`
`UNITED STATES DISTRICT COURT
`SOUTHERN DISTRICT OF NEW YORK
`
`
`
`
`AMERICAN COUNCIL OF THE BLIND OF
`NEW YORK, INC., MICHAEL GOLFO, and
`CHRISTINA CURRY, on behalf of themselves
`and all others similarly situated,
`
`
`Plaintiffs,
`
`
`
`v.
`
`CITY OF NEW YORK, NEW YORK CITY
`DEPARTMENT OF TRANSPORTATION,
`BILL DE BLASIO, in his official capacity as
`Mayor of the City of New York, and POLLY
`TROTTENBERG, in her official capacity as
`Commissioner of the New York City Department
`of Transportation,
`
`
`Defendants.
`
`
`
`
`
`
`
`
`
`
`
`
`
`18 Civ. 5792 (PAE)
`
`OPINION &
`ORDER
`
`
`PAUL A. ENGELMAYER, District Judge:
`
`
`This case involves a challenge under the Americans with Disabilities Act of 1990
`
`(“ADA”) and related statutes to the accessibility of New York City’s signalized crosswalks
`
`to blind and low-vision pedestrians.
`
`On behalf of a certified class of blind and low-vision New York City pedestrians,
`
`plaintiffs—the American Council of the Blind of New York, Inc. (“ACBNY”), Michael Golfo,
`
`and Christina Curry—have sued the City of New York, the New York City Department of
`
`Transportation (“DOT”), Mayor Bill de Blasio, and DOT Commissioner Polly Trottenberg
`
`(collectively, the “City” or “defendants”). Plaintiffs allege that the City has long failed to
`
`provide non-visual crossing information at the vast majority of its signalized intersections, i.e.,
`
`those which provide visual crossing information to sighted pedestrians. Plaintiffs allege that the
`
`City’s failure to accommodate blind and low-vision pedestrians violates Title II of the ADA,
`
`
`
`Case 1:18-cv-05792-PAE Document 126 Filed 10/20/20 Page 2 of 67
`
`
`
`42 U.S.C. § 12132; section 504 of the Rehabilitation Act of 1973, 29 U.S.C. § 794(a)
`
`(“Rehabilitation Act”); and the New York City Human Rights Law, N.Y.C. Admin. Code
`
`§ 8-107(4)(a) (“NYCHRL”).
`
`With discovery complete, plaintiffs now move for summary judgment on all claims, but
`
`solely as to liability, without yet seeking a judicial remedy. Plaintiffs principally argue that
`
`(1) the absence of non-visual crossing information at the vast majority of the City’s signalized
`
`intersections denies blind and low-vision pedestrians meaningful access to those intersections
`
`and the pedestrian grid, in violation of the ADA, the Rehabilitation Act, and the NYCHRL; and
`
`(2) the City’s failure to add non-visual crossing information on occasions when it has done
`
`construction at, or upgraded aspects of, the same intersections is also unlawful.
`
`For the following reasons, the Court grants plaintiffs’ summary judgment motion in
`
`principal part. Most significantly, the Court finds, on the undisputed facts, that the near-total
`
`absence at the City’s signalized intersections of crossing information accessible to blind and low-
`
`vision pedestrians denies such persons meaningful access to these intersections, in violation of
`
`all three statutes cited above. The Court also grants plaintiffs’ motion as to liability on their
`
`claim that the City’s failure to add non-visual street-crossing information at particular
`
`intersections at which it installed new traffic signals after June 27, 2015, violates the ADA and
`
`Rehabilitation Act. The Court otherwise denies plaintiffs’ motion for summary judgment. The
`
`case will now proceed promptly forward on two tracks: (1) to determine the remedy for the
`
`violations that have been established; and (2) to resolve plaintiffs’ open claims.
`
`
`
`2
`
`
`
`Case 1:18-cv-05792-PAE Document 126 Filed 10/20/20 Page 3 of 67
`
`
`
`I.
`
`Background
`A.
`
`Factual Background1
`
`According to the U.S. Census Bureau’s 2017 American Community Survey 1-Year
`
`Estimates, among New York City’s non-institutionalized population, 205,212 persons are blind
`
`or have other vision difficulties. Atkinson Decl., Ex. 1 (“Census Community Survey”) at 2.
`
`That amounts to approximately 2.4% of the City’s population. See id.
