Printer Friendly

Damages under the Privacy Act: sovereign immunity and a call for legislative reform.

INTRODUCTION
  I. THE NARROW CONSTRUCTION CANON:
     HISTORY, CRITIQUE, AND RECENT USE
 II. STATUTORY CONSTRUCTION
     AND STARE DECISIS
III. THE PRIVACY ACT CONTEXT
     A. The District Court Decision in Cooper
     B. The Ninth Circuit Decision in Cooper
     C. The Reasonableness of Both
        Cooper Decisions
     D. Pre-Cooper Decisions and
        Residual Ambiguity
        1. Actual Damages and the Text
        2. Beyond the Damages Provision Text
           a. Deterrence, Compensation, and
              Citizen Enforcement
           b. Limited Government Liability
           c. Broad Protection from
              Dignitary Harms
           d. Shared Purpose with
              Privacy Torts
           e. Vindication of the Constitutional
              Right to Privacy
           f. Additional Explicit
              Legislative History
           g. Inferences from Legislative
              History: Punitive Damages and
              Theories of Compromise
           h. Legislative History and
              the Privacy Protection
              Study Commission
        3. Analogies to Other Statutes
        4. Recap: Actual Damages Under the
           Standard Modes of Interpretation
 IV. DOE V. CHAO
     A. Combining the Chao Holding with a
        Narrow Reading of Actual Damages
     B. The Chao Methodology and Its Possible
        Implications
        1. Justice Souter on the Text
        2. Justice Ginsburg on the Text
        3. Possible Implications of Chao for
           Interpreting Actual Damages
CONCLUSION: A CALL FOR
  LEGISLATIVE REFORM


INTRODUCTION

In 1974, in the wake of Watergate, Congress passed the Privacy Act (the Act). (1) Broadly, the purpose of the Act is to regulate the treatment of personal information by the federal government. While the Act places certain requirements on the government, (2) it also enables individuals to take affirmative steps to ensure that their information is being protected. (3) Specifically, individuals may generally access personal records stored by the government, (4) request that the government correct errors in personal records, (5) request review of government refusals to correct alleged errors, (6) and bring civil lawsuits for any failures by the government to comply with the Act that result in an individual adverse effect. (7)

While the general distrust of the federal government in the aftermath of Watergate created a political climate ripe for passing the Act, (8) concerns about the federal government's increasing use of computers to collect, store, manipulate, and distribute personal information also played a role in Congress approving the Act. (9) The Act received strong bipartisan support (10) and passed with unusual speed because of time pressure created by the stormy political climate. (11) Considering the politics of urgency surrounding the Act when it was passed, it is perhaps no surprise that the Act is both supported by limited legislative history (12) and commonly targeted by critics, (13) a situation bound to produce work for the courts.

Among the provisions of the Act that have generated significant litigation, perhaps the most important for the Act's efficacy is the provision explaining the damages available to an adversely affected individual who brings suit against the federal government for mistreatment of personal information or other failures to comply with the Act. (14) This damages provision reads:
   In any suit brought under the provisions of subsection g(1)(C) or
   (D) of this section in which the court determines that the agency
   acted in a manner which was intentional or willful, the United
   States shall be liable to the individual in an amount equal to the
   sum of--(A) actual damages sustained by the individual as a result
   of the refusal or failure, but in no case shall a person entitled
   to recovery receive less than the sum of $1,000; and (B) the costs
   of the action together with reasonable attorney fees as determined
   by the court. (15)


One ongoing debate about the damages provision is whether "actual damages" is limited to "pecuniary," "out-of-pocket" losses, or whether a more expansive definition of "actual damages" is appropriate. (16) This question has been brought to court repeatedly since at least 1982, (17) and the circuits remain split. (18) The Supreme Court acknowledged this split while deciding another case involving interpretation of the Privacy Act damages provision, but declined to offer the circuits any guidance because the issue of defining actual damages was not properly before the Court. (19)

The question of how to define actual damages may seem like a minor point, but it has real importance for the remedial efficacy of the Act. Privacy Act violations often cause nonpecuniary harms, like fear and anxiety, without resulting in any out-of-pocket losses. (20) Identity theft, a possible result of many Privacy Act violations, results in no pecuniary loss in most cases, even in cases where victims must deal with hours of frustration and endless worries before the wrong done by the government can be righted. (21) Similarly, reputational harm--often permanent and wide-reaching in today's computerized world (22)--is a predictable result of Privacy Act violations and can be difficult to translate into pecuniary loss. If actual damages are limited to only pecuniary losses, many individuals who are significantly harmed by Privacy Act violations may receive no compensation from the government. According to one commentator, a sufficiently narrow interpretation of actual damages could render the entire Act "[t]oothless." (23) According to Daniel Solove, a narrow reading of actual damages would "mak[e] a total mockery of the Privacy Act." (24) Unless our trust in the government has changed so much since Watergate that there is no longer any concern about how the government treats personal information, the definition of actual damages really matters.

The question returned to court most recently in August 2008, when the District Court for the District of Northern California decided Cooper v. Federal Aviation Administration. (25) This Article is not about the Cooper case, but the case will be used as a lens for considering the general issue of how to interpret the "actual damages" language of the Act. The facts in Cooper are compelling. (26)

The plaintiff, Stanmore Cawthon Cooper, was a recreational pilot (27) who first obtained his pilot's license in 1964. (28) In order to fly legally, most pilots must have a valid "airman medical certificate" in addition to a pilot certificate, (29) and obtaining such a medical certificate requires the completion of an application to be filed with the Federal Aviation Administration (FAA). (30) Airman medical certificates must be renewed periodically, with an individual's renewal period depending on various factors. (31)

Cooper regularly renewed his medical certificate until about 1981, when he began to fear he was HIV-positive and no longer wanted to share his medical information. (32) His fears were confirmed in 1985 when he was diagnosed as HIV-positive, (33) and he did not again apply to renew his medical certificate until 1998. (34) In the interim, due to worsening HIV symptoms, he began collecting Social Security disability benefits in 1996. (35) When his symptoms eased in 1998, Cooper stopped collecting disabil- ity benefits. (36) On his application for a renewed medical certificate from the FAA in 1998, his first such application since 1985, Cooper failed to disclose his HIV status or the medication he was taking on account of the infection. (37) Similarly, he failed to disclose any of this information on his subsequent applications in 2000, 2002, and 2004. (38)

In 2002, the Department of Transportation (DOT) and Social Security Administration (SSA) began considering a joint investigation, know as Operation Safe Pilot, which would involve the interagency sharing of information (between DOT and SSA, and eventually the FAA as well) to prevent various types of fraud against the agencies. (39) Among the frauds targeted by the proposed investigation was one that described Cooper's situation: "Pilots that are claiming a debilitating condition with the SSA and claim good health to obtain a FAA medical certificate." (40)

Operation Safe Pilot was eventually implemented as a regional investigation in Northern California, (41) and the information sharing eventually implicated Cooper. (42) Cooper's pilot certificate was revoked, and he was indicted, eventually pleading guilty to a misdemeanor, "making and delivering a false official writing." (43)

Cooper brought a suit in the Northern District of California, claiming that the interagency sharing of his personal information violated [section] 552a(b) of the Privacy Act, (44) which requires written consent from an individual before an agency may share his records with another person or agency, with some exceptions, most notably the "routine use" exception. (45) The only damages that Cooper tried to demonstrate were emotional damages, largely resulting from the public disclosure of his HIV-positive status by the press after his indictment. (46) He offered no evidence of any pecuniary damages. (47)

In Cooper, the district court held that actual damages are limited to pecuniary harms. (48) Given the circuit split, this result was not overwhelmingly surprising. The district court's rationale for its decision, however, was surprising. While most courts attempting to define actual damages under the Act have treated the inquiry like a routine question of statutory interpretation, the district court found it dispositive that the damages provision of the Act is, at least to some extent, a waiver of federal sovereign immunity. Citing the long-debated canon of statutory construction that says federal waivers of sovereign immunity must be narrowly construed in favor of the government, (49) the district court held that, because the text of the damages clause is ambiguous, (50) a ruling in favor of the government was compelled. (51) In so holding, the court asserted that the narrow construction canon precludes any recourse to legislative history even when the text is ambiguous. (52) The court looked at the statutory language, decided it was ambiguous, and thus ruled for the government. The issue of defining actual damages under the Act, according to the district court, "must be decided by the rule that when 'analyzing whether Congress has waived the immunity of the United States, [courts] must construe waivers strictly in favor of the sovereign ... and not enlarge the waiver beyond what the language requires.'" (53)

Commentators were quick to assert that the Cooper district court decision took an ahistorical approach in applying the narrow construction canon. (54) The district court cited just two cases related to sovereign immunity: one decided by the Supreme Court in 1986 and one decided by the Ninth Circuit in 2004. (55) Although the narrow construction canon has been in use since at least Schillinger v. United States, (56) decided by the Supreme Court in 1894, deployment of the canon has varied substantially over the years. (57)

Most notably, influenced by Justice Scalia's general theory of statutory interpretation, (58) the stringency of the narrow construction carton's use seemed to climax in the early to mid-1990s in the cases of United States v. Nordic Village, Inc. (59) and Lane v. Pena. (60) In Nordic Village and Lane, not only did the Court emphasize that waivers of federal sovereign immunity must be "unequivocally expressed," (61) but it also stated with clarity that the only possible source of this unequivocal expression is the statutory text. (62) Ambiguity, the Court said, could not be resolved by recourse to legislative history or other nontextual sources; instead, any textual ambiguity must result in a decision in favor of the government. This, of course, sounds very similar to the Cooper district court's sovereign immunity discussion.

But the Court's sovereign immunity jurisprudence has not stagnated since Nordic Village and Lane. On the surface, perhaps the most significant exemplar of a different approach to using the narrow construction canon is Richlin Security Service Co. v. Chertoff, (63) decided only a few months before the Cooper district court decision was issued. In Richlin, after acknowledging that "some ambiguity subsists in the statutory text" (64) of the waiver in question, the Court proceeded to consider first prior Court precedents (65) and then legislative history and policy considerations. (66) Only then did the Court come to the narrow construction canon, and what it said seems to bear little resemblance to Nordic Village, Lane, or the district court decision in Cooper:
   The sovereign immunity canon is just that--a canon of construction.
   It is a tool for interpreting the law, and we have
   never held that it displaces the other traditional tools of
   statutory construction.... In this case, traditional tools of
   statutory construction.... [leave] no need for us to resort to
   the sovereign immunity canon because there is no ambiguity
   left for us to construe. (67)


Taking this approach--one used by the Supreme Court quite recently--would seem to result in a different outcome in Cooper than that reached by the district court, and this was borne out when the Ninth Circuit reversed the Cooper decision on appeal. (68) Citing Richlin, the Ninth Circuit asserted:
   The district court erred by failing to consider the full panoply of
   sources available to it for evaluating the scope of the
   Government's waiver of sovereign immunity under the Act [and]
   rely[ing] on the sovereign immunity canon alone, to the exclusion
   of the traditional tools of statutory construction. (69)


Considering the "text, purpose, and structure of the Act, as well as how actual damages has been construed in other closely analogous federal statutes," (70) the Ninth Circuit held that "actual damages" was clearly intended to include damages for nonpecuniary harms. Recourse to the sovereign immunity narrow construction canon was not needed because the standard modes of statutory interpretation resolved all ambiguity. (71) This Article makes two significant positive claims. First, neither the district court nor the Ninth Circuit was unreasonable or in violation of precedent in applying the narrow construction canon in their Cooper decisions. Use of the canon has varied significantly, even in Supreme Court opinions, and moreover, the application of canons of statutory construction is not governed by stare decisis. In construing the actual damages language of the Privacy Act, a court is free to take a variety of approaches to using the canon, ranging from the Nordic Village approach that will result in a narrow actual damages definition to the Richlin approach, which, as demonstrated by the Ninth Circuit, can reasonably result in an expansive definition.

Second, even if a court does take the Richlin approach, an expansive definition of actual damages will not necessarily result. Even if a court applies all of the standard tools of statutory construction, it could reasonably come to the conclusion that residual ambiguity remains and then apply the narrow construction canon and hold in favor of a narrow definition.

Based on all the uncertainty inherent in these two positive claims, and especially in light of the Supreme Court's damages-restricting 2004 Privacy Act decision in Doe v. Chao, (72) the normative thrust of this Article is that Congress should amend the Privacy Act's damages provision to make the statutory text clearly allow for the recovery of damages for nonpecuniary harms. Cases in which the various tools of statutory interpretation, including both the standard tools of construction and unique canons of construction, yield uncertain results are not uncommon, but this is a case in which Congress is perhaps particularly well positioned to make a change for the better. Unlike the subjects of many disputes over statutory language, the underlying question here requires little expert knowledge. Congress simply has to decide whether to allow damages for nonpecuniary harms that result from government misuse of personal information. Especially in light of ever-increasing concerns over identity theft, widespread reputational harm, and the computerization of information, Congress should decide in favor of recovery for such damages, and should amend the Privacy Act accordingly.

This Article proceeds as follows. Part I expounds upon the history, critique, and use of the narrow construction doctrine, focusing on the Supreme Court's federal sovereign immunity jurisprudence. Part II discusses the relationship between canons of construction and stare decisis, concluding that a court could take any number of approaches to applying the sovereign immunity narrow construction canon to the damages provision of the Privacy Act without inappropriately disrespecting precedent. Part III uses the district court and Ninth Circuit opinions in Cooper to exemplify two very different reasonable approaches to applying narrow construction. It then presents an extensive discussion and analysis of the non-sovereign-immunity arguments considered by the courts in cases involving the interpretation of actual damages under the Act, concluding that a court could reasonably determine that ambiguity remains after applying the standard tools of interpretation, meaning sovereign immunity could reasonably be deemed dispositive even under the Richlin approach used by the Ninth Circuit in its Cooper decision. Part IV discusses the Supreme Court's 2004 decision in Doe v. Chao, (73) explaining how that decision contributes to the need for legislative reform and also supports the view that an application of the standard tools of interpretation may leave residual ambiguity about the meaning of actual damages. Part V advocates for legislative reform and concludes.

I. THE NARROW CONSTRUCTION CANON: HISTORY, CRITIQUE, AND RECENT USE

The idea that waivers of federal sovereign immunity must be narrowly construed and expressed by clear statutory language dates back to at least the 1894 Supreme Court decision in Schillinger v. United States. (74) In Schillinger, the federal government was initially sued in the Court of Claims for allegedly violating a patent that covered a "mode of constructing concrete pavements." (75) The plaintiff, Schillinger, claimed that the statutes conferring jurisdiction upon the Court of Claims waived the government's sovereign immunity in this context. The Supreme Court disagreed, holding that the statutes re- quired "some element of contractual liability ... at the foundation of every action" brought in the Court of Claims, and denying jurisdiction for "mere torts" like patent infringement. (76)

At the very beginning of the Court's opinion, in explaining the Court's methodology, Justice Brewer stated:
   The United States cannot be sued in their courts without
   their consent, and in granting such consent Congress has an
   absolute discretion to specify the cases and contingencies in
   which the liability of the Government is submitted to the
   courts for judicial determination. Beyond the letter of such
   consent, the courts may not go, no matter how beneficial
   they may deem or in fact might be their possession of a larger
   jurisdiction over the liabilities of the Government. (77)


Five years later, in Price v. United States, (78) Justice Brewer similarly wrote for the Court that "[t]he Government is not liable to suit unless it consents thereto, and its liability in suit cannot be extended beyond the plain language of the statute authorizing it." (79) Justice Brewer instructed the reader to "[s]ee, among other cases, Schillinger v. United States," (80) but provided no further discussion on the origin of this principle.

