Printer Friendly
The Free Library
22,741,889 articles and books

Connecting self-representation to civil Gideon: what existing data reveal about when counsel is most needed.

Introduction

  I. Unmet Legal Needs, Unrepresented Litigants, & Civil Gideon
 II. Reviewing Reports: The Impact of Representation
       A. Preliminary Considerations: Methodologies in the Reports
       B. Courts
          1. Housing Cases
          2. Family Law Cases
          3. Small Claims Cases
       C. Administrative Agencies
          1. Social Security Disability Appeals
          2. Unemployment Cases
          3. Immigration Cases
          4. Other Administrative Appeals
III. Assistance Short of Full Representation
       A. Pro Se Clinics, Self-Help Centers and Hotlines
       B. Assessing the Early Evaluation Efforts of Assistance
          Programs
 IV. Observations and Answers
       A. Beyond Representation
          1. The Substantive Law
          2. The Complexity of the Procedures
          3. The Judge: Individual Practices and Perspectives
          4. The Court's Operation
       B. Representation: What We Know
          1. The Importance of Power
          2. Not Just Any Advocate: The Importance of Knowledge
             and Expertise
       C. Applying What We Know From The Reports
          1. Implications for Moving Forward
          2. The Need for Clarity of Goals and Values
  V. Questions and Complications
       A. Questioning the Data
       B. When Positive Outcomes Could Be Negative Ones
       C. The Role of Expectations
       D. The Thorny Question of Resources
       E. Short-Term Versus Long-Term Change, Stability, Fluidity,
          and Backlash
Conclusion

INTRODUCTION

Over the past decade, the phenomenon of self-representation in civil cases has received increased scrutiny. The courts' struggles with huge numbers of cases involving at least one party without counsel have led to questions about the proper role of the key players in the court system and the development of programs designed to facilitate self-representation. States have created Access to Justice Commissions to respond to the problems of those without access to legal representation. A revitalized movement seeking to establish a civil right to counsel has emerged, pressing for the expansion of the availability of counsel for the poor. The activity of the past decade comes against the backdrop of unmet legal needs, the inadequacy of funding for legal services for the poor, and reports demonstrating that litigants without counsel often fare poorly even where basic needs are at stake in the proceedings. (1)

Viewed one way, facilitating self-representation and establishing a civil right to counsel could conflict. In this view, the response to the problems facing those without counsel is to provide assistance improving their ability to self-represent, not to provide counsel. The responses could instead be part of the same Access to Justice agenda. Proponents of self representation might believe that, even with robust assistance programs, some cases still are not appropriate for self-representation. Proponents of a civil right to counsel might acknowledge that, even with an expansion of the availability of counsel, no current proposal contends that counsel should be provided at public expense for all litigants in all civil cases.

An approach viewing self-representation and civil Gideon as part of a common Access to Justice agenda suggests a common inquiry: what are the scenarios in which full representation by counsel is most needed? Part of this question involves policy choices as to the importance of what is at stake in the proceeding. Not surprisingly, initiatives seeking an expanded right to counsel focus on scenarios in which basic human needs, such as child custody and shelter, are at stake. (2)

Part of the question, however, is a research question: what does the data reveal about the characteristics of litigants, cases, courts, or agencies that help identify the cases in which the presence of counsel is most likely to impact the outcome of the case?

Many reports explore the problems facing those without counsel in the "poor people's courts," typically handling family, housing, and consumer cases. Other reports shed light on the impact of assistance programs short of full representation by counsel. Before we embark on new research, we should assess what we know from existing reports. (3)

Part I provides the background for the reports, including unmet legal needs, unrepresented litigants, and civil Gideon. Part II explores what the reports reveal about the correlation between representation and success rates in court. Notwithstanding methodological differences, reports consistently show that representation is a significant variable affecting a claimant's chances for success in eviction, custody, and debt collection cases. That finding also applies to administrative proceedings. Rebecca Sandefur's meta-analysis of studies of the effects of representation reports that parties represented by lawyers are between 17% and 1380% more likely to receive favorable outcomes in adjudication than are parties appearing pro se. (4) With programs facilitating self-representation, litigants and court personnel report high levels of satisfaction; the programs' impact on case outcomes is less clear.

Part III discusses key variables beyond representation that impact case outcomes, before identifying two crucial conclusions regarding what we do know about representation. The first is the importance of power. The greater the power opposing a litigant, and the more that litigant lacks power, the greater will be the need for representation; the section therefore explores sources of power that line up against litigants without counsel, and factors that act as barriers to representation. (5) The second theme is the importance of having not just any advocate, but of having a skilled advocate with knowledge and expertise relevant to the proceeding. The balance of the article discusses programmatic implications that flow from the conclusions and complicating questions.

While gaps in our knowledge suggest the need for additional research, (6) the consequences of waiting, without acting, until all new research is complete, are devastating. As we gain new information, the insights will raise new questions as well; we may never be certain. More importantly, powerless litigants across the country suffer harmful outcomes in cases involving basic needs, with devastating consequences for them and their families. Where we can already identify likely starting points for reform, the price of delay outweighs the costs of uncertainty.

I. UNMET LEGAL NEEDS, UNREPRESENTED LITIGANTS, & CIVIL GIDEON

The phenomenon of unrepresented litigants in the courts is nothing new. Legal Needs studies consistently show that 70-90% of the legal needs of the poor go unaddressed. (7) Many unmet legal needs involve housing, family, and consumer issues. (8) Legal services offices only represent a fraction of eligible clients seeking assistance. (9) Unrepresented litigants often fare poorly in the courts. (10)

During the 1990's, the problems facing, and caused by, unrepresented litigants-also referred to as self-represented, pro se, and, in California, pro per--increasingly gained attention across the country. (11) Most family law cases involve at least one party without counsel, and often two. (12) Most tenants, many landlords, and most debtors appear in court without counsel. (13) Unrepresented litigants disproportionately are minorities and typically are poor. (14) They often identify an inability to pay for a lawyer as the primary reason for appearing without counsel. (15)

Conferences, publications, and websites have focused attention on problems involving cases with unrepresented litigants. (16) One focus is the changing roles for judges, mediators, and clerks in courts with a high volume of unrepresented litigants. (17) Innovative assistance programs are another focus. The programs include hotlines, technological assistance, clinics, pro se clerks offices, "lawyer-of-the-day" programs, and self-help centers, developed to provide assistance to litigants who otherwise would receive no help at all. (18) Conferences of Judges and State Court Administrators adopted resolutions calling for courts to provide meaningful Access to Justice. (19) The number of state Access to Justice Commissions, comprised of stakeholders from the courts, legal services programs, and private bar, expanded rapidly. The commissions are charged with broad mandates to assess the civil legal needs of low-income citizens and design initiatives to respond to those needs. (20)

As the courts, bar associations, and Access to Justice Commissions grapple with the problems involved with unrepresented litigants, a renewed call for a civil right to counsel, or civil Gideon, has gained momentum. The years after 2003, which marked the 40th Anniversary of Gideon v. Wainwright, (21) saw a sharp increase in the number of articles, (22) conferences, (23) and websites dedicated to the issue, (24) as well as a surge in membership in the newly-created National Coalition for a Civil Right to Counsel. (25) Some advocates pursued test case strategies attempting to establish the right to counsel by court decision, (26) while others pursued a legislative strategy. (27) Bar associations and Access to Justice Commissions joined the call, inspired by the unanimous passage of an American Bar Association resolution urging the provision of legal counsel as a matter of right, at public expense, to low-income persons in those categories of adversarial proceedings where basic human needs are at stake, such as those involving shelter, sustenance, safety, health, or child custody, as determined by each jurisdiction. (28)

II. REVIEWING REPORTS: THE IMPACT OF REPRESENTATION

A. Preliminary Considerations: Methodologies in the Reports

As demonstrated in the following sections, consistent themes emerge from the various reports analyzing the courts and administrative agencies before which the cases of many unrepresented, indigent litigants are heard. The themes emerge despite the use of varying methodologies that at times makes the comparison risky. The studies of courts handling housing cases, discussed in the next section, illustrate the point.

Some reports focus primarily on adversarial hearings in the courtroom, others focus on the dynamic in the hallways, while still others include a discussion of the various means of dispute resolution. In terms of the factors used to evaluate the impact of counsel, consistent factors across the jurisdictions include which party was awarded possession, whether a judgment was entered for rent, whether a rent abatement was ordered, and whether the landlord was ordered to make repairs. Other reports measure the number of days between the court date and the date a tenant is to be evicted, as well as the number of court appearances, answers interposed, and motions filed, including post-judgment motions. (29)

The methodologies employed in the studies reflect a range of techniques. The most rigorous is the study by Carroll Seton, Gregg Van Ryzin, Martin Frankel, and Jean Kovath of New York City's Housing Court, using means of a randomized experiment involving a treatment group of legal aid-eligible tenants that was targeted to receive legal counsel and a control group that was not. (30) Other studies, while less formal and not involving a randomized experiment, obtained results through file reviews, interviews with participants, observations or some combination. (31) The picture of the operation of some courts is confirmed or enhanced by newspaper reports and other articles telling the story of particular litigants and cases, judicial decisions, or the combined experiences of advocates familiar with the forum. (32)

Studies of other forums also differ enough in their inquiries and methodologies to make caution advisable in any comparative analysis. Studies of domestic disputes and consumer cases often focus on different proceedings and measure different outcomes. (33) In contrast, the studies of administrative agencies use more parallel methodologies, with more straightforward measures of success rates. (34) Many of the early studies of assistance programs identify successful outcomes by gauging the perceptions of actors involved in the process, often without examining case files to identify the results of the case. (35)

A more detailed analysis of the methodologies is beyond the scope of this article. Despite the methodological differences, we may draw important conclusions from the research that exists. The questions that remain unanswered provide guidance as to how to frame future questions for inquiry, and how to design future research initiatives.

B. Courts

1. Housing Cases

Courts that handle housing cases have been the focus of countless reports across the country over the past three decades. (36) The titles capture the perilous fate awaiting unrepresented tenants: Injustice In No Time, (37) No Time for Justice, (38) Judgment Landlord, (39) Justice Evicted, (40) and 5 Minute Justice or "Aint [sic] Nothing Going on But the Rent!" (41) In addition, Silence in the Court." Participation and Subordination of Poor Tenants' Voices in Legal Process captures the powerlessness of tenants, (42) while Alone in the Hallway speaks to the perils of hallway negotiations. (43)

Despite some variation in details, the core features of the courts seem remarkably consistent. The courts are high-volume courts, with few cases going to trial, and the vast majority resolved by default or settlement, typically the result of hallway negotiations. Tenants rarely are represented by counsel, while the representation rate of landlords varies from low rates in some courts, to highs of 85-90% in others; (44) where landlord representation is high, the typical case pits a represented landlord against an unrepresented tenant. The demographics of the tenants reveal a vulnerable group of litigants, typically poor, often women, and disproportionately racial and ethnic minorities. (45)

While the details of eviction procedures vary, the common outcome measurements include possession, rent abatement, and repairs. Regardless of whether tenants appear or default, settle or go to trial, raise defenses or do not, the result invariably is a judgment for the landlord. Typically, the results are unaffected by whether the landlord is represented by counsel. (46) The unrepresented tenant faces swift eviction, and with minimal judicial involvement. (47)

One variable that often can halt the swift judgment for the landlord is representation for the tenant, with the likelihood of eviction dropping precipitously. Some reports discuss winning generally, showing tenants three, (48) six, (49) ten, (50) or even nineteen (51) times as likely to win if they are represented by counsel, in comparison to unrepresented tenants. Others talk in terms of represented tenants faring better "[a]t every stage of the proceeding" or more generally in avoiding having judgments entered against them. (52) Studies providing specific data show that represented tenants default less often, (53) obtain better settlements, (54) or win more often at trial. (55)

Unrepresented tenants are steamrolled by the courts' operation. Some reports highlight the manner in which unrepresented tenants are silenced by the court. (56) Others report that the courts do the work for landlords, while holding tenants to higher standards of proof. (57) Thus, Barbara Bezdek refers to the Baltimore Rent Court's systematic bias in favor of landlords, (58) while Paris Baldacci describes New York City Housing Court as "a largely one-sided eviction apparatus" in which pro se litigants are systematically silenced by the court system. (59) Describing the Philadelphia courts, David Eldridge's analysis finds that there is little landlords can do to undermine their position of strength, and that they appear to win independently of representation: "all landlords and their attorneys act as repeat players, reinforcing judges' orientation to eviction by virtue of their complaint." (60)

2. Family Law Cases

Available studies consistently show that representation is an important variable affecting case outcomes in the area of family law. Regarding legal custody, in their landmark study, Eleanor Maccoby and Robert Mnookin found that parents represented by counsel were more likely to request joint legal custody; not surprisingly, where both parties were represented, the outcome involved joint legal custody 92% of the time, compared to 77% of the time when only one parent was represented. (61) Regarding physical custody, the most common outcome was that the mother obtained physical custody, but representation remained a significant variable. When the father alone was represented, the mother obtained physical custody only 49% of the time, while that figure rose to 86% when the mother alone was represented. (62)

Two other studies provide data consistent with aspects of the Maccoby-Mnookin findings on representation and custody. Jane Ellis found that parents were more likely to elect some form of shared decisionmaking when both parties were represented by counsel. (63) Overnight and daytime visits scheduled per year with the nonprimary residential parent also occurred most frequently when both parties were represented. (64)

The Maryland study, Families in Transition, reports custody regression results correlating favorable outcomes for sole custody with representation. (65) Sole custody was awarded to the mother in 54.8% of the cases in which only the mother was represented, compared to 13.4% of the cases in which only the father was represented; the father won sole custody in 16.2% of the cases in which only the father was represented, compared to 7.1% of the cases where neither party was represented. (66) Joint legal custody with physical custody to the mother was the most common result when both parties were represented, while joint legal custody with physical custody to the father rarely occurred unless the father alone was represented. (67)

In contrast to the data on custody, Families in Transition reports that representation correlated differently, with an increased chance of obtaining a financial award. For example, women received an alimony award in 18% of the cases in which both parties were represented, but only 2-4% of the cases where one or both parties was without counsel; similarly, women were most likely to receive other monetary awards or a share of the pension when both parties were represented. (68) While the results may seem inconsistent with the typical correlation providing an advantage where a represented party faces an unrepresented one, a different correlation more likely explains the results. As the studies of unmet legal needs and unrepresented litigants reveal, the most common reason for appearing without counsel is the inability to pay for counsel. (69) If more money means a greater likelihood of retaining counsel, it likely also means a greater ability to pay a financial award to the opposing party; poorer litigants can be expected both to appear without counsel at a higher rate and to be less able to pay a financial award.

When the primary questions shift from divorce, custody, and financial awards to issues focused on domestic violence, the results continue to reflect the importance of counsel. For example, Jane Murphy's study of women seeking intervention in the legal system for domestic violence reveals that having an attorney substantially increased the rate of success in obtaining a protection order: 83% of women who had an attorney were successful in getting the order, compared to only 32% of women without an attorney. (70) In a different inquiry, Amy Farmer and Jill Tiefenthaler sought to explain the 21% decline in violence against women by intimate partners between 1993 and 1998. (71) The authors concluded that one of the three significant factors explaining the decline is the increased provision of legal services for victims of intimate partner abuse. (72)

While the studies consistently demonstrate the important role of counsel, the analysis is complicated for a variety of reasons. "Domestic disputes, unlike other civil disputes, are difficult to assess regarding winners per se." (73) Moreover, the studies do not always examine the same types of proceedings. (74) Complicating the analysis further are variables including gender, violence, whether the relationship is classified as high or low conflict, changing standards and presumptions in the area of custody, and attempts to assess the resolution over time, rather than solely at the time of the divorce judgment. (75) Even the extent to which attorneys play a positive or negative role is a matter of dispute. (76) Despite the analytical difficulties, however, the presence of counsel is a key variable impacting the outcomes of family law cases.

3. Small Claims Cases

A third area of unmet needs involves consumer cases, often heard in small claims courts. (77) The plethora of studies of small claims courts, many from the 1960s, '70s, and '80s, reveal a pervasive trend where unrepresented individuals are severely disadvantaged by the court process. (78) The studies consistently show a high rate of default among debtors. (79) Debtors who do not default often are pressured into settling their cases without trial. (80) Whether the cases were resolved by settlement or trial, representation played a crucial role in impacting which party obtained favorable judgments, and the size of the award for plaintiffs. (81) The picture of the unrepresented litigant pitted against the skilled lawyer informs the raw data regarding the impact of counsel on case outcomes.

The problems persist for the typical unrepresented individual in small claims courts in the 21st century, (82) despite the inordinate attention garnered by the courts in the 1970s and early 1980s and repeated calls for reform. (83) Paula Hannaford-Agor and Nicole Mott's study of cases from Lake County Circuit Court in Waukegan, Washington, found that the represented litigant was more likely to receive a judgment in his or her favor. (84) Justice Howard Dana's 2005 study of 288 small claims cases in Portland, Maine, found that lawyers increased the plaintiffs' success rate by 25% (measured in terms of increasing "great" outcomes and reducing "poor" ones) when they faced unrepresented defendants, while a lawyer for the defendant increased the success rate by 67% when pitted against an unrepresented plaintiff. (85) A four-part Boston Globe Spotlight Series, aptly titled Debtors' Hell, portrays a small claims court system in the Greater Boston area in which the dignity and rights of debtors are streamrolled by proceedings tilted toward creditors. (86) When repeat players sue individuals in small claims courts, representation dramatically increases the defendants' likelihood of success.

C. Administrative Agencies

Available studies consistently reveal the importance of representation as a crucial variable in improving the success rate of appeals. (87) Data from Social Security disability appeals, unemployment appeals, immigration appeals, and other administrative appeals typically show that the success rate is 15-30% greater when the claimant is represented. (88) While the jump is consistent across many studies, the levels of success vary not only based on the type of benefit case involved, but the grounds for appeal, nature of the claim, and procedural posture as well. As a result, the increased chances for success with representation can range from a 50% increase according to some studies involving unemployment and social security disability appeals, to increases in which representation makes the chances of success at least two or three times more likely in certain immigration appeals. (89)

1. Social Security Disability Appeals

Herbert Kritzer, reporting success rates approximately 15-30% higher for represented claimants, (90) the Maryland Action Plan (24% gap), (91) and William Popkin (23% gap), (92) provide consistent data confirming that represented claimants fare far better than unrepresented ones in their social security disability appeals. Kritzer reports on another study from the 1970s, the Boyd Report, finding that appellants represented by attorneys were successful in 78.4% of the cases, while unrepresented appellants succeeded in only 28.3% of the cases. (93)

The presence of counsel, of course, is not the sole factor influencing the outcome of cases, as reflected by Popkin's data on benefits cases at the reconsideration stage. In those cases, both represented and unrepresented claimants fared poorly (20% and 23% success rates, respectively). (94) Finally, the impact of gender, race, and culture are relevant to the discussion of outcomes. "Recent studies indicate that racial, gender, and cultural biases may play a role in a significant number of ALJ determinations, particularly in benefits hearings that affect some of society's most disadvantaged members." (95) The Ninth Circuit Gender Bias Task Force concluded that "many of the 'facially neutral' aspects of SSA disability determinations may have gender-differentiated impacts," with credibility determinations of particular concern. (96) The components of gender and race presumptively impact the discussion of representation and success rates for the same reasons people generally appear without counsel, highlighting the underlying connections between race, gender, and poverty.

