubling. When Congress or a government
agency has enacted strict environmental protection standards, negotiations that compromise these strict standards with weaker standards
result in the application of values that are simply inconsistent with
the rule of law. Furthermore, environmental mediation and negotiation present the danger that environmental standards will be set by
private groups without the democratic checks of governmental institutions. Professor Schoenbrod recently has written of an impressive
environmental mediation involving the settlement of disputes concerning the Hudson River. According to Schoenbrod, in that case private
parties bypassed federal and state agencies, reached an accommodation on environmental issues, and then presented the settlement to
governmental regulators. The alternative to approval of the settlement was continued litigation, which was already in its seventeenth
year, with no end in sight.32
The resulting agreement may have been laudable in bringing an
end to protracted litigation. But surely the mere resolution of a
29 Fiss, Against Settlement, 93 YALE L.J. I073, io85 (I984).
30 See J. LuKAS, COMMON GROUND: A TURBULENT DECADE IN THE LIVES OF THREE

AMERICAN FAMILIES I27 (I985); Lukas, Community and Equality in Conflict, N.Y. Times,

Sept. 8, I985, at E5, col. 2.

31 See Wald, Negotiation of Environmental Disputes: A New Role for the Courts?, IO COLUM.
J. ENVTL. L. i, 8 (1985).

32 See Schoenbrod, Limits and Dangers of Environmental Mediation: A Review Essay, 58

N.Y.U. L. REv. I453, I466-67 (I983).

This content downloaded from 210.212.249.225 on Tue, 10 May 2016 14:34:08 UTC
All use subject to http://about.jstor.org/terms

678 HARVARD LAW REVIEW [Vol. 99:668

dispute is not proof that the public interest has been served. This is
not to say that private settlements can never produce results that are
consistent with the public interest; rather, it is to say that private
settlements are troubling when we have no assurance that the legislative- or agency-mandated standards have been followed, and when

we have no satisfactory explanation as to why there may have been

a variance from the rule of law.
In the Hudson River example, we should be concerned if private
negotiators settled the environmental dispute without any meaningful
input or participation from government regulators, or if the private

parties negotiated a settlement at variance with the environmental

standard that had been established by government agencies. If, however, government agencies promulgated the governing environmental
standards pursuant to legislatively established rulemaking procedures
(which, of course, involve public participation), and if the private
parties negotiated a settlement in accordance with these agency standards and subject to agency approval, then the ADR process may be
seen to have worked well in conjunction with the rule of law. Indeed,
the environmental negotiators may have facilitated the implementation
of the rule of law by doing what agency regulators had been unable
to achieve for seventeen years.
A subtle variation on this problem of private application of public
standards is the acceptance by many ADR advocates of the "brokentelephone" theory of dispute resolution that suggests that disputes are
simply "failures to communicate" and will therefore yield to "repair

service by the expert 'facilitator."'33 This broken-telephone theory was

implicitly illustrated in a speech by Rosalynn Carter describing the
admittedly important work of the Carter Center at Emory University
in Atlanta. The Carter Center recently conducted a seminar that

brought together people on both sides of the tobacco controversy.

According to Rosalynn Carter, "when those people got together, I

won't say they hated each other, but they were enemies. But in the
end, they were bringing up ideas about how they could work together. "34

This result is praiseworthy - mutual understanding and good

feeling among disputants obviously facilitates intelligent dispute resolution - but there are some disputes that cannot be resolved simply

by mutual agreement and good faith. It is a fact of political life that

many disputes reflect sharply contrasting views about fundamental
public values that can never be eliminated by techniques that encourage disputants to "understand" each other. Indeed, many disputants understand their opponents all too well. Those who view to33 C. KRAUTHAMMER, CUTTING EDGES: MAKING SENSE OF THE EIGHTIES 4-5 (1985).
34 May, Ex-First Lady Tells of Work of Carter Center, Detroit Free Press, Sept. 13, I985,
at 8B.

This content downloaded from 210.212.249.225 on Tue, 10 May 2016 14:34:08 UTC
All use subject to http://about.jstor.org/terms

i986] ALTERNATIVE DISPUTE RESOLUTION 679

bacco as an unacceptable health risk, for example, can never fully

reconcile their differences with the tobacco industry, and we should
not assume otherwise. One essential function of law is to reflect the

public resolution of such irreconcilable differences; lawmakers are

forced to choose among these differing visions of the public good. A
potential danger of ADR is that disputants who seek only understanding and reconciliation may treat as irrelevant the choices made by our
lawmakers and may, as a result, ignore public values reflected in rules
of law.

We must also be concerned lest ADR becomes a tool for diminish-

ing the judicial development of legal rights for the disadvantaged.