`
`1.
`
`The Parties
`
`Plaintiff ACBNY is a New York non-profit corporation. JSF ¶ 1. Its purpose is to
`
`“support and promote the educational, vocational and social advancement of people with vision
`
`disabilities.” Id. Its members include individuals with vision disabilities within the meaning of
`
`the ADA, Rehabilitation Act, and NYCHRL, including approximately 45 members in its Greater
`
`New York Chapter, which includes New York City. Id. ¶¶ 2, 20.
`
`
`1 The Court draws its account of the underlying facts from the parties’ respective submissions on
`the motion for summary judgment, including: the parties’ joint statement of undisputed facts,
`Dkt. 91 (“JSF”); plaintiffs’ Local Rule 56.1 statement, Dkt. 93 (“Pl. 56.1”); defendants’ Local
`Rule 56.1 counter-statement, Dkt. 105 (“Def. 56.1”); the declaration of Lori Scharff, Dkt. 95
`(“Scharff Decl.”); the declaration of Christina Curry, Dkt. 96 (“Curry Decl.”); the declaration of
`Michael Golfo, Dkt. 97 (“Golfo Decl.”); the declaration of Torie Atkinson, Esq., Dkt. 98
`(“Atkinson Decl.”), and supporting exhibits; the corrected declaration of Joshua Benson,
`Dkt. 107 (“Benson Decl.”), and supporting exhibits; the corrected declaration of James
`Celentano, Dkt. 108 (“Celentano Decl.”), and supporting exhibits; and the declaration of Pamela
`A. Koplik, Esq., Dkt. 102 (“Koplik Decl.”), and supporting exhibits.
`
`Citations to a party’s 56.1 statement incorporate the evidentiary materials cited therein. When
`facts stated in a party’s 56.1 statement are supported by testimonial, video, or documentary
`evidence and not denied by the other party, or denied by a party without citation to conflicting
`admissible evidence, the Court finds such facts to be true. See S.D.N.Y. Local Civil Rule 56.1(c)
`(“Each numbered paragraph in the statement of material facts set forth in the statement required
`to be served by the moving party will be deemed to be admitted for purposes of the motion
`unless specifically controverted by a correspondingly numbered paragraph in statement required
`to be served by the opposing party.”); id. Rule 56.1(d) (“Each statement by the movant or
`opponent . . . controverting any statement of material fact[] must be followed by citation to
`evidence which would be admissible, set forth as required by Fed. R. Civ. P. 56(c).”).
`3
`
`
`
`
`
`Case 1:18-cv-05792-PAE Document 126 Filed 10/20/20 Page 4 of 67
`
`
`
`Plaintiff Christina Curry is deaf, legally blind, and uses a forearm crutch as a mobility
`
`aid. Id. ¶ 3. As a result, she cannot see traffic in intersections unless it is “very close,” and
`
`cannot rely on visual street signals to cross the street. Id. ¶ 31. Because of her visual and
`
`auditory disabilities, she requires “tactile information” to use pedestrian signals. Id. She lives in
`
`the Bronx and regularly walks on New York City sidewalks to commute and as a part of her job
`
`as Executive Director of the Harlem Independent Living Center. Id. ¶¶ 32–34, 36. She avers that
`
`during her frequent pedestrian travel throughout New York City, she risks being hit by vehicles,
`
`fears for her life, is often grabbed by well-meaning pedestrians, and uses circuitous, sometimes
`
`costly, alternatives to walking to avoid such incidents—all because she cannot use the visual traffic
`
`signals that are available to sighted pedestrians. See id. ¶ 60; Curry Decl. ¶¶ 9–13.
`
`Plaintiff Michael Golfo is a blind resident of Tarrytown, New York, who previously
`
`commuted to Manhattan daily for work and now walks on New York City sidewalks about once
`
`per week to visit his doctors and friends. JSF ¶¶ 63–64. As a result of his disability, he relies on
`
`his hearing and guide dog to navigate the New York City streets. Id. ¶ 65. Even with these aids,
`
`Golfo finds it difficult to traverse the City; he has almost been hit by cars on many occasions
`
`while crossing the street. Id. ¶ 74. Like Curry, Golfo often must rely on sighted persons to help
`
`him make such crossings. Id. ¶¶ 75–76. He also often relies on expensive taxis or car services to
`
`navigate the City, or takes elongated routes to avoid difficult intersections. Id. ¶¶ 79–80.