2. Unemployment Cases

Studies of unemployment compensation appeals consistently show that represented claimants win more often than unrepresented ones. Kritzer's comprehensive study of unemployment compensation appeals in Wisconsin demonstrated that represented claimants fared better regardless of the nature of the claim. (97) Represented claimants won 44.2% of the cases in which they appealed, compared to 29.7% for unrepresented claimants; when broken down by the nature of the claim, a comparable gap remained for appeals involving misconduct cases, with a smaller gap for cases in which the employee allegedly quit. (98) The same did not hold true for employers. While represented employers prevailed in 58.4% of the appeals, unrepresented employers prevailed almost as often (57.3%). (99) Earlier studies also reported both a 15% gap between represented and unrepresented claimants, but no variation in success rates based on representation for employers. (100)

Kritzer theorizes that the representatives for claimants might be more likely to accept cases with potential merit, whereas employers might be more likely to retain an outside representative in problematic cases, partially explaining the difference in the impact of representation between employers and claimants. (101) He nonetheless concludes that, "absent an independent judgment about the merits of individual cases, we can only find patterns that seem consistent with the view that representation in unemployment appeals hearings enhances the represented parties' chance of success." (102)

3. Immigration Cases (103)

Donald Kerwin reports that represented immigrants obtain relief in removal proceedings at significantly higher levels than those without representation, regardless of the nature of the proceedings. (104) For example, in asylum cases, success rates were 39% for represented, non-detained persons, compared to 14% for unrepresented, non-detained persons, dropping to 18% and 3%, respectively, where the persons were detained. In suspension of deportation cases, 62% of represented, non-detained persons received relief, compared to only 17% of unrepresented, non-detained persons, with the figures dropping to 33% and 0%, respectively, where the persons were detained. (105)

Andrew Schoenholtz and Jonathan Jacobs similarly found a dramatic difference in the success rates for represented asylum seekers as opposed to unrepresented ones: represented asylum seekers referred through the affirmative process, in which the applicant applies for asylum prior to the initiation of removal proceedings, were six times more likely to be granted asylum than unrepresented ones. (106) Those placed in the defensive posture, which occurs if the INS apprehends an individual before he files an affirmative application, were four times more likely to be granted asylum if they were represented. (107)

Beyond the data analyzed by Kerwin and by Schoenholtz and Jacobs, the importance of counsel is well known to those familiar with INS proceedings. The U.S. Commission on International Religious Freedom found success rates of 25% in cases of asylum seekers with legal representation who were initially caught at a port of entry without proper documents, compared to 2% approval rates for those without representation. (108) Law review articles consistently note that "the intervention of counsel seems likely to increase an alien's chances of prevailing against the INS." (109) Efforts to expand the right to counsel to various immigration proceedings underscore the importance of counsel and the barriers facing unrepresented litigants, (110) with the call most urgent for immigrant children. (111) Even immigration judges recognize the need for appointment of counsel for unaccompanied minors:

   What is necessary, then to insure fairness for the unaccompanied
   juvenile who appears in Immigration Court charged with being
   removable from the United States? ... Because Immigration Court
   proceedings are essentially legal in nature, the first and most
   obvious answer to that question seems to be that legal
   representation be made available for the juvenile alien. (112)

4. Other Administrative Appeals

Studies of state tax appeals, (113) Federal Employees Compensation Act ("FECA") claims, (114) and claims for veterans benefits, (115) all report a jump in success rates of 15% to 20% for represented claimants compared to unrepresented ones. Regarding welfare "fair hearings", the data is both murkier and dated. Two reports describe minimal or no benefit from attorney representation, (116) while others show a sizable benefit. (117) Recent data from Massachusetts falls within the 15-20% gap in success rates. (118)

At first blush, Karl Monsma and Richard Lempert's data of representation in public housing evictions over a twenty-year period in Oahu, Hawaii, suggests that represented claimants fare worse than unrepresented ones. (119) A closer evaluation of the data, however, suggests a more complex correlation. Tenants were more likely to be represented in behavioral cases, with more serious consequences, than in nonpayment or falsification cases. (120) The outcomes varied based on type of case; moreover, while the legal aid lawyers fared poorly early in the study, their success rate increased during the period of the study. (121)

Monsma and Lempert offer two theories to explain the increased lawyer effectiveness. First, they speculate that the lawyers had more success as the board's proceedings became more formal. (122) An alternative, but by no means mutually exclusive, explanation for the increase in lawyer effectiveness over time is that it reflects the increasing experience of legal aid lawyers with the eviction board. In the early years, the eviction board was an unfamiliar forum for Legal Aid Society of Honolulu ("LASH") lawyers, and different cases were handled by different lawyers. In the later periods first one paralegal and then another handled most cases. Thus, they gained the advantage of being repeat players before the board. (123)

The connection between the success rate of representation and the skill and/or strategies of the lawyers is consistent with at least two other reports involving administrative agencies. Ronald Hammer and Joseph Hartley speculate that the absence of a gap in the success rate for represented claimants in welfare cases was tied to the test-case strategy pursued by Milwaukee Legal Services during the period studied. (124) Kritzer interprets his data throughout as showing that skilled, expert representation, as opposed to any form of representation, best explains the gaps in success rates in his data. (125)

III. ASSISTANCE SHORT OF FULL REPRESENTATION

A. Pro Se Clinics, Self-Help Centers and Hotlines

Efforts to develop reliable evaluation tools for limited assistance programs have produced a growing number of reports shedding light on the operation of various programs, their impact, and the need for continued evaluation. Until the past few years, the effectiveness of limited assistance initiatives seemed to be measured by the impressions people had of programs, rather than hard data. Beth Henschen's American Judicature Society Report of activities in rural jurisdictions captured the flavor by noting that, while most program partners report that their jobs had been made easier by the pro se assistance programs, very few programs evaluated their programs in any formal way from the perspective of litigants. (126) Inside legal services, one 2001 compilation of pro se clinics in Massachusetts and New England yielded information on an array of programs, but with little hard data. (127)

Existing data showed more promise in the family area than the housing area, the two most common subject areas for pro se clinics. Clinics in the family areas reported successes in helping clients obtain divorces and orders of protection. (128) In the housing area, while some programs reported that participants got "better results" or had a high level of client satisfaction, at least two other programs concluded that their clinics, in which tenants were advised how to handle their cases pro se, were largely ineffective unless paired with assistance in court. (129) Neil Steiner found that, while tenants assisted by a clinic at the Legal Services Center in Jamaica Plain retained possession more often than unrepresented tenants without help, represented tenants still fared far better than those who received clinic help. (130) A different program found that, while 15% of tenants retained possession after pro se instruction alone, when attorneys subsequently assisted the tenant in court-based mediation under limited representation agreements, the figure jumped to 58%. (131) Recognizing the need for assistance in court, programs turned increasingly to court-based lawyer-for-the-day programs, and continued to find that advice alone in the hallways was typically insufficient to help pro se tenants, while only representation in mediation and in the courtroom affected case outcomes. (132)

Reports beyond Massachusetts similarly showed that court-based programs could make a significant impact, particularly in areas of family law, in terms of reaching large numbers of litigants, providing them access into the legal system, and obtaining tangible results. Deborah Rhode and Ralph Cavanaugh, examining uncontested divorces in Connecticut in the mid-970s, found no substantial benefit from lawyers in the pool of cases studied, where the litigants largely agreed on custody, financial issues, and the need for the dissolution. (133) Michael Millemann described the Maryland Family Law Assisted Pro Se Experiment in which law students, supervised by lawyers, assisted 4,400 people over a seventeen-month period. (134) Millemann concluded that consumer satisfaction was highest when largely mechanical justice was involved, in the sense of helping people complete paperwork needed to perform relatively simple tasks. (135) Participation of legal services lawyers in the Family Law Access Partnership Program ("FLAPP"), providing assistance at Facilitator Locations in Riverside, California, found a high level of success in terms of the percentage of clients reporting that they obtained the outcome they sought and dollar benefits achieved in terms of child support. (136) Early reports of the Self Help Access Center in Sonoma County (137) and three pilot programs involving Family Law Information Centers in California revealed that a large number of litigants received assistance, felt positively about their experience and the outcomes received, and followed through on the advice given. (138) The reports, however, do not include specific data measuring case outcomes beyond the reports from the clients.

Comprehensive evaluations of the Van Nuys Legal Self-Help Center in Los Angeles, California, found that visitors, many of whom are poor and have limited educational backgrounds and ability to communicate in English, were very satisfied with the assistance, while court clerks and staff reported that the Center reduced demands on them. (139) In terms of case outcomes, the 2001 report tentatively concluded that, in the family area, visitors were more likely to complete their dissolutions, and in a timely manner, than non-Center visitors. In the eviction cases, however, while visitors were more likely to file answers and get to trial, "[c]enter visitors fare no better (and no worse) in UD cases than the general population." (140) By 2003, the researchers were able to confirm the high level of satisfaction and reduction in confusion and anxiety amongst litigants assisted. (141) In terms of outcomes, Center-assisted litigants fared better than unassisted litigants in family cases, (142) yet the report again found that outcomes experienced by Center-assistance litigants are indistinguishable from those of unassisted pro per litigants. (143) The one striking difference supported by the data was that Center-assisted litigants consistently agree to pay landlords higher amounts of back rent than unassisted pro per litigants. (144)

The 2007 Domestic Abuse Self-Help Project ("DASH") from Los Angeles reflects many of the same trends. Ninety-eight percent of litigants responding to a survey described the DASH services to be good or very good, and court personnel spoke highly of the program as well. Yet, only 10% of the litigants provided responses to the survey. (145) Moreover, the impact on case outcomes is unclear, and the evaluation found concerns about the effectiveness of self-help assistance when the respondent is represented by an attorney in court:

   In domestic violence cases when the balance of power between the
   parties is extremely unequal to begin with ... the respondent
   having legal representation at the hearing has a tendency to
   intimidate a petitioner/victim even more, effectively vitiating any
   self-help preparation. (146)

Beyond pro se clinics and self-help centers, hotlines also provide assistance short of full representation. (147) As hotlines became an explicit focus of the Legal Services Corporation, the need to assess their effectiveness became paramount. (148) Jessica Pearson and Lanae Davis reported that when outcomes could be determined, hotline cases were roughly evenly split between successful (48%) and unsuccessful (52%), with callers tending to prevail when they understood what they were told and followed the advice. (149) Clients who rated their outcomes most favorably "were significantly more likely to be white, English-speaking, [and] educated at least to the eighth-grade." (150) Brief services, coaching a client on how to deal with a private party or proceed in court, or providing written information were more likely to yield favorable outcomes. (151) In terms of substantive areas, housing and consumer cases yielded more favorable outcomes, while family cases were more apt to be pending. (152) Other hotline surveys reported favorable outcomes for a majority of cases. (153)

Critics, however, raise questions both about how the data should be interpreted and the framework in which hotlines are being evaluated in the first place. For example, Ross Dolloff reads the Pearson and Davis data as producing a less than 8% success rate for callers in terms of case outcomes. (154) He also articulates the inherent risks in relying on reported satisfaction without knowing what clients mean by favorable, and whether their expectations were marginal in the first place. Dolloff's ultimate concern is that the focus on a single delivery mechanism will divert attention from the more important inquiry about determining overall effectiveness of a delivery system, and that such a discussion needs to flow from the values and goals sought for the system as a whole. (155)

B. Assessing the Early Evaluation Efforts of Assistance Programs

The perspectives of John Greacen, Richard Zorza, and Bonnie Hough, leaders in the efforts nationally to expand and improve initiatives involving self-representation, are crucial in making sense of the early evaluation efforts. Although Greacen's 2002 summary of "what we know" includes the conclusions that self-represented litigants universally appreciate the programs, are highly satisfied with the services, and report that their understanding of the law and legal processes is improved, Greacen notes that those reports are not very reliable. Court staff universally appreciate the programs, as do most judges; and, for the most part, self-represented litigants are less likely to require hearings and proceed through the court much faster. (156) Greacen discusses not simply the extent self-represented litigants use the courts and programs, but the difficulties in determining the "justness" of court outcomes. (157)

Hough's 2005 assessment concludes that: (1) self-help centers are heavily used; (2) they increase Access to Justice but are far from filling the gap; (3) customer satisfaction is extremely high; and (4) self-help programs help the court become more efficient and effective. In an effort to target where self-help programs can do the most good, Hough encourages identification of factors that make self-help unrealistic. (158) Zorza similarly identifies goals beyond serving large numbers of litigants and improving customer satisfaction, including increasing access to the justice system relative to community need, preserving the community's perception of the increase in Access to Justice, and raising the rate of outcomes that are "just." (159)

Despite the comparatively nascent state of the existence and evaluation of self-help programs, and the dangers of comparing answers produced from different questions, the reports provide important insights for the inquiry of where and when counsel is most needed in legal proceedings. The court-based programs, primarily serving litigants in the housing and family areas, seem to have their greatest success in terms of providing some access to litigants who otherwise would have none, easing the strain on the court system, and leaving its customers with a high level of satisfaction with the services received. The impact on ease outcomes is harder to gauge, with successes more common in the family area than in the housing area, absent a greater level of attorney involvement.

IV. OBSERVATIONS AND ANSWERS

A. Beyond Representation

While the presence of counsel can dramatically affect case outcomes, that factor is only one variable. Other key variables include the substantive law, the complexity of the procedures, the individual judges, and the over-all operation of the forum.

1. The Substantive Law

In theory, the substantive rights should be the primary determination of the case outcome: where the law and facts favor a party, that party should obtain a favorable outcome. Tenants stand a better chance of avoiding eviction where the law affords them protections through the implied warranty of habitability, the right to quiet enjoyment, consumer protection statutes, and existence of rent control. Vulnerable mothers stand a better chance of maintaining custody and making ends meet where the law values their past role as primary caretaker, considers exposure to domestic violence in the custody evaluation and awards child support and alimony regularly and at higher levels.

The reports are replete with examples of the interplay between the substantive rights and the outcome. Chester Hartman and David Robinson discuss the inadequacy of legal protections for tenants in many jurisdictions as a crucial factor in the phenomenon of widespread displacement and homelessness, compounded by the weakening of tenant protections in recent years. (160) Advocates in Massachusetts reported a sharp increase in eviction proceedings following the elimination of rent control. (161) The evaluation of the Van Nuys Self-Help Center attributed unfavorable outcomes in many eviction cases to the fact that the users of the center simply did not have a case. (162) Michael Millemann's account of the Maryland Family Law Assisted Pro Se Experiment includes a call for law reform on substantive issues. (163) The data from social security, unemployment, and immigration appeals, while consistently showing a disparity in outcomes based on representation, also show different success rates based on the type of proceeding. (164)

At the same time, however, the reports reveal that unrepresented litigants often are unable to obtain favorable outcomes even when they appear to have the law on their side. The stunning regularity with which unrepresented tenants lose in housing courts, independent of whether they raise defenses or not, underscores this point. (165) The powerlessness of the litigants in courts such as the Baltimore Rent Court, and the litigants' inability to articulate their claims, provide a particularly poignant illustration as to why meritorious claims might not be considered by a court. (166) The high rate of default in debt collection cases likely means that some debtors with viable defenses are not even gaining access to the legal system. (167) The disparity reported by Jane Murphy in the importance of counsel in obtaining protective orders (83% as compared to 32%) (168) is too stark to support a conclusion that only women with meritorious claims find their way to counsel. The law matters, but by no means fully explains what happens in court.

2. The Complexity of the Procedures

Simplification has become an important theme, with the increased focus on self-representation and the changes within the court system over the past decade. (169) Millemann's analysis of the need for law reform includes a discussion of procedural reform. (170) Reports from self-help centers and hotlines suggest that limited assistance is most effective where the assistance needed primarily involved help completing forms to obtain access to the court process.

Relatively simple procedures do not, alone, provide a forum in which all claimants with meritorious claims prevail. Small claims courts are designed to provide simple access to the courts, yet poorer claimants routinely are steamrolled during the course of the process. Eviction cases would be greatly simplified if the landlord did not have to prove his prima facie case, or tenants were not allowed to raise defenses. Simplified child support or alimony calculations help only if the levels of awards are higher than they would have been absent the standards. As with the substantive law, the complexity of the procedures is only one variable in the analysis.

3. The Judge: Individual Practices and Perspectives

Assuming cases are resolved with judicial involvement, the judge is an important player, impacting case outcomes. Kritzer's descriptions of the hearing processes for social security and unemployment cases reflect the heavy role played by the individual judges in shaping the flow of evidence, not to mention the conclusions drawn from the evidence. (171) Data from the U.S. Department of Homeland Security on immigration cases reveal dramatically the disparity in outcomes based on the identity of the judge. (172) Eldridge's study of Philadelphia's housing court found that two of the four significant independent variables related to the judge. (173) The practices of individual judges likely helped to explain variations among counties in Missouri in their response to domestic violence. (174)

Since most cases in court settle, individual judges affect cases beyond their decisions on the merits. The extent to which a judge is flexible in assisting litigants without counsel varies tremendously, and greatly impacts the proceeding. (175) Whether judges pressure unrepresented litigants to settle their cases in the unmonitored hallway settings, or make comments designed to induce settlement is crucial as well. (176)

4. The Court's Operation

In addition to the practices of individual judges, the overall operation of the court and the messages the court conveys to litigants play important roles in affecting case outcomes. Reports of housing cases describe the extent to which housing courts operate as eviction mills, routinely producing swift judgments in landlords' favor, whether by default, consent, or after hearing; Eldridge refers to the "judges' orientation to eviction by virtue of [the landlords'] complaint" while Rand describes the system as one created "to work in a landlord's favor." (177) Barbara Yngvesson and Patricia Hennessey observed a similar favoritism afforded to plaintiffs in small claims courts, which they attribute to the joint assumptions that small claims cases were by their nature simple, and that plaintiffs' complaints were presumptively valid. (178) That an employer's representation bears little correlation to the case outcomes may suggest a presumptive orientation in unemployment hearings as well. (179) The Missouri study examining the differing responses by communities to domestic violence described the interrelated roles of law enforcement officials, prosecutors, and judges, operating as "part of an interdependent system." (180)

Beyond its presumptive favoritism to certain categories of claimants, the court's operation is relevant in terms of the system-wide procedures relating to self-representation. Some courts have adopted procedures that move them toward what Zorza terms "The Self-Help Friendly Court." (181) Beyond the user-friendly nature of forms and procedures, the manner in which the critical players in the system beyond the judges--such as clerks and court-connected mediators--perform their roles is also a crucial component of a court's operation. (182) Finally, oversight by top court officials will impact the behavior of individual judges, positively or negatively, sending intended or unintended messages to judges about how to manage their dockets. (183)

B. Representation: What We Know

1. The Importance of Power

A unifying theme of the reports and analyses regarding representation and limited assistance programs is the importance of power. The variables that provide advantages to some parties and disadvantages to others can be understood as sources of power, or the lack of power. At a minimum, each variable discussed above provides a source of power. Where the law favors landlords, creditors, employers, or the government, that source of power will be stacked against tenants, debtors, and claimants. (184) Where the procedural rules are complex, those familiar with the forum or with representation will better navigate the system, while those unfamiliar and unrepresented will be tripped up. (185) Where judges favor one category of litigants, such as landlords or employers, that dynamic provides a third source of power. Where housing courts seem geared to provide judgments for the landlords, and small claims courts operate to benefit business plaintiffs, the orientation of the forum provides a fourth source of power.

The litigants themselves arrive at the court with varying degrees of power. The business interests in small claims courts, employers in unemployment hearings and larger landlords in eviction cases stand in a position of power. Repeat players are more likely to wield financial power, utilize a forum that serves their interests, benefit from the substantive law, and be familiar with the procedure. (186)

Unrepresented litigants often are vulnerable: typically poor, disproportionately people of color, and often unfamiliar with the forum. Many unrepresented litigants face barriers in terms of language, literacy, education, disability, or age. Single parents face the added hurdle of navigating the court system with children in tow. Where ongoing relationships are involved, the power dynamic between the parties is also important. (187) Finally, legal representation provides a source of power. Skilled lawyers utilize the substantive law, navigate the procedures, and tailor their cases to the particular judges. They avoid certain judges, maneuver their cases to others, and understand the dynamics of a particular forum. Lawyers may be superfluous where certain landlords, employers, and business interests already benefit from the operation of housing courts, agencies hearing unemployment cases, and small claims courts. Yet, a skilled lawyer can also neutralize the power that the unrepresented litigant typically encounters, providing a vulnerable, one-shot litigant with the benefits of repeat-player status.

