Concept
Contraception and Privacy
Fundamental rights
Contraception and Privacy
The family-rights cases just studied protect what parents decide for children who already exist. The contraception cases protect the earlier decision whether to have a child at all, and the Court had treated that decision as basic before it ever reached contraception. Buck v. Bell upheld Virginia’s compulsory sterilization of institutionalized persons by reasoning that the principle supporting compulsory vaccination was broad enough to cover the procedure, and the decision has not endured, morally or legally.1 In Skinner v. Oklahoma, decided during World War II, the Court struck down Oklahoma’s sterilization law for habitual criminals on equal protection grounds. It refused to apply deferential review to legislation touching a basic civil right, because “[m]arriage and procreation are fundamental to the very existence and survival of the race.”2 A state that decides who may reproduce holds power over the family at its earliest stage, and that power explains why the Court was willing to scrutinize laws in this area.
Three decisions then widen the protection, and each answers a question the last left open. Griswold asked where the interest lives and protected married couples. Eisenstadt asked who may claim it and answered that the individual may, married or single. Carey asked what counts as interference and answered that substantial limits on access count, not only bans on use. The sections below take the cases in that order.
The sequence also matters beyond contraception. In the 1960s the Court moved into contraception, then a somewhat controversial issue, and the public generally accepted its intervention. In the 1970s it moved into abortion and placed sharp constitutional limits on laws restricting access; those limits changed over time but endured for almost fifty years.3 Dobbs v. Jackson Women’s Health Organization ended that regime, as the last section of this chapter explains. The same privacy reasoning later reached same-sex intimacy in Lawrence and, in Obergefell, produced a right to marry grounded in both due process and equal protection.4 The three cases ahead are the foundation for each of those developments.
Griswold and Marital Privacy
Griswold v. Connecticut holds that the Constitution protects a zone of marital privacy emanating from several Bill of Rights guarantees, so a state may not forbid married couples to use contraceptives. A law that operates directly on the intimate marital relationship in this way must satisfy heightened scrutiny, and Connecticut’s could not.
Connecticut made it a crime to use any drug, medicinal article, or instrument for the purpose of preventing conception, and a companion statute punished anyone who assisted, abetted, or counseled another to commit an offense as though the assister were the principal.5 No statute prohibited the sale of birth control devices.6 Estelle Griswold, executive director of the Planned Parenthood League of Connecticut, and Lee Buxton, a Yale Medical School professor who served as the League’s medical director, gave individualized birth control advice to women at the League’s New Haven center, and each was convicted as an accessory and fined $100.7 Because the defendants were providers, the interest the Court vindicated belonged to the married couples the center served.
The statute violated no specific provision of the Bill of Rights, and the Court had said only two years earlier, in Ferguson v. Skrupa, that courts do not substitute their social and economic beliefs for the judgment of elected legislatures.8 Justice Stewart later recalled that this history led the majority to do its best to avoid resting on the Due Process Clause.9 The seven Justices in the majority therefore offered different bases for the right, yet all agreed that the Court may protect liberty interests that appear nowhere in the constitutional text.10 Read the opinions as three attempts to give one liberty a constitutional address. The dissenters denied that any address existed.
