Constitutional Law II — Notes for Review — Fall 2003

            Please note:  These notes are fallible and incomplete.  Please do not rely on them to the exclusion of your own notes from class or to the exclusion of the material assigned or referred to in the textbook.  Also, to the extent you disagree with anything in these notes, please do not assume that I am correct and you are not.

            Constitutional provisions.  In this course, we talked primarily about the following provisions of the Constitution: (1) the Contract Clause; (2) the Establishment Clause of the First Amendment; (3) the Free Exercise Clause of the First Amendment; (4) the Free Speech and Press Clauses of the First Amendment; (5) the Due Process Clause of the Fifth Amendment (including its “equal protection component,” see Bolling v. Sharpe); (6) the Takings Clause of the Fifth Amendment; (7) the Ninth Amendment; (8) the Thirteenth Amendment; (9) the Citizenship Clause of the Fourteenth Amendment; (10) the Privileges or Immunities Clause of the Fourteenth Amendment; (11) the Due Process Clause Fourteenth Amendment; (12) the Equal Protection Clause of the Fourteenth Amendment; and (13) the Enforcement Clause of the Fourteenth Amendment (Section 5 of that amendment).  We also spoke at various times about: (1)  the Commerce Clause; and (2) the Privileges and Immunities Clause of Article IV.

            General areas.  As a matter of doctrine, we talked about the following areas of constitutional law: (1) the scope of the Citizenship Clause of the Fourteenth Amendment; (2) the scope of the Privileges or Immunities Clause of the Fourteenth Amendment; (3) economic substantive due process (so-called “Lochnerism”); (4) footnote 4 of Carolene Products and the idea of “discrete and insular minorities” meriting special protection from the judiciary; (5) selective incorporation of provisions of the Bill of Rights into the Due Process Clause of the Fourteenth Amendment; (6) procedural due process; (7) the scope of the Contract Clause; (8) the scope of the Takings Clause; (9) equal protection challenges to laws that regulate on the basis of economic or social distinctions; (10) equal protection challenges to laws that regulate on the basis of race, ethnicity, or religion; (11) equal protection challenges to laws that regulate on the basis of sex; (12) equal protection challenges to laws that regulate on the basis of wealth, age, and mental capacity; (13) equal protection and/or due process challenges to laws that impair the exercise of unenumerated rights protected to some degree by the Constitution (including so-called “fundamental rights”); (14) equal protection challenges to laws that regulate on the basis of sexual orientation; (15) the state-action doctrine; (16) freedom of speech and of the press; (17) freedom of association; (18) establishment of religion; and (19) free exercise of religion.

            Historical themes.  We also discussed several periods in the history of the United States in which the Constitution changed, either pursuant to formal amendment or pursuant to judicial adaptation, in response to events in the political or economic arena.  These events included: (1) the revolutionary war; (2) slavery, the abolitionist movement, Dred Scott, and the Civil War; (3) Reconstruction; (4) the economic expansion of the late nineteenth and early twentieth century and “Lochnerism”; (5) the New Deal; (6) the Cold War; (7) the Civil Rights Movement; (8) the Vietnam War; and (9) the “sexual revolution.”  With regard to the period immediately following the Civil War, we discussed the following sequence of events: (1) the proposal and ratification of the Thirteenth Amendment; (2) the enactment of “Black Codes”; (3) the enactment of the Civil Rights Act of 1866; (4) doubts about the constitutionality of that act; and (5) the proposal and ratification of the Fourteenth Amendment.

            Methods of interpreting the Constitution.  As an abstract matter, we talked about several methods of interpreting the Constitution, including: (1) interpreting the document according to its “plain meaning”; (2) interpreting it according to the intentions of the people who wrote, proposed, defended, and ratified it; and (3) interpreting it according to free-standing principles of right and wrong, such as “natural law”; as well as (4) interpreting it according to hybrids of these approaches.  We talked about some of the advantages and disadvantages of these approaches.

            The Citizenship Clause of the Fourteenth Amendment.  This clause overruled Dred Scott, allowing an African American to be a citizen of a state for purposes of the federal Constitution.  This clause also played at least a supporting role in the Supreme Court’s 1999 decision Saenz v. Roe.  In that decision, the Court struck down a California rule pursuant to which people who had resided in the state for less than a year and who applied for welfare received the benefits that would have been available in the state from which they had come.  The Court justified its decision at least in part on the ground that the Citizenship Clause forbids establishing two levels of citizenship, one for “newcomers” and another for “oldtimers.”

            The Privileges or Immunities Clause of the Fourteenth Amendment.  The Supreme Court rendered this clause a virtual dead letter in The Slaughter-House Cases (1873).  As of that decision, the clause protected a small set of rights implicit in the idea of national citizenship.  See pages 467-68.  In Saenz v. Roe, however, the Supreme Court may have reinvigorated this clause.  In this decision, the Court struck down a California law that provided that people who moved to the state and who were eligible for welfare received the benefits of the state they lived in before they moved to California for one year.  The Court based its decision in part on the effect this law had on people’s right to travel to another state and to remain there if they so chose.  The Court reasoned that this implicated the Privileges or Immunities Clause of the Fourteenth Amendment.  See 526 U.S. 489, 502 (1999).

            Selective incorporation.  We discussed the phenomenon of “selective incorporation” of various provisions of the Bill of Rights into the Due Process Clause of the Fourteenth Amendment.  We read the general rule of Duncan v. Louisiana (1968), see page 487, that the Due Process Clause incorporates those clauses of the Bill of Rights that are “fundamental to the American scheme of justice.”