2. Not Just Any Advocate: The Importance of Knowledge and Expertise

Kritzer's observations about social security and unemployment hearings provide an important window into the importance of having not just any advocate, but one with specialized expertise. While represented claimants fare better than unrepresented ones in both settings, representatives did not perform equally. In the social security setting, Kritzer describes the importance of a "knowledgeable, experienced" advocate and articulates the differences between the inexperienced, experienced, and "very best" advocates. (188) With unemployment cases, noting that inexperienced law students and low-paid advocates for employers tend to be less effective than other representatives, Kritzer concludes that it is the combination of general advocacy skills, knowledge of specific hearing practices and players (e.g., how the administrative law judges run hearings), and substantive knowledge of the statutes, regulations, and decisions governing the payment of unemployment compensation ("UC") that characterizes the most effective advocates. Formal training (in the law) is less crucial than day-to-day experience in the UC setting. (189)

Millemann describes a model in which law students operate under the supervision of experts in the relevant area of practice. (190) Jan Hagen's study of welfare fair hearings illustrates how representatives help claimants make use of procedural due process in advancing their claims. In addition, "the relevance and effectiveness of petitioners' presentation of oral evidence were rated lower for petitioners without legal representation than for those with legal representation." (191) Farmer and Tiefenthaler tie the decline in domestic violence, in part, not simply to the expansion of the availability of lawyers generally, but to the dramatic increase in legal services programs specifically serving victims of domestic violence. (192) Monsma and Lempert theorize that the increase in the success rate for represented litigants in public housing hearings was due in part to the growing expertise of the representatives. (193) Sandefur's meta-analysis posits that "relational expertise"--skill at negotiating the interpersonal environments in which professional work takes place--may be a crucial component explaining the success representatives have in assisting vulnerable litigants who otherwise might be ignored by decisionmakers. (194)

Advocates with the skills described by Kritzer--general advocacy skills, knowledge of the forum, and knowledge of the law--are repeat players. Representation for poorer tenants is almost exclusively provided by legal services lawyers familiar with housing cases; in contrast, efforts to increase representation for tenants by utilizing pro bono lawyers from large law firms have been hampered by "the complexity of housing law and Housing Court practice and the volunteers' lack of expertise in these areas." (195)

C. Applying What We Know From The Reports

1. Implications for Moving Forward

One component of the question of where counsel is most needed involves the importance of the interests at stake. A second involves a prediction of effectiveness: counsel would be provided in cases in which full representation is likely to affect the case outcome, but lesser forms of assistance will not. Matching appropriate levels of assistance to the power needed for litigants is a crucial starting point in understanding where counsel is most important or where self-help programs might suffice.

A civil right to counsel should be developed as a component to a coherent Access to Justice strategy. (196) I articulate elsewhere a three-pronged approach, involving:

(1) the expansion of the roles of the court system's key players, such as judges, court-connected mediators and clerks, requiring them to assist unrepresented litigants as necessary to prevent forfeiture of important rights;

(2) the use of assistance programs, rigorously evaluated to identify which most effectively protect litigants from the forfeiture of rights; and

(3) the adoption of a civil right to counsel where the expansion of the roles of the key players and the assistance programs do not provide the necessary help to vulnerable litigants. (197)

While we may not know the full range of cases in which nothing short of counsel will suffice, we know that the greater the power lined up against a litigant and the more vulnerable the litigant, the greater the likelihood litigants will forfeit important rights, absent representation. (198) Based on power dynamics alone, counsel presumptively would be more important in a custody case where the opposing party is represented by counsel and where the litigant is a victim of domestic violence. Counsel in small claims cases would be more important where the plaintiff is a business interest or other repeat player. Counsel presumptively would be needed in a higher percentage of eviction cases given the orientation of the housing courts, and the data showing that unrepresented tenants lose swiftly regardless of the landlord's representation. Limited representation models vary in terms of the extent of the assistance provided and who provides assistance. The greater the power imbalance, the more extensive the assistance and the greater the skill level of the advocate need to be. (199) If fewer hurdles face unrepresented litigants, a lower level of involvement might suffice.

We know further that not just any representative will suffice where litigants are most vulnerable. Attorneys with strong advocacy skills, knowledge of the forum, and knowledge of the law presumptively will be the most effective in a particular scenario. The importance of skilled advocates increases relative to the power stacked against the unrepresented litigants. Pro bono lawyers operating in their specific areas of expertise would count among those ranks. With law students, or pro bono lawyers operating outside their areas of expertise, unskilled and untrained advocates are less effective absent appropriate supervision.

With lay advocates, Kritzer's data indicates that the success rate of skilled lay advocates can rival that of skilled attorneys in certain settings. Ralph Cavanaugh and Deborah Rhode found no empirical justification for the unauthorized practice of law restrictions in the area of uncontested divorces. (200) Authors of one study from England conclude that "specialization, rather than professional status, seems to be the best guarantee of such protection." (201) Given the organized bar's repeated resistance to lay advocacy, the use of lay advocates will involve political considerations as well as concerns about competence. (202)

2. The Need for Clarity of Goals and Values

The reports underscore the need for a careful discussion of goals and values at every turn. Case outcomes are not the sole valid evaluative measure for assessment. Proponents of limited-assistance programs identify alternative goals involved in the establishment of self-representation initiatives, including achieving customer satisfaction, easing strains on the court, and increasing litigants' understanding of court processes. (203) Nothing in this Article is intended to undercut the validity of those goals. Yet, a system that achieves a high level of satisfaction is not necessarily a response to the inquiry of where counsel is most essential, absent a decision that satisfaction trumps case outcomes. (204) The decision involves the allocation of scarce resources, parallel to those such as how to structure a legal services office to best meet the community's needs. The mix of an office's programs and priorities will vary according to its goals and mission. (205) Even where the goal is defined in terms of case outcomes, programs might be structured differently, depending on how we assess the importance of the rights at stake. For example, right to counsel initiatives have focused recently on family and housing areas in recognition of the importance of those rights to litigants. (206) Yet, even within the areas of family law and housing, the level of child support is not necessarily of equivalent importance to custody, nor is the need for repairs presumptively as important as possession of one's home. Similarly, while debtors in small claims court may be as powerless as tenants in housing court, we may not value the avoidance of debt collection as highly as we do the preservation of one's home.

Beyond prioritizing the interests at stake and identifying the features that suggest counsel is most likely to affect the outcome, additional choices involve how to articulate the threshold below which an outcome is unacceptable. For example, the United States Supreme Court's due process analysis suggests the need to evaluate not simply based on the interest at stake as well as the government's interest, but the "risk of error" absent the protections. (207) A program designed to fill that void would be trying to identify cases in which there is a risk of error absent the presence of counsel.

Different cases might receive counsel, however, with different formulations of the outcomes to be avoided. Instead of using a "risk of error" analysis, programs might decide counsel is most important: (1) where counsel is likely to impact the outcome of the cases; (2) where the result absent counsel would be "unjust"; (208) or (3) where basic needs are at stake (the standard articulated in the ABA Resolution). Each threshold involves a different set of cases from either the "risk of error" or "impact the outcome" approaches. The potential differences in standards might affect both the structure of programs and the design of future research.

V. QUESTIONS AND COMPLICATIONS

While the reports shed light on the impact of counsel in case outcomes, they raise questions as well that complicate the analysis. One question involves whether the results might reveal less about the impact of counsel, and more about how clients decide when to retain lawyers, and how lawyers decide which cases to accept. A second question involves the possibility that positive outcomes might be viewed as negative ones from a different perspective. Other questions involve the role expectations play in the analysis and the thorny question of resources. A final complication relates to differences in impacts in the short and long term. The balance of this section discusses these questions.

A. Questioning the Data

The reports suggest that the disparities in case outcomes between represented parties and unrepresented ones show that lawyers impact case outcomes. An alternative reading might focus instead on lawyer and client behavior. Lawyers might "cherry-pick," selecting their cases based on the likelihood of success, thereby taking cases in which litigants might succeed even without counsel. Clients might make decisions to retain counsel based on the merits of the case, so that clients with weaker cases are less likely to retain counsel.

Regarding the factor of case selection by lawyers, a few studies, such as Seron's randomized study of New York City Housing Court, control for this variable in program design and still reveal a significant improvement in results for represented parties. (209) Kritzer and Sandefur, while acknowledging that cherry-picking might play a role, conclude that the disparities are too stark to provide a complete explanation. (210)

Regarding decisions by clients, the most common reason given by litigants for their decisions to appear without counsel is that they could not afford counsel, not that they felt there was no merit to their case. (211) Even assuming that behavioral choices of both lawyers and clients explain some portion of the data, those characteristics do not undermine the overall findings supported by the studies. Those concerns simply underscore the analysis in this article that representation is only one of a number of variables explaining case outcomes. (212)

B. When Positive Outcomes Could Be Negative Ones

Related to complications that might arise from unarticulated or competing goals is the reality that particular outcomes that are positive from one perspective might be negative from another, if counsel increases the likelihood that women obtain the restraining orders they seek, that will be viewed as a positive for some, but a negative for those believing that restraining orders are too freely given. (213) A similar problem arises in analyzing the question of whether self-representation clogs the courts. John Greacen's tentative conclusion from the existing data in 2002 was that hearings and trials in domestic relations cases take less time when self-represented litigants are involved. (214) The question remains as to whether speed is a good or bad thing. Where unrepresented litigants are steamrolled in housing court, slowing down the system is an important goal.

Whether one believes that lawyers complicate cases or preserve litigants' rights, affects the assessment as to whether the lawyer's role is positive or negative. The question of speed exposes conflict among the multiple goals identified by Zorza and others. "Efficiency" and easing the strain for overburdened court personnel might, depending on the prescription, come at the price of empowerment and Access to Justice.

C. The Role of Expectations

The role of expectations also complicates the discussion. As noted above, customer satisfaction is an important evaluative tool for measuring the success of hotlines and self-help programs. Whether it is a positive or negative result that a program increases customer satisfaction without impacting case outcomes is part of the conversation on values and goals. (215) If tenants expect to lose in housing court, and landlords expect to win, advice to the tenant that explains the process but fails to affect the outcome might lead to satisfied landlords and tenants. Representation by counsel for tenants might decrease the satisfaction of landlords expecting to regain possession.

The 2003 evaluation of the Van Nuys Center revealed the curious finding that center-assisted litigants consistently agreed to pay landlords higher amounts of back rent than unassisted litigants. (216) Professor Gary Blasi, involved in the first Van Nuys evaluation, suggests that the center raised expectations by explaining how the legal system should work, as opposed to how it likely would work. (217) The Center's possible role of what Blasi terms the "dispenser of norms," contrasts starkly with, for example, Gary Bellow's description of "political lawyering," which includes the willingness to challenge the status quo. (218) Whether our programmatic choices, by design or unintended consequence, dispense norms or challenge them bears watching.

D. The Thorny Question of Resources

At the heart of programmatic decisions is the question of how best to allocate resources. Within legal services programs, more money allocated toward hotlines or limited-assistance programs will lead to more clients being served. However, where assistance does not affect case outcomes to the extent full representation does, questions of values and trade-offs arise. Moreover, a comprehensive Access to Justice approach does not necessarily involve only the resources allocated to legal services organizations. As courts embrace their role in assisting unrepresented litigants, the trade-offs involve larger pools of resources, including the courts and private bar.

For example, is it sensible to add additional judges to high-volume courts? If the goal is to reduce the number of cases per judge, allowing more time per case, the answer might well be yes. But if the implication is that unrepresented litigants will thereby receive better treatment in the courts, the questions of goals, values and measures of success reappear, if judges with lower caseloads are unwilling or unable to equalize imbalances of power to the extent needed to impact case outcomes, the additional resources might be misallocated. Whether the resources are better directed to self-help programs, or appointment of counsel, depends on the dynamics of the particular court.

The question of resources affects the programmatic decisions involved in targeting the expansion of counsel as well. The more complicated the analysis, the more resources will be devoted to sorting; as a result, fewer resources will be available for representation. (219) Moreover, the longer the sorting process, the later in the case counsel will appear, likely decreasing the effectiveness of counsel. (220)

Finally, the costs relating to the harm that flows from the absence of counsel must be considered. Douglas Colbert, studying bail hearings, focused on the link between counsel at bail hearings and reduction in incarceration. (221) Calculating the costs involved with preventable evictions or lost custody is far more complicated. Yet, even if the exact dollar savings for eviction defense is open to question, the connection between eviction defense and homelessness prevention is sufficiently established so that the cost savings are far higher than $0. (222) Articles by Laura Abel, and by John Roman, Mischelle Van Brakle, and William Turner underscore the need to develop a reliable methodology in assessing the difficult question of cost estimates, and therefore cost-benefit evaluations. (223) Despite the difficulties, an evaluation of costs involved in representation requires an equivalent assessment of savings, if one method reduces the cost per client, without assessing cost savings achieved from the differences in case outcomes, the analysis is incomplete, and often misleading. (224)

E. Short-Term Versus Long-Term Change, Stability, Fluidity, and Backlash

A final complicating factor is that programmatic changes that yield benefits in the short-run may have different impacts over time. Marc Galanter's analysis of the advantages of repeat player status reflects this reality. (225) Even where repeat players suffer short-term losses, they adapt their behavior to realize long-term gains in a system. (226) The same analysis likely applies to the impact of counsel. Counsel for tenants can be effective in part because the tenant is unrepresented in most cases. Landlords dedicate few resources to each case because they typically prevail, regardless of the quality of their lawyer, the persuasiveness of their evidence, or the presence of a viable defense. Counsel for tenants can be effective, in part, because the landlords are not prepared to prevail in a true adversarial trial, and more willing to lose a case or two than change their practices.

Dramatic changes might have a greater impact in the short term than in the long term unless the underlying power dynamic is altered as well. If most tenants suddenly appear with counsel, the manner in which the court routinely processes evictions would grind to a halt. In the long run, however, powerful landlords can be expected to adapt, the court procedures would change, and the repeat player likely would regain control. (227) Representation is only one important variable affecting case outcomes, along with the substantive law, the procedures, the decisionmaker, and the operation of the forum. (228)

CONCLUSION

The amount of assistance that is appropriate and necessary requires assessment both of the importance of what is at stake in a given proceeding and of where an advocate is most likely to affect the outcome. Those working in Access to Justice communities, including in the courts and in self-representation programs, have a common interest in identifying the scenarios in which the presence of counsel is most likely to impact case outcomes. Programs providing assistance short of full representation need to understand where their assistance is most meaningful, and where a referral to programs providing counsel is a better use of resources. Programs providing counsel, whether in legal services offices, through the private bar, or through other mechanisms, similarly need to understand when counsel has the greatest impact.

For the civil Gideon movement, accurate data helps identify promising starting points for expanding a civil right to counsel. Those insights should inform an Access to Justice strategy, making counsel available where basic needs are at stake and nothing short of full representation can protect those needs. Existing studies shed light on the scenarios in which counsel matters most and suggest areas for future research.

(1.) Paul Marvy and Rebecca Sandefur have put together the most extensive compilation of reports. Marvy was Coordinator of the Committee for Indigent Representation and Civil Legal Equality ("CIRCLE") of the Northwest Justice Project. Northwest Justice Project, http://www.nwjustice.org/(last visited Nov. 13, 2009). Sandefur lists reports as references in her forthcoming article. Rebecca Sandefur, Elements of Expertise: Lawyers' Impact on Civil Trial and Hearing Outcomes 34-42 (Mar. 26, 2008) (unpublished manuscript, under review, on file with author). For a discussion of many of the reports, see Russell Engler, And Justice for All--Including the Unrepresented Poor: Revisiting the Roles of the Judges, Mediators and Clerks, 67 FORDHAM L. REV. 1987 (1999).

(2.) See infra Part I.

(3.) As always, care must be taken when using data and studies. For a discussion of pitfalls and tips to avoid them, see Sarah H. Ramsey & Robert F. Kelly, Assessing Social Science Studies: Eleven Tips for Judges and Lawyers, 40 FAM. L.Q. 367 (2006). The tips include identifying the research questions, being cautious with causal claims, considering how the data were analyzed, and assessing the practical significance of the research findings. Id.

(4.) Sandefur, supra note 1, at 24.

(5.) The focus on power is hardly a novel concept. See, e.g., LESLIE BENDER & DAAN BRAVEMAN, POWER, PRIVILEGE AND LAW: A CIVIL RIGHTS READER (1995).

(6.) Laura Abel of the Brennan Center for Justice has identified the following questions: 1) To what extent does legal representation lead to more accurate decision making in different types of cases?; 2) When will assistance short of full legal representation suffice?; 3) What are the costs and what are the savings of various approaches to increasing access to the courts? Laura K. Abel, Pressing Questions Encountered by the Access to Justice and Civil Right to Counsel Movements, Law & Society Association Presentation 2, 5, 8 (July 7, 2006) (unpublished manuscript, on file with author).

(7.) LEGAL SERVS. CORP., DOCUMENTING THE JUSTICE GAP IN AMERICA: THE CURRENT UNMET CIVIL LEGAL NEEDS OF LOW-INCOME AMERICANS 9-13 (2005) [hereinafter LEGAL SERVS. CORP. 2005 REPORT], available at http://www.lsc.gov/justicegap.pdf. Virtually "all of the recent state studies found a level of need substantially higher than the level found in [a] 1994 ABA study." Id. at 13. The worst recession since the Great Depression has dramatically increased the number of Americans whose basic human needs are at issue in legal proceedings and need counsel. LEGAL SERVICES CORP., DOCUMENTING THE JUSTICE GAP IN AMERICA: THE CURRENT UNMET CIVIL LEGAL NEEDS OF LOW-INCOME AMERICANS--AN UPDATED REPORT OF THE LEGAL SERVICES CORPORATION 5 (2009), available at http://www.lsc.gov/pdfs/documenting_the_justice_gap_in_america_2009.pdf ("The current economic crisis, with its attendant problems of high unemployment, home foreclosures, and family stress has resulted in legal problems relating to consumer credit, housing, employment, bankruptcies, domestic violence, and child support, and has pushed many families into poverty for the first time.").

(8.) LEGAL SERVS. CORP. 2005 REPORT, supra note 7, at 11.

(9.) See Alan W. Houseman, The Future of Legal Aid." A National Perspective, 10 UDC/DCSL L. REV. 35, 43-46 (2007).

(10.) See infra Part II.

(11.) See, e.g., Engler, supra note 1; JONA GOLDSCHMIDT ET AL., MEETING THE CHALLENGE OF PRO SE LITIGATION: A REPORT AND GUIDEBOOK FOR JUDGES AND COURT MANAGERS 8-10 (1998).

(12.) JOHN M. GREACEN, SELF-REPRESENTED LITIGANTS AND COURT AND LEGAL SERVICES RESPONSES TO THEIR NEEDS: WHAT WE KNOW 3-6 (2002), available at http://www.courtinfo.ca.gov/programs/cfcc/pdffiles/SRLwhatweknow.pdf [hereinafter GREACEN, WHAT WE KNOW]. For an edited version of the article, see John M. Greacen, An Administrator's Perspective: The Impact of Self-Represented Litigants on Trial Courts--Testing Stereotypes Against Real Data, 41(3) JUDGES' J. 32 (2002) [hereinafter Greacen, An Administrator's Perspective]. See also Engler, supra note 1, at 2047-52. Surveys of litigants in New York City found that 75% of the litigants in the New York City Family Court and 90% of the tenants in Housing Court appeared without counsel. OFFICE OF THE DEPUTY CHIEF ADMIN. JUDGE FOR JUSTICE INITIATIVES, SELF-REPRESENTED LITIGANTS: CHARACTERISTICS, NEEDS, SERVICES: THE RESULTS OF TWO SURVEYS 1 (2005) [hereinafter Two SURVEYS].

(13.) See infra Parts II.A, II.C.

(14.) See, e.g., Engler, supra note 1, at 2048; GOLDSCHMIDT ET AL., supra note 11, at 11; GREACEN, WHAT WE KNOW, supra note 12, at 3-6. The 2005 surveys from New York City found that 79% of the self-represented litigants in Family and Housing Court were African-American or Hispanic. Two SURVEYS, supra note 12, at 3. Regarding income, 21% had incomes below $10,000, an additional 36% had incomes between $10,000 and $20,000, and another 26% had incomes between $21,000 and $30,000; thus 83% of the self-represented litigants had incomes below $30,000. Id. at 4. Spanish-speaking litigants had less formal education than English-speaking litigants. Id. at 5.

