Nowhere to Now, Where? Reconciling Public Cannabis Use in a Public Health Legal Framework

Penn State Law Review Penn State Law Review

Article 2

10-1-2021

Nowhere to Now, Where? Reconciling Public Cannabis Use in a Nowhere to Now, Where? Reconciling Public Cannabis Use in a

Public Health Legal Framework Public Health Legal Framework

Daniel Orenstein

Now here to Now, Where? Reconciling Public Cannabis Use in a Public Health Legal Framework

Daniel G. Orenstein*

ABSTRACT

As states continue to legalize recreational cannabis, most have continued to heavily restrict where consumption of newly licit cannabis is permitted. In 2021, New York became the first state to permit open, outdoor public consumption of cannabis. All other legalizing states either restrict lawful cannabis use to private property or allow a small number of licensed venues for consumption outside of public view, the latter approach borrowed from alcohol control. In contrast, a few legalizing jurisdictions outside of the United States have adopted an approach adapted from tobacco control that allows limited outdoor public use but prohibits indoor public use.

Each regulatory option presents individual and population health risks that reflect the complex intersection of health, social inequities, and community norms. Cannabis consumers face uncertain, but potentially significant, health risks from use. Additionally, the relative availability of lawful use locations is inextricable from existing inequities in policing practices and housing. Those who do not use cannabis but are exposed to others' use face possible harms from secondhand smoke and intoxicated behaviors due to use. Such risks are similarly prone to inequitable distribution due to existing employment and housing patterns. Communities as a whole also face risks. Among other risks, changing cannabis norms may increase the prevalence or intensity of cannabis use, and concentration of cannabis outlets in under-resourced communities may also prove detrimental to community health, as seen in the distribution of tobacco and alcohol outlets.

*Visiting Assistant Professor, Indiana University Robert H. McKinney School of Law. J.D., 2011, Arizona State University Sandra Day O'Connor College of Law; M.P.H., 2016, Harvard T.H. Chan School of Public Health; B.A., 2005, University of Arizona. The author thanks Richard Weinmeyer, J.D., M.A., MPhil, for his kind feedback on the manuscript and the students of the Penn State Law Review for their diligent editorial work.

Each public use approach to cannabis carries attendant risks, but a regulatory framework based on the tobacco control model best balances the protection of public health and the promotion of equity and social justice. This model recognizes the parallels between cannabis and tobacco (in addition to those between cannabis and alcohol). Restricting indoor public cannabis use while allowing limited outdoor public use provides a pathway to mitigate the public health risks of cannabis legalization by leveraging an approach that has proven effective in the tobacco context at reducing secondhand exposures and denormalizing smoking behavior.

Table of Contents

Smoke is smoke. This messaging permeates advocacy from public health-oriented anti-smoking organizations, state agencies, and others seeking to tightly regulate cannabis and educate the public on its use, health effects, and similarity to tobacco.' Although this framing typically describes where cannabis consumption should be prohibited, rarely does it address where consumption should be permitted. Prior to New York's legalization in March 2021,2 every U.S. state that legalized recreational (or "adult use") cannabis prohibited open public consumption, instead restricting lawful use to private residences or a small number of licensed locations.3 In practice, however, such restrictions have not inhibited more brazen cannabis consumers from using cannabis on public sidewalks or in similar outdoor settings where it would be common to encounter tobacco use. For example, in San Francisco, California, a jurisdiction with notably permissive enforcement policies,4 public cannabis consumption is a

law prohibition.6 As

legalization common sight,' despite a state reconfigures legal frameworks and social norms across a growing number of states, the sight of public cannabis consumption may become as unremarkable as the sight of a cigarette or vaping device. Despite legitimate public health concerns, this would be a positive development. This Article argues that legally authorizing public cannabis consumption in the same limited outdoor locations as tobacco use is the best legal approach to balancing public health risks and equity in the context of adult use cannabis legalization.

Cannabis legalization is no longer the audacious goal of a small group of activists or a mere hypothetical for legal or public health theorizing. Advocates have succeeded in shifting public support to favor legalization. Between 1999 and 2019, the proportion of U.S. adults supporting legalization more than doubled from 31% to 67%, now reflecting majority support among all generations born after 1945.' Support for cannabis legalization has also extended to the ballot box. Between 2012 and 2019, voters in eleven states approved nine of the twelve adult use legalization initiatives that appeared on state ballots." In 2020, all four legalization measures that appeared on state ballots passed by comfortable margins.9 Voter support for three ofthese measures equaled or outperformed support

for either presidential candidate' 0 amid the highest turnout election in American history." As of April 2021, more than four in ten Americans live in a state that has legalized adult use cannabis, and more than seven in ten live in a state that has legalized medical cannabis.1 2 Legalization is thus already a reality in much of the country and likely to continue to proliferate. However, this is not the end of the debate, but rather the beginning of the more challenging task of choosing between various regulatory models.1 3 Accordingly, this Article bypasses consideration of

whether cannabis should be legalized in favor of attention to how to do so without unduly jeopardizing public health.

Cannabis 14

is unique from a regulatory standpoint because it

combines elements of tobacco (primarily consumed by smoking)," alcohol (intoxicating),1 6 and pharmaceutical drugs (medical utility). Additionally, cannabis's regulatory history is steeped in racism.18 These distinctive characteristics further compel an approach of regulatory adaptation, rather than transplantation of parallel approaches to analogous substances into cannabis policy in toto.19 Concerning specific policy questions, however, reference points in existing law are highly instructive. In the case ofpublic use, tobacco and alcohol control models are especially pertinent, but they take diametrically opposed paths. For example, most jurisdictions prohibit the public consumption of alcohol, shifting use to private residences and commercial establishments such as bars and restaurants. 20 Conversely, and mainly due to tobacco control advocates' efforts, a large and growing number ofjurisdictions have steered toward prohibiting tobacco consumption in most indoor and some outdoor public

locations. 2 outdoor locations such as sidewalks, curbs, and alleyways.

1These prohibitions have shifted tobacco use to less desirable

superficial

resemblance,

thus making public

The relationship between cannabis and tobacco consumption, particularly considerations of secondhand smoke exposure, extends far beyond cannabis consumption one of the most significant and complex aspects of legalization from a public health perspective. 22 Consequently, cannabis legalization presents extensive legal and policy challenges. These challenges are exacerbated by the underdeveloped evidence base regarding the drug's health effects 23 and the general paucity of examples of operative, comprehensive, and successful legal models for adult use.24 Therefore, parallel frameworks and evidence related to other substances provide crucial context for constructing sound cannabis laws.

As with tobacco, alcohol, and many pharmaceuticals, the fact that cannabis products can be harmful to individual and population health is not inherently fatal to an argument for legalization. 25 However, the

potential for such harms does necessitate balancing costs and benefits in cannabis policymaking. Regulation of products and their use is a spectrum rather than a binary, much like the harmfulness or healthfulness of such products themselves. 26 Against this backdrop, this Article uses a public health law lens to assess the benefits and detriments of four general models of regulating public cannabis use: (1) consumption in private residences (3) only; (2) consumption consumption in limited outdoor locations (tobacco model); and (4) unrestricted public use.

in limited indoor sites (alcohol model);

Based on this analysis, the tobacco model provides the best guidance for public cannabis use. Permitting cannabis consumption in the same limited outdoor locations where tobacco use is permitted is legally sound and consistent with public health goals. This conclusion is undeniably controversial from a public health perspective. 27 Even so, this conclusion flows from the foundational goals of tobacco and alcohol control laws, the social justice implications of available legal options, and the demonstrated efficacy of denormalization in the tobacco control context.

This Article proceeds in four parts. PartI outlines and contextualizes the prevailing public use models in U.S. jurisdictions for tobacco and alcohol, provides background on existing cannabis public use laws in legalizing states and internationally, and constructs the four-part typology of policy models for regulating public cannabis consumption noted above. Part II explains and distinguishes the potential risks of each model. These potential risks encompass risks to cannabis consumers, risks to others, and indirect social and public health risks. Part III evaluates the balance of benefits and harms under each approach. Part IV argues that, based on current evidence, adopting a tobacco control approach to cannabis by

permitting open public cannabis consumption in limited outdoor locations best serves the goals of public health and social justice.28

Cannabis remains illegal under federal law,29 yet legalization has rapidly galloped from the political fringe to the mainstream. Over the past twenty-five years, thirty-six U.S. states have legalized cannabis for medical purposes, and eighteen of those states have also legalized recreational ("adult use") cannabis. 30 The federal government has the legal authority to extinguish state-legal cannabis industries by enforcing the Controlled Substances Act,31but, in practice, the federal government has not significantly interfered with states' cannabis legalization efforts for the past decade. 32 Few accurately predicted this tectonic legal shift. Despite vocal activism in the 1960s, only 12% of Americans supported cannabis legalization at the end of that decade.33 Support increased haltingly over the next few decades, but reached only 31% as of the early 2000s.34 By 2020, however, two-thirds of Americans favored legalization, 35 federal legalization considerable congressional support,36 and entities such as the National Cannabis

decriminalization

attracted

bills

Industry Association publicly proclaimed to represent thousands of businesses and professionals in the nascent legal cannabis industry.37

Amidst the tumult of rapidly changing state laws, policymakers and regulators have often had to adapt quickly and work within the boundaries set by voter initiatives-the primary avenue for adult use legalization.38 Consequently, most legalizing states' regulatory frameworks are at best imperfect-and often severely deficient-when judged against public health best practices. 39 Public use looms as one of the most challenging considerations in forming cannabis legalization policy. However, in answering the public use question, principles and lessons drawn from tobacco control, alcohol control, and existing cannabis laws in other states and countries are enlightening.