Justice Douglas’s opinion for the Court took the route that avoided the due process label. He declined to treat Lochner as a guide, because the Court does not sit as a super-legislature over economic problems, business affairs, or social conditions, whereas this law operated directly on an intimate relation of husband and wife.11 The Constitution nowhere mentions the right of association, the right to educate a child in a school of the parents’ choice, or the right to study a foreign language, he observed, yet the Court had protected each, because without those peripheral rights the specific guarantees would be less secure. The logic is practical: a guarantee of free speech means little if government may forbid a person to read or to join with others, so protection must extend a short distance past the words. Douglas generalized the point into the claim that specific Bill of Rights guarantees have penumbras, formed by emanations from those guarantees, that give them life and substance, and that several of them create zones of privacy: the First Amendment’s right of association, the Third Amendment’s bar on quartering soldiers in any house in peacetime, the Fourth Amendment’s protection of persons, houses, papers, and effects, the Fifth Amendment’s Self-Incrimination Clause, and the Ninth Amendment.12
Applied to Connecticut’s law, the penumbral reasoning turned on the choice of means. The statute concerned a relationship within the zone of privacy, and because it forbade the use of contraceptives instead of regulating their manufacture or sale, it pursued its goal by means with a “maximum destructive impact” on that relationship. A governmental purpose may not be achieved by means that sweep unnecessarily broadly into protected freedoms, and this law did. Use happens at home, so enforcement would require officials to look there. Douglas asked whether the Court would allow police to search the marital bedroom for telltale signs of contraceptive use and called the idea repulsive to the notions of privacy surrounding the marriage relationship. He closed by describing marriage as a coming together that is “intimate to the degree of being sacred” and as an association for as noble a purpose as any in the Court’s prior decisions.13
Justice Goldberg, joined by Chief Justice Warren and Justice Brennan, supplied the second route. He agreed that the concept of liberty protects fundamental personal rights and is not confined to the specific terms of the Bill of Rights, and he turned to the Ninth Amendment to show why.14 That Amendment provides that the enumeration of certain rights shall not be construed to deny or disparage others retained by the people. Madison proposed it to answer the objection that listing particular rights might imply that the unlisted ones had been handed to the government.15 In Goldberg’s view, holding that a right as basic as privacy in marriage may be infringed because the first eight amendments do not mention it would give the Ninth Amendment no effect whatsoever.16 He drew the limits carefully. The Amendment is not an independent source of rights, and he did not claim that the Fourteenth Amendment applies it against the States. It shows that the list of rights was never meant to be exhaustive, and so it supports reading the Fourteenth Amendment’s liberty beyond the listed rights.17
To keep judges from deciding by personal preference, Goldberg said they must look to the traditions and collective conscience of the people and ask whether a principle is so rooted there as to be ranked as fundamental, an inquiry that anticipates the history-and-tradition test taught earlier in this chapter.18 Applying heightened scrutiny, he found that the State, which argued at most that the ban discouraged extramarital relations, could show neither a compelling subordinating interest nor necessity. A more discriminately tailored statute could serve marital fidelity, and Connecticut already had statutes against adultery and fornication.19
Justice Harlan supplied the third route. He refused to join Douglas’s opinion because it implied that due process reaches only rights assured by the letter or penumbra of the Bill of Rights.20 For Harlan the question was whether the statute violates basic values “implicit in the concept of ordered liberty,” and the Due Process Clause stands “on its own bottom,” aided by the Bill of Rights but not dependent on it.21 In his dissent in Poe v. Ullman, on which the Griswold concurrence expressly relied, he explained why such a statute deserved more than a plausibility test: it intruded on the privacy of the home in its most basic sense, and that required strict scrutiny.22 He also answered the worry that this approach lets judges roam. Restraint, he argued, comes from continual insistence on the teachings of history, the basic values underlying society, and the roles of federalism and separation of powers, and the incorporation formula that Black and Stewart advanced would not supply it.23
Justice White, concurring in the judgment, also rested on liberty alone. He noted that the liberty the Fourteenth Amendment protects includes the right to marry, establish a home, and bring up children, and that a statute dealing so substantially with the marriage relationship bears a substantial burden of justification.24 Connecticut offered one justification, its policy against illicit sexual relationships. White accepted that policy as legitimate but could not see it advanced by a ban on marital use, so he found the broad ban of marginal utility to the declared objective.25
Each opinion tied the protection to marriage, and that premise shaped what came next. Both the majority and the concurring opinions emphasize the importance of marriage to the holding.26 Harlan’s Poe dissent said he would not suggest that adultery, homosexuality, fornication, and incest are immune from criminal inquiry however privately practiced, because it is one thing for the State to forbid extramarital sexuality or to say who may marry and another, after acknowledging a marriage, to regulate the details of its intimacy by criminal law.27 Goldberg quoted the same passage to make clear that the holding in no way interferes with a state’s proper regulation of sexual promiscuity or misconduct.28 Because marriage was the premise, the decisions that followed had to ask whether the interest survives outside marriage.
Despite their different addresses, the routes demand much the same of the state. Douglas refused to allow means that sweep unnecessarily broadly, Goldberg required a compelling interest and necessity, Harlan required strict scrutiny, and White required a substantial burden of justification. The holding states the common result as heightened scrutiny of state interference with the intimate marital relationship.