(15.) See, e.g., Two SURVEYS, supra note 12, at 7 (60% of the litigants reported that they could not afford counsel); Engler, supra note 1, at 2027; BOSTON BAR ASS'N TASK FORCE ON UNREPRESENTED LITIGANTS, REPORT ON PRO SE LITIGATION 17 (1998), available at http://www.bostonbar.org/prs/reports/unrepresented0898.pdf ("Most of the unrepresented litigants [in the Boston Housing Court] reported that they wanted an attorney but felt they could not afford one."); N.H. SUPREME COURT TASK FORCE ON SELF-REPRESENTATION, CHALLENGE TO JUSTICE: A REPORT ON SELF-REPRESENTED LITIGANTS IN NEW HAMPSHIRE 2, 4 (2004), available at http://www.ajs.org/prose/pdfs/NH%20report.pdf ("A sample of self-represented litigants in New Hampshire showed that most of them were in court on their own because they could not afford to hire or continue to pay a lawyer."). Data from the British Columbia Supreme Court's Self-Help Center found that 75.9% of the Center's users listed "I cannot afford to hire a lawyer" as the reason for not having a lawyer. JOHN MALCOLMSON & GAYLA REID, BC SUPREME COURT SELF-HELP INFO. CTR., FINAL EVALUATION REPORT 31 (2006), available at http://www.pblo.org/lawyers/search/item.123035.

(16.) Conferences addressing the issue of unrepresented litigants included the National Conference on Self-Represented Litigants Appearing in Court (Scottsdale, AZ, 1999); The Changing Face of Legal Practice: A National Conference on Unbundled Legal Services, (Baltimore, MD, 2000); the Massachusetts Statewide Conference on Unrepresented Litigants (Worcester, MA, 2001); the New York State Unified Court System Access to Justice Conference (Albany, NY, 2001); and the Eastern Regional Conferenee on Access to Justice for the Self-Represented Litigant (White Plains, NY, 2006).

Publications include those by the American Judicature Society and State Justice Institute, such as GOLDSCHMIDT ET AL., supra note 11; JONA GOLDSCHMIDT & LISA MILORD, JUDICIAL SETTLEMENT ETHICS: JUDGES' GUIDE (1996) [hereinafter JUDGES' GUIDE]; CYNTHIA GRAY, REACHING OUT OR OVERREACHING: JUDICIAL ETHICS AND SELF-REPRESENTED LITIGANTS (2005), available at http://www.ajs.org/prose/pdfs/Pro%20se%20litigants%20final.pdf; BETH M. HENSCHEN, LESSONS FROM THE COUNTRY: SERVING SELF-REPRESENTED LITIGANTS IN RURAL JURISDICTIONS (2002). See also RICHARD ZORZA, THE SELF-HELP FRIENDLY COURT: DESIGNED FROM THE GROUND UP TO WORK FOR PEOPLE WITHOUT LAWYERS (2002).

For information on the web, see the Pro Se Resources page of the American Judicature Society's website, AJS Publications and Resources, http://www.ajs.org/prose/proresources.asp (last visited Nov. 13, 2009); the website funded by the State Justice Institute serving as a network for practitioners of self-help programs, Self Help Support, http://www.SelfHelpSupport.org (last visited Nov. 13, 2009); and Self-Represented Litigation Network, http://www.srln.org (last visited Nov. 13, 2009).

(17.) See, e.g., GOLDSCHMIDT ET AL., supra note 11, at 25; GRAY, supra note 16; Rebecca A. Albrecht, John M. Greacen, Bonnie Rose Hough and Richard Zorza, Judicial Techniques for Cases Involving Self-Represented Litigants, 42(1) JUDGES' J. 16, 18 (Winter 2003); Engler, supra note 1, passim; John M. Greacen, Legal Information vs. Legal Advice-Developments During the Last Five Years, 84 JUDICATURE 198, available at http://www.ajs.org/prose/pro_greacen.asp.

(18.) Engler, supra note I, at 2003-06; Houseman, supra note 9, at 40-43. For resources available on the web, see supra note 16.

(19.) In 2000, the Conference of State Court Administrators called on the courts to provide Access to Justice for those without counsel. CONFERENCE OF STATE COURT ADMINISTRATORS, POSITION PAPER ON SELF-REPRESENTED LITIGATION (2000), available at http.//cosca.nesc.dni.us/WhitePapers/selfreplitigation.pdf. The following year, the Conference of Chief Justices promulgated Resolution 23, titled "Leadership to Promote Equal Justice," resolving in part to "[r]emove impediments to access to the justice system, including physical, economic, psychological and language barriers." Resolution 23, http://ccj.ncsc.dni.us/AccessToJusticeResolutions/reso123Leadership.html (last visited July 17, 2006). In 2002, the two Conferences jointly issued Resolution 31, resolving that "courts have an affirmative obligation to ensure that all litigants have meaningful access to the courts, regardless of representation status. Resolution 31, http://ecj.nesc.dni.us/resol31AsstPgmsSlfLitigants.html (last visited Oct. 28, 2009).

(20.) See, e.g., Robert Echols, The Rapid Expansion of "State Access to Justice" Commissions, MGMT. INFO. EXCHANGE J., Summer 2005, at 41; Houseman, supra note 9, at 60-62. For more information, see the supporting webpages of the American Bar Association, ABA Standing Committee on Legal Aid & Indigent Defendants, http://www.abanet.org/legalservices/sclaid/atjresourcecenter/atjmainpage.html (last visited Nov. 13, 2009), and National Legal Aid & Defender Association (NLADA), http://www.atjsupport.org/SPAN_Library/SPAN_Report (last visited Nov. 13, 2009).

(21.) 372 U.S. 335 (1963).

(22.) See, e.g., Special Issue, A Right to a Lawyer? Momentum Grows, 40 CLEARINGHOUSE REV. 163 (2006) [hereinafter Momentum Grows]; Symposium, Civil Gideon." Creating a Constitutional Right to Counsel in the Civil Context, 15 TEMP. POE. & CIV. RTS. L. REV. 697 (2006).

(23.) Conferences across the country included panels on civil Gideon as part of the broader discussion of Access to Justice in civil cases. For example, the 2002 Washington State Access to Justice Conferenee included a civil Gideon panel, while the Keynote Speaker for New York's 2001 Access to Justice Conference was Justice Earl Johnson, Jr., who dedicated his remarks to the topic of civil Gideon (conference materials on file with author).

(24.) Brennan Center for Justice, Civil Right to Counsel, http://www.brennancenter.org/content/section/eategory/civil_right_tocounsel/ (last visited Nov. 13, 2009); National Coalition for a Civil Right to Counsel, http://www.eivilrighttoconnsel.org/(last visited Nov. 13, 2009).

(25.) See, e.g., Paul Marvy & Debra Gardner, A Civil Right to Counsel for the Poor, 32 HUM. RTS. 8, 9 (2005).

(26.) See, e.g., King v. King, 174 P.3d 659 (Wash. 2007); Frase v. Barnhart, 840 A.2d 114 (Md. 2003).

(27.) See, e.g., H.R. 2124, 79th Leg., Reg. Sess. (Tex. 2005) (relating to appointment of counsel in appeals of certain eviction suits). Advocates in California are drafting a model statute. See Clare Pastore, The California Model Statute Task Force, 40 CLEARINGHOUSE REV. 176 (2006). Advocates in New York recently expanded the reach of the statutory right to counsel in divorce cases to the Supreme Court, and drafted and arranged for the filing of legislation initiatives covering the elderly in eviction and foreclosure cases. See N.Y. JUD. LAW [section] 35(8) (2007); Manny Fernandez, Free Legal Aid Sought for the Elderly, N.Y. TIMES, Nov. 16, 2007, available at http://www.nytimes.com/2007/11/16/nyregion/ 16housing.html?_r=l&oref-=slogin.

(28.) The ABA House of Delegates adopted the resolution on August 7, 2006, at its Annual Meeting. AM. BAR ASS'N, RESOLUTION ON CML RIGHT TO COUNSEL, available at http://www.abanet.org/media/issues/ civiljustice/civiljustice_abapolicy.pdf.

(29.) See infra Part II.B.1.

(30.) Carroll Seron et al., The Impact of Legal Counsel on Outcomes for Poor Tenants in New York City's Housing Court: Results of a Randomized Experiment, 35 LAW & SOC'Y REV. 419, 423-26 (2001). By prior agreement with the Administrative Judge of the Civil Court, control and treatment cases were rotated through the same three judges. Id. at 424. Five variables were selected to test the effect of the program on substantive legal outcomes, and four additional variables were selected to measure the effect of the program on the efficiency of the Court. Id. at 426.

(31.) See, e.g., infra notes 36-40.

(32.) See, e.g., infra notes 41, 56.

(33.) See infra Parts II.B.2-3.

(34.) See infra Part II.C.

(35.) See infra Part III.

(36.) For an overview of studies discussing the eviction process, see Chester Hartman & David Robinson, Evictions: The Hidden Housing Problem, 14 HOUSING POL'Y DEBATE 461, 477-78 (2003).

(37.) THE WILLIAM E. MORRIS INST. FOR JUSTICE, INJUSTICE IN NO TIME: THE EXPERIENCE OF TENANTS IN MARICOPA COUNTY JUSTICE COURTS (2005) [hereinafter INJUSTICE IN NO TIME].

(38.) LAWYER'S COMM. FOR BETrER HOUSING, NO TIME FOR JUSTICE: A STUDY OF CHICAGO'S EVICTION COURT (2003) [hereinafter No TIME FOR JUSTICE].

(39.) Anthony J. Fusco, Jr. et al., Chicago's Eviction Court: A Tenant's Court of No Resort, 17 URB. L. ANN. 93, 114-16 (1979) (noting a study of Chicago's Eviction Court covering 1976 through 1978 estimating that landlords were represented in 80.3% of the cases, while tenants were represented in 7.1% of the cases). The article is based on the findings of the 1976 report published as J. BIRNBAUM, N. COLLINS & A. FUSCO JR., JUDGMENT LANDLORD: A STUDY OF EVICTION COURT IN CHICAGO (1978).

(40.) AM. CIVIL LIBERTIES UNION, ACCESS TO JUSTICE PROJECT, JUSTICE EVICTED: AN INQUIRY INTO HOUSING COURT PROBLEMS (1987) [hereinafter JUSTICE EVICTED].

(41.) MONITORING SUBCOMM. OF THE CITY WIDE TASK FORCE ON HOUSING COURT, 5 MINUTE JUSTICE OR "Aim" [SICI NOTHING GOING ON BUT THE RENT!" (1986) [hereinafter 5 MINUTE JUSTICE]. 5 MINUTE JUSTICE and JUSTICE EVICTED are only two of many reports on or accounts of New York City Housing Court over the past thirty-five years. See, e.g., 144 Woodruff Corp. v. Lacrete, 585 N.Y.S.2d 956, 960 (N.Y. Civ. Ct. 1992); 5 MINUTE JUSTICE, supra, at 43, 57, 59-60, 73-74; COAL. FOR THE HOMELESS ET AL., STEMMING THE TIDE OF DISPLACEMENT: HOUSING POLICIES FOR PREVENTING HOMELESSNESS 63-64 (1986); COMM. TRAINING AND RESOURCE CTR. & CITY-WIDE TASK FORCE ON HOUSING COURT, INC., HOUSING COURT, EVICTIONS AND HOMELESSNESS: THE COSTS AND BENEFITS OF ESTABLISHING A RIGHT TO COUNSEL iii (1993); COURT STUDY GROUP OF THE JUNIOR LEAGUE OF BROOKLYN, REPORT ON A STUDY OF THE BROOKLYN LANDLORD AND TENANT COURT 21 (1973) (finding that 32% of represented tenants obtained favorable outcomes, compared to 18% of unrepresented ones; in contrast, representation had minimal impact on outcomes for landlords); JUSTICE EVICTED, supra note 40, at 26; KIRA KRENICHYN & NICOLE SCHAEFER-MCDANIEL, RESULTS FROM THREE SURVEYS IN NEW YORK CITY HOUSING COURTS (2007), available at https://www.policyarchive.org/bitstream/handle/10207/8683/ threesurveys.pdf?sequence=1; Steven Brill, The Stench of Room 202, AM. LAW., Apr. 1987, at 1; Jan Hoffman, Chaos Presides in New York Housing Courts, N.Y. TIMES, Dec. 28, 1994, at Al. Despite a 1997 program by top New York State court officials designed to effectuate sweeping changes in processing pf cases in the Housing Court, JUDITH S. KAYE & JONATHAN LIPPMAN, BREAKING NEW GROUND (1997), the continued plight pf unrepresented tenants were reflected not only in Seron's study but the 2004 Conference and 2005 Report of the New York County Lawyers' Association. N.Y. COUNTY LAWYERS ASS'N, THE NEW YORK CITY HOUSING COURT IN THE 21ST CENTURY: CAN IT BETTER ADDRESS THE PROBLEMS BEFORE IT? (2005).

(42.) Barbara Bezdek, Silence in The Court: Participation and Subordination of Poor Tenants' Voices in Legal Process, 20 HOFSTRA L. REV. 533 (1992).

(43.) Erica L. Fox, Alone in the Hallway." Challenges to Effective Self-Representation in Negotiation, 1 HART. NEOOT. L. REV. 85 (1996) (discussing the disadvantages of hallway settlements between the parties themselves).

(44.) For example, in Maricopa County, Arizona, while approximately 87% of landlords were represented by counsel, not a single tenant was represented. INJUSTICE IN NO TIME, supra note 37, at 1-2, 8-9. In a Berkeley, California study, only 20.4% of tenants were represented, as opposed to 83.4% of landlords. Hartman & Robinson, supra note 36, at 477 (citing REBECCA HALL, EVICTION PREVENTION AS HOMELESSNESS PREVENTION: THE NEED FOR ACCESS TO LEGAL REPRESENTATION FOR LOW-INCOME TENANTS (1991)). A 2005 study from Boston found a greater disparity in representation rates than previously reported, with landlords represented in 85% of the cases, compared to 7% for tenants. MASS. LAW REFORM INST., SUMMARY PROCESS SURVEY 14 (2005) [hereinafter MLRI, 2005 SURVEY]. A similar study from 1995 found that 67.8% of landlords were represented, while no tenants in the study were represented. MASS. LAW REFORM INST., SUMMARY PROCESS SURVEY 14 (1995) [hereinafter MLRI, 1995 SURVEY]. These figures compared as follows to representation rates statewide: 66.7% versus 11.5% (1995), 52% versus 7% (1998-89), and 66% versus 6% (2005). See MLRI, 2005 SURVEY at 3; MASS. LAW REFORM INST., SUMMARY PROCESS SURVEY 9 (1998-99); MLRI, 1995 SURVEY at 5. In Cambridge, Massachusetts, a recent examination of 365 cases found that landlords were represented in 355 of them (97.3%), compared to tenants represented in 39 (10.7%). Jennifer Greenwood et al., Tenancy at Risk: Leveling the Playing Field 16 (May 2008) (unpublished report, on file with author). In Hartford, Connecticut, 16% of tenants were represented, compared to 85% of landlords. Hartman & Robinson, supra note 36, at 477-78 (citing RAPHAEL L. PODOLSKY & STEVEN O'BRIEN, A STUDY OF EVICTION CASES IN HARTFORD: A FOLLOW-UP REVIEW OF THE HARTFORD HOUSING COURT (1995)). Numerous studies of the New York City Housing Court found an imbalance in representation rates (90-95% for landlords, 5-10% for tenants). See supra note 41. In a New Haven, Connecticut study, while 73% of landlords were represented by counsel, 83% of tenants were not. Steven Gunn, Note, Eviction Defense for Poor Tenants: Costly Compassion or Justice Served?, 13 YALE L. & POL'Y REV. 385, 411 (1995).

(45.) Bezdek's study of Rent Court in Baltimore, Maryland, found that the tenants are typically unrepresented poor black women. Bezdek, supra note 42, at 560. A New Haven, Connecticut study found that tenants often are extremely poor members of minority groups and women. Gunn, supra note 44, at 393. The various New York City studies have found the same--a vulnerable population of unrepresented tenants, often poor women of color, and many with a limited understanding of English. See id.

(46.) For example, in Chicago, landlords won at the same rate regardless of whether they were represented. Fusco et al., supra note 39, at 116. In Detroit, represented landlords obtained completed judgments at almost the same rate as unrepresented ones (86% versus 83.6%). Marilyn Miller Mosier & Robert A. Soble, Modern Legislation, Metropolitan Court, Miniscule Results: A Study of Detroit's Landlord-Tenant Court, 7 U. MICH. J.L. REFORM 8, 38 fig.13 (1973).

(47.) Regarding the speed of the proceedings, see supra notes 38-44. In Maricopa County, for example, most cases took less than a minute to hear, and many cases were heard in less than twenty seconds, with judgments overwhelmingly favoring landlords. INJUSTICE IN NO TIME, supra note 37, at 1-2, 8-9. In Chicago, the 2003 study reported that hearings lasted an average of 1 minute and 44 seconds, down 50% from the 3 minutes observed in 1996. No TIME FOR JUSTICE, supra note 38, at 11; see also id. at 9, 18 (discussing LAWYER'S COMM. FOR BETTER HOUSING, TIME TO MOVE: THE DENIAL OF TENANTS' RIGHTS IN CHICAGO'S EVICTION COURT (1996)).

Regarding the minimal judicial oversight, reports, articles, and judicial decisions from the 1970s, 1980s and 1990s, the New York City Housing Court consistently found the resolution of most cases by means of pressured, unmonitored hallway settlements between a landlord's lawyer and unrepresented tenant, and minimal judicial oversight. Fox's 1995 observations of hallway settlements reveal the difficulties unrepresented tenants face attempting to assert their rights in the unmonitored, hallway negotiations with landlords' lawyers. Fox, supra note 43. According to an op-ed piece in The Washington Post, most litigants facing eviction in Washington, D.C., settle their cases through a "lopsided consent judgment process" which pits unrepresented tenants against landlords represented by lawyers, with the result that many tenants sign away their rights, and the court does little or nothing to protect them. Julie Becker, Gimme Shelter, WASH. POST, Oct. 27, 2002, at B8.

(48.) Gunn, supra note 44, at 414 tbl. 18 (finding that tenants represented by legal services lawyers were more than three times as likely to avoid eviction as were unrepresented tenants (23% versus 7%)).

(49.) No TIME FOR JUSTICE, supra note 38, at 18 (finding that tenants with legal representation were "six times more likely to prevail"). A study of 763 cases from 2002 found that, while represented tenants obtained continuances more often than unrepresented ones (32% versus 13%), in the small sample of cases that went to trial, tenants were evicted regardless of representation: "in all eases in which a defense was raised the tenant lost." Id. at 5, 18.

(50.) HALL, supra note 44, at 2 (finding that represented tenants were ten times more likely than unrepresented ones to win in court).

(51.) David L. Eldridge, The Making of a Courtroom: Landlord-Tenant Trials in Philadelphia's Municipal Court 65-69, 130-42 (2001) (unpublished Ph.D. dissertation, University of Pennsylvania, on file with author) (studying 153 hearings in Philadelphia's Landlord-Tenant Court). Eldridge found that tenants were nineteen times more likely to win their cases when an attorney represented them. Id. at 135-37. He employed a multi-method design to test the significance of eight variables, four of which ultimately proved to be significant. Id. at 135. The strongest association was between tenant representation and hearing outcome. Id. at 135-37. Two other significant variables related to the judge, where identity or particular comments indicated a pro-landlord orientation, Id. at 135. Eldridge's earlier study (1996) of a smaller sample found no apparent association between whether the tenant raised a defense and the hearing outcome, with judges finding in favor of landlords in 95% of contested evictions. For a description of the 1996 "TAG" Study, see id. at 41.