            The Due Process Clause of the Fourteenth Amendment.  Note:  Virtually any rule promulgated by the courts under this clause would apply to the federal government under the Due Process Clause of the Fifth Amendment.

            Procedural due process.  This aspect of the Due Process Clause protects people against inadequate procedures.  This aspect focuses on the procedures by which the government deprives people of life, liberty, or property, not on the fundamental fairness of the government’s reasons for doing so.  (That is the focus of substantive due process.)

            First question.  The first question in procedural due process analysis is whether the “thing” of which a person claims deprivation is “life, liberty, or property” for purposes of the clause.  This is not always easy to determine.  The following “things” are clearly property: real property; personal property (such as a toaster); a contractual right; an entitlement to receive benefits under a program like Medicaid; a license; and tenure.  But a benefit does not become property for purposes of the clause simply because one subjectively hopes for it.  One does not look to how much a person wants or even needs something to determine whether it qualifies as property, etc., under the clause.  Instead, one looks to the nature of the thing and to the extent to which the government creates a reasonable expectation of entitlement based on positive law.  See the last paragraph on page 490 and the second paragraph on page 491, both parts of the Court’s decision in Roth.

            Second question.  The second question in procedural due process analysis — which one only reaches if the “thing” at issue is life, liberty, or property — is whether the procedures provided by the government are adequate.  To answer this question, one applies the “flexible balancing test” of Mathews v. Eldridge.  See the second full paragraph on page 504.  This looks like standard cost/benefit analysis.  On can compare this test to the balancing test set forth by Judge Learned Hand in United States v. Carroll Towing, which may be familiar to you from torts.

            Note:  It is often said that procedural due process requires, at a minimum, notice and an opportunity to be heard, but the government may, in cases of emergency, deprive a person of something before providing notice and a hearing, and provide only a post-deprivation hearing.

            Another note.  Often, the issue in a procedural due process case will be whether the government or the private claimant should bear the risk of error between a preliminary adjudication and subsequent, more extensive review.  Specifically, the government and claimants often argue over whether, if the government prevails at the preliminary stage, the “thing” at issue should be denied to the claimant pending full review.  If the “thing” is not denied to the claimant pending full review, the claimant has every incentive to request such review, and the preliminary review will not be particularly meaningful, at least from the government’s point of view.  On the other hand, if the “thing” is denied to the claimant pending full review, the claimant may suffer greatly, as well as unjustifiably, if it turns out that the claimant remains entitled to the “thing.”  Factors that courts might take into account in deciding who should bear the risk of error between adjudications include: (1) the degree to which the claimant depends on the “thing”; (2) the extent to which the claimant can use the “thing” to hurt others (this factor might be pertinent with respect to a license); and (3) the scarcity of the “thing,” such that someone other than the claimant could put it to better use.  Some of these factors were at issue in Matthews v. Eldridge.

            Mount Healthy burden-shifting.  Often the government has both a good and a bad reason for acting.  Under Mount Healthy, if a person aggrieved by a governmental action shows that a “bad reason” was a “motivating factor” in the government’s action, the government must then show by a preponderance of the evidence that it would have acted as it did with or without the bad reason.  The judge then makes the final call.  See note 5 on page 513.

            [Note: We did not discuss Mt. Healthy in 2003.]

            The Contract Clause.  Note:  This clause applies only to the states.  If one is upset with the federal government’s interference with a contractual expectation, one must look to the Due Process Clause of the Fifth Amendment.

            Energy Reserves Group test.  The general rule regarding contracts between private parties is set forth in Energy Reserves Group v. Kansas Power & Light, 459 U.S. 400, 411-13 (1983).  Under this test, one first asks whether there is a valid contract.  If so, one next asks whether the law at issue substantially impairs an obligation arising under that contract.  If it does, one then asks whether the law reasonably serves a significant and legitimate governmental purpose.  If so, the law does not violate the Contract Clause.  If not, it does.  On this third question, the courts will defer to the legislature.  See Energy Reserves Group, 459 U.S. at 412-13.  The decision whether a law “reasonably” serves a “significant” governmental purpose is somewhat subjective, but there are some considerations that play a role in the determination.  For example, one might ask whether the state enacted the law in response to an emergency, whether the law will expire when the emergency ends, how deeply the law impairs the contractual obligation at issue, and whether the law regulates an activity that was already subject to heavy regulation.  See Blaisdell.

            Where the state is a party to the contract.  If a state enacts a law that impairs one of its own contractual obligations, the courts will subject the law to “greater scrutiny” than would be the case regarding a law that only interferes with private obligations.  See United States Trust v. New Jersey, pages 537-38.  When this is the case, the courts will not show as much deference toward the legislature’s determinations regarding the “reasonability” of the law as would be the case were the state not a party to the contract at issue.  See Energy Reserves Group, 459 U.S. at 412-13.

            The Takings Clause.  There are two kinds of takings — possessory and regulatory.

            Possessory takings.  Almost any permanent physical occupation or confiscation of property by the government is a per se taking.  For example, if the government builds a road through one’s land, or builds a dam and floods one’s land, or even requires a homeowner to set aside a cubic foot of space for a cable box (whether or not the homeowner subscribes to cable, that is), there is a taking.  The analysis is more complex, however, when the government requires private parties to allow other private parties on their property.  This requires some kind of factor-based analysis.  Compare PruneYard Shopping Center (no taking where California required the owner of a mall to allow people to distribute leaflets in the mall) with Kaiser Aetna (taking where the government required the owners of a private marina to allow others into the marina).