Without explicitly acknowledging either of the aforementioned opinions, the Court reasserted this position in the 1927 case of Eastern Transportation Co. v. United States, (81) stating that "[t]he sovereignty of the United States raises a presumption against its suability, unless it is clearly shown; nor should a court enlarge its liability to suit beyond what the language requires." (82)

By 1931, in United States v. Michel, (83) the Court cited Price and Eastern Transportation for the assertion that is was "well estab- lished that suit may not be maintained against the United States in any case not clearly within the terms of the statute by which it consents to be sued." (84) The Court in 1940 referred to "the postulate that without specific statutory consent, no suit may be brought against the United States." (85) A year later, in United States v. Sherwood, the Court modified the language while expressing the same idea, explaining that any statutory waiver of federal sovereign immunity must be "strictly interpreted." (86)

None of these early cases explicitly preclude the use of nontextual modes of statutory interpretation. They insist on "statutory consent," "strict interpretation" and sticking to "plain language," but they do not explicitly say that legislative history or other nontextual modes of interpretation may not be used. This was also true of many other cases decided in this period. (87) In fact, one need not stray far in the opinions containing these assertions of the narrow construction canon to see how the Court felt about using legislative history to interpret waivers. In Michel, for example, the sentence following the assertion that waivers must be limited to the "terms of the statute" begins: "There is nothing in the legislative history of the provision to indicate...." (88) Similarly, in Sherwood, the paragraph containing the call for strict interpretation begins by noting that the Court has given "due regard to the words of [the statute] and to its legislative history." (89)

By 1945, the Supreme Court had openly acknowledged its leniency in applying the narrow construction canon. In Canadian Aviator, Ltd. v. United States, (90) the Court said:
   While the general history of the Act as outlined above does not
   establish that the statute necessarily extends to the noncollision
   cases in view of the rule of strict construction of statutory
   waiver of sovereign immunity, we think Congressional adoption of
   broad statutory language authorizing suit was deliberate and is
   not to be thwarted by an unduly restrictive interpretation. (91)


Although it does not explicitly mention legislative history, the Canadian Aviator opinion asserts that the waiver text should be considered in light of implicit congressional intent. At least one other Supreme Court case from the same time period seems to support this assertion. (92) Thus, for much of the first century of its existence, the narrow construction canon was treated as a tool to be used in conjunction with the standard modes of statutory interpretation, not an independent arbiter of cases to be used to the exclusion of the standard modes. In fact, some have doubted that the canon was even playing a significant analytical role by the end of its first century. In 1989, Professor Cass Sunstein classified narrow construction of federal sovereign immunity waivers as "obsolete" in his taxonomy of canons. (93) Sunstein offered no support or explanation for his conclusion, perhaps he meant that, even though courts continued to cite it, the narrow construction canon seemed to do little work in most cases. The standard modes of interpretation were still all in play, and for the most part no explanation was given of how exactly narrow construction was cabining interpretation. (94)

The Court's usage of the canon changed, however, once Justice Scalia became involved. In 1992, in United States v. Nordic Village, Inc., (95) Justice Scalia brought his general aversion to legislative history and other nontextual modes of interpretation (96) to the sovereign immunity context. Although Justice Scalia began by repeating the standard tale about unequivocalness and narrow construction, he went on to add:
   [L]egislative history has no bearing on the ambiguity point.
   As in the Eleventh Amendment context, the "unequivocal expression"
   of elimination of sovereign immunity that we insist
   upon is an expression in statutory text. If clarity does not exist
   there, it cannot be supplied by a committee report. (97)


In Justice Scalia's defense, as he suggests with his reference to the Court's Eleventh Amendment jurisprudence, his disregard for nontextual sources in interpreting waivers of sovereign immunity was not without precedent. By the time of Nordic Village, it was well established that nontextual sources would not be considered in evaluating potential waivers of the state sovereign immunity found in the Eleventh Amendment, and Justice Scalia was not responsible for authoring any of those key decisions. (98)

The Eleventh Amendment cases offer some support for Justice Scalia's position, but his Nordic Village decision does not merely follow what had been done in the past. In the Eleventh Amendment cases, the Court repeatedly emphasized "the principles of federalism that inform Eleventh Amendment doctrine." (99) Those cases involved states being sued in federal courts, and in the words of Justice Powell, "[a] State's constitutional interest in immunity encompasses not merely whether it may be sued, but where it may be sued." (100) In adopting the Eleventh Amendment doctrine's disregard for nontextual sources to the federal sovereign immunity context, Justice Scalia ignored this distinction. Given that Justice Scalia is generally a champion of federalism, (101) it cannot be that he simply overlooked this distinction. Rather, the best explanation for his choice to ignore the distinction here is perhaps that he views avoiding reliance on legislative history as more fundamental to proper jurisprudence than recognizing any difference between state governments and the federal government in this context.

Justice Scalia's modified use of the narrow construction canon did not go unnoticed. In dissent, after commenting more generally on "the Court's love affair with the doctrine of sovereign immunity," (102) Justice Stevens honed in on the exclusion of legislative history:
   [O]ne must ask what valid reason supports a construction of
   the waiver ... that is so "strict" that the Court will not even
   examine its legislative history.

   Surely the interest in requiring the Congress to draft its
   legislation with greater clarity or precision does not justify a
   refusal to make a good-faith effort to ascertain the actual meaning
   of the message it tried to convey in a statutory provision that is
   already on the books. (103)


Justice Stevens's criticism of Justice Scalia's opinion did not stop there. In a law review piece published the year after Nordic Village, (104) he derisively described the exclusion of legislative history as "still another piece of judicially-crafted armor plate," (105) and described Nordic Village as "another step down the path charted by Thrasymachus." (106) With more than a tinge of irony, Justice Stevens quoted "an unusually perceptive article written by an associate professor of law at the University of Virginia in 1970" (l07)--Antonin Scalia--describing federal sovereign immunity doctrine as making "blatant affront[s] to the basic precepts of justice," (l08) and lamenting the fact that "recently the scope of the doctrine has in some important respects been extended beyond what it was in 1882." (109)

In 1996, when the Court--in an opinion authored by Justice O'Connor, not Justice Scalia--reiterated its stringent Nordic Village approach to using the narrow construction canon in Lane v. Pena, (110) Justice Stevens again dissented, this time joined by Justice Breyer. Although he declined to reiterate his general objection, Justice Stevens did note, referring to the date of origin of the statutory language in question, that "Congress had no reason to suspect in 1978 that 14 years later this Court would adopt (and apply retroactively) a radically new and unforgiving approach to waivers of sovereign immunity." (111)

Justice Stevens was hardly alone in his critique of Justice Scalia's exclusion of nontextual sources from waiver considerations, or in his more general critique of the Court's waiver doctrine in the era of Nordic Village and Lane. Professors William Eskridge, Philip Frickey, and Elizabeth Garrett characterized these decisions as a "great expansion in the power of the [narrow construction] canon," and asserted that the Court had failed to justify this expansion, noting the "tension between the sovereign immunity canon and contemporary values supporting compensation for victims of wrongdoing." (112) Professor Frickey, writing alone, called the narrow construction canon and other similar canons "judicially created requirements that may dislodge an interpretation consistent with ordinary meaning." (113) Ann Madden objected that "[a]s a result of Nordic Vil- lage, when a statute lacks an explicit waiver of sovereign immunity, courts must find the government immune from compensatory damages, even when it is guilty of overt discrimination." (114) Professor Donald Doernberg claimed that a conflict existed between the Court's immunity doctrine and constitutional notions of protection from the government. (115)

More generally, some viewed the Court's expansion of the power of the narrow construction doctrine as exemplifying the danger of "loose canons." (116) Professors Eskridge, Frickey, and Garrett wrote:
   As the sovereign immunity example indicates, hoary canons
   sit like loaded guns, to be picked up and even modified to
   produce more firepower by skillful advocates and opportunistic
   judges. Regrettably, because the canons are rule-like in
   form, judges may rely upon them without acknowledging
   that the canons are rooted in controversial values, maybe
   have been phrased in different ways in prior cases, and can
   evolve over time on a case-by-case basis. (117)


The debate over narrow construction is, of course, part of the broader longstanding debate over the general role that canons of construction should play, (118) and many viewed Nordic Village and Lane as perfect ammunition for the anti-canon position. (119) Fortunately, at least in the eyes of the critics, the Court began backing away from Nordic Village before Lane was even de- cided. In the 1995 case of United States v. Williams, (120) Justice Ginsburg's majority opinion began by quoting Nordic Village and citing another case that also trumpets strict construction, (121) but then proceeded to take what has accurately been described as "a purposive approach at odds with the formal textualist approach of those earlier cases." (122) Noting that a ruling in favor of the government would conflict with the Court's "preference for commonsense inquiries over formalism," (123) Justice Ginsburg concluded that sovereign immunity had been waived. Although one should not attach too much significance to every decision a Justice makes about whether to join, concur, or dissent, it is somewhat surprising that Justice Scalia concurred, especially considering that three other Justices dissented. While Justice Scalia endorsed Nordic Village in his concurrence and claimed that the language of the waiver was simply unequivocal, he did not comment at all on Justice Ginsburg's methods. (124) Moreover, he did not engage at all with the dissent's assertion that the majority opinion was "an unusual departure from the bedrock principle that waivers of sovereign immunity must be unequivocally expressed." (125)

The Court's 1999 decision in West v. Gibson (126) further demonstrated the Court's declining use of stringent narrow construction. In Gibson, the Court did not discuss the nature of narrow construction, but it explicitly considered both legislative history and purpose in construing a waiver. (127) Justice Scalia was among the dissenters, but Justice O'Connor-the author of Lane--joined the majority.

If Williams and Gibson are indicative of the gradual erosion of the Court's stringent use of narrow construction since the mid1990s, Richlin Security Service Co. v. Chertoff (128) displays the full magnitude of the movement. Language from Richlin, decided just a few months before the district court decision in Cooper, was quoted in the Introduction, but it is worth repeating again. Justice Alito, writing for a unanimous Court, after considering other cases, legislative history, and policy arguments to resolve the textual ambiguity of a possible immunity waiver, said:
   The Sovereign immunity canon is just that--a canon of construction.
   It is a tool for interpreting the law, and we have
   never held that it displaces the other traditional tools of
   statutory construction.... In this case, traditional tools of
   statutory construction [leave] no need for us to resort to the
   sovereign immunity canon because there is no ambiguity
   left for us to construe. (129)


In spite of Justice Alito's claim to the contrary, this is a far cry from Nordic Village. Justice Scalia did not sign on to the part of the opinion considering legislative history, and Justice Thomas did not sign on to the parts considering legislative history, other cases, or policy arguments, but the entire Court signed on to the language quoted above.

The Richlin opinion, particularly the language about "traditional tools of statutory construction [leaving] no need ... to resort to the sovereign immunity canon," (130) presents a picture that is very different from narrow construction circa Nordic Village. In Richlin, the narrow construction canon is treated as a last resort when the traditional modes of statutory construction fail to resolve ambiguities.

Although it may be tempting to interpret the changing use of the canon from Nordic Village to Richlin as akin to a change in precedent, the two are not the same. As the next Part discusses, the generally held view is that the use of canons of statutory construction is not governed by stare decisis.

II. STATUTORY CONSTRUCTION AND STARE DECISIS

The vast majority of scholars agree that stare decisis does not apply to doctrines of statutory construction, (131) including when and how to use the many canons of construction that are available to judges. (132) This conversation has taken place mainly in the academy, and the Supreme Court has not explicitly commented on whether statutory interpretation methodology is governed by stare decisis, but the Court has come very close to saying that it is not. (133) In the context of canons of interpretation, for example, the Court has described the canons as rules of thumb that are not "mandatory rules" but rather "guides." (134)

Moreover, the reality of the situation is readily observable on the bench. Even though current members of the Supreme Court, for instance, disagree significantly in their approaches to statutory interpretation, Justices are not accused of violating precedent solely on the basis of their interpretation methods in the same manner that they are accused of violating precedent when there are other types of disagreements. (135) Sydney Foster has highlighted two observable differences between the Court's treatment of questions of statutory interpretation methodology and its treatment of issues that are governed by stare decisis. (136)

First, although the Court insists on a "special justification" for overruling prior decisions governed by stare decisis, no such justification has been required for diverging from previously prevalent methodologies of statutory interpretation. (137) Second, although Justices sometimes allow precedent to be the but-for cause for deciding an issue a certain way where stare decisis governs, there are no examples of Justices using a method of statutory interpretation with which they disagree solely because the Court had used it in the past. (138)

For purposes of this Article, the importance of stare decisis not governing statutory interpretation methods is that what may at first seem like a change in the precedential law from Nordic Village to Richlin should not be viewed in that manner. While the Court has undoubtedly taken different approaches to using the narrow construction canon over time, and this might be useful for predicting how the Court will approach a particular case, the Court is not bound to use the canon in any particular way. Even though the Court has apparently moved toward the Richlin approach recently, this does not mean that it should or will abandon the approach in Nordic Village. In fact, the Court has already shown that it can move back and forth very quickly in its use of the canon. As was discussed in Part I, the Court showed signs of using a less stringent approach after Nordic Village in Williams, decided in 1995, only then to hand down the stringent Lane decision in 1996, and then to return to a less stringent approach in West in 1999. The lack of stare decisis treatment of the use of interpretive canons facilitates such oscillation. A court deciding how to construe actual damages under the Privacy Act may adopt the Nordic Village approach or it may adopt the Richlin approach. (139) In the next Part, a more detailed look at the Cooper case and the cases that came before it illustrates how these two different available approaches play out in the Privacy Act context.

III. THE PRIVACY ACT CONTEXT

The opinions of the district court and the Ninth Circuit in Cooper exemplify two very different but reasonable approaches to applying narrow construction in the Privacy Act context. The district court applied narrow construction strictly and considered nothing beyond the Act's text. The Ninth Circuit took the broader Richlin approach. This Part examines in detail these disparate decisions in Cooper, and then extensively discusses and analyzes the non-sovereign-immunity arguments considered by pre-Cooper courts in cases involving the interpretation of actual damages under the Act, concluding that a court could reasonably determine that ambiguity remains even under the Richlin approach used by the Ninth Circuit in its Cooper decision.

A. The District Court Decision in Cooper

After determining that Stanmore Cooper's rights under Section 552a(b) of the Privacy Act had in fact been violated by the interagency sharing of his personal information under Operation Safe Pilot, (140) the district court considered the question of damages. The court first stated that the term "actual damages" is "facially ambiguous." (141) The court then cited Ninth Circuit cases that interpreted "actual damages" to mean economic damages in the context of other statutes, and a Ninth Circuit case that interpreted the term actual damages in the Fair Credit Reporting Act to include "emotional distress and humiliation." (142)

After acknowledging that the dispute over the Privacy Act issue was based largely on disagreement about the meaning of the legislative history, the district court asserted:
   The court need not ... conduct its own analysis of the legislative
   history [because] the issue must be decided by the rule
   that when "analyzing whether Congress has waived the
   immunity of the United States, [courts] must construe waivers
   strictly in favor of the sovereign ... and not enlarge the
   waiver beyond what the languages requires." (143)


Following this "rule," the district court held that actual damages under the Privacy Act do not include nonpecuniary damages, and thus granted summary judgment in favor of the government. (144)

B. The Ninth Circuit Decision in Cooper

On appeal, in an opinion written by Judge Milan D. Smith, Jr., the Ninth Circuit first addressed whether the sovereign immunity canon was applicable at all and then addressed the question of how to apply it. (145) Mr. Cooper had argued that the canon only applied if there was a question as to the initial waiver of sovereign immunity and that here there was no such question because the Privacy Act clearly allows for a private action against the federal government. (146) Although Mr. Cooper's position reflected at least one well-respected scholar's views on how the narrow construction canon has been applied in recent years, (147) the court rejected it, relying primarily on

Nordic Village, in which the court applied the canon to prevent the recovery of monetary damages in a bankruptcy action under a certain section of Chapter 11 even though that section allowed for other forms of relief. (148) The only other case the Ninth Circuit cited on this point was Lane. (149)

Having decided that the narrow construction canon was applicable, the court then criticized the district court's approach to applying the canon. Citing Richlin, the court said that "[t]he district court erred by failing to consider the full panoply of sources available to it for evaluating the scope of the Government's waiver." (150) The court also cited the Supreme Court's 2001 decision in Chickasaw Nation v. United States for language indicating that "canons are 'not mandatory rules' but guides 'designed to help judges determine the Legislature's intent' [and] 'other circumstances evincing congressional intent can overcome their force.'" (151)

Elaborating on its view on how the canon should be applied, the court insisted that the policy underlying the Act could not be ignored in determining the scope of an immunity waiver. (152) It advocated tailoring the available remedies to the protection of the privacy right in question, (153) an approach that originated in the judicial shaping of remedies for violations of constitutional rights. (154)

The court considered the legislative history of the Privacy Act but decided it was "not a reliable source for the meaning of ac- tual damages because both sides of the argument can readily find support for their respective positions in that history." (155) The court therefore decided the case on the basis of "text, purpose, and structure of the Act, as well as how actual damages has been construed in other closely analogous federal statutes.'' (156)

The Ninth Circuit's approach was thus in clear contrast to the majority decision in Nordic Village, which looked to the text and nothing else before ruling for the government. In Lane, the Supreme Court's primary reaffirmation of the Nordic Village approach, the Court did entertain arguments about the "larger statutory scheme," (157) but ultimately declined to rely on the fact that narrow construction would create what it acknowledged to be a "somewhat bewildering" (158) statutory scheme, "[g]iven the existence of a statutory provision that is directed precisely to the [question at hand of the] remedies available." (159) The Court did consider a statute other than the one being construed but only insofar as the language in that statute was possibly amending the scope of the waiver in question. (160)

The Ninth Circuit's approach in Cooper clearly mimics Richlin and not Nordic Village or Lane. (161) The court concluded that "the term actual damages is unambiguous" in the Privacy Act (162) and that "Congress intended the term actual damages in the Act to encompass both pecuniary and nonpecuniary injuries." (163) Although the court recognized that the term "actual damages" had no plain meaning that could be discerned from regular or legal dictionaries, (164) it found this clarity elsewhere.

First, the court emphasized the broad statement of purpose found in the Congressional Findings and Statement of Purpose attached to the Act, which said that the Act was written "to provide certain safeguards for an individual against an invasion of personal privacy by requiring federal agencies ... to ... be subject to civil suit for any damages which occur as a result of willful or intentional action which violates any individual's rights under this Act." (165) Considering the frequency and importance of nonpecuniary harms stemming from Act violations, (166) the court found it "difficult to see how Congress's stated goal ... could be fully realized unless the Act encompasses both pecuniary and nonpecuniary injuries." (167)

Second, the court cited other sections of the Act in which "Congress signaled its intent.., to extend monetary recovery beyond pure economic loss." (168) In one section, the Act requires agencies to maintain systems of records that protect against threats or hazards to records that "could result in substantial harm, embarrassment, inconvenience, or unfairness to any individual on whom information is [being] maintained." (169) In another section, the Act requires agencies to maintain records so as to "assure fairness in any determination relating to the ... character ... of ... the individual that may be made on the basis of such record[s]." (170) The concern shown for nonpecuniary harm in these sections of the Act, according to the court, supports a broad reading of "actual damages."