(52.) Russell Engler & Craig S. Bloomgarden, Summary Process Actions in Boston Housing Court: An Empirical Study and Recommendations for Reform 5 (May 20, 1983) (unpublished manuscript, on file with author). See also Fusco et al., supra note 39, at 114-16 (finding that represented tenants avoided summary possession at the initial hearing at a greater rate than unrepresented tenants (61.3% versus 15.8%), obtained continuances more easily (34.7% versus 2.9%), obtained possession more often (13.3% versus 6%), and had a greater likelihood of obtaining dismissals without adverse action (17.3% versus 7.8%)); Mosier & Soble, supra note 46, at 35-38 (finding that represented tenants lost completely to the landlord at the initial heating less frequently (27.6% versus 81.9%), obtained more dismissals (31.2% versus 3.7%), and more frequently avoided a judgment for possession or all rent to the landlord (87.7% versus 46.2%) compared to unrepresented tenants); Seron et al., supra note 30, at 426 tbl.2 (finding that represented tenants were less likely to have judgments entered against them (31.8% versus 52%) and have warrants of eviction entered against them (24.1% versus 43.5%)). In addition, landlords obtained possession in over 97% of all cases, including almost 85% on contested cases. Paula Hannaford-Agor & Nicole Mott, Research on Self-Represented Litigation, Preliminary Results and Methodological Considerations, 24 JUST. SYS. J. 163, 171 (2003) (finding that represented parties were more likely to see judgments in their favor, with dismissals or judgments for plaintiffs most likely if neither side was represented). The Hannaford-Agor & Mott report used data from 2000 as part of a study of data from five courts, and including landlord-tenant disputes in Lake County, Illinois. "Because the landlord is likely to be the plaintiff, the power relationship between landlord and tenant outside the courtroom is perpetuated in the courtroom, especially without representation." Id.

(53.) See, e.g., Seron et al., supra note 30, at 426 (finding that fewer represented tenants defaulted, 15.8-28.2%); Engler & Bloomgarden, supra note 52, at 5, 29, 51 (studying 500 eviction cases in Boston Housing Court and finding that virtually all defaulting tenants were unrepresented).

(54.) See, e.g., PODOLSKY & O'BRIEN, supra note 44 (finding that represented tenants obtained stipulations of settlement more advantageous to their interests); Seton et al., supra note 30, at 429 (finding that represented tenants were more likely to obtain settlement terms that required rent abatements or repairs); Engler & Bloomgarden, supra note 52 (finding that regarding stipulations, represented tenants were more likely to have judgment entered in their favor (31.7% versus 5.1%), or to have the agreements include an order for repairs, a waiver or reduction of rent, or even the payment of money to the tenant; the disparities applied whether the settlement occurred in the presence of court mediators (Specialists) or in the hallway).

(55.) Blue Ribbon Citizens' Committee on Slum Housing, Executive Summary 2-3 (n.d.) (unpublished manuscript, on file with author) (empirical investigation of the use of the implied warranty of habitability in the Municipal Court of the Los Angeles Judicial District, reviewing data from 1997 & 1998). Of fifty-one tenants who attempted to defend their eviction based on conditions, not a single tenant proceeding unrepresented was successful; only tenants with lawyers won at trial, Id. For a discussion of this data, see Mary Helen McNeal, Having One Oar or Being Without a Boat: Reflections on the Fordham Recommendations on Limited Legal Assistance, 67 FORDHAM L. REV. 2617, 2642 (1999). See also Engler & Bloomgarden, supra note 52, at 8 (The disparities were less dramatic at trial than in settlement or with defaulting tenants, although the small number of represented tenants still prevailed more frequently than unrepresented ones. Of landlords, 73.8% were represented, compared to 13.6% of tenants.). A 1997 study of the effects of full and limited representation in Boston Housing Court found that represented tenants retained possession 41.3% of the time, compared to 12.5% for pro se tenants; for tenants not retaining possession, represented tenants were given an average of 129.7 days before execution issued, compared to 53.7 for unrepresented tenants. BOSTON BAR ASS'N TASK FORCE ON UNREPRESENTED LITIGANTS, supra note 15, at 16-18 (citing Neil Steiner, An Analysis of the Effectiveness of a Limited Assistance Outreach Project to Low-Income Tenants Facing Eviction (Oct. 14, 1997) (unpublished manuscript, on file with author)).

David Grossman's recent data on Boston Housing Court appears to challenge some of these findings. An examination of 1000 case files found tenants maintaining possession approximately 50% of the time, a far higher rate than found by the MLRI 2005 survey (24%), and doing so at the same rates whether represented or not. See MLRI, 2005 SURVEY, supra note 44; David Grossman, Data (n.d.) (unpublished data, on file with author). Grossman found a clear advantage for represented tenants at trial (50% versus 11%), and also in terms of the length of time, displaced tenants could stay in possession, with shorter stays for unrepresented tenants. Grossman was unable to measure the extent to which representation impacted the waiver of back rent, and identifies definitional differences he used that make comparisons to earlier reports more complicated. Nonetheless, the data appeared to reflect greater success in maintaining possession for unrepresented tenants in Boston Housing Court, which, if true, could be partially explained by changes in the court's operations, changes in personnel, changes in the housing stock, and changes in the nature of the landlords using the court and their goals. See MLRI, 2005 SURVEY, supra note 44; David Grossman, supra.

(56.) Bezdek's study of Baltimore's Rent Court analyzes "the functional voicelessness of virtually all tenants in this forum," arguing that the operational premise of the rent court serves to reinforce the rights of landlords while obscuring those of tenants. Bezdek, supra note 42, at 533-35. Tenants' social powerlessness compounds the institutional barriers raised by the rent. Id. Frank Bloch reports an in-court observation in Nashville, Tennessee, consistent with the "silencing" of an unrepresented tenant along the lines Bezdek describes. Interrupting the pro se tenant's attempt at trial to question a bookkeeper as to why her rent increased, the judge snapped, "Rent goes up all the time. Any more questions?" The tenant had none, and the landlord obtained judgment. Frank S. Bloch, Framing the Clinical Experience: Lessons on Turning Points and the Dynamics of Lawyering, 64 TENN. L. REV. 989, 999 (1997) (describing the silencing of an unrepresented tenant in an eviction case in General Sessions Court, Davidson County, Tennessee). In the subsequent case, where the tenant was represented by counsel, the same judge, after an extended colloquy, informed the landlord's lawyer that his notice was invalid and that he would have to bring a new eviction proceeding, while informing tenant's counsel that "they're going to get possession anyway, eventually." Id.

(57.) See INJUSTICE IN NO TIME, supra note 37, at 1-2, 8-9 (reporting that when tenants did appear, while the justices did not require landlords to prove their claim for possession or monetary award, the justices held tenants to a higher standard of proof for their defenses. Fewer than 20% of the tenants even appeared in court; tenants rarely had the cases against them dismissed).

(58.) See generally Bezdek, supra note 42.

(59.) N.Y. COUNTY LAWYERS ASS'N, supra note 41, at 3 (summarizing Paris Baldacci's paper, later published as Paris R. Baldacci, Assuring Access to Justice: The Role of the Judge in Assisting Pro Se Litigants in Litigating Their Cases in New York City's Housing Court, 3 CARDOZO PUB. L. POL'Y & ETHICS J. 659 (2006)).

(60.) Eldridge, supra note 51, at 142. Spencer Rand describes the system as one that reflects "that it was created by landlords to work in a landlord's favor; ... [i]t is rare that I have found a client proceed pro se and not end up with a possession order against her." Spencer Rand, Teaching Law Students to Practice Social Justice: An Interdisciplinary Search for Help Through Social Work's Empowerment, 13 CLINICAL L. REV. 459, 496-97 (2006); see, e.g., Marc Galanter, Why the "Haves" Come Out Ahead: Speculations on the Limits of Legal Change, 9 LAW & SOC'Y REV. 95, 98-103 (1974) (describing relative success of repeat players versus one-shot players in court).

(61.) ELEANOR E. MACCOBY & ROBERT H. MNOOKIN, DIVIDING THE CHILD: SOCIAL AND LEGAL DILEMMAS OF CUSTODY 108-09, 300 (1992). When the mother alone was represented, the outcome involved joint legal custody in 73% of the cases, and 88% of the cases in which the father only was represented; joint custody resulted in 51% of the cases where neither party had a lawyer, Id. at app. A tbl.5.A2. The authors studied approximately 1100 California families as they made post-separation arrangements for their children, Id. at 13. All parents filed for divorce between September 1984 and April 1985, and all had at least one child under age sixteen at the outset of the study. Id. The authors focused their study on clusters of questions surrounding gender role differentiation, legal conflict, contact in terms of maintenance and change as time passes, and co-parenting relationships. Id. at 10; see also Robert H. Mnookin et al., Private Ordering Revisited." What Custodial Arrangements Are Parents Negotiating, in DIVORCE REFORM AT THE CROSSROADS 37, 61-65 (Stephen D. Sugarman & Henna Hill Kay eds., 1990).

(62.) MACCOBY & MNOOKIN, supra note 61, at 109-10. While the authors found that mothers nearly always requested physical custody regardless of whether they were represented by counsel, the same was not true with fathers: only 21% of unrepresented fathers sought physical custody, compared to 80% of represented fathers. Id. at 111.

(63.) Jane Ellis, Plans, Protections, and Professional Intervention: Innovations in Divorce Custody Reform and the Role of Legal Professionals, 24 U. MICH. J.L. REFORM 65, 132 (1990) (studying 1988 data from Washington State). That result occurred in 91% of the cases in which both sides were represented, compared to 70% where only one party was represented and 77% when neither party had counsel. Id.

(64.) Id. at 132-33. The mean number of total visits was 120 per year where both parents had counsel, compared to 88 where only one side had counsel and 97 where neither party was represented, Id.

(65.) THE WOMEN'S LAW CTR. OF MD., INC., FAMILIES IN TRANSITION: A FOLLOW-UP STUDY EXPLORING FAMILY LAW ISSUES IN MARYLAND (2006), available at http://www.wlcmd.org/pdf/FamilieslnTransition.pdf.

(66.) Id. at 48 tbl.16. Sole custody was also awarded to the mother in 17.8% of the cases in which both parties were represented, and 41% of the cases in which neither side was represented. Id. Sole custody was also awarded to the father in 4.4% of the cases where both parties were represented, and 0 cases in which only the mother was represented. Id.

(67.) See id. Joint legal custody with physical custody to the mother resulted in 44.7% of the cases when both parties were represented, compared to 26.8% of the cases in which the mother alone was represented, 29.6% of the cases when the father alone was represented, and 25.6% of the cases when neither party was represented. Id. The 44.7% figure was by far the most common result when both parties were represented, with joint legal and physical custody the next most common result, at 20.7% of the cases. Id. Joint legal custody with physical custody to the father occurred in 15.5% of the cases when only the father was represented, compared to 7.3% when both parties were represented, 5.7% when neither party was represented, and 2% when only the mother was represented. Id.

(68.) Id. at 47 tbl.15. Alimony was awarded in 2% of the eases where only the woman was represented or neither side was represented, and 4% of the cases in which only the man was represented, Id. When both parties were represented, women received financial awards other than alimony in 28% of the eases and a share of the pension in 25% of the cases; the percentages were lower when only the woman was represented (11% and 6% respectively), only the man was represented (9% and 6% respectively), or neither party was represented (3% and 2% respectively), Id. Maccoby and Mnookin explored financial issues but without separating for the variable of representation. MACCOBY & MNOOKIN, supra note 61, at 115-31.

(69.) See supra note 16.

(70.) Jane Murphy, Engaging with the State: The Growing Reliance on Lawyers and Judges to Protect Battered Women, 11 AM. U. J. GENDER Soc. POL'Y & L. 499, 511-12 (2003).

(71.) AMY FARMER & JILL TIEFENTHALER, EXPLAINING THE RECENT DECLINE IN DOMESTIC VIOLENCE 18 (2003), available at http://4allnc.ncbar.org/media/ 4984/explainingrecentdeclineindomesticviolencestudy.pdf. The number of violent incidents dropped from 1.1 million to 876,340. Id.

(72.) Id. at 10-11. The other two significant factors were improved educational and economic status for women and demographic trends, such as the aging of the population and increase in racial diversity, Id. at 11, 15-16.

(73.) Harmaford-Agor & Mott, supra note 52, at 171. For example, whereas the question of restraining orders might lend itself to a more straightforward examination as to whether the restraining order was or was not issued, the potential outcomes in divorce cases become far more complicated.

(74.) See, e.g., supra notes 62, 71, 72 and accompanying text. With custody alone, the analysis can focus on legal or physical custody, each with the potential results of joint custody, custody for the mother alone, or custody to the father alone. See supra note 61 and accompanying text. Whether the custody requests of the parents conflict or coincide affects the analysis as well. See supra note 62 and accompanying text. Where financial issues are involved, the array of issues can include child support, spousal support, shares of pensions, and health issues. See supra note 68 and accompanying text. Counsel can affect the grounds for divorce that a party pleads, the nature of the custodial arrangement that a party seeks, and whether a party seeks a financial award. See supra note 68 and accompanying text. The presence of counsel can impact the choice of process for dispute resolution, such as whether the parties resort to an adversarial heating, reach a negotiated agreement on their own, or resolve their disputes in mediation. See supra notes 70-72 and accompanying text.

(75.) Whether the courts are biased against or in favor of women in the issuance of restraining orders and custody determinations, and whether joint parenting relationships in high conflict situations are indications of positive or negative outcomes, all remain matters of dispute. For a thoughtful review of twenty-five years of research, analysis, and discussion, see Clare Dalton, When Paradigms Collide: Protecting Battered Parents and Their Children in the Family Court System, 37 FAM. & CONCILIATION CTS. REV. 273 (1999).

(76.) For example, where attorneys help one party obtain sole custody rather than joint custody, the chance increases either that a party will put in a financial claim, or that a party will obtain a restraining order. No easy consensus exists as to whether those results are positive or negative. Moreover, divorce lawyers are notorious for aggressive litigation tactics believed to heighten the level of conflict. See, e.g., Richard E. Crouch, The Matter of Bombers: Unfair Tactics and the Problem of Defining Unethical Behavior in Divorce Litigation, 20 FAM. L.Q. 413, 413 (1986). Crouch is the former chairman of the Committee on Ethical Practices & Procedures of the American Bar Association's Section on Family. Id. Not surprisingly, the growth of the collaborative law movement and the push generally for therapeutic justice are commonly tied to the family law area. See, e.g., PAULINE H. TESLER, COLLABORATIVE LAW" ACHIEVING EFFECTIVE RESOLUTION IN DIVORCE WITHOUT LITIGATION (2001); Barbara A. Babb, Fashioning an Interdisciplinary Framework for Court Reform in Family Law: A Blueprint to Construct a Unified Family Court, 71 S. CAL. L. REV. 469, 477-78 (1998) ("Incorporation of therapeutic jurisprudence as the underlying goal of the court's operation provides an organizational philosophy around which to design the system's components."). A recent study from North Carolina found that cases were more likely to settle, either on their own or through mediation, when both parties were represented by counsel, whereas a higher percentage of cases involving a represented plaintiff and unrepresented defendant were resolved through litigation. Suzanne Reynolds, Catherine T. Harris & Ralph A. Peoples, Back to the Future: An Empirical Study of Child Custody Outcomes, 85 N.C.L. REV. 1629, 1670-71 (2007).

(77.) While many studies of small claims courts and of debt collection cases exist, some caveats must be considered before drawing conclusions from the data. First, not all consumer cases are debt collection cases. Second, not all small claims cases are consumer cases. Third, small claims courts are only one venue in which debt collection cases are heard. Despite these caveats, the combined review of reports of debt collection cases and small claims courts is quite revealing regarding the impact of counsel.

(78.) The greatest number of studies of small claims courts comes from the late 1960s into the early 1980s. See Suzanne E. Elwell & Christopher D. Carlson, The Iowa Small Claims Court: An Empirical Analysis, 75 IOWA L. REV. 433, 440 n.54 (1990) (listing twenty separate studies, some focused on particular states, and others comparing courts in different states). According to the authors of a 1975 review of small claims literature to that time, small claims is in fact the least forgotten of the lower courts today. No other lower court has received such widespread attention from lawyers, social scientists, and the concerned public in the past decade. Barbara Yngvesson & Patricia Hennessey, Small Claims, Complex Disputes: A Review of the Small Claims Literature, 9 LAW & SOC'Y REV. 219, 220 (1975). The studies typically use a combination of case file reviews, surveys or interviews of litigants and others in the system, and in-court observations; the reports paint a picture with troubling implications for a system designed to be accessible to the little person. The pervasive trend in which individuals are disadvantaged by the court process emerges from a broad analysis that: 1) looks beyond trial outcomes; 2) separates business parties from individual ones; 3) examines the types of claims; and 4) explores the process beginning with the service of the court pleadings.

(79.) David Caplovitz's classic study of consumer cases in New York, Chicago, and Detroit courts found a default rate of over 90%. DAVID CAPLOVITZ, CONSUMERS IN TROUBLE: A STUDY OF DEBTORS IN DEFAULT 8-9, 215-21 (1974). His findings are consistent with contemporaneous data from the Boston Municipal Court. See CTR. FOR AUTO SAFETY, LITTLE INJUSTICES: SMALL CLAIMS COURTS AND THE AMERICAN CONSUMER 98 (1972). Authors of an Ohio study, which found that represented proprietors were twice as likely to obtain default judgments as unrepresented ones, speculated that "counsel may act as a factor discouraging the defendant from appearing at trial." Earl Hollingsworth, William G. Feldman & David C. Clark, The Ohio Small Claims Court: An Empirical Study, 42 U. CIN. L. REV. 469, 482 (1973). Caplovitz reported that out-of-court conversations between unrepresented debtors and the creditor's lawyer often led directly to the debtors default. CAPLOVITZ, supra, at 204-05. Some debtors reported that they "[t]ried to settle and thought the court action was discontinued," while others were "[a]dvised not to go to court by plaintiff's attorney or own attorney." Id. at 205.

(80.) Caplovitz describes the typical dynamic in court:

   When the debtor does appear for a trial, he is usually summoned to
   the bench by the judge, who is anxious to clear his calendar, and
   is told to go out into the hall and work out a settlement with the
   plaintiff's lawyer. In this fashion, even debtors with valid
   defenses are pressured to make some payment.

CAPLOVITZ, supra note 79, at 218-19. A study of small claims courts in Boston and Cambridge from the early 1970s similarly found that judges encouraged "unrepresented litigant[s] to go outside and attempt a 'settlement' with the attorney representing the other side." CTR. FOR AUTO SAFETY, supra note 79, at 110.

(81.) Austin Sarat's study of New York small claims cases found that represented plaintiffs facing unrepresented defendants recovered two-thirds or more of their claimed awards 71.4% of the time, a figure that dropped to 13.5% when an unrepresented plaintiff faced a represented defendant. Austin Sarat, Alternatives in Dispute Processing." Litigation in a Small Claims Court, 10 LAW & SOC'Y REV. 341, 367 (1976). The figure was 64.1% when neither party was represented, and 40% when both parties were represented. Id. In Ohio, businesses and proprietorships were represented more often than individual plaintiffs, and prevailed more often as well. Hollingsworth, Feldman & Clark, supra note 79, at 478-82. Most plaintiffs were businesses or proprietorships, while most defendants were individuals. Id. In one county, individuals received favorable judgments in 74% of the cases going to judgment, while all but two of the 295 business plaintiffs prevailed. Id. Even where attorneys are excluded from the process, business interests still dominated the forum, prevailing against ill-equipped individuals. Id. at 478-79. Beatrice Moulton's study of the California small claims courts found that poor individuals appeared most frequently as defendants, while business interests and government agencies, many of whom filed multiple claims as part of their collection practices, were the real beneficiaries of the court. Beatrice A. Moulton, Note, The Persecution and Intimidation of the Low-Income Litigant as Performed by the Small Claims Court in California, 21 STAN. L. REV. 1657, 1659 (1969). Moulton describes a process in which the low-income defendant is intimidated by the process, "often unaware of his rights and unable to articulate his side of the story." Id. at 1664-65.