A. Historyand Context ofthe Tobacco ControlModel

The history and context of the tobacco control model revolve around the fact that exposure to secondhand tobacco smoke leads to wellestablished adverse health effects. Secondhand smoke causes premature death in children and nonsmoking adults, increases children's risk of numerous ailments (e.g., sudden infant death syndrome, acute respiratory infections, ear problems, and severe asthma), negatively affects cardiovascular functions, and causes coronary heart disease and lung cancer.4 0 In the 1950s, evidence of the health harms of smoking began to accumulate at a time when nearly half ofU.S. adults (including half ofall

physicians) were smokers. 4 1 Although the landmark 1964 Surgeon General's Report initiated a shift in behavior and public opinion against tobacco smoking, 42 smoking nevertheless remained permissible in the vast majority of public spaces until the late 20th century. Among countless examples, in 1979 only 1.2% of U.S. hospitals banned smoking, 4 3 and a comprehensive ban on smoking on domestic airline flights arrived only in 1990.44

Tobacco control activists have fought an enduring battle to remove smoking from public locations such as bars, restaurants, and other workplaces. 45 Modern tobacco control best practices prohibit smoking in all indoor public places and outdoor locations where members of the public are more likely to be exposed, such as near building entrances, public transit, outdoor seating areas, playgrounds, and outdoor public events. 46 More recently, such smoking restrictions have extended to cover newly popular electronic smoking devices, such as e-cigarettes and similar products. These devices do not emit actual smoke but rather what is commonly (albeit erroneously)47 called "vapor," which nevertheless contains potentially harmful aerosolized particles. 48

The World Health Organization ("WHO") Framework Convention on Tobacco Control ("FCTC")49 and its implementing guidelines" together represent a global consensus standard on tobacco control best practices." Although the FCTC was not designed to apply to cannabis, its evidence-based public health approach is nevertheless instructive for cannabis policy. Protecting persons from tobacco smoke follows from a duty implicit in "the right to life and the right to the highest attainable standard of health," as recognized in various national constitutions, international legal instruments, and the preamble of the FCTC itself 52 The FCTC's implementing guidelines explicitly note that the government's duty to protect all persons from threats to fundamental rights and freedoms includes protecting these persons from tobacco smoke.53

Accordingly, the FCTC and its implementing guidelines take an unambiguously strong position against public tobacco smoke exposure. Article 8 of the FCTC obligates Parties to adopt and implement measures to protect persons from passive tobacco smoke exposure "in indoor workplaces, public transport, indoor public places and, as appropriate, other public places."5 4 The FCTC's guidelines also declare that "[a]ll indoor workplaces and indoor public places should be smoke free."55 The guidelines recommend adopting a broad definition of "public places," including any places accessible to the public or used by the public, regardless of legal ownership or access rights.56 Regarding outdoor and quasi-outdoor public places, the guidelines instruct Parties to consider evidence of potential health hazards and "act to adopt the most effective protection against [tobacco smoke] exposure . ."7 Alternative approaches falling short of a comprehensive smokefree environment, "including ventilation, air filtration[,] and the use of designated smoking areas[,]" are regarded as ineffective and inadequate under the FCTC's guidelines.58

However, a public health approach to public tobacco smoking and passive exposure can accommodate some limited outdoor public use. For example, in its Model Ordinance, leading U.S. tobacco control advocacy group Americans for Nonsmokers' Rights ("ANR") provides for the prohibition of smoking in a wide variety of enclosed public places 59 and some outdoor spaces, such as shopping malls, parking structures, and amphitheaters. 60 ANR's Model Ordinance also prohibits smoking on city and county property and adjacent land, as well as "[w]ithin a reasonable distance" (e.g., 15-25 feet) "outside entrances, operable windows, and ventilation systems of enclosed areas where smoking is prohibited." 6 1 Furthermore, ANR's Model Ordinance imposes similar distance-based restrictions for a wide range of outdoor areas, including restaurants, bars, gambling facilities, arenas, stadiums, amphitheaters, recreational areas and their parking lots, playgrounds, public events, transportation facilities, and outdoor common areas of various residential facilities.62 Although these restrictions are far-reaching, outdoor smoking in locations beyond the specified distance is implicitly unrestricted. 63 ANR's Model Ordinance is thus a comprehensive regulatory plan rather than a blanket ban on public smoking.

to secondhand

smoke [,]" 64 explicitly citing

In its Model Ordinance, ANR declares that "there is no safe level of the highly exposure influential 65 ventilation standard for indoor air quality published by the American Society of Heating, Refrigerating and Air-Conditioning Engineers ("ASHRAE"). ASHRAE takes the position that only a complete ban on indoor smoking can effectively eliminate the health risk of indoor is no current authority for identifying an exposure because

there

acceptable level of environmental tobacco smoke. 66 ASHRAE's current Standard 62.1 (Ventilation for Acceptable Indoor Air Quality) defines environmental tobacco smoke or secondhand smoke as including sidestream smoke (from the lit end of a tobacco product), exhaled mainstream smoke (exhaled by the smoker), and the emissions of electronic smoking devices. 67 Importantly, ASHRAE also includes smoke and emissions from cannabis products in this definition. 68

Tobacco control advocates have vigorously pursued a public health approach in seeking to eliminate smoking from bars, restaurants, and other indoor public places and workplaces. Through significant effort, these advocates have achieved widespread adoption of comprehensive smokefree laws in states and localities around the country. As of April 2021, twenty-seven U.S. states, the District of Columbia, Puerto Rico, the U.S. Virgin Islands, and over 1,100 cities and counties, covering over 60% of the U.S. population, have laws in place that meet a 100% smokefree standard for all non-hospitality workplaces, restaurants, and bars.69 Less comprehensive casinos) cover many additional jurisdictions.70

(e.g., exempting

restrictions

Smokefree laws, such as those noted above, operate by targeting either individual smokers or persons and entities in control of public areas, such as store owners and property managers. Such laws may impose civil fines on smokers for use in a prohibited area or on owners for failing to adopt a smokefree policy and take reasonable enforcement actions. 71 These laws reduce smoking, in part because they frame smoking as a disfavored and antisocial act.72 While smokefree laws do not directly prohibit a person from smoking, these laws constrain the act to less

desirable areas. This constraint makes smoking both less convenient for the smoker and less visible to others, thereby expressing a measure of societal disapproval 3 without total prohibition.

B. Historyand Context ofthe Alcohol ControlModel

In contrast to the tobacco control model, the prevailing U.S. alcohol control model generally restricts outdoor public use, thereby shifting lawful use of alcohol primarily to indoor locations, such as bars and restaurants. While U.S. alcohol laws are complex, most states generally prohibit consumption ofalcohol in public places either directly under state law or indirectly through the aggregation of local prohibitions.74 Eleven states have specific exceptions allowing public consumption in defined "entertainment districts," for example Savannah's Historic District and New Orleans' French Quarter.75 Prohibitions on public alcohol consumption did not become commonplace in the U.S. until the 1970s, and adoption, enforcement, and penalties have varied widely. 76 In large part, these public consumption laws replaced former prohibitions on vagrancy and public drunkenness when the latter came under judicial scrutiny for criminalizing status (e.g., alcoholism) rather than conduct.7 7 Thus, the underlying crime shifted from the status ofbeing drunk in public to the act of drinking in public.

Like tobacco use, alcohol use is associated with a host of individual and public health harms. Among other harms, excessive alcohol use increases the risk ofinjuries, violence, alcohol poisoning, and risky sexual behaviors. 78 Long-term risks include development of chronic diseases, various cancers, and other problems affecting social functioning, mental health, learning, memory, the liver, the brain, and the cardiovascular, is also associated with digestive, and

immune systems.79 Alcohol

significant harms to others, including interpersonal violence and driving under the influence, the latter responsible for over 10,000 annual deaths in the U.S.80 Unlike tobacco exposure, however, such alcohol-related harms to others require additional action beyond consumption, such as physical violence or driving a car.

Beyond its myriad social and behavioral impacts, the COVID-19 pandemic of2020-2021 also precipitated several legal changes, including in alcohol control. Several jurisdictions altered or suspended alcohol restrictions by permitting bars and restaurants to serve takeout alcohol, allowing open public consumption, or loosening rules for alcohol delivery."i These changes were intended as temporary interventions to assist service-based industries suffering economic losses due to social distancing measures,82 but some jurisdictions may never return to prior rules.83 These changes, along with the ubiquity of alcohol consumption locations, illustrate the general absence of significant societal disapproval of alcohol consumption, standing in stark contrast to the tobacco control approach.

C. ExistingAdult Use CannabisLaws

Since 2012, a significant number of U.S. states and a small number of countries have legalized adult use cannabis and have taken a variety of approaches to public cannabis use. In the U.S., alcohol control has provided the dominant policy model, but a number of states have instead adopted a more restrictive approach that limits lawful cannabis use to private property. In contrast, the tobacco control model has been more influential in shaping cannabis policy in legalizing jurisdictions outside the U.S.

1. U.S. Jurisdictions

As of August 2021, possession and use of non-medical "adult use" cannabis is legal8 4 in eighteen states, the District of Columbia, and two

U.S. territories. 85 Colorado and Washington passed the first adult use cannabis laws by voter initiative in 2012, followed by eleven successful initiatives in other states from 2014-2020.86 New Jersey voters approved a legislatively referred constitutional amendment to legalize "retail marijuana" in 2020, with the legislature finalizing enabling legislation in February 2021.87 Legislatures also enacted adult use laws in Vermont, in Illinois, and Connecticut, New York, Virginia, and New Mexico followed suit in the first six months of2021.89

in 2018-2019,88

legislatures

territories

two

In many jurisdictions, the political push for legalization explicitly referenced alcohol policy, asking voters to "regulate marijuana like alcohol." 90 Consequently, many cannabis laws are based on state alcohol laws. 91 This basis is apparent in that consumption in visible public locations is prohibited in every legalizing state except New York (and, to an extent, Connecticut, as discussed below). Thus, in the current legal environment, cannabis consumption is generally prohibited in all locations where tobacco use is prohibited and in most locations where alcohol use is prohibited.92

However, some jurisdictions do allow cannabis consumption in specific, limited sites. 93 These limited sites include either independently operating businesses or sites connected to a cannabis retailer. Most existing and recently proposed laws that take this approach prohibit tobacco or alcohol consumption in cannabis consumption areas. 94 Of the eighteen states with adult use cannabis laws as of August 2021:

legislation

legalization

Examining each of these states' cannabis consumption laws in turn, New York's 2021 is unique. New York's legislation not only allows for cannabis consumption at licensed on-site locations, 95 but also at many other locations where tobacco consumption is permitted, such as designated outdoor dining areas, hotel rooms, and membership associations. 96 New York's law also generally allows outdoor public use, unless otherwise prohibited by the state's smokefree air law. 97 Connecticut will generally allow municipalities to regulate public cannabis consumption on property owned by or under the control of the municipality. 98 However, the state will also require municipalities with

populations over 50,000 to designate a place within the municipality where public consumption is permitted. 99 Connecticut's approach to on-site consumption has yet to be determined, with agency recommendations due by January 2023.100

California permits on-site cannabis consumption areas in retailers, lounges, and tourism venues (e.g., tour buses and specific public events).' 0' The state delegates authority to local jurisdictions to approve on-site consumption locations, which several have done. 0 2 For example, San Francisco has authorized multiple cannabis lounges 0 3 and provided temporary event permits for at least eight large outdoor events, including two highly attended annual concerts at the city's Golden Gate Park and an annual event centered around the park's "Hippie Hill" celebrating the unofficial cannabis holiday "4/20." 104

Alaska permits consumption in cannabis retail establishments, 05 issuing its first two approvals for such sites in early 2020.106 Alaska's rules for on-site consumption require that the designated area be outdoors or physically separated from the retail space with a secure door and separate ventilation system. 0 7 Employees must be able to monitor the consumption site from a smoke-free area.1 08 Consumers may not bring their own cannabis for consumption or use cannabis concentrates, alcohol, or tobacco products.1 09 Localities can ban operation ofthese sites or any other cannabis businesses." 0

Illinois similarly approved its first cannabis consumption lounge in early 2020."1 Notably, Illinois allows localities to authorize consumption not only at cannabis retailers, but also at certain tobacco retailers."l2

Michigan

authorizes

licensure of "designated consumption establishments," as well as "temporary marihuana events.""1 3 Designated consumption establishments must physically separate consumption areas from areas where smoking is prohibited and provide a smoke-free space from which employees can monitor the consumption area. Designated consumption establishments must also have a ventilation system that directs air from the consumption area to outside the building and incorporates filtration adequate to remove visible smoke and eliminate odor at the property line." 4

Similarly, New Mexico will license outdoor and indoor "cannabis consumption areas," but smoking is only permitted if the area "is in a designated smoking area or in a standalone building from which smoke does not infiltrate other indoor workplaces or other indoor public places where smoking is otherwise prohibited.""