The dissenters rejected the whole enterprise. Justice Black, joined by Justice Stewart, found the law as offensive as the majority did and still would not strike it. He thought privacy a “broad, abstract and ambiguous” concept that can be shrunk or expanded at will, believed the Constitution protects privacy only through specific provisions such as the Fourth Amendment, and concluded that government may invade privacy unless some specific provision forbids it.29 He read the Ninth Amendment as a limit on the federal government, adopted to assure the people that it held only the powers granted, and denied that anything in its language gave judges authority to strike down laws they thought contrary to the collective conscience of the people.30 The worry underneath is restraint. The Court has no machinery for measuring that conscience, so a judge who consults it will end up consulting his own values.31 The majority’s answer was that a reading limited to listed liberties would give the Due Process Clause an extremely narrow meaning and would ignore the purpose of the Ninth Amendment.32
Justice Stewart, who joined Black’s dissent and also wrote his own, called the statute “an uncommonly silly law” but would have let it stand, on the ground that courts do not substitute their social and economic beliefs for the judgment of legislative bodies.33 Silliness was not enough, because the wisdom of a law is a political question that voters can correct, while invalidity requires a constitutional prohibition. A Court that struck down every law it thought foolish would be exercising the very judgment Skrupa had returned to legislatures. Goldberg replied that this reasoning would sustain even a law requiring married couples to be sterilized after two children, since such a law, however objectionable, violates no specific provision either.34
The later history shows which route lasted. Stewart himself wrote in his Roe v. Wade concurrence that Griswold can be rationally understood only as a holding that the statute substantively invaded the liberty protected by the Due Process Clause, and he said he now accepted it as part of that line of cases.35 The penumbral explanation gave no way to tell which other guarantees lay in the shadows, and the Court has since disavowed it and acknowledged that its task is to give meaning to the Fourteenth Amendment’s concept of personal liberty.36 Harlan’s route is therefore the one later decisions follow, and a student asked for the source of the privacy right today should name the liberty protected by the Due Process Clause.37 Lawrence v. Texas later called Griswold the most pertinent beginning point for its own analysis and described the protected interest as a right to privacy that emphasized the marriage relation and the protected space of the marital bedroom.38 The next step is Eisenstadt’s: “[i]f the right of privacy means anything, it is the right of the individual, married or single, to be free from unwarranted governmental intrusion into matters so fundamentally affecting a person as the decision whether to bear or beget a child.”39
Run on Connecticut’s statute, step one finds a ban on use, step two finds married couples, step three finds due process liberty, and step four fails at the means. The State’s only asserted aim of discouraging extramarital relations could be served by narrower laws, and Connecticut already had statutes against adultery and fornication.40
Eisenstadt and Individual Privacy
Eisenstadt v. Baird carried the contraception line from the married couple to the single person. It held that a Massachusetts statute giving married persons access to contraceptives while denying it to unmarried persons violated the Equal Protection Clause, and it announced that the right of privacy belongs to the individual. William Baird lectured on birth control at Boston University and gave a woman a contraceptive device, and he was convicted.42 The statute made it a crime to distribute any article for the prevention of conception, but it excepted registered physicians who prescribe such articles for married persons and registered pharmacists who dispense them on prescription.43 Distribution to single persons was a felony, and distribution to married persons was a misdemeanor unless it ran through physicians and pharmacists.44 Griswold had protected a married couple’s use of contraceptives, so the question was whether an unmarried person held a comparable interest.
The Court resolved the case under the Equal Protection Clause, and that choice shaped the opinion. Equal protection leaves a state free to treat different classes of persons differently, but a classification must be reasonable, not arbitrary, and must rest on some ground of difference having a fair and substantial relation to the object of the legislation, so that all persons similarly circumstanced are treated alike.45 The Court therefore asked whether any ground of difference rationally explained treating married and unmarried persons differently, and it found none. It added that if the statute had impinged on fundamental freedoms under Griswold, the classification would have had to be necessary to a compelling state interest, but the law failed even the more lenient standard, so the stricter one never had to be applied.46 Deciding on the narrower ground let the Court invalidate the law without first settling how far Griswold reaches, which is why a student should describe Eisenstadt as an ordinary equal protection decision and not as a strict scrutiny decision. The clause it applied is the same one the Court had used in the 1950s and 1960s to advance civil rights in the race cases, a reminder that one clause can carry very different kinds of claims.47
Each justification Massachusetts offered failed against the statute’s own structure. Deterring premarital sex could not be the law’s rational purpose, because married persons could obtain contraceptives whatever use they intended, and the statute had no rational relation to that goal.48 The health justification fell for a similar reason. The same devices the state regulated when sold to prevent pregnancy were available without any controls when sold to prevent disease, and the Court found it inconceivable that the need for health controls varies with the purpose of use when the physical act is one and the same.49 A blanket rule fares no better when more precise tests are available. The Court pointed to Carrington v. Rash, where it had refused to tolerate a blanket exclusion of all servicemen from voting even though most had no intention of remaining in the State, because some were bona fide residents and more precise tests could separate them from the rest.50 The common mechanism is that a stated purpose can explain a line only if the line advances it, and a statute whose exceptions leave the supposed harm untouched shows that the stated purpose does not account for the line the statute draws.