Examining fifteen small claims courts across the country, John Ruhnka et al. reported murkier data regarding plaintiffs. See JOHN C. RUHNKA ET AL., SMALL CLAIMS COURTS: A NATIONAL EXAMINATION, NATIONAL CENTER FOR STATE COURTS (1978). Plaintiffs represented at trial or having attorney assistance in trial preparation won a higher percentage of cases than those without any help or with only advice as to how to prepare. However, when measuring the percentage of the dollar value recovered in comparison to the claim, attorney services made little difference, Id. at 65 ("[P]laintiffs ... tended to win almost as often without [attorneys]."). When measured in terms of win/loss rate, represented plaintiffs (93%) and those with attorney help preparing for trial (88%) had higher success rates than those receiving advice (80%) or turned down by an attorney (81%); the overall success rate was 86%. Id. at 66 tbl.3.34. However, in terms of percentage of the claim recovered, the figure for all cases (89%) was nearly identical to attorney representation (88%), with other forms of attorney assistance correlating to a lower percentage of awards: 84% (lawyer help preparing for trial), 81% (lawyer told litigant how to prepare), and 82% (lawyer told litigant he would not handle case), Id. at 66 tbl. 3.35. Yngvesson & Hennessey's explanation of the small claims court as an operation that exhibits favoritism to plaintiffs helps explain this data. Yngvesson & Hennessey, supra note 78, at 226.

(82.) For example, an Iowa study of 1986 data showed that the majority of small claims cases filed involved business plaintiffs suing individual defendants. Represented defendants defaulted less often than unrepresented ones, and won a higher percentage of their cases overall (13% versus 3%); represented plaintiffs fared slightly better than unrepresented ones overall (98% versus 95%), with the margin spreading for cases at trial (93% versus 86%).

Elwell & Carlson, supra note 78, at 511. The authors found their data typical of reports of other small claims courts, with "plaintiff-dominated judgments, a high rate of default judgments, and suggestions that businesses and those with prior experience or attorney representation fare better in the small claims court." Id.

Sterling & Schrag's study of small claims cases filed against consumers in Washington, D.C. in 1988 found a default rate of 74%. Hillard M. Sterling & Philip G. Schrag, Default Judgments Against Consumers: Has the System Failed?, 67 DENV. U. L. REV. 357, 361 (1990). Most of the defaulting debtors later interviewed were women, and almost all were black, Id. at 364. None of the nondefaulting cases went to trial, with most resulting in a stipulated agreement in which the defendant met the plaintiff's demands, Id. at 361. A 1994 unpublished account of small claims cases in Boston reported that over 90% of collection cases resulted in default judgments; when debtors actually appeared, collection lawyers typically continued the case to a later date. Glen Hannington, Small Claims Court Abuse by Attorneys 10-12 (1994) (unpublished manuscript, on file with author). Most non-defaulting cases settled under pressure from the court with the defendant agreeing to a payment schedule. Id. at 10.

(83.) Proposals to increase jurisdictional limits, decrease or eliminate the presence of lawyers, prohibit assignees or collection companies from filing claims in small claims courts, impose limitations on the number of claims a person can file in small claims court each year, and increase the use of mediation in small claims court are all designed to return the court to its status of providing an accessible forum for low- and moderate-income citizens to obtain access to a forum for redressing common wrongs. See, e.g., Tal Finney & Joel Yanovich, Expanding Social Justice Through the "'People's Court", 39 LOY. L.A.L. REV. 769 (2006); James C. Turner & Joyce A. McGee, Small Claims Reform: A Means of Expanding Access to the American Civil Justice System, 5 UDC/DCSL L. REV. 177, 182-88 (2000); see also HALT (Help Abolish Legal Tyranny), HALT Small Claims Court Best Practices, http://www.halt.org/lic/art.php?aid=89 (last visited June 21, 2007).

(84.) Hannaford-Agor & Mott, supra note 52, at 171.

(85.) Memorandum from Justice Howard Dana on Small Claims (Portland District Court 2005--Cases 1-288) 2-3, (Oct. 26, 2006) (on file with author). Excluding the cases in which the plaintiff prevailed by default, Justice Dana found that lawyers increased the plaintiffs' success rate by 32%, while lawyers for defendants decreased the pro se plaintiffs' success rate by 43%. Id. at 3-4.

(86.) Debtors" Hell, Parts I-IV, BOSTON GLOBE, July 30-Aug. 2, 2006, http://www.boston.com/news/special/spotlight_debt/part1/page1.html.

(87.) In contrast to some of the methodological difficulties in measuring what constitutes a favorable outcome in certain court cases, the measurement of success before administrative agencies often is more straightforward. Typically, claimants are appealing the denial of benefits, and the evaluation may occur through analysis of the results of the hearings. While differences in the nature of the claim, the forum, and the parties can complicate the comparison across types of hearings, the variable of representation typically increases the chances for success, often dramatically.

(88.) See infra Parts II.C.1-4.

(89.) See infra Parts II.C.1-4. As discussed more fully below, some data also shows that the success varies based on the skill of the representative, and that, in certain scenarios, lay advocates can perform as well as lawyers in impacting the success rate for represented claimants. See infra Part II.B.2.

(90.) HERBERT M. KRITZER, LEGAL ADVOCACY: LAWYERS AND NONLAWYERS AT WORK 111-20 (1998) (studying data from Wisconsin from the mid-1970s to the mid-1990s). Kritzer found that claimants represented by an attorney were successful in 60-70% of their appeals, while unrepresented claimants succeeded at rates as low as 30% and as high as 55%. Since the higher rates in each category typically occurred in the same year, the gap in the representation status ranged from 15-30% each year. Id. at 117 chart. Kritzer also reported data regarding the success rates for claimants appearing with nonattorney representatives. For each year reported, claimants represented by nonattorneys fared far better than unrepresented ones, but slightly poorer than those represented by attorneys, Id.

(91.) Advisory Council of the Md. Legal Servs. Corp., Action Plan for Legal Services to Maryland's Poor 12, App. (Jan. 1988) (unpublished report, on file with the University of Baltimore Law Library) [hereinafter Maryland Action Plan]. The Plan discusses data from the 1980s and reported a success rate of 60% for represented claimants, and only 36% for unrepresented ones. Id.

(92.) William D. Popkin, The Effect of Representation in Nonadversary Proceedings--A Study of Three Disability Programs, 62 CORNELL L. REV. 989, 1024 tbl.1A (1977) (discussing data from the 1970s). Popkin's data from the 1970s show a success rate of 71% for represented claimants and only 48% for unrepresented ones. Id.

(93.) KRITZER, supra note 90, at 114-15. Unlike Popkin's article, the Boyd Report separately reported the success rate for nonlawyer advocates, which was 51.5%.

(94.) Popkin, supra note 92, at 1024-27. Popkin explains this difference by noting the differences in the issues involved, with the emphasis predominately on medical issues, Id. On the other hand, the benefits of counsel extend beyond the administrative appeal stage and into federal court, with the Milton Heumarm and J.L. Pottenger report of cases assigned to a pro bono panel showing that represented claimants won a remand in 81% of the cases (23% of which were with a panel attorney and 58% with other counsel), compared to only 10% of unrepresented ones. Milton Heumann & J.L. Pottenger, Jr., The Civil Pro Bono Panel of the United States District Court for the Eastern District of New York: An Evaluation 43 tbl.6 (1984) (unpublished study, on file with the Fordham Urban Law Journal). The report suggested further that representation by counsel produces judicial efficiencies by reducing the length of time a case is litigated, Id. at 45. While the disparity likely remains today, it is worth remembering that the period studied by Heumann & Pottenger covered the early years of the Reagan Administration, during which an inordinate number of claims were rejected at the agency level, leading to a high rate of reversal by the courts.

(95.) Elaine Golin, Solving the Problem of Gender and Racial Bias in Administrative Adjudication, 95 COLUM. L. REV. 1532, 1533 (1995). Racial difference was larger at the administrative appeal level than at any other level, with ALJ's allowing disability insurance benefits for 66% of white appellants in the Social Security Disability Insurance ("SSDI") program, but only 55% of black appellants. Id. at 1546-47. A comparable gap existed with the Supplemental Security Income ("SSI") program, with 51% of black applicants and 60% of white applicants receiving benefits, Id. at 1547 n. 107 (citing GEN. ACCOUNTING OFFICE, SOCIAL SECURITY: RACIAL DIFFERENCE IN DISABILITY DECISIONS WARRANTS FURTHER INVESTIGATION (1992)). For the most part, "this difference was unexplainable by demographic factors or severity and type of impairment." Id. at 1547.

(96.) Id. at 1544-45 (quoting Linda G. Mills, A Calculus for Bias: How Malingering Females and Dependent Housewives Fare in the Social Security Disability System, 16 HARV. WOMEN'S L.J. 211, 218 (1993)).

(97.) KRITZER, supra note 90, at 23-77.

(98.) Id. at 34-39. In misconduct cases, represented claimants prevailed in 50.8% of their appeals, compared to a 37.9% success rate for unrepresented claimants, Id. at 37. With "Quit" cases, the overall success of the claimants dropped, but the benefits of representation remained, as 26.3% of represented claimants prevailed in these cases, while only 17.4% of claimants won on appeal. Id. at 38. Considering all appeals, and not simply those in which the claimant appealed, represented claimants won 50.4% of the cases, compared to unrepresented claimants who prevailed in 41.5% of the cases, Id. at 34.

(99.) Id. at 34.

(100.) Murray Rubin found that representation for claimants increased the success rate by about half, from an overall success rate of 30.8% to a success rate of 45.4% for represented claimants, Id. at 32 (citing Murray Rubin, The Appeals System, in UNEMPLOYMENT COMPENSATION STUDIES AND RESEARCH 628 (1980) (analyzing 10,972 unemployment compensation appeals decided during April 1979). Maurice Emsellem and Monica Halas found a disparity of 45% success rate for represented claimants versus 34% for unrepresented claimants. Maurice Emsellem & Monica Halas, Representation of Claimants at Unemployment Compensation Proceedings: Identifying Models and Proposed Solutions, 29 U. MICH. J.L. REFORM 289, 292 (1996). Kritzer concludes that the three studies are largely consistent, despite some variation in numbers. He notes that the 1994 Ohio study was more limited, while the 1979 study did not separate data based on the nature of the issue on appeal, or which side was making the appeal, factors that impact which side has the burden of proof and might indicate the merits of the case. KRITZER, supra note 90, at 34. Regarding the impact of representation for employers, Emsellem and Halas report: "[i]n contrast, the success rate for employers remained precisely the same, sixty-five percent, whether or not they were represented." Emsellem & Halas, supra, at 292. Rubin found that the success rate actually fell for employers, from 69.2% to 54.1%, when they brought a representative. See KRITZER, supra note 90, at 32 (citing Rubin, supra).

(101.) KRITZER, supra note 90, at 32.

(102.) Id. at 37. In identifying the characteristics of an effective advocate, Kritzer ultimately concludes that formal legal training is less crucial than day-to-day experience in the unemployment cases setting, and that it is the combination of general advocacy skills, knowledge of specific hearing practices and players, and substantive knowledge of the relevant law that characterizes the most effective advocates. Id. at 76. For discussion of this point, see infra Part IV.B.2.

(103.) I include Immigration Cases in the section on Administrative Proceedings because the cases are heard in the executive, rather than in the judicial, branch. See, e.g., STEPHEN H. LEGOMSKY, IMMIGRATION AND REFUGEE LAW AND POLICY 6 (4th ed. 2005).

(104.) Data from the Executive Office of Immigration Review for Fiscal Year 2003 on approval rates in removal proceedings "reveal a strong correlation between legal representation and positive case outcomes." Donald Kerwin, Charitable Legal Programs for Immigrants: What They Do, Why They Matter and How They Can Be Expanded, IMMIGR. BRIEFINGS, No. 04-06, June 2004, at 6. The numbers are higher in each instance for nondetained immigrant persons than for detained ones. Id.

(105.) Id. Other cases similarly revealed a disparity in outcomes:

in adjustment of status cases, the figures were 87% for represented, non-detained persons compared to 70% of unrepresented, non-detained persons, dropping to 41% and 21%, respectively, where the persons were detained, id.;

in INS [section] 212(c) cases, involving waivers for legal permanent residents stopped at the border attempting to return to the United States, represented, non-detained persons won 75% of the time compared to 49% for unrepresented persons, with the figures dropping to 56% and 34%, respectively, for detained persons, id.;

in cancellation of removal cases for lawful permanent residents, 68% of the represented, non-detained immigrants received relief compared to 60% of unrepresented ones, with the figures dropping to 59% and 55%, respectively, where the persons were detained, id.;

in cancellation of removal cases for non-lawful permanent residents, 6% of represented persons prevailed while only 2% of unrepresented ones prevailed where persons were non-detained; detained persons lost all such cases, regardless of whether they were represented, id.

(106.) Andrew I. Schoenholtz & Jonathan Jacobs, The State of Asylum Representation: Ideas for Change, 16 GEO. IMMIGR. L.J. 739, 743 (2002) (analyzing data provided by the INS Asylum Office for fiscal years 1998 and 1999, and the first seven months of 2000).

(107.) Id. at 743. The impact of representation carried across the various nationalities:

              Asylum Granted:
Nationality   Relaresented      Unrelaresented

Armenian      50%               25%
Somali        54%               22%
Chinese       46%               13%
India         31%               1%
Liberia       60%               8%

Id.

(108.) Charles H. Kuck, Legal Assistance for Asylum Seekers in Expedited Removal: A Survey of Alternative Practices, in STUDY ON ASYLUM SEEKERS IN EXPEDITED REMOVAL: AS AUTHORIZED BY SECTION 605 OF THE INTERNATIONAL RELIGIOUS FREEDOM ACT OF 1998 232, 239 (2004).

(109.) Sharon Finkel, Note, Voice of Justice: Promoting Fairness Through Appointed Counsel for Immigrant Children, 17 N.Y.L. SCH. J. HUM. RTS. 1105, 1125 (2001) (quoting Margaret H. Taylor, Promoting Legal Representation for Detained Aliens: Litigation and Administrative Reform, 29 CONN. L. REV. 1647, 1665 (1997)); see also Beth J. Werlin, Renewing the Call: Immigrants' Right to Appointed Counsel in Deportation Proceedings, 20 B.C. THIRD WORLD L.J. 393 (2000); Note, A Second Chance: The Right to Effective Assistance of Counsel in Immigration Removal Proceedings, 120 HARV. L. REV. 1544 (2007) (arguing that not simply the right to counsel, but the right to effective assistance of counsel is needed in such proceedings, given the nature of the proceedings and the consequences of an adverse decision).

(110.) See, e.g., Garcia v. Hayes, No. CV07-3239 (C.D. Cal. May 16, 2007) (seeking an order requiring constitutionally adequate hearings for aliens who are detained pursuant to immigration statutes for longer than six months); AM. BAR ASS'N COMM. ON IMMIGRATION, REPORT TO THE HOUSE OF DELEGATES 1, 5-9 (2006), available at http://www.abanet.org/publicserv/immigration/107a_right_to_counsel.pdf (urging the ABA to support the "due process right to counsel for all persons in removal proceedings," given the complexity of the proceedings, the disparity in case outcomes depending on whether the asylum-seeker has legal representation, the hardships facing those seeking asylum, the systemic costs involved due to the lack of representation, and the potentially small number of persons eligible for relief). However, since the Commission on Immigration's proposal seems to call simply for the right to appear with counsel, rather than to have counsel appointed at public expense, the proposal provides limited guidance.

(111.) This is the context of litigation in Washington State seeking to establish such a right. See Machado v. Ashcroft, No. CS-02-0066-FVS (E.D. Wash. Mar. 14, 2002). According to counsel involved in the efforts to obtain class-wide relief, the Machado Court granted the temporary restraining order, ordering that counsel be assigned, before the plaintiff elected to return voluntarily; the court subsequently granted the government's Motion to Dismiss. See E-mail from Matt Adams, Legal Director, Northwest Immigrant Rights Project, to Russell Engler, Professor of Law & Director of Clinical Programs, New England School of Law (June 18, 2007, 14:07:07 EST) (on file with author); see also Finkel, supra note 109.

(112.) Machado, No. CS-02-0066-FVS, at 11 (quoting MICHAEL F. RAHILL, OFFICE OF THE CHIEF IMMIGRATION JUDGE, WHAT CHILD IS THIS? HOW IMMIGRATION COURTS RESPOND TO UNACCOMPANIED MINORS (2000)). AS Professor Taylor declared in the papers supporting the case, "[I]t is inconceivable that a child with limited English proficiency could represent his or herself [sic] in removal proceedings. I believe that legal representation is absolutely essential to safeguard the due process rights of children who face removal from the United States." Id. at 8.

(113.) Kritzer's study of state tax appeals found that represented taxpayers won 36% of their appeals, compared to 20% for unrepresented ones. KRITZER, supra note 90, at 83. Taxpayers represented by nonattorneys actually fared worse than unrepresented taxpayers, winning only 15% of their appeals, a result Kritzer attributes to a lack of procedural expertise among nonlawyer representatives. Id. at 83, 108.

(114.) Popkin, supra note 92, at 1024. Popkin's study of FECA claims found that represented claimants in FECA won 72% of their hearings, compared to 57% of claimants appearing without a representative, Id.

(115.) Id. at 1025. For the veterans programs, Popkin reports that those appearing with a representative won 74% of their claims, while unrepresented claimants won only 55% of the time. Id. When Popkin reclassifies cases where service-connection, but no disability, was established as losses as opposed wins, the figures change to a 65-71% success rate for represented claimants and a 55% success rate for unrepresented ones. Id.

(116.) Ronald Hammer and Joseph Hartley, examining cases in Wisconsin between 1965 and 1974, found only a slight benefit to attorney representation, with represented claimants prevailing in 48% of the cases, compared to 44% of the unrepresented ones. Ronald Hammer & Joseph Hartley, Procedural Due Process and the Welfare Recipient: A Statistical Study of AFDC Fair Hearings in Wisconsin, 1978 WIS. L. REV. 145, 207. Laura Cooper's report on Minnesota cases similarly found that in most situations, those with attorney representation were not more successful than those without attorney representation. Laura Cooper, Goldberg's Forgotten Footnote: Is There a Due Process Right to a Hearing Prior to the Termination of Welfare Benefits When the Only Issue Raised Is a Question of Law?, 64 MINN. L. REV. 1107, 1170 (1980).

(117.) Jan Hagen found that petitioners with counsel won 30% of their appeals, compared to the 13% won by those without counsel. In terms of losses, given the large numbers of continuances and withdrawals during the period studied, petitioners with counsel lost 46% of the time compared to 72% for those without counsel. Jan L. Hagen, Justice for the Welfare Recipient: Another Look at Welfare Fair Hearings, 57 SOC. SERV. REV. 177, 189 (1983).

The Maryland Department of Human Resources estimated a success rate of 70-80% for represented claimants, who susually were represented by Maryland Legal Aid Bureau attorneys or paralegals at Aid to Families with Dependent Children ("AFDC") hearings, compared to a 40-45% success rate for claimants without counsel. Maryland Action Plan, supra note 91, at 12 (involving data for public benefit applications for Fiscal Year 1986 for the primary public entitlement programs affecting low income persons in Maryland). For Medicaid hearings, the Department of Health and Mental Hygiene staff provided specific data indicating that counsel was present at 21% of the hearings, with a 76% reversal rate, compared to a 46% reversal rate when claimants were not represented by counsel. Id.

(118.) ALCAN RODGERS, STATISTICAL APPENDIX TO REPORT TO BOSTON BAR ASSOCIATION TASK FORCE ON EXPANSION OF THE CIVIL RIGHT TO COUNSEL, ON LEGAL REPRESENTATION IN PUBLIC BENEFITS AGENCY HEARINGS 2 (2008) (reporting a 44.6% success rate for represented claimants, compared to a 27.6% success rate overall).