Professor Tony Amsterdam has aptly observed that ADR may result
in the reduction of possibilities for legal redress of wrongs suffered by
the poor and underprivileged, "in the name of increased access to
justice and judicial efficiency."35 Inexpensive, expeditious, and informal adjudication is not always synonymous with fair and just adjudication. The decisionmakers may not understand the values at stake
and parties to disputes do not always possess equal power and resources. Sometimes because of this inequality and sometimes because
of deficiencies in informal processes lacking procedural protections,
the use of alternative mechanisms will produce nothing more than
inexpensive and ill-informed decisions. And these decisions may
merely legitimate decisions made by the existing power structure
within society. Additionally, by diverting particular types of cases
away from adjudication, we may stifle the development of law in
certain disfavored areas of law. Imagine, for example, the impoverished nature of civil rights law that would have resulted had all race
discrimination cases in the sixties and seventies been mediated rather
than adjudicated. The wholesale diversion of cases involving the legal
rights of the poor may result in the definition of these rights by the
powerful in our society rather than by the application of fundamental
societal values reflected in the rule of law.
Family law offers one example of this concern that ADR will lead
to "second-class justice." In the last ten years, women have belatedly
gained many new rights, including new laws to protect battered
women and new mechanisms to ensure the enforcement of childsupport awards. There is a real danger, however, that these new
rights will become simply a mirage if all "family law" disputes are
blindly pushed into mediation. The issues presented extend beyond
questions of unequal bargaining power. For example, battered women
often need the batterer ordered out of the home or arrested - goals
fundamentally inconsistent with mediation.36
35 Address by Professor Anthony G. Amsterdam, Judicial Conference, D.C. Circuit (May
2I, I984), reprinted in I05 F.R.D. 25I, 291 (I985).

36 As Carol Lefcourt of the National Center on Women and Family Law explains:

This content downloaded from 210.212.249.225 on Tue, 10 May 2016 14:34:08 UTC
All use subject to http://about.jstor.org/terms

68o HARVARD LAW REVIEW [Vol. 99:668

Some forms of mediation, however, would protect the public values at stake. Professors Mnookin and Kornhauser suggest, for example, that divorce settlements can be be mediated successfully despite disparities in bargaining power by requiring court review of

settlements that deviate from a predefined norm.37 Additionally, some

disputes that are not otherwise subject to court review also might be
well suited for mediation.38 Many cases, however, may require nothing less than judicial resolution. At the very least we must carefully
evaluate the appropriateness of ADR in the resolution of particular
disputes.
Even with these concerns, however, there are a number of promising areas in which we might employ ADR in lieu of traditional
litigation. Once a body of law is well developed, arbitration and other
ADR mechanisms can be structured in such a way that public rights
and duties would not be defined and delimited by private groups.
The recent experience of labor arbitrators in the federal sector, who
are required to police compliance with laws, rules, and regulations,
suggests that the interpretation and application of law may not lie

outside the competence of arbitrators.39 So long as we restrict arbitrators to the application of clearly defined rules of law, and strictly
confine the articulation of public law to our courts, ADR can be an
effective means of reducing mushrooming caseloads. Employment

discrimination cases offer a promising example. Many employment

discrimination cases are highly fact-bound and can be resolved by
applying established principles of law. Others, however, present novel
questions that should be resolved by a court. If the more routine
cases could be certified to an effective alternative dispute resolution
system that would have the authority to make some final determinations, the courts could devote greater attention to novel legal questions, and the overall efficiency of an anti-discrimination law might
be enhanced.40
The goals of mediation - communication, reasonable discourse, and joint resolution of
adverse interests - work against the most immediate relief the battered woman requires.
The goals she seeks are protection from violence, compensation, possession of her home
without the batterer, and security for her children. Only the judicial system has the
power to remove the batterer from the home, to arrest when necessary, and to enforce
the terms of any decree if a new assault occurs. The empirical data now show that the
therapeutic model for handling battering is ineffective and that firm law enforcement
including imprisonment is required to deter wife abuse.
Lefcourt, Women, Mediation and Family Law, i8 CLEARINGHOUSE REV. 266, 268 (I984).

37 See Mnookin & Kornhauser, Bargaining in the Shadow of the Law: The Case of Divorce,

88 YALE L.J. 950, 993 (I979).

38 See R. Abel, The Contradictions in Informal Justice, in THE POLITICS OF INFORMAL
JUSTICE 309 (R. Abel ed. I982).

39 See, e.g., Devine v. White, 697 F.2d 421, 438-39 (D.C. Cir. i983) (suggesting that labor
arbitrators are just as q-alified to interpret statutes governing personnel relations as they are to
interpret labor contracts).