Third, the court noted that actual damages under the Act are available to any party that can show that a violation of the Act caused them to suffer an "adverse effect," a term that has been interpreted by many courts to include nonpecuniary harms. (171) It would be an unreasonable reading of the damages provision of the Act, the court said, "[t]o recognize that the Act entitles one to actual damages for an adverse effect related to one's mental or emotional well-being, or one's character ... while holding that one injured under the Act cannot recover actual damages for nonpecuniary injuries." (172)

Finally, the court relied on the similar actual damages language found in the Fair Credit Reporting Act (FCRA), (173) passed just four years prior to the Privacy Act, (174) and the prior interpretations of that FCRA language by various circuit courts. Because both the Privacy Act and the FCRA were passed to protect privacy rights against private information disclosure, and because both statutes provide similar remedies, the court deemed it fair to presume Congress intended "actual damages" to have the same meaning in both laws. (175) Thus, based on text, purpose, structure, and prior interpretations of similar language in an analogous statute, the Ninth Circuit held that the term "actual damages" in the Privacy Act unambiguously allows recovery for nonpecuniary harms.

The Ninth Circuit declined to hear the Cooper case en banc. (176) Seven judges, however, joined a sharp dissent to the denial of rehearing, authored by Judge Diarmuid O'Scannlain. (177) Citing Lane and Nordic Village, Judge O'Scannlain asserted that "[t]he Supreme Court has consistently held that the sovereign immunity of the United States may be waived only by an unequivocal expression in statutory text" (178) and accused the majority of "neglect[ing] this principle." (179) Judge Smith defended his original opinion in a concurrence filed with the order denying rehearing. He again cited Richlin, (180) and reasserted the propriety of "look[ing] to several sources manifesting the Act's overall objective" (181) in determining the scope of the waiver.

C. The Reasonableness of Both Cooper Decisions

The district court and the Ninth Circuit came to very different decisions in the Cooper case in large part because of their differing applications of the narrow construction canon. Although both courts deemed the canon applicable, applying the canon entailed the consideration of different sets of sources for the two courts. The district court applied the narrow construction canon to end its inquiry after determining that the text had no clear meaning. The Ninth Circuit, on the other hand, after agreeing with the district court that the text of the immunity waiver itself was not enough to achieve clarity, went on to explore statutory purpose and structure and prior interpretations of an analogous statute.

As was established in Part I, both of these methods of applying the narrow construction canon have historical precedent in Supreme Court decisions. Although more recent Supreme Court decisions may lean toward the method employed by the Ninth Circuit, methods of statutory interpretation, including the use of canons, do not fall under the umbrella of stare decisis, as was discussed in Part II. The district court was entitled to take the Nordic Village approach; the Ninth Circuit was entitled to take the Richlin approach; and another court deciding the case tomorrow in a circuit with no substantive precedents would be entitled to take any approach it sees fit to applying the canon. If the Nordic Village approach is used, the decision of the district court in Cooper is the correct one. If the Richlin approach is used, the Ninth Circuit's decision does not seem unreasonable. Thus, depending on a court's approach, the actual damages available under the Privacy Act can reasonably be interpreted either to include or not to include recovery for nonpecuniary harm.

D. Pre-Cooper Decisions and Residual Ambiguity

But the situation is even murkier than the Cooper decisions acknowledged. Prior to Cooper, the courts that considered the interpretation of "actual damages" under the Privacy Act largely ignored sovereign immunity.182 Thus, the courts treated these cases as matters of standard statutory interpretation, tacitly using Richlin's expansive approach to the available sources of authority in interpreting the Privacy Act's immunity waiver. Although the reason for the complete omission of references to sovereign immunity is unclear, the best explanation might be that almost all of the cases were decided prior to Nordic Village and therefore litigants and judges may not have seen sovereign immunity and the narrow construction canon as having much bite. Weighing the available sources of authority and methods of interpretation differently, courts arrived at differing opinions before Cooper, none of which are necessarily unreasonable. Taken together, however, the differing opinions suggest that ambiguity remains even when all the standard modes of interpretation are exhausted, and therefore that the best decision under the Richlin approach may be that sovereign immunity is dispositive. This section argues for that position by mining the arguments presented by pre-Cooper courts and by scholars who considered the actual damages interpretive question prior to Cooper.

1. Actual Damages and the Text

Courts and scholars agree on at least one point: The text of the damages provision provides little guidance on the scope of "actual damages" under the Act. The phrase "actual damages" has no consistent legal interpretation,183 and the damages pro- vision does nothing to narrow down the array of interpretations of the phrase existent in other areas of the law.1s4 There is seemingly unanimous agreement that the text of the damages provision alone cannot resolve the interpretation of the "actual damages" language. As was discussed above, both Cooper decisions agree with this position.

2. Beyond the Damages Provision Text

Most courts and scholars that examined the issue prior to Cooper ultimately relied on statutory purpose, legislative history, and policy considerations in choosing the best interpretation of "actual damages." This subsection considers the various arguments that go beyond the text of the waiver provision, and concludes that no argument provides clear interpretive guidance. Even if one argument seems to provide clear guidance, when the arguments on both sides are considered, it is difficult to say that no ambiguity remains.

a. Deterrence, Compensation, and Citizen Enforcement

Two clear purposes of the Privacy Act are deterring government mistreatment of personal information and compensating individuals whose information is mistreated. (185) Those in favor of an expansive interpretation of actual damages put substantial reliance on these broad purposes.186 If the government is not punished enough for the mistreatment of information, or if individuals do not receive enough compensation when their information is mistreated, the proponents of expansive interpretation argue that these purposes will be frustrated. One should not look at these purposes in isolation, however, when considering how to interpret actual damages. That deterrence and compensation are broad purposes of the Act does not mean that every ambiguity should be construed so as to increase deterrence and compensation.

Although it is true that the recovery of actual damages through civil suits is the primary mechanism for deterrence under the Act (187) the Act does provide for criminal penalties for certain violating government employees. (188) Admittedly, the range of activities that are criminally punishable is considerably narrower than the range of activities for which civil suit may be brought. (189) Nonetheless, the specter of criminal punishment is a significant deterrent.

It is also true that most harms associated with the mistreatment of personal information do not involve only monetary loss. As was discussed earlier, (190) mistreatment of information often results in nothing more than anxiety and frustration tied to the fear of identity theft. Often in such cases there is no monetary loss. On the other hand, there are many privacy breaches that do result in monetary loss. Such breaches presumably were more significant, and thus more salient to legislators, when the Privacy Act was passed in 1974, a time before identity theft had become an ever-present threat. For such breaches, the denial of recovery for nonpecuniary harms does not frustrate the compensation of victims entirely.

It does not follow from these arguments that a narrow reading of actual damages to include only monetary loss does not frustrate the purposes of the Act at all--it surely does. (191) But this purposive argument is not dispositive. If every ambiguity in the Act were construed in favor of deterrence and compensation, overdeterrence and excessive compensation would result. To assume that Congress sought to achieve these purposes through a broad interpretation of actual damages rather than through other possible means is to assume the answer to the question under consideration. At best, these broad purposive arguments are a thumb on the scale in favor of including nonpecuniary harms in actual damages, and as we will see, there are thumbs on both sides.

b. Limited Government Liability

Those in favor of a broad reading of actual damages rely on purposive arguments about compensation and deterrence, but those on the other side of the debate argue that these purposes were to be pursued against a backdrop of limited government liability. For example, in Fitzpatrick v. IRS, (192) the Eleventh Circuit explained that "the evolution and structure of the damage provisions indicate Congressional intent to restrict damage liability to a maximum.... Throughout the Privacy Act debate, a central concern was the scope of potential government liability for damages." (193)

The Fitzpatrick court and others were correct in finding that Congress was concerned with limiting liability; even parties in favor of a broad interpretation of actual damages acknowledge this point. (194) But although there is general agreement that Congress wanted to limit government liability, there is no such agreement as to whether or not the requirement that damages be "actual" was an avenue through which Congress sought to protect the federal coffers. As Frederick Z. Lodge points out in his article, the congressional discussion of curbing potentially excessive government liability focused on calibrating the level of conduct required for culpability and evaluating the possibility of punitive damages, not on the breadth of actual dam- ages. (195) Even with an expansive definition of actual damages in place, Congress placed significant hurdles in the way of excessive liability under the Privacy Act. (196) Most notably, a victim must prove that the mistreatment was willful or intentional, a very difficult burden to carry.

Between the passage of the Privacy Act in 1974 and August 1983, sixty damages claims were brought against the government. (197) Of those sixty, all but four were dismissed before the issue of damages was even considered, and courts awarded actual damages in only two cases. (198) Clearly, the Privacy Act contains other mechanisms for preventing runaway government liability. This position is buoyed by the significantly lower costs associated with the Privacy Act than have been estimated by the Office of Management and Budget since the Act's inception. (199) Just as the broad concerns about deterrence and compensation are not determinative support for including recovery for nonpecuniary harms in actual damages, concerns about excessive government liability do not settle the matter in the other direction--they are just a thumb on the other side of the scale.

c. Broad Protection from Dignitary Harms

Like the Ninth Circuit in Cooper, at least one pre-Cooper commentator has pointed to the seemingly expansive language found in another part of the Privacy Act as a reason to interpret the damages provision expansively. Mr. Lodge points to the provision of the Privacy Act that urges the government to take precautions to prevent "substantial harm, embarrassment, inconvenience, or unfairness to any individual on whom information is maintained" (200) as support for the position that the "actual damages" language must be read broadly. (201) Because the Privacy Act urges protection of "dignitary interests" in this other provision, Mr. Lodge argues, the damages provision of the Act should be read to allow for suits to recover damages for harm to dignitary interests. (202)

Although this language about "embarrassment, inconvenience, or unfairness" is suggestive of an expansive view of remediable harms, it does not relate to remedies. The section of the Act containing this language describes the types of "administrative, technical, and physical safeguards" (203) that agencies must put in place, not what remedies are available when willful or intentional mistreatment of information occurs despite these safeguards. Because such willful and intentional mistreatment is the only concern of the damages provision, the "embarrassment, inconvenience, or unfairness" language is nothing more than suggestive.

d. Shared Purpose with Privacy Torts

The Privacy Act is, of course, not the only legal protection against the improper treatment of personal information. Tort causes of action are available for disclosure of private information, appropriation of a name or likeness, and other misuses of personal information. (204) At common law, "[o]ne can recover damages for emotional distress, personal humiliation, and nonpecuniary loss" for these privacy torts. (205) Given the common purpose of the Privacy Act and privacy torts, Congress was surely aware of the availability of these expansive remedies for privacy torts when it passed the Act, (206) and some have argued that this fact weighs in favor of an expansive reading of the Act. (207) In Johnson v. Department of Treasury, for instance, the Fifth Circuit explained that "[i]f Congress borrowed from the common law in designing the remedy provision, surely Congress was not unmindful of the common law when it selected the words 'actual damages' for the accompanying damages provision." (208)

Those favoring broad damage recovery under the Privacy Act also note that the legislature considering the Act in 1974 was presumably aware of the Supreme Court's decisions in two cases, Gertz v. Robert Welch, (209) and, to a lesser extent, Time, Inc. v. Hill. (210) In Gertz, which came down just a few months before the Act was passed, the Supreme Court held that if a state has a lower standard for the intent element of defamation than "actual malice," compensation for the defamed party must be limited to "actual injury" in order for the state defamation law not to violate the First Amendment. (211) Although the Court refused to define "actual injury," (212) it provided at least some guidance, stating that "actual injury is not limited to out-of-pocket loss." (213) Since Congress was presumably aware of Gertz when it passed the Act, some have argued that "actual damages" in the damages provision should not be limited to out-of-pocket loss either. (214)

In Time, the Court applied First Amendment principles to limit the application of a New York statute protecting the use of individuals' names and likenesses for commercial purposes. (215) Specifically, the Court held that Time was not liable for publishing, in an account of a play based on a true story, many of the same inaccuracies that the play itself contained, in the absence of a showing of knowledge of the inaccuracies or recklessness on the magazine's part. (216) In a sidebar discussion about the similarities between the interests protected by privacy torts and causes of action for libel and slander, the Court mentioned that "[i]n the 'right of privacy' [tort] cases, the primary damage is ... mental distress." (217) Even though this language appeared only in a footnote and was not central to the Court's opinion, it is another example of Supreme Court recognition of the prevalence of nonpecuniary harms when privacy is compromised. At least one court interpreting actual damages expansively under the Act has presumed congressional awareness of Time. (218)

Congressional awareness of expansive remedies or judicial acknowledgement of nonpecuniary harms in the tort context is, of course, not controlling; it is, at best, suggestive. In considering the value of this evidence for interpreting the damages provision of the Privacy Act, one federal court deemed any reliance on Gertz "misplaced," arguing that the actual legislative history of the Privacy Act is a more appropriate source for determining congressional intent. (219) Again, this might be a piece of evidence in favor of a broad interpretation of actual damages, but if the goal is to rid "actual damages" of its latent ambiguity, these arguments do not seem up to the task.

e. Vindication of the Constitutional Right to Privacy

In Johnson v. Department of Treasury, (220) the Fifth Circuit, in arguing for a broad interpretation of actual damages, cited language from the 1978 Supreme Court case Carey v. Piphus (221) about compensating injuries caused by the deprivation of constitutional rights. In Carey, the Court held that "the rules governing compensation for injuries caused by the deprivation of constitutional rights should be tailored to the interests protected by the particular right in question." (222) The court in Johnson argued that because privacy is a constitutional right and thus falls within the range of interests countenanced by the Supreme Court in Carey, "[l]ogically ... Congress, in choosing the term 'actual damages' was tailoring the compensation to the nature of the privacy interest protected. Obviously, mental distress is the normal and typical damage resulting from an invasion of privacy," (223) So Congress must have desired to compensate such harm when it authorized suit for actual damages. (224)

There are at least two problems with this argument. First, the court in Johnson assumed that Congress had in mind a principle set forth in a Supreme Court decision in 1978 while considering passing the Privacy Act in 1974. The Carey court may have been stating a principle it believed to exist in law prior to 1974, but just because the principle could be derived from prior law does not mean that courts should assume Congress had knowledge of it.

Second, and more importantly, although courts have held that there is a constitutional "right to privacy," (225) not every privacy right is a constitutionally protected interest. Courts disagree about the existence and extent of a constitutional right to information privacy. The circuits are not of one mind with regard to whether the right to privacy emanating from the Fourteenth Amendment includes protection against inappropriate use or dissemination of information, (226) and "the Supreme Court has not yet decided whether informational privacy is constitutionally protected." (227) The Court has at least hinted at the existence

of an information privacy branch of the constitutional right to privacy. Most notably, in Whalen v. Roe, (228) the Court stated: "The cases sometimes characterized as protecting 'privacy' have in fact involved at least two different kinds of interests. One is the individual interest in avoiding disclosure of personal matters, and another is the interest in independence in making certain kinds of important decisions." (229) The first kind of interest identified in Whalen certainly sounds like a constitutionally protected interest in information privacy, but the Court in Whalen did not elaborate on the existence of such a constitutional interest, nor did it find that interest violated by the New York healthcare information statute at issue in the case. The Whalen language is thus generally considered dictum, (230) though some lower courts have seized upon it in finding a constitutional right to information privacy. (231)

The argument that broad interpretation of the damages provision is appropriate because one of the purposes of the Privacy Act is to vindicate a constitutional right, and because Congress was following the Carey method of tailoring remedies to constitutional violations, is thus debatable at best.

f. Additional Explicit Legislative History

The prior subsections make reference to legislative history to support purposive arguments about the interpretation of the "actual damages" language. Although neither the district court nor the Ninth Circuit in Cooper put any weight on the Privacy Act's legislative history, pre-Cooper courts and scholars, to a limited extent, relied on legislative history even when it was not relevant to any of the above purposive arguments.

The Johnson court, for instance, discussed comments made by Representative McCloskey during the House debate. (232) Representative McCloskey was concerned that "it is often quite difficult

to prove actual damages," (233) and the court in Johnson argued that Representative McCloskey's statement suggests a broad definition of actual damages since out-of-pocket losses are generally easy to prove. (234) Even the court in Johnson, however, recognized the uncertainty of this jump in reasoning. (235) Out-of-pocket damages, just like nonpecuniary damages, may be difficult to prove for Privacy Act violations. To establish a violation, causation must always be demonstrated, and many responses to the mistreatment of personal information are not easy to tie to that mistreatment. (236)

Representative McCloskey also referred to actual damages as "ordinary damages" during the House debates over the Act. (237) According to the Johnson court, "[c]ertainly 'ordinary damages' are compensatory damages--not damages for out-of-pocket loss alone." (238) A closer look at Representative McCloskey's words in context, however, reveals that he was simply drawing the distinction between ordinary damages and punitive damages. (239) He used "ordinary" to mean other-than-punitive. His statement offers no insight into the proper interpretation of actual damages.