            Nor can the government require a landowner to give up property in exchange for some form of zoning relief, unless there is an appropriate relationship between the impact of the landowner’s proposed change, were the relief granted, and the condition imposed.  First, there must be an “essential nexus” between these two things.  See Nollan v. California Coastal Comm’n.  In other words, the landowner’s intentions must implicate the same concerns that cause the government to impose the condition.   Second, there must be some “rough proportionality” between the condition and the impact of the landowner’s proposed change.  See Dolan v. City of Tigard.  This is especially true where the law applied is not one of general applicability, but instead is imposed on a case-by-case basis.

            Regulatory takings.  No possession is required to establish a regulatory taking.  There are at least two kinds of regulatory takings.  First, there are “per se regulatory takings.”  A per se regulatory taking occurs if the government deprives a landowner of all economically beneficial or productive use of his or her land.  See Lucas, pages 553-54.  This would not, however, include uses that would be classified as nuisances at common law.  Even if a landowner cannot establish that a governmental action deprived him or her of all economically beneficial use of a piece of property, the landowner may still try to establish a regulatory taking under the amorphous test of Penn Central.  See page 551.  Under this test, one looks to the economic impact of the regulation, the existence or non-existence of “distinct investment-backed expectations” regarding what the landowner could build on the site, and the character of the governmental action at issue.  (Frankly, I don’t think the third factor — regarding the “character” of the government’s action — adds anything to the test, because Justice Brennan appeared to be using this factor in Penn Central to distinguish regulatory from possessory takings, whereas nowadays courts appear to use Penn Central only for cases involving regulatory takings.)  In Penn Central, the Court found no taking where a city would not allow Penn Central to build an office tower on top of a landmark train station.

            Eastern Enterprises v. Apfel.  We talked a little about this case, which arguably depends more on a substantive due process rationale than on a takings rationale.

            Personal property.  Keep in mind that destruction of all economic value of personal, as opposed to real, property will not constitute a taking, provided non-economic value remains.  See Andrus v. Allard (the case involving feathers of a protected bird that Congress said could not be sold).  See page 555.

            Equal protection.  When analyzing a regulation under the Equal Protection Clause, one must first ask whether the regulation divides the universe of entities subject to regulation into more than one group on the basis of a characteristic.  If not, then the regulation does not implicate equal protection.  If so, one then asks what the nature of the distinction is.

            Suspect classifications and strict scrutiny.  If a regulation makes a “suspect classification,” it will provoke strict scrutiny.  When subjecting a regulation to this kind of scrutiny, courts will presume that it is unconstitutional, and will only uphold it if it is necessary and narrowly tailored to serve a compelling governmental interest.  “Narrow tailoring” in this context probably means that the regulation must be the least restrictive means available to the government to accomplish a compelling objective.  Few regulations will satisfy this test.  Distinctions based on race, ethnicity, religion, and alienage (for the most part) are examples of suspect classifications.

            Quasi-suspect classifications and intermediate scrutiny.  If a regulation makes a “quasi-suspect classification,” it will provoke intermediate scrutiny.  When subjecting a regulation to this kind of scrutiny, courts will (probably) presume that the regulation is unconstitutional, and will only uphold it if it is substantially related to an important governmental interest.  Distinctions based on sex and legitimacy of birth are examples of quasi-suspect classifications.  Also, many would argue that the Court covertly subjects distinctions based on mental capacity and sexual orientation to intermediate scrutiny, or at least to some form of scrutiny more rigorous than the traditional “minimum rationality” test.

            Non-suspect classifications and minimum rationality.  If a regulation makes a distinction based on a non-suspect classification, it will provoke the so-called “minimum rationality” test.  When applying this test, courts will presume that the regulation is constitutional, and they will uphold it so long as it is rationally related to a legitimate governmental interest.  Few regulations would fail this test.  In addition, the government need not have come up with a rational basis for the regulation at the time of enactment.  It suffices if the government comes up with such a basis in time to present it to a reviewing court.  Distinctions based on economic or social characteristics, as well as distinctions based on age and wealth, are examples of non-suspect classifications.  In addition, the Court has treated distinctions based on mental capacity and sexual orientation as non-suspect, although many would argue that the Court covertly treats such distinctions as quasi-suspect.

            “Underinclusive” and “overinclusive” regulation.  Once the government has identified the evil it seeks to eradicate, any failure on its part to regulate all sources of the evil constitutes “underinclusive” regulation.  Similarly, any failure on its part to limit its regulation to sources of the evil constitutes “overinclusive” regulation.  This is not generally a problem with minimum rationality review, because courts will generally allow the government to take “one step at a time” against an identified evil, thus justifying underinclusiveness, and will also generally allow the government to paint with a broad brush where using a fine brush would be too expensive or administratively inconvenient, thus justifying overinclusiveness.  Regulations that are under- or overinclusive and that make suspect or quasi-suspect classifications would probably fail scrutiny, however, because the fit between ends and means would not be close enough.

            Treating similarly situated entities similarly.  This can be a difficult concept to work with, but it relates to the discussion of underinclusive and overinclusive regulation.  If two entities are similarly situated, in the sense that they cause the same evil, a regulation that affects only one of them will fail to treat similarly situated entities similarly, which gives rise to an argument that the regulation’s means and end are not commensurate.

            Treating dissimilarly situated entities dissimilarly.  One can also talk about a obligation of the government to treat dissimilarly situated entities dissimilarly, but this is not nearly as clear cut as the government’s obligation to treat similarly situated entities similarly.  If two entities are dissimilar in a material way, one could theoretically argue that the government must take this dissimilarity into account in making regulations.  For the most part, however, courts reject this analysis.  Thus, the government may prohibit both rich and poor from sleeping under bridges and stealing bread.  But the argument often arises.

            Footnote 4 of Carolene Products.  Given the three levels of scrutiny under the Equal Protection Clause, plaintiffs naturally want to establish that the government has made a suspect or quasi-suspect classification, and defendants naturally want to establish that the government has made only a non-suspect classification.  Footnote 4 of Carolene Products is helpful, although not dispositive, in deciding what kind of classification the government has made.  As the Court noted in this footnote, although it will generally presume the constitutionality of legislation, it will not do so when reviewing legislation that: (1) impairs a right explicitly protected by the Constitution; (2) impairs a right that is somehow integral to the political process, on the ground that the political process, being “injured” by the impairment, would be unable to repair the damage; and (3) isolates for adverse treatment a minority that cannot adequately defend itself in the political process.  In deciding what groups fit into this last category, one looks to whether the minority is “discrete and insular,” whether members of the minority bear “immutable characteristics,” and whether the minority has been subjected to a “history of purposeful unequal treatment.”  Justices Powell and Brennan talked about this kind of analysis in Bakke.  See pages 675 (second paragraph) and 682 (carryover paragraph at the bottom of the page).

            Intent versus impact.  A plaintiff seeking to prove a violation of the Equal Protection Clause must establish that the defendant intended to discriminate along improper lines.  See Washington v. Davis; Arlington Heights.  A showing of disproportionate impact alone will almost never suffice, unless the disproportion is so striking as to preclude any conclusion other than one of intentional discrimination.  See Yick Wo for an example of such strikingly disproportionate impact.  The law need not discriminate on its face, however.  Discriminatory application or administration of a law also implicates equal protection.

            Arlington Heights.  Look to Arlington Heights for examples of ways to prove intentional discrimination with circumstantial evidence.  As the Court noted in that case, one can look to: (1) the historical context of the action at issue; (2) the sequence of events leading up to the action; (3) departures from normal procedures; (4) substantive departures (actions out of the ordinary); and (5) the legislative or administrative history of the action, such as contemporaneous statements by the proponents of the action.

            The Equal Protection “Component” of the Due Process Clause of the Fifth Amendment .  The Equal Protection Clause of the Fourteenth Amendment does not apply to the federal government, but the Supreme Court has held that the Due Process Clause of the Fifth Amendment, which does apply to the federal government, includes an “equal protection component.”  See Bolling v. Sharpe.  In general, whatever would be unconstitutional for a state or local government under the Fourteenth Amendment’s Equal Protection Clause would also be unconstitutional for the federal government under this component of the Fifth Amendment’s Due Process Clause.  It is possible, however, that Congress retains special powers, beyond those possessed by the states, to adopt legislation designed to promote the goals of the Equal Protection Clause, pursuant to Section 5 of the Fourteenth Amendment.  See Adarand Constructors.  To the extent this is true, it indicates that Congress may have greater power than the states and local governments to adopt affirmative action programs when regulating conduct by the states and their subdivisions.  But this remains to be seen, and this thinking had no discernible impact on the Court’s analysis in Adarand.

            Affirmative action.  All race-based distinctions provoke strict scrutiny — even distinctions designed to help racial minorities.  Similarly, all sex-based distinctions provoke the same level of scrutiny, even if they are designed to help women.  One might say that there’s a certain “me too” aspect to equal protection and suspect classifications.  In other words, once the courts identify a discrete and insular minority meriting heightened protection under the Equal Protection Clause, the courts then automatically extend similar protection to the “majority” that remains once the “minority” is identified.

            Not necessarily fatal.  Remember that affirmative action programs can and often do survive scrutiny.  In Grutter v. Bollinger (2003), for example, the Supreme Court expressly held that the goal of diversity in education can be compelling.  In accordance with this decision, quotas and firm goals are not permitted, but use of race as a factor in admissions is allowed.

            Remedial measures.  Remember also that a court may make explicit race-based distinctions in its orders when it is seeking to remediate a violation of the Equal Protection Clause.  It does this as part of its ordinary equitable jurisdiction.  Similarly, parties to litigation involving the Equal Protection Clause may include explicit race-based distinctions in consent decrees that they submit to the court for approval.  In addition, in accordance with Justice Powell’s opinion in Bakke and later decisions of the Court, properly constituted governmental bodies may adopt affirmative action programs designed to remediate past discrimination, provided they satisfy a set of fairly rigorous criteria.  First, the body must be appropriate to the task.  Second, it must do its homework regarding identifying the past discrimination and determining its extent.  Finally, the remedy it designs must not exceed the scope of the identified violation.  Thus, if a properly constituted body, having thoroughly examined a particular situation, discovered an existing violation of the Equal Protection Clause in that situation, it could devise an affirmative action plan designed to address the precise violation discovered.  Also, in analyzing such a plan, the courts would look to its flexibility.  For example, the courts would ask whether the plan allows for waivers and relaxation of presumptions where the actual facts differ from the facts presumed to justify the plan.  Similarly, a court might look to the duration of the plan, that is, to whether it would end when the identified violations have been addressed.  This was an important part of the Supreme Court’s analysis in Adarand Constructors.

            Fundamental rights.  Laws that impair so-called “fundamental rights” are unconstitutional unless they are necessary and narrowly tailored to serve a compelling governmental interest, and are the least restrictive means of doing so.  One can go through the cases to identify the specific fundamental rights that the Supreme Court has identified.  They include, but are by no means limited to, a right to procreate (see Skinner), a right to use contraceptives (see Griswold and Eisenstadt), a right to abort a pregnancy, within certain broad limits (see Roe, Casey, and Stenberg), a right to marry (see Loving and Zablocki), a right to “parent” (referred to in various cases), (most likely) a right to refuse medical treatment (assumed to exist in Cruzan), and a right to engage in sodomy (Lawrence v. Texas).  These rights specifically do not include a right to obtain the assistance of a physician in committing suicide, unless (perhaps) a person is terminally ill, in great pain, and facing imminent death (see Glucksberg and Quill).  [Note: We did not read Stenberg in 2002.]  One should also note that the Court appeared to refrain from literally referring to the rights at issue in Casey and Lawrence as “fundamental.”

            The “undue burdens” test.  Remember that the right to abort a pregnancy is actually protected by the “undue burdens” test, rather than strict scrutiny per se.  Under this test, a regulation of abortion is unconstitutional if it imposes an undue burden on the exercise of that right, that is, if its purpose or effect is to place a substantial obstacle between a woman and her ability to choose that procedure.  See the fourth paragraph on page 851, which is part of the opinion of Justices O’Connor, Kennedy, and Souter.

            The government need not pay for the exercise of a fundamental right, however.  Thus, the government need not pay for abortions, see Maher v. Roe and Harris v. McRae, or provide facilities at which they can be performed, see Webster v. Reproductive Health Servs.

            Abstract notes regarding substantive due process (fundamental rights analysis):  First, remember the Ninth Amendment, which may or may not confer legitimacy to the idea of “unenumerated” constitutional rights.  Second, keep in mind the contest among the justices regarding the appropriate “level of generality (or specificity)” at which to identify historically protected rights.  If one identifies historically protected rights at a highly specific level, applications of the doctrine of substantive due process to new phenomena will appear relatively radical, and a court will be less disposed to recognize “new” rights.  Conversely, if one identifies historically protected rights at a high level of abstraction or generality, recognition of “new” rights will appear commonplace, as long as the new right fits within the general pre-existing right.

            One good description of the Court’s approach to substantive due process can be found in Casey, on page 847 (regarding the exercise of “reasoned judgment”), in the first full paragraph.  Another can be found in Justice Harlan’s opinion in Griswold, on page 828, in the last full paragraph of that opinion.  See also Justice Holmes’ dissent in Lochner, on page 476 (last paragraph of the opinion).  For a presentation of the more conservative justices’ approach, see the opinion of the Court in Glucksberg, on page 900.

            Romer v. Evans is a hard case to classify logically.  Reading between the lines, however, the Court seemed to subject laws that distinguish on the basis of sexual orientation to some form of heightened scrutiny, even though it claimed to use only minimum rationality.  The Court described the amendment at issue in that case as one that swept so broadly that it could not have rested on anything but invidious hatred toward homosexual people, which is, after all, an illegitimate governmental interest.

            The Cincinnati case (Equality, 128 F.3d 289 (6th Cir. 1997)) is quite similar, but different in one or two small ways.  First, the law at issue in Equality was an amendment to a city charter.  The law at issue in Romer was an amendment to a state constitution.  Some have argued that this made a difference.  Second, the law at issue in Equality did not purport to deny “protected status,” as did the law at issue in Romer.  Instead, it purported to deny “special status.”  There are other differences between the two provisions.

            State action.  Very few of the provisions of the Constitution apply to private conduct.  In particular, the key substantive provisions of the Fourteenth Amendment — the Due Process Clause and the Equal Protection Clause — apply only to state action.  Similarly, the First Amendment applies directly (instead of via the Due Process Clause of the Fourteenth Amendment) only to actions by the federal government.  But this simple assertion conceals a complex analysis of when an ostensibly private actor performs a state action.  In fact, this is an amorphous area of law.

            There are three basic sub-doctrines of the state action doctrine: (1) the “public function” doctrine; (2) the “symbiotic relationship” doctrine; and (3) the “nexus” doctrine (for lack of a better name).

            Public function doctrine.  Under the public function doctrine, if a private entity or person performs a function traditionally and exclusively performed by the government, he, she, or it is a state actor.  Because this is a difficult standard to satisfy, few entities have been deemed state actors under this doctrine.  See Marsh v. Alabama (company town); Terry v. Adams (primary election); and Evans v. Newton (park).  The Supreme Court has never decided whether a private police officer is performing a public function in accordance with this doctrine.  See footnote 14 of Flagg Bros. v. Brooks, on page 611.

            Symbiotic relationship doctrine.  A symbiotic relationship exists where the government and a private entity have such a high degree of interdependence that the private entity’s operation could be considered integral to the government’s operations, and vice versa.  See Burton v. Wilmington Parking Authority.

            Nexus.  This could also be described as an “encouragement” or “facilitation” test.  In applying this test, which acts as a catch-all category, one looks to whether the government is sufficiently implicated in private conduct such that the private conduct can be attributed to the government.  To illustrate, one could compare the facts of Flagg Brothers v. Brooks with the facts of Lugar v. Edmondson Oil Company.  [Note: We did not read either of these cases in 2003.]  In Flagg Brothers, a warehouse proposed to sell goods entrusted to it for storage because the owner of the goods had failed to pay her bill.  The Court held that the action of the warehouse was not state action.  In Lugar, by contrast, the Court found state action where a sheriff attached Lugar’s property pursuant to an ex parte application for an attachment by Edmondson.  The difference perhaps lay in the sheriff’s role in the attachment.  The government was in the background in Flagg Brothers.

            Shelley v. Kraemer is another example of a situation in which the government’s role in a particular situation was deemed prominent enough to implicate the Equal Protection Clause.  In this case, private parties had agreed to a racially discriminatory restrictive covenant.  But the covenant could only be enforced pursuant to an order of a court.  The Supreme Court held that judicial enforcement of the covenant would constitute state action and violate the clause.  One wonders, however, whether judicial enforcement of an agreement between two private parties not to discuss a particular subject, etc., would constitute state action and violate the First Amendment as a regulation of speech on the basis of content.  As a doctrinal matter, many defend Shelley v. Kraemer against this criticism by noting that, in Shelley, a willing seller and a willing buyer were being prevented from transacting business, whereas in the case of a reciprocal agreement not to disparage, etc., the two sides at least initially chose to relinquish their right to speak freely on the subject at issue.  Another scholarly argument is that Shelley v. Kraemer simply reflects our national commitment to eradicate racial discrimination.  Some also argue that there is actually a hierarchy of constitutional values, with equal protection and freedom of speech at the top, and procedural due process at the bottom — at least for purposes of state action analysis.  On this view, the courts will be more likely to find state action where discrimination is at issue, and less likely to find it where a denial of procedural due process is at issue.  Compare Shelley with Jackson v. Metropolitan Edison.

            Licensing and regulation.  Being licensed or subject to regulation by the government — even if the regulation is quite close – is usually not enough to render an entity a state actor.  See Moose Lodge (liquor license); Jackson (regulated utility).

            The First Amendment.

            Why protect expression?  There are several justifications for protecting expression.  One is that expression is the mechanism by which the people inform themselves.  In a democracy, the government is the people.  Therefore, when the people talk to each other, they are engaging in a process of deliberation without which effective government would be impossible.  An advantage of this theory is that it has obvious support in the structure of the Constitution.  A disadvantage of this theory is that the political significance of much of the expression protected by the First Amendment is at best remote.  Take, for example, a work of instrumental music, or a work of abstract art.  Although such works may have ultimate political significance, their primary significance is emotional or aesthetic.  Another theory for the protection of expression is that expression is part of “self-actualization.”  In other words, by expressing ourselves, we become more fully the people we are destined to be.

            Incitement — the Brandenburg Test.  Government may not punish abstract advocacy of violent or illegal acts.  Instead, government may only punish speech that is: (1) directed to inciting or producing imminent lawless conduct; and (2) likely to do so.  Notwithstanding the apparent clarity of this test, however, there are some unresolved issues pertaining to it.  For example, it isn’t clear what “direction” means.  For example, does one need to parse the words of the speaker?  Should one consider the gravity of the potential harm at issue?

            Prior restraint.  The Supreme Court has said that prior restraints are subject to the strictest of scrutiny, because, if a prior restraint is imposed, the speech subject to the restraint is never available to the public, which, the theory goes, needs to hear the speech to fulfill its role as the real government in a democratic system.

            National security.  National security can justify a prior restraint in rare instances, but the government would have to satisfy a very rigorous standard.  To get a feel for the standard, see Justice Brennan’s concurring opinion in New York Times v. United States, in which he said that a prior restraint must rest upon at least “allegation and proof that publication must inevitably, directly, and immediately cause the occurrence of an event kindred to imperiling the safety of a transport already at sea.”  See page 980.  Similarly, Justice Stewart indicated that the Constitution would require a showing that “direct, immediate, and irreparable damage” would result from publication of the information at issue.  See page 980.  Several justices indicated, however, that a carefully drawn act of Congress authorizing a prior restraint in particular instances might have helped the government’s case.

            Licensing of parades, etc.  Licensing statutes and ordinances must have narrow, objective, and definite standards for administration, to prevent inappropriate exercise of discretion by governmental officials.  See Shuttlesworth, page 1000.

            The collateral bar rule.  Generally speaking, one must adhere to a court-ordered prior restraint, even if the restraint is predicated upon an unconstitutional statute, unless the order is “transparently invalid” or has only a “frivolous pretense to validity.”  See Walker, page 996.

            Prior restraints and pre-trial legal proceedings.  We did not discuss this in 2003, but any court imposing such a restraint must bear a heavy burden to justify such action.  See generally Nebraska Press, page 1125.

            The public forum, etc.

            New traditional public forums?  We are not likely to get new TPF’s, in light of the Court’s language in ISKCON, see page 1029, indicating that the government does not create a traditional public forum by inaction.

            Fighting words.  Technically, the government may punish fighting words, but prosecutions under fighting words statutes historically fail because of vagueness in the statute.  See the Chaplinsky formulation for the definition of fighting words, as set forth in Cohen, page 1071, where the Court describes fighting words as “personally abusive epithets which, when addressed to the ordinary citizen are, as a matter of common knowledge, inherently likely to provoke violent reaction.”

            Vagueness.  If a reasonable person cannot tell what a law forbids or allows, the law is unconstitutionally vague and it violates due process.  [Note: We did not discuss this concept in 2003.]

            Commercial speech.  This is a form of “lesser protected” speech.  There is no single definition of commercial speech.  Many people look to the factors set forth by Justice Marshall in Bolger v. Youngs Drug Products Corp., 463 U.S. 60 (1983).  These factors are: (1) whether the material is conceded to be an advertisement; (2) whether it refers to a specific product; and (3) whether the speaker has an economic incentive for the speech.  The test for regulation of commercial speech is the somewhat lenient test of Central Hudson.  Under this test, one first asks whether the speech concerns lawful activity and is not misleading.  If it concerns unlawful activity, or it is misleading, it is not protected by the First Amendment and the government may ban it outright.  If it concerns lawful activity and is not misleading, the next question is whether the government’s interest is substantial.  If the government’s interest is not substantial, the regulation fails scrutiny.  If it is substantial, one next asks whether the regulation “directly advances” the government’s interest.  If it does not directly advance that interest, the regulation fails scrutiny.  If it does directly advance that interest, one next asks whether the regulation is no more extensive than necessary to serve that interest.  If it’s more extensive than necessary, the regulation fails scrutiny.  If not, it’s constitutional.  Note that the Court has held that the “least restrictive means” test is not part of the Central Hudson test.  See Fox, page 1148-49.