40 See Edwards, supra note 5, at 925-26, 930. But cf. Getman, Labor Arbitration and

This content downloaded from 210.212.249.225 on Tue, 10 May 2016 14:34:08 UTC
All use subject to http://about.jstor.org/terms

986] ALTERNATIVE DISPUTE RESOLUTION 68i

In other areas, we could capitalize on the substantive expertise

and standards developed by well-established ADR mechanisms. For
example, the experience and standards developed through decades of
labor arbitration and mediation could prove particularly useful in
settling disputes between nonunionized employees and their employers

in cases of "unjust dismissal."'41 Labor arbitrators have developed

fine-tuned standards for just-cause terminations, which they could
easily transfer to the nonunion workplace, thus providing similar
protection to nonunion employees. Similarly, the expertise developed
over the years by commercial arbitrators could be used to settle other
business disputes, which now often require years of litigation. We
should also encourage more private parties to accept binding arbitration voluntarily. Recently, the SEC and the securities industry developed a system of securities arbitration used in thousands of securities law cases.42 If this system is fair to investors and to broker-

dealers, perhaps we should permit investors to commit themselves by

contract to binding arbitration.
Additionally, the qualities of labor arbitration that make it so
successful in the context of collective bargaining are readily transferable to other fields of law. The presence of a skilled neutral with
substantive expertise, the avoidance of issue-obscuring procedural
rules, the arbitrator's freedom to exercise common sense, the selection

of arbitrators by the parties, and the tradition of limited judicial

review of arbitral decisions factors that make arbitration superior
to litigation in labor cases would make arbitration superior to
litigation in other contexts as well. Although the labor context has
the benefit of a collective bargaining agreement providing rules not
subject to arbitrary change by one party,43 the experience with federal
employees demonstrates that arbitration can achieve substantial ben-

efits even when it is limited to the interpretation of rules imposed

unilaterally. 44 Perhaps arbitration could prove useful in moderating
disagreements between citizens, in resolving grievances of citizens
Dispute Resolution, 88 YALE L.J. 9I6 (I979) (suggesting that labor arbitration owes its success
to the collective bargaining relationship).

41 There has been a movement in state courts to protect even nonunionized employees from

unjust dismissals. See Note, Protecting Employees at Will Against Wrongful Discharge: The
Public Policy Exception, 96 HARV. L. REv. 193I, I931 (I983); Note, Protecting at Will Employees Against Wrongful Discharge: The Duty to Terminate Only in Good Faith, 93 HARV. L.
REV. i8i6 (I980).

42 See Katsoris, The Arbitration of a Public Securities Dispute, 53 FORDHAM L. REV. 279
(I984). In Wilko v. Swan, 346 U.S. 427 (I953), the Supreme Court held arbitration agreements
between investors and broker-dealers nonenforceable. Under SEC rule I5C2-2, such arbitration

agreements are illegal. Hence, in order for arbitration to be effective in the securities area,
either Congress must change the law to permit arbitration agreements or the procedures developed by the industry must be attractive to securities plaintiffs.
43 See Getman, supra note 40.

44 See Edwards, supra note 5, at 932.

This content downloaded from 210.212.249.225 on Tue, 10 May 2016 14:34:08 UTC
All use subject to http://about.jstor.org/terms

682 HARVARD LAW REVIEW [Vol. 99:668

against social service agencies, and in resolving complaints of prisoners

over conditions of confinement.
Finally, there are some disputes in which community values
coupled with the rule of law - may be a rich source of justice.
Mediation of disputes between parents and schools about special ed-

ucation programs for handicapped children has been very successful.

A majority of disputes have been settled by mediation, and parents
are generally positive about both the outcome and the process.45 At
issue in these mediations is the appropriate education for a child, a
matter best resolved by parents and educators - not courts.46 Similarly, many landlord-tenant disputes can ultimately be resolved only

by negotiation. Most tenant "rights" are merely procedural rather than

substantive. Yet tenants desire substantive improvement in housing
conditions or assurances that they will not be evicted. Mediation of
landlord-tenant disputes, therefore, can be very successful - often
more successful than adjudication - because both parties have much
to gain by agreement.47
In both of these examples, however, the option of ultimate resort
to adjudication is essential. It is only because handicapped children
have a statutory right to education that parent-school mediation is
successful. It is only because tenants have procedural rights that
landlords will bargain at all.
ADR can thus play a vital role in constructing a judicial system
that is both more manageable and more responsive to the needs of
our citizens. It is essential - as the foregoing examples illustrate that this role of ADR be strictly limited to prevent the resolution of
important constitutional and public law issues by ADR mechanisms

that are independent of our courts.48 Fortunately, few ADR programs

have attempted to remove public law issues from the courts. Although
this may merely reflect the relative youth of the ADR movement, it
may also manifest an awareness of the danger of public law resolution

in nonjudicial fora.
IV. OVERRIDING CONSIDERATIONS

Apart from the issues concerning the appropriate application of

ADR mechanisms, two additional overriding considerations should
45 See L. SINGER & E. NACE, MEDIATION IN SPECIAL EDUCATION: Two STATES' EXPERIENCES ( I985).

46 See id. at 15.

47 See Janes, The Role of Legal Services Programs in Establishing and Operating Mediation
Programs for Poor People, i8 CLEARINGHOUSE REV. 520, 521 (1984).

48 In order to ensure that public law issues are not resolved in private fora, we must permit
litigants who raise issues of public or constitutional law to use courts even if private ADR
systems have already settled the dispute. Cf. Alexander v. Gardner-Denver Co., 415 U.S. 36
(1974) (holding that a title VII claimant does not waive his right to proceed in federal court by
virtue of an adverse decision in grievance-arbitration).

This content downloaded from 210.212.249.225 on Tue, 10 May 2016 14:34:08 UTC
All use subject to http://about.jstor.org/terms

i986] ALTERNATIVE DISPUTE RESOLUTION 683

affect the employment of ADR. One has to do with research and

appraisal, the other with the training and expertise of those who will
serve as neutrals in ADR systems.
Because the ADR movement is still in the formative stage, there
is much to learn about the feasibility of alternatives to litigation. ADR
is, as yet, a highly speculative endeavor. We do not know whether
ADR programs can be adequately staffed and funded over the longterm; whether private litigants will use ADR in lieu of or merely in

addition to litigation; what effect ADR may have on our judicial

caseload; whether we can avoid problems of "second class" justice for
the poor; and whether we can avoid the improper resolution of public
law questions in wholly private fora. In light of these and other
uncertainties about ADR, we should continue to view alternative

dispute resolution as a conditional venture, subject to further study

and adjustment. Every new ADR system should include a formal
program for self-appraisal and some type of "sunset" arrangement to
ensure that the system is evaluated after a reasonable time before
becoming permanently established.
In addition to continued research and appraisal, we must ensure
the quality of the suddenly emerging ADR "industry." Most participants in the ADR movement have joined with pure motives, but this
is not true of everyone. There are now a number of self-proclaimed
ADR "experts," with business cards in hand and consulting firms in
the yellow pages, advertising an ability to solve any dispute. Unfortunately, those who seek to prey on a new idea may wreak havoc
with our systems of justice and destroy the legitimacy of the ADR
movement at its inception. One way to limit this problem is to train
potential neutrals to ensure their expertise in both substantive areas
and in dispute resolution techniques.
There are a number of ADR proponents who appear to believe
that a good neutral can resolve any issue without regard to substantive
expertise. Our experience with arbitrators and mediators in collective
bargaining proves the folly of this notion. The best neutrals are those
who understand the field in which they work. Yet, the ADR movement often seeks to replace issue-oriented dispute resolution mechanisms with more generic mechanisms without considering the importance of substantive expertise.49
Some would respond that judges are generalists and yet we trust
our state and federal judiciary to resolve a broad range of disputes.
This argument, however, is deceptive because judges are specialists
in resolving issues of law. Law aims to resolve disputes on the basis
of rules, whereas alternative dispute resolution mechanisms turn to
nonlegal values.50 If disputes are to be resolved by rules of law, the
49 See Edelman, supra note i, at I38-39.

50 I recognize that legal values may not be wholly absent from ADR mechanisms. See
Eisenberg, Private Ordering Through Negotiation, 89 HARV. L. REV. 637 (I976).

This content downloaded from 210.212.249.225 on Tue, 10 May 2016 14:34:08 UTC
All use subject to http://about.jstor.org/terms

684 HARVARD LAW REVIEW [Vol. 99:668

legal experts designated by our state and federal constitutions - that

is, the judges - should resolve them. If nonlegal values are to resolve

disputes, we should recognize the need for substantive expertise.

As we reflect, above all we must remember that the overarching
goal of alternative dispute resolution is to provide equal justice to all.
"If . . . reform benefits only judges, then it isn't worth pursuing. If
it holds out progress only for the legal profession, then it isn't worth
pursuing. It is worth pursuing only if it helps to redeem the promise
of America. "51 So long as this remains the paramount goal of ADR
and we continue to focus on the essential role of public values reflected
in law, the progress of the ADR movement in the next decade will
surely surpass that of the last.

51 Higginbotham, The Priority of Human Rights in Court Reform, 70 F.R.D. 134, 138 (1976).

This content downloaded from 210.212.249.225 on Tue, 10 May 2016 14:34:08 UTC
All use subject to http://about.jstor.org/terms