Likewise, New Jersey will license "cannabis consumption areas," which must be on the premises of a cannabis retailer and may be either outdoors or in a physically separated area.i 6

Colorado allows consumption on cannabis bus tours and, pursuant to a 2019 law, will begin authorizing consumption licenses for restaurants, hotels, and other businesses ("marijuana hospitality businesses") that will allow for indoor smoking and vaping.ii? However, as of January 2021, it does not appear that any such licenses have been granted.iS

Similarly, Massachusetts plans to license consumption lounges.ii 9 Yet, the state has not done so as ofAugust 2021. One existing consumption venue has occupied a legal gray area since opening in 2018, operating as

a private club open only to dues-paying members12 and drawing the ire of some local officials.121

Montana has yet to finalize enabling regulations following a 2020 ballot initiative, but language in the state's measure appears to allow the responsible state regulatory body to authorize on-site consumption.1 22

Washington,123 Oregon,1 24 Vermont,125 Maine,126 Arizona,127 and Virginia1 28 prohibit on-site consumption. Nevada also prohibits on-site consumption, but this provision expires in 2021.129 Nevada's prohibition has been a point of contention regarding tourism, a major state industry,1 30 as visitors who purchase cannabis products lack places to lawfully consume them. The only current exception to Nevada's prohibition is a dispensary located near downtown Las Vegas on land within a sovereign "colony" of the Las Vegas Paiute Tribe. Because of its location, this dispensary is not subject to the state's prohibition and allows on-site consumption. 131 Nevada passed legislation in June 2021 that will allow onsite consumption lounges beginning October 1, 2021.132

2. Non-U.S. Jurisdictions

Turning to jurisdictions outside of the United States, Uruguay in 2013 and promulgated regulations in 2014.133 legalized cannabis is highly centered on government Uruguay's form of legalization involvement. Except for individual home cultivation and limited cooperatives, the government oversees all cannabis production from seed to sale and limits commercial purchases to licensed pharmacies.13 4 Smoking or lighting cannabis products is prohibited in enclosed public spaces, workplaces, public transit vehicles, health care facilities, and schools.'3 5 Outdoor smoking spaces are allowed with restrictions on permissible enclosures.1 36 These rules generally parallel Uruguay's rules for tobacco.

some provinces,

such as Ontario,

Canada's 2018 national adult use cannabis legalization framework permits provinces to set many of their own rules, including for public cannabis consumption. 137 consumption is generally permitted in locations where tobacco use is allowed. 38 Ontario permits cannabis smoking and vaping in most outdoor public locations, unless otherwise restricted by municipal bylaws.1 39 However, Ontario prohibits smoking or vaping cannabis in enclosed public places, enclosed work places, public areas within twenty meters of places where children gather (e.g., school grounds, playgrounds), publicly-owned sports fields other than golf courses, reserved seating areas of outdoor sports and entertainment locations, bus shelters, and outdoor grounds ofgovernment office buildings, among other locations. 40 Similarly, Alberta'41 and British Columbia14 2 allow cannabis smoking and vaping in most, though not all, locations where tobacco smoking and vaping are permitted. Local laws, however, may be more restrictive.

Locations where cannabis use is prohibited, but tobacco use may be permitted include sports fields, playgrounds, hospital or school property, zoos, and outdoor pools. 43 Additionally, several Canadian cities have long had quasi-legal cannabis lounges and Netherlands-style "coffee shops," though their status under Canadian law is questionable at best.144

The Netherlands' coffee shop1 45 model is popularly thought of as legalization, but a more accurate picture is simply one of tolerance and formalized non-enforcement.1 46 Many restrictions apply under the Netherlands' model, and local authority is key. Most municipalities do not allow coffee shops, and those that do require licensure. 47 Other than in the coffee shops, public cannabis use is illegal. 48

Somewhat similarly, Portugal decriminalized cannabis and other drugs in 2001, but drug possession remains an administrative offense that is "energetically" enforced.1 49 As such, public cannabis use, like use and possession generally, remains illegal, albeit not a criminal offense.

D. Typologyfor Future Public CannabisUse Regulation

"Legalization" is an umbrella term that encompasses a wide array of policy options.50 While there are many similarities between existing state cannabis regulatory regimes, there are also critical differences, and future frameworks may further diverge. With respect to public consumption, four general models represent the viable legal options: (1) private residences only; (2) licensed indoor public locations (alcohol model); (3) limited outdoor public locations (tobacco model); and (4) unrestricted public use. Although some aspects of these models are not mutually exclusive in

practice,' 5' each category will be considered separately to better illustrate relevant principles.

The first model, permitting cannabis use only in private residences, legalizes cannabis use but theoretically hides it from public view. Six adult use states currently use this model. By prohibiting consumption in any location visible to the public while not allowing on-site consumption, these states confine lawful cannabis use to private residences. The second model, permitting cannabis use in licensed indoor public locations, replicates the alcohol control model. This model is the current approach used in eleven states that allow or plan to allow on-site consumption in age-restricted establishments.s2 The third model permits cannabis use in limited outdoor public locations and replicates the tobacco model by moving use away from indoor locations to less favorable outdoor sites.1 53 This model is operative in Uruguay and some Canadian provinces. The fourth model, unrestricted public use, would resemble the bygone era of widespread tobacco consumption in the U.S. but is currently hypothetical. Two caveats are appropriate to note at this point. First, this Article's analysis applies primarily to inhaled cannabis products rather than alternatives such as edibles, which do not produce secondhand smoke or aerosols. Legalization of only non-inhaled cannabis products is a potentially credible alternative but for two issues: First, too little is currently known about the health effects ofsuch products vis-i-vis smoked cannabis. 5 4 Second, smoking and vaping remain the dominant modes of cannabis use."5 When the people or legislature ofa state legalize cannabis, smoked flower is most likely what they envision. Restricting legal products or public use to a small subset of largely unfamiliar products with unverified health effects is neither wise public health policy nor consistent with the intent of such laws.

Second, this typology excludes a total prohibition (i.e., "nowhere") because this Article proceeds from an assumption of legalization. Until Colorado and Washington's landmark 2012 initiatives, cannabis use was

prohibited in all locations across the United States. Public opinion appears firmly in favor of cannabis legalization in the abstract, 5 6 and medical and adult use legalization initiatives have been met with increasing success over the past two decades.1'" It is always possible that political winds may shift, as they did between the 1970s and 1980s,1 58 and the results of ongoing state legalization experiments are still developing. However, the U.S. appears to be on a trajectory toward some form of cannabis legalization. Moreover, from a public health standpoint conscious of social justice, the prohibition approach has been disastrous. Cannabis prohibition (and the War on Drugs more broadly) has occasioned appalling enforcement inequities 159 and, even at that exceedingly high price, has largely failed to curb consumption.1 60

II.

Cannabis use presents at least three distinct categories of risks moderated by location of use and operative legal framework. First, cannabis consumers face direct health risks associated with cannabis use and indirect risks posed by encounters with law enforcement and threats to housing security, both of which are inextricably linked to social justice and equity. Second, cannabis use presents risks to others, including direct harms from secondhand exposure to cannabis product emissions and indirect harms due to intoxicated behavior, such as motor vehicle accidents and social disturbances. Third, cannabis use presents indirect societal risks such as increased use prevalence or frequency, changing norms for cannabis use and smoking generally, erosion of existing clean indoor air

laws, and potential adverse effects of concentrating cannabis outlets in vulnerable communities.

A. Risks to Consumers

1. Negative Health Effects of Cannabis Use

The short-and long-term negative health effects of cannabis use are a point of contention and remain understudied for such a widely used drug. The gap in research is due primarily to research restrictions created by the drug's Schedule I status under federal law.' 6' However, existing evidence is sufficient to warrant caution in cannabis policymaking. The National Academies of Sciences, Engineering & Medicine's milestone 2017 report on the state of scientific evidence found substantial or moderate evidence of associations between cannabis use and a variety of adverse health effects, including the development of schizophrenia and other psychoses, risk of motor vehicle accidents, lower birth weight (with maternal use), worse respiratory symptoms and bronchitis episodes, risk of pediatric overdose injuries, cognitive impairment, incidence of certain mental health outcomes and conditions, and development of use disorders for various substances.I6 2

Additionally, daily or near-daily cannabis use appears

to be associated with impaired cognition, poorer education attainment, increased risk of psychotic symptoms or diagnosis, and cardiovascular risks.1 63 Such statistical relationships are not necessarily causal, but they necessitate a careful approach to legalization. Other risks are also likely to emerge. Cannabis smoke and tobacco smoke are highly similar, differing primarily in the presence or absence of cannabinoids and nicotine.1 64 Further research may well identify that some, or many, ofthe myriad risks of tobacco smoking 6 ' also apply to cannabis smoking, despite the current

underdeveloped status of such evidence. As the aphorism cautions, "absence of evidence is not evidence of absence."

2. Potential for Law Enforcement Encounters

Enforcement of cannabis laws has been persistently unequal. Despite similar use rates across racial and ethnic groups,166 racial arrest disparities are pervasive nationally. On average, a Black person is nearly four times more likely than a white person to be arrested for a cannabis-related crime. 167 Recent legalization and decriminalization efforts have reduced cannabis arrest rates but have not eliminated disparities. Black persons remain more likely to be arrested for cannabis possession in every state.i68 For example, arrest rates dropped dramatically following Colorado's 2012 legalization. Yet five years later the arrest rate for cannabis-related crimes remained five times higher for Black persons compared to white persons.1 69 Law enforcement disparities are deeply entrenched and tied to numerous systemic factors; 170 minor revisions to the criminal code will not eradicate them.

remaining prohibitions,

In legalizing jurisdictions, law enforcement may shift focus from including public cannabis possession consumption. This shift in focus potentially undermines the social justice rationales for cannabis legalization. For example, in 2014, Washington, D.C. decriminalized cannabis possession before legalizing it in 2015.171 Cannabis arrests markedly declined, but arrests for public consumption later increased, with greater increases for Black persons.1 72 This increase in arrests of Black persons quickly caused arrest rates to return to pre-

legalization levels.1 73 Black persons represented 80% ofthose arrested for public consumption following legalization, while only representing 47% of D.C.'s population. 74

Beyond such arrests is the potential that enforcement of even minor regulatory prohibitions may escalate into dangerous or deadly situations. Consider the deaths of Eric Garner and George Floyd. In 2014, New York police officers in street clothes attempted to arrest Mr. Garner for selling untaxed loose cigarettes. Garner briefly slipped free, and an officer applied a prohibited chokehold that continued despite what became Garner's dying words, "I can't breathe." 7 7 George Floyd uttered the same plea in 2020 before dying under the knee of a Minneapolis police officer during an arrest for using a counterfeit $20 bill at a convenience store. The words have become a rallying cry for widespread protests against police violence.1 76 Both encounters were prompted by minor violations but led to tragic ends. These high-profile examples sadly represent a mere fraction of the true impact of such encounters."?