`
`Defendants the City of New York and the DOT are responsible for and have broad
`
`authority over the City’s streets, and oversee the installation, repair, and maintenance of its
`
`traffic signals, sidewalks, crosswalks, and roadways. Id. ¶ 6. Each receives federal funds,
`
`including from the Federal Highway Administration (“FHWA”). Id. ¶ 5.
`
`
`
`4
`
`
`
`Case 1:18-cv-05792-PAE Document 126 Filed 10/20/20 Page 5 of 67
`
`
`
`Defendant Bill de Blasio, sued in his official capacity, is Mayor of the City of New York.
`
`Id. ¶ 7. Defendant Polly Trottenberg, also sued in her official capacity, is Commissioner of the
`
`DOT. Id. ¶ 8.
`
`2.
`
`New York City’s Pedestrian Landscape
`
`New York City has the highest population density of any major American city. Id. ¶ 9.
`
`Walking “is a major form of transportation in the city, and access to sidewalks is an important
`
`component of city life.” Id.
`
`To allow pedestrians safely to navigate its sidewalks and traffic intersections, the City has
`
`installed roughly 120,000 pedestrian control signals—i.e., devices informing pedestrians when to
`
`cross the street and when to wait—at about 13,200 of its 45,000 intersections. Id. ¶¶ 11, 13. Of
`
`those 13,200 signalized intersections, all but 443 (roughly 96.6%) communicate crossing
`
`information exclusively in a visual format, id. ¶ 19, with an image of a mid-stride white stick
`
`figure indicating “walk,” and an upraised orange hand indicating “don’t walk,” id. ¶ 108. These
`
`visual signals are inaccessible to the blind. Id. ¶ 138.
`
`3.
`
`Difficulties Facing Blind Pedestrians
`
`As the parties agree, any pedestrian attempting to cross a street must perform four tasks:
`
`(1) locate the edge of the street and crossing point; (2) align to cross the street; (3) decide when
`
`to begin crossing; and (4) maintain alignment (i.e., the correct direction) while crossing the street.
`
`Pl. 56.1 ¶ 6. Those with a visual disability, however, have difficulty accomplishing each of these
`
`tasks. Id. ¶¶ 8–15.
`
`Typically, a blind person traversing New York City sidewalks will stop when he or she
`
`encounters a curb or other detectable warning surface, and will assume that a crossing point is
`
`located there. Id. ¶ 8. However, without assistance, blind pedestrians begin crossing from
`
`outside the crosswalk (i.e., in a traffic lane) nearly 30% of the time. Id. ¶ 9. And once blind
`
`
`
`5
`
`
`
`Case 1:18-cv-05792-PAE Document 126 Filed 10/20/20 Page 6 of 67
`
`
`
`pedestrians locate a crossing point, they cannot receive the visual “walk” or “don’t walk”
`
`information communicated to sighted individuals by the standard pedestrian control signals
`
`described above. JSF ¶ 138. Instead, they generally must rely on other cues, such as fellow
`
`pedestrians’ movements or the sound of traffic parallel to the crosswalk, to ascertain when it is
`
`safe to walk. Pl. 56.1 ¶¶ 13–14. But picking the correct audible cues out of many traffic lanes
`
`can be difficult. Id. ¶¶ 11–12. That difficulty is often compounded, especially in the City, by
`
`idiosyncratic crosswalk architecture and quiet traffic such as hybrid cars and bicycles. Id. As a
`
`result, the assumption that a pedestrian has the “walk” signal when parallel traffic is moving is
`
`unreliable, leading blind pedestrians to attempt to cross the street against moving traffic up to
`
`half the time. Id. ¶ 14. Likewise, when following sighted pedestrians, a blind person might not
`
`realize that those pedestrians were in fact jaywalking until they hear onrushing traffic. See JSF
`
`¶¶ 56, 76.
`
`As a result, absent any accommodations, pedestrians with visual disabilities often
`
`encounter danger, inconvenience, and humiliation while attempting to use the City’s crosswalks.
`
`Unable to reliably locate crosswalks or time their crossing, they risk being hit by cars, JSF
`
`¶¶ 25, 58, 73–74, and becoming stranded in the middle of intersections, id. ¶ 73, and may be
`
`unwillingly grabbed by strangers hoping to assist them, id. ¶¶ 26, 59, 77; see also Scharff Decl.