(119.) See generally Karl Monsma & Richard Lempert, The Value of Counsel: 20 Years of Representation before a Public Housing Eviction Board, 26 LAW & SOC'Y REV. 627 (1992).

(120.) Id. at 645-47.

(121.) Id. at 649-51. The success rate was in the range of 9.6% to 21.9%. Id. at 651.

(122.) Id. at 650.

(123.) Id.

(124.) Hammer & Hartley, supra note 116, at 207-08. The authors speculate that the test case strategy being pursued by Milwaukee Legal Services during some of the period studied led to a disproportionately high rate of loss for represented claimants, affecting the overall numbers, Id. at 208. Moreover, given the relatively small number of cases in which attorneys appeared at all, the authors simply concluded that attorneys were not a factor, but that no conclusion could be drawn as to whether a greater effort should be made to provide representation. Id. at 208-09.

(125.) KRITZER, supra note 90, passim.

(126.) HENSCHEN, supra note 16, at 40-44; see generally GOLDSCHMIDT ET AL., supra note 11.

(127.) Allan Rodgers, Mass. Legal Services Program Clinics & Similar Efforts to Help Unrepresented Litigants Going to Court, 1-20 (2001) (unpublished chart, on file with author). Of the twenty-eight Massachusetts programs listed, the column "Measures of Effectiveness" was blank or indicated "no available data" for sixteen programs, with most of the others filled with impressions or conclusory language, Id. Similarly, virtually all of the information on the fifty programs from the New England states yielded little concrete data, with existing evaluations described in terms of "helpful" or "effective." Id.

(128.) Id. A divorce clinic in Neighborhood Legal Services in Lynn reported that 80% of participants succeeded in court, while a similar clinic in Worcester, Massachusetts showed that, for clients who filed a petition, 90% of those helped by the clinic obtained divorces; the office provided representation where complications arose, Id. at 2. In Lowell, Massachusetts, where advocates operated a clinic in District Court, helping walk-in clients seeking 209A orders of protection by helping them fill out the papers and accompanying them into the courtroom, a review of cases showed that nearly all participants received protection orders. Id. at 2,4,9.

(129.) Id. at 12-15; Ross Dolloff & Patricio Rossi, Mediation Project Gets Results for North Shore Tenants, LEGAL SERVICES REP., May 2006, at 1, 12 (finding no appreciable difference for clients who received advice and education from pro se clinics alone); E-mail from Ross Dolloff, Director, Neighborhood Legal Services, Inc., to Russell Engler, Professor of Law & Director of Clinical Programs, New England School of Law (June 11, 2007, 15:45:22 EST) (on file with author) [hereinafter Dollof E-mail]; E-mail from Dan Manning, Director of Litigation, Greater Boston Legal Services, to Russell Engler, Professor of Law & Director of Clinical Programs, New England School of Law (June 14, 2007, 15:53:37 EST) (on file with author) [hereinafter Manning E-mail].

(130.) Steiner, supra note 55, at 21-27. Comparing unrepresented tenants with clinic help to those without, Steiner found improvements both in terms of who retained possession, (31.9% versus 22.7%) or, for tenants losing position, in terms of the length of time prior to eviction (91.5 days versus 80 days), Id. Represented tenants from the clinic, however, retained possession in 36.7% of the cases, compared to 16.4% for unrepresented ones. Id.

(131.) BOSTON BAR ASS'N TASK FORCE ON UNREPRESENTED LITIGANTS, supra note 15, at 17 (citing results from the "Grant Report: Neighborhood Legal Services Eviction Defense Mediation Services," which conducted experimental representation in the Northeast Housing Court).

(132.) See Dolloff & Rossi, supra note 129, at 12-14. Support for this description also comes from my conversations and correspondence with Dan Manning of Greater Boston Legal Services, Faye Rachlin of the Legal Assistance Corporation of Central Massachusetts, Tina Sanchez of Western Mass Legal Services, Ross Dollof, formerly of Neighborhood Legal Services, and Lynn Girton of the Volunteer Lawyers Project. See Dolloff E-mail, supra note 129; E-mail from Lynn Girton, Chief Counsel, Volunteer Lawyers Project, to Russell Engler, Professor of Law & Director of Clinical Programs, New England School of Law (June 15, 2007, 14:09:56 EST) (on file with author); Manning E-mail, supra note 129; E-mail from Faye Rachlin, Managing Attorney, Legal Assistance Corporation of Central Mass., to Russell Engler, Professor of Law & Director of Clinical Programs, New England School of Law (June 14, 2007, 15:28:33 EST) (on file with author); Telephone Interview with Tina Sanchez, Staff Attorney, Western Massachusetts Legal Services (June 22, 2007). Consistent with these findings, an analysis of the Toronto East Representation Pilot Project found that "[w]hen matters proceeded to a hearing ... the results show tenant outcomes definitely improved in almost all aspects...." GENE FILICE, TORONTO EAST REPRESENTATION PILOT PROJECT 18 (2006), available at http://www.cleonet.ca/instance.php?instance_id=3513.

(133.) Ralph C. Cavanaugh & Deborah L. Rhode, The Unauthorized Practice of Law and Pro Se Divorce: An Empirical Analysis, 86 YALE L.J. 104, 130-41 (1976) (critiquing the stated justifications for the unauthorized practice of law restrictions in the pro se divorce area).

(134.) Michael Millemann et al., Rethinking the Full-Service Legal Representational Model: A Maryland Experiment, 31 CLEARINGHOUSE REV. 1178, 1178 (1997) (describing "an experimental project in which law students provide legal information and advice to otherwise unrepresented parties in family law cases").

(135.) Id. at 1186. The level of satisfaction dropped slightly when limited discretion was involved, as in applying the "material change of circumstances" test to support modification orders, and was at its lowest when more complex judgments were involved, such as the "best interests of the child standard" for custody cases, Id.

(136.) KEN SMITH, EVALUATION REVIEW 2 (n.d.) (on file with author) (summarizing the report "Family Law Access Partnership Project ("FLAPP"): Outcomes Evaluation Report for 2001 Grant Year" and reporting an 80% success rate).

(137.) IOLTA INFO. SERVS. & SONOMA COUNTY LEGAL AID, SHAC: THE FIRST SIX MONTHS (2001), available at http://www.calegaladvocates.org/search/download.94544. The program was implemented in mid-2000.

(138.) JUDICIAL COUNCIL OF CAL., ADMIN. OFFICE OF THE COURTS, FAMILY LAW INFORMATION CENTERS: AN EVALUATION OF THREE PILOT PROGRAMS (2003), available at http://www.courtinfo.ca.gov/programs/cfcc/resources/publications/FLICrpt.htm. According to the report, the programs served a large number of litigants (45,000 during the fiscal year studied), customer satisfaction was high (78% to 95% satisfied, depending on the question), and most judges were extremely satisfied with the program as well. Id. at 4, 79, 80-82.

(139.) THE EMPIRICAL RESEARCH GROUP, UCLA LAW SCH., EVALUATION OF THE VAN NUYS LEGAL SELF-HELP CENTER FINAL REPORT 2, 3, 16 (2001) [hereinafter 2001 VAN NUYS REPORT]. The 2001 and 2003 reports by the UCLA School of Law Empirical Research Group are the most comprehensive, early evaluations. Using observations, reviews of case files and center applications, surveys and interviews, researchers sought to examine whether the Center made a difference in case outcomes, whether litigants felt more satisfied with their experiences, whether the court was satisfied with the assistance provided, and what demonstrable methods for improvement existed. Half of the litigants sought help on family law problems, and another quarter for help defending unlawful detainer cases, Id. at 5.

(140.) Id. at 3.

(141.) THE EMPIRICAL RESEARCH GROUP, UCLA LAW SCH., EVALUATION OF THE VAN NUYS LEGAL SELF-HELP CENTER FINAL REPORT 1 (2003) [hereinafter 2003 VAN NUYS REPORT]. The report found that 88% of visitors were "Extremely Satisfied" or "Very Satisfied" with the services they received, that Center visitors are from the low end of the economic scale, and that the Center reduces by a measurable amount the anxiety and confusion that pro per litigants experience in the courthouse, thus reducing barriers that often prevent low-income litigants from accessing justice, Id. at 6. In terms of substantive areas, family law remained the dominant focus (46%), with one-quarter of the visits related to landlord-tenant matters. Id. at 4.

(142.) Id. at 1, 16. For example, a higher percentage of Center-assisted litigants receive child support (50% versus 32%), and with higher average payments (mean of $509 versus $226).

(143.) Id. at 11. The authors explain this conclusion in part by their sense that very few tenants have viable defenses. Elsewhere, the report concludes that the adoptions of two recommendations from the prior report--the development of instructional videotapes and trial preparation clinics--offered substantial benefits, and gave litigants the ability to defend themselves in court against unlawful evictions, Id. at 1, 12-14. Support for this conclusion appears to come from interviews with visitors, combined with the reasoning that lower-income tenants might otherwise not participate in the system absent help. Id. at 10-11.

(144.) Id. at 11; see also E-mail from Gary Blasi, Professor of Law, UCLA School of Law, to Russell Engler, Professor of Law & Director of Clinical Programs, New England School of Law (June 8, 2007, 14:40:18 EST) (on file with author) [hereinafter Blasi E-mail].

(145.) Neighborhood Legal Servs. of L.A. County, DASH Final Evaluation Report 3-4 (Mar. 29, 2007) (unpublished report, on file with author).

(146.) Id. at 23.

(147.) See JESSICA PEARSON & LANAE DAVIS, THE HOTLINE OUTCOMES ASSESSMENT STUDY, FINAL REPORT--PHASE III: FULL-SCALE TELEPHONE SURVEY 1 (2002), available at http://www.nlada.org/DMS/Documents/1037903536.22/finalhlreport.pdf ("Pioneered by legal services programs for the elderly, Hotlines have been adopted for use by a growing number of legal services programs that serve a general low-income population. Historically, more than two-thirds of the cases handled by the Legal Services Corporation (LSC)-funded legal services programs are for advice and counsel, referral, or brief services. The theory behind Hotlines is that these tasks can be performed effectively through a telephone-based system, supported by appropriate computer software and staffed by advocates specially trained in the provision of advice and referral services.").

(148.) See Houseman, supra note 9, at 43. Pearson and Davis, of the Center for Policy Research, completed their comprehensive final report in 2002.

(149.) See PEARSON & DAVIS, supra note 147, at i-iii.

(150.) Id. at ii.

(151.) Id.

(152.) Id.

(153.) Surveys conducted by the American Association of Retired Persons ("AARP") also found that significant numbers of callers followed up on the advice given, with a majority of cases reporting favorable outcomes. See Center for Elder Rights Advocacy, Reports and Statistics, http://legalhotlines.org/reports.php (last visited Nov. 26, 2009). For example, the follow-up survey for consumer cases found that 72% of the callers took the action advised, with 55% reporting a change for the better, and 41% of those reporting "no change" still awaiting results. SHOSHANNA EHRLICH & LANAE DAVIS, LEGAL HOTLINES CLIENT OUTCOME STUDY 2006 5 (2006). With benefits cases, 78% of the callers took the recommended actions, with 65% reporting a change for the better; 52% of the latter group reported an average benefit of $1,150 per month, with a range of $79-$3,500. Id. at 6.

(154.) Ross Dolloff, Let's Talk About Values, Not Systems, MGMT. INFO. EXCHANGE J., Summer 2003, at 41. 50% of the clients were advised to take no action at all, only 55% of those advised to take action took the recommended action, and only 32% of those completely resolved their problem. 32% of 55% of 50% is less than 8% of those who received services who actually took successful action, Id.

(155.) Id. at 38. Dolloff also reminds the reader that hotlines can serve two very different functions: screening and intake, on the one hand, and telephone advice sometimes followed by written materials, Id. at 39.

(156.) GREACEN, WHAT WE KNOW, supra note 12, at 2. Greacen surveyed data from a variety of self-representation reports regarding who were the self-represented litigants, the types of proceedings in which litigants self-represented, and the impact on court processes. The article also explored how clients learned about assistance programs, whether they used the services, how they rated the services, and whether they understood the information provided. While there is some evidence that self-represented litigants who receive assistance are better prepared for court and more confident, there is little evidence on whether self-represented litigants who receive assistance are more likely to obtain a favorable outcome. Id.

(157.) Id. at 29; John M. Greacen, Self-Represented Litigants: Learning from Ten Years of Experience in Family Courts, 44(1) JUDGES' J. 24 (2005).

(158.) BONNIE ROSE HOUGH, EVALUATION OF INNOVATIONS DESIGNED TO INCREASE ACCESS TO JUSTICE FOR SELF-REPRESENTED LITIGANTS 3-5, available at http://www.ajs.org/prose/Midwest%20Notebook%20Contents/Tab%207/Paper%209.pdf. Hough wrote her article for the Summit on the Future of Self-Represented Litigation. The article provides a brief history of evaluation efforts, summarizes the lessons learned, and identifies key research questions. Her key directions for the future identify the need to share results of evaluations and measure what programs are trying to achieve. Id. at 5-6. Hough also encourages assessment of, among other items, the extent of cost savings, the impact of systems changes, the extent to which self-help centers improve the quality of justice, and the continuing identification of factors that make self-help unrealistic. Id. at 8-10.

(159.) Richard Zorza, Evaluation of Pro Se Innovation (1992) (PowerPoint Presentation, on file with author). Zorza's overall questions for evaluation include not only increasing the effectiveness and efficiency of the court system, cost-effectiveness and increasing satisfaction and empowerment of individuals, but also the justice questions mentioned. Id.

(160.) Hartman & Robinson, supra note 36, at 474-76, 480-82.

(161.) Id. at 471-72. According to one account, the number of eviction proceedings filed jumped from 4937 in 1992 to 7120 in 1997, following the elimination of rent control, Id.

(162.) 2003 VAN NUYS REPORT, supra note 141, at 11.

(163.) Michael Millemann et al., Rethinking the Full-Service Legal Representational Model." A Maryland Experiment, 31 CLEARINGHOUSE REV. 1178, 1184, 1189 (1997) (describing "an experimental project in which law students provide legal information and advice to otherwise unrepresented parties in family law cases"). Millemann endorses Jane Murphy's work showing that clear rules in the area of child support reduced judicial discretion, helping pro se litigants enforce their legal rights. He supports Murphy's call for similar reform in the areas of alimony and marital property. Id.

(164.) See supra Part II.C.

(165.) See supra Part II.A.

(166.) See, e.g., Bezdek, supra note 42. Frank Bloch describes his observation of a judge in General Sessions Court, Davidson County, Tennessee, handle a case involving an unrepresented tenant consistent with the "silencing" along the lines Bezdek describes. Interrupting the pro se tenant's attempt at trial to question the bookkeeper as to why her rent increased, the judge snapped "Rent goes up all the time. Any more questions?" The tenant had none, and the landlord obtained judgment. Bloch, supra note 56, at 999.

(167.) See CAPLOVITZ, supra note 79, at 205. While Erik Larson's study of defaults in one housing court suggests a correlation between the weakness of the merits and default, he identifies other variables that also correlate. He cautions against simply concluding that the correlation proves a hypothesis that tenants decide to default based on the strengths of their cases. Erik Larson, Case Characteristics and Defendant Tenant Default in a Housing Court, 3 J. EMPIRICAL L. STUD. 121 (2006).

(168.) See supra note 70 and accompanying text.

(169.) The theme of procedural complexity relates to the variable of the court's operation, and has also been a consistent theme in Justice Earl Johnson's civil Gideon writings. See, e.g., Earl Johnson, Jr., Will Gideon's Trumpet Sound A New Melody? The Globalization of Constitutional Values and its Implications .for a Right to Equal Justice in Civil Cases, 2 SEATTLE U. J. SOC. JUST. 201, 219-20 (2003).

(170.) Millemann quotes Deborah Rhode and David Luban, who encourage a strategy of simplification or modification of legal rules or processes, "including Plain-English statutes, no-fault insurance schemes, computerized title searches, standardized forms for simple wills and uncontested divorces, and automatic wage garnishment for obligations such as child support payments." Millemann et al., supra note 163, at 1189-90 (quoting DEBORAH L. RHODE & DAVID LUBAN, LEOAL ETHICS 737 (2d. ed. 1995)).

(171.) See KRITZER, supra note 90, at 26-32, 127-32.

(172.) See, e.g., Julia Preston, Wide Disparities Found in Judging of Asylum Cases, N.Y. TIMES, May 31, 2007, at Al (summarizing the findings analyzed more fully in Jaya Ramji-Nojales et al., Refugee Roulette: Disparities in Asylum Adjudication, 60 STAN. L. REV. 295 (2007)).

(173.) Eldridge, supra note 51, at 135. The two variables were whether a case was heard by Judge "J" and whether a judge makes a pro-landlord argument throughout the course of the hearing. Id.

(174.) Brandi L. Byrd et al., Investigating the Justice System Response to Domestic Violence in Missouri, 63 J. Mo. B. 222, 225-26 (2007).

(175.) GOLDSCHMIDT ET AL., supra note 11, at 3-4; GRAY, supra note 16, at 1.

(176.) Bloch reports an in-court observation in Nashville, Tennessee, where the tenant was represented by counsel, and the judge, after an extended colloquy, informed the landlord's lawyer that his notice was invalid and they would have to bring a new eviction proceeding, while then informing tenant's counsel that "they're going to get possession anyway, eventually." Bloch, supra note 56, at 999. The Judge's declaration that a tenant will have to move anyway illustrates how a judge's comments not only influence whether a case settles, but on what terms. Id.

(177.) See supra note 60.

(178.) Yngvesson & Hennessey, supra note 78, at 226.

(179.) See supra Part II.C.

(180.) Byrd et al., supra note 174, at 226.

(181.) See generally ZORZA, supra note 16. The increasing attention to the problems related to self-representation has revealed a wide variety of practices in courts' handling of cases with unrepresented litigants. See, e.g., GOLDSCHMIDT ET AL., supra note 11; GRAY, supra note 16.

(182.) For a detailed exploration of the roles of the judges, mediators, and clerks, and the ways in which they are connected, see Engler, supra note 1.

(183.) For example, a solitary "pro-tenant" judge out of step with "landlord-friendly" colleagues will be the target of pressure to conform, as will the lone ranger of a judge adopting procedures that facilitate self-representation but are very different from those of his or her colleagues. See, e.g., GOLDSCHMIDT ET AL., supra note 1 l; GRAY, supra note 16. Consider the different impact of inquiries from top court officials that focus on cases that have lingered on the docket as opposed to inquiries about steps taken to promote Access to Justice and prevent the unknowing waiver of rights.

(184.) Nothing in Sandefur's meta-analysis, finding a negative correlation between the complexity of the substantive law and lawyer advantage, undercuts this point. See Sandefur, supra note l, at 26-27. First, an inquiry as to whether the law favors one side over the other is not necessarily the same as the inquiry into the complexity of the substantive law. Second, whether a lawyer might use the law to his or her advantage also is not tied to the complexity of the substantive law. Third, Sandefur's meta-analysis studies only hearings and trials, id. at 8, muting the impact of lawyers on the default rate and settlements. Finally, despite the negative correlation, Sandefur still concludes that knowledge of substantive law plays a role in lawyer advantage. Id. at 32.

(185.) As Sandefur notes, it is a lawyer's knowledge of both substantive law and "legal procedure" that contributes to the lawyer's advantage. Id.

(186.) The idea that "the poor are less likely than the rich to enjoy the benefits and protections afforded by law" is, of course, not a new concept. See, e.g., Jerome E. Carlin, Jan Howard & Sheldon L. Messinger, Civil Justice and the Poor: Issues for Sociological Research, 1 LAW & SOC'Y REV. 9, 10 (1966). The authors consider: "(1) the character of the law as it concerns the poor; (2) the treatment of the poor by courts and other legal agencies; (3) the legal representation received by the poor; and (4) the capacity of the poor to use the legal system." Id.

(187.) One example is where domestic violence is present in the relationship involving litigants.

(188.) KRITZER, supra note 90, at 133-49. The difference is notable not simply because the types of cases they accept, but the greater credibility afforded by judges to the best advocates and the strengths of these advocates in presenting their cases. Id.