            Keep in mind that, under Central Hudson, the government generally may not single out commercial speech for regulation on a ground that applies equally to non-commercial speech.  See Discovery Network, page 1149.

            Defamation.  The First Amendment does not protect certain forms of defamatory speech.  See generally New York Times Co. v. Sullivan.  The law in this area is fairly complicated, and depends on several factors.  If the plaintiff is a public official, for example, he or she cannot recover damages for a defamatory falsehood relating to his or her official conduct unless he or she “proves that the statement was made with “actual malice” — that is, with knowledge that it was false or with reckless disregard of whether it was false or not.”  See page 1174.  Moreover, the plaintiff must make this showing with “convincing clarity.”  See page 1174.  “Public officials” includes candidates for public office.  It also includes individuals fairly far down the “food chain” of government, although it does not include all governmental officials.

            The standard for public officials also applies to so-called “public figures,” that is, individuals who for all purposes or for limited purposes somehow become involved in public controversies.  (By the way, some surmise that there can never be an “involuntary” public figure.  In other words, some believe that a person can become a public figure only by voluntarily inserting him or herself into the maelstrom of public opinion.)

            If the plaintiff is not a public official or figure, but the story at issue relates to a matter of public concern, the standard for liability for compensatory damages in an action for defamation need not be the actual malice standard of New York Times v. Sullivan.  Instead, it can be whatever standard a state sets, so long as the standard is not strict liability.  In these circumstances, however, the standard for presumed or punitive damages is the actual malice standard of Sullivan.  See Gertz v. Robert Welch, Inc.

            If the plaintiff is not a public official or figure, and the story at issue does not relate to a matter of public concern, the state may be able to impose liability for compensatory damages in an action for defamation on the basis of strict liability.  (In other words, the state may be able to provide that, if the “published” statement is false, of and concerning the plaintiff, and injurious to character, etc., liability for compensatory damages attaches, without regard to scienter.)  I say “may be” because the Court did not expressly resolve this issue in Dun & Bradstreet, Inc. v. Greenmoss Builders, Inc.  In that case, the Court did, however, hold that a private plaintiff bringing an action for defamation with respect to speech not relating to a matter of public concern need not meet the standard of New York Times v. Sullivan to obtain punitive damages.  The Court did not state what the standard must be, however.  Presumably the plaintiff would have to establish some form of fault.

            Freedom of association.  Under the opinion of the majority in Roberts, the government may not require a private expressive association to accept members they do not desire, unless the requirement serves a compelling public interest and the government lacks a significantly less restrictive means of accomplishing its objective.  In practice, however, this analysis amounts to an assessment of whether requiring a particular association to accept a particular member, etc., would impose a sufficiently heavy burden on the association’s ability to express itself.  In Roberts itself, which involved the Jaycees, the Court was skeptical as to how requiring the Jaycees to admit women would impair its message.  In Dale, by contrast, the Court was far more willing than it had been in Roberts to defer to the Boy Scouts’ representation that having to allow Dale, an avowed homosexual and advocate for homosexuals, to serve as an assistant scoutmaster would impair its ability to convey a message.

            We also talked a little about Justice O’Connor’s separate opinion in Roberts.  Under her approach, when assessing the constitutionality of a statute or ordinance that threatens to interfere with the membership of a private association, one would first ask whether the organization is predominantly commercial or not.  If the organization is predominantly commercial, government may regulate its membership as long as the regulation satisfies minimum rationality.  If, on the other hand, the organization is not predominantly commercial, but instead is expressive, she appears to argue that the government may not regulate its membership at all.

            Symbolic speech.  This kind of speech is protected under O’Brien.  Under this test, which is essentially the same as the test for content-neutral regulations of the time, place, or manner of speech, a regulation is constitutional if: (1) it is based on a legitimate power of the government (such as the power to raise an army); (2) it promotes an important or substantial public interest; (3) the interest it serves is unrelated to the suppression of expression; and (4) the incidental restriction on expression is no greater than is essential to promote the government’s interest.

            Obscenity.  See the Miller test, on page 1296.  This test has several parts: (a) whether the average person, applying contemporary community standards, would find that the work, taken as a whole, appeals to the prurient interest; (b) whether the average person, applying contemporary community standards, would find that the work depicts or describes, in a patently offensive way, sexual conduct that is specifically defined by applicable state law; and (c) whether the work, taken as a whole, applying the standard of a reasonable person (not community standards), lacks serious literary, artistic, political, or scientific value.  The Supreme Court has given examples of the kind of “sexual conduct” that the state can define for purposes of part (b) of the test.  See pages 1296-97.  Also, independent appellate review must be available when necessary to hear constitutional arguments relating to the status of speech as obscene or protected.  Please also bear in mind that, in a criminal prosecution for obscenity, the government must establish that the defendant knew the contents of the work at issue.

            Child pornography.  The test for child pornography is much more lenient, from the government’s point of view, than the test for obscenity not involving depictions of actual children.  The trier of fact need not determine that the material appeals to the prurient interest of the average person; the state need not require that sexual conduct be portrayed in a patently offensive manner; and the work need not be considered as a whole.  But the material must involve a live performance or some kind of visual reproduction of such a performance.  The forbidden acts must be defined with sufficient precision, to put people on notice of what is forbidden.  The Court has never resolved whether a depiction of a child engaging in sexual conduct would be protected if it had serious artistic value, etc.