As an example on the other side of the debate, the court in Houston v. Department of Treasury (240) put significant weight on Representative Eckhardt's comments. (241) During the House debate, Representative Eckhardt said: "There is nothing in this that would provide for any damages beyond ... actual out-of-pocket expenses...." (242) Relying on this statement, and the lack of a challenge to it during the debates, the court in Houston asserted that "Congress ... intended to limit its 'actual damages' to 'out-of-pocket' expenses." (243) Although Representative Eckhardt's statement at first seems convincing, and the language certainly favors a narrow reading of actual damages, context again is key. Whereas the court in Johnson failed to notice that Representative McCloskey's seemingly expansive language was in fact just a way of making a distinction between punitive damages and other damages, it readily noted that Representative Eckhardt's apparently restrictive language--emphasized by the court in Houston--was about that very same distinction. (244) When Representative Eckhardt said "[t]here is nothing in this that would provide for any damages beyond ... actual out-of-pocket expenses," the "this" he had in mind was not the Privacy Act as a whole or the damages provision of the Act, but rather a proposed amendment to add punitive damages. (245)

Although the language used by Representative Eckhardt may seem more telling on the surface than that used by Representative McCloskey, neither congressman's comments offer legitimate insight. Aside from the bits of legislative history mentioned above in relation to the more purposive arguments, there simply is not much probative explicit legislative history. After all, "the drafters of the [Act] never directly discussed emotional distress." (246) The remarks of Representatives Eckhardt and McCloskey appear at first glance to be useful and on point, but in the end the explicit legislative history is just not convincing. (247) As another commentator has observed, efforts to use explicit legislative history to discern the meaning of actual damages demonstrate that "the exercise of divining what was meant from a number of related, but not exact, statements [is] just as difficult as interpreting the language of the statute itself." (248) Perhaps for this reason, courts and scholars have often turned to what may be inferred from this explicit history.

g. Inferences from Legislative History: Punitive Damages and Theories of Compromise

At the heart of much of the pre-Cooper jurisprudence and scholarship on the meaning of actual damages in the Privacy Act is a discussion of what can be inferred from the changes to the Act as it progressed through Congress. Specifically, the focus has been on two main issues: the consideration of including punitive damages over the course of the process, and the scope of a final compromise between the Senate and the House on the wording of the damages provision.

Both the House and the Senate allowed for punitive damages in the original versions of the bills that were eventually reconciled to create the Privacy Act. (249) As these two bills went through the House and the Senate, both were modified to remove punitive damages before the two houses of Congress had to get together to compromise on a final statute. (250) The House bill did not reach its final pre-compromise stage, however, without a vigorous debate over an amendment, mentioned in the prior subsection, that would have allowed for punitive damages in certain instances. (251)

According to Mr. Lodge, "[t]he existence of such strong sentiment in favor of punitive damages" makes it unlikely that the final compromise was so narrow as to allow only for out-of-pocket losses. (252) The court in Johnson also pointed to the debate over punitive damages as evidence in favor of broad interpretation, (253) but for a different reason than Mr. Lodge. The court in Johnson used the debate over punitive damages, along with a House debate over damages for government negligence, to suggest that Congress was explicit and vocal when it was concerned about the inclusion of a certain type of damages in the Act. (254) Since Congress did not explicitly or vocally express similar concern over damages merely in excess of out-of-pocket losses, the Johnson court refused to believe that it should read such concern into the Act. (255)

Inferring a broad interpretation of actual damages from the legislative history related to punitive damages is perhaps a reasonable suggestion, but it is no more than that--a suggestion. One could argue on the opposite side that the decision to remove punitive damages after serious consideration reflects anxiety about excessive government liability, or that the legislative history did not express concern for nonpecuniary damages because they were simply not on the table. At the risk of condemning implicit arguments about legislative history in general, it seems that Mr. Lodge and the Johnson court are just guessing here.

Once the internal cameral debates over punitive damages and other matters were complete, the House and Senate had to reconcile two bills with significant differences, some of which were found in the two damages provisions. In the Fitzpatrick case, the Eleventh Circuit thrust the compromise involved in this reconciliation into the center of the debate over actual damages.

The court in Fitzpatrick asserted that there was an "obvious quid pro quo" (256) in the compromise over damages. The House's final pre-compromise damages provision allowed recovery for only actual damages and required a plaintiff to demonstrate "willful, arbitrary or capricious" government action to recover. (257) The final Senate pre-compromise version, on the other hand, allowed for both "actual and general damages" and required merely government negligence for recovery. The House version was thus more restrictive with respect to both the types of recoverable damages and the government conduct required for recovery. (258) According to the Fitzpatrick decision, the "obvious quid pro quo" involved the House agreeing to a less restrictive standard of conduct--"willful or intentional" instead of "willful, arbitrary or capricious"--in return for the Senate conceding to the House's desire to limit damages to "actual damages" rather than "actual and general damages." (259) Taking "general damages" to mean the full range of provable compensatory damages, the court in Fitzpatrick concluded that the Senate sacrificed the availability of such damages as part of this compromise, leaving only actual damages, which "as used in the Act cannot be synonymous with common law general damages, and must refer to pecuniary loss." (260)

The validity of the conclusions that the court in Fitzpatrick drew from this inferred compromise turns entirely on how one understands the Senate's use of "general damages." The elimination of general damages via compromise suggests that the full range of provable compensatory damages is not recoverable under the Act only if general damages is properly understood to refer to that full range of provable damages. Although the Fitzpatrick decision offered some support for this understanding, (261) it has not gone uncontested. Most notably, the court in Johnson argued that the "general damages" removed from the Senate's final version of the bill were not a type of provable damages. Rather, "general damages" as used by the Senate meant "presumed damages." (262) In support of this position, the court noted that "[t]he first definition of 'general damages' in Black's [Law Dictionary] is 'such as the law itself implies or presumes to have accrued from the wrong complained of." (263) Additionally, the court noted that the Senate's assumed awareness of the expansive use of "actual injury" in Gertz v. Robert Welch is inconsistent with the conclusions drawn from the supposed compromise in Fitzpatrick. (264)

One could level a broader objection against this compromise argument--that the compromise did not even exist, at least not as described in Fitzpatrick. The court in Fitzpatrick inferred the compromise from the changes made to the final versions of the House and Senate bills, but "[t]here is no reference ... in the legislative history to such a compromise." (265) In fact, Mr. Lodge suggests that it is just as likely that, if there was a compromise, it was smaller than the court in Fitzpatrick suggested and offers no indication of how to interpret actual damages. Although the court in Fitzpatrick claimed that one side gave way on the scope of damages while the other gave way on the standard of conduct, Mr. Lodge notes that the final standard of conduct itself was actually a compromise. (266) The agreed upon "willful or intentional" standard was a compromise between the House's "willful, arbitrary, and capricious" standard and the Senate's position that any breach--intentional or not--should be a violation. (267)

Mr. Lodge is not alone in this view. The court in Johnson agreed, citing a telling statement by Senator Ervin. (268) Senator Ervin stated:
   The standard for recovery of damages under the House bill
   would have rested on the determination by a court that the
   agency acted in a manner which was willful, arbitrary, or
   capricious. The Senate bill would have permitted recovery
   against an agency on a finding that the agency was negligent
   in handling his records.


These amendments represent a compromise between the two positions.... (269)

Although Senator Ervin's statement is not dispositive on the nature of any compromise between the two houses of Congress, it is stronger evidence than that mustered by the court in Fitzpatrick or any other authority that has relied on a theory of compromise to ascertain the meaning of actual damages. Because the compromise suggested by Senator Ervin and the court in Johnson offers no insight into the breadth of available damages, it seems prudent to draw no conclusions about actual damages on the basis of any possible compromise.

h. Legislative History and the Privacy Protection Study Commission

A final argument based on legislative history, not easily categorized as either "explicit" or "implied," centers around the Privacy Protection Study Commission (PPSC), a group established by the Privacy Act to advise Congress on privacy issues. (270) During the hearings on the Act, Senator Ervin described the role of the PPSC as follows: "[T]he commission is directed to examine certain issues which are not included in the compromise between the House and Senate bill [sic], such as ... a question of whether the federal government should be liable for general damages." (271) Interpreting "general damages" to mean the full extent of compensatory damages, pecuniary or otherwise, the court in Fitzpatrick argued that Senator Ervin's statement is a strong indication that Congress did not intend the Privacy Act to include such a broad range of damages when enacted. (272) The court offered strong support for this position, or at least support that the PPSC itself believed this to be the case, by quoting a PPSC report saying that "Congress meant to restrict recovery to specific pecuniary losses until the [PPSC] could weigh the propriety of extending the standard of recovery." (273)

As was discussed earlier, (274) however, the court's interpretation in Fitzpatrick of general damages is controversial. The court in Johnson argued that the general damages that Senator Ervin expected the PPSC to study were not broad compensatory damages, but rather presumed damages. (275) Moreover, although the PPSC interpreted Senator Ervin to refer to broad compensatory damages, its interpretation is not conclusive, as the PPSC had no special insight into what the senator had in mind. (276) While the court in Fitzpatrick was right to assert that the PPSC interpreted Senator Ervin's statement to mean that the Privacy Act was not meant to allow for nonpecuniary damages when enacted, the court in Johnson rightly pointed out that the PPSC's uninformed understanding of one senator's words is far from ironclad proof of what the Act's drafters had in mind. Ultimately, the PPSC's discussion offers little interpretative value.

3. Analogies to Other Statutes

The Ninth Circuit in Cooper was not the first court to consider interpretations of analogous statutes when interpreting the "actual damages" language in the Privacy Act. Most notably, in Johnson, the Fifth Circuit presaged the Ninth Circuit in pointing to the interpretations of the "actual damages" language of the FCRA that allow recovery for nonpecuniary harms, also noting the similar underlying purposes of the two statutes. (277) The court in Johnson also pointed out that Congress was aware of the FCRA and had it in mind when debating the terms of the Privacy Act, as evidenced by at least two mentions of the FCRA during the Privacy Act debates. (278)

Unlike the court in Cooper, however, the court in Johnson recognized that "it would not be fair to charge Congress with knowledge of the case law construing 'actual damages' under the Fair Credit Reporting Act" (279) considering that "[f]or the most part, the cases [interpreting "actual damages" in the FCRA] were decided after the Privacy Act became law." (280) The court in Johnson could identify only one district court case, Millstone v. O'Hanlon Reports, Inc., (281) decided before the Privacy Act was passed on December 31, 1974, that broadly construed actual damages under the FCRA. (282) Even Millstone was decided just two months before the Act was passed, while Congress was already in the middle of its debates. The court's finding in Johnson that the FCRA analogy was "persuasive" (283) seems difficult to support given its acknowledgment that it was unfair to assume Congress had knowledge of later or concurrent FCRA interpretations. If one's goal is to determine the intent of Congress, then once one acknowledges that Congress cannot be charged with knowledge of interpretations of a similar statute, it does not seem sensible to give any weight to those interpretations.

In Johnson, the court also analogized between the Privacy Act and the Fair Housing Act (FHA), (284) noting that the Seventh Circuit and the Tenth Circuit had both interpreted "actual damages" language in the FHA to include recovery for emotional distress and humiliation. (285) Unlike the FCRA cases, the FHA cases were decided before the Privacy Act debates in Congress. (286) In the FHA context, however, the court in Johnson could not offer any discussion of the FHA during the Privacy Act debates to suggest that Congress was actually aware of the FHA interpretations. The court instead relied merely on the assertion that "[i]t is appropriate to charge Congress in the exercise of its legislative function with knowledge of ... federal court precedents." (287) Moreover, the court did not allege that Congress had passed this analogous statute with the same purpose in mind as in the case of the Privacy Act. The FHA was passed as part of a "coordinated scheme of federal civil rights laws enacted to end discrimination." (288) The "overarching goal of the FHA [is] to allow protected individuals ... 'to live in the residence of their choice in the community." (289) In the FHA case, then, the analogous statutory argument loses much of its force because--unlike in the FCRA case--the FHA is not a statute with the protection of private information as a primary purpose. (290) Congress may have meant one thing by actual damages in the context of a statute designed to prevent discrimination and something else entirely in the context of a statute designed to protect information privacy. Neither the FCRA analogy nor the FHA analogy is particularly helpful or convincing.

4. Recap: Actual Damages Under the Standard Modes of Interpretation

The text of the damages clause offers no clear indication of how to interpret "actual damages." Although courts and scholars prior to the Cooper case placed significant emphasis on legislative history, this has proven to be a "slippery and murky path [and] neither [side has] presented the history in a persuasive manner adequate to determine a clear victor." (291) The Ninth Circuit acknowledged this fact in its Cooper decision.

While I sympathize with the Ninth Circuit and commentators in wanting to emphasize the privacy-protecting purposes of the Act in deciding the types of damages that should be available, it seems to me that the purposive arguments are quite murky as well. It is true that deterrence and compensation were key goals of the Act, but discussion of limited government liability is also prevalent in the congressional debates. Although at least one commentator has found the purposive arguments convincing even after acknowledging the lack of value in the legislative history, (292) the various purposive arguments only offer thumbs on both sides of the scale, not a decisive answer. Even if deterrence and compensation were the key purposes of the Privacy Act, it does not follow that other purposes should be disregarded or that Congress intended to further those purposes by making available an expansive range of damages under the Act. There is no evidence that draws a clear connection between the goals of deterrence and compensation and the "actual damages" language of the damages provision. Analogies to other statutes seem promising at first, but there is no case in which a statute with both a similar purpose to the Privacy Act and the same "actual damages" language was interpreted expansively prior to the enactment of the Privacy Act.

The best answer, which should come as no surprise given the conditions under which the Act was passed, (293) is probably that there is legitimate ambiguity here, even after all standard modes of interpretation are considered. Thus, even under Richlin's characterization of the narrow construction canon as a last resort, to be used when the standard tools of interpretation have been exhausted, this is an issue that could very well be decided by narrow construction. If legitimate ambiguity remains, the canon dictates that actual damages be construed narrowly in favor of the government. That is, the canon dictates excluding nonpecuniary harms. The district court in Cooper may have come to exactly the right result, even under the Richlin approach.

Of course, courts often proclaim that statutory language is unambiguous when it would seem apparent to any uninterested party that reasonable minds could differ over interpretation. Under administrative law's Chevron doctrine, for instance, courts often decide cases at "step one," ruling that the statutory language at question is unambiguous, even in cases in which there is a dissenting opinion that strongly disagrees with the majority's interpretation or in which other courts have expressed mixed views about how to interpret the language. (294) Ultimately, even when the doctrine specifies that judges should rely exclusively on statutory text only if the text is unambiguous, courts still sometimes seem to interpret statutes as they normally would. This judicial whimsy layers yet another level of uncertainty on top of what is already a muddled picture in the Privacy Act context.

IV. DOE V. CHAO

This Part offers a brief discussion of the Supreme Court's 2004 decision in another case about the Privacy Act damages clause, Doe v. Chao. (295) This decision is worth discussing here for two reasons. First, its holding that the statutory minimum damages are available only if the plaintiff can make some showing of actual damages makes the ambiguity about the scope of actual damages under the Act even more significant and troubling. Second, the Court's methods in Chao point toward a possible decision in favor of the narrow reading of actual damages should the Cooper question reach the Court, a real possibility given the circuit split and Judge O'Scannlain's strong dissent to the denial of rehearing en banc in Cooper.

In Chao, the Court faced a different issue related to the Privacy Act damages provision. Specifically, the question presented was whether a showing of actual damages--however defined--was necessary for a plaintiff to collect the statutory minimum damages of $1000. (296) The circuit courts were divided on this issue, (297) with the majority believing that no such showing was necessary. (298)

In a majority opinion by Justice Souter, the Court sided with the minority of the circuit courts, holding that a showing of actual damages is in fact necessary to collect the statutory minimum damages. (299) Justice Ginsburg filed a dissent, joined by Justices Stevens and Breyer, and Justice Breyer also filed a dissent of his own. Even though the Court was briefed extensively on the role of the narrow construction canon, (300) none of the opinions even alluded to sovereign immunity. The majority--like most of the lower courts and unlike any court trying to interpret actual damages--asserted that the text of the damages provision was dispositive, (301) considering nontextual points only to respond to points raised by the dissenters and the petitioner. (302) In this Part, I first discuss how the Chao holding compounds the possible negative results of a narrow reading of actual damages. Then, I explain why the Court's methodology in Chao might suggest such a narrow reading if the Cooper question were to reach the Court.