Having rejected those purposes, the Court asked whether the statute could stand “simply as a prohibition on contraception.” It declined to decide that question, because “whatever the rights of the individual to access to contraceptives may be, the rights must be the same for the unmarried and the married alike.”51 The reasoning is a dilemma with two horns. If Griswold forbids a ban on distribution to married persons, a ban on distribution to unmarried persons is equally impermissible. If Griswold is no bar to a ban on distribution, Massachusetts still could not outlaw distribution to the unmarried while allowing it to the married, because the evil the state perceived would be identical in both cases and the underinclusion, meaning the law’s failure to reach all of the conduct it targets, would be invidious.52 Either way the marital-status classification falls, which is why the Court never had to say how much protection access to contraceptives receives.
The first horn depends on an answer to the objection that Griswold’s privacy belonged to marriage. The Court conceded that the right of privacy in Griswold inhered in the marital relationship, and it then observed that the married couple lacks “a mind and heart of its own” because it is an association of two individuals, each with a separate intellectual and emotional makeup.53 From that premise came the sentence later cases quote: “If the right of privacy means anything, it is the right of the individual, married or single, to be free from unwarranted governmental intrusion into matters so fundamentally affecting a person as the decision whether to bear or beget a child.”54 The sentence cites Stanley v. Georgia, whose own statement that freedom from unwanted governmental intrusion into one’s privacy is fundamental makes no mention of marital status.55
The reasoning has two parts, and each has its own logic. The first part concerns whose interest is at stake. A decision whether to bear or beget a child is made by a person, so if the interest belonged only to the marital unit, the same decision would lose protection the moment one person made it alone. Protecting the decision itself keeps the right attached to what makes it important. The second part concerns the political check on lawmakers. The Court borrowed Justice Jackson’s point that the surest practical guard against arbitrary government is to require that rules officials impose on a minority be imposed generally, because officials who can pick a few people to burden escape the political price they would pay if larger numbers were affected.56 The unmarried were that minority in Massachusetts, and a law that reached everyone would have drawn opposition from everyone.
A reading of Griswold that stops at the marital bedroom would have left the unmarried with nothing to invoke, since the majority and concurring opinions in Griswold all stressed the importance of marriage.57 Eisenstadt displaced that narrow reading on the question of who may claim the interest, and it did so in a case about distribution, while Griswold had concerned use. Justice White, joined by Justice Blackmun, concurred in the result on a narrower ground that shows what the decision does not require. Because Griswold protected married persons’ use of contraceptives, he reasoned, a restriction on distribution that burdens that right cannot be accepted on faith: the record must show that the restriction is essential to the statute’s purpose, or the relevant facts must be such that a court can take judicial notice of them. Baird had given away vaginal foam, the inquiry does not end merely because some contraceptives are harmful and their distribution may be restricted, and the record contained no evidence that restricting foam was essential to health, so White would not let the conviction stand.58
The first state’s rule raises the question of access that the next section takes up. The later cases show which part of Eisenstadt endured. Planned Parenthood of Southeastern Pennsylvania v. Casey described the freedom to use contraceptives as later guaranteed, under the Equal Protection Clause, for unmarried couples, and it quoted the individual-privacy sentence with emphasis on the word “individual.”59 Lawrence v. Texas agreed that Eisenstadt was decided under the Equal Protection Clause, but it added that the Court went on to state the fundamental proposition that the law impaired the exercise of unmarried persons’ personal rights, and it concluded that after Griswold the right to make certain decisions regarding sexual conduct extends beyond the marital relationship.60 Justice Scalia’s dissent in Lawrence read the same opinion the other way. He called Eisenstadt an equal protection decision with nothing to do with substantive due process and treated its privacy language as dictum that referred back to the penumbral right recognized in Griswold.61 The dissent in Bowers v. Hardwick had likewise read Eisenstadt as extending protection to intimate choices by unmarried as well as married persons.62 The difference matters because a holding about unequal treatment supports only an equal protection claim, while a statement about the individual’s liberty can support a due process claim. The later sections of this chapter, including the one on Dobbs, ask which of those the contraception decisions establish.
Run on the Massachusetts statute, step one finds a line between married and unmarried persons, and step two finds no rational ground for it, because the statute’s own exceptions defeat the deterrence and health purposes the state offered. The analysis therefore ends at the equal protection violation.