            Pacifica.  The standard of review for content-based regulations of the broadcast media appears to be more lenient than the standard of review for books, newspapers, movies, or speech in the park.  This is based on a variety of considerations — the scarcity of frequencies in the electromagnetic spectrum (although this is becoming increasingly irrelevant), the ease with which children can see or hear indecent, etc., language or imagery broadcast via these media, and the broadcast media’s long history of regulation.  There was no majority opinion in Pacifica, however.  The Court may have elevated the broadcast media to the same level of protection as other media in dictum in United States v. Playboy Entertainment Group Inc.  See 529 U.S. 803 (2000), but it probably did not.

            Zoning adult entertainment.  The Supreme Court finally came up with a majority rationale for allowing the zoning of adult entertainment according to its content in Renton v. Playtime Theatres, Inc., although the Court technically does not consider such zoning to be “content-based.”  Under the “secondary effects” rationale of Renton, the Court evaluates facially content-based zoning of adult entertainment establishments according to the test for content-neutral regulations set forth in O’Brien, on the ground that the government does not seek to suppress the speech itself, but only to regulate its unwanted secondary effects — diminution in property values, etc.  The Court also held in Renton that a municipality need not conduct its own empirical study to justify such an ordinance, but instead may rely on the studies of others.  Keep in mind the difference between “secondary” and “primary effects” of speech.  In Boos v. Barry, the Court reasoned that regulation of speech to prevent embarrassment or offense would be regulation according to speech’s primary effect, and would not qualify for the Renton test.

            The Establishment Clause.  The primary test for evaluating laws challenged under the Establishment Clause is the Lemon test, named after the 1971 decision Lemon v. Kurtzman.  The test has three prongs, each of which the law at issue must satisfy: (1) the law must have a secular purpose; (2) the law’s primary effect must be neither to promote nor to inhibit religion; and (3) the law must not foster excessive entanglements between church and state.  In the 1997 decision Agostini v. Felton, the Court appeared to fold the entanglements prong into the effects prong.  See 521 U.S. 203, 232.  It also quoted some additional language that appeared to elaborate on these two combined prongs.  Under this additional language, the Court looks to “the character and purposes of the institutions that are benefitted, the nature of the aid that the [government] provides, and the resulting relationship between the government and religious authority.”  Id. (quoting Lemon).  This language indicates that the “effects” test is much less empirical than one might otherwise imagine.

            In the 2002 decision Zelman v. Simmons-Harris, the Supreme Court upheld a school-choice plan in Cleveland, Ohio, in which certain children in low-income families were given the option of attending any school, public or private, sectarian or non-sectarian, that participated in the program.  Although over 80% of the private schools that participated had a religious affiliation, and although 96% of the students who used the program to attend a private school attended a sectarian school (at least in one year studied), the Court upheld the program against challenge under the Establishment Clause.  The Court reasoned that the criteria for the program were formally neutral regarding the religious affiliation (or lack thereof) of participating private schools, and the program had sufficient breadth to make the choice of parents to send their children to sectarian schools truly private, and not that of the government.

            Equal treatment or “neutrality”.  This approach to the Establishment Clause is gaining steam.  Under some justices’ view of neutrality, an even-handed law that makes no distinctions on the basis of religion does not violate the Establishment Clause.  One can safely put the Chief Justice and Justices Scalia and Thomas in this camp.  One can probably also put Justice Kennedy in this camp.  Justice O’Connor, however, seems to adhere to a more “substantive” version of neutrality, pursuant to which she is concerned somewhat about how the government’s conduct would be perceived.  Justices Souter and Breyer also consider this important.  Justices Stevens and Ginsburg have such a lenient standard for perceptions, however, that virtually any official practice that redounds to the benefit of religion, however even-handed the practice might be, would violate the clause.  See generally Capitol Square Review and Advisory Board v. Pinette.

            Coercion.  The Court actually based its decision in Lee v. Weisman, pages 1389-90, on a “psychological coercion” theory.  This case involved a non-sectarian benediction at a high school graduation.  (The Court struck down the practice.)  The dissenting justices (the Chief Justice and Justices White, Scalia, and Thomas) would have imposed a requirement of legal coercion, which they said was not present in the case.  It is unlikely, however, that a majority of the Court would adhere to a coercion test today.

            [Note: We did not read or discuss Pinette or Lee in 2003.]

            The Free Exercise Clause.  Under Smith, neutral laws of general applicability are subject only to minimum rationality review, even if they incidentally impose burdens on the exercise of religion.  There are exceptions to Smith, however.  First, a law must be truly neutral and generally applicable to qualify for Smith.  If there are non-religious exceptions to the law, an argument can be made that the law is not neutral toward religion.  Second, no law may regulate pure belief.  Third, laws that impair so-called “hybrid” rights, such as a combination of free exercise and free speech, or free exercise and the right to parent, will still be subject to strict scrutiny.  Finally, there’s something left of Sherbert, although not much.  Perhaps there’s something left of Sherbert with regard to unemployment benefits, or with regard to cases in which individualized assessments are feasible.  In addition, the Religious Freedom Restoration Act, which purported to “overrule” Smith and which the Court struck down in City of Boerne v. Flores, may still apply to the federal government.  The standard under RFRA is strict scrutiny, with a least restrictive means test.  In striking down some (or all) of RFRA, the Boerne Court said that Congress lacked the power under Section 5 of the Fourteenth Amendment to enact RFRA because there was no “congruence and proportionality” between historical interference with religious free exercise and RFRA.

            Courts do not like to assess the extent of a burden on religious exercise, nor will they question the wisdom of beliefs.  But they will question the sincerity of beliefs, if necessary.