A. Combining the Chao Holding with a Narrow Reading of Actual Damages

Even taken on its own, the Chao holding is a significant hindrance to the collection of damages under the Privacy Act. According to one commentator, "[t]he Supreme Court's narrow construction of the Privacy Act [in Chao] will force individuals to overcome unrealistic hurdles." (303) Whereas before Chao it seemed that just demonstrating the government's willful and intentional mistreatment of personal information and claiming adverse effect might be enough to obtain the $1000 statutory minimum damages, Chao made it clear that something more than that is needed. Given the expense of bringing a lawsuit and the difficulty of proving damages (even some types of pecuniary damages), the requirement of a showing of actual damages--even if the term is broadly interpreted--in effect means that many small injuries will go unremedied because the expense of bringing the suit exceeds the expected recovery. According to critics of Chao, the Act's private enforcement mechanism was weak before the Court's holding, and this decision only made matters worse. (304)

A narrow reading of actual damages would compound the damage-limiting nature of the Chao decision. Although nonpecuniary damages are often not the easiest damages to demonstrate, it is not hard to imagine a case where they might be. For instance, in the case of an identity theft allegedly caused by the government's misuse of personal information, it might be easy to demonstrate that the misuse of information caused hours of stress and worry for the victim. At the same time, it might be very difficult to prove a causal relationship between the government's misuse of the personal information and the pecuniary losses resulting from the alleged identity theft. The victim would have to either demonstrate that the government's misuse of information led to the information falling into the wrong hands or prove the negative fact that neither the victim nor anyone else with access to the relevant information ever put the information's security in jeopardy. In such a case, Chao would require that the victim demonstrate actual damages to recover even the $1000 statutory minimum damages. Under a narrow reading of actual damages, simply showing stress or worry would not suffice even to open the door.

Even if the Privacy Act was not intended to offer unlimited private enforcement, as the requirements for recovery build up, the private enforcement mechanism risks being completely undermined. The text of the Act makes it clear that a victim must clear the difficult hurdle of showing willfulness or intent, and Chao clarified that a showing of actual damages is required for any recovery. Further constricting recovery by defining actual damages to include only pecuniary damages would risk crippling the Act's private enforcement mechanism. As Daniel Solove has said, it could make a mockery of the Act. (305)

B. The Chao Methodology and Its Possible Implications

In Chao, Justice Souter's majority opinion found the text of the damages provision to be unambiguous and thus dispositive. The opinion did not need to include any discussion of the sovereign immunity canon because it is unnecessary to distinguish between various approaches to the canon when the text is unambiguous. Justice Ginsburg's dissent also found the text to be unambiguous, but the clear meaning that she found was completely different from Justice Souter's reading. In this section, I first explain the two contradictory readings of the "unambiguous" text, and then discuss what the Court's methodology in Chao might mean should certiorari be granted on the Cooper question.

1. Justice Souter on the Text

Justice Souter's majority opinion in Chao argues that textual analysis demonstrates the need to show actual damages in order to collect the statutory minimum of $1000. (306) Since it is helpful to have the text of the damages provision in front of you when considering the arguments that follow, I reproduce it here.
   In any suit brought under the provisions of subsection
   (g)(1)(C) or (D) of this section in which the court determines
   that the agency acted in a manner which was intentional or
   willful, the United States shall be liable to the individual in an
   amount equal to the sum of--(A) actual damages sustained
   by the individual as a result of the refusal or failure, but in no
   case shall a person entitled to recovery receive less than the
   sum of $1,000; and (B) the costs of the action together with
   reasonable attorney fees as determined by the court. (307)


The essence of Justice Souter's argument is that the clause "a person entitled to recovery" refers to a person who has demonstrated actual damages. Thus, the provision that "in no case shall a person entitled to recovery receive less than the sum of $1,000," taken in full, simply means that no person who has demonstrated actual damages should receive less than $1000, even if the amount of actual damages demonstrated is a smaller figure. (308) Justice Souter reads the first part of the provision, everything up to the word "failure," to set forth the full extent of the group of individuals who may recover--those who have sustained actual damages as a result of qualifying government action--and he reads the rest of the provision to set a minimum recovery of $1000 for any individual in that group. Under Justice Souter's reading, the $1000 minimum offers no help to those who cannot even show actual damages. (309)

Although Justice Souter insists that the language speaks for itsell he went on to make additional textual arguments, asserting that the damages provision could easily have been worded differently if its drafters had meant for the statutory minimum damages to be available in the absence of a showing of actual damages. For instance, he argued that the phrase "person entitled to recovery" could have been introduced prior to the mention of actual damages if the actual damages requirement were not meant to limit the group of eligible persons. (310) Alternatively, Justice Souter argued that the "person entitled to recovery" language could have been omitted entirely if it was not meant to refer back to the actual damages requirement. (311) He objects to the assumption that the text was written with this superfluity, suggesting that the language simply could have allowed for the recovery of actual damages "but in no case ... less than the sum of $1,000" if the intent was to provide a damages floor not conditioned on a showing of actual damages. (312)

2. Justice Ginsburg on the Text

Like Justice Souter in the majority opinion, Justice Ginsburg began with the text, arguing that it unambiguously supports her position. She argued that the "person entitled to recovery" language is most sensibly read to include anyone who experienced an adverse effect resulting from qualifying government action. (313) Under her reading, the beginning of the provision--everything up to the word "willful"--creates two conditions for government liability: (1) adverse government action, and (2) willfulness or intent. The remainder of the provision does not create additional conditions for recovery, but rather specifies the consequences if the two conditions set forth above are met. (314) In defense of this position, Justice Ginsburg pointed out that the language immediately after "willful" is "shall be liable." By interpreting the "person entitled to recovery" language in the next clause as an additional condition for recovery of any damages, she argued the majority turned "shall be liable" into "may be liable," disregarding what the text actually says. (315) She also argues that the majority turned "person entitled to recovery" into "person entitled to actual damages" for no good reason. (316) The "person entitled to recovery" language suggests greater breadth than "person entitled to actual damages" would, and there is no reason to equate the two phrases. (317)

Like Justice Souter, Justice Ginsburg suggested that Congress easily could have phrased the damages provision differently if it had intended a different meaning. Rather than providing that an adversely affected individual is entitled to "actual damages ... but in no case shall a person entitled to recovery receive less than ... $1,000," the adversely affected individual could have been said to be entitled to "actual damages ... but in no case shall a person who proves such damages [in any amount] receive less than $1,000." (318)

Again mirroring the majority opinion, Justice Ginsburg made a superfluity argument. She argued that the language "person entitled to recovery" does no work under the majority's reading, as evidenced by her aforementioned alternative formulation. (319) Moreover, she suggested that the majority's interpretation renders superfluous certain language found in another provision of the Privacy Act that the damages provision directly references. The damages provision, on its own terms, applies to "any suit brought under the provisions of subsection (g)(1)(C) or (D) of this section [5 U.S.C. [section] 552a]." Subsection (g)(1)(D) allows for civil suit whenever the government "fails to comply with any provision of this section [552a], or any rule promulgated thereunder, in such a way as to have an adverse effect on an individual." (320) Justice Ginsburg argued that, under Justice Souter's reading, "the liability-determining element 'adverse effect' becomes superfluous, swallowed up by the 'actual damages' requirement." (321) If actual damages must be shown for any recovery to be made, the adverse effect requirement in g(1)(D) might as well have been left out.

The majority responded directly to this last superfluity argument, suggesting that "adverse effect" is a term of art used to delimit the group of individuals who have standing to sue, (322) but Justice Ginsburg dismissed this argument, deeming it worthy of only a footnote. (323) She found it unacceptable to interpret the statute in such a way that g(1)(D) expressly gives standing to a particular group of people only to have the damages provision tell a subset of that group that they in fact have no avenue to recovery. Under the majority reading, "the open door for plaintiffs like Buck Doe is an illusion: what one hand opens, the other shuts," (324) and Justice Ginsburg could not accept this interpretation as reasonable.

3. Possible Implications of Chao for Interpreting Actual Damages

In considering the implications of Chao for a possible Supreme Court decision on the question presented in Cooper, I should first note that the membership of the Court has turned over significantly since Chao was decided in 2004. Justices Souter and O'Connor and Chief Justice Rehnquist, all members of the Chao majority, are no longer on the Court. Justice Stevens, who dissented in Chao, has also retired. Unless the Cooper question reaches the Court in the very near future, it is perhaps also likely that Justice Ginsburg, the author of the primary dissent in Chao, will no longer be a member of the Court when the case is heard. (325) This significant turnover limits the predictive value of analyzing the methodology used in Chao, but it does not eliminate it. Given the similarities between the two cases--both are interpreting the same provision of the same statute and the same purpose-based arguments are made in both contexts--the Chao decision will almost undoubtedly be used to frame the Cooper question if it reaches the Court.

Of course, the Chao decision's usefulness for predicting how the Court will decide the Cooper question seems to be even more severely limited because Chao was decided on the basis of clear text. The majority found the text to unambiguously dictate its holding that actual damages must be demonstrated before the $1000 statutory minimum damages can be collected. Therefore, it did not have to wrestle with the question of how to apply the narrow construction canon, and it did not have to consider the purpose-based arguments regarding issues like broad private enforcement and deterrence. A decision on the Cooper question would hinge on the Court's approach to narrow construction, and, if the Court took the Richlin approach, also on the Court's views on the various relevant nontextual arguments. Because the Court did not deal in Chao with the narrow construction canon or engage with the nontextual arguments common to Chao and Cooper, it would seem hard to argue that Chao has any implications for an eventual decision on the Cooper question.

To cease the analysis here, however, would be to ignore reality to some extent. The Court was not blind to the purposive arguments against narrowing the scope of damages available under the Privacy Act when deciding Chao. In fact, Justice Ginsburg, though relying primarily on the text, made essentially all of the same purpose-based arguments in her Chao dissent that can be found in the Ninth Circuit decision in Cooper and the other circuit court opinions and academic works that favor a broad interpretation of the damages provision of the Act. (326) Justice Ginsburg also made arguments about other analogous statutes that are very similar to those arguments accepted by the Ninth Circuit in Cooper. (327)

Moreover, although both Justice Souter's majority opinion and Justice Ginsburg's dissenting opinion in Chao insisted that the text was dispositive, it is somewhat hard to imagine that they actually believed the text to be so clear that no consideration could be given to these purpose-based arguments. As a first point, I reiterate that the majority claimed that the text provided an unambiguous answer in one direction, while the dissent claimed that the same text provided an equally unambiguous but opposite answer. While the fact that the two blocs felt so strongly about conflicting interpretations of the text is not by itself enough to demonstrate that the text is in fact ambiguous, it is certainly suggestive. Still, it is worth considering whether either side simply took an untenable position.

In contending that the text is clear on its face, neither side proffers an unreasonable interpretation. The phrase "a person entitled to recover" can be read to refer back to the "actual damages" language, which can be understood as a condition for recovery. At the same time, the language up to "willful" can be understood as the exclusive home for recovery conditions, and what follows--including the "actual damages" language--can be read as a list of consequences if the recovery conditions are met. If one considers only what the language says, it is hard to see how one side can be declared a clear winner.

As for the other textual arguments, both sides suggest that the damages clause could easily have been written otherwise if the opposing viewpoint was the drafters' intent, and both sides also suggest that language that was included is superfluous under the opposing interpretation. This is a rare instance where everyone can be correct, and everyone in fact is correct. Regardless of the intended meaning, the damages provision could have been drafted better. Clarifying language could have been added, and unhelpful language could have been omitted. These arguments from the majority and the dissent in Chao drive home a key point: No matter what Congress was trying to express, it did not get the message across clearly.

Although both sides claim to rely on the text, it is hard to believe that anyone actually finds the text so unambiguous on the question decided in Chao as to feel compelled to ignore all other arguments presented. Thus, it seems fair to suggest that the majority in Chao did not completely ignore the purpose-based arguments made by Justice Ginsburg in dissent. Most likely, the majority simply did not find these arguments compelling enough to be decisive. The same can be said for Justice Ginsburg's analogies to other statutes. Considering that these are in large part the same types of arguments that the Ninth Circuit found convincing in Cooper, the lesson from Chao may be that the Supreme Court will not be as impressed with these arguments as the Ninth Circuit was.

If the Court grants certiorari on the Cooper question, even if it uses the Richlin approach, it might be less inclined to find clarity in purpose and analogous statutes than to find residual ambiguity; thus, it might rule in favor of the government. This will be true to the extent that the present Court sees eye-to-eye with the 2004 Court, or to the extent that the Cooper question is framed in light of the Chao decision and the Court does not want to rely on arguments similar to those to which the Court gave short shrift in 2004. (328)

CONCLUSION: A CALL FOR LEGISLATIVE REFORM

The scope of the damages provision of the Privacy Act will be a crucial determinant of the Act's continuing efficacy in ensuring that the government treats personal information appropriately. As the language of the Privacy Act now stands, the interpretation of one of its crucial terms--actual damages--is dependent on which approach courts use in applying a judicially-made canon, and how courts weigh various factors if they use the more liberal approach of applying that canon. The Supreme Court has already restricted the scope of available damages in its 2004 decision in Doe v. Chao, and the methodology used by the Court in that case offers at least some evidence that a further restriction may be forthcoming if the Court chooses to answer the question derided by the Ninth Circuit in Cooper in the coming years.

There is no reason to subject the efficacy of the Privacy Act to the whims of the courts. Even if the Supreme Court answers this question in the near future, there is no reason even to leave this decision to the highest court in the land. The Act protects important interests in information privacy, and even though knowledge of this purpose is of limited use to courts interpreting the damage provisions, it should be a guiding force behind legislative review and amendment of the Act. The fear and anxiety caused by government mistreatment of personal information is a real harm, and the ability of victims of such mistreatment to be compensated for such harm should not be dependent upon the interpretive whims of any court. This is particularly true in the Internet age, an age in which information spreads more quickly and reaches the wrong hands more swiftly than the Privacy Act's drafters could possibly have imagined. Identity theft and widespread permanent reputational harm are as real as pecuniary damages, and they should be treated as such.

An interpretation of the term "actual damages" that excludes recovery for nonpecuniary harms, especially when coupled with the requirement to show actual damages to obtain even the $1000 statutory minimum damages, would in fact render the Privacy Act toothless. The world has not changed so much since the days of Watergate that the government should once again be given free rein to use personal information with no fear of repercussions, and Congress should act accordingly by amending the language of the damages provision of the Privacy Act to clearly allow for recovery on the basis of nonpecuniary harms.

(1.) Pub. L. No. 93-579, 88 stat. 1096 (codified as amended at 5 U.S.C. [section] 552a (2006)).

(2.) 5 U.S.C. [section] 552a(e).

(3.) Id. [section] 552a(d).

(4.) Id. [section] 552a(d)(1).

(5.) Id. [section] 552a(d)(2).

(6.) Id. [section] 552a(d)(3).

(7.) Id. [section] 552a(g)(1).

(8.) In introducing the Senate version of the Privacy Act on May 1, 1974, Senator Ervin remarked:
   If we have learned anything in this last year of Watergate, it is
   that there must be limits upon what the Government can know about
   each of its citizens. Each time we give up a bit of information
   about ourselves to the Government, we give up some of our freedom.


120 CONG. REC. $6741 (daily ed. May 1, 1974) (statement of Sen. Ervin), reprinted in LEGISLATIVE HISTORY OF THE PRIVACY ACT OF 1974: SOURCE BOOK ON PRWACY 3, 4 (Joint Comm. on Gov't Operations ed., 1976) [hereinafter SOURCE BOOK]; see also Major John F. Joyce, The Privacy Act: A Sword and a Shield but Sometimes Neither, 99 MIL. L. REV. 113, 122 (1983) ("The Watergate scandal which created a political crisis unparalleled in the history of this country also kindled a firestorm of interest in the protection of personal privacy from governmental intrusions.").

(9.) For Congressional discussion of the concerns about increased computerization, see, for example, S. REP. NO. 93-1183, at 1-2 (1974), reprinted in SOURCE BOOK, supra note 8, at 154-55. See also Haeji Hong, Dismantling the Private Enforcement of the Privacy Act of 1974: Doe v. Chao, 38 AKRON L. REV. 71, 80-81 (2005) (describing the public and congressional concerns over increased electronic surveillance that led to the passing of the Privacy Act, with detailed references to the congressional record).

(10.) Letter from Marc Rotenberg, Exec. Dir., Elec. Privacy lnfo. Ctr., and Chris lay Hoofnagle, Deputy Counsel, Elec. Privacy Info. Ctr., to Adam Putnam, Chair, House Gov't Reform Subcomm. on Tech., Info. Policy, Intergovernmental Relations & the Census, and William Clay, Ranking Member, House Gov't Reform Subcomm. on Tech., Info. Policy, Intergovernmental Relations & the Census (Mar. 25, 2003), http://www.epic.org/privacy/profiling/datamining3.25.03.html ("In 1974, the Congress, with broad bipartisan support, enacted comprehensive legislation [the Privacy Act] to prevent precisely the type of data profiling that is now under consideration by several federal agencies."); see also S. REP. NO. 93-1183, at 10 (1974), reprinted in SOURCE BOOK, supra note 8, at 163 ("[T]he concern for privacy is a bipartisan issue and knows no political boundaries.").

(11.) For details on the time pressure and frenzy surrounding the passage of the Privacy Act, see, for example, Joyce, supra note 8, at 122-23.

(12.) Hong, supra note 9, at 86 ("[B]ecause of Congress' swift passage of the Privacy Act, there are no detailed legislative comments or committee reports to explain many of the key provisions...."); see also ANITA L. ALLEN, PRIVACY LAW AND SOCIETY 502 (2007) (quoting the Department of Justice website as saying that "the Act's imprecise language, limited legislative history, and somewhat outdated regulatory guidelines have rendered it a difficult statute to decipher and apply").

(13.) See Hong, supra note 9, at 72 n.6 (listing some sources that have criticized the Privacy Act since it became law).

(14.) 5 U.S.C. [section] 552a(g)(4) (2006).

