480 F.3d 1
MR. I., as parent and next friend of L.I., a minor; Mrs. I., as parent and next friend of L.I., a Minor, Plaintiffs, Appellees/Cross-Appellants,
MAINE SCHOOL ADMINISTRATIVE DISTRICT NO. 55, Defendant, Appellant/Cross-Appellee.
United States Court of Appeals, First Circuit.
Heard December 5, 2006.
Decided March 5, 2007.
Eric R. Herlan with whom Drummond Woodsum & MacMahon, was on brief, for appellant/cross appellant.
Richard L. O’Meara with whom Amy M. Sneirson, Staci K. Converse and Murray, Plumb & Murray, were on brief, for appellees/cross-appellees.
Diane C. Smith, on brief for amici curiae Autism Society of Maine, Council of Parent Advocates and Attorneys, Disability Rights Center, and National Disability Rights Network.
Brendan P. Rielly and Jensen Baird Gardner & Henry, on brief for amici curiae Maine School Management Association, Maine Education Association, Maine Administrators of Services for Children with Disabilities, and Maine Principals’ Association.
Frank D’Alessandro and Kids Legal at Pine Tree Legal Assistance, on brief for amici curiae Asperger’s Association of New England.
Before TORRUELLA, Circuit Judge, CYR, Senior Circuit Judge, and HOWARD, Circuit Judge.
HOWARD, Circuit Judge.
This case presents an issue of eligibility for benefits under the Individuals with Disabilities Education Act, 20 U.S.C. ? 1400 et seq. (Supp.2006) (the “IDEA”). We have previously noted that such issues can require a “difficult and sensitive” analysis. Greenland Sch. Dist. v. Amy N., 358 F.3d 150, 162 (1st Cir.2004) (not reaching the eligibility question). This case is no exception. The appellant, Maine School Administrative District No. 55 (“the district”), appeals the district court’s determination that the appellees’ daughter (“LI”) qualifies as a “child with a disability” eligible for special education and related services under the IDEA as a result of her Asperger’s Syndrome. The appellees (“Mr. and Mrs. I” or “the parents”) cross-appeal the district court’s rulings that (1) even though LI was entitled to IDEA services, her parents were not entitled to reimbursement of their expenses in unilaterally placing LI in a private school following the district’s refusal to provide those services and (2) the district would not be separately ordered to provide compensatory education services to reverse the effects of that decision on LI’s progress. We affirm the judgment of the district court.
We begin with an overview of the statutory framework. The IDEA provides funding to each state “to assist [it] to provide special education and related services to children with disabilities,” 20 U.S.C. ? 1411(a)(1), provided that “[a] free appropriate public education is available to all children with disabilities residing in the state. . . .” Id. ? 1412(a)(1)(A). In this sense, a “free appropriate public education” encompasses “special education and related services,” id. ? 1401(9), including “specially designed instruction, at no cost to parents, to meet the unique needs of a child with a disability. . . .” Id. ? 1401(29).
To receive special education and related services under the IDEA, a child must qualify as a “child with a disability.” In relevant part, a “child with a disability” is a child
(i) with mental retardation, hearing impairments (including deafness), speech or language impairments, visual impairments (including blindness), serious emotional disturbance (referred to in this chapter as “emotional disturbance”), orthopedic impairments, autism, traumatic brain injury, other health impairments, or specific learning disabilities; and
(ii) who, by reason thereof, needs special education and related services.
Id. ? 1401(3)(A). The Secretary of Education has promulgated a regulation defining each of the categories of disability set forth in ? 1401(3)(A)(i). Those definitions, so far as they are relevant here, require that each of the enumerated conditions “adversely affect[ ] a child’s educational performance” to constitute a disability. 34 C.F.R. ?? 300.8(c)(1)(i) (2006) (autism), (c)(4)(i) (emotional disturbance), (c)(9)(ii) (other health impairment).1
The IDEA places the burden of identifying children with disabilities upon each state. 20 U.S.C. ? 1412(a)(3)(A). In deciding whether a particular student has a disability under the IDEA, Maine uses a “pupil evaluation team,” or “PET,” 05-071-101 Me.Code. R. ? 9.4 (2006), consisting of the student’s parents, a representative from the school district, and a number of educational and other professionals. Id. ? 8.6; see also 20 U.S.C. ? 1414(d)(1)(B). Though the members of the PET attempt to achieve consensus on this issue, the school district retains the “ultimate responsibility to ensure that a student is appropriately evaluated” for IDEA eligibility. 05-071-101 Me.Code. R. ? 8.11(C).
The parents of a child deemed ineligible for IDEA benefits can challenge that determination before an impartial hearing officer. 20 U.S.C. ?? 1415(b)(6), (f)(1)(A), (f)(3)(A). After the hearing, the officer issues a final administrative decision, accompanied by findings of fact. Id. ?? 1415(h)(4), (i)(1)(A). Any party aggrieved by the decision can then file a civil action in federal district court. Id. ? 1415(i)(2)(A). Then the “trial court must make an independent ruling based on the preponderance of the evidence, but the Act contemplates that the source of that evidence generally will be the administrative hearing record, with some supplementation at trial.” Town of Burlington v. Dep’t of Educ., 736 F.2d 773, 790 (1st Cir.1984), aff’d sub nom. Sch. Comm. v. Dep’t of Educ., 471 U.S. 359, 105 S.Ct. 1996, 85 L.Ed.2d 385 (1986) (“Burlington“); see also 20 U.S.C. ? 1415(i)(2)(C).
In keeping with this approach, the district court referred the case to a magistrate judge for proposed findings and a recommended disposition, see 28 U.S.C. ? 636(b)(1)(B) (2006), which were made based on the facts adduced at the due process hearing and supplemental evidence submitted by the parents. The district court, in the absence of an objection from either side, accepted the magistrate’s proposed findings wholesale. In the continued absence of any challenge to these factual findings, we take the same tack.
LI attended Cornish Elementary School in Cornish, Maine, until 2003. Though she excelled academically, by the fourth grade she began to experience sadness, anxiety, and difficulty with peer relationships. These problems persisted into the fifth grade, when LI sought to distance herself physically from most of her classmates. Her parents sought psychological counseling for LI and she started taking a prescription anti-depressant. Her grades also dropped from “high honors” to “honors.” As the school year progressed, however, LI became more successful at interacting with her peers and participating in class.
During the summer recess preceding sixth grade, LI asked her mother, as she had the previous summer, to allow her to be home-schooled. LI also expressed her desire to attend The Community School (“TCS”), a private school in South Tamworth, New Hampshire, where her older sister had matriculated. Nevertheless, LI started the 2003-2004 school year at Cornish, where Mrs. I believed her daughter would benefit, in particular, from her assigned sixth grade teacher.
By mid-September, however, LI was “slacking off” in her academic work and regularly missing school, prompting a meeting between her teacher and Mrs. I. At this meeting, also attended by LI, Mrs. I noticed cuts or scratches on her daughter’s arms; the teacher offered that LI might have inflicted those wounds on herself during her “lengthy bathroom breaks” from class. According to the teacher, LI was also having continued trouble relating with her peers due to a “serious lack of awareness” of their social and emotional states, which bordered on “hostility.” The teacher added that she could not “reach” LI, who had refused to complete assignments and shown a “passive resistance to meeting learning goals.” Yet the teacher considered LI “a very bright young girl with strong language and math skills . . . capable of powerful insights in her reading and writing. . . .”
The teacher and Mrs. I came up with a “contract” that would have entitled LI to study more advanced topics in her areas of interest in November if she satisfactorily completed her assignments for October. As October approached, however, LI refused to sign the contract and stayed home from school on both September 30 and October 1. On October 1, following an argument with Mrs. I over one of LI’s academic assignments, LI deliberately ingested excessive quantities of one her prescription drugs and two over-the-counter medications in a suicide attempt.
LI spent the balance of the day in the emergency room at a nearby hospital and was discharged with instructions to remain out of school for two days under high safety precautions. The hospital social worker also directed Mr. and Mrs. I to “share with [LI] something that would change in her life, and produce a positive impact on her emotional functioning.” Based on LI’s comments to hospital personnel that she hated school, Mr. and Mrs. I told her that she would not have to return to Cornish Elementary and discussed enrollment at TCS as an alternative.
In the wake of her attempted suicide, LI met with a new counselor, who, suspecting that LI might suffer from Asperger’s Syndrome, referred her to Dr. Ellen Popenoe for neuropsychological testing.2 Mr. and Mrs. I conveyed this information, as well as the news of LI’s suicide attempt, to the district’s director of special services, Jim McDevitt. They added that LI would not return to Cornish Elementary “for the time being” and that they were looking at other options, including TCS. McDevitt explained the process for seeking reimbursement from the district for placing LI in a private school and also told the parents that the district planned to convene a pupil evaluation team for LI at the end of the month. At that meeting, the PET decided that LI should receive up to ten hours of tutoring outside of school each week pending completion of her neuropsychological testing.
The testing, finished by early November, further suggested that LI had Asperger’s Syndrome, as well as adjustment disorder with depressed mood.3 Popeneo, the neuropsychologist, observed that LI “experiences significant limitations in many areas of adaptive skills” and executive skills, “which likely contribute[s] to her behavioral and emotional difficulties.”4 These behavioral difficulties, particularly LI’s poor pragmatic language abilities and restricted range of social interests, supported a diagnosis of Asperger’s. Popeneo recommended that LI begin seeing both a social skills coach, who would help her develop social abilities and judgment, and a therapist familiar with Asperger’s, who would use a cognitive-behavioral approach.5
Popeneo also recommended that LI undergo a speech-language evaluation, which was completed in January 2004 by Amber Lambke, a speech-language pathologist. Lambke observed that LI suffered “significant social understanding deficits which impact her overall emotional and social well-being.” Like Popeneo, Lambke recommended that LI receive direct teaching of social skills.
In the meantime, McDevitt told Mrs. I that he would attempt to find LI a tutor in accordance with the PET’s decision. Mrs. I had not heard back from him by November 10, however, so she started home-schooling LI. Despite additional prodding by Mrs. I in November and December, the district never provided a tutor as ordered by the PET, nor explained its failure to do so. While LI preferred home-schooling to attending Cornish Elementary, Mrs. I was having trouble getting LI to complete her assignments, and her counselor believed that LI should resume formal schooling.
On January 5, 2004, LI began attending TCS. Although she was withdrawn and isolated at the outset, over time LI developed positive relationships with some of her peers. She also thrived academically, completing assignments at the seventh? and eighth-grade level with ease. TCS, however, provided LI with neither the direct teaching of social skills nor the cognitive behavioral therapy that had been recommended as treatment for her Asperger’s.
When the PET reconvened in early March, it accepted Popenoe’s conclusion that LI suffered from both Asperger’s and adjustment disorder with depressed mood. The PET also agreed that LI needed social skills and pragmatic language instruction. The PET, however, could not reach consensus on whether LI qualified as a “child with a disability” under the IDEA. The district’s representatives argued that LI’s condition, whether denominated “autism,” “emotional disturbance,” or “other health impairment,” 20 U.S.C. ? 1401(3)(A)(i), had not affected her academic performance “to a marked degree” or “over a long period of time,” which they deemed essential to IDEA eligibility. The district then issued a “prior written notice,” id. ? 1415(b)(3), announcing its refusal to offer special education services on the stated basis of “no significant adverse effect on education.” The district instead asked the PET to consider LI’s eligibility for services under the Rehabilitation Act, 29 U.S.C. ? 794 (2000).
At its next meeting, the PET identified LI as a “qualified individual with a disability” under the Rehabilitation Act, id. ? 794(a), and recommended an array of services. These included close supervision throughout the school day; instruction in “social pragmatics”; access to the district’s existing gifted and talented programming as well as additional programming provided through a consultant to be hired by the district; and placement in any elementary school within the district. The district also offered to supply a tutor to work with LI for three hours each day to ease her eventual transition back to the classroom.
Mr. and Mrs. I objected to this proposal as inadequate and unduly restrictive, given LI’s success in a classroom environment at TCS and her apprehension over returning to public school. They wanted LI to remain at TCS for the balance of the academic year with a view toward beginning her transition back to public school in September 2004, and notified the district that they intended to seek reimbursement under the IDEA for LI’s attendance at TCS. LI completed the 2003-2004 academic year at TCS, and stayed on for the 2004-2005 and 2005-2006 school years as well. While she has done well academically, she continues to experience “atypical” peer relationships and spent the summer of 2004 shunning her TCS classmates in favor of solitary pursuits. LI also generally refuses to go outdoors or to eat more than a severely limited variety of foods. Her current social worker believes that, without social skills coaching, LI is unlikely to master the flexible thinking, problem solving, teamwork, and communication abilities she will need for employment in the future.
After the final PET decision, Mr. and Mrs. I requested a due process hearing to challenge the district’s refusal to identify LI as a child with a disability under the IDEA. The hearing officer upheld the district’s decision that LI was ineligible for IDEA services. The hearing officer noted the parties’ agreement that LI had Asperger’s and a depressive disorder, making her “a troubled young woman,” but further observed that she was not entitled to IDEA benefits unless these disabilities “`adversely affect[ed]’ [her] educational performance.”
The hearing officer recognized that both the IDEA and Maine’s implementing regulations define “educational performance” to include more than just academic proficiency, but concluded that the IDEA does not call for services “to address social and emotional needs when there are no academic needs.” Accordingly, because LI “completes homework independently, is well behaved in class, is successful at test taking and successfully completes projects,” the hearing officer determined that “neither the [IDEA] nor the Maine Special Education Regulations require a school district to provide special education services to address what is essentially a mental health issue.”
In response, Mr. and Mrs. I commenced an action in the district court, which, as we have noted, referred the case to a magistrate judge.6 The magistrate judge determined that the hearing officer erred in treating LI’s lack of academic needs as dispositive of her IDEA eligibility when the correct standard, he believed, is whether a disability “manifest[s] itself in an adverse effect on the child’s ability to learn.” Nevertheless, the magistrate judge ruled that LI did not meet this standard because her condition did not adversely affect her achievements as measured by any of the criteria Maine uses to define “educational performance.” While the magistrate judge recognized that LI had fallen short of these benchmarks during the period in the fall of 2003 when she had repeatedly missed school and attempted suicide, he considered this episode too short-lived “to trigger eligibility for special-education services.”
The district court, however, rejected the magistrate judge’s recommended decision, concluding that LI’s “condition did adversely affect her educational performance as Maine defines that term and that the events of the fall of 2003 cannot be isolated from [her] underlying condition.” 416 F.Supp.2d 147, 152 (D.Me.2006). The district court determined that LI’s Asperger’s had exerted an adverse effect on her educational performance as measured by state criteria, most significantly in the areas of socialization and communication. The district court also disagreed with the view that any downturn in LI’s educational performance was too fleeting to constitute an “adverse effect.” Reasoning that neither the Maine regulations defining the disabilities listed in ? 1401(3)(A)(i) nor their federal counterparts used any restrictive modifier in conjunction with the term “adversely affects,” the district court ruled that “any negative effect should be sufficient” to constitute a disability under the IDEA. 416 F.Supp.2d at 160 (emphasis added).
Turning to the second prong of the IDEA’s eligibility standard, 20 U.S.C. ? 1401(3)(A)(ii), the district court concluded that LI needed special education and related services by reason of her disability. First, the district court found that the PET had agreed to provide LI with a number of accommodations that fit the definition of “special education” under both the IDEA and Maine law, including one-on-one tutoring and instruction in social pragmatics. Second, observing that “the PET, the experts, the School District and the parents all initially believed that [LI] `needed’ the identified services,” the district court decided to “hold the parties to their original understandings” and therefore treated “need” as an uncontested issue. 416 F.Supp.2d at 167. Based on its determination that LI satisfied both elements of the IDEA eligibility test, the district court ordered the district “to convene a PET meeting . . . to develop an IEP for [LI] that meets her unique needs as a student with Asperger’s Syndrome and a depressive disorder.”7 Id. at 168.
The district court also considered the parents’ requests for additional relief: reimbursement of their expenses in unilaterally placing LI at TCS, and compensatory education to make up for the district’s failure to identify her as eligible under the IDEA. Though the district court found that Mr. and Mrs. I had given the requisite notice of the unilateral placement under Maine law, the court also ruled that their decision to enroll LI at TCS was not “`reasonably calculated to enable [her] to receive educational benefits'” so as to entitle them to reimbursement. 416 F.Supp.2d at 172 (quoting Florence County Sch. Dist. Four v. Carter ex rel. Carter, 510 U.S. 7, 11, 114 S.Ct. 361, 126 L.Ed.2d 284 (1993) (further internal quotation marks omitted by district court)). Finally, reasoning that LI’s “IEP will necessarily take into account the effect of the school district’s failure to identify and offer [LI] special education services earlier,” the district court did not separately grant the parents’ request for compensatory education. Id. at 173.
The district challenges the district court’s conclusion that LI qualifies as a “child with a disability” under the IDEA. While we have never expressly set forth our standard of review for a district court’s decision on IDEA eligibility, we have treated “ultimate determinations in cases under the Act” as mixed questions of fact and law. Roland M. v. Concord Sch. Comm., 910 F.2d 983, 990 (1st Cir. 1990); see also Ms. M ex rel. K.M. v. Portland Sch. Comm., 360 F.3d 267, 272 (1st Cir.2004); Kathleen H. v. Mass. Dep’t of Educ., 154 F.3d 8, 13 (1st Cir. 1998). We agree with the parties that whether a student qualifies as “a child with a disability” under ? 1401(3) also poses a mixed legal and factual inquiry. See J.D. ex rel. J.D. v. Pawlet Sch. Dist., 224 F.3d 60, 64 (2d Cir.2000); Yankton Sch. Dist. v. Schramm, 93 F.3d 1369, 1374 (8th Cir.1996). Mixed questions generally “fall along a degree-of-deference continuum, ranging from non-deferential plenary review for law-dominated questions, to deferential review for fact-dominated questions.” In re PolyMedica Corp. Sec. Litig., 432 F.3d 1, 4 (1st Cir. 2005). But we need not decide at the moment where along the continuum the question of IDEA eligibility falls, as the parties agree that we should review the question for clear error.
The district maintains, however, that the district court arrived at its conclusion that LI is a “child with a disability” only through a series of legal errors. First, the district argues that the district court misread the terms “adversely affects” and “educational performance” as they appear in the regulatory definitions of the disabilities attributed to LI, improperly extending the breadth of ? 1401(3)(A)(i). Second, the district claims that the district court similarly misinterpreted the term “special education” as it appears in ? 1401(3)(A)(ii), the second prong of the test for IDEA eligibility. The district also challenges the determination that it effectively waived the opportunity to dispute LI’s need for special education. We review these rulings of law de novo. Greenland, 358 F.3d at 156; Manchester Sch. Dist. v. Crisman, 306 F.3d 1, 9 (1st Cir.2002).
Though the IDEA “establishes a basic floor of education” for children with disabilities, guaranteeing them “[a] free appropriate public education,” 20 U.S.C. ? 1412(a)(1)(A), it does not displace the states from their traditional role in setting their own educational policy. Burlington, 736 F.2d at 788-89; see also J.D., 224 F.3d at 65; Daniel R.R. v. State Bd. of Educ., 874 F.2d 1036, 1044 (5th Cir.1989). Each state thus remains free to calibrate its own educational standards, provided it does not set them below the minimum level prescribed by the statute. Roland M., 910 F.2d at 987; Burlington, 736 F.2d at 788-89.
As we have seen, the right to special education and related services under the IDEA extends to children “with” one or more of a variety of disabilities, 20 U.S.C. ? 1401(3)(A)(i), “who, by reason thereof, need[ ] special education and related services.” Id. ? 1401(3)(A)(ii). The IDEA does not itself define any of the qualifying disabilities listed in ? 1401(3)(A)(i), though the Department of Education has issued a regulation fleshing them out. 34 C.F.R. ? 300.8(c). The regulatory definitions, with one exception not relevant here, state, among other requirements, that each condition must “adversely affect[ ] a child’s educational performance.” Id. ? 300.8(c)(1)-(c)(13). In keeping with the IDEA’s respect for state policy judgments, however, the regulation does not expand upon this phrase, “leaving it to each State to give substance to these terms.” J.D., 224 F.3d at 65; see also Greenland Sch. Dist. v. Amy N., No. 02-136-JD, 2003 WL 1343023, at *8 (D.N.H. Mar. 19, 2003), aff’d on other grounds, 358 F.3d 150 (1st Cir. 2005).
It is here that the district’s argument as to the proper scope of ? 1401(3)(A) begins to encounter difficulty. While Maine’s Department of Education has promulgated its own regulation defining the disabilities recognized under the IDEA, those definitions simply ape their federal counterparts, including the requirement that a disability “adversely affect[ ] the student’s educational performance.” 05-071-101 Me.Code. R. ?? 3.2-3.14 (2006). The regulation, like its federal cousin, also does not further elaborate on this phrase, although Maine has adopted its own definition of “educational performance” for IDEA purposes:
The term “educational performance” includes academic areas (reading, math, communication, etc.), non-academic areas (daily life activities, mobility, etc.), extracurricular activities, progress in meeting goals established for the general curriculum, and performance on State-wide and local assessments.
Id. ? 2.7. Despite this expansive notion of educational performance, and in the absence of any regulatory guidance as to the term “adversely affects,” the district asks us to hold that a child meets the first criterion of IDEA eligibility in Maine “only if the student’s condition imposes a significant negative impact on the child’s educational performance . . . limited to those areas of performance actually being measured and assessed by the local unit, in accordance with law.” We decline to do so.
At the outset, Maine does not look only at “areas of performance actually being measured and assessed by the local unit” when determining whether a child has a disability under the IDEA. That much is clear from the regulatory definition of “educational performance” itself, which counts “performance on state-wide and local assessments” as just one of a number of different indicators embraced by the concept. As the district points out, the term “general curriculum,” which also appears in the definition of educational performance, has a narrower meaning under the regulations, i.e., “the school administrative unit’s local curriculum for grades K-12 which incorporate the content standards and performance indicators of the Learning Results.”8 05-071-101 Me.Code. R. ? 2.11. Based on this definition, the district argues that “educational performance” encompasses only those “performance indicators” measured as part of the local curriculum.9 Even if the district’s reading of “general curriculum” is correct, however, the fact remains that a student’s progress in that regard comprises but one of the aspects of “educational performance” as defined by the regulation. More far-ranging measurements, such as “academic areas” and “non-academic areas,” are also included.
As the magistrate judge and the district court observed, Maine’s broad definition of “educational performance” squares with the broad purpose behind the IDEA: “to ensure that all children with disabilities have available to them a free and appropriate public education that emphasizes special education and related services designed to meet their unique needs and prepare them for further education, employment, and independent living.” 20 U.S.C. ? 1400(d)(1)(A) (emphases added). We have likewise held that the IDEA entitles qualifying children to services that “target `all of [their] special needs,’ whether they be academic, physical, emotional, or social.” Lenn v. Portland Sch. Comm., 998 F.2d 1083, 1089 (1st Cir.1993) (quoting Burlington, 736 F.2d at 788). It is true that we have also stated that IDEA services need not address “problems truly `distinct’ from learning problems.” Gonzalez v. P.R. Dep’t of Educ., 254 F.3d 350, 352 (1st Cir.2001); see also Rome Sch. Comm. v. Mrs. B., 247 F.3d 29, 33 n.3 (1st Cir.2001) (noting that, in determining adequacy of IEP for emotionally disturbed boy, “[t]he question is whether [his] behavioral disturbances interfered with the child’s ability to learn”). But it does not follow, as the hearing officer wrongly concluded, that a child without “academic needs” is per se ineligible for IDEA benefits, especially when the state has conditioned eligibility on a standard that explicitly takes “non-academic areas” into account. See Weixel v. Bd. of Educ., 287 F.3d 138, 150 (2d Cir.2002) (“IDEA’s coverage is not limited to students with `learning disabilities’. . . .”). In other words, as the district admits, “educational performance in Maine is more than just academics.”
In light of Maine’s broad notion of “educational performance” as the standard of IDEA eligibility, we see no basis for restricting that standard to “areas of performance actually being measured and assessed by the local unit.” Indeed, “there is nothing in IDEA or its legislative history that supports the conclusion that . . . `educational performance’ is limited only to performance that is graded.” See Robert A. Garda, Jr., Untangling Eligibility Requirements Under the Individuals with Disabilities Education Act, 69 Mo. L.Rev. 441, 471 (2003). To be sure, some states have adopted more circumscribed criteria for identifying children with disabilities under the IDEA, requiring, for example, that a student perform poorly in a specific area of “basic skills.” See J.D., 224 F.3d at 66 (discussing prior version of 22-000-06 Vt. Code Reg. ?? 2362(a)(2), (f) (2006)). Maine, however, has chosen not to do so.10 We therefore decline the district’s invitation to reformulate state educational policy by narrowing the indicia of educational performance used as the test for IDEA eligibility under Maine law. The district court properly articulated this standard as “whether [LI’s] condition adversely affected her performance in any of the educational areas Maine has identified.” 416 F.Supp.2d at 159 (footnote omitted).
The district also argues that the district court misconstrued the “adversely affects” component of the test to include disabilities with “any adverse effect on educational performance, however slight. . . .” Id. at 160. The correct formulation, the district urges, requires “some significant impact on educational performance.” In rejecting this proposal, the district court reasoned that the phrase “adversely affects,” as it appears in the relevant regulations, “has no qualifier such as `substantial,’ `significant,’ or `marked,'” and declined to infer such a limitation “from Maine’s regulatory silence.” Id. We agree with this interpretation of the “adversely affects” standard.
Though the district marshals a number of arguments in support of its contrary position, they all sound a common theme: that an unlimited definition of “adversely affects” will qualify every child with one of the listed disabilities?no matter how minor?for IDEA benefits. This contention, however, overlooks the structure of the IDEA’s eligibility standard, which requires not only that a child have one of the listed conditions, ? 1401(3)(A)(i), but also that, “by reason thereof,” the child “needs special education and related services,” id. ? 1401(3)(A)(ii). So a finding that a child meets the first criterion because his or her disability adversely affects educational performance?to whatever degree?does not itself entitle the child to special education and related services under the IDEA. See Mark C. Weber, Special Education Law and Litigation Treatise ? 2.2(1), at 2:4 (2d ed.2002); Garda, supra, at 490-91. The child must also need special education and related services by reason of the disability.11
In fact, an adverse effect on educational performance, standing alone, does not even satisfy the first prong of the eligibility test. The child’s condition must also possess the additional characteristics required by the regulatory definitions of each of the disabilities enumerated in ? 1401(3)(A)(i). See 34 C.F.R. ?? 300.8(c)(1)-(c)(13); 05-071-101 Me.Code. R. ?? 3.2-3.14. For example, to meet the first part of the eligibility standard on the basis of autism, a child must have “ a developmental disability  significantly affecting  verbal and  nonverbal communication and  social interaction,  generally evident before age three,  that adversely affects a child’s educational performance.” 34 C.F.R. ? 300.8(c)(1)(i); 05-071-101 Me.Code. R. ? 3.2. Thus, the “adversely affects educational performance” requirement serves as but one of a list of factors that must be present for a child’s condition to qualify as a disability under ? 1401(3)(A)(i)?and, to receive IDEA benefits, the child must also need special education and related services by reason of the disability under ? 1401(3)(A)(ii). The district court’s interpretation of “adversely affects,” then, is unlikely to loose the torrent of IDEA claims forecast by the district and its amici.
The district’s specific arguments fare no better. The district contends that ? 1401(3)(a)(i) fails to put the states on notice that, as a condition of accepting federal money under the IDEA, they are required to provide benefits to children whose conditions have merely an “adverse effect” on their educational performance. It is true that “when Congress attaches conditions to a State’s acceptance of federal funds” pursuant to its Spending Clause authority, “the conditions must be set out unambiguously” so that each state can intelligently decide whether to take the money and its accompanying obligations. Arlington Cent. Sch. Dist. Bd. of Ed. v. Murphy, ___ U.S. ___, ___, 126 S.Ct. 2455, 2459, 165 L.Ed.2d 526 (2006) (internal quotation marks omitted). Based on this principle, the Supreme Court has held that whether the IDEA imposes a particular obligation on the states depends, at the outset, on whether the IDEA “furnishes clear notice regarding the liability at issue. . . .” Id.
The principal place to look for such notice, of course, is the text of the IDEA itself. Id. The district asserts that the language of ? 1401(3)(A)(i) fails to clarify that a state’s duty to provide IDEA benefits extends to children with disabilities having only an adverse effect on educational performance. In fact, the district argues, the statute?through its use of the term “disability”?limits that duty to children whose conditions “significantly impact educational performance.” We disagree.
To properly understand “disability” as it appears in the IDEA, we do not, as the district implores, resort to dictionary definitions of the word “disable,” but to ? 1401(3)(A)(i), which functions as the first part of the statutory definition of “child with a disability.” Section 1401(3)(A)(i), as the district court observed, does not include the qualifying language urged upon us by the district, but simply defines “child with a disability” as a child “with” one of a number of specific conditions.12
The district also directs us to the more restrictive meaning of the term “disability” under Title II of the Americans with Disabilities Act and the Rehabilitation Act. Because the IDEA contains its own definition of the term, however, its appearance in other acts of Congress is of little moment. See United States v. Meade, 175 F.3d 215, 220-21 (1st Cir.1999). Putting aside the difference between the legislative goals of the IDEA and these other acts, then, the IDEA simply defines “disability” differently than they do. Compare 20 U.S.C. ? 1401(3)(A) with 29 U.S.C. ? 705(9)(B) and 42 U.S.C. ? 12102(2)(A) (defining “disability” as “physical or mental impairment that substantially limits one or more major life activities”). This clear disparity in text puts the district’s suggestion that we look to those other acts in construing the term “disability” here on par with comparing “plums and pomegranates.” Meade, 175 F.3d at 221.
Given the express definition of “disability” set forth in ? 1401(3)(A)(i), we need look no further to conclude that the statute sufficiently articulates the first prong of the standard for IDEA eligibility and, in so doing, adequately informs the states of the extent of their obligations. Murphy, 126 S.Ct. at 2463. The district and its amici nevertheless argue that this standard, as interpreted by the district court, flies in the face of congressional admonishments against identifying too many students as “children with disabilities” under the IDEA. It is true that, in amending the Act in 1997, Congress voiced concern about “over identifying children as disabled when they may not be truly disabled . . . particularly in urban schools with high proportions of minority students . . . .” H.R.Rep. No. 105-95, at 89 (1997), reprinted in 1997 U.S.C.C.A.N. 78, 86. To remedy this problem, Congress changed the formula for calculating the funds due each state under the IDEA from one “based on the number of children with disabilities to a formula based on census and poverty. . . .” Id. at 88, 1997 U.S.C.C.A.N. at 85.
Notably, though, Congress thought this shift?rather than any alteration to the eligibility criteria?sufficient to address the over-identification problem. Id. at 89, 1997 U.S.C.C.A.N. at 87. Congress specifically stated, in fact, that the change to the funding formula “should in no way be construed to modify the obligation of educational agencies to identify and serve students with disabilities.” Id. at 88, 1997 U.S.C.C.A.N. at 85. Congress eschewed any change to the eligibility standard not only in 1997, but also in 2004, when it amended the IDEA again. Individuals with Disabilities Education Improvement Act of 2004, Pub.L. No. 108-446, ? 602(3)(A), 118 Stat. 2647, 2652, codified at 20 U.S.C. ? 1401(3)(A). The Department of Education similarly declined, by and large, to tinker with its definitions of the ? 1401(3)(A)(i) disabilities when it issued regulations in response to the amended Act. 71 Fed.Reg. 46,540, 46,549-46,551 (Aug. 14, 2006). Thus, although the district and its amici argue that an over-identification problem persists, we cannot tighten the standard for IDEA eligibility when Congress itself has chosen not to do so.13
The legislative history, then, only strengthens our conviction that ? 1401(3)(A)(i), as construed by the district court, does not offend the Spending Clause by springing hidden liabilities upon participating states. Furthermore, as the district acknowledges, states deciding whether to enter into the IDEA bargain also have the benefit of the federal regulation defining the disabilities set forth in ? 1401(3)(A)(i). Those definitions, again, specifically require that each disability (save one) “adversely affect[ ] a child’s educational performance.” 34 C.F.R. ?? 300.8(c)(1)-(c)(13). They do not contain the limiting language urged by the district, i.e., “significantly impacts educational performance.”
We reject the district’s argument that such a limitation lurks in the term “adversely,” which the district equates with “calamitously” or “perniciously” on the authority of an unabridged dictionary. We think it considerably more likely that federal regulators used “adverse” in its ordinary sense, namely “against.” Black’s Law Dictionary 58 (8th ed.2004); see also Webster’s Third New International Dictionary of the English Language (Unabridged) 31 (1993) (giving primary definition of “adverse” as “acting against or in a contrary direction”).14 In this way, the regulation sensibly demands that a disability cannot qualify a child for IDEA benefits unless it has a negative effect on educational performance; no effect, or a positive one, will not do.15 The regulation does not, however, put any quantitative limit, “significant” or otherwise, on the disability.
Maine’s regulation, cribbed from 34 C.F.R. ? 300.8, also requires no particular degree of impact on educational performance. 05-071-101 Me.Code. R. ?? 3.2-3.14. This fact alone distinguishes this case from the decisions of other courts, cited by the district, which derived a higher standard from state law. See J.D., 224 F.3d at 66-67; Gregory M. ex rel. Ernest M. v. State Bd. of Educ., 891 F.Supp. 695, 702 (D.Conn.1995); Doe ex rel. Doe v. Bd. of Educ., 753 F.Supp. 65, 70 & n. 9 (D.Conn. 1990).16 In J.D., for example, the Second Circuit considered a Vermont regulation that defined “adverse effect of the disability on educational performance” to require a determination “that the student is functioning significantly below expected age or grade norms, in one or more of the basic skills.” 224 F.3d at 66 (internal quotation marks omitted). This provision further required that the “determination of adverse effect, usually defined as 1.0 standard deviation or its equivalent, shall be documented and supported by two or more measures of school performance,” which were themselves specified by the regulation. Id. Based on this standard, the Second Circuit concluded that the child did not qualify for IDEA benefits because he was “unable to identify at least two school performance measures that point to an adverse effect,” despite his emotional-behavioral disability. Id. at 67.
The Second Circuit reached its decision in J.D., then, by applying the highly specific definition of “adversely affects educational performance” set forth in state law, not by imposing its own gloss on that language, as the district invites us to do here. For the reasons we have stated, we decline that invitation. States wishing to put meat on the bones of the “adversely affects” standard are free to do so?provided, of course, they do not transgress the “floor” of substantive protection set by the IDEA.17 See generally Burlington, 736 F.2d at 788-89. On its own, however, the federal regulation does not contain the “significant impact” requirement the district desires, and we cannot put it there. The district court correctly ruled that any negative impact, regardless of degree, qualifies as an “adverse effect” under the relevant federal and state regulations defining the disabilities listed in ? 1401(3)(A)(i).
Because the district court applied the right standard, we review its determination that LI has one of the disabilities included in ? 1401(3)(A)(i) “for clear error on the record as a whole.” Ms. M., 360 F.3d at 272. We find none. As the hearing officer noted, the parties agree that LI suffers from Asperger’s, manifested in her poor pragmatic language skills and social understanding difficulties, as well as from a depressive disorder brought on by the stress of managing these problems; indeed, the district has never questioned the opinions of LI’s neuropsychologist and speech therapist in this regard. The parties disagree, however, on whether these conditions have adversely affected LI’s educational performance in light of her strong grades, generally nondisruptive classroom behavior, and what the district court called her “undisputed intellectual ability.” 416 F.Supp.2d at 161. In a lengthy written opinion, the district court tackled this issue head on, ultimately finding that, despite LI’s above-average academic performance, “many of [her] social and communication deficits, including her isolation, inflexibility, and self-mutilation during schooltime, are precisely in the content areas and skills that Maine mandates educationally.” Id. at 163. This finding, the district court reasoned, compelled the conclusion that LI’s disability had exerted an adverse effect on her educational performance under the governing standard.
Much of the district’s challenge to this outcome relies on its contention that the district court applied the “adversely affects educational performance” test too leniently, which we have already rejected. A few of the district’s supplemental points, however, merit additional discussion. First, the district argues that the district court mistakenly gauged LI’s educational performance on the basis of selected “performance indicators,” see 05-071-131 Me. Code R. ?? 1-8, that Maine has developed to measure students’ proficiency in various “content standard subject areas.” Me. Rev.Stat. Ann. tit. 20-A, ? 6209. This was error, the district asserts, because Maine does not mandate the actual use of the performance indicators by local school districts, but has simply instructed them to develop their own “local assessment systems.” Id. ? 6202-A. While we have our doubts about this proposition, see 05-071-127 Me.Code R. ? 4.02 (requiring each district to “implement a local assessment system as the measure of student progress on achievement of the content standards of the system of Learning Results established in” 05-071-131 Me.Code R. ?? 1-8), the district court did not assess LI’s educational performance solely by reference to the performance indicators. Our review of the record convinces us that, even if the district court erred by also using the performance indicators to measure LI’s educational performance, the error did not affect the outcome of its analysis.
In particular, the district court found that LI had difficulty with “communication,” an area of “educational performance” specifically incorporated in Maine’s definition of that term for IDEA purposes. 416 F.Supp.2d at 162 & n. 8 (quoting 05-071-101 Me.Code R. ? 2.7). The district disputes this finding, emphasizing certain aspects of both her educators’ observations and the results of the testing conducted by Popeneo and Lambke. The district court, however, focused on other aspects of those materials, such as the educators’ reports of LI’s “distancing” herself from her teachers and peers and, most significantly, the experts’ express conclusions that LI had “poor pragmatic language skills” and “significant social understanding deficits.” 416 F.Supp.2d at 161-63. The district court was by no means required to second-guess these conclusions, especially after they had been unreservedly accepted by both the districts’ representatives at the PET and the hearing officer. Though the record of the administrative hearing might permit a different view, the district court did not commit clear error in finding that LI’s Asperger’s has impaired her ability to communicate.
Moreover, the district court’s ruling that LI had demonstrated an adverse effect on her educational performance did not rest solely on her deficits in communication, but also on other difficulties implicating “the career preparation component of the Maine general curriculum.” Id. at 162. The district does not question that “career preparation”?which comprises one of the “content standards” dictated by statute, Me.Rev.Stat. Ann. tit. 20-A, ? 6209(2)(A), rather than one of the “performance indicators” established by regulation?is irrelevant to the “educational performance” inquiry for purposes of the IDEA. Indeed, the IDEA exists, in part, to ensure children with disabilities receive an education preparing them for employment. 20 U.S.C. ? 1400(d)(1)(A). Nor does the district question the lower court’s specific finding, consistent with the opinion of LI’s current social worker, that a number of LI’s symptoms have hindered her in this area. 416 F.Supp.2d at 162. This finding was itself an adequate basis for the district’s court’s conclusion that LI’s educational performance has suffered, even if, as the district argues, her condition has not impacted her communication skills.
Second, the district argues that the impact of LI’s condition on her educational performance, which it sees as limited to her suicide attempt and the events immediately preceding it in the fall of 2003, was not sustained enough to constitute an adverse effect. Though the magistrate judge accepted this point of view, the district court disagreed, treating the suicide attempt as simply the darkest point in the spectrum of LI’s educational difficulties. There is ample support for this approach in the record. The signs of LI’s Asperger’s revealed themselves in the fourth grade, when she began experiencing difficulty with peer relationships, and first translated into a measurable impact on her schoolwork in the fifth and sixth grades, when her grades declined. More importantly, there is every indication that these symptoms will persist, to one degree or another: they have not completely subsided since LI’s enrollment at TCS, and both Popeneo and LI’s current social worker believe that continued intervention is essential to LI’s long-term success. In light of this evidence, the district’s argument that LI’s suicide attempt did not adversely affect her educational performance is beside the point; as Popeneo explained, the suicide attempt was but a manifestation of LI’s Asperger’s and associated depression. The district court properly treated these disorders, rather than the suicide attempt, as the relevant condition for assessing the impact of LI’s disability upon her educational performance.
Third, the district charges that the district court “committed legal error” by ruling that LI met the first prong of the standard for IDEA eligibility without assigning her one of the disabilities listed in ? 1401(3)(A)(i). As we have pointed out, that a condition “adversely affects a child’s educational performance” functions as just one of the essential elements of each of the qualifying disabilities as defined in the regulation, so a determination that a child has one of those disabilities would ordinarily demand a showing as to each of those elements. Here, however, the district court specifically noted that, while the parties were at odds as to whether LI’s condition adversely affected her educational performance, they were in agreement that her condition otherwise “fit[ ] within those enumerated” by ? 1401(3)(A)(i). 416 F.Supp.2d at 156. The district has not questioned this observation. Because the district did not dispute below whether LI satisfied the additional criteria of any of the relevant disability categories, its argument that the district court should have chosen from among those categories is forfeit. See, e.g., States Res. Corp. v. Arch. Team, Inc., 433 F.3d 73, 85 (1st Cir.2005) (“This circuit religiously follows the rule that issues not presented to the district court cannot be raised on appeal.”) (internal quotation marks omitted). There was no error in the district court’s ? 1401(3)(A)(i) analysis.
The district also argues that the district court misapplied ? 1401(3)(A)(ii), which requires that a child “need[ ] special education and related services” as a result of his or her disability in order to qualify for them under the IDEA. The district asserts two errors: first, the district court used the wrong definition of “special education,” and, second, it found that the district had waived any argument that LI does not “need” special education based on the position it took before the PET and the hearing officer. We believe the district court correctly defined “special education” under ? 1401(3)(A)(ii). We do not decide, however, whether the district court properly treated the “need” issue as waived, because the district has not adequately explained to us why LI does not need special education, even under its view of the proper standard for making that determination.
The IDEA defines “special education,” in relevant part, as “specially designed instruction, at no cost to parents, to meet the unique needs of a child with a disability . . . .” 20 U.S.C. ? 1401(29). A federal regulation, promulgated by the Department of Education, elaborates:
Specially designed instruction means adapting, as appropriate to the needs of an eligible child . . . , the content, methodology, or delivery of instruction?
(i) To address the unique needs of the child that result from the child’s disability; and
(ii) To ensure access of the child to the general curriculum, so that the child can meet the educational standards within the jurisdiction of the public agency that apply to all children.
34 C.F.R. ? 300.39(b)(3) (2006).18 As the district court noted, Maine law also contains its own definition of “special education”: “classroom, home, hospital, institutional or other instruction; educational diagnosis and evaluation; transportation and other supportive assistance, services, activities, or programs, as defined by the commissioner [of education], required by exceptional students.”19 Me.Rev.Stat. Ann. tit. 20-A, ? 7001(5) (1993).
The district court ruled that a number of the interventions recommended by Popeneo and Lambke, and included in the services offered by the PET under the Rehabilitation Act, were “special education” within the meaning of federal law as well as “under Maine’s broader definition.” 416 F.Supp.2d at 166. In challenging this conclusion, the district principally argues that the district court misinterpreted Maine law to exceed IDEA requirements as to the definition of “special education.” We have little trouble with the district court’s interpretation, given the expansive language of Me.Rev.Stat. Ann. tit. 20-A, ? 7001(5), but, in any event, that provision was not essential to the district court’s view that LI needs special education. The district court also specifically ruled that certain of the services recommended for LI constituted “special education” as defined by federal law. 416 F.Supp.2d at 166.
Most significantly, the district court reasoned that “extra instructional offerings such as social-skills and pragmatic-language instruction are `specially designed instruction’ to ensure [LI’s] `access . . . to the general curriculum.'” Id. (quoting 34 C.F.R. ? 300.39(b)(3)). The district protests that its proffered “social pragmatics instruction,” which “was aimed more at counseling LI at how she could better interact with others” than at traditional “speech services,” qualifies as a “related service,” not “special education,” under the IDEA. The district has it backwards, however. While “speech-language pathology services” comprise a category of “related services,” 20 U.S.C. ? 1401(26)(A), directly teaching social skills and pragmatic language to LI amounts to adapting the content of the usual instruction to address her unique needs and to ensure that she meets state educational standards, viz., those defining educational performance to include “communication” and requiring progress in “career preparation.”20 See Part III.A.2, supra. The district court did not err in ruling that the services recommended for LI by her neuropsychologist and speech-language pathologist, and agreed to by the PET as part of its Rehabilitation Act plan, are “special education.”
The district also challenges the district court’s resolution of whether LI “needs” the special education in question. The district court made no finding on this point, electing to “hold the parties to their original understandings” that “`[n]eed’ is not a contested issue.” 416 F.Supp.2d at 168. In support of this course of action, the district court noted that “the factual record on need is poorly developed” because “the PET meetings proceeded on the basis that everyone agreed that LI `needed’ and should be afforded what the experts recommended for her” and because the district gave no indication that it disputed LI’s need for special education in either the prior written notice heralding its denial of IDEA benefits or its brief filed in advance of the due process hearing. Id. at 167. Accordingly, the district court reasoned that “[w]hether or not waiver is the correct term,” it had no sensible option but to conclude that LI “`needed’ the identified services” as the parties “all initially believed.” Id.
The district insists that it preserved the issue of LI’s need for special education by presenting argument and evidence on that score at the due process hearing. We agree with the district that it adduced some evidence at the hearing, in the form of testimony from McDevitt, as to LI’s need for special education. Specifically, in response to a question from counsel for the district on whether he believed that LI “requires special education, specialized instruction, take your pick, to do acceptably well in school,” McDevitt replied, “No, I don’t.” He went on to state his view that LI was “having a successful time” at TCS, even without “special services.” McDevitt then explained why the district’s offer of Rehabilitation Act services should not be construed as its opinion that “these interventions are necessary interventions for [LI] to participate meaningfully in public school,” i.e., because they did not constitute special education, but accommodations intended to make LI and her parents feel comfortable about her return to Cornish.
We need not decide whether this presentation came too late to raise the issue of LI’s need for special education, as the district court ruled, because the district does not explain why LI does not need special education under the standard it urges us to follow in making that determination. The district contends that “whether a child needs special education for IDEA eligibility should depend on whether that child requires special education to benefit in those areas of educational performance that are adversely affected,” but does not argue that LI does not pass that test. Instead, the district argues, based on McDevitt’s testimony and LI’s performance at TCS, that she does not need special education “to benefit from school” or “to do well in school.”
But whether a child requires special education “to do well in school,” or even “to benefit from school,” presents a different question from whether the child requires special education “to benefit in those areas of educational performance that are adversely affected by her disability.” The former inquiry considers the effect of special education on the child’s overall achievement in school, while the latter focuses on the effect of special education on the components of that achievement hampered by the child’s disability. See Garda, supra, at 498-99 (positing “which of the child’s performance areas must need special education?” as a crucial question in developing the test for IDEA eligibility under ? 1401(3)(A)(ii)). Indeed, a child may “do well in school” without special education, accumulating a high grade point average, but may nevertheless perform below acceptable levels in other areas, such as behavior. See, e.g., In re Monrovia Unified Sch. Dist., 38 Inds. with Disabilities Educ. L. Reptr. (LRP Publ’ns) 84, at 342-43 (Cal. State Educ. Agency Nov. 27, 2002) (finding student to “require special education to address social, behavioral, and written expression needs” despite “good academic work”). The questions of whether such a child “needs special education” under a proper interpretation of ? 1401(3)(A)(ii)?and how to articulate that interpretation in the first instance? have generated a cacophony of different answers. See Garda, supra, at 491-507 (surveying divergent authority).
We do not attempt to compose the correct standard of “need” here. We simply note the significant variance between the standard the district urges us to adopt and the standard it argues has been satisfied. McDevitt’s testimony may have supported a finding that LI does not require special education “to do well in school,” had the district court not ruled that the issue had been waived. But the district does not explain how such a finding would support the conclusion that LI does not “need special education” under the IDEA and, in fact, argues that the proper inquiry incorporates a substantially different standard, i.e., whether LI “requires special education to benefit in those areas of educational performance that are adversely affected.” Conversely, the district does not explain how the evidence received at the due process hearing falls short of that standard. The district has therefore failed to show that the district court’s treatment of the “need” issue as settled had any effect on its ultimate conclusion that LI qualified for IDEA benefits. Even if the district court erred in finding the district had waived the “need” argument, then, the error was harmless. See Hampton Sch. Dist. v. Dobrowolski, 976 F.2d 48, 54 (1st Cir.1992) (treating district court’s mistaken ruling that parents waived procedural objections to development of IEPs as harmless error where alleged procedural flaws did not meaningfully affect substance of IEPs).
The district has not directed us to any error undermining the district court’s determination that LI meets the second prong of the standard for IDEA eligibility, 20 U.S.C. ? 1401(3)(A)(ii).21 Having found that the district court’s ruling as to the first prong also holds up, Part III.A, supra, we affirm the district court’s decision that LI is eligible for services under the IDEA.
In their cross-appeal, the parents challenge the adequacy of the relief given as a remedy for the district’s failure to provide LI with IDEA benefits. First, they argue that the district court wrongfully denied them reimbursement for the costs of enrolling LI at TCS on the ground that it is not an educationally appropriate placement. Second, they argue that the district court should have explicitly ordered the district to provide LI with a compensatory education as a remedy for its denial of IDEA services, rather than leaving that matter for the PET to decide in the first instance. We address these contentions in turn.
The IDEA authorizes a district court reviewing the outcome of a due process hearing to “grant such relief as the court determines is appropriate.” 20 U.S.C. ? 1415(i)(2)(C)(iii). The Supreme Court has read this provision, as it appeared in the predecessor to the IDEA, as empowering a court “to order school authorities to reimburse parents for their expenditures on private special education for a child if the court ultimately determines that such placement, rather than a proposed IEP, is proper under the Act.” Burlington, 471 U.S. at 369, 105 S.Ct. 1996. In accordance with this holding, parents “are entitled to reimbursement only if a federal court concludes both that the public placement violated IDEA and that the private school placement was proper under the Act.” Florence County, 510 U.S. at 15, 114 S.Ct. 361.
We have identified reimbursement under the IDEA as “`a matter of equitable relief, committed to the sound discretion of the trial court.'” Roland M., 910 F.2d at 999 (quoting Burlington, 736 F.2d at 801). Ordinarily, we review a district court’s decision to award or withhold equitable relief for an abuse of that discretion. See, e.g., Valentin-Almeyda v. Municipality of Aguadilla, 447 F.3d 85, 104 (1st Cir.2006). But, as the Court made clear in Florence County, the right to reimbursement of private special education expenses depends in the first instance on whether the private school placement was “proper.” We consider this threshold inquiry, like other conclusions demanded by the IDEA, as a mixed question of fact and law. Part III, supra. As we did with the question of LI’s eligibility for IDEA benefits, we will review the propriety of her enrollment at TCS for clear error based on the parties’ accession to that standard. Id.
Mr. and Mrs. I, however, claim that the district court applied the wrong test in deciding that TCS was not a “proper” placement for LI, a question we review de novo. Id. Again, the district court ruled that the parents’ decision to enroll LI at TCS was not “`reasonably calculated to enable [her] to receive educational benefits'” so as to entitle them to reimbursement. 416 F.Supp.2d at 172 (quoting Florence County, 510 U.S. at 11, 114 S.Ct. 361 (further internal quotation marks omitted by district court)). Despite the district court’s recitation of this test, the parents insist that it actually applied what they describe as a more restrictive standard, derived from the Sixth Circuit’s decision in Berger v. Medina City Sch. Dist., 348 F.3d 513 (6th Cir.2003). The parents contend that this standard, which disallows reimbursement for a unilateral private placement that “does not, at a minimum, provide some element of special education services in which the public school placement was deficient,” id. at 523, is at odds with Florence County.22
Like the district court, we do not detect any tension between this aspect of the Sixth Circuit’s holding in Berger and the principle that a private school placement is improper unless it is reasonably calculated to enable the child to receive educational benefit. 416 F.Supp. at 172; see also Frank G. v. Bd. of Educ., 459 F.3d 356, 364-65 (2d Cir.2006) (applying “reasonably calculated” test while discussing and distinguishing, but not criticizing, Berger); Berger, 348 F.3d at 522 (quoting “reasonably calculated” test). In Burlington, the Supreme Court reasoned that because “parents who disagree with the proposed IEP are faced with a choice: go along with the IEP to the detriment of their child if it turns out to be appropriate or pay for what they consider to be the appropriate placement,” they are entitled to reimbursement of the expenses of that placement if it turns out they were right in choosing it. 471 U.S. at 370, 105 S.Ct. 1996. Implicit in this reasoning is the notion that parents rightfully decide on a private placement when it addresses, at least in part, their child’s special educational requirements, while the IEP does not.
We do not see, then, how the decision to reject public education in favor of enrolling a child in private school can be described as “reasonably calculated to enable the child to receive educational benefit” if the private school does not offer at least “some element of special education services in which the public school placement was deficient.” Berger, 348 F.3d at 523. To hold otherwise would, in essence, embrace the argument we explicitly rejected in Rafferty: that the IDEA entitles a parent, at public expense, to “seek any alternative school she wishes if the public education is inadequate.” 315 F.3d at 27.
Accordingly, the district court did not apply the wrong standard in finding that TCS is not an appropriate private school placement under the IDEA because it “does not offer any of the special education services recommended by the experts or the PET.” 416 F.Supp.2d at 173. We are left, then, to review this finding for clear error, and discern none. Although both of the experts who examined LI, as well as her present social worker, have stressed that LI needs direct teaching of social skills to manage the effects of her Asperger’s, it is undisputed that TCS has never provided her with this service, or any roughly equivalent intervention. TCS has also not supplied the cognitive behavioral therapy recommended by Popeneo or the close supervision or one-on-one tutoring offered by the PET as part of the Rehabilitation Act plan. The district court did not clearly err in judging TCS an inappropriate private placement in the absence of any of these special education services.
The parents resist this conclusion on two principal grounds. First, they liken certain of TCS’s distinguishing features to the interventions recommended for LI and found by the district court to constitute “special education” in assessing LI’s eligibility under the IDEA. As we have recognized, a private placement need provide only “some element of the special education services” missing from the public alternative in order to qualify as reasonably calculated to enable the child to receive educational benefit. Berger, 348 F.3d at 523 (emphasis added). Nor must the placement meet every last one of the child’s special education needs. Frank G., 459 F.3d at 365. But the reasonableness of the private placement necessarily depends on the nexus between the special education required and the special education provided. Here, the connection between, for example, the one-on-one tutoring recommended for LI and the relatively small student-faculty ratio boasted by TCS was more than remote enough to support the district court’s conclusion that the choice of the private school was not reasonably calculated to ensure that LI received educational benefit?particularly in light of the fact that, as we have just discussed, TCS did not offer anything approaching the direct teaching of social skills unanimously endorsed by the professionals who have tested and treated LI.
Second, the parents protest that, laboring under the trauma of LI’s suicide attempt and facing a lack of cooperation from the district, they acted reasonably by any measure in unilaterally placing LI at TCS. We sympathize with the family’s emotional upheaval, and we certainly do not condone the district’s apparent inattention to the task of locating a tutor for LI as it repeatedly promised it would. And we cannot doubt that TCS, where LI’s sister had prospered and where LI herself had expressed interest in attending even before the events of the fall of 2003, must have seemed an attractive solution to an exceedingly difficult set of circumstances. But these considerations cannot change the fact that TCS, where LI has remained for more than two full academic years, simply does not provide the special education services that LI’s mental health professionals have prescribed. The district court did not commit clear error when it found that TCS is not an appropriate private placement under the IDEA.23
Finally, Mr. and Mrs. I challenge the district court’s refusal to order the district to provide LI with compensatory education. We have recognized that, as another form of “appropriate relief” available under ? 1415(i)(2)(C)(iii), a court may require “compensatory education” in the form of “further services, in compensation for past deprivations” of IDEA benefits. Me. Sch. Admin. Dist. No. 35 v. Mr. R., 321 F.3d 9, 17-18 (1st Cir.2003). Compensatory education, like reimbursement, is a form of equitable relief. G ex rel. RG v. Fort Bragg Dependent Schs., 343 F.3d 295, 309 (4th Cir.2003); accord Ms. M, 360 F.3d at 273-74. Accordingly, we review the district court’s decision on compensatory education for abuse of discretion. Part IV.A, supra.
The district court considered the parents’ request for compensatory education in light of the other relief granted, namely its order to the district “to convene a PET meeting in accordance with State and Federal law to develop an IEP for [LI] that meets her unique needs as a student with Asperger’s Syndrome and a depressive disorder.” 416 F.Supp.2d at 168. Noting that “[t]he IEP necessarily will take into account the effect of the School District’s failure to identify and offer special education services earlier,” the district court declined to order compensatory education on the theory that the PET could better assess “what special education [LI] needs at this point. . . .” Id. at 173.
This approach strikes us as sensible and, moreover, not an abuse of the district court’s discretion. As the parents acknowledge, it is not unheard of for a compensatory education claim to be remanded to the responsible educational authority for consideration, particularly where “the district court does not believe that the record is sufficient to permit it to make the highly nuanced judgments necessary to resolve the claim….” Mr. R, 321 F.3d at 20. The parents, in fact, do not appear to object to such an approach here, provided we “ensure at the very least that guidelines governing the type, form, intensity, and duration of services are specified to assist the parties in moving forward without confusion or acrimony.” This is a worthy objective, to be sure, but we are not up to the task.
Like the district court, we confront an administrative record naturally devoid of any evidence as to the effect of the district’s failure to offer IDEA services to LI over the past two years and counting, since LI’s eligibility for those services was precisely what was at issue in the due process hearing. As a result, any “guidelines” that we might set forth to “govern” the resolution of the compensatory education claim would amount to an improper advisory opinion, just as it would have been a highly speculative exercise for the district court to attempt to resolve the claim on its merits. The district court ordered the district to convene a PET, in accordance with applicable law, for the purpose of formulating an IEP for LI that meets her needs, and further recognized that this task would necessarily require resolution of the compensatory education inquiry. We do not view this as an abuse of discretion, and Mr. and Mrs. I have not provided us with any authority to the contrary.
For the foregoing reasons, we affirm the judgment of the district court in its entirety.
Although this regulation was amended during the pendency of this appeal, none of the amendments affects our analysisSee generally Assistance to States for the Education of Children With Disabilities and Preschool Grants for Children with Disabilities, 71 Fed. Reg. 46,540 (Aug. 14, 2006) (codified at 34 C.F.R. pt. 300). For ease of reference, then, we cite to the current version of the regulation throughout.
“Asperger’s disorder is a developmental disability on the autism spectrum that is associated with significant misperceptions of otherwise routine elements of daily life. It is a permanent condition that is not treatable with medication.”Greenland, 358 F.3d at 154.
Adjustment disorder with depressed mood is characterized by a psychological response to an identifiable stressor that results in the development of clinically significant emotional or behavioral symptoms,i.e., depressed mood, tearfulness, or feelings of hopelessness. Am. Psychiatric Ass’n, Diagnostic and Statistical Manual of Mental Disorders 679 (4th ed. 2000) (“DSM-IV”).
Adaptive skills are those necessary to cope with common life demands and meet the standards of personal independence appropriate for one’s age, sociocultural background, and community setting. DSM-IV at 42. Executive skills are those necessary to think abstractly and to plan, initiate, monitor, and stop complex behaviorId. at 149.
In general, cognitive-behavioral therapy seeks to identify the thinking associated with unwanted feelings and behaviors in order to replace it with thoughts leading to more desirable reactions. Nat’l Ass’n of Cognitive-Behavioral Therapists, http://www.nacbt.org/ whatiscbt.htm (last visited January 26, 2007)
In addition to seeking review of the hearing officer’s decision under 20 U.S.C. ? 1415(i)(2), Mr. and Mrs. I also asserted a claim for relief under the Rehabilitation Act, which was rejected by the district court. They have not pursued this claim on appeal
The state must develop and implement an “individualized education program,” or “IEP,” to meet the particularized needs of each child with a disability. 20 U.S.C. ?? 1412(a)(4), 1414(d)(1)(A)(i)
Maine’s “Learning Results” are a “state-wide system” developed to “establish high academic standards at all grade levels” in eight different subject areas. Me.Rev.Stat. Ann. tit. 20-A, ? 6209 (Supp.2006)
More specifically, the district argues that, while Maine has developed extensive performance indicators to measure progress toward the “Learning Results,” 05-071-131 Me.Code R. ?? 1-8, state law does not require the use of these criteria, leaving measurement of student achievement to a “local assessment system.” Me.Rev.Stat. Ann. tit. 20-A, ? 6202-A. We need not, and do not, pass upon this argumentSee Part III.A.2, infra.
The Maine Department of Education has proposed amending its special education regulations to insert,inter alia, a requirement that “[a] child’s disability must result in an adverse affect [sic] on the child’s ability to learn and/or perform the academic, daily living, and/or age-relevant tasks required to demonstrate educational progress in the general curriculum.” Maine Unified Special Education Regulation ? VII.3 (proposed Nov. 2006), to be codified at 05-071-101 Me.Code R. ? 1 et. seq., available at http://www.maine. gov/education/rulechanges.htm (last visited Feb. 21, 2006). The proposed regulations also restrict the definition of “educational performance” for children older than five to “academic areas (written literacy skills, math, communication, etc.) [and] functional areas of performance (daily life activities). . . .” Id. ? II.9. These draft regulations, still in the public comment period, are not before us.
In attacking this reasoning, the district argues that the second part of the definition of “child with a disability” is too broad to function as a meaningful filter for IDEA eligibility. For the reasons stated in Part III.B.2,infra, we do not have occasion to address the scope of that provision here. For the moment, we note only that the district recognized, in its brief to the hearing officer, that the first and second prongs of ? 1401(3)(A) do operate in conjunction to determine eligibility. Post-Hrg. Memo. at 6 (“`adverse effect’ and the child’s `need for special education are intertwined. . . .'”).
Contrary to the district’s suggestion, that ? 1401(3)(A)(i) uses the words “impairment” or “serious” in naming some of the disabilities set forth provides no basis for inferring thatany condition must be a “serious impairment” to meet the statutory standard, let alone a “significant impact” requirement.
Moreover, Congress took this course of action despite the presidential committee report touted by the district and its amici in support of their proffered standard. President’s Comm’n on Excellence in Special Educ.,A New Era: Revitalizing Special Education for Children and Their Families (2002), available at http://www.ed.gov/inits/comm issionsboards/whspecialeducation/index.html (last visited Jan. 19, 2007). This report not only further expressed concern about over-identification, as the district and its amici point out, but strongly criticized the regulatory definitions of the disabilities recognized by the IDEA. Id. at 22. Because neither Congress nor the Department of Education appears to have acted on the commission’s recommendations, however, the report is of little use in construing the eligibility standards that have endured.
One district court recently used the secondary definition of “adverse” from a different dictionary?”causing harm”?to interpret the “adversely affects” requirement, concluding that, when a student “experiences only a slight impact on his educational performance, it cannot be said that the student is harmed.”Ashli & Gordon C. ex rel. Sidney C. v. Hawaii, No. 05-00429-HG-KSC, 2007 WL 247761, at *9 (D.Hawai’i Jan.23, 2007). In fact, however, the student is still “harmed”?if only slightly?so the court’s conclusion does not follow from the definition it cites. As a result, Ashli & Gordon C. does not persuasively address the absence of any qualitative limitation in the regulatory language.
The “adversely affects” test also serves an additional function: ensuring that it is the “enumerated disability, and not other factors” that impacts educational performanceSee Garda, supra, at 486.
The Connecticut decisions applied a nowsuperseded state regulation defining “socially and emotionally maladjusted” in part as “a condition which `significantly impedes the child’s rate of educational development.'”Doe, 753 F.Supp. at 70 n. 9 (quoting Conn. Agencies Regs. ? 10-76a-1(m) (1989)); see also Gregory M., 891 F.Supp. at 702 (articulating same test).
Maine recently passed emergency legislation, effective May 30, 2006, defining “child with a disability,” in relevant part, as: “[f]or children at least 3 years of age and under 20 years of age evaluated in accordance with [20 U.S.C. ?? 1414(a)-(c)] as measured by both standardized, norm-referenced diagnostic instruments and appropriate procedures with delays or impairments such that the children need special education . . . with at least one” of a number of specified conditions. An Act To Improve Early Childhood Education, 2006 Me. Legis. Serv. 662, sec. A-15, ? 7001(1-B)(B),to be codified at Me.Rev.Stat. Ann. tit. 20-A, ? 7001(1-B)(B). The Maine Department of Education has also proposed a regulation imposing a number of requirements, similar to Vermont’s, on the adverse effect determination. See Maine Unified Special Education Regulation, supra, ? VII.3 These versions of the Maine definitions are not before us, however, and we express no opinion on them.
Again, this regulation was amended effective October 13, 2006, but the amendment does not affect our analysis, so we cite to the current versionSee note 1, supra.
This definition is set forth as part of a statute requiring each school district to,inter alia, “[p]rovide special education for each exceptional student within its jurisdiction.” Me.Rev.Stat. Ann. tit. 20-A, ? 7202(5) (1993). The statute defines “exceptional student” as a person between the ages of five and twenty who “[r]equires special education because of an impairment in one or more” specified functions. Id. ? 7001(2). As we have observed, supra note 17, this provision was recently amended, as were ?? 7001(5) and 7202(5), but the amended versions are not before us.
Contrary to the district’s reading, the Second Circuit inJ.D. did not “conclude” that “training in peer relationship skills … is more akin to a related service rather than special education.” Rather, as we have discussed, the court in J.D. ruled that the student did not qualify for IDEA benefits because his condition did not adversely affect his educational performance in the manner required for IDEA eligibility under Vermont law. 224 F.3d at 67-68. The court in J.D. therefore had no occasion to define “training in peer relationship skills,” which the defendant had offered as part of a Rehabilitation Act plan, as either special education or a related service under the IDEA, and did not do so.
In its reply brief, the district contends that, even if LI in fact needed the services deemed special education by the district court, she did not need them “by reason of” her condition as required by ? 1401(3)(A)(ii). Because the district did not raise this argument in its principal brief (or, for that matter, before the district court), we do not consider itSee, e.g., Forcier v. Metro. Life Ins. Co., 469 F.3d 178, 183 (1st Cir.2006).
It should be noted thatFlorence County does not hold that a private school placement must be “reasonably calculated to enable the child to receive educational benefits” to give rise to reimbursement under the IDEA; in fact, whether the private placement there was proper had been “settled” by the time the case arrived at the Court. 510 U.S. at 12-13, 114 S.Ct. 361. While the “reasonably calculated” language appears in the opinion, the Court used it only in quoting from the lower court decision being reviewed, which in turn took the language from the Court’s earlier opinion in Bd. of Educ. v. Rowley, 458 U.S. 176, 102 S.Ct. 3034, 73 L.Ed.2d 690 (1982). Rowley itself set forth the “reasonably calculated” test as the measure of the adequacy of an IEP, not the propriety of a private school placement. 458 U.S. at 206-07, 102 S.Ct. 3034. Nevertheless, we have previously held, based on Florence County, that “a private school placement must be reasonably calculated to enable the child to receive educational benefits” to constitute a proper placement. Rafferty v. Cranston Pub. Sch. Comm., 315 F.3d 21, 26 (1st Cir.2002) (internal quotation marks omitted). We are bound by this prior holding, see, e.g., United States v. Malouf, 466 F.3d 21, 26-27 (1st Cir.2006), which, in any event, the parties do not question.
We do not reach, then, the district’s alternative arguments for affirmance: that the parents failed to provide the requisite notice of their intent to enroll LI in private school at public expense, and that the district offered LI a free and appropriate public education in the form of its Rehabilitation Act plan