(189.) Id. at 76. Even the descriptions of the perils facing unrepresented debtors involve the descriptions of behavior by the creditors' lawyers intimately familiar with the operation of the court system.

(190.) Millemann et al., supra note 134, at 1181. Millemann's illustration of the critical importance of the initial diagnostic interview underscores the importance of having appropriate expertise, through the supervisor if not the advocate, for undertaking the crucial interviewing role.

(191.) Hagen, supra note 117, at 189-90.

(192.) FARMER & TIEFENTHALER, supra note 71, at 3.

(193.) See Monsma & Lempert, supra note 119, at 663. Hammer & Hartley also offer theories to explain their data that relate to the lawyers, speculating that the lack of benefit from representation in their study related to the case selection strategy adopted by the legal services office. See Hammer & Hartley, supra note 116, at 206 n.356.

(194.) See Sandefur, supra note 1, at 30-32.

(195.) COMM. ON LEGAL ASSISTANCE OF THE ASS'N OF THE BAR OF THE CITY OF N.Y., HOUSING COURT PRO BONO PROJECT OF THE ASSOCIATION OF THE BAR 33 (1988). "Although the Project's representation was outstanding it also was inefficient and reached only 158 families. The participants emerged from the Project with the conviction that pro bono counsel did not offer a large scale, long range solution to the problem of tenant unrepresentation in Housing Court, and that an infusion of full-time funded legal services attorneys was required to meet this pressing need." Id. at 8-9.

(196.) See, e.g., Russell Engler, Towards a Context-Based Civil Gideon Through Access to Justice Initiatives, 40 CLEARINGHOUSE REV. 196, 197 (2006); Houseman, supra note 9, at 62-67; Momentum Grows, supra note 22.

(197.) Engler, supra note 196, at 197. For a discussion of the importance of understanding the civil Gideon initiative as an exercise in effectuating social change rather than framing legal claims, see Russell Engler, Shaping a Context-Based Civil Gideon from the Dynamics of Social Change, 15 TEMP. POE. & CIV. RTS. L. REV. 697 (2006).

(198.) In identifying the scenarios in which referral to an attorney might be most important, Hough has identified the following triggers: the complexity of the legal issues; language barriers; characteristics of the litigant; and the judge. Telephone Interview with Bonnie Rose Hough, Judicial Council of California, Administrative Office of the Courts, San Francisco (June 12, 2006).

(199.) The Massachusetts Housing Court programs illustrate this point. See supra notes 129-132.

(200.) See Cavanaugh & Rhode, supra note 133, at 166.

(201.) Richard Moorhead, Avrom Sherr & Alan Paterson, Contesting Professionalism: Legal Aid and Nonlawyers in England and Wales, 37 LAW & SOC'Y REV. 765, 799 (2003). The authors go further, concluding that in some settings, lay advocates perform better than lawyers. For example, in discussing the peer review results, the authors observe:

   Levels of work below threshold competence (the level at which
   contractees should be performing) were very similar, although
   solicitors had more cases falling below the level of inadequate
   professional services as "poor." This suggests that, if anything,
   solicitors posed slightly more of a risk to the public than
   nonlawyers. More marked, however, was the difference in the number
   of cases handled at the higher levels of quality. Here the
   nonlawyers performed much more strongly.

Id. at 788. In their conclusions, the authors assert: "[t]his enables a confident assertion that taken as a group, nonlawyers perform to higher standards than lawyers." Id. at 795. This finding may or may not transfer accurately across the Atlantic. While not backing away from their conclusions, the authors immediately identify other variables that they were not able to analyze and control for, including the greater likelihood that the nonlawyers were in settings that were non-profit-making, Id. In addition, the nonlawyers often were specialists, while the lawyers were not. Id.

(202.) Not surprisingly, calls for an expanded use of lay advocates typically carry with them the promise of careful regulation, including attorney supervision. See, e.g., AM. BAR ASS'N, COMM'N ON NONLAWYER PRACTICE, NONLAWYER ACTIVITY 1N LAW-RELATED SITUATIONS 136-37 (1995) (suggesting criteria for assessing whether and how a specific kind of nonlawyer activity should be regulated). For an excellent overview of the history and current trends in lay advocacy, see id. at 13-72; Deborah L. Rhode, The Delivery of Legal Services by Non-Lawyers, 4 GEO. J. LEGAL ETHICS 209, 216-21 (1990) (discussing the extent to which enforcement of the prohibition against the unauthorized practice of law has diminished). As Gary Bellow and Jeanne Kettleson observed thirty years ago, "[m]uch more effective lay ... representation would be a necessary component of any significant expansion of access...." Gary Bellow & Jeanne Kettleson, From Ethics to Politics: Confronting Scarcity and Fairness in Public Interest Practice, 58 B.U.L. REV. 337, 386 n. 193 (1978).

(203.) For example, John Greacen, Bonnie Hough, and Richard Zorza, leaders in the efforts nationally to expand and improve initiatives involving self-representation, identify other goals including the success in achieving customer satisfaction and increasing their understanding of court processes, achieving the satisfaction of other participants in the court system, easing the pressures on the court system dealing with large numbers of unrepresented litigants, and achieving the need to maintain community support for the integrity of the dispute resolution system. See GREACEN, WHAT WE KNOW, supra note 12; HOUGH, supra note 158; Zorza, supra note 159.

(204.) If so, counsel is most important where customers are least satisfied. If not, customer satisfaction is an important, but separate, goal. Even where the goal is "satisfaction," the same scenario might yield different levels of satisfaction depending on what is being measured. Confused litigants may appreciate help from a self-help center while still questioning the overall fairness of a system with a palpable imbalance. As the concurring justices in the Maryland case seeking a right to counsel in certain cases observed: "Each of us knows, I believe, that ... the poor, unrepresented parent faced with experienced counsel on the other side is at a great, system-built-in, disadvantage." Frase v. Barnhart, 840 A.2d 114, 134-35 (Md. 2003) (Cathell, J., concurring). Quoting Cicero, the concurrence notes further, "by what justice can an association of citizens be held together when there is no equality among the citizens? ... For what is a State except an association or partnership in justice?" Id. at 130.

(205.) For a sampling of literature critiquing the operation of legal services and legal aid programs, see, for example, Gary Bellow & Jeanne Charn, Paths Not Yet Taken: Some Comments on Feldman's Critique of Legal Services Practice, 83 GEO. L.J. 1633 (1995); Marc Feldman, Political Lessons: Legal Services for the Poor, 83 GEO. L.J. 1529 (1995). For insight into considerations in structuring legal services offices, see, for example, Bellow & Charn, supra; Jeanne Charn and Randi Youells, A Question of Quality, EQUAL JUSTICE, Winter 2004, at 33-34, available at http://www.lsc.gov/pdfs/ejm05c-quality.pdf; Feldman, supra. For a flavor of the discussion regarding legal services triage, compare Paul R. Tremblay, Acting "A Very Moral Type of God": Triage Among Poor Clients, 67 FORDHAM L. REV. 2475 (1999), with Justice A. Dunlap, I Don't Want to Play God--A Response to Professor Tremblay, 67 FORDHAM L. REV. 260l (1999).

(206.) See, e.g., Michael Greco, Court Access Should Not Be Rationed." Defined Right to Counsel in Civil Cases Is an Issue Whose Time has Come, 91 A.B.A.J. 6 (2005). The need for a right to counsel in the civil area reflects the recognition that interests in our physical liberty are not always more important than interests in family and shelter. Most parents would prefer to serve thirty days in prison than lose custody of their children or render their family homeless.

(207.) Lassiter v. Dep't of Soc. Servs., 452 U.S. 18, 27 (1981). The Lassiter court relied on the three elements articulated in Matthews v. Eldridge, 42 U.S. 319 (1976): "the private interests at stake, the government's interest, and the risk that the procedures used will lead to erroneous decisions." 452 U.S. at 27. Those elements are to be balanced against the presumption against a right to appointed counsel unless "the indigent, if he is unsuccessful, may lose his personal freedom." Id.

(208.) Of course, enormous definitional problems occur in identifying the standard for "just" and "unjust." See GREACEN, WHAT WE KNOW, supra note 12, at 29-30.

(209.) Seron et al., supra note 30, at 423-24.

(210.) KRITZER, supra note 90, at 33-37; Sandefur, supra note 1, at 15 ("[L]awyers select cases on the basis of, among other things, whether they think their potential client has a chance of actually winning based on their own past experience with similar eases...."). As noted above, despite the potential bias in case selection, Sandefur concludes that "[t]he impact of lawyer representation on civil case outcomes is potentially quite large. Id. at 32. Moreover, as also noted above, since Sandefur's meta-analysis focuses only on eases where actual hearings took place, the role lawyers play in assisting clients in steps leading up to hearings, and in turn achieving more favorable outcomes for them, may be underestimated. Id. at 8.

(211.) See supra note 15 and accompanying text. Given the complexity of some legal claims and defenses, such as with some defenses for tenants in eviction proceedings, a client's sense of the strength or weakness of his or her own case does not necessarily correlate to the actual strength of the merits of the case.

(212.) In analyzing defaults in one housing court in Michigan, Larson identifies other variables to explain why tenants might not appear. Larson, supra note 167. Larson does not read his data as reflecting a decision by tenants not to show up when they are likely to lose. He does find "a clear association between issue characteristics and likelihood of default, with more contestable issues being associated with a substantially lower likelihood of default." Id. at 140. Larson identifies other variables that correlate positively with changes in the default rate. He observes that the interpretation fails to consider the eventual disposition of the case or a set of interests that might be served by an appearance, such as a negotiated agreement, that go beyond merely winning or losing, Id. at 140 n.14. The other variables that correlated positively with changes in the default rate include whether the property involved was in an area characterized by a high concentration of poverty and amount of monthly rent. Id at 140-41.

(213.) Gender roles complicate this analysis, amid some reports that the issuing of restraining orders follows traditional gender roles, with heterosexual men, and those in gay and lesbian relationships, less likely to obtain relief through the courts. See, e.g., ELIZABETH A. SCHNEIDER ET AL., DOMESTIC VIOLENCE AND THE LAW: THEORY AND PRACTICE 71 (2d ed. 2008) ("It has often been claimed that domestic violence by men is significantly underreported .... In a society that stresses male self-reliance, however, it may be even more difficult for a man to call the police to report that his female partner is abusing him." (citations omitted)). "Same-sex domestic violence is believed to occur at a rate equal to heterosexual domestic violence, estimated in 25-33% of relationships.... Yet, despite the prevalence of same-sex domestic violence, the issue has traditionally been plagued by invisibility and lack of societal awareness." Id. at 86. See generally Joanna Bunker Rohrbaugh, Domestic Violence in Same-Gender Relationships, 44 FAM. CT. REV. 287 (2006).

(214.) Greacen, An Administrator's Perspective, supra note 12, at 32. As the title reflects, the purpose of the inquiry is to test the stereotype that self-representation clogs the courts and the findings that self-represented cases take less time is presumptively positive, Id. Greacen also found that self-represented litigants are less likely to require hearings, Id. He offers three very different interpretations, one suggesting that self-represented litigants make rational decisions as to when to ask for a hearing, another that the presence of lawyers complicates and prolongs cases, and a third that self-represented litigants are unable to use the rules of procedure and evidence to present their cases or obtain the remedies they seek. Id.

(215.) The literature exploring the extent to which perceptions of fairness correlate to outcomes is an important piece of the puzzle. See, e.g., Tom R. Tyler, The Role of Perceived Injustice in Defendants' Evaluation of Their Courtroom Experience, 18 LAW & SOC'Y REV. 51, 56-57 (1984) ("The first [question of study] concerns the relationship between perceptions of absolute outcomes, relative outcomes, distributive fairness and procedural fairness.... The second issue to be explored concerns the influence of outcome levels and judgments of distributive and procedural fairness upon outcome satisfaction and attitudes toward legal authorities."). Recent data suggest that, at least in the family law area, the manner in which cases are resolved is related to satisfaction as measured by the durability of the resulting order. A recent Maryland study, for example, found that "when custody arrangements were resolved through judicial intervention, the rate of subsequent litigation are [sic] twice as high as the rate when custody arrangements were resolved by agreement...." THE WOMEN'S LAW CTR. OF MD., INC., supra note 65, at 40.

(216.) 2003 VAN NUYS REPORT, supra note 141, at 11. The review of reports from housing courts assumed that positive outcomes for tenants included not only retaining possession, but obtaining repair orders and rent deductions. This assistance program apparently led tenants to pay more rent, not less.

(217.) In this category of cases, assisted tenants were actually less satisfied than unassisted ones, since unassisted tenants had no illusions as to how the system would operate. See Blasi E-mail, supra note 144 ("I have a general theory, that is consistent with both this finding and the finding that assisted tenants were less happy with their court experience than unassisted tenants. That is that the Pro Per center is a dispenser of norms and the belief that the legal system works as it should. Thus assisted tenants were more often told what their rights were and the law is (e.g., the warranty of habitability) than what was likely to happen in court as an empirical matter. Unassisted tenants had no illusions and an appropriate degree of cynicism. Unassisted tenants thus probably also are more likely to have a more zero sum, adversary attitude (independent of legal norms) while assisted tenants are likely to have more faith in 'the system,' including the system where tenants pay what they owe."). For a more detailed discussion by Blasi of system justification theory, see generally Gary Blasi & John T. Jost, System Justification Theory and Research: Implications for Law, Legal Advocacy, and Social Justice, 94 CAL. L. REV. 1119 (2006).

(218.) Gary Bellow, Steady Work: A Practitioner's Reflections on Political Lawyering, 31 HARV. C.R.-C.L.L. REV. 297, 305 (1996). Reflecting on seven separate strategies of political lawyering, Bellow identifies common threads that include "an oppositionist social vision" and "persistent engagement with adversaries and decisionmakers." Id. at 300-06.

(219.) The simpler the screening process, the greater the chances of assigning a benefit-counsel in this case--to a scenario that might not fit the original goals so clearly. Counsel will either be appointed to cases in which it might not have been necessary, or might not be appointed to other cases where the need was not immediately clear. The need to consider the costs of implementing and administering programs in cost-benefit analyses is not a new idea. See, e.g., Joel F. Handler, "Ending Welfare As We Know It"--Wrong for Welfare, Wrong for Poverty, 2 GEO. J. ON POVERTY L. & POL'Y 3, 29 (1994) (discussing the enormous price tag in implementing workfare programs in the early Clinton Administration); Camilla E. Watson, Machiavelli and the Politics of Welfare, National Health, and OM Age: A Comparative Perspective of the Policies of the United States and Canada, 1993 UTAH L. REV. 1337, 1370-73 (discussing the costs and administrative inconvenience involved in tax with holding laws). In the words of Theresa Funiciello, in her critique of the American social welfare system and support for some form of guaranteed income, "The money's all there-it's just being spent on an army of social welfare professionals whose interests are protected by the elected officials they help put into office and by the press that naively reports on all of it at face value." THERESA FUNICIELLO, TYRANNY OF KINDNESS: DISMANTLING THE WELFARE SYSTEM TO END POVERTY IN AMERICA xix-xx (1993).

(220.) Each of the recent test cases involving claims for appointed counsel illustrates the difficulties unrepresented litigants face as the case unfolds, and how, by the time of trial, many of the procedural and substantive decisions that shape the case will have been made. See, e.g., King v. King, 174 P.3d 659 (Wash. 2007); Frase v. Barnhart, 840 A.2d 114 (Md. 2003).

(221.) Douglas L. Colbert et al., Do Attorneys Really Matter? The Empirical and Legal Case for the Right of Counsel at Bail, 23 CARDOZO L. REV. 1719 (2002).

(222.) One report from New York City in 1993 estimated that, while the cost of providing counsel for every tenant in New York City would total $84 million, the savings to the city in reduced shelter costs would exceed $150 million, saving the city an estimated $67 million. See JOHN ROMAN ET AL., ESTIMATING COSTS AND BENEFITS OF PRO SE LITIGATION 18 n.5 (2003).

(223.) See, e.g., ROMAN ET. AL., supra note 222; Abel, supra note 6, at 8 ("What are the costs and what are the savings of various approaches to increasing access to the courts?").

(224.) One report from New York City calculated the costs of the failure to provide counsel by referencing the costs of sheltering homeless families. COMM. TRAINING AND RESOURCE CTR. & CITY-WIDE TASK FORCE ON HOUSING COURT, INC., supra note 41, at iii. Roman et al. criticize that approach, noting that the study assumes that each tenant evicted would require a full year of emergency shelter; as a result, they assume instead that the costs are about half of what the paper estimated. ROMAN ET AL., supra note 222, at 18. Yet, those revised costs, in turn, underestimate the costs of homelessness by focusing solely on costs to the city. The authors ultimately conclude that the cost-benefit ratio is unclear. Id.

(225.) See Galanter, supra note 60.

(226.) For example, assuming Galanter is correct, changes in small claims court that restrict the ability of business interests to file and easily prevail in large numbers of cases would eventually be offset by changes in collection practices, either in small claims court or in a different forum.

(227.) For example, if a robust warranty of habitability defense were enforced by skilled counsel for the tenants, it remains to be seen whether the long-term result would be a decline in evictions and an increase in repairs, or an evisceration of the defense, by judicial decision or legislative act. At the 1993 Conference on Social Change, held at Seton Hall Law School, Gary Bellow analyzed strategies to effectuate social change. After describing the process of developing strategies, mobilizing clients, and launching the law reform initiatives, Bellow described the predictable backlash that resulted in terms of pressure from courts, and ultimately legislatures, to modify the behavior and strategies. Gary Bellow, Remarks at the Seton Hall University School of Law Conference on Social Change (Feb. 26, 1993) (author's notes, on file with author).

(228.) Of course, there exists a right to appointed counsel on the criminal side, yet poorer defendants still face long odds in the criminal courts given the power of the prosecution and police, and the behavior of many judges. For example, the ABA Standing Committee on Legal Aid and Indigent Defendants ("SCLAID") sponsored hearings in 2003, which marked the 40th Anniversary of the Gideon decision, and thereafter issued its report, aptly titled Gideon's Broken Promise: America's Continuing Quest for Equal Justice. ABA STANDING COMM. ON LEGAL AID AND INDIGENT DEFENDANTS, GIDEON'S BROKEN PROMISE: AMERICA'S CONTINUING QUEST FOR EQUAL JUSTICE (2004), available at http://www.abanet.org/legalservices/sclaid/defender/brokenpromise/. The ABA issued its ten principles for indigent defense in an effort to respond to rampant problems in the delivery system. ABA STANDING COMM. ON LEGAL AID AND INDIGENT DEFENDANTS, TEN PRINCIPLES OF A PUBLIC DEFENSE DELIVERY SYSTEM (2002), available at http://www.abanet.org/legalservices/ downloads/sclaid/indigentdefense/tenprinciplesbooklet.pdf. None of this is to suggest that the expanded right to counsel is not an important goal worth achieving, but rather to serve as a reminder that no single step is likely to be a panacea.

Russell Engler, Professor of Law and Director of Clinical Programs, New England School of Law. I am grateful for the helpful feedback I received from Laura Abel, Paula Galowitz, Bonnie Hough, Herbert Kritzer, Paul Marvy, Tracy Miller, Clare Pastore, Allan Rodgers, Rebecca Sandefur, Jeff Selbin, Carroll Seron, Gerry Singsen, and Richard Zorza. This work was supported by a stipend from the Board of Trustees of New England School of Law.

COPYRIGHT 2010 Fordham Urban Law Journal
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2010 Gale, Cengage Learning. All rights reserved.

 Reader Opinion

Title:

Comment:



 

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:2008 ABA Section of Litigation Access to Justice Symposium
Author:Engler, Russell
Publication:Fordham Urban Law Journal
Date:Feb 1, 2010
Words:28707
Previous Article:The need for a national civil justice survey of incidence and claiming behavior.
Next Article:Lawyering in juvenile court: lessons from a civil Gideon experiment.
Topics:

Terms of use | Copyright © 2014 Farlex, Inc. | Feedback | For webmasters