(15.) Id. Subsection g(1)(C), referred to in the damages provision, allows individuals aggrieved by certain government failures to comply with the Act to bring civil suit. Id. [section] 552a(g)(1)(C). Subsection g(1)(D), also mentioned, allows for civil suit whenever the government "fails to comply with any other provision of this section, or any rule promulgated thereunder, in such a way as to have an adverse effect on an individual." Id. [section] 552a(g)(1)(D). Essentially, any government failure to comply with its responsibilities under the Act allows the adversely affected individuals to bring civil suit.

(16.) The idea of limiting damages for privacy violations to pecuniary or out-ofpocket losses may seem obviously inappropriate to some, grounded as privacy is in concepts of emotional space, personality, autonomy, and dignity. The very origin of the right to privacy in American jurisprudence lies in concerns over the protection of "an inviolate personality." Samuel D. Warren & Louis D. Brandeis, The Right to Privacy, 4 HARV. L. REV. 193, 205 (1890). According to Robert Post:
   [T]he distinction [Warren and Brandeis] were most concerned to
   press involved the question of damages.... By reconceptualizing
   personality in terms of emotional integrity, Warren and Brandeis
   desired to make damages available ... for distress and anguish.
   They thus advocated that damages be seen as compensation for "the
   value of mental suffering."


Robert C. Post, Rereading Warren and Brandeis: Privacy, Property, and Appropriation, 41 CASE W. RES. L. REV. 647, 664-65 (1991) (quoting Warren & Brandeis, supra, at 213). Courts, at times, seem to forge an inseparable link between privacy violations and nonpecuniary damages. Over twenty years before Congress passed the Privacy Act, an Illinois court said: "Basically, recognition of the right to privacy means that the law will take cognizance of an injury, even though no right of property or contract may be involved and even though the damages resulting are exclusively those of mental anguish." Eick v. Perk Dog Food Co., 106 N.E.2d 742, 745 (Ill. App. Ct. 1952). Nonetheless, considering the conceptual muddle that is theoretical privacy scholarship, it seems best to set this broad conceptual issue aside and focus on the Privacy Act, not "privacy" writ large.

(17.) See Johnson v. Dep't of Treasury, 700 F.2d 971 (5th Cir. 1983).

(18.) Compare Fanin v. U.S. Dep't of Veterans Affairs, 572 F.3d 868, 872 (11th Cir. 2009) (holding that "[o]btaining monetary damages under [the Act] requires proof of 'actual damages," and in this circuit that means pecuniary losses"), and Hudson v. Reno, 130 F.3d 1193, 1.207 (6th Cir. 1997) (holding that "actual damages under the Privacy Act do not include recovery for 'mental injuries, loss of reputation, embarrassment or other nonquantifiable injuries'"), and Fitzpatrick v. IRS, 665 F.2d 327, 328 (11th Cir. 1982) (holding that "damages under the Privacy Act are recoverable only for proven out-of-pocket losses"), with Johnson, 700 F.2d at 972 (holding that "actual damages" include damages for mental injuries). There are many district court decisions on both sides of this debate as well. Compare, e.g., DiMura v. FBI, 823 F. Supp. 45, 47 (D. Mass. 1993) (holding that "'actual damages' does not encompass emotional damages"), and Albright v. United States, 558 F. Supp. 260, 264 (D.D.C. 1982) (holding that "[a]ctual damages under the Privacy Act are limited to 'out-of-pocket' expenses and do not include damages for emotional trauma, anger, fright or fear"), with Rice v. United States, 211 F.R.D. 10, 13 (D.D.C. 2002) (holding that "emotional trauma alone is sufficient to qualify as an 'adverse effect'" and "that the Privacy Act does not require proof of pecuniary, economic, or 'spedal' damages").

(19.) Doe v. Chao, 540 U.S. 614, 627 n.12 (2004).

(20.) See id. at 634 (Ginsburg, J., dissenting); see also Daniel Solove, The Nature of Privacy Harms: Financial and Physical Harm vs. Emotional and Mental Harm, CONCURRING OPINIONS (Jan. 15, 2010, 8:01 AM), http://www.concurringopinions.com/archives/2010/01/the-nature-of-privacy- harms-financial-and-physical-harm-vs-emotionaland-mental-harm.html ("Privacy Act violations often involve harms that are not akin to traditional types of injuries. Privacy harms caused by misuse or improper dissemination of information are more abstract in nature and often can't be directly linked to financial losses or physical injury.").

(21.) See Hong, supra note 9, at 107q)8.

(22.) See, e.g., David S. Ardia, Reputation in a Networked World: Revisiting the Social Foundations of Defamation Law, 45 HARV. C.R.-C.L.L. REV. 261,262 (2010) ("Reputation is more enduring because information about us, whether good or bad, can exist--and be easily retrievable - forever. Powerful search engines scour and index photos, videos, and text. Semantic connections link previously disparate pieces of information to individuals and to each other. In the past, much personal information was publicly inaccessible because of practical impediments to its access. The Internet is largely eliminating these impediments."). Not only is reputational harm more permanent and wide-reaching in a networked world, but it also is less susceptible to traditional legal protections. See id. at 263 ("[T]raditional approaches to protecting reputation that were blunt and ineffective before the networked self arose ... are even less effective today."). Moreover, even if our legal system has a mechanism to compensate those whose reputations are inappropriately injured, the actual corrective information offered by the system may not be widely disseminated. See id. at 306 ("Because our judicial system largely exists outside the network, its ability to influence our collective view of the world and of each other is limited."). For more on the relationship between the internet and reputational harm, see generally DANIEL J. SOLOVE, THE FUTURE OF REPUTATION: GOSSIP, RUMOR, AND PRIVACY ON THE INTERNET (2007).

(23.) Anne S. Kimbol, The Privacy Act May Be Toothless, HEALTH L. PERSP., Sept. 2008, at 4, available at http://law.uh.edu/healthlaw/perspectives/.

(24.) Solove, supra note 20.

(25.) Cooper v. FAA, No. C07-1383VRW, 2008 U.S. Dist. LEXIS 116149 (N.D. Cal. Aug. 22, 2008).

(26.) The district court in Cooper offered a good description of the intrigue created by the facts of the case: "A good many laudable policies collide in the facts at bar. These include policies to ensure the safety of the nation's airways, to root out waste, fraud and abuse in the Social Security system and to secure personal privacy of citizens with a leitmotif of policies against discrimination." Id. at *1.

(27.) Dan Slater, Ct: Emotional Distress Not Enough To Recover for Privacy Act Violations, WSJ LAW BLOG (Aug. 28, 2008 9:13 AM), http://blogs.wsj.com/law/2008/08/28/ ct-emotional-distress-not-enough-to-recover-for-privacy-act-violations/.

(28.) Cooper, 2008 U.S. Dist. LEXIS 116149, at *2.

(29.) 14 C.F.R. [section] 61.3(c) (2010); see also 14 C.F.R. [section] 61.23(a)-(b) (2010) (describing the types of flight operations that do and do not require medical certificates).

(30.) Cooper, 2008 U.S. Dist. LEXIS 116149, at *2.

(31.) See 14 C.F.R. [section] 61.23(d) (2010).

(32.) Cooper, 2008 U.S. Dist. LEXIS 116149, at *2.

(33.) Id.

(34.) Id. at *3.

(35.) Id.

(36.) Id.

(37.) Id.

(38.) Id.

(39.) Id. at *3-4.

(40.) Id. at *4.

(41.) Id. at *5.

(42.) Id. at *6-8.

(43.) Id. at *10.

(44.) Section 552a(b) provides:
   No agency shall disclose any record which is contained in a system
   of records by any means of communication to any person, or to
   another agency, except pursuant to a written request by, or with
   the prior written consent of, the individual to whom the record
   pertains, unless disclosure of the record would ... [satisfy one of
   several exceptions].


5 U.S.C. [section] 552a(b)(2006).

(45.) See id. [section] 552a(b)(3).

(46.) Cooper, 2008 U.S. Dist. LEXIS 116149, at *28-29.

(47.) Id. at *29.

(48.) Id. at *34.

(49.) For a discussion of the history, critique, and usage of this canon of construction, see infra Part I.

(50.) Cooper, 2008 U.S. Dist. LEXIS 116149, at *33-34.

(51.) Id. at *34.

(52.) Id. at *32-33.

(53.) Id. at *33 (quoting Library of Cong. v. Shaw, 478 U.S. 310, 318 (1986)).

(54.) See Nicole M. Quallen, Recent Development, Damages Under the Privacy Act: Is Emotional Harm "Actual"?, 88 N.C.L. REV. 334, 341 (2009) (arguing that "the Cooper court neglected to note that the canon [of narrow construction of sovereign immunity waivers] has recently been questioned").

(55.) Cooper, 2008 U.S. Dist. LEXIS 116149, at *33-34. The two cases cited are Library of Congress v. Shaw, 478 U.S. 310 (1986), and Siddiqui v. United States, 359 F.3d 1200 (9th Cir. 2004).

(56.) 155 U.S. 163 (1894).

(57.) See generally Gregory C. Sisk, The Continuing Drift of Federal Sovereign Immunity Jurisprudence, 50 WM. & MARY L. REV. 517 (2008).

(58.) See Antonin Scalia, Common-Law Courts in a Civil Law System: The Role of United States Federal Courts in Interpreting the Constitution and Laws, in A MATTER OF INTERPRETATION 3 (Amy Gutmann ed., 1997).

(59.) 503 U.S. 30 (1992).

(60.) 518 U.S. 187, 190 (1996).

(61.) Id. at 192; Nordic Village, 503 U.S. at 33.

(62.) Lane, 518 U.S. at 192; Nordic Village, 503 U.S. at 37.

(63.) 553 U.S. 571 (2008).

(64.) Id. at 580.

(65.) Id. at 580-83.

(66.) Id. at 583-89.

(67.) Id. at 589-90.

(68.) Cooper v. FAA, 596 F.3d 538 (9th Cir. 2010).

(69.) Id. at 550.

(70.) Id. at 549.

(71.) See id.

(72.) 540 U.S. 614 (2004).

(73.) Id.

(74.) 155 U.S. 163 (1894). Multiple sources suggest that the Court's decision in Eastern Transportation Co. v. United States, 272 U.S. 675 (1927), is where the narrow construction doctrine began. See, e.g., Timothy J. Simeone, Rule 11 and Federal Sovereign Immunity: Respecting the Explicit Waiver Requirement, 60 U. CHI. L. REV. 1043, 1047 (1993); Ann M. Madden, Note, Lane v. Pena: How Federal Governmental Agencies Can Discriminate and Not Be Held Accountable, 7 WIDENER J. PUB. L. 143, 149 (1997). This is likely due to the fact that Eastern Transportation does not cite Schillinger or Price v. United States, discussed in the text accompanying notes 7882. For a sprawling history of the narrow construction canon's use, see Sisk, supra note 57.

(75.) 155 U.S. at 171.

(76.) Id. at 167.

(77.) Id. at 166. Justice Brewer cited no authority for this proposition, but perhaps that should come as no surprise. After all, as Justice Stevens has emphasized, "the doctrine of sovereign immunity is nothing but a judge-made rule." United States v. Nordic Village, Inc., 503 U.S. 30, 42 (1992) (Stevens, J., dissenting).

(78.) 174 U.S. 373 (1899).

(79.) Id. at 375-76.

(80.) Id. at 376.

(81.) 272 U.S. 675 (1927).

(82.) Id. at 686. The Court concluded in Eastern Transportation that its view on limiting waivers of sovereign immunity led to its prior decision in Blamberg Bros. v. United States, 260 U.S. 452 (1923), but Chief Justice Taft's opinion in Blamberg contains no clear statement of the sovereign immunity doctrine.

(83.) 282 U.S. 656 (1931).

(84.) Id. at 659.

(85.) United States v. Shaw, 309 U.S. 495, 500-01 (1940).

(86.) 312 U.S. 584, 590 (1941).

(87.) Other Supreme Court cases that insist on strict interpretation and statutory consent without explicitly ruling out nontextual modes of interpretation (and often make use of them) include: Library of Congress v. Shaw, 478 U.S. 310, 318 (1986); Ruckelshaus v. Sierra Club, 463 U.S. 680, 685-86 (1983); McMahon v. United States, 342 U.S. 25, 27 (1951); and United States v. N.Y. Rayon Importing Co., 329 U.S. 654 (1947).

(88.) 282 U.S. at 659.

(89.) 312 U.S. at 590 (emphasis added).

(90.) 324 U.S. 215 (1945).

(91.) Id. at 222 (citations omitted) (emphasis added).

(92.) See Indian Towing Co. v. United States, 350 U.S. 61, 68-69 (1955). One commentator has suggested that the Court's opinion in Middlesex County Sewerage Authority v. National Sea Clammers Ass'n, 453 U.S. 1 (1981), also evidences a lenient reading of the narrow construction doctrine. Madden, supra note 74, at 149 & n.50. This seems to be an incorrect reading of Justice Powell's opinion in National Sea Clammers. The question of sovereign immunity had already been resolved and was not before the Court. National Sea Clammers, 453 U.S. at 9-10 & n.14. Rather, the Court was considering whether any private right of action was available to the plaintiff under the statutes in question, and thus followed the standard private right of action doctrine, which focuses on congressional intent. Id. at 13.

(93.) Cass R. Sunstein, Interpreting Statutes in the Regulatory State, 103 HARV. L. REV. 405, 506 (1989).

(94.) Even including the time period when narrow construction was at its peak, after Nordic Village and Lane, some have argued that courts too often have stated the canon without explaining its meaning or orienting it with respect to other modes of interpretation. See Sisk, supra note 57, at 561.

(95.) 503 U.S. 30 (1992).

(96.) See Scalia, supra note 58, at 3.

(97.) Nordic Village, 503 U.S. at 37 (citations omitted).

(98.) See, e.g., Dellmuth v. Muth, 491 U.S. 223, 230 (1989) (Kennedy, J.) ("[E]vidence of congressional intent must be both unequivocal and textual.... Legislative history generally will be irrelevant to a judicial inquiry into whether Congress intended to abrogate the Eleventh Amendment."); Hoffman v. Conn. Dep't of Income Maintenance, 492 U.S. 96, 104 (1989) (White, J.) ("[L]egislative history generally will be irrelevant to a judicial inquiry into whether Congress intended to abrogate the Eleventh Amendment.... Similarly, the attempts ... to construe [the potential statutory waiver] in light of the policies underlying the Bankruptcy Code are unavailing. These arguments are not based in the text of the statute and so, too, are not helpful...." (internal citation and quotation marks omitted)); Atascadero State Hosp. v. Scanlon, 473 U.S. 234, 242 (1985) (Powell, J.) (refusing to accept reliance on legislative history because "Congress may abrogate the States' constitutionally secured immunity from suit in federal court only by making its intention unmistakably clear in the language of the statute").

(99.) Dellmuth, 491 U.S. at 227; see also Atascadero, 473 U.S. at 238 (asserting that the abrogation of state sovereign immunity implicates "the fundamental constitutional balance between the Federal Government and the States").

(100.) Pennhurst State Sch. & Hosp. v. Halderman, 465 U.S. 89, 99 (1984).

(101.) See Sandra Zellmer, Preemption by Stealth, 45 HOUS. L. REV. 1659, 1670 (2009) (describing Justice Scalia, former Chief Justice Rehnquist, and former Justice O'Connor as "[c]hampions of federalism"); see also Antonin Scalia, Foreword: The Importance of Structure in Consitutional Interpretation, 83 NOTRE DAME L. REV. 1417, 1417 (2008) (identifying federalism and separation of powers as the two principal "structural provisions of the Constitution").

(102.) Nordic Village, 503 U.S. at 42 (Stevens, J., dissenting).

(103.) Id. at 45.

(104.) John Paul Stevens, Is Justice Irrelevant?, 87 NW. U. L. REV. 1121 (1993).

(105.) Id. at 1128.

(106.) Id. at 1127. Here, Justice Stevens refers to Plato's Republic. Specifically, he has in mind Thrasymachus's position that "justice or right is simply what is in the interest of the stronger party." PLATO, THE REPUBLIC 77 (Desmond Lee trans., Penguin Books 2d ed. 1974); see Stevens, supra note 104, at 1 & nn. 1-2.

(107.) Stevens, supra note 104, at 1129. The article to which he refers is Antonin Scalia, Sovereign Immunity and Nonstatutory Review of Federal Administrative Action: Some Conclusions from the Public-Lands Cases, 68 MICH. L. REV. 867 (1970).

(108.) Stevens, supra note 104, at 1129 (quoting Scalia, supra note 107, at 869).

(109.) Id. (quoting Scalia, supra note 107, at 868).

(110.) 518 U.S. 187 (1996).

(111.) Id. at 210 (Stevens, J., dissenting).

(112.) WILLIAM N. ESKRIDGE, JR., PHILIP P. FRICKEY & ELIZABETH GARRETT, LEGISLATION AND STATUTORY INTERPRETATION 338 (2000).

(113.) Philip P. Frickey, Revisiting the Revival of Theory in Statutory Interpretation: A Lecture in Honor of Irving Younger, 84 MINN. L. REV. 199, 206 (1999).

(114.) Madden, supra note 74, at 165.

(115.) See DONALD L. DOERNBERG, SOVEREIGN IMMUNITY OR THE RULE OF LAW: THE NEW FEDERALISM'S CHOICE 124-28 (2005).