`
`¶ 18 (“I feel that I and every other blind and deafblind pedestrian are taking our lives into our
`
`hands every time we walk out into the community.”). They are often forced to take longer, less
`
`convenient routes than they otherwise would to avoid troublesome intersections, JSF ¶¶ 27–28, 57,
`
`may wait up to 20 minutes at a single intersection to make sure they are crossing with other
`
`pedestrians, id. ¶¶ 24, 55, 75, and may forgo walking altogether in favor of more expensive, but
`
`safer, transit methods, such as taxis or car services, id. ¶¶ 27, 60, 79; Golfo Decl. ¶¶ 2, 9 (“These
`
`
`
`6
`
`
`
`Case 1:18-cv-05792-PAE Document 126 Filed 10/20/20 Page 7 of 67
`
`
`
`experiences are inconvenient, frustrating, often humiliating, and deny me my dignity and
`
`independence.”). Alternatively, rather than undertaking such fraught excursions, they may
`
`simply decline to participate in activities or visit portions of the City at all. See Golfo Decl. ¶ 8.
`
`4.
`
`Accessible Pedestrian Signals
`
`Accessible Pedestrian Signals (“APS”) are devices that communicate “walk” and “don’t
`
`walk” signals to pedestrians in a non-visual format, through audible tones, speech messages,
`
`and/or vibrating surfaces. JSF ¶ 14. According to plaintiffs’ expert, Janet M. Barlow, APS
`
`greatly increase blind pedestrians’ ability to safely cross intersections. Pl. 56.1 ¶¶ 14, 17.
`
`Where installed in New York City, APS function as follows. At an intersection with
`
`APS, devices on each street corner emit a soft “locator tone” every second, which allows blind
`
`and low-vision pedestrians to locate them when they approach the intersection. JSF ¶ 16. Each
`
`device also has a pushbutton with a raised arrow pointing at the crossing with which it is
`
`associated. Id. ¶ 15. Because each APS device points to a specific crosswalk from each side of
`
`the street, a standard four-cornered intersection equipped with APS typically requires two
`
`devices per curb, and eight per intersection. See id.
`
`When an APS button is pressed, the device responds with an audible speech message. Id.
`
`¶ 17. If the visual “walk” signal is not on for the crosswalk associated with that APS device, the
`
`device audibly says “wait.” Id. If, on the other hand, pedestrians do have a “walk” signal for the
`
`relevant crosswalk, then the APS device either says “walk sign is on” or emits a rapid ticking
`
`sound. Id. ¶ 18. The pushbutton itself also vibrates, to alert pedestrians who are both deaf and
`
`blind that it is safe to walk. Id. APS thus inform blind individuals, solely through audio and
`
`tactile cues, both where and when to cross the street safely.
`
`
`
`7
`
`
`
`Case 1:18-cv-05792-PAE Document 126 Filed 10/20/20 Page 8 of 67
`
`
`
`5.
`
`New York City’s APS Program
`
`The City installed its first APS device in 1957, but as of 2003 had only added about a
`
`dozen throughout the City. JSF ¶¶ 87–88. Around 2004, the DOT established an informal APS
`
`Program, pursuant to which it installed APS upon request, but only near organizations or
`
`programs that served large numbers of blind and low-vision persons. Id. ¶¶ 90, 93, 111.
`
`Each year between 2004 and 2011, the DOT received more requests for APS than it
`
`installed. Id. ¶ 109. It chose among these requests without any formal ranking system. Id.
`
`Beginning in 2011, however, the DOT developed a Prioritization Tool for ranking these
`
`requests. Id. ¶ 113. The Prioritization Tool is a “worksheet completed by traffic engineers,”
`
`which evaluates the characteristics of an intersection, including its “geometry, crossing width,
`
`the presence of bike lanes or signal timing changes, traffic volume, and proximity to high-
`
`pedestrian areas,” and determines that intersection’s priority for the installation of APS. Id.2
`
`Each characteristic is assigned a numeric score, with features that make a crosswalk more
`
`challenging for blind pedestrians producing a higher score. Id. ¶ 117. Cost is not a factor in this
`
`ranking process. Id. ¶ 125. Once the DOT evaluates an intersection, it generally enters its
`
`ranking into the City’s database with other ranked locations, to determine how it compares with
`
`other intersections. Id. ¶ 118. Because newly ranked intersections may be ranked as higher
`
`priorities than ones already on the list, a ranked intersection may remain on the ranking list for
`
`years. Id. ¶¶ 127–128. The DOT does not appear to maintain a policy for deciding which
`
`intersections to rank at any given time, but today it evaluates and ranks at least all intersections
`
`for which it receives a new APS request. Pl. 56.1 ¶ 65; JSF ¶ 121. As of June 2019, the DOT
`
`had ranked approximately 1,600 intersections using the Prioritization Tool. JSF ¶ 129.