Carey and Contraceptive Access
Carey v. Population Services International extends the contraception line from the question of who holds the privacy right to the question of what counts as interference with it. A New York statute let nonmedical contraceptives be distributed to adults only through licensed pharmacists and barred their sale to anyone under sixteen without a physician’s approval, and the Court held that both restrictions unduly burden the privacy right concerning contraception. Neither rule forbade anyone from using a contraceptive, so the case tests whether Griswold and Eisenstadt protect only the decision to use contraceptives or also the means of acting on that decision. Casey later described Carey as extending constitutional protection to the sale and distribution of contraceptives.64 The challenger was a vendor, Population Planning Associates, and the Court let it assert its potential customers’ rights as well as its own, following Craig v. Boren, because the statute inflicted injury in fact on the vendor itself.65
The Court first placed the decision at the center of a recognized group of liberties. The Due Process Clause protects a right of personal privacy that includes independence in making certain important decisions, among them decisions about marriage, procreation, contraception, family relationships, and child rearing and education. The decision whether to beget or bear a child is “at the very heart of this cluster of constitutionally protected choices.”66 The list is the family of rights this chapter has been teaching, so the Court was treating contraception as one more application of a principle already in place and was not announcing a new interest.
New York defended its law by arguing that the earlier decisions did not reach distribution. Griswold struck down a ban on use and had no occasion to discuss laws regulating manufacture or sale, and Eisenstadt rested on equal protection and said only that whatever the right of access may be, it must be the same for the unmarried as for the married.67 The Court rejected that argument because it overlooks the premise of both decisions, which is the individual’s freedom from unwarranted governmental intrusion into the decision whether to bear or beget a child. Griswold’s stress on the sacred precincts of the marital bedroom explained why that statute was especially repulsive, but later decisions made clear that constitutional protection of autonomy in childbearing does not depend on that element.68 Griswold therefore can no longer be read as holding only that a state may not prohibit a married couple’s use of contraceptives. Read in light of its progeny, it teaches that the Constitution protects individual decisions in matters of childbearing from unjustified intrusion by the State.69
A restriction on sale burdens that decision for a practical reason. A decision to avoid conception that a person cannot carry out is a decision in name only, and the Court observed that a total prohibition on sales would intrude on the decision as harshly as a direct ban on use, and might do more damage because a sales ban is more easily and less offensively enforced.70 That observation fits Griswold’s own reasoning closely. A ban on use can be enforced only by inquiry into private life, which is what made the Connecticut statute repugnant, and that discomfort works as a natural brake on how often a state will enforce it. A ban on sale is enforced at the counter. If the right protected people only against the bedroom-search form of intrusion, a state could reach the same result by choosing the more comfortable form.
From that premise the Court drew its standard by analogy to the abortion decisions after Roe v. Wade, which struck down statutes that did not ban abortion outright but limited access to it in various ways. Doe v. Bolton and Planned Parenthood of Central Missouri v. Danforth establish that the same test applies to regulations that burden the decision by substantially limiting access to the means of carrying it out as applies to statutes that prohibit the decision entirely: the regulation may be justified only by a compelling state interest and must be narrowly drawn to express only the legitimate interests at stake.71 The Court located the source of that scrutiny in the childbearing decision, with access protected because it is essential to exercising the decision, and it denied that it was recognizing an independent fundamental right of access to contraceptives.72 That framing explains why only some regulations trigger the scrutiny. The business of manufacturing and selling contraceptives may still be regulated in ways that leave protected choices alone, and even a burdensome regulation can be sustained by a sufficiently compelling interest.73
Applied to New York’s pharmacist-only rule for adults, the analysis found a significant burden. The burden is smaller than under a total ban, yet confining sales to a small fraction of the possible retail outlets makes contraceptives considerably less accessible, reduces the opportunity for privacy of selection and purchase, and lessens the possibility of price competition.74 The Court compared Doe v. Bolton, which struck down a requirement that abortions be performed only in accredited hospitals when no proof showed that the requirement was substantially related to protecting patients’ health, and it found the same infirmity in the pharmacist rule. The State then needed a compelling interest, and the interests recognized in the abortion cases did not fit, because the rule bore no relation to health as applied to nonhazardous contraceptives and the interest in protecting potential life is not implicated in regulating contraceptives.75 Justice Powell, concurring, would have asked first whether the pharmacist-only rule actually imposes a heavy burden on adults, and he saw no basis for applying the Griswold and Roe standard of review without that showing.76 That limit matches the threshold taught at the start of this chapter: the state must be shown to interfere directly and substantially with the liberty before the compelling-interest inquiry begins.