(116.) On "loose canons" generally, see William N. Eskridge, Jr., Textualism, The Unknown Ideal?, 96 MICH. L. REV. 1509, 1515, 1542-48 (1998) (reviewing Scalia, supra note 58).

(117.) ESKRIDGE ET AL., supra note 112, at 339.

(118.) See, e.g., KARL N. LLEWELLYN, THE COMMON LAW TRADITION: DECIDING APPEALS 521-35 (1960); Felix Frankfurter, Some Reflections on the Reading of Statutes, 47 COLUM. L. REV. 527 (1947); Henry J. Friendly, Mr. Justice Frankfurter and the Reading of Statutes, in FELIX FRANKFURTER: THE JUDGE 30 (Wallace Mendelson ed., 1964); Karl N. Llewellyn, Remarks on the Theory of Appellate Decision and the Rules or Canons About How Statutes Are To Be Construed, 3 VAND. L. REV. 395, 401-06 (1950); Robert J. Martineau, Craft and Technique, Not Canons and Grand Theories: A NeoRealist View of Statutory Construction, 62 GEO. WASH. L. REV. 1 (1993); Richard A. Posner, Statutory Interpretation--in the Classroom and in the Courtroom, 50 U. CHI. L. REV. 800, 805-22 (1983).

(119.) See Bradford C. Mank, Textualism's Selective Canons of Statutory Construction: Reinvigorating Individual Liberties, Legislative Authority, and Deference to Executive Agencies, 86 KY. L.J. 527, 558-64 (1998).

(120.) 514 U.S. 527 (1995).

(121.) Id. at 531.

(122.) Mank, supra note 119, at 562.

(123.) 514 U.S. at 536.

(124.) Id. at 541 (Scalia, J., concurring).

(125.) Id. at 541 (Rehnquist, C.J., dissenting) (internal quotation marks omitted).

(126.) 527 U.S. 212 (1999).

(127.) Id. at 222.

(128.) 553 U.S. 571 (2008).

(129.) Id. at 589-90.

(130.) Id. at 590.

(131.) See, e.g., Sydney Foster, Should Courts Give Stare Decisis Effect to Statutory Interpretation Methodology?, 96 GEO. L.J. 1863, 1872-84 (2008); Philip P. Frickey, Interpretive-Regime Change, 38 LOY. L.A.L. REV. 1971, 1976 (2005); Joseph A. Grundfest & A.C. Pritchard, Statutes with Multiple Personality Disorders: The Value of Ambiguity in Statutory Design and Interpretation, 54 STAN. L. REV. 627, 682 n.183 (2002); Morell E. Mullins, Sr., Tools, Not Rules: The Heuristic Nature of Statutory Interpretation, 30 J. LEGIS. 1, 68 (2003); Nicholas Quinn Rosenkranz, Federal Rules of Statutory Interpretation, 115 HARV. L. REV. 2085, 2144 (2002); Jonathan R. Siegel Judicial Interpretation in the Cost-Benefit Crucible, 92 MINN. L. REV. 387, 396 (2007); Jonathan R. Siegel, The Polymorphic Principle and the Judicial Role in Statutory Interpretation, 84 TEX. L. REV. 339, 385-86 (2005); Robert Weisberg, The Calabresian Judicial Artist: Statutes and the New Legal Process, 35 STAN. L. REV. 213, 213-14 (1983). For an argument that courts should give not just ordinary but "extra-strong" stare decisis effect to statutory interpretation methodology, see Foster, supra, at 1884-97.

(132.) See Foster, supra note 131, at 1866.

(133.) See id. at 1874.

(134.) Chickasaw Nation v. United States, 534 U.S. 84, 94 (2001).

(135.) See Foster, supra note 131, at 1865-66.

(136.) Id.

(137.) Id. at 1875-77. Foster cites Bowers v. Hardwick, 478 U.S. 186 (1986) and Lawrence v. Texas, 539 U.S. 558 (2000) as two noteworthy cases in which such a special justification was required. Foster, supra note 131, at 1875. A particularly relevant example mentioned by Foster of diverging from prevalent interpretation methodologies without requiring a special justification is the shift to requiring an unequivocal textual waiver of state sovereign immunity in the Eleventh Amendment context. Id. at 1879-80; see also supra notes 97-100 and accompanying text.

(138.) See Foster, supra note 131, at 1876, 1881.

(139.) A court need not even adopt either of these approaches. It can apply the canon however it sees fit, or not at all. For additional discussion of this last possibility in the Privacy Act context, see infra Part III.

(140.) Cooper v. FAA, No. C07-1383VRW, 2008 U.S. Dist. LEXIS 116149, at *15-26 (N.D. Cal. Aug. 22, 2008).

(141.) Id. at *31.

(142.) Id.

(143.) Id. at *32-33 (quoting Library of Cong. v. Shaw, 478 U.S. 310, 318 (1986)).

(144.) Id. at *34.

(145.) Cooper v. FAA, 596 F.3d 538, 549 (9th Cir. 2010).

(146.) Id.; see also Opening Brief for Plaintiff-Appellant Stanmore Cawthon Cooper at 48-50, Cooper v. FAA, 596 F.3d 538 (9th Cir. 2010) (No. 08-17074).

(147.) According to Gregory Sisk, the application of the narrow construction canon to a particular question has increasingly depended on whether the question engages the core issue of whether a statutory waiver exists at all. The basis for this alleged shift is that courts have increasingly viewed the narrow construction canon as a jurisdictional safeguard, preventing the government from being subjected to jurisdiction unless immunity has explicitly been waived. Jurisdiction is not generally implicated unless the question at hand is whether there is a waiver at all. See Sisk, supra note 57, at 522, 574-87. In the words of Professor Sisk: "As the distance grows between a statutory standard or limitation and the core substance of the waiver, presumptions in favor of the government fade and statutory construction assumes an ordinary shape." Id. at 522. For instance, in interpreting statutes of limitations--a context somewhat removed from the core question of whether a waiver exists or not--the Supreme Court has, in recent times, explicitly ignored the narrow construction canon on account of the theory that the canon was not meant to apply to each detail of a statutory immunity waiver. See id. at 581-87 (discussing Irwin v. Department of Veteran Affairs, 498 U.S. 89 (1990) and its progeny). A key potential counterexample is John R. Sand & Gravel Co. v. United States, 552 U.S. 130, 133-39 (2008), which relies on the narrow construction canon to rule against the equitable tolling of a statute of limitations. Professor Sisk has extensively discussed the discordance between John R. Sand and the general shift away from applying the canon in cases that do not deal with the core question of waiver existence. Sisk, supra note 57, at 587-605; see also Gregory C. Sisk, The Inevitability of Federal Sovereign Immunity, 55 VILL. L. REV. 899, 923-24 (2010) (reiterating Professor Sisk's views about the shift away from applying the strict construction canon in cases unrelated to the question of waiver existence). Applying Professor Sisk's framework to the question of interpreting "actual damages" under the Privacy Act is not perfectly straightforward. Although Professor Sisk's emphasis is on the question of waiver existence and the jurisdictional nature of this question, he does allow for the possibility of questions of remedy still being subject to narrow construction. In one formulation, Sisk asserts that "It]he traditional rule of strict construction in favor of the sovereign has become more attentively focused upon the general scope of the waiver in terms of the cause of action and remedy allowed against the government." Sisk, supra note 57, at 522 (emphasis added). This would seem to suggest that the actual damages question is still likely subject to narrow construction, but the best reading of Professor Sisk's discussion of applying the canon to questions of remedy is probably not so broad. In discussing what has become the mainstream use of the canon today, he cites Molzof v. United States, 502 U.S. 301 (1992), in which the Supreme Court refused to apply strict construction in interpreting the exclusion of punitive damages from the remedies available against the government under the Federal Tort Claims Act. Sisk, supra note 57, at 556-57. The Molzof case seems like a reasonable analogy for the "actual damages" question, so Professor Sisk's framework likely suggests that a court would not apply narrow construction to the question in Cooper. Of course, as was discussed in Part II, courts are not bound by their prior use of the canons, so Professor Sisk is just discussing an observed pattern--not anything precedential. Both the district court and the Ninth Circuit in Cooper diverged from the pattern he observed.

(148.) Cooper, 596 F.3d at 549.

(149.) Id.

(150.) Id. at 550.

(151.) Id. (quoting Chickasaw Nation v. United States, 534 U.S. 84, 94 (2001)).

(152.) Id.

(153.) Id. at 550-51.

(154.) See Carey v. Piphus, 435 U.S. 247, 258-59 (1978) ("[T]o further the purpose of [section] 1983, the rules governing compensation for injuries caused by the deprivation of constitutional rights should be tailored to the interests protected by the particular right in question.").

(155.) Cooper, 596 F.3d at 547.

(156.) Id. at 549.

(157.) Lane v. Pena, 518 U.S. 187, 195-96 (1996).

(158.) Id. at 196.

(159.) Id. at 197.

(160.) Id. at 197-200.

(161.) At least to the extent that the Ninth Circuit considered congressional intent and policy in the Cooper decision, it was also in keeping with past Ninth Circuit precedents. See, e.g., Hopi Tribe v. Navajo Tribe, 46 F.3d 908, 921-22 (9th Cir. 1995) (holding a narrow construction of a sovereign immunity waiver improper because of "the overriding congressional purpose behind" the legislation in question); United States v. Oregon, 44 F.3d 758, 766 (9th Cir. 1994) ("The Supreme Court has repeatedly looked to indicia of Congressional intent in order to construe the scope of the unequivocally expressed waiver of immunity in the McCarran Amendment."); In re Town & Country Home Nursing Servs., Inc., 963 F.2d 1146, 1151 (9th Cir. 1991) ("It is well established that when the federal government waives its immunity, the scope of the waiver is construed to achieve its remedial purpose."). The Ninth Circuit focused on this, among other considerations, when it eventually issued an order denying rehearing en banc. Cooper v. FAA, 622 F.3d 1016, 1020 (9th Cir. 2010).

(162.) Cooper v. FAA, 596 F.3d 538, 550 (9th Cir. 2010).

(163.) Id. at 549.

(164.) Id. at 544-45.

(165.) Id. at 545 (emphasis added by court) (quoting Pub. L. No. 93-579, [section] 2(b)(6), 88 Stat. 1896 (1974) (codified at 5 U.S.C. [section] 552a (2006))).

(166.) Id. at 545-46.

(167.) Id. at 546.

(168.) Id.

(169.) Id. (emphasis omitted) (quoting 5 U.S.C. [section] 552a(e)(10)).

(170.) Id. (emphasis omitted) (quoting 5 U.S.C. [section] 552a(g)(1)(C)).

(171.) Id.

(172.) Id. at 546-47.

(173.) 15 U.S.C. [section][section] 1681-1681x (2006).

(174.) Pub. L. No. 91-508, [section][section] 601-622, 84 Stat. 1114, 1228-36 (1970).

(175.) Cooper v. FAA, 596 F.3d 538, 548 (9th Cir. 2010) (citing Smith v. City of Jackson, 544 U.S. 228, 233 (2005), for the proposition that "[w]hen Congress uses the same language in two statutes having similar purposes, particularly when one is enacted shortly after the other, it is appropriate to presume that Congress intended that text to have the same meaning in both statutes").

(176.) Cooper v. FAA, 622 F.3d 1016 (9th Cir. 2010).

(177.) Id. at 1022 (O'Scannlain, J., dissenting from the order denying a rehearing en banc).

(178.) Id.

(179.) Id.

(180.) Id. at 1020 (Smith, J., concurring in the order denying a rehearing en banc).

(181.) Id.

(182.) One notable exception is DiMura v. FBI, 823 F. Supp. 45 (D. Mass. 1993), a case decided shortly after Nordic Village. In DiMura, the district court cited Nordic Village in construing "actual damages" under the Privacy Act narrowly on sovereign immunity grounds, ld. at 47-48.

(183.) See, e.g., Johnson v. Dep't of Treasury, 700 F.2d 971, 974 (5th Cir. 1983) (asserting that "'actual damages' has no plain meaning or consistent legal interpretation"); Fitzpatrick v. IRS, 665 F.2d 327, 329 (11th Cir. 1982) ("'[A]ctual damages' has no consistent legal interpretation."); Cooper v. FAA, No. C07-1383VRW, 2008 U.S. Dist. LEXIS 116149, at "30-31 (N.D. Cal. Aug. 22, 2008) (reviewing Ninth Circuit decisions to show that "actual damages" is textually ambiguous); DiMura, 823 F. Supp. at 47 ("IT]he phrase 'actual damages' in the Act is ambiguous."); Houston v. Dep't of Treasury, 494 F. Supp. 24, 30 n.12 (D.D.C. 1979) (stating that "actual damages" is not a generally well-defined phrase); Frederick Z. Lodge, Note, Damages Under the Privacy Act of 1974: Compensation and Deterrence, 52 FORDHAM L. REV. 611, 612 (1984) ("The term 'actual damages' ... has no generally accepted legal definition, and is not clearly defined in the Privacy Act.").

(184.) For a list of citations to varying interpretations of actual damages from different areas of the law, see Johnson, 700 F.2d at 985-86 n.42.

(185.) See Lodge, supra note 183, at 619-22.

(186.) Id. at 621-22.

(187.) In passing the Act, the legislature envisioned remedies resulting from civil suits brought by individual citizens as a "vital element of the Privacy Act's enforcement scheme." Id. at 622 (citing legislative history highlighting the centrality of individual civil enforcement). The Senate said of its original version of the Act that it was "designed to encourage the widest possible citizen enforcement through the judicial process." S. REP. NO. 93-1183, at 83 (1974), reprinted in SOURCE BOOK, supra note 8, at 236. This statement has been read by some as strong evidence in favor of a broad interpretation of actual damages, but the legislative history actually shows that the statement was about standing. Later in the same paragraph, elaborating on the point about "the widest possible citizen enforcement," the Senate Report explains that the intent is "to afford the widest possible standing consistent with the constitutional requirement of 'case or controversy' in Article III, Sec. 2 of the Constitution." Id.

(188.) See 5 U.S.C. [section] 552a(i) (2006).

(189.) Id.

(190.) See supra notes 20-21 and accompanying text.

(191.) The concern here is not just that a victorious suit will result in insufficient recovery or that too many cases with real harm will offer no potential recovery. Rather, an additional concern is that even those suits with the potential for some recovery will not offer enough to overcome the costs of bringing the suit. This problem is exacerbated by the rule that only individuals--not associations or groups--can bring suit under the Act. Am. Fed'n of Gov't Emps. v. Hawley, 543 F. Supp. 2d 44, 49 (D.D.C. 2008) ("[O]nly individuals have standing to bring Privacy Act claims."). The problem is exacerbated even further by the Supreme Court's decision in Doe v. Chao, 540 U.S. 614, 616 (2004), which required a showing of "actual damages" before statutory minimum damages can be recovered. This decision is discussed extensively in infra Part IV.

(192.) 665 F.2d 327 (11th Cir. 1982).

(193.) Id. at 330.

(194.) See, e.g., Lodge, supra note 183, at 626.

(195.) Id.

(196.) Id. at 632.

(197.) OFFICE OF INFO. & PRIVACY, U.S. DEP'T OF JUSTICE, FREEDOM OF INFORMATION 133-50 (1983).

(198.) Id.

(199.) See Hong, supra note 9, at 110. These statistics can be interpreted as indicating that courts are generally thwarting the congressional aim of providing protection and compensation.

(200.) 5 U.S.C. [section] 552a(e)(10) (2006).

(201.) Lodge, supra note 183, at 621-22.

(202.) Id.

(203.) 5 U.S.C. [section] 552a(e)(10).

(204.) Traditionally, privacy torts are divided into four categories: (1) intrusion upon seclusion; (2) appropriation of name or likeness; (3) disclosure of private information; and (4) publicity placing another in a false light. See RESTATEMENT (SECOND) OF TORTS [section] 652A(2) (1977). For the origin of this categorization, see generally William L. Prosser, Privacy, 48 CALIF. L. REV. 383, 388-89 (1960).

(205.) Hong, supra note 9, at 76; see West v. Media Gen. Convergence, Inc., 53 S.W.3d 640, 648 (Tenn. 2001) (holding, in the context of a false light claim, that "[t]he plaintiff need not prove special damages or out of pocket losses necessarily, as evidence of injury to standing in the community, humiliation, or emotional distress is sufficient").

(206.) See Orekoya v. Mooney, 330 F.3d 1, 9 (1st Cir. 2003).

(207.) See, e.g., Johnson v. Dep't of Treasury, 700 F.2d 971, 977 (5th Cir. 1983).

(208.) Id.

(209.) 418 U.S. 323 (1974).

(210.) 385 U.S. 374 (1967).

(211.) 418 U.S. at 349.

(212.) Id. at 349-50.

(213.) Id. at 350.

(214.) See, e.g., Lodge, supra note 183, at 627.

(215.) 385 U.S. at 387-88 ("We hold that the constitutional protections for speech and press preclude the application of the New York statute to redress false reports of matters of public interest in the absence of proof that the defendant published the report with knowledge of its falsity or in reckless disregard of the truth.").

(216.) Id. at 378-79, 387-88.

(217.) Id. at 385 n.9.