`
`
`2 The Prioritization Tool was adapted from a similar tool developed by the National Cooperative
`Highway Research Program, overseen by the National Academy of Sciences. JSF ¶ 114.
`8
`
`
`
`
`
`Case 1:18-cv-05792-PAE Document 126 Filed 10/20/20 Page 9 of 67
`
`
`
`In 2012, the New York City Council passed a law requiring the City to install APS at 25
`
`locations per year. Id. ¶ 95. Between 2012 and 2015, the DOT installed APS at slightly more
`
`than 25 locations each year, bringing the total number of intersections with APS to 131 by
`
`November 1, 2015. Id. ¶¶ 97–100.
`
`In 2014, the New York City Council passed a new law requiring the City to install APS
`
`at 75 intersections annually, effective January 1, 2016. Id. ¶ 101. Between 2016 and 2018, the
`
`DOT installed APS at slightly more than 75 intersections each year, raising the total number of
`
`the City’s intersections with APS to 371 by 2019. Id. ¶¶ 102–104.
`
`Since 2019, the DOT has received appropriations to install more APS than are required
`
`under City law, with funding to install APS at 150 intersections per year in 2019–2020, 235 per
`
`year in 2020–2021, and 305 per year in 2021–2024. Id. ¶¶ 106–107. As of September 10, 2019,
`
`443 intersections within New York City were equipped with APS. Id. ¶ 108.
`
`Further, in November 2019, after plaintiffs had moved for summary judgment in this
`
`lawsuit, the New York City Council passed, and the Mayor signed, a law—the “Safe Streets
`
`Legislation”—mandating “master plans” for city streets, sidewalks, and pedestrian spaces. See
`
`N.Y.C. Admin. Code § 19-199.1; Koplik Decl., Ex. B. That law requires the DOT to issue a
`
`five-year master plan every five years, beginning December 1, 2021, and by December 1 of
`
`every fifth year thereafter. N.Y.C. Admin. Code § 19-199.1(b)(2), (c). It further sets certain
`
`benchmarks that each master plan must include. Id. § 19-199.1(c)(2). For both the 2021
`
`and 2026 master plans, the law requires the DOT to set, as a benchmark, the installation of APS
`
`at no fewer than 2,500 intersections within the next five years, with installation of such signals at
`
`no fewer than 500 intersections during each year of the plan. Id. § 19-199.1(c)(2)(vi), (3)(iii).
`
`Thus, in sum, the 2019 Safe Streets Legislation requires the installation of APS at 5,000
`
`
`
`9
`
`
`
`Case 1:18-cv-05792-PAE Document 126 Filed 10/20/20 Page 10 of 67
`
`
`
`intersections by 2031. The legislation does not, however, appear to allocate funding or resources
`
`to meet these goals. See Def. Mem. at 12 n.5.
`
`On October 21, 2019—the day plaintiffs moved for partial summary judgment—the DOT
`
`also issued a memorandum entitled “Accessible Pedestrian Signals Policy.” Koplik Decl., Ex. A
`
`(“2019 APS Memo”). That memorandum purports to “clarify” the City’s existing policy
`
`regarding the installation of APS. Id. at 1. First, it states that “[e]ach newly approved signal will
`
`be designed and constructed with APS.” Id.3 Second, it provides that APS will be installed
`
`whenever either a Capital Street Project or Street Improvement Project (“SIP”)4 involves the
`
`relocation or reconstruction of any traffic signal pole foundation. Id. at 2.
`
`6.