While most

law enforcement encounters do not follow this heartbreaking trajectory, there are far too many that do. These tragic lead to devastating consequences, particularly for Black encounters communities and other communities of color. The looming potential for such escalation of even trivial criminal matters is an issue of profound moral concern. Like other mala prohibita, public cannabis use is not inherently wrongful or immoral; it is punishable only as part of a broader regulatory framework to protect public welfare.1 78 Much more can be, should be, and has been said about the relationship between law enforcement and the communities that law enforcement is meant to serve and protect. Being that one ofthe explicit goals of cannabis legalization is inequitable outcomes associated with to reduce criminalization of cannabis,179 legalization should at minimum be structured to decrease-and certainly not to increase-potential law enforcement encounters.

the negative and

3. Cannabis, Housing, and Equity

In addition to inequities related directly to law enforcement, cannabis policymaking must also consider the broader context of housing inequity. U.S. homeownership rates declined from 2004 to 2016.180 However, the rate of decline and overall trajectory varied significantly by race and ethnicity. 8" The most significant decline over that period was among Black households (49.7% to 42.2%), while declines were smaller among white (76.0% to 71.9%) and Hispanic (48.1% to 46.0%) households. 8 2 Among these three groups, not only did Black households have the lowest overall rate and largest decline, but they were also the only group to experience a decline from 2015-2016.183 Additionally, new units added to the home rental supply have primarily been priced at the upper end ofthe market.18 4 Consequently, the overall rental supply has not kept pace with increased demand, contributing to rising rents and a persistent shortage of low-cost rentals. 85 These market

longstanding structural disparities, suggesting that policies that harm renters will disproportionately impact communities of color and others who have suffered inequities due to the War on Drugs. Cannabis consumers whose landlords prohibit cannabis use lease non-renewal, termination, or eviction if their use is discovered, threatening to worsen existing housing disparities.1 86 Those seeking to avoid such outcomes by consuming outside of their rental face risk of detection by law enforcement, again with differential impacts due to existing housing disparities. As explained by the Massachusetts Cannabis Control Commission:

trends exacerbate underlying and

face potential

Living in federally subsidized housing exacerbates cannabis-related legal problems. Under federal law, persons using, possessing, or selling cannabis (or other controlled substances) can be evicted from public housing, often through expedited procedures.' Moreover, evictions can follow from activities of persons considered to be under a tenant's control (i.e., visitors and guests).1 89 Such policies are rooted in the pinnacle ofthe War on Drugs in the 1980s and 1990s and continue to have substantial negative impacts today due to persistent public housing shortages.1 90

to be using drugs.1 92 Some exclusions

The Anti-Drug Abuse Act of 1988 required public housing leases to include clauses allowing eviction for drug use and other behaviors deemed to threaten other tenants' safety.191 Subsequent legislation imposed additional mandatory and discretionary restrictions, including a mandatory three-year ban on readmission of tenants evicted for drug-related crimes with discretion to apply longer exclusions and to exclude applicants believed extend to entire households, including mandatory denial of admission for three years if any household member has been evicted from federally assisted housing for drug-related criminal activity.193 Officials may (but are not required to) admit a household if the offending member has successfully completed an approved drug rehabilitation program, is imprisoned, or has died.1 94 Officials must also prohibit admission if any household member is "currentlyengaging in illegal use of a drug" or if there is "reasonable cause to believe that a household member's illegal use or pattern of illegal use of a drug may threaten the health, safety, or right to peaceful enjoyment of the premises by other residents."195

Similar restrictions apply to federal housing vouchers.1 96 Persons in need face substantial barriers to accessing federal rental assistance programs. Fewer than one in four eligible low-income renter households receives such assistance; among households without children, this number drops to fewer than one in ten.1 97 Compounding these formal legal barriers, many public housing authorities apply policies that appear to exceed even the strict rules dictated by federal law.1 98

In contrast, public housing admission and voucher restrictions relating to alcohol apply only to "abuse or pattern of abuse," rather than mere use.1 99 This divergence is unsurprising given the difference in federal legal status, but the stark legal contrast between use and "abuse"2 oo is revealing. Federal law and policy, and that of most states, continues to acknowledge that some harmful substances, such as alcohol, may be lawfully used up to the point that related behaviors necessitate legal intervention. For other substances, including cannabis, the law makes no such distinction, treating all use equivalently irrespective of frequency, dosage, level of intoxication, or resulting behavior. Absent changes in federal law, cannabis use restrictions are thus likely to contribute to existing inequalities intertwined amongst the complex intersection ofrace, poverty, and affordable housing access.

A related but distinct set of legal concerns attach to persons experiencing homelessness. Nationally, chronic homelessness declined from 2010 to 2016, particularly for families. 20i Non-chronic homelessness also declined to a lesser extent.20 2 However, in 2016, some metropolitan areas experienced record-high homeless populations. 203 Homelessness from 2016 rose each year thereafter through 2019.204 Although in in a majority of states, homelessness has declined homelessness in other states, notably California, have more than offset

increases

such improvements in the aggregate.2 0 s While homelessness overall has decreased since the last recession,206 the 2020 economic collapse due to the COVID-19 pandemic is likely to severely exacerbate challenges in this area with damaging impacts on health equity. 20 7

Persons who lack consistent housing access are frequently at risk for law enforcement contact for numerous reasons.208 Should these persons use cannabis in public, they risk running afoul of state and/or local law, which could precipitate law enforcement contacts with potentially damaging implications. For example, law enforcement may find other contraband or may inaccurately assess the person's mental health or capacity, leading to unnecessary incarceration or inappropriate civil commitment.209 Even small civil fines for violating public cannabis use restrictions could be highly detrimental for members ofthis population, as a lack of funds can compound legal debt and contribute to continued homelessness.2io Prohibitions on public cannabis consumption, as applied to persons experiencing homelessness, thus represent another facet ofthe criminalization of homelessness. The criminalization of homelessness includes legal restrictions on a variety of behaviors that would not be criminal if performed in private (e.g., bathing, urinating). 211 Public cannabis consumption does not share the essential character of some

activities prohibited by antihomeless laws (e.g., sleeping). However, because persons experiencing homelessness "do not have the luxury of privacy[] and must carry out their private lives in public places," 21 2 this cannabis consumption restrictions may still adversely population.

target

B. Risks to Others

1. Secondhand Exposure

Secondhand smoke is a well-known risk associated with tobacco products. 213 There are few studies of secondhand cannabis smoke, and most have focused on detecting uptake of cannabinoids, rather than other health effects. 214 However, existing research on cannabis smoke has pointed to negative cardiovascular effects 215 and identified multiple carcinogenic compounds similar to those in tobacco smoke.216 Given the similarity of cannabis and tobacco smoke, comparable secondhand exposure harms are biologically plausible and logically probable, despite presently limited verification.

There is even less evidence to date regarding secondhand harms of vaporized, vaped, or dabbed cannabis, but risks are at least conceivable. Cannabinoids are detectable on surfaces exposed to secondhand cannabis vaporizer emissions, which has implications not only for secondhand exposure from inhaling emissions but also thirdhand exposure from contact with chemicals on surfaces.217 Other emission components may similarly transmit via inhalation or contact and may be harmful to health. For example, similar devices used for tobacco products produce emissions with high concentrations of ultrafine particles and other potentially harmful substances. 21 ' Additionally, a 2019-2020 outbreak of lung

injuries associated with vaping cannabis products resulted in over 2,800 hospitalizations and 68 deaths apparently linked to the additive Vitamin E acetate. 219

Secondhand cannabis exposure risks are relevant for at least three locations: workplaces, homes, and public sites. One ofthe primary drivers of smokefree laws and voluntary tobacco policies has been protecting employees exposed to tobacco smoke at work. 220 To the extent that cannabis consumption is permitted in workplaces, such as consumption lounges, worker exposure represents a similarly substantial public health concern. Magnifying such concerns, cannabis businesses may be more likely to hire employees from populations negatively impacted by the War on Drugs, as several state and local social equity programs provide incentives to do so. 22 1 Cannabis businesses may also more readily accept applicants with past criminal drug activity due to the nature of the business.222 These facets raise meaningful questions of fair allocation of public health burdens if employees exposed to health risks come disproportionately inequities. Employees who have faced employment challenges due to criminal history, systemic racism, or other barriers may also be more reticent to advocate or organize for their rights, including the right to breathe clean air at work. This potential reluctance and the possibility ofemployer abuse further increase the importance of protecting employees' rights through law rather than voluntary employer policies.

from populations

to other

subject

Cannabis use in the home presents secondhand exposure risks to all persons present. Although both indoor and outdoor smoke pose health risks, pollutant concentrations are typically higher indoors due to the relatively smaller volume of air and absence of circulation. 2 23 Residences may present additional risks due to the time spent living and sleeping in

Such risks may be particularly acute for partners, children,225 the space. or others who lack the capacity or power to object to use inside the home. In multi-unit housing, secondhand exposures can also negatively impact persons in neighboring units.226

Use in outdoor public sites potentially exposes passersby and persons in neighboring residences or businesses to cannabis emissions. Such exposures may be fleeting and limited to unpleasant odor, which is highly subjective.227 However, these exposures could also be frequent and more significant. For example, some persons may be at higher risk for adverse effects due to respiratory illness, pregnancy, or other conditions. The impact and potential legal consequences of these exposures will depend on the duration and concentration ofuse, as well as proximity to neighbors. Still, drifting smoke could conceivably be the foundation of a private nuisance claim, among other actions, 228 regardless of whether the source is tobacco or cannabis.

2.

Intoxicated Behavior

Cannabis use can result in intoxication based primarily on the presence of the cannabinoid A9-tetrahydrocannabinol ("THC"). Cannabis intoxication has subjective effects on sociability, sensitivity to stimuli, perception oftime, and appetite. However, cannabis intoxication can also cause more significant effects, such as panic attacks, paranoia, and hallucinations. 229 Like alcohol intoxication, acute cannabis intoxication impairs perception and motor skills.2 30 Moreover, using cannabis before

driving increases the risk of motor vehicle accidents. 2 3 1 Motor vehicle accidents are a significant cause of death in the U.S. and the leading cause of death among those ages 16 to 25.232 Thus, cannabis use and intoxication raise similar concerns to alcohol with respect to impaired driving.2 33

There is some evidence that cannabis intoxication impairs driving differently than alcohol. Alcohol and cannabis intoxication are both dose dependent. However, the onset, duration, and effects of cannabis are highly variable between individuals based on tolerance, THC absorption, and consumption method, whereas alcohol's effects are relatively more predictable.234 Intoxication effects may also differ between the two substances. Cannabis and alcohol both impair essential skills and performance (e.g., tracking and attentiveness). However, persons using alcohol tend to increase risk-taking behaviors, while persons using cannabis tend to overestimate impairment and employ compensatory strategies, such as decreasing speed. 235 Despite such differences, both substances impair the ability to drive, and the effects are worse when combining the substances.2 36 To the extent that the location of cannabis use affects the decision to drive, allowing public use could increase the prevalence of impaired driving, though evidence of such an effect has not yet been established.

In addition to risks from impaired driving, cannabis intoxication may also lead to social disturbances. Public intoxication laws generally seek to protect public safety and preserve the peace by preventing intoxicated persons from "bothering and/or threatening the safety of other people in

public places" and reducing the "annoyances and deleterious effects which may and do occur because of the presence of persons who are in an intoxicated condition." 237 The offense of public intoxication typically requires disfavored conduct, "such as boisterousness; an indecent condition or act; . .. vulgar, profane, loud, or unbecoming language; or ... a public disturbance[.]" In some cases, public intoxication requires the creation of an actual, rather than merely possible, danger to either the intoxicated person or another party. 23 8 However, compared to cannabis, alcohol is more likely to precipitate such conduct. Dysfunctional behaviors associated with acute alcohol intoxication, per its ICD-10 definition, include disinhibition, argumentativeness, aggression, and mood lability (mood swings), 239 all of which are easily associated with behaviors targeted by public In contrast, acute cannabis intoxication, per its ICD-10 definition, involves dysfunctional behavior or perceptual abnormalities that include anxiety, agitation, paranoid ideation, and impaired judgment, among others.240 While undesirable, such effects are less recognizably connected to the behaviors targeted by public intoxication laws.

intoxication

laws.

C. IndirectSocietalRisks

1. Social Norms

Norms are behavioral regulations imposed by society informally, rather than by law. Though there are manifest debates over what is or is not a norm and the pathways through which norms operate, norms generally function as internal and external pressures toward conformity. Norms affect behaviors ranging from simple and beneficial (e.g., queuing in alcohol consumption). 2 4 1 The interactions between legal interventions and social norms are convoluted.2 42 However, there is substantial evidence that smokefree air laws for tobacco have contributed to changing cultural attitudes and population-level smoking behaviors. This process has denormalized smoking, particularly public smoking, from a behavior once

to complex and deleterious

(e.g., excessive

line)

common and viewed as fashionable or glamorous to one now largely uncommon and even stigmatizing.241

Smoking denormalization through smokefree laws differs from some other types of potentially stigmatizing public health interventions. First, restrictions on public smoking are "demeaning but not degrading" and "temporary rather than enduring" forms of stigmatization. 2 44 Smokers are not rendered "less than" by their exclusion, and such exclusion lasts only for the duration of the disfavored behavior. Second, there is a critical distinction between social or legal disapproval of a behavior in general as compared to its practice in a public setting. For example, sexual activity between consenting adults is not only legally accepted but constitutionally protected,2 45 while its practice in public is neither.2 46 Third, smokefree laws target a behavior because of its effects on others, rather than a latter paternalistic justification may be valid, albeit controversial, in some circumstances, but the former is a broadly accepted facet ofpublic health regulation.2 48

the actor's health.2 47 The

in protecting

interest

Any of the four public use policy models considered in this Article would represent an expansion of locations for lawful cannabis use compared to the baseline oftotal prohibition. The interaction between law, norms, and stigma may thus not precisely follow the path of tobacco, which has seen a steady contraction in permissible use locations over the past few decades. Because cannabis use has previously been a criminal act and not merely a disfavored one, social norms for use under legalization will also change in some respects regardless ofprevailing public use laws. In addition to changing norms for cannabis use, cannabis public use may also impact norms for other substances, including tobacco. Cannabis is not tobacco. However, like tobacco, smoking and vaping remain the most common consumption methods of cannabis. 249 Social acceptance of

cannabis smoking and vaping in public places may therefore lead to the reacceptance of smoking and vaping of tobacco products. To the extent that public cannabis use becomes legally permissible and socially tolerated in locations where tobacco smoking has been eliminated, this may undermine legal protections for clean air250 and renormalize smoking behaviors,251 potentially undoing decades ofpublic health advances.

2. Outlet Density

Alcohol outlet density

(the concentration of locations where consumers may purchase alcohol in a given area) is associated with excessive consumption and various related harms.252 Tobacco outlet density is associated with higher youth smoking rates253 and affects adult smoking behaviors.254 Similarly, a high concentration of cannabis imposing an businesses, particularly consumption sites, may risk inequitable burden on the health of surrounding populations. This risk is especially unpalatable given the history of inequitable impacts ofthe War on Drugs on communities of color and communities with lower socioeconomic status.255

Medical cannabis dispensaries and recreational cannabis retailers already demonstrate this troubling pattern. 25 6 This is predominantly the result of restrictive licensing requirements for cannabis businesses, which typically require a minimum distance from schools and other sensitive locations. Consequently, such restrictions often exclude wide swaths of urban and suburban areas. Additionally, many localities oppose cannabis businesses, but some localities lead more successful opposition efforts than others due to relative social power and organization. This imbalance may further contribute to the concentration of outlets amongst underresourced populations.257

The operative legal environment will strongly influence how the multifaceted risks outlined in the previous Section manifest. This Section explores how each of the four policy models for public cannabis consumption-private property, alcohol, tobacco, and unrestricted-are likely to shape the intersecting risks for consumers, others, and society. Each model is considered in turn, followed by a global "risk matrix" that collects and summarizes the conclusions for ease of comparison.

A. Private Property Model

The private property model presents mixed risks for cannabis consumers. This model is unlikely to encourage greater or more frequent use because it severely restricts lawful locations. Cannabis use inside a private home is already unlikely to be detected under prohibition. Because this model does not expand lawful use locations outside the home, it largely preserves the status quo in legal restrictions and social norms for

public use and for smoking generally. 258 However, a private property model also maintains existing risks of law enforcement encounters, thus perpetuating or exacerbating associated inequities. Persons who are homeowners will be able to avail themselves of adult use cannabis laws. Due to existing disparities in home ownership, such persons are more likely to be white and ofhigher socioeconomic status, attaching significant social justice concerns to this approach. Renters and persons experiencing housing insecurity will face the risk of law enforcement encounters if they consume cannabis in public, and they will face risks to their housing security if they use cannabis in a rented unit, shelter, or similar space where the owner prohibits cannabis use. For many, legalization under this model is thus either an illusion or a trap.

The private property model strongly protects most members of the public by keeping lawful use away from public spaces. Similarly, because the model creates no new businesses in which cannabis consumption is permitted, it produces no added risks for workers. Similar to tobacco, however, cannabis also presents concerns regarding how to control exposure risks within homes and drifting smoke between residences in multi-unit housing. Restricting cannabis use to private residences also does not mitigate (and may actually intensify) exposure risks for persons who lack the power to enforce nonsmoking norms within the home, including partners and children. If legalization increases the frequency or prevalence of cannabis use, the impacts of increased exposure would be shifted to homes, likely falling disproportionately on those with less power or who face other barriers. Illustrating this in the context of tobacco, secondhand exposure has decreased overall, 25 9 but indoor exposure to

smoke continues secondhand communities and persons living in poverty. 260

to disproportionately

affect Black

laws,

Nationally, under current

to secondhand cannabis smoke

reported recent outdoor-only is higher among persons exposure identifying as Black, Hispanic, or another race compared to those identifying as white.2 61 Given similar rates of cannabis use across racial and ethnic groups,2 62 one explanation for this difference in reported exposure is that because communities of color have lower average rates of home ownership, 263 members of these communities consume cannabis outdoors to avoid detection by neighbors or landlords. Unfortunately, this also places members of these communities at higher risk for detection by law enforcement. 264 Simultaneously, however, Black and Hispanic respondents report higher rates of indoor secondhand cannabis smoke exposure than white respondents, so it may instead be that Black and Hispanic cannabis smokers consume indoors to avoid law enforcement detection. The reasons underlying these exposure disparities (and their apparent contradiction) are unclear and may also reflect factors such as study limitations or demographic confounders. Regardless, restricting legal cannabis use to private residences is likely to inequitably affect communities of color along the same pathways that existing law does precisely because the practical changes of the approach would be limited. While the private property model appears to pose multiple clear risks, this approach may decrease risks associated with cannabis intoxication. Particularly, persons using cannabis at home may be less likely to drive following consumption. Alcohol provides evidence for this potential outcome in that driving after binge drinking is more common for alcohol purchased for on-premises consumption, especially at bars.26 Similarly, persons consuming cannabis at home may be less likely to be intoxicated in public simply by virtue ofnot consuming in public in the first instance.

Ultimately, the private property model is likely to have only limited impact on indirect societal risks. As noted above, cannabis use in private homes goes largely undetected even under prohibition. Therefore, legalizing this act is unlikely to meaningfully alter behavior beyond the impact of legalization itself. The absence of lawful public use locations limits the visibility of use and thus the potential for significant changes to social norms. Additionally, the private property model curbs some risks of outlet density by restricting outlets retailers without on-site consumption,266 analogous to permitting liquor stores but not bars.

B. AlcoholModel -Licensed IndoorLocations

locations

The Alcohol Model

restaurants, and other on-site

imposes only limited risks for cannabis consumers. Although this approach may encourage increases in intensity of use, these effects are unknown and difficult to predict. For alcohol, generational differences in preferred drinking location have shifted away from bars, toward home consumption.267 However, closures of many alcohol consumption locations during the COVID-19 pandemic led to substantial increases in alcohol sales from other outlets.268 Thus, the absence of on-site consumption locations does not necessarily reduce consumption. For the novelty of legalization encourages experimentation cannabis, regardless of consumption location, making it challenging to separate trends. Though an imperfect analogue, the Netherlands' coffee shop model provides some predictive guidance. The number of cannabis users increased following adoption of the model, but then stabilized and subsequently decreased when the number of shops was reduced.269 The availability ofthese locations does not appear to have translated to heavier use or longer lifetime use compared to other European countries, and other factors related to price and commercialization also play an important role in shaping behavior.270

To some extent, the alcohol model improves on equity-related risks to cannabis consumers compared to the private property model because on-site consumption locations would-theoretically-be open to all. However, the added expense of patronizing such locations in terms of transportation, admission or cover charges, and price mark-ups 271indicates that this approach does not actually level the playing field. Lounges attached to cannabis retailers may require a minimum purchase to access the lounges. For example, San Francisco's Barbary Coast reportedly requires a $40 purchase to use the attached lounge. 272 Freestanding consumption lounges that do not profit from cannabis product sales would need to charge an entrance fee or sell other products to generate revenue. In practice, such costs will exclude persons of lower socioeconomic status, leaving them to consume at home or illegally in public. This exclusion subjects these persons to the same risks as under the private property model, including threats to housing security and potential risks from law enforcement encounters.

The alcohol model presents significant risks to workers. Ventilation provides insufficient protection, particularly for those exposed to secondhand emissions for extended periods in businesses where smoking is permitted. This risk is evident with respect to tobacco smoke, as reflected in the ASHRAE standards, WHO FCTC, and ANR Model Smokefree Ordinance. 273 The potential that cannabis employers may hire disproportionately from communities impacted by War on Drugs further enhances the need for adequate worker protections. Some may argue that cannabis employees expect to work in proximity to cannabis use and enter such arrangements voluntarily. This may be true for some; however, structural factors strongly influence employees' choice of where to work and under what conditions. 274 Particularly for persons with reduced job prospects due to a prior drug conviction or other factors, the job "chosen"

is likely to be whichever one makes an offer, rather than a reflection of true choice.

While outdoor consumption

locations would reduce worker exposure, two problems limit the utility of such arrangements. First, concentrations of secondhand smoke particulates can be high even in outdoor smoking areas, 275 and drifting smoke can reduce air quality in nearby indoor areas,276 meaning workers would still face potentially unhealthy exposure risks. Second, the business viability of outdoor consumptions locations is influenced by climate, available space, and urbanicity, among other factors. If outdoor spaces are not desirable, they will tend not to be profitable, and thus the private market may not generate them in sufficient numbers to accommodate consumer demand. This lack of supply will produce regression to the default of cannabis use either in other (prohibited) public locations or at home, presenting the same risk profile as the private property model.

The alcohol model likely limits exposure risks for other members of the public. Some drifting smoke from consumption locations is possible but likely remediable by requiring locations to be sufficiently distant from neighboring businesses and residences. Filtering escaping air, as required by Michigan's regulations for on-site consumption licensee, 277 may provide additional protection for the public, though it is insufficient for those inside per ASHRAE standards.278 However, consumption locations may present additional risks if they attractnon-cannabis-using patrons. For example, consumption sites that do not sell cannabis (and possibly those that do) will likely require additional revenue streams, such as live entertainment or food and beverage sales. 279 If such offerings entice visitors who would not otherwise use or be exposed to cannabis (generally or on a particular occasion), increase exposure risks and this will contribute to normalizing cannabis use. To illustrate, consider a person who does not use cannabis but accompanies friends to a consumption lounge because it has appealing food or a show by a popular act. In

(particularly given

that alcohol consumption

addition to secondhand exposure within the venue, such a person may face to consume cannabis. Many non-drinkers similarly social pressure frequent bars and other alcohol-serving venues and face social pressure to is highly consume normalized). However, these persons are not directly exposed to alcohol by their mere presence, as they would be in a cannabis consumption venue. Intoxication risks under the alcohol model are likely to be significant compared to other models. For alcohol consumption, driving after binge is more common for alcohol purchased for on-premises drinking consumption, especially at bars.280 To the extent that the same is true for cannabis, which is currently unknown, the alcohol model may increase rates of driving under the influence. Although the risks appear to be somewhat different for cannabis than for alcohol, they are nevertheless a substantial public health concern. Similarly, public disturbances resulting from cannabis intoxication may be similar to those resulting from alcohol under this model. However, the specific consequences may be less severe given differences in intoxication effects on behavior.

Cannabis use under the alcohol model would be more visible than under the private property model, and the existence of lawful businesses allowing cannabis consumption would normalize cannabis use to some extent. However, appropriate state regulations or local ordinances could limit the external visibility of use and restrict the number and density of such locations. Ifjurisdictions adopt such checks, consumption under this limiting approach would normalization. Additionally, this approach would not reintroduce smoking to other public spaces and would thus be less likely to renormalize smoking generally.

largely occur outside of public view,

C. TobaccoModel -Limited Outdoor Locations

such

Under the tobacco model, cannabis consumption would be permitted in public locations where public health best practices allow tobacco use. Generally, locations sufficiently distant from business and residential entrances and windows and would exclude areas such as public recreation sites and transit centers. 281The potential effects of allowing cannabis use in these locations are largely unknown because the only examples to date come from the

include only outdoor

locations would

very brief history of legalization in Uruguay and a few Canadian provinces.

respectively); 282 however,

Among the many unknown effects of expanding public use are its impacts on use prevalence and frequency. It is too soon to evaluate the potential effects of the tobacco model in Canadian provinces. However, cannabis prevalence in Canada as a whole is generally lower than in legalizing U.S. states. According to a 2018 survey, legalizing U.S. states, which all prohibited public use, had higher prevalence of daily (11.3%), weekly (18.2%), and monthly (25.0%) cannabis use compared to Canada (8.9%, 14.1%, and 19.0%, there are considerable challenges to direct comparisons between Canadian and U.S. jurisdictions. 283 Although daily or near-daily use rates were stable, early data shows a small but noticeable increase in overall cannabis use in legalization. 284 Further Canada complicating trend analysis, cannabis use had been increasing in Canada prior to legalization. 285 It is plausible that future effects in the U.S. may be more substantial, depending on whether existing differences reflect the impact of legalization, baseline dissimilarity, or other confounding factors. The availability of various alternative product types may also be relevant. Canada's initial legalization framework permitted only dried flower, with edibles, extracts, and other products allowed only as of late 2019.286 In contrast, such products have typically been available immediately in legalizing U.S. jurisdictions, which may affect preferences in mode and location ofuse.

from 14.9%

to 16.8%

following

The effects of the tobacco model on consumption habits are uncertain, but this approach addresses many social justice concerns raised by the private property and alcohol models. Although access to outdoor spaces is not equal, 287 the amount of space required for cannabis consumption is not nearly as great as for other important uses, such as exercise. Despite inequalities, access to outdoor space is more equitable than private home ownership or access to stable housing. 2 The tobacco

model is at least facially equitable and introduces no inherent inequality in its application, as compared to the likely interactions between preceding models and existing housing inequalities. However, enforcement couldand unfortunately likely would-remain inequitable even under the tobacco model in the absence of more systematic reforms. This shortcoming is common to all approaches, as no public use framework is sufficient to remedy pervasive policing inequities. One potential option to mitigate enforcement inequities is to avoid penalizing smokers themselves and instead focus on requiring property owners to adopt appropriate smoking policies and take reasonable steps to enforce these policies. Though not a panacea, this can help shift the focus of enforcement away from individuals' behavior toward responsible property management.

The potential impacts ofthe tobacco model on secondhand exposures are complex. Unlike the alcohol model, the tobacco model creates no new cannabis businesses at which consumption occurs and thus does not create additive risks of secondhand employee exposure. The availability of lawful outdoor consumption locations also provides a generally costless and accessible alternative to home consumption, potentially reducing secondhand exposure risks for others in the home. While many will undoubtedly continue to consume at home, this approach removes fear of enforcement as a barrier to behavioral change.

However, the tobacco model may increase secondhand exposure risks for members ofthe public generally. Passersby would be exposed to cannabis smoke or emissions from persons consuming on sidewalks and in other locations where one commonly encounters tobacco smoke, and smoke may also drift further distances. However, such exposures would be more transient and occasional than exposures in the home or workplace and would diffuse risks more broadly across the population. Nevertheless, there are non-negligible public health risks to this model, many of which go beyond direct public exposure. In addition to their direct effects, prohibitions on tobacco smoking in public spaces encourage the adoption of voluntary smokefree policies in the home that are linked to reduced tobacco initiation, increased cessation, and denormalization of use. 28 9

Unless carefully tailored to replicate best practices, policies permitting cannabis use in public locations may reduce the potential to realize similar positive externalities of smokefree laws with respect to cannabis.

Applying the tobacco model to cannabis could also replicate some undesirable and inequitable patterns of tobacco policy. As noted above, secondhand tobacco smoke exposure has decreased generally. 290 Yet, indoor exposures remain a serious problem disproportionately affecting Black communities and persons living in poverty. 291 A hypothetically similar pattern for secondhand cannabis exposure would present a troubling social justice problem, illustrating the complex intersections of these issues and the careful balancing of risks and benefits required.

With respect to intoxication, early data following Canada's 2018 legalization shows no increase in the prevalence of driving within two hours of cannabis consumption and a decline in riding in a vehicle with a driver who has consumed.2 92 Being that several, but not all, Canadian provinces have adopted the tobacco model for public cannabis use, this general trend does not represent a distinction by public use model. However, three of the four most populous provinces (Ontario, British Columbia, and Alberta) have adopted the tobacco model. 293 As such, lack of an aggregate increase suggests that public use has not radically increased motor vehicle accidents. Although some legalizing U.S. states have observed increases in motor vehicle accidents, findings have been mixed. 294 Additionally, it is difficult to separate effects that may be related to visitors from non-legalizing (which would decrease as legalization proliferates). Other than New York (as of 2021), all U.S. jurisdictions also prohibit open public use, preventing analysis to date of any potential impact of the tobacco model as an alternative.

states

The tobacco model may contribute to public intoxication, even if its effects on driving behavior are unclear. The tobacco model allows consumption to move out of private residences but does not create defined consumption locations that might impose their own rules and restrictions on behavior. Thus, for unmoderated cannabis is potential consumption in public locations. Mitigating this, however, are the differences between cannabis intoxication and alcohol intoxication. As

there

the effects of cannabis

discussed above, 2 95 intoxication may be undesirable, but they are not as clearly related to the conduct public intoxication laws are intended to prevent. At least one state supreme court has recognized and given weight to this distinction. 2 96 Additionally, there is no inherent conflict in permitting public cannabis consumption while maintaining prohibitions on public intoxication, including for cannabis. Even jurisdictions that punish public alcohol intoxication typically allow alcohol consumption in many public locations, such as sports stadiums, entertainment venues, and restaurants. Punishment for public intoxication is tied to the acts of intoxicated persons rather than to consumption itself.

Applying the tobacco model to cannabis consumption preserves existing clean indoor air protections because it does not create any new locations where smoking ofany type is permissible. Similarly, the tobacco model does not contribute to cannabis outlet density because it creates no outlets beyond those selling for off-site consumption, which would exist under any of the four models. However, consumption under the tobacco model would be more visible than consumption under the private property or alcohol models because it would occur outdoors. This may normalize cannabis use and smoking in general. An unknown number ofpersons who currently use cannabis at home or covertly may alter their behavior under the tobacco model, allowing for their consumption to be seen by others. Still, the tobacco model does not allow consumption in all public locations, only in less favorable sites where tobacco use is permitted. This has been a leading factor in denormalizing tobacco use and is likely to mitigate normalization of cannabis use and renormalization of smoking generally under this model.

Additional research in Canadian provinces adopting the tobacco model for public cannabis use, such as Ontario, will provide crucial information on the likely trajectory of cannabis and tobacco use under the tobacco model approach. Cigarette and cannabis co-use among cannabis users declined in Ontario from nearly 60% in 1996 to approximately 42% in 2017, just prior to cannabis legalization, 297 but post-legalization trends are not yet known. New York will similarly provide key data in the U.S.

context under its 2021 law, but this will take time and will be muddled by the state's additional allowance of on-site consumption.

Generally, however, best practice restrictions on tobacco smoking in public places have contributed to dramatically cutting smoking prevalence in the U.S. and elsewhere by making nonsmoking a social norm. 298 These restrictions have succeeded without prohibiting limited (but still available and accessible) locations for tobacco consumption. Even California, which boasts among the lowest state smoking rates,2 99 permits smoking in most non-enclosed outdoor locations and some exempted indoor locations. 300 Localities frequently have more robust restrictions. For example, San Francisco has a highly restrictive ordinance prohibiting smoking at public events on city property,301 in city parks and recreation areas,312 and numerous other locations.303 Yet even San Francisco allows smoking at the curb of the nearest street, sidewalk, or alley (or, if there is no curb, fifteen feet from an entrance, exit, window, or vent). 304 This illustrates the multiplicity of options that state and local authorities could exercise under the tobacco model to permit cannabis use in limited public locations without acquiescing to widespread public use.

D. UnrestrictedModel

Unrestricted public use would impose few or no restrictions on the location of cannabis use, relying purely on social norms to regulate behavior. This position represents an extreme that neglects both the considerable police power authority of the government to regulate personal behaviors harmful to others and the bleak history ofwidespread, unrestricted tobacco consumption. Few, if any, organized legalization advocates openly seek a complete absence of restrictions on cannabis consumption. However, some advocates appear to take a position that personal courtesy is sufficient to regulate cannabis smoking behavior. For example, NORML's "Principles of Responsible Cannabis Use" advise the putative "responsible cannabis user" to "carefully consider his/her set and setting, regulating use accordingly," to "respect [the] rights of others," to "observe[] accepted standards of courtesy and public propriety," and to

"respect[] the preferences ofthose who wish to avoid cannabis entirely."305 Such reliance on courtesy and responsibility implies the absence of legal restrictions. However, in subsequently setting out the "Core Attributes of Adult Access Regulations[,]" NORML endorsed an alcohol model of "designated social use spaces (e.g., licensed social clubs)" with continued prohibition on use in "public, non-designated spaces (e.g., parks, city streets)" to be enforced by civil fines.306

Given the absence of advocacy, the unrestricted model may initially appear to be a Straw Man. Yet, it is worth recalling how few years have passed since tobacco smoke was ubiquitous. Despite clear evidence of tobacco's direct and indirect harms as early as the 1960s, establishing even non-smoking sections in restaurants remained a high water mark of antismoking advocacy in the 1970s.307 Through most of the 1980s, smoking was permitted on airplanes (and remained so on international flights until 2000).308 Even in the modern era, many states' clean indoor air laws remain weaker than local equivalents and, in some cases, preempt stronger local protections for workplaces, restaurants, and bars.309 The potential for retrenchment on this issue is neither remote nor hypothetical. Though public sentiment has come to strongly favor restrictions on public smoking, 310 new fronts in that conflict continue to arise, such as extending restrictions to e-cigarettes and similar products. 311 As such, this model requires similar analysis.

The unrestricted model would increase cannabis use prevalence and frequency. The decrease in U.S. tobacco prevalence generally tracks the decrease in public locations where smoking is permitted. Over 40% of the U.S. adult population were cigarette smokers in 1965, compared to less than 15% in 2018.312 The average number of cigarettes smoked per day also decreased among those who smoke, with the proportion of smokers who consume more than 24 cigarettes (a little more than a pack) per day decreasing from 25% in 1975 to just 6% in 2018.313 Many studies have also found that workplace smoking restrictions are associated with decreased tobacco use prevalence. 314 An unrestricted model probably would not drive cannabis use prevalence or frequency to the heights of mid-twentieth century-U.S. tobacco use, but the history of tobacco control cautions the relationship between available locations and use is robust. Increased prevalence and consumption frequency bring increased health risks both known and yet to be discovered.

that

The unrestricted model is, however, at least superficially equitable insofar as the absence of legal restrictions on cannabis public use presents fewer opportunities for inequitable enforcement. Still, at least one remaining concern under the unrestricted model would be the enforcement of age restrictions. Overall, the unrestricted approach might reduce potential law enforcement encounters and threats to housing security, but other approaches to achieve those goals would likely be both more effective and less damaging to public health.

The apparent fairness of the unrestricted approach is revealed as superficial when the risks that it presents to others are taken into account. public will Persons disproportionately be those in service industries and members of the general public with comparatively less structural and social power and influence who are therefore less able to efficaciously enforce social norms

secondhand

subjected

exposure

restrictions

to enforce smoking

to prevent smoking in their vicinity. In addition, the potential ubiquity of public use may reduce the authority of persons living with cannabis consumers the home. Again, prohibitions on smoking in public locations can also encourage the adoption of voluntary smoke-free policies in the home that are linked to reduced use denormalization.3 15 The lack of such restrictions on cannabis may produce the inverse, thereby increasing exposures for persons in the home, with impacts likely to be inequitably distributed given patterns observed in tobacco control.

cessation,

initiation,

increased

tobacco

The potential impacts of an unrestricted model on intoxication and its related harms are unclear. As noted above, concerns on this point are somewhat distinct for cannabis as compared to alcohol. However, if the prevalence and frequency of cannabis use do increase significantly, a rise in the frequency of intoxicated driving and disturbances related to other intoxicated behavior would likely follow, even without concentration of use in particular locations.

Many of the aforementioned risks flow from the indirect social consequences of unrestricted public cannabis use. If the prior history of tobacco use is a reasonable guide, an absence of legal restrictions would precipitate highly permissive social norms and normalization of cannabis use. Similarities between cannabis smoking and tobacco smoking would subsequently resurrect widespread public smoking generally, as it would prove challenging to draw a defensible social distinction between cannabis and tobacco smoke. However, most young adults correctly attribute considerable risks to tobacco use, even if they view the risks of cannabis use as more dubious.316 This could limit the impact of cannabis normalization on tobacco use, though substance co-use poses additional challenges.

E. Risk-Benefit Matrix

Combining the risks and advantages of each model yields the matrix of risks below in Figure 1. The risks of the private property model fall most heavily on cannabis consumers themselves, with comparatively fewer direct or indirect risks to others as a result of the strict constraints on permissible use locations. The alcohol model produces risks both for

the consumer and others, combining the consumer risks of the private property model with additional risks from the creation of indoor businesses that put workers at risk and normalize consumption. The tobacco model produces fewer risks to the individual consumer, though it does create some public risks due to the public nature ofthe consumption. The unrestricted model fares poorly on most accounts, producing considerable risks of nearly all types other than those relating to enforcement and outlet density.

Each regulatory model presents benefits and detriments from a public policy perspective, leaving the question of how to properly weight them in determining the best approach among imperfect options. Three major factors lead to the conclusion that the tobacco model is the best choice based on current evidence. First, the history of cannabis prohibition and the War on Drugs necessitates an approach that prioritizes social justice and equity interests. The tobacco model best balances risks and benefits in this context. Second, an approach that merges major aspects of tobacco and cannabis control provides advantages in administrative simplicity, enforceability, and credibility that will enhance effectiveness. Third, the history of tobacco control demonstrates that this model is an effective public health approach when incorporating best practices. Additionally,

the denormalizing effect of restricting use to limited outdoor locations is a powerful public health tool, albeit one that should be wielded with care.

A. PrioritizingEquity

Relevant considerations of public health ethics include producing benefits (beneficence), avoiding and removing harms (nonmaleficence), maximizing the balance of benefits over harms (utility), distributing benefits and burdens fairly (distributive justice), and respecting personal liberty (autonomy). 317 Each public use model presents distinct advantages and disadvantages related to these principles. As is typically the case, the principles must be balanced against one another and weighted appropriately for the given context.318 Among the factors that help resolve conflicts between these considerations are principles of effectiveness and proportionality. Effectiveness that infringement of moral requires considerations be likely to protect public health, and proportionality requires that probable public health benefits outweigh the infringement. 319 To the extent that any of the models offer "benefits" from a public health perspective, such benefits are only relative to other options or as compared to prohibition. Cannabis use carries at least some risks to health, so any increase in use or expansion of use locations is not beneficial per se, but rather reflects a choice to accept some set of harms in order to mitigate or avoid others. The risks of each approach vary in type, acuity, and equity of distribution. Because the unique history of cannabis prohibition in the U.S. has produced profound, pervasive, and persistent inequities, a legalization framework grounded in public health must prioritize equity in evaluating each model's balance of risks and benefits. The private property model produces inequities across all three types of risk. Existing inequities in housing and law enforcement cause this model to produce disproportionate effects on some cannabis users in exposure to law enforcement, threats to their housing security, and harms to those who live with them. Although this approach minimizes several important risks, such as public exposure and use normalization, its exacerbation of existing enforcement inequities renders it incompatible with social justice. Restricting cannabis use to private property results in a bifurcated "legal for me, but not for thee" system that hollows out the promise that legalization would advance equity.

The alcohol model similarly creates inequalities in the distribution of the benefits and risks of legalization. Those with greater resources (i.e., persons who can afford entrance fees or on-site price premiums) can avail

themselves of the benefits of the law. Those with fewer resources lack access to such benefits and face greater likelihood of law enforcement encounters, as they do under the private property model. Such concerns are not grounded in any putative "right to smoke" 32 0-though such arguments might plausibly be made for medical cannabis use in some contexts-but rather in social justice and fundamental fairness. As illustrated in legalizing jurisdictions to date, following legalization the enforcement of remaining restrictions, such as public consumption prohibitions, is likely to continue apace or increase, allowing enforcement inequities to persist.32 1 Any benefits of the alcohol model in terms of business or job creation and limited general public exposure are also outweighed by the risks to workers from secondhand exposure and the likely under-resourced neighborhoods. Both of these risks are likely to disproportionately affect vulnerable communities already harmed by the War on Drugs. risks,

tobacco model unquestionably presents to

including the general public and potential secondhand exposure normalization of cannabis use and smoking generally. However, the tobacco model distributes risks more equitably than other approaches. The burdens of this model do not fall inherently on particular populations, though continual monitoring for such impacts would be essential, given that the burdens of tobacco have been inequitable due to a variety of factors, including marketing practices.322

of cannabis

concentration

outlets

risks

The

The unrestricted model theoretically promises equity, as it imposes no restrictions to be inequitably applied. However, while equity should be prioritized, it does not follow that other considerations should be ignored. The normalization and secondhand exposure risks of this approach are

manifestly unacceptable from a public health perspective, given the appalling toll of widespread tobacco use and the likelihood of repetition for cannabis under this model. Moreover, negative effects of the model will still be inequitable despite their apparent evenhandedness, given that the public health damage of tobacco has been and continues to be a major source of health disparities.3 23

B. Simplicity, Enforceability,and Credibility

Merging regulatory frameworks for public use of cannabis and tobacco will improve the administrative simplicity and enforceability of related laws, enhancing the credibility of public health interventions for cannabis. Despite their differences, tobacco and cannabis are similar in many ways, including the primacy of smoking as a consumption method and, most likely, the associated risks of consumption. A unified approach to regulating public smoking of both substances is a logical next step that applies the hard-won lessons oftobacco control to a product with a similar risk profile. The introduction of newer product types further aligns policy concerns for the two substances. For example, electronic devices for vaporizing THC, CBD, or nicotine products are highly similar, and some devices can be used interchangeably amongst these different products. 32 4 Some of these products have already led to significant public health harms, such as the 2019-2020 outbreak of EVALI (e-cigarette or vaping remain under associated investigation, comprehensive regulation is crucial to minimizing future harms, and a unified approach to public use ofsuch products supports such efforts.

injury). 325 While

root causes

lung

the

Unifying cannabis and tobacco control also presents the opportunity to advance tobacco control. Adopting the tobacco model for public cannabis use does not mean accepting whatever tobacco policy may be in a particular jurisdiction, but rather adopting tobacco control best practices. In jurisdictions that have not adopted comprehensive smokefree laws for tobacco, cannabis represents an occasion for health advocates to press a uniform approach applying best practices to all smoking and vaping products, potentially overcoming objections to stricter tobacco control policies standing alone.326 For example, a provision of Connecticut's 2021 legalization bill setting local authority to regulate public cannabis use

this authority

incorporated tobacco consumption.327 Language added by the bill also expanded the scope of local authority over tobacco smoking and vaping and imposed additional restrictions on permissible consumption locations.328

into existing

control over

Prohibition has been the dominant U.S. cannabis policy model for the better part of a century, and persons who use cannabis are therefore accustomed to furtive use, particularly in public.329 Cannabis is also often mixed with tobacco (e.g., spliffs and blunts), and vaped cannabis products can easily be disguised as tobacco products.330 Distinguishing between cannabis and tobacco use may therefore be exceedingly difficult. In contrast, smoking as a general behavior is more difficult to conceal, though this is somewhat less true for vaping. 33i Divergent rules for cannabis and tobacco use require investigation to distinguish between the substances. While cannabis is generally considered to have a distinctive odor, this is not necessarily true of all forms, such as vaporized oils, and may not account for mixed use of cannabis and tobacco. Under the tobacco model, rules for both substances are the same, allowing enforcement to focus on objective and easily verified concerns such as whether smoking is permitted in a particular location3 3 2 or whether the user is of legal age, rather than subjective considerations such as odor or appearance.