The ban on distribution to minors produced the harder case, and the Justices divided. Justice Brennan’s plurality opinion, joined by Justices Stewart, Marshall, and Blackmun, applied the same logic to persons under sixteen: strict scrutiny governs restrictions on access because access is essential to the right of decision in matters of childbearing that underlies Griswold, Eisenstadt, and Roe.77 The plurality accepted that minors have some constitutional rights while state power over children is greater than over adults, and against that background it was unpersuaded that the state’s interest in discouraging sexual activity among the young was sufficiently “significant.”78 Only four Justices joined that reasoning, so a student should describe the minors holding as a plurality’s and should not present its reasoning as a majority rule.
The plurality’s reasoning turned on evidence. New York argued that making contraceptives hard for minors to obtain would deter early sexual activity by increasing its hazards. The plurality answered that the same argument would support a ban on abortions for minors, or a prohibition on abortion or contraceptive access for the unmarried, whose sexual activity many states also disapprove, and the Court had rejected the argument in each of those settings.79 It also saw substantial reason to doubt that limiting access deters early sexual behavior, since the State had conceded in the District Court that there was no evidence that teenage sexual activity rises with the availability of contraceptives and offered none.80 The consequence is a point about proof: a state that burdens a fundamental right and defends the burden as a rational means to some significant policy must offer “more than a bare assertion” that the burden is connected to that policy.81 The narrow exception for physicians did not rescue the rule, because less than total restrictions that significantly burden the decision must also pass scrutiny, and the deterrence rationale suited the rule poorly: it delegated the State’s disapproval of minors’ sexual conduct to physicians who might apply it arbitrarily.82
The separate opinions show how much the result depended on the weakness of the State’s proof. Justice White concurred only in the result because the State had not demonstrated that the prohibition measurably contributes to the deterrent purpose it advanced.83 Justice Stevens found the means irrational, since a state cannot dramatize its disapproval of teenage sex by exposing minors to a greater risk of pregnancy and disease, and he compared the approach to forbidding the use of safety helmets in order to discourage motorcycling.84 Justice Powell found the provision defective for a different reason: it prohibited parents from distributing contraceptives to their own children and so interfered with parental interests in rearing them.85 Justice Rehnquist dissented, arguing that the Court had denied a state the power to legislate on public morality as it concerns minors.86 The concurrences thus reach the judgment through proof, means-end fit, and parental interests, which is why the case settles less about minors’ own rights than the plurality’s language suggests.
The decision leaves one question open. A restriction on minors’ access that a state could back with evidence that it measurably advances the deterrent goal is not the restriction Carey reviewed, and Justice White’s rationale rested on the absence of exactly that showing.87 The plurality’s reference to “significant” interests, in place of the “compelling” interests that strict scrutiny usually demands, raises the question whether a lesser standard applies to regulation of minors.88 Carey does not answer it. On an exam, treat it as an open question: state that the plurality accepted greater state power over minors, say that the Court has not settled whether that lowers the standard of review, and then apply the evidence requirement, which the plurality and the concurrences all enforced.
Later decisions read Carey in opposite directions, and the difference lies in which part of the opinion each emphasized. In Bowers v. Hardwick the Court relied on the catalogue of privacy decisions that Carey recited, which tied child rearing and education, family relationships, procreation, marriage, and contraception to distinct subjects. It concluded that none of the rights announced in those cases resembled a claimed right of homosexuals to engage in sodomy, found no connection between family, marriage, or procreation and homosexual activity, and called unsupportable any claim that the cases insulate all private sexual conduct between consenting adults from state proscription.89 Lawrence v. Texas looked instead at Carey’s result and concluded that Eisenstadt and Carey, along with Roe, confirmed that the reasoning of Griswold could not be confined to the protection of married adults.90 The next section follows that disagreement. A separate question, taken up in the discussion of Dobbs and other rights, is what remains of the access reasoning now that the Roe line it drew on no longer stands.
Run on New York’s pharmacist-only rule, step one finds a limit on who may sell, step two finds fewer outlets, less privacy, and weaker price competition, step three requires a compelling interest, and step four finds none, because the rule bore no relation to health for nonhazardous products and the interest in protecting potential life was not implicated.91 Run on the ban for those under sixteen, the same steps fail at the last stage, because the State offered an assertion of deterrence without evidence.