`
`Installing APS
`
`Even at intersections that already have pedestrian control signals, the process of installing
`
`APS is more burdensome than merely affixing a pushbutton unit to the signal. Because most
`
`intersections do not have extant signals at all the locations necessary to effectively install APS,
`
`“[t]he vast majority of locations require at least one or two new poles per corner.” Celentano
`
`Decl. ¶ 6(f). Installing such poles requires the excavation of part of the sidewalk, cutting into the
`
`roadway to install conduits and signal cables, and the installation of new foundation poles, as
`
`well as restoration of the roadway, sidewalk, and roadway markings. Id. ¶ 6(e)–(k). As a result,
`
`according to the DOT’s 30(b)(6) witness, the average cost per intersection to install APS
`
`
`3 As discussed below, that policy reverses the City’s historic practice, which was generally not to
`do so. Between 1990 and 2014, no new traffic-signal installations included APS. JSF ¶ 133.
`Beginning in 2014, the DOT began ranking new traffic signals for the addition of APS, but
`continued to install new signals at a faster rate than it was adding APS. Compare id. ¶ 135
`(between 100 and 120 new traffic signals each year), with id. ¶¶ 95–104 (between 26 and 83 new
`intersections equipped with APS each year since 2014).
`
` These terms are defined and discussed more fully below.
`10
`
` 4
`
`
`
`
`
`Case 1:18-cv-05792-PAE Document 126 Filed 10/20/20 Page 11 of 67
`
`
`
`averaged approximately $60,930 in 2018. Benson Decl. ¶ 17. For the 85 intersections that were
`
`equipped with APS in 2018, the total cost was $5.18 million. Id.
`
`7.
`
`Upgrades to Pedestrian Signals Other than APS
`
`Over the years, the City has improved and modified various aspects of its streets, pedestrian
`
`signals, and crosswalks, while not simultaneously installing APS. As discussed more fully
`
`below, plaintiffs allege that making such improvements without also including APS violates the
`
`ADA and Rehabilitation Act. The Court briefly describes each type of modification.
`
`a.
`
`New Signal Installations
`
`In a typical year, the City installs between 100 and 120 traffic signals. JSF ¶ 135.
`
`Between 1990 and 2014, none of these installations included APS. Id. ¶ 133. Between 2014
`
`and 2019, the City ranked new traffic signals using the Prioritization Tool, but did not
`
`consistently include APS with newly constructed signals. Id. ¶ 132. As of July 2019, DOT
`
`purports to have adopted a policy requiring the installation of APS whenever a new traffic signal
`
`is installed. See 2019 APS Memo at 1.
`
`b.
`
`Capital Street Projects and SIPs
`
`Capital Street Projects, or “Capital Projects,” are major reconstruction projects, which
`
`can range from repaving to full reconstruction of a roadbed, sidewalks, and utilities, among other
`
`eclectic projects. JSF ¶ 178. They are planned, funded, and initiated by the DOT, but actually
`
`constructed by the Department of Design and Construction (“DDC”) on the DOT’s behalf. Id. ¶ 179.
`
`A SIP is generally of a similar scale—and can include redesigning or reconfiguring
`
`intersection geometry, adding or removing lanes of traffic, adding or removing traffic and
`
`pedestrian control signals, and adding or removing bike lanes or crosswalks. Id. ¶ 171. They
`
`may also include “some level of signal work.” Id. ¶ 174. SIPs are carried out within the DOT,
`
`without the involvement of the DDC, id. ¶ 171, and, unlike Capital Projects, do not involve the
`
`
`
`11
`
`
`
`Case 1:18-cv-05792-PAE Document 126 Filed 10/20/20 Page 12 of 67
`
`
`
`full reconstruction of a roadbed or the utility lines underneath, see Benson Decl. ¶ 13. Even
`
`though “a good majority” of all SIPs involve “some level of signal work,” Atkinson Decl., Ex. 3
`
`(“Celentano Tr.”) at 48, only 14 of the 778 SIPs completed between 2009 and 2019 included the
`
`installation of APS at the intersection(s) affected, JSF ¶ 173.
`
`In the 2019 APS Memo, the DOT purports to have clarified its policy that it will install
`
`APS whenever a Capital Project or SIP involves the relocation or reconstruction of any traffic
`
`signal pole foundation. See 2019 APS Memo at 2.
`
`c.