(218.) Johnson v. Dep't of Treasury, 700 F.2d 971, 977 (5th Cir. 1983).

(219.) Houston v. Dep't of Treasury, 494 F. Supp. 24, 30 n.16 (D.D.C. 1979).

(220.) 700 F.2d at 977.

(221.) 435 U.S. 247 (1978).

(222.) Id. at 259.

(223.) Johnson, 700 F.2d at 977.

(224.) See Lodge, supra note 183, at 628 (arguing for a broad reading of the damages provision because the Privacy Act vindicates a constitutional right and courts should be flexible and use the best remedies available when constitutional rights are at risk).

(225.) See, e.g., Griswold v. Connecticut, 381 U.S. 479, 485 (1965).

(226.) See Joseph L. Bierwirth, Jr., A Constitutional Right to Privacy in Confidential Information: Does it Exist in the Circuit Court?, 41 B.B.J. 4, 4 (1997) ("In some cases around the country, the existence of the so-called 'confidentiality branch' of the constitutional right to privacy has been presumed. In others, its existence has been rejected or questioned."). Compare A.L.A. v. West Valley City, 26 F.3d 989, 990 (10th Cir. 1994) (finding a constitutional right to confidential medical information), Doe v. City of New York, 15 F.3d 264, 267 (2d Cir. 1994) (finding a constitutional right to the confidentiality of information about HIV status), and United States v. Westinghouse Elec. Corp., 638 F.2d 570, 577 (3d Cir. 1980) (finding a constitutional right to the confidentiality of employee medical records), with, e.g., J.P. v. DeSanti, 653 F.2d 1080, 1089 (6th Cir. 1981) ("Absent a clear indication from the Supreme Court we will not construe isolated statements in Whalen ... more broadly than their context allows to recognize a general constitutional right to have disclosure of private information measured against the need for disclosure."), and DiMura v. FBI, 823 F. Supp. 45, 47 n.3 (D. Mass. 1993) ("[T]he constitutional right of privacy is not implicated by government disclosures of personal information." (citing Borucki v. Ryan, 827 F.2d 836, 842-43 (1st Cir. 1987))).

(227.) Hong, supra note 9, at 76; see also Bierwirth, supra note 226, at 4 ("Neither the United States Supreme Court nor the First Circuit Court of Appeals has ever squarely addressed the issue [of the existence of an information privacy branch of the constitutional right to privacy]. It remains an open question as to whether, if pressed, either would recognize such a right.").

(228.) 429 U.S. 589 (1977).

(229.) Id. at 598-600.

(230.) See, e.g., Russell v. Gregoire, 124 F.3d 1079, 1093-94 (9th Cir. 1997); Am. Fed'n of Gov't Emps. v. Dep't of Hous. & Urban Dev., 118 F.3d 786. 791 (D.C. Cir. 1997); Adams v. Drew, 906 F. Supp. 1050, 1054-55 (E.D. Va. 1995).

(231.) E.g., City of New York, 15 F.3d at 267; Westinghouse, 638 F.2d at 577.

(232.) Johnson v. Dep't of Treasury, 700 F.2d 971, 979-80 (5th Cir. 1983).

(233.) 120 CONG. REC. 36,659 (1974) (statement of Rep. McCloskey), reprinted in SOURCE BOOK, supra note 8, at 923.

(234.) Johnson, 700 F.2d at 979-80.

(235.) Id. at 980 n.22.

(236.) Id. Adding to the difficulty of proving even pecuniary damages is that the appropriate mental state--either willfulness or intent--must be shown.

(237.) 120 CONG. REC. 36,659 (1974) (statement of Rep. McCloskey), reprinted in SOURCE BOOK, supra note 8, at 921.

(238.) Johnson, 700 F.2d at 980.

(239.) 120 CONG. REC. 36,659 (1974) (statement of Rep. McCloskey), reprinted in SOURCE BOOK, supra note 8, at 921 ("[I]f he reveals information improperly, he may subject his agency to punitive damages. If he withholds the information, on the other hand, he is subject only to ordinary damages, attorneys' fees, and costs.").

(240.) 494 F. Supp. 24 (D.D.C. 1979).

(241.) Id. at 30 n.13.

(242.) 120 CONG. REC. 36,956 (1974) (statement of Rep. Eckhardt), reprinted in SOURCE BOOK, supra note 8, at 926.

(243.) Houston, 494 F. Supp at 30.

(244.) Johnson v. Dep't of Treasury, 700 F.2d 971, 980-81 n.24 (5th Cir. 1983).

(245.) Id.

(246.) Quallen, supra note 54, at 345.

(247.) More complete chronicles of the relevant legislative history are available elsewhere. See, e.g., Lodge, supra note 183, at 622-28.

(248.) Quallen, supra note 54, at 345.

(249.) For the original House bill, see H.R. 16,373, 93d Cong. (1974), reprinted in SOURCE BOOK, supra note 8, at 239. The clause allowing for recovery of punitive damages in the original House bill was section 552a(f)(3)(A)(ii). See SOURCE BOOK, supra, at 250. For the original Senate bill, see S. 3418, 93d Cong. (1974), reprinted in SOURCE BOOK, supra note 8, at 9. The clause allowing for recovery of punitive damages in the original Senate bill was section 304(b)(2). See SOURCE BOOK, supra note 8, at 27.

(250.) See Lodge, supra note 183, at 623-25.

(251.) Id. at 623. For part of this debate, see, for example, 120 CONG. REC. 36,658-60 (1974), reprinted in SOURCE BOOK, supra note 8, at 919-24.

(252.) Lodge, supra note 183, at 623.

(253.) Johnson v. Dep't of Treasury, 700 F.2d 971, 978-79 (5th Cir. 1983).

(254.) See id.

(255.) See id. at 980.

(256.) Fitzpatrick v. IRS, 665 F.2d 327, 330 (11th Cir. 1982).

(257.) See id.

(258.) See id.

(259.) Id.

(260.) Id.

(261.) Id.

(262.) Johnson v. Dep't of Treasury, 700 F.2d 971, 982 (5th Cir. 1983).

(263.) Id. at 982 n.32 (quoting BLACK'S LAW DICTIONARY 352 (5th Ed. 1979)).

(264.) Id. at 982.

(265.) Lodge, supra note 183, at 625.

(266.) Id.

(267.) Id.

(268.) Johnson v. Dep't of Treasury, 700 F.2d 971, 981-82 (5th Cir. 1983).

(269.) 120 CONG. REC. 40,406 (1974) (statement of Sen. Ervin), reprinted in SOURCE BOOK, supra note 8, at 862.

(270.) See Privacy Act of 1974, Pub. L. No. 93-579, [section] 5, 88 Stat. 1896, 1905-09.

(271.) 120 GONG. REC. 40,405 (1974) (remarks of Sen. Ervin), reprinted in SOURCE BOOK, supra note 8, at 859.

(272.) Fitzpatrick v. IRS, 665 F.2d 327, 330 (11th Cir. 1982).

(273.) PRIVACY PROT. STUDY COMM'N, PERSONAL PRIVACY IN AN INFORMATION SOCIETY 530 (1977).

(274.) See supra notes 262-66 and accompanying text.

(275.) Johnson v. Dep't of Treasury, 700 F.2d 971, 982 (5th Cir. 1983).

(276.) Id. at 983 n.33.

(277.) Id. at 983-84.

(278.) Id. at 984.

(279.) Id.

(280.) Id. at 984 n.37.

(281.) 383 F. Supp. 269 (E.D. Mo. 1974).

(282.) Johnson, 700 F.2d at 984 n.37.

(283.) Id. at 984.

(284.) 42 U.S.C. [section][section] 3601-3619 (2006).

(285.) Johnson, 700 F.2d at 983.

(286.) For example, the court in Johnson pointed to Smith v. Sol D. Adler Realty Co., 436 F.2d 344 (7th Cir. 1970), and Steele v. Title Realty Co., 478 F.2d 380 (10th Cir. 1973).

(287.) Johnson, 700 F.2d at 983 (citing Cannon v. Univ. of Chi., 441 U.S. 677 (1979)).

(288.) Huntington Branch, NAACP v. Town of Huntington, 844 F.2d 926, 935 (2d Cir. 1988).

(289.) Human Res. Research & Mgmt. Grp. v. Cnty. of Suffolk, 687 F. Supp. 2d 237, 260 n.21 (E.D.N.Y. 2010) (quoting City of Edmonds v. Wash. State Bldg. Code Council, 18 F.3d 802, 806 (9th Cir. 1994)); see also Brian Patrick Larkin, Note, The Forty-Year "First Step": The Fair Housing Act as an Incomplete Tool for Suburban Integration, 107 COLUM. L. REV. 1617, 1624-30 (2007).

(290.) It is worth noting, however, that the FHA is not without concern for privacy generally, or for information privacy specifically. The FHA features exemptions to its broad anti-discrimination rules for the sale and rental of housing when the seller or landlord would be Riving in the housing alongside or nearby the purchaser or tenant. See 42 U.S.C. [section] 3603(b) (2006). These exemptions are for the benefit of defendants, whereas the Privacy Act is concerned with the privacy of potential plaintiffs, but the exemptions do show concern for privacy, and to the extent that personal information can be gleaned from living in close proximity to someone, they reflect concern for information privacy. Still, information privacy is far removed from the overarching anti-discrimination policies behind the FHA and its sanctioned remedies.

(291.) Quallen, supra note 54, at 348.

(292.) See id.

(293.) See supra notes 8-11 and accompanying text.

(294.) See, e.g., Massachusetts v. EPA, 549 U.S. 497 (2007).

(295.) 540 U.S. 614 (2004).

(296.) Id. at 616.

(297.) Compare Doe v. Chao, 306 F.3d 170, 176-79 (4th Cir. 2002) (holding that the statutory minimum damages are available only to plaintiffs who suffered actual damages), and Hudson v. Reno, 130 F.3d 1193, 1207 (6th Cir. 1997) (citing failure to show actual damages as a reason to affirm dismissal of Privacy Act claims), with Orekoya v. Mooney, 330 F.3d 1, 5 (1st Cir. 2003) (describing actual damages and statutory minimum damages as available independent of one another), Wilborn v. Dep't of Health & Human Servs., 49 F.3d 597, 603 (9th Cir. 1995) (requiring no showing of damages to grant the $1000 statutory minimum), Waters v. Thornburgh, 888 F.2d 870, 872 (D.C. Cir. 1989) (stating that a plaintiff who has shown an intentional or willful failure by the government to protect privacy information "is entitled to the greater of $1,000 or the actual damages sustained"), Johnson v. Dep't of Treasury, 700 F.2d 971, 977 n.12 (5th Cir. 1983) (stating that the "statutory minimum of $1000, of course, is recoverable," even though no actual damages had been shown), and Fitzpatrick v. IRS, 665 F.2d 327, 330 (11th Cir. 1982) (explaining that the $1000 statutory minimum is to provide those with no "provable damages" with incentive to sue).

(298.) Chao, 540 U.S. at 631 (Ginsburg, J., dissenting) (describing the position that actual damages need not be shown to recover the statutory minimum damages as "the interpretation ... advanced by most courts of appeals").

(299.) Id. at 618 (majority opinion).

(300.) See, e.g., Brief for the Respondent at 17-25, Doe v. Chao, 540 U.S. 614 (2004) (No. 02-1377).

(301.) Chao, 540 U.S. at 620.

(302.) Id. at 624-27.

(303.) Hong, supra note 9, at 98.

(304.) Id. at 102-03.

(305.) Daniel Solove, The Nature of Privacy Harms: Financial and Physical Harm vs. Emotional and Mental Harm, CONCURRING OPINIONS (Jan. 15, 2010, 8:01 AM), http://www.concurringopinions.com/archives/2010/01/the-nature-of-privacy- harmsfinancial-and-physical-harm-vs-emotional- and-mental-harm.html.

(306.) Chao, 540 U.S. at 620.

(307.) 5 U.S.C. [section] 552a(g)(4) (2006).

(308.) Chao, 540 U.S. at 620.

(309.) At least one lower court has relied significantly on this straightforward textual argument in denying statutory minimum damages. In DiMura v. FBI, 823 F. Supp. 45, 48 (D. Mass. 1993), after determining that the plaintiff had not suffered "actual damages," the court said: "[g]iven that plaintiff is not 'a person entitled to recovery,' he is not entitled to the $1,000 statutory minimum damages." For a detailed academic discussion arguing that Justice Souter's textual analysis is not the most natural reading, see Hong, supra note 9, at 99.

(310.) 540 U.S. at 621 n.2.

(311.) Id. at 623.

(312.) Id. Justice Souter also addressed a less obvious text-based argument. The dissent in the Fourth Circuit Chao decision argued that the "person entitled to recovery" language does not prevent individuals unable to show actual damages from collecting the statutory minimum because the damages provision also provides for recovery for costs and attorney's fees. An individual who is unable to show actual damages is still a "person entitled to recovery" because he is entitled to recover costs and attorney's fees as long as he shows adverse effect. See Doe v. Chao, 306 F.3d 170, 188-89 (4th Cir. 2002) (Michael, J., concurring in part and dissenting in part). In Justice Souter's words, "[i]nstead of treating damages as a recovery entitling a plaintiff to costs and fees, this analysis would treat costs and fees as the recovery entitling a plaintiff to minimum damages; it would get the cart before the horse." Chao, 540 U.S. at 625 n.9 (citations omitted).

(313.) Chao, 540 U.S. at 628 (Ginsburg, J., dissenting).

(314.) Id. at 629.

(315.) Id. at 631; see also Hong, supra note 9, at 99 (agreeing with Justice Ginsburg on this point),

(316.) 540 U.S. at 631.

(317.) Id. at 630

(318.) Id.

(319.) Id. at 631.

(320.) 5 U.S.C. [section] 552a(g)(1)(D) (2006).

(321.) Chao, 540 U.S. at 631.

(322.) Id. at 624 (majority opinion). For more on this issue of interpreting "adverse effect," see infra note 328 and accompanying text.

(323.) 540 U.S. at 631 n.1 (Ginsburg, J., dissenting).

(324.) Id. This point had previously been made by the First Circuit in Orekoya v. Mooney, 330 F.3d 1, 18 (1st Cir. 2003).

(325.) But see Tom Goldstein, Call Off the RBG Retirement Watch Before it Starts, SCOTUSBLOG (Feb. 5, 2009, 3:17 PM), http://www.scotusblog.com/2009/02/call-off-the-rbg-retirement-watch-before-it- starts/.

(326.) See, e.g., Chao, 540 U.S. at 634 (Ginsburg, J., dissenting) (discussing the provision of recovery for all damages as one purpose of the Privacy Act, and stressing that many violations of the Act will only result in fear, anxiety, and other nonpecuniary harms); id. at 634-35 n.4 (arguing that the similar purposes that underlie the Act and privacy torts counsel in favor of a broad interpretation of the damages provision); id. at 635 (suggesting that the key purpose of the Act was to create a private enforcement mechanism to deter government misuse of personal information, and that the majority's narrow reading of the damages provision gives many who suffer real harm inadequate incentive to bring suit).

(327.) Id. at 639 (analogizing the Privacy Act to the Tax Reform Act and the Electronic Communication Privacy Act).

(328.) In his dissent to the denial of rehearing en banc in Cooper, Judge O'Scannlain accurately pointed out that one part of the Ninth Circuit's Cooper opinion actually seems inconsistent with the Supreme Court's holding in Chao. Cooper v. FAA, 622 F.3d 1016, 1022 (9th Cir. 2010) (O'Scannlain, J., dissenting from the order denying rehearing en banc). In Chao, Justice Souter wrote that the "adverse effect" language found in 5 U.S.C. [section] 552a(g)(1)(D) and referenced in the damages provision of the Act was used just to limit recovery to those who had standing to bring suit. See supra note 322 and accompanying text. The Ninth Circuit in Cooper, however, aligned itself with Justice Ginsburg's dissenting view in Chao, arguing that the majority's reading of the Act "seems unreasonable in light of how we and other courts have ... recognized that a nonpecuniary harm, such as emotional distress, may constitute an adverse effect under the Act." Cooper v. FAA, 596 F.3d 538, 546 (9th Cir. 2010). Judge O'Scannlain is correct in asserting that this part of the Ninth Circuit opinion in Cooper "conflicts with the Supreme Court's interpetation" in Chao. Cooper, 622 F.3d at 1022. The Ninth Circuit's conclusion, however, is not dependent on this aspect of its opinion, so it would be incorrect to assert--and Judge O'Scannlain does not assert--that the Ninth Circuit majority's holding in Cooper was in direct contradiction to the Supreme Court's holding in Chao.

ALEX KARDON, J.D., Yale Law School, 2010; B.A., Columbia University, 2007. I thank Tom Donnelly, Christine Jolls, and Gregory Sisk for tremendously helpful comments on earlier drafts of this Article. Additionally, I thank William Eskridge for sharing his insightful thoughts in a discussion about statutory interpretation when this Article was in its early stages.
COPYRIGHT 2011 Harvard Society for Law and Public Policy, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2011 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Kardon, Alex
Publication:Harvard Journal of Law & Public Policy
Date:Mar 22, 2011
Words:25686
Previous Article:Foreign nationals, electoral spending, and the First Amendment.
Next Article:Protecting religions from "defamation": a threat to universal human rights standards.
Topics:

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters