On March 31, 1983, the Legal Aid Society filed McCain v. Koch, which would turn out to be a sort of analogue to Callahan v. Carey in that it sought for New York City and New York State to provide shelter for homeless families. It was a class action suit brought before the New York State Supreme Court in Manhattan, with Edward Greenfield as judge. The essence of the suit was that even though the city provided shelter for single men (under the Callahan consent decree) and women (under Eldredge v. Koch), the city failed to provide housing for homeless families. Other charges claimed that city decisions on providing shelter were arbitrary and provided no written notice or opportunity for a hearing; that shelter could be terminated without notice or hearing; that the city relocated homeless families in squalid rooms far from their children’s schools; that the city shuttled families among various sites; that the city failed to provide relocation benefits or transportation funds; that families in emergency shelter lost their public assistance benefits; and that families were coerced into accepting unsafe and inadequate shelter. At that time the after-business-hours entry point to the family shelter system was the Emergency Assistance Units (EAUs), of which there were four, one in each borough except Staten Island. (By 1993 there was only one EAU, located in the Bronx.)
The plaintiffs made many legal claims to support their demands. They claimed that under the Emergency Assistance to Needy Families with Children (EAF) program, the equal protection clauses of the U.S. and New York State constitutions, and article 17, section 1 of the state constitution, the city must provide sufficient shelter for all families in need, just as it provided such shelter for men under Callahan. Plaintiffs also argued that the due process provisions of the U.S. and New York State constitutions prohibited decision making by the EAU without notice and hearings. Finally, the plaintiffs claimed that under article 17 of the New York State Constitution, and various portions of the Social Security Act, New York State Social Service Law, and the New York State Code of Rules and Regulations (volume 18, Social Services) the city had to provide emergency housing that met minimal standards. The plaintiffs also asked the court to certify them as a class, meaning that the suit would be on behalf of not just the named plaintiffs but all similarly situated families.
The initial result of the suit was that the court issued an interim order, dated June 20, 1983, which required that “when a family is not denied emergency housing, assistance, and services,” the shelter provided must meet certain minimal standards, which were spelled out by the court and were similar to those of the Callahan consent decree but adapted to the needs of families. Thus cribs had to be provided for infants, rooms had to have window guards, and placements had to be made with children’s educational needs in mind.1 Note that the interim order did not establish a right to shelter but only specified that when shelter was provided it had to meet certain standards. These standards were later incorporated into regulations issued by the New York State Department of Social Services (DSS).2 This was 18 NYCRR 352.3 (g)-(h) (1983).
Susan Demers points out in an interesting article that the regulations in fact expanded considerably on the interim order.3 Thus the interim order required a room that “has accessible to it a sanitary bathroom with hot water.” The DSS regulation required that “[e]ach family must have a private bathroom. At a minimum this shall include a toilet, a sink and a shower or bathtub, all of which shall be properly maintained with hot and cold running water.” Or again, the interim order required that the shelter provided be “sufficiently heated pursuant to City law.” The state regulations required that
[a] heating system shall be permanently installed and operated in accordance with applicable local law. Where local law or code does not govern the provision of heat, the system will provide heat to maintain a temperature of 69°F (20°C) in all occupied parts of the building, including corridors. Where windows do not open, proper ventilation, including but not limited to air conditioning, shall be operational.
The interim order called for “a clean mattress and pillow and . . . clean and sufficient blankets . . . [and] a sufficient number of clean towels.” State regulations ordered that
[a]ll mattresses and bedding material shall be clean. Each bed shall have at least two clean sheets, adequate clean blankets, clean pillows and pillowcases. A complete change of linens shall be made by hotel/motel staff at least once a week and more often where individual circumstances warrant or when a new family occupies the unit. Each unit shall be supplied with towels, soap and toilet tissues. A clean towel shall be provided daily to each resident.
The regulations also dealt with issues not mentioned in the interim order at all. For example, the regulations required the facility to make arrangements for
removal of garbage; maintenance of floor coverings, draperies and furniture; repainting of the facility at least once every five years; maintenance and inspection of the electrical system; maintenance of plumbing and plumbing fixtures; maintenance and inspection of heating, ventilation and air conditioning systems; a regular vermin control program; and provision to insure that entrances, exits, steps and walkways are kept clear of garbage, ice, snow and other hazards.
Further, the interim order required that its standards be met only “so far as is practicable,” while state regulations threatened that “[n]o family shall be referred to a hotel/motel, nor shall any reimbursement be made for costs incurred from such referral unless all of the requirements set forth . . . are met.” Thus the required standards took a quantum leap in moving from the court interim order to the state regulations.
Part of the reason for this quantum leap was that the state was issuing the regulations with an eye to resolving the McCain litigation. DSS consulted with the plaintiffs and asked them what the regulations would have to contain to resolve the case. The plaintiffs wanted a set of regulations that would require hotels to meet the standards of the Henry Street Settlement, a long-established settlement house known for the quality of its accommodations. The final version of 18 NYCRR 352.3 (g)-(h) (1983) gave the plaintiffs some of what they wanted, but not all. And there remained the issue of whether the regulations would be complied with. So the litigation continued.4
Another fateful result of the interim order was that on September 29, 1983, DSS issued Administrative Directive 83 ADM-47, which also incorporated terms of the interim order and required that “[e]mergency housing must either be provided immediately if a homeless person is determined eligible or written notice must be given that no assistance will be provided where a homeless person is determined ineligible.”5
The plaintiffs in the McCain case had input into the writing of 83 ADM-47. There was back-and-forth discussion between the DSS and the plaintiffs and words such as “delay” and “quickly” were tossed about. But the key word “immediately” came from DSS, not the plaintiffs.6 The directive left unclear exactly what eligibility meant. It noted only that “[i]n no case, however, shall the district deny public assistance solely on the basis that a homeless person has no permanent address.”7 Eligibility standards were thus left unclear, and were not clarified with the following double negative: “when the individual is determined not to be ineligible, an emergency placement shall be made and other needs met.” This lack of clarity would have far-reaching implications for policy. And a single word, “immediately,” would especially turn out to have extensive consequences for the city.
In June of 1984 the McCain plaintiffs asked Justice Greenfield for a preliminary injunction that would finalize the interim order and require the city to provide “safe and adequate emergency housing and assistance suitable for families and dependent children.”8 The court continued to skirt the issue of whether the New York State Constitution implied a right to shelter. The court noted the language of article 17 of the constitution, which it held was “mandatory and imposes upon the State an affirmative duty to aid the needy.” Nonetheless, the court acknowledged that “[n]either the Constitution nor Social Service Law . . . provide that emergency shelter shall be given to the needy in explicit terms.” Then the court continued:
However, once the defendants have undertaken to provide emergency shelter . . . the issue becomes whether the shelter provided should meet reasonable minimum standards. . . . The equitable powers of this court may be invoked to compel compliance with minimal standards. If convicted criminals have such rights, the homeless who become interim wards of a governmental entity are entitled to no less.9
Greenfield converted the interim order to a preliminary injunction, but he denied class certification, finding that many issues of fact and law were not common to all potential members of the class and implying that class certification was “inappropriate, and particularly where governmental operations are concerned.” Both sides appealed Greenfield’s decision, which was scheduled to be taken up by the state Appellate Division in the fall of 1986.
While the McCain litigation made its way through the court system, Legal Aid brought another action against the city. This was Lamboy v. Gross, which was heard by Justice Helen Freedman of the New York State Supreme Court in Manhattan. Lamboy was very similar to McCain, as it concerned homeless families that had to sleep overnight in an EAU when space in a family shelter or a motel could not be found for them. Gross was George Gross, Human Resources Administration (HRA) commissioner as of September 18, 1984. Marie Lamboy and fellow petitioner Oscar Serrano and their three children applied for emergency shelter on May 9, 1985. The city was unable immediately to find shelter for the five-member family, who therefore spent three nights at an EAU. On their first night the family was offered housing suitable for a family of four, but they turned it down and did not receive a placement on the next two nights.10 An affidavit filed by the city indicated that Ms. Lamboy had been a challenging client. Before she applied to the EAU, she had been arrested for stealing at one shelter and banned from another for violating curfew. Another hotel had given her five hundred dollars to leave.11 The father of the family was brain damaged as the result of a mugging and the family’s six-year-old child attended special classes for emotionally handicapped children.12 Another plaintiff was Wilma Acevedo, who had an epileptic six-year-old son. According to the Times, “the Acevedos spent a single night at the Bronx emergency office at 437 East 147th Street, where they slept on plastic chairs and Formica tables under fluorescent light. They were not offered beds at a group shelter because of her son’s health problems.”13
At issue was the question of whether the families had been granted shelter “immediately,” as was specified in Administrative Directive 83 ADM-47. In the court’s opinion, the issue at hand did not concern the state or federal constitutions but was “a much more narrowly circumscribed one, dealing simply with the question of whether the city appellants have properly complied with a State-issued administrative directive [83 ADM-47].”14 An expedited fair hearing found that the directive had not been violated because the Lamboy-Serrano family was provided with cots and a crib and some food at the EAU. Petitioners requested a preliminary injunction preventing the respondents from denying emergency shelter for reasons other than those permitted by 83 ADM-47. Petitioners also requested to be certified as a class. Justice Freedman decided not to await a decision in the very similar McCain case and issued a decision on August 26, 1985.
The city argued that it had tried but simply could not find appropriate accommodations for Ms. Lamboy and her family. Judge Freedman wrote that “[t]he city seems to claim that under the circumstances it did the best it could for the Lamboy-Serrano and Acevedo families and that it attempted to comply with 83 ADM-47 but failed.”15 Below is Judge Freedman’s response to this argument in full:
The expedited fair hearing decision in the Lamboy-Serrano matter, presumably approved by the State Commissioner, suggests that the city complied with 83 ADM-47. Although there was no dispute at the fair hearing that the family was eligible for housing and was at no time informed that it was ineligible, the fair hearing officer found that it could not be determined under all of the circumstances of this case that the family was denied emergency housing assistance from May 9 through 11 because the family was furnished cots and a crib and given some food at the EAU. The officer noted that the medical condition of the emotionally handicapped child made referral to a family shelter inappropriate and described other problems relating to keeping the entire family together. Although the court recognizes that this family, because of its size, past history relating to other shelters and composition was difficult to place it does not agree that furnishing cots and a crib in an emergency assistance unit constitutes compliance with 83 ADM-47.16
In other words, no matter how difficult the task at hand, nothing justifies letting homeless families stay in the EAU. A few months later, on October 2, 1985, Freedman specifically held that “[i]n no event shall provision of overnight accommodations in the city respondents’ welfare offices (including Income Maintenance Centers and Emergency Assistance Units) constitute the provision of emergency housing pursuant to the requirements of Administrative Directive 83 ADM-47 or this Order.”17 The court issued the requested preliminary injunction requiring the city to provide shelter, other than a stay in the EAU, immediately.
The city was appalled. Assistant corporation counsel Judith A. Levitt responded to Freedman’s ruling as follows: “We are doing everything we possibly can right now within reason, and I think this goes beyond reason. This is the real world, not fantasy land.” The city and state both appealed Judge Freedman’s decision to the Appellate Division.
But was Judge Freedman correct that Administrative Directive 83 ADM-47 made no provision for the possibility that appropriate shelter could not be found and that letting a family stay in the EAU might therefore be unavoidable? She was correct, though the reason why was more complicated than her decision suggested. On June 19, 1985, the New York State Department of Social Service, which was the agency that promulgated 83 ADM-47, issued a letter to social service providers clarifying the regulation. The agency wrote,
A determination [italics in the original] to place a family in an emergency assistance unit, even for one night, would be in violation of Department policy. However, there will be circumstances which make it necessary for a family to spend the night in an emergency assistance unit. Such circumstances would include emergencies arising from natural disasters (fires, floods, explosions, etc.) and isolated occasions where the local districts’ good faith efforts to locate preferable alternative housing are unsuccessful. Reasonable efforts must be made by the local districts to plan for emergency needs and consistent and meaningful attempts by staff at the emergency units must have been made to locate emergency housing for the local district to meet its obligations under the ADM.18
The letter acknowledges that when “good faith efforts to locate preferable alternative housing are unsuccessful,” families may be allowed to stay in EAUs. Demers, in “The Failures of Litigation as a Tool for the Development of Social Welfare Policy,” cites this passage in her argument that the city policy of regularly placing families in EAUs should have been allowed by the court.19 But the letter specifies that EAU placements are acceptable only during natural disasters or on “isolated occasions.” At the time Judge Freedman’s decision was made, scores of families were on many nights left in the EAUs. Judge Freedman, and the Appellate Division to which her decision in this case was appealed,20 thus rightly concluded that regular use of the EAUs was not permitted under 83 ADM-47. The Appellate Division would quote the letter of June 19, 1985, and point out that “[i]t is, of course, well settled that the construction given statutes and regulations by the agency responsible for their administration, if not irrational or unreasonable, should be upheld.”21 Moreover, much to the satisfaction of Steven Banks, the plaintiffs’ attorney, the decision contained “a wonderful passage . . . which says the word ‘immediate’ is a word of consummate clarity, it means right away, without delay.”22
But what about a situation like the one in which the city found itself in which good faith efforts to locate preferable alternative housing are regularly unsuccessful? The regulation simply does not allow for such a state of affairs. Perhaps the regulators foresaw that the city would be sued for its systematic failure and that under court pressure the city would eventually improve its operation and comply. This reasoning perhaps was correct: After about twenty years of court orders and millions of dollars in fines, the city would eventually develop a system that not only did not place families in the EAUs, but did away with EAUs entirely. Whether this process was the most efficient way to reform the family-intake system is doubtful, but eventually it worked.
As the McCain proceedings rolled along, developments concerning the Callahan case continued to unfold. Robert Trobe and Bonnie Stone, who were responsible for finding shelter space, were convinced that the plumbing ratios specified in the consent decree made locating compliant buildings very hard and required adding unnecessary fixtures when otherwise suitable space was found. Stone eventually decided to try to document empirically that the ratios forced the city to build into the shelters shower and toilet capacity that was never used. Stephen Levine, executive assistant of the Bureau of Management Systems of HRA, who had a Ph.D. in history, was assigned the task of designing and implementing a study of shower and toilet use in the men’s shelters. The result was “An Observational Study of Toilet and Shower Utilization at Three Men’s Shelters,” which was completed in early February 1985. Including an addendum and a follow-up, the study in fact looked at four large men’s shelters, the Fort Washington Armory, the Franklin Armory, the Bedford-Atlantic Shelter, and the Schwarz Building at the Charles M. Gay Shelter on Ward’s Island. Levine’s methodology was to station observers at shelter toilets and showers during what he supposed to be peak usage periods in the morning and at dinner time. In total, fourteen peak-time periods of between seventy-five minutes and three hours were observed. At fifteen-minute intervals the observers counted the number of showers, water closets, and urinals that were being used.
The results seemed to confirm the city’s position. Maximum peak-time use of water closets at a given point ranged from 70 percent to 15 percent. Maximum peak-time use of showers at a given point ranged from 72.2 percent to 7.4 percent. For urinals, maximum peak-time use at a given point ranged from 100 percent to 12 percent. (The study notes that during the one moment of 100 percent use of the urinals, there were fifteen empty water closets.) The study also reported that at no point were clients ever observed waiting to use a water closet, shower, or urinal. At all shelters the capacity of toilet and shower facilities “greatly exceeds demand, even during the peak hours of utilization.”23 The city thus moved in late 1986 for a modification of the plumbing ratios. The New York State Department of Social Services, whose regulations were consistent with those of the Callahan decree as they had been modified, told the court that it would grant the city a waiver of its regulations, but only if the court further modified the ratios in the decree.24 The city proposed to the court a scheme in which the allowed plumbing ratios would rise as the population of a given shelter rose. The Callahan decree ratios would apply to shelters of up to two hundred clients, but, for example, for shelters of five hundred clients the ratio of water closets to clients would be one to eighteen and the ratio of showers to clients would be one to twenty-six. Such a sliding scale would, according to the city, acknowledge the “frequently demonstrated principle” that “the need for additional resources . . . does not always increase in identical proportions to an increase in the number of users.”25
The plaintiffs felt that the debate over plumbing ratios had been definitively resolved back in 1982. They told the court,
The only apparent difference between the City defendants’ prior motions and this one is their proffer of a “study” which purportedly establishes “actual experience” with the plumbing fixtures in the shelters. However this study or survey is seriously flawed and comes nowhere near to a demonstration of “actual experience.” Defendants’ survey must be rejected out of hand as inherently unreliable . . . [I]t should be disregarded in its entirety.26
As the use of scare quotes around the word “study” suggests, the plaintiffs hardly hid their contempt for Levine’s work and made a wide range of arguments against the study. They pointed out that the study itself acknowledged that one morning observation had been done while hot water was scarce; Levine himself agreed on deposition that those results ought to be discarded. They asked why observations were made at fifteen-minute intervals when continuous observation of the facilities was possible. They objected strongly to the fact that shelter clients had not been interviewed about their experience with shelter facilities. They discovered upon deposing the study’s authors and implementers that one observation had been made while a movie was being shown at the shelter and another observation was conducted while a World Series game was on television. They argued that the study’s results could not be “projected,” that is, that it was impossible to generalize from the relatively small set of observations made to the general usage of the facilities. They questioned the qualifications of Levine, pointing out that his Ph.D. was in history and that most of his previous work for the city involved interviews or questionnaires. And they implied that the study had been done on the cheap as they got Levine to estimate that the whole project cost the city only $972. The plaintiffs concluded, “In sum, the city defendants, despite their ‘study’ and ‘formula,’ have not made the requisite showing to justify modification of the decree.”27 In his decision of October 1, 1987, Judge Sklar (who had replaced Judge Wallach) largely agreed with the plaintiffs. “[M]uch of the methodology used in conducting the surveys appears to be haphazard,” he wrote. He specifically rejected “the city’s premise that as population increases proportionally, fewer plumbing fixtures are needed,” commenting that “[t]his assumption is not borne out by the studies.” Sklar concluded that the study was “patently inadequate.”28
The plumbing-usage study certainly had its faults, many of which the plaintiffs accurately pointed out. But the fact is that the formula the city cited and that the plaintiffs and court rejected—the formula that “the need for additional resources . . . does not always increase in identical proportions to an increase in the number of users”—is a truth cited in basic textbooks on queuing theory, which is the mathematical science of waiting in lines. For example, the queuing theory expert Robert B. Cooper acknowledges in an encyclopedia article “the important fact that large systems are more efficient than small ones.”29 At my request the queuing theory expert Prof. Kardi Technomo has calculated the optimum number of plumbing fixtures for shelters of various sizes. Some of the results are included in table 2.1:30
Sources: City-proposed ratios: Callahan v. Carey, Index No. 42582/79 Sup. Ct. NY County decision of Judge Stanley Sklar, October 1, 1987, p. 6; optimum ratios: personal correspondence with Kardi Technomo, 9/25/13; Callahan decree ratio: Callahan v. Carey, Index No. 42582/79 Sup. Ct. NY County, decision of Judge Richard Wallach, November 4, 1982, p. 3.
As can be seen, applying queuing theory to this problem always calls for fewer showers than the modified Callahan ratios would require. Optimizing showers implies that the number of clients for each shower head goes up, just as the city had proposed, although in some cases optimum ratios are higher than the city-proposed ratios and would require installing more showers than the city proposal called for.
One wishes that there were more systematic documentation of actual conditions at the shelters’ plumbing facilities. In this context the plaintiffs made a fair point when they objected to the failure of the city’s study to include interviews with clients. But the plaintiffs’ arguments make an assumption that needs to be questioned. That assumption is that plumbing ratios are a good proxy for overall conditions in the shelters. As we saw, Robert Hayes testified that when he was rebuffed in his efforts to include a limit of two hundred on the population of a shelter, he fell back on the low plumbing ratios to limit the size and inhumanity of the shelters. But plumbing ratios were not what drove the city’s decisions on shelter size. The need to accommodate large numbers of clients quickly, and the political difficulties involved in acquiring buildings that were not already owned by the city or state, were the factors that forced the city to rely on armories and other large structures. The city hardly rejected the use of large shelters because they entailed installing too much plumbing. Instead, the city went with the large shelter strategy because there were no ready alternatives and built in excess plumbing. The result was the worst of all possible worlds: big shelters (which the plaintiffs didn’t want) with too much plumbing (which the city didn’t want). Both sides would have been better off if the questions of plumbing ratios, shelter size, and quality of life had been disentangled and each dealt with in its own right. The debate over toilet ratios was perhaps a minor matter. It is worth focusing on, however, because through the use of queuing theory it is—unlike most controversies over shelter management—susceptible to something like an “objective” answer. In this small instance, the court seems to have allowed itself to descend unnecessarily into the minutiae of shelter operations and to have come up with a bad answer.
Meanwhile, the appeals of Greenfield’s preliminary injunction in McCain were being considered, and on May 13, 1986, the Appellate Division of the state supreme court, in a thirty-two-page decision written by Justice Ernest H. Rosenberger, unanimously granted a preliminary injunction “barring the denial of emergency shelter to homeless families.” In doing this the Appellate Division did what Justice Greenfield of the supreme court would not do, that is, “reach plaintiffs’ constitutional claims”31 and conclude that homeless families had a right to shelter under article 17 of the state constitution. Or, more exactly, the court found that “[i]t is also likely that plaintiffs will succeed on their claim that the NY Constitution article XVII obligates defendants to provide emergency shelter for homeless families.”32
However, at the same time, the Appellate Division reversed the lower court’s decision that shelter, once provided, must meet minimal standards. After reviewing what the court of appeals—New York State’s highest court—had previously said about this matter, the Appellate Division held that
the adequacy of the level of welfare benefits is a matter committed to the discretion of the Legislature. . . . In light of the broad discretion vested in the Legislature, we cannot conclude that plaintiffs are likely to prove that article XVII substantively guarantees minimal physical standards of cleanliness, warmth, space and rudimentary convenience in emergency shelter.33
The Appellate Division came to this conclusion “reluctantly” and hoped that when its decision was appealed, it would be “time for the Court of Appeals to reexamine and, hopefully, change its prior holdings in this area.”34 The Appellate Division thus vacated the preliminary injunction issued by Justice Greenfield in June of 1984 that required the city, when it had decided to provide shelter, to provide shelter that met minimal standards.
This finding by the Appellate Division received considerable attention. The New York Times ran an article on the decision entitled “Ruling Widens Shelter Rights for Homeless,” which quoted a Legal Aid Society lawyer as saying, “This case is for poor families what Brown v. Board of Education was for discrimination.”35 This response was probably due to the Appellate Division’s finding that it was likely that plaintiffs would establish a right to shelter under article 17 of the New York State Constitution. But what the Appellate Division gave with one hand it, for practical purposes took away with another when it ruled that courts could not set minimal shelter-condition requirements. The city was in any case committed to providing emergency housing to families. The most pressing policy issue was how to improve the quality of that shelter. The HRA commissioner was closer to the mark when he noted that “[w]e understand that the Appellate Division in its decision today found that homeless families have a right to this [city-provided] shelter. At the same time, the court also found that the nature of the shelter provided is up to the city.”36
On July 3, 1986, plaintiffs obtained an order from Justice Freedman to require the city to comply with the preliminary injunction issued by the Appellate Division on May 13, 1986. The order required that the city “[p]rovide lawful emergency housing to all eligible homeless families with children, such emergency housing not to include overnight accommodations at Emergency Assistance Units or Income Maintenance Centers.”37
However, this July 3, 1986, order added an element that was not present in the Appellate Division’s preliminary injunction of May 13, 1986. That preliminary injunction said nothing that would prevent the city from housing families overnight in an EAU or IMC. The provision of the order that prevented the city from placing families overnight in the EAUs would later in the litigation turn out to be crucial when, in the nineties, 83 ADM-47 was rescinded, and the provision for “immediate” placement was then dropped and the question came before the court how long the city could leave families waiting in the EAU.
Plaintiffs appealed the vacating of the preliminary injunction about the minimum standards to the New York State Court of Appeals. In other words, McCain’s lawyers asked the highest court in New York State to hold that courts do, in fact, have the power to compel shelter providers to meet minimal standards once they decide to provide shelter. On April 22, 1987, the court of appeals held that courts do indeed have that power. In reaching this decision, the court of appeals expressly said that it was not necessary to decide the question of whether there was a right to shelter. According to the court of appeals,
the sole issue is whether the court has the power to issue a preliminary injunction requiring . . . [the state and city] . . . when they have undertaken to provide emergency housing for homeless families with children to provide housing which satisfies minimum standards of sanitation, safety and decency. We hold that the Supreme Court has such power.38
The court of appeals further said, “Thus to decide the narrow issue here, it is not necessary to resolve questions pertaining to the underlying obligation to furnish ‘emergency shelter to eligible homeless families with children.’ . . . We do not reach them.”39
According to the decision, the New York State Supreme Court had correctly reasoned that shelter that did not meet minimal standards was no shelter at all. Since no regulations covering the quality of shelter then existed, the court had to fashion its own standards, which it developed in the interim order. When, after the state DSS commissioner issued minimal shelter standards, these were more detailed than the standards in the interim order and therefore did not conflict with them even when the interim order was finalized as a preliminary injunction. There remained, therefore, the questions of whether the city and state were meeting the standards set forth in the regulations and whether the preliminary injunction was rendered moot by the issuance of the regulations.
The court of appeals thus sent the case back to the lower Appellate Division to decide those issues. The Appellate Division found that the preliminary injunction was not moot, because the state social service regulations were binding on only Department of Social Service employees and not, therefore, on the city Department of Housing Preservation and Development, which was also involved in housing homeless families. Questions as to whether the city was meeting the requirements of the state regulations and the preliminary injunction were sent back to the lower supreme court. Thus the McCain case was consolidated with Lamboy (another case, Slade v. Koch, which involved accommodations of pregnant clients, was also consolidated with Lamboy) and ended up back before the Supreme Court of the State of New York with Helen Freedman as judge.
Where, then, did these early steps of McCain and Lamboy leave the question of whether there was a right to shelter in New York State? The Appellate Division did issue a preliminary injunction based on article 17 of the New York State Constitution that barred the city from denying shelter to homeless families. But this finding was not considered by the court of appeals, leaving the question of a constitutional right to shelter unsettled. Nonetheless, we have in these preliminary actions an example of how a right can be created without any one person or organization explicitly creating one. We also have an example of how the American policy process, fragmented as it is and thus often held to be resistant to dramatic change because of its multiple “veto points,” can, precisely because of its fragmentation, generate rapid change.40 One political scientist described this in another context: “what is perhaps most intriguing . . . is how policy became more ambitious as it traveled from one institution to another. . . . Each institution made incremental changes that seemed small when viewed individually but that constituted major, rapid change when put together.”41
Neither the framers of the New York State Constitution nor the courts in Callahan, Lamboy, or (finally) McCain nor the state regulators explicitly endorsed a right to shelter. But a combination of mutually reinforcing incremental changes produced a nonincremental change: an operational right to shelter for both singles and families, with detailed quality specifications, that would empower the advocates of the homeless to drive policy for years to come.
The obvious solution to homelessness was more permanent housing, but where would the permanent housing come from? In his useful biography of Ed Koch, Jonathan Soffer sets the background for a path-breaking housing program that would become Koch’s “most enduring achievement”:
New York was facing an affordable housing crisis. Between 1978 and 1981 the number of apartments renting for $200 month declined by 43 percent, with 67 percent of renters spending more than 35 percent of their income on rent. And in Koch’s first term eighty-one thousand units of housing stock, mostly at the lower end of the scale, disappeared.42
In his 1985 State of the City address, and then some weeks later in a more developed form, Koch announced his intention to initiate what would be called the “Ten-Year Housing Plan,” which would by 1997 produce more than 150,000 units of housing for low- to middle-income families at a price of about $5 billion.43 About 10 percent of these units would eventually be occupied by formerly homeless families.44
It was a daring plan. The money was city money raised with general obligation bonds. In years past, most of the money for developing urban low-cost housing came from the federal government. But under the Reagan administration, such development funds were mostly eliminated, and by 1986, as Koch said, “Federal support of housing is now just a memory.”45 Thus annual city spending on housing rose from $25 million in 1985 to more than $850 million in 1989 before tapering off to about $300 million in 1995.46 No other city in the country would come close to matching this effort. For years after the 1975 fiscal crisis, raising such funds was impossible because bond markets were closed to the city. However, by 1986 investors had confidence in the city once again, making the Ten-Year Plan possible. The shift from federal to local dollars brought with it a major organizational challenge. The plan was to be directed by the city’s Department of Housing Preservation and Development (HPD). Felice Michetti, Koch’s deputy commissioner of HPD and commissioner of the department under Dinkins, noted that
[e]ssentially HPD went from [being] at the beginning of the Koch Ten-Year Plan . . . a housing agency that historically relied on federal money and federal programs to carry out its mission . . . and really didn’t do direct developments . . . to a city agency that was charged with spending $5.1 billion and spending it efficiently and free of corruption.47
Also making the plan possible was the fact that the city already owned thousands of potentially viable housing units. These were the in rem stock, buildings that the city acquired from tax-delinquent owners. In 1986, the city owned more than fifty-three thousand units of occupied in rem housing, and forty-nine thousands units in vacant in rem housing.
Koch described the development of the plan as follows:
I think the biggest challenge [for a chief executive] is making an appropriate decision as to how to spend your money. . . . One point, I think, had a great impact on the city. I called in my deputy mayor on Operations, who was Nat Leventhal, and I said, we have to create a housing program [because] the federal people are not providing housing at all. All they do now is repair existing federal low-income housing. The state isn’t providing any money.
I want to have a program to create housing for low-income, middle-income, and for homeless people. A program that will provide housing and use just abandoned housing, which we have lots of. At one point we owned 8 percent of all the housing in the City of New York, which was crazy. Crazy, unless you’re a communist. And there were people who said we want the city to own all the housing. I remember that very, very well. And I said, over my dead body.
So we created 150,000 either new units or units that were new based on gutting an old building [and] rebuilding it. It cost approximately a hundred thousand dollars to create such a unit, a two-bedroom unit.
I’m very proud of that program. And in that program we had units for the homeless and we also had units for people who had AIDS and we had [units for the] elderly. I mean we tried in a responsible way to address the needs of lots of people without being able to address the needs of all the people.48
The Ten-Year Housing Plan would be continued under the Dinkins administration, which fatefully used the apartment units it generated to rehouse the homeless, with various unintended consequences. Robert Hayes later suggested that the plan was ultimately the result of the pressure that legal advocacy had placed on the city: “The most significant thing about Callahan was that it convinced the Koch administration to start renovating in rem housing in earnest. . . . It was a radical shift—here were all these buildings in tax arrears that the city had been trying to get rid of, and now there was a use for them.”49 But this is to exaggerate the impact of homelessness on the Ten-Year Plan and of the Ten-Year Plan on homelessness. Much of the appeal of the plan to Koch was, no doubt, that it would benefit not just the homeless but also the key lower-middle-class, white constituency he was courting. Thus, while it is true that 10 percent of the families housed by the program were formerly homeless, about a third were either moderate- or middle-income families.50
While its shelter system had been expanding, the city also developed other services for the homeless, including street outreach programs. In October 1982 Project HELP (Homeless Emergency Liaison Project) began as a pilot program under the auspices of NYC Health and Hospitals Corporation. The project was a mobile psychiatric team, staffed with social workers, nurses, and psychiatrists, that focused on trying to convince homeless people with mental disabilities voluntarily to accept help, from offers of coffee and food to mental health services, including transportation to a shelter or emergency room. Project workers brought only twelve to fifteen homeless people to a hospital each year for the first few years of the program’s operation.51
In 1985, in an effort to bring in more of the city’s street-dwelling homeless, Sara Kellerman, commissioner of mental health in New York City, granted psychiatrists on Project HELP teams the power to designate individuals as committable. Psychiatrists thus had the power to authorize involuntary transportation of mentally impaired homeless people to emergency rooms for evaluation and admission. This designation authority was given to project psychiatrists under section 9.37 of the New York State Mental Hygiene Law and made the project into something like a mobile emergency room. However, the city continued to interpret the law as allowing involuntary transportation only in cases where danger to the client or others was “imminent.”
Here’s an example of Project HELP’s outreach efforts:
R.V., an elderly white man of Eastern European background, has been living in front of the United Nations for several years. Most of his time is spent shouting at the United Nations building in a language that team members were initially unable to identify. For more than a year, as various Project H.E.L.P. team members approached him, R.V. would become increasingly agitated. A team member would stand at some distance and attempt to make contact with the patient. After more than a year of such efforts, R.V. became less agitated, and team members were finally able to approach him more closely. The language in which he was shouting was determined, and a Project H.E.L.P. psychiatrist began to converse with him. R.V. has never accepted any project services, but he is assessed on a regular basis. If any significant deterioration in his condition is noted, he may be involuntarily transported to a hospital.52
Project HELP kept track of 1,309 street homeless people in its first thirty months, only about 3 percent of whom were involuntarily transported to an emergency room. Jane Putnam, the director of Project HELP, believed that the strict interpretation of the Mental Hygiene Law was preventing her team from assisting many people who needed care. In a coauthored article, she wrote,
The legal procedures relating to both inpatient and outpatient care must be examined. Some patients will need “asylum” for the rest of their lives. Others may need involuntary treatment in diverse secure settings. Application of the “dangerousness” or “imminent risk” criteria is denying necessary and appropriate treatment to vast numbers of gravely disabled patients not competent to judge their needs for care and treatment.53
In short, according to Putnam, “We have been seeing people die with their rights on.”54
Mayor Koch agreed with Putnam’s evaluation and from 1986 to 1987 supported bills introduced in the New York State legislature that would modify the “harm to self or others” criteria and allow involuntary transportation in cases of self-neglect and grave disability. However, the bills failed to pass.55
As Koch considered how to organize the city’s street outreach efforts, in late January of 1986, Ellen Baxter’s efforts, as director of the Committee for the Heights Inwood Homeless (CHIH), to develop decent, permanent shelter for single, homeless adults was partly realized with the opening of a residence known as “The Heights” at 530 West 179th Street in the mostly Hispanic neighborhood of Washington Heights, Manhattan. The Heights was a path-breaking and highly influential project in several ways. It was the first SRO opened specifically for homeless people in New York City, and the first to develop new SRO housing of any type in the city in many years. It was the first housing project for the homeless in the nation to be syndicated, meaning that it developed a temporary network of several funding sources to make loans for its single purpose, and the first to receive a commercial bank loan. Even a set of partners at the later infamous Goldman and Sachs would eventually invest. Seven different funding sources would eventually be put together. The Heights’s commitment to maintaining a mixture of different types of residents—the mentally ill, low-income people, substance abusers, people living with AIDS, as well as former shelter clients and people who had been living on the street—was also an innovation. The residents of the Heights were also protected by signing a lease, something then unknown in other forms of housing for the homeless.
Also new, and critical to the project’s success, various types of services, including case management and help in gaining and maintaining entitlements residents qualified for, were provided by a separate organization, Columbia University Community Services (CUCS).56 One of the most important features of the Heights, according to Baxter, was that residents staffed most of the residence’s nonprofessional positions, including front desk management and security:
The tenant self-management approach I think is very helpful to ensuring the quality of the housing that’s delivered. I think that sometimes not-for-profit housing sponsors forget their not-for-profit mission. Being the landlord and your interest in paying your bills and serving people who have very limited income, sometimes we have to grapple with the contradictions of that. And I think that having tenants who live in the building as part of the management helps to preserve the mission. [Residents] have more of an investment in their own housing if they have a stake in it that’s meaningful.57
Finally, in anticipation of what would be called the “New York/New York Agreements,” Baxter succeeded in getting both the city and the state, who at the time were each insisting that the other move first on helping the mentally ill homeless, to cooperate on developing the fifty-five units of permanent housing at the Heights. In effect, Baxter had developed one of the first supportive housing facilities, that is, housing for a range of low-income and disabled clients who would probably otherwise be homeless that provided on-site services that made living independently in the community possible. It was a crucial innovation. Ted Houghton, in his essential account of the supportive housing movement, writes,
The Heights . . . paved the way for all the supportive housing developers that followed, by showing how unrelated government and private funding streams could be cobbled together to build and operate housing with services. . . . CUCS staff . . . made a subtle but significant shift from clinical and treatment-oriented services to a focus on providing the supports necessary to maintain the tenants’ long-term stability in housing.58
Eventually it would dawn on policymakers that the only way to get homeless people off the streets and out of shelters was to have some form of permanent housing where their various special needs would be met. For street dwellers and repeat shelter users, who would come to be called the “chronically homeless,” supportive housing would provide that permanent base. Cynthia Stuart, chief operating officer of the Supportive Housing Network of New York—a set of nonprofit organizations that by 2015 provided about thirty thousand supportive housing units in the city—put the matter this way: “Callahan . . . propelled the creation of so much supportive housing because the City of New York had to provide shelter to all these people and . . . [they] weren’t going to get out if they didn’t have a decent apartment. So that is what supportive housing was designed around.”59
In May 1987, Mayor Koch went out with the Project HELP team on its rounds of Manhattan and saw a number of the program’s subjects. One of them was a woman who was known to Project HELP as both Ann Smith and Billie Boggs and who was later identified as one Joyce Brown. Koch described these events:
I’d been out on the street a number of times with our HELP unit that kept in touch with people who were really out of it mentally and wouldn’t go to shelters. But [HELP] kept track of them, helped to feed them, bringing them sandwiches. . . . Under the law [you] could commit them if they were in danger of injuring themselves or others, immediate danger. That was the law. But most of them were not in immediate danger [yet] they were doing things that are terrible. One for example was Billie Boggs.
HELP . . . showed me Billie Boggs. She was on Second Avenue living, for long periods of time, on a grate that had some heat coming out. And she was sitting in her own feces and she would be yelling at passersby and when they gave her money she would burn it. It was a very sorry sight.
I remember the [mental health] commissioner saying to me, people have a right to do what they want; even if it’s terrible and you don’t like it, they want to live on the streets. And I thought to myself, that’s crazy. I think it’s crazy. I said to corporation counsel, this can’t go on. So the corporate counsel thought about it and . . . weeks, maybe months passed and then the corporation counsel came in to see me and . . . they said, we have found a way. . . . We think we can establish the premises that if they will be in danger to themselves or to others in the foreseeable—that’s the keyword—future you could pick them up. You don’t have to wait until they are going to die.60
This new interpretation of the Mental Hygiene Law was communicated to Project HELP and HHC psychiatrists in a memo of September 9, 1987, from HHC vice president for legal affairs John E. Linville and vice president for mental hygiene Luis Marcos. The memo urged clinicians that “the law recognizes a concept of ‘serious harm’ that is significantly broader than actively suicidal conduct. Significant, passive self-neglect meets the ‘serious harm’ standard as well.” The memo further urged doctors to commit those patients they believed were “in danger of serious harm within the foreseeable future.”61 Koch planned to transport up to five hundred mentally ill street dwellers in a year and dedicated twenty-eight beds in Bellevue Hospital to accommodate them.62
On October 28, 1987, relying on this new, broader interpretation of the Mental Hygiene Law, Project HELP involuntary transported six individuals to the psychiatric emergency room at Bellevue Hospital. One of those was Joyce Brown. Project documents described her as follows:
Female in her 40’s, known to Project HELP. She has been living against the wall of a restaurant on the Upper East Side for many months. She is dirty, disheveled and malodorous. She urinates and defecates in her location. She is delusional, withdrawn and unpredictable. Without provocation, she can get very agitated and explosive. She has a history of having assaulted passersby. She accepts coins from passersby but will tear up or burn bills. She has paranoid ideation, ideas of reference and inappropriate affect. Her judgment is grossly impaired. She has no awareness of illness or insight. Her physical condition is poor. The preliminary diagnosis is acute psychosis, possibly associated with paranoid schizophrenia. Because of her mental condition including lack of judgment, she requires hospitalization. She is also a danger to self.63
At Bellevue Joyce Brown was examined by a psychiatrist who diagnosed her as schizophrenic, injected her with the antipsychotic medication Haldol and a fast-acting tranquilizer, Ativan, and committed her. About a day later another psychiatrist examined Brown and confirmed her emergency admission. However, Brown chose to contest her commitment. Soon after her admission to Bellevue, she contacted the New York Civil Liberties Union and asked them to represent her in court.
On November 2, 1987, the court heard Brown’s case. According to Luis R. Marcos,
At the hearing, four psychiatrists testified consistently that she suffered from chronic schizophrenia, that she was delusional, that her thinking and judgment were seriously impaired, and that she was in need of hospitalization because of mental illness and dangerousness to herself. . . . Three psychiatrists retained by the NYCLU disputed the testimony of the hospital psychiatrists point by point, maintaining that Joyce was neither seriously mentally ill nor a danger to herself or others; they described her life-style as simply reflecting an eccentric personality, not psychopathology.64
For every apparently disturbed behavior of Brown’s cited by the city psychiatrists, the defense psychiatrists had a response. City psychiatrists pointed to evidence gathered by Project HELP workers that Brown threw away or tore up and urinated on money that was given to her by passersby. The NYCLU’s psychiatrists saw nothing problematic in this. Brown had the small amount she thought she needed, and she told passersby that she didn’t want money and that she was afraid of having money on her at nightfall. Moreover, money was sometimes thrown at her by policemen who would call her a whore. So rejecting and urinating on money was appropriate. City psychiatrists thought it was delusional that Brown used aliases. NYCLU psychiatrists pointed out that Brown was trying to escape detection by her sisters, who lived in New Jersey and were looking for her and had once had her involuntarily committed. Brown defecated on the street, city psychiatrists noted. This was due to the fact, said the NYCLU psychiatrists, that local stores would not give her access to their bathrooms. Brown’s clothes were dirty, tattered, malodorous, and too thin for the cold weather on the streets. NYCLU psychiatrists argued that Brown’s clothes were weather appropriate because she lived on top of a steam grate and therefore heavy clothing would have been too warm. Brown reacted to a Project HELP worker by chanting meaningless rhymes that referred to her genitals, city psychiatrists noted. This was done on purpose, Brown and her psychiatrists claimed, to ignore the worker.
Justice Lippmann issued his decision on November 12, 1987. The judge found that “the psychiatric experts . . . are nearly diametrically opposed in their assessment of her [Brown’s] mental condition and in their prediction as to whether she is likely to cause herself or others harm.”65 Lippmann therefore relied primarily on his impressions of Joyce Brown as she was on the stand and concluded that the city did not have the right to commit Joyce Brown. He argued,
I am aware that her mode of existence does not conform to conventional standards, that it is an offence to aesthetic sense. [Nevertheless,] she copes, she is fit, she survives. . . . She refuses to be housed in a shelter. That may reveal more about conditions in shelters than about Joyce Brown’s mental state. It might, in fact, prove she is quite sane. [Also,] there must be some civilized alternatives other than involuntary hospitalization or the streets.66
HHC appealed to a five-judge panel of the Appellate Division, which reversed Lippmann’s decision. A three-person majority found that Joyce Brown’s deteriorating condition did indeed qualify as harm to herself. Brown and her lawyers appealed the panel’s decision to New York State’s Court of Appeals, the state’s highest court.
While Brown’s case was on appeal, HHC petitioned the New York Supreme Court for permission to medicate her against her will. There were two days of hearings before Justice Kirschenbaum before the court appointed an independent psychiatrist, one Francine Cournos, to determine if Brown should be involuntarily medicated. Cournos advised against involuntarily medication. She judged that antipsychotic medication would improve Brown’s condition, but that Brown had not so far harmed herself or others, and that involuntarily medication now might alienate Brown from treatment and cause her to reject medication later on when it might be more necessary. Therefore, before the court of appeals had ruled on her case, Justice Kirschenbaum ordered Brown released and HHC did so a few days later, once they had located a unit for her in a Single Room Occupancy hotel. The court of appeals then considered the Brown case moot and declined to rule on the legality of her commitment.
Sam Tsemberis, who would later be known for developing the “Housing First” model of accommodating street dwellers, began working at Project HELP in August of 1988, soon after the Joyce Brown affair. He described his time at the project, and how it led to his changed thinking about street outreach, as follows:
I was working at the Columbia Psychiatric Institute at the time doing HIV prevention with the runaway kids. . . . When the job [at Project HELP] opened up I was living in East Village. . . . If you remember, it was very painful to walk around the city those days—this is before Giuliani locked everybody up at Rikers—just to see the number of homeless people, especially homeless people with mental illness. So when this job [at Project HELP] opened up . . . I met with [Richard C. Surles,] who was vice-president at Health and Hospitals Corporation because the program was run by HHC out of Gouverneur and Bellevue [hospitals]. . . . I met with him in July, I guess it was, a very hot day I remember. I went there in a short-sleeve shirt. He was a little taken aback. I was like no tie, short-sleeve shirt . . . applying for the director’s job. He was a very corporate kind of a person. . . .
As we’re talking, I tell him . . . I’m like more of, a humanist kind of a guy, R. D. Laing was one of my heroes. I don’t actually believe in involuntarily hospitalization. And, then he [said], wait—you’re applying for this job and you don’t believe in involuntary hospitalization? But he was intrigued by this. I don’t think he knew exactly what to make of it. He called me back for a second interview [and] we had a meeting of the minds, I think he understood that deep caring didn’t necessarily mean involuntary commitment. And he thought that I would be a good balance to a team that was too gung-ho to bring people to the hospital against their will. And so that’s how I started working there [in] August of ‘88.
Well, it was intriguing. . . . The only usefulness I saw in the program was where we would bring people to the [hospital]. There were people on the street that were medically compromised. People that were coughing up blood. . . . Remember the TB epidemic at that time. There was blood in their sputum. Other people had been standing or had not slept lying down for so long that their feet had swollen and they had cellulitis of the skin that had broken through and there was pus oozing through their sneakers. And then there were people that had lost limbs from gangrene. I mean there was a lot of bad stuff going on on the street [that was] painful to see.
And we would take people to the hospital, make up a mental illness excuse because of the way they have self-neglected themselves. You know, like, their lives were at risk because of their passivity. I mean, in that sense, there were many days when we thought like we had saved somebody’s leg or stopped or prevented something worse from happening to them [like] pneumonia. This was quite gratifying in a way, but there were many other days where the psychiatrists were just chasing down people just because they were delusional.
And that never had made any sense to me. Initially it would make sense to me, but it would make less sense to me when after being in Bellevue for a month they would be discharged back on to the street . . . usually carrying the same shopping bag that they had gone into Bellevue with. . . . There was twenty-one days of treatment, but the disposition never worked. They never got to a housing program or a residential program. And, those people, the returners—that wasn’t everybody, you know, but that was like a good third or a half of the people we were bringing in—began to really, really trouble me.
Because [the] cyclical aspect of . . . this intervention was not really working for them. Those people had characteristics that precluded admission . . . to housing. The kind of housing that was being built then—around ‘91, ‘92—was the first wave of New York/New York housing, and the housing strategy was basically to build mental health housing, not to build affordable housing like we actually need. They built boutique programs of twenty to thirty [units of] . . . supportive housing, [with] social workers on site, and they looked like re-institutions.
Now, in order to run those places, you needed to have people coming into them that were clean and sober, and compliant with psychiatric medication. And ready to follow the rules of these SROs. That was not going to be the people we are dealing with at Project HELP, I mean, of course, when you build one of those places they would immediately fill to capacity because there was so many homeless people that you could find people that could make those requirements. But the group I was dealing [with] could not get in. They . . . didn’t want to be associated with a mental health program . . . [and] stigmatized as living in a building that’s solely for mentally ill people. And even if they did want to get in they couldn’t stay clean long enough or agree to take medication because the medication had terrible side effects. It was not for them. Or they couldn’t believe they actually had a mental illness that required medication. So, there was a real disconnect there for us and that was a very frustrating aspect of the work.67
Before August of 1988 the most striking battles over homelessness policy were the court confrontations of Callahan and McCain and related cases. Homeless people themselves participated in these battles by proxy, being represented by advocates like Robert Hayes and Steven Banks. Outside of the courtroom, on the streets, the role played by the homeless was a passive one: through their public presence they were seen by other New Yorkers and projected powerful images of want and suffering into the city’s consciousness. I will discuss how the visibility of the homeless, and their disruption of public expectations of order, pushed their predicament onto the political agenda. But in the late summer of 1988 and off and on through the early nineties, the homeless, some of them, became the center of a more active form of disruption: the disturbances surrounding Tompkins Square Park in the Lower East Side of Manhattan.
Tompkins Square Park lies between Avenues A and B, and East Seventh and East Tenth streets, in Manhattan’s Lower East Side. By the late 1980s important neighborhood groups included yuppies, who were beginning the process of gentrifying the neighborhood; squatters, mostly young and radical, who were living in abandoned buildings and trying to keep them out of the hands of speculators and the city, which was often the official owner; older, working-class residents, the last representatives of a period when most of the neighborhood was Ukrainian; and partygoers and tourists who visited the many music clubs in the area.
In June, members of Avenue A Block Association, a community group that represented the interests of local owners and renters, had gone to a meeting of Community Board 3 to call for enforcement of noise ordinances, a greater police presence, and the imposition of a curfew. Accounts differ as to whether the board accepted, or merely made note of, the call for a curfew.68 In any case, Ninth Precinct commander Captain Gerald McNamara began enforcing a 1:00 a.m. curfew on July 11. But the curfew was enforced inconsistently, with partiers and club goers being asked to leave but the homeless generally being allowed to stay. Nonetheless, protests against the curfew began to develop. A relatively small skirmish between police and neighborhood residents took place on the evening of July 31. The big riot broke loose on August 6–7. A journalist at the park that night described the events as follows:
It started before midnight with a ragged little rally directed at the park faithful—the unlucky, the unruly, your tired, your poor. Near the entrance at St. Marks and Avenue A a plump balding man in tie-dye exhorted about a hundred punks, politicos, and curious neighbors through a tinny speaker system: “Yuppies and real estate magnates have declared war on the people of Tompkins Square Park!” . . . The cops were going to shut the park down as they had every night that week. . . . The grim cops at the gate, the chants of “Die Yuppie scum,” the M80s exploding deeper in the park—all added to the aura of latent violence.
Suddenly the cops had their riot helmets on and clubs out. Someone in front of a bar threw a bottle toward the mounted police massed at 7th Avenue and the cops backed up. . . . Another bottle smashed on the pavement. And another. The mounted police backed up again. The foot patrolmen stood shoulder-to-shoulder at the park entrance. . . . By now the crowd numbered in the hundreds.
About 12:55, I heard an explosion and the mounted police suddenly charged up Avenue A, scattering the knot of demonstrators still in the streets. I ducked behind a car. The police were radiating hysteria. . . . They were sweeping 9th Avenue and it didn’t matter if you were press or walking home from the movies or sitting on your stoop to catch a breeze. You were going to move. At First Avenue, I watched two cops on horseback gallop up on the sidewalk and grab a guy by his long hair, pulling him across the street between them. Minutes later, the same guy was down on the sidewalk in front of Stromboli’s, bleeding.
The cops seemed bizarrely out of control, levitating with some hatred I didn’t understand. They’d taken a relatively small protest and fanned it out over the neighborhood, inflaming hundreds of people who’d never gone near the park to begin with. They’d called in a chopper. And they would eventually call 450 officers.69
In response to the riot, Mayor Koch decided to suspend enforcement of the curfew that had been the focus of the demonstration that turned ugly.70 There began a cycle that stretched over the next few years wherein the number of the homeless in the park would grow, the city would crack down and drive them out, but they would soon return again. With the curfew no longer being enforced, over the next few months the homeless began returning to the park, until by early July 1989 about one to two hundred people made their home in the park, and had constructed a set of makeshift lean-tos and tents for shelter.71 On July 5, 1989, a Parks Department regulation that prohibited permanent structures in the parks was enforced by more than 250 police in riot gear who smashed down the shanties with axes and sledgehammers. Even as they did so, police acknowledged that the homeless would soon make their homes again in the park, as only the construction of permanent shelters, not living in the park, was being prohibited. And indeed, by November 1989 the homeless returned to the park again, and the Koch administration announced a final effort to clear the park. This time, it was hoped, the affair would be definitively settled as a support center would be built on the outskirts of the park to offer the homeless people living there various services and placements that were expected to remove them from the park.72 It was not until the beginning of the Dinkins administration that those hopes would be tested.
A theory of the politics of poverty, developed by Frances Piven and Richard Cloward, holds that disruptive action by the poor is the key to the expansion of social provision.73 The incidents surrounding Tompkins Square Park do not bear out this hypothesis. As we shall see, the upshot of these disruptions was not an expansion of rights but restrictions on the use of city parks by the homeless. The riot of August 6–7, 1988, was not so much a riot by the homeless for their rights as a riot by the police against the homeless and other occupiers of the park, so it is unsurprising that that event did not advance the cause of the homeless. But neither did the many other disruptions around the park by the homeless and their supporters contribute to the establishment or expansion of a right to shelter. In this case disruption resulted in retrogression rather than progress.
At the very end of the Koch administration, work began on an accord that would be described as a “landmark” and a “watershed” in the effort to house the homeless.74 This was the New York/New York Agreement, signed on August 22, 1990, which was a city and state program to provide 3,314 units of supported housing for the mentally ill homeless, with a capacity to house 5,225 people, by June 30, 1992.75 Between them the city and the state allocated $194.7 million in capital funds to build the agreement’s housing.76 This was far more money than had ever been spent on a project of this type. Years later, when the state official who negotiated the agreement, Cindy Freidmutter, presented these numbers to mental health policy makers of other states, they would be shocked at the gigantic size of the New York effort.77
As the name suggests, the agreement was between New York City and New York State. Years of frustration among advocates and policymakers over the lack of permanent housing for the mentally ill was one reason the agreement was widely regarded as a great breakthrough. Since, as often happens under American federalism, the two levels of government had failed to coordinate for so long, the eventual agreement came to many as a great surprise.
For years, the city had argued that the problem of the mentally ill homeless should be dealt with by the state, because mental health policy had traditionally been a state concern and because it had been the deinstitutionalization of state mental hospitals that had left many mentally ill people without a place to stay. Moreover, the city was reluctant to commit to providing housing for the homeless mentally ill because it had already undertaken to provide shelter and permanent housing for homeless families and AIDS patients and feared setting the precedent of aiding an unending string of special-needs constituencies.78
Meanwhile, the state argued that it was the city’s failure to preserve SRO hotels and other sources of inexpensive housing that had created the crisis. Further, the state was highly defensive on the subject of the mentally ill homeless. Landmark studies by psychiatrist Elmer Struening and colleagues had shown that about one-third of the residents of the city’s shelters for men were mentally ill. On that basis the city had floated proposals to have the state assume responsibility for all mentally ill shelter clients. Badly stung, the State Office of Mental Health had pronounced the studies by Struening and others as “political documents” ginned up by Koch to support his reelection efforts. Moreover, the state was faced with a lawsuit, Klostermann v. Cuomo, that claimed that the state’s responsibility for the mentally ill extended to providing them with safe housing. The state was therefore eager to disclaim any responsibility for the homeless mentally ill.79 As a result, well after the deinstitutionalization of the fifties and early sixties, and the low-income housing destruction of the seventies, little public-sector progress had been made on housing the homeless mentally ill.
The New York/New York Agreement was developed so late in the Koch administration that the final step of signing the agreement took place under Dinkins. However, all important negotiations and decisions were made while Koch was still in office. For the city, the process was overseen by William Grinker, HRA commissioner, and actual negotiations were done by his deputy administrator for policy and program development, Diane Baillargeon, and by Frank Lipton, a psychiatrist in charge of the department’s Office of Psychiatry. For the state, discussions took place under the aegis of Richard C. Surles, commissioner of the State Office of Mental Health. His chief negotiator was Cindy Freidmutter.
As was noted earlier, many political analysts emphasize the difficulty of achieving change. What then made possible this “monumental advance”?80 On the city’s part, the tough primary challenge Koch was facing from David Dinkins, who was emphasizing the weakness of the mayor’s homelessness policy, made Koch eager to achieve some progress he could claim credit for. As for the state, years of stinging criticism for its treatment finally began to have an effect. Throughout the mid- and late 1980s the state responded to critics by opening up community residences—transitional group residences for mentally ill clients waiting to be placed into permanent housing—throughout the state. Governor Mario Cuomo continued this responsive trend by appointing a progressive director to the State Office of Mental Health, Richard C. Surles, who was given a mandate to further address the issue of the homeless mentally ill. It also helped that the city and state had recently come to an agreement on the difficult issue of crowded city emergency rooms referring inpatients to state psychiatric facilities. The trust thus established encouraged progress on larger issues.81
Perhaps surprisingly, given the exasperation both sides had felt toward each other on the issue of homelessness, negotiations went easily. Baillargeon recalled that negotiations were not difficult. According to her, “We weren’t negotiating with ‘somebody’s going to win, somebody’s going to lose’ . . . we were professionals saying this is a positive outcome all the way around.”82 That account squares with Freidmutter’s:
I think they picked two pragmatic people who—Diane and I—were problem solvers. We were there to solve the problem, not to duke each other out. We were two women, building a consensus rather than trying to win the fight. And we figured out pretty early that we had people behind us who had very strong ideological views and that bringing them into a room was not going to make this work. So we did agree . . . to really take our points back and try to work out this agreement point by point—going back to our principles without bringing in everybody. I’m not sure [there was] ever a meeting with everyone at the table. I don’t remember one.83
But why did negotiations go so smoothly on this fractious issue? The negotiators agree that the key was a combination of timing, receptivity, and preparedness. In the fragmented American policymaking system, one sometimes simply has to wait for a moment of syzygy to develop when all the factors are aligned. Freidmutter remarked,
I’ve been in government . . . more than thirty years, and I’ve been around for a lot of those moments. I’ve seen these kinds of change issues and have worked with government on a lot of these issues. They just come. Really . . . you could try to say it rationally, but sometimes the moment comes when you can do something transformational. And often those moments pass because people don’t work fast enough or they can’t reach agreements, and they let the moment pass. And the timing is off.84
Thus Freidmutter credits Surles because “he understood that when the moment is right for something, you don’t walk away on principle, you figure out a way to do it in the best way possible.”85
Baillargeon emphasizes the need to be prepared when the moment comes. Much has been said in recent years about the importance of ideas in public policy, but one needs to be ready with more than an idea. Baillargeon had been responsible for developing a plan for homelessness policy. Having a plan turned out to be crucial:
My experience with this kind of stuff is that there are just these moments in time when . . . it’s just the right moment. . . . And I think that part of the moment was that we were doing this plan. . . . Plans have a way of often getting put on a shelf. . . . But for the most part, when you’re doing a plan and you’re in government, people pay a little bit of attention. And also you don’t want [to] do a plan that’s boring, that has nothing new in it. And so from my point of view, I was shopping for useful, interesting, cool stuff that might get us out of the shelter business and into something better. . . . And I think the result of that was . . . a kind of openness that doesn’t always happen just because one person has a good idea. . . . The plan had to be vetted by OMB, it had to be vetted by City Hall, it had to be vetted all over the place. And so the idea was raised to a higher level, rather than just an isolated good idea.86
The terms of the agreement mostly lived up to the hyperbole with which it was greeted. First, the schedule was ambitious: two years to produce more than three thousand housing units. It is true that a series of extensions resulted in the last residences not being opened until October 1998, which was still an impressive accomplishment given the parties’ fiscal problems and the challenges of siting housing for disabled populations in New York City. Also, New York/New York was the first housing program targeted exclusively at the mentally ill homeless. Given the heat it had taken for so long for ignoring the deinstitutionalized and noninstitutionalized, the state was adamant that this new housing should go only to the severely mentally ill. At the time, this was ground-breaking attention to a long-neglected population, but it meant that people with substance abuse problems, people with chronic medical conditions, people living with HIV and AIDS, and people suffering from personality disorders were excluded from the agreement unless they were also severely mentally ill. (These groups would eventually be covered by a later city/state accord negotiated during the Bloomberg administration.) The city was equally insistent that, under the agreement, housing would go to the literally homeless only; people living in doubled-up or otherwise unsatisfactory situations were not eligible. Another innovative feature of the agreement designated a not-for-profit service provider, the Center for Urban Community Services, to serve as an advisor to the shelters, outreach teams, and other organizations that would refer potential clients to agreement-provided housing. The idea was to provide both large and small referral agencies with assistance in negotiating a complex bureaucracy.
Implementation of the agreement is generally thought to have been a success. Ted Houghton, in his definitive historical account, expresses the consensus opinion: “Despite some not so inconsequential delays, the NY/NY Agreement to House Homeless Mentally Ill Individuals of 1990 has been remarkably successful at achieving its promise.” By December 31, 2000, a total of 7,774 persons had at one point or another been placed in housing sponsored by the first New York/New York Agreement.87
By the end of the Koch administration, the entitlement stage of homelessness policy had been consolidated by being extended beyond homeless individuals to homeless families. That consolidation had been achieved mostly through the efforts of advocates for the homeless in the courts. But the Koch administration, though it never succeeded in raising the quality of all shelters to an acceptable level, deserves considerable credit for creatively muddling through the necessarily challenging implementation of that entitlement. Also, during the Koch years certain limitations and dilemmas, real and apparent, of the entitlement phase began to manifest themselves. The Keener and “New Arrivals” studies suggested that certain limitations on the right to shelter might eventually be necessary. The existence of a small but salient population of street dwellers who would not take advantage of the right to shelter was addressed, unsuccessfully, in the involuntary-transportation policy of Project HELP. These and other challenges of the entitlement stage would have to be dealt with by the Dinkins administration. The policy learning process had begun and would continue.