`
`Signal Lens Replacements
`
`Before 2000, New York City’s pedestrian crossing signals displayed textual messages
`
`reading “walk” and “don’t walk” to indicate whether pedestrians should or should not cross the
`
`street. JSF ¶ 137. Between 2000 and 2004, the DOT replaced the displays in every pedestrian
`
`signal in the City to instead show a mid-stride white stick figure, signaling “walk,” and an
`
`upraised orange hand, signaling “don’t walk.” Id. ¶¶ 137, 139. According to the DOT’s 30(b)(6)
`
`witness, the City undertook that project to improve the “clarity of communication” to
`
`pedestrians, especially for those who do not speak English, but did not also install APS at the
`
`same time. Pl. 56.1 ¶ 70; Atkinson Decl., Ex. 2 (“Benson Tr.”) at 63. The DOT has conceded that
`
`the clarity provided by such a display “is not available to pedestrians who cannot see the
`
`display.” Def. 56.1 ¶ 71; Benson Tr. at 64. Unlike APS installation, the replacement of a signal
`
`lens requires only the removal of one lens and the attachment of a new one. It does not involve
`
`any construction with regard to the sidewalk, roadway, or any electrical cables or conduits.
`
`Celentano Decl. ¶¶ 21–22.
`
`Similarly, between 2012 and 2014, the DOT replaced over half of the City’s pedestrian
`
`signal displays to include a “countdown clock” indicating how much time a pedestrian has to
`
`finish crossing the street. JSF ¶ 141. As with the signal lenses discussed above, the City did not
`12
`
`
`
`
`
`Case 1:18-cv-05792-PAE Document 126 Filed 10/20/20 Page 13 of 67
`
`
`
`concurrently add APS. Id. ¶ 142. The benefit of a flashing visual countdown clock is not
`
`available to blind pedestrians. Id. ¶ 143. The installation of countdown clocks also involves
`
`only removing one lens from the crossing signal and attaching a new one. It does not involve
`
`any other physical alterations to the traffic signal or excavation of any roadway or sidewalk.
`
`Celentano Decl. ¶¶ 17–18.
`
`d.
`
`Leading Pedestrian Intervals and Exclusive Pedestrian Phases
`
`The City has also implemented two forms of timing changes at certain intersections,
`
`which give pedestrians additional time to cross the street without any vehicles entering the
`
`intersection.
`
`First, a Leading Pedestrian Interval (“LPI”) is a “signal timing treatment” that gives
`
`pedestrians an advanced “walk” sign for about seven to 10 seconds before the parallel vehicular
`
`traffic signal turns green. JSF ¶ 145. This allows pedestrians to enter a crosswalk before turning
`
`vehicles have an opportunity to do so. It thereby “makes pedestrians more visible to drivers and
`
`discourages motorists from aggressively turning before pedestrians begin crossing.” Benson
`
`Decl. ¶ 22. The DOT’s 30(b)(6) witness avers that the safety benefits of LPIs inure to sighted
`
`and blind pedestrians alike because an LPI is a “training tool” that reminds motorists to “exercise
`
`greater caution when driving around pedestrians.” Id. However, he also testified at his
`
`deposition that LPIs “can make an intersection more challenging for a blind or low vision
`
`pedestrian[]” because, during an LPI, “the lack of sound of parallel traffic movement is, you
`
`know, mysterious and confusing to someone who is not able to see.” JSF ¶ 147; see also id.
`
`¶ 149 (DOT received complaints that LPIs are “unsafe for the visually impaired”). By
`
`August 2019, the City had implemented LPIs at 3,951 intersections, only 113 of which also have
`
`APS. Id. ¶¶ 158–159. The presence of an LPI increases an intersection’s ranking—i.e., its
`
`priority for being upgraded with APS—under the DOT’s Prioritization Tool. Id. ¶ 162.
`13
`
`
`
`
`
`Case 1:18-cv-05792-PAE Document 126 Filed 10/20/20 Page 14 of 67
`
`
`
`Second, an Exclusive Pedestrian Phase (“EPP”), or “Barnes Dance,” is a timing treatment
`
`in which pedestrians at all corners of an intersection have an exclusive interval during which to
`
`cross using any of the crosswalks within an intersection. Id. ¶ 163. In other words, all vehicular
`
`traffic has a red light, and all pedestrians at the intersection have a “walk” sign. Similar to LPIs,
`
`EPPs can cause confusion and disorientation to blind and low-vision pedestrians who, without
`
`APS, generally rely on the sound of parallel traffic to determine when it is safe to cross. Id.
`
`¶ 164 (Benson: “My understanding of it is that a lot of people with vision impairments have
`
`either learned on their own or received training to use the sound of traffic movements to help
`
`navigate. So in the absence of those sounds, it becomes disorienting.”). As of June 2019, 98
`
`intersections in New York City have EPPs, none of which also are equipped with APS. Id.
`
`¶¶ 168–169. As with LPIs, the presence of an EPP increases an intersection’s ranking under the
`
`DOT’s Prioritization Tool. Id. ¶ 170.
`
`“No physical alterations” are performed when implementing an LPI or EPP at a given
`
`intersection. Celentano Decl. ¶¶ 9, 14. Rather, each requires only that a new “phasing
`
`document” be uploaded to the intersection’s traffic controller via a flash drive with the new
`
`timing information. Id. ¶¶ 8, 13.
`
`8.
`
`Regulatory Guidance on the Installation of APS
`
`The U.S. Department of Justice (“DOJ”) is responsible for promulgating regulations
`
`under the ADA. See 42 U.S.C. § 12134(a). However, Congress has also tasked a separate
`
`federal agency, the Architectural and Transportation Barriers Compliance Board (“Access
`
`Board”) with issuing “minimum guidelines” under the ADA, id. § 12204(a), with which the
`
`DOJ’s regulations “shall be consistent,” id. § 12134(c). These guidelines “do not have any
`
`binding effect on their own, but instead help shape the Attorney General’s regulations, which
`
`
`
`14
`
`
`
`Case 1:18-cv-05792-PAE Document 126 Filed 10/20/20 Page 15 of 67
`
`
`
`must be ‘consistent’ with the Board’s guidelines.” Paralyzed Veterans of Am. v. Ellerbe Becket
`
`Architects & Eng’rs, P.C., 950 F. Supp. 389, 390 (D.D.C. 1996).
`
`In 1999, the Access Board began the rulemaking process for guidelines related to
`
`pedestrian facilities in “public rights-of-way,” and in 2002 and 2005 released proposed drafts of
`
`such guidelines. See Accessibility Guidelines for Pedestrian Facilities in the Public Right-of-
`
`Way, 76 Fed. Reg. 44664, 44667 (July 26, 2011) (notice of proposed rulemaking summarizing
`
`past draft guidelines and regulatory history).
`
`On July 26, 2011, the Access Board published in the Federal Register a notice of
`
`proposed rulemaking, proposing its Accessibility Guidelines for Pedestrian Facilities in the
`
`Public Right-of-Way, or “PROWAG.” Id.5 However, the Access Board has not formally
`
`promulgated a final rule adopting the PROWAG. And the DOJ has never taken action
`
`conforming its regulations to the PROWAG’s recommendation. See, e.g., Sarfaty v. City of Los
`
`Angeles, No. 17 Civ. 3594 (SVW), 2020 WL 1078804, at *5 (C.D. Cal. Feb. 7, 2020) (“The Court
`
`declines to consider the PROWAG standards in its analysis, because (as noted by the City), they
`
`have not yet been adopted by the DOJ, the agency responsible for implementing ADA
`
`regulations.”); Scharff v. County. of Nassau, No. 10 Civ. 4208 (DRH), 2014 WL 2454639, at *12
`
`(E.D.N.Y. June 2, 2014) (noting that Access Board has “proposed, but not yet promulgated,
`
`standards for public rights-of-way”).
`
`The PROWAG contains some guidance on the installation of APS. Relevant here, it
`
`states at R209.1 that “[w]here pedestrian signals are provided at pedestrian street crossings, they
`
`shall include accessible pedestrian signals and pedestrian pushbuttons.” 76 Fed. Reg. at 44690.
`
`
`5 Although the Access Board’s guidelines do not use this language, the acronym PROWAG
`stands for “Proposed Right-of-Way Access Guidelines,” and reflects common usage. See, e.g.,
`Def. 56.1 ¶¶ 22–23.
`
`
`
`15
`
`
`
`Case 1:18-cv-05792-PAE Document 126 Filed 10/20/20 Page 16 of 67
`
`
`
`That specification appears to relate only to newly installed pedestrian signals, because R209.2
`
`further states that “[e]xisting pedestrian signals shall comply with R209.1 when the signal
`
`controller and software are altered, or the signal head is replaced.” Id