Cigarette smoking has declined precipitously over the past few decades. 333 In no small measure, this decline is due to the highly effective

efforts of public health messaging regarding the dangers of smoking. 334 Many cannabis users correctly recognize the harms of smoking and secondhand tobacco smoke, but do not associate similar harms with cannabis smoke.335 Additionally, the use ofvaping products as a smoking alternative is increasingly popular for both tobacco and cannabis, particularly among younger users. Merging cannabis and tobacco policy can improve public understanding that "smoke is smoke" and facilitate informed consideration of the risks associated with both products.

Leveraging effective tobacco control strategies to craft cannabis regulatory frameworks is well-supported based on existing evidence and advocated by many in health policy. 336 However, for public health messaging to be effective, the public must find it credible. 337 Due to the history of drug law enforcement in the U.S. and other countries, many citizens already view cannabis restrictions as unjustified. Efforts to limited form of prohibition by banning public maintain a more consumption will likely be met with suspicion. To the extent consumption bans are easy to evade, they will further erode confidence in and compliance with not only consumption rules themselves but also restrictions. Cannabis regulatory potentially other related cannabis frameworks should be rigorous in order to protect public health, but they must also be reasonable, evidence-based, and practicable in order to foster credibility and regain public trust.

C. The PowerofDenormalization

If "smoke is smoke" and cannabis and tobacco policy are thusly linked, then the tobacco control approach should apply not only to where cannabis use is prohibited, but also to where it is permitted. To date, most public health advocates have endorsed the former but typically not the latter. Such hesitancy is understandable. Tobacco control advocates have expended decades of effort to achieve meaningful change in public

tobacco use and are justifiably concerned that condoning public cannabis smoking may reverse hard-fought gains by renormalizing smoking behavior.338 This is particularly worrisome with respect to cannabis because many people, especially younger generations, view cannabis as less harmful and more socially acceptable than tobacco.339 If populationlevel reductions in tobacco smoking are replaced with increased cannabis smoking, this would be a regrettable erosion of an important public health achievement.

However, the value of denormalization as a public health tool is not so easily diminished, and the tobacco model allows it to be brought to bear on cannabis. Tobacco smoking rates have declined dramatically in large part because of the concerted efforts of public health advocates to denormalize smoking behavior. One of the key aspects of this effort has been removing tobacco use from indoor public spaces. This has the immediate effect of creating cleaner air in those locations, but it also provides a secondary effect by signaling that the behavior is disfavored. This new norm represents a profound change from the prior norm of smoking anywhere at any time, subject only to the boundaries of courtesy-if even those. Similarly excluding cannabis consumers when the appropriate balance between they choose normalization and stigma by neither affirmatively creating locations that encourage use nor punishing use that does not affect others. The slope from legalization to normalization may be a slippery one, but there are a number of safe landings along the path.

to partake

strikes

D. Limitations, Clarifications,and TemporaryMeasures

Nothing in this argument should be construed as either endorsing or condemning cannabis use per se. The argument for applying a tobacco control approach to public cannabis use presupposes that some portion of the population will choose to consume cannabis and that society has legally sanctioned the activity as a general matter through legalization. The argument turns then to the question of which potential consumption locations produce the most favorable balance of risks and benefits. The tobacco model will still produce public health harms. These harms are a cost to society and should not be ignored or dismissed without due consideration, but they are overcome by the societal costs of the alternatives, including the status quo of prohibition.

Applying the tobacco model to public cannabis use and permitting consumption in restricted outdoor locations presents the best balance of risks and benefits under a public health law approach based on current

evidence. However, a private property model may be an acceptable shortterm approach despite its flaws. First, it is easier to loosen an overly restrictive policy than to ratchet up restrictions on an overly liberal one. For example, the FDA initially hesitated to regulate e-cigarettes, resulting in proliferation of the devices, skyrocketing youth initiation, and the arrival of myriad flavor variants that had been outlawed in traditional cigarettes for a decade.340 When cities, states, and eventually the federal government attempted to impose new restrictions on e-cigarettes in response to public health concerns, these entities faced forceful resistance from an emboldened industry that had previously operated with little oversight. In the span of a few short years, a ban on the sale of "Unicorn Poop"-flavored nicotine products became to some an assault on personal freedom epitomizing government overreach.3 41 Restricting tobacco smoking locations required decades of concerted advocacy efforts and has still not produced uniformly strong laws across all states. 342 Ifthe tobacco model is not feasible initially, the private property model could provide a temporary stopgap that does not expand public use too broadly to undo. In contrast, temporary adoption ofthe alcohol model would create a new type of business entity with a vested interest in maintaining its position and viability.

Second, the current disjuncture between state and federal law alerts states to remain wary of increased federal enforcement. Cannabis remains unquestionably illegal under federal law, and only the limits of discretion and personnel constrain enforcement and allow state experimentation to proceed without substantial interference.14' A state adopting a permissive approach to open public use might well find that it has exceeded the limits of federal forbearance toward the laboratories of democracy.

From a public health law perspective, cannabis legalization presents numerous challenges. Among the most difficult is determining where consumption should be permitted. Risks to cannabis consumers, others exposed to cannabis use, and society as a whole are inevitable under any approach to public use. However, when considering the central importance of social equity, a tobacco control model that allows cannabis consumption in limited outdoor locations provides the best available balance of risks and benefits. Merging tobacco and cannabis policy also yields a more enforceable, credible, and effective approach that allows cannabis to be included in comprehensive smokefree air laws while recognizing that the central idea that "smoke is smoke" should inform the full range of cannabis policies. Tobacco control best practices have successfully reduced the population health burdens of tobacco use. Applying the tobacco control approach can leverage hard-won lessons to prevent similar harms from cannabis.

Cannabis policy neither begins nor ends with legalization, and it reflects the full and troubled history of drug policy generally, as well as the need for continued refinement. Further scientific and social evidence should inform policymaking in this area. However, the rapid and inexorable drive toward legalization across states over the past decade has made it infeasible to wait for all ofthe evidence before addressing the most difficult questions of law and policy. The shameful history of the War on Drugs compels cannabis legalization to address social justice concerns and avoid contributing to existing inequities or creating new ones. Prioritizing these issues leads to the conclusion that limited open public consumption is the best choice among imperfect options.

--- FOOTNOTES ---

INTRODUCTION.....................................................................................................61 I.

PUBLIC USE REGULATORY FRAMEWORKS....................................................67 A. History and Context of the Tobacco ControlModel...........................68 B. History and Context of the Alcohol Control Model ........................ 73 C. ExistingAdult Use CannabisLaws................................................ 74 1. U .S. Jurisdictions................................................................ . 74 2. Non-U.S. Jurisdictions .............................................................. 80 D. Typology for FuturePublicCannabis Use Regulation ...................... 81 II. POTENTIAL RISKS OF PUBLIC CANNABIS USE...............................................83

A. Risks to Consumers............................................................................. 84 1. Negative Health Effects of Cannabis Use ................................. 84 2. Potential for Law Enforcement Encounters .............................. 85 3. Cannabis, Housing, and Equity ................................................. 87 B. Risks to Others.................................................................................... 91 1. Secondhand Exposure ............................................................... 91 2.

Intoxicated Behavior.................................................................93

C. IndirectSocietal Risks ........................................................................ 95 1. Social N orms ............................................................................. 9 5 2. Outlet Density ........................................................................... 97 III. APPLICATION TO CANNABIS PUBLIC USE TYPOLOGY ................................... 98 A. PrivatePropertyModel ...................................................................... 98

1. Risks to Consumers...................................................................98 2. Risks to Others .......................................................................... 99 3.

Indirect Societal Risks.............................................................101 B. Alcohol Model -Licensed IndoorLocations.................................... 101 1. Risks to Consumers.................................................................101 2. Risks to Others ........................................................................ 102 Indirect Societal Risks.............................................................104 3. C. TobaccoModel -Limited OutdoorLocations..................................104

1. Risks to Consumers.................................................................104 2. Risks to Others ........................................................................ 106 Indirect Societal Risks.............................................................108 3. D. UnrestrictedModel...........................................................................109

1. Risks to Consumers.................................................................111 2. Risks to Others ........................................................................ 111 Indirect Societal Risks.............................................................112 3.

E. Risk-Benefit Matrix........................................................................... 112

IV. AN ARGUMENT FOR OPEN PUBLIC USE IN LIMITED OUTDOOR

LOCATIONS ............................................................................................ 113 A. PrioritizingEquity ............................................................................ 114 B. Simplicity, Enforceability, and Credibility....................................... 116 C. The Power ofDenormalization.........................................................118 D. Limitations, Clarifications,and TemporaryMeasures.................... 119

CONCLUSION......................................................................................................121

INTRODUCTION

1. See, e.g., MarijuanaSmoke: SecondhandMarijuanaSmoke is Not Healthy, AMs.

FOR NONSMOKERS' RTS., https://bit.ly/3f9wkhC (last visited Apr. 30, 2021); Protecting Nonsmokers from Secondhand Marijuana Smoke, AM. NONSMOKERS' RTS. FOUND. 1 (2019), https://bit.ly/3bMHoiE; Memorandum from State of California Tobacco Education and Research Oversight Committee Chairperson Michael Ong (Aug. 20, 2019) (https://bit.ly/3f9EQgE) ("It is important for tobacco control programs to make it clear to the public that smoke is smoke. Both tobacco and marijuana produce secondhand smoke and both are associated with adverse health outcomes."); see also What You Need to Know About All Secondhand Smoke, Vape andMarijuana,TOBACCO FREE CA (June 19, 2020), https://bit.ly/2SmdElN (discussing the similarity of risks from all sources of secondhand smoke, including cannabis).

2. See Marihuana Regulation and Taxation Act, 2021 N.Y. Laws ch. 92 (S. 854-A);

see also discussion infra Section I.C.1.

3. See discussion infra Section I.C.1. 4. San Francisco officially designated cannabis-related offenses by adults the lowest priority for law enforcement tenyears before California passed its adult use law. See CITY & COUNTY OF S.F., CAL., ORDINANCE No. 297-06 (2006).

5. While merely anecdotal,

the author personally witnessed this on numerous occasions and in various locations as a resident of San Francisco between 2017 and 2019, experiences which served as inspiration for this Article.

6. See CAL. HEALTH & SAFETY CODE § 11362.3(a)(1) (West 2017) ("Section 11362.1 does not permit any person to ... [s]moke or ingest cannabis or cannabis products in a public place .... ").

7. Andrew Daniller, Two-Thirds ofAmericans SupportMarijuanaLegalization,PEW RES. CTR. (Nov. 14, 2019), https://pewrsr.ch/3hQX6gz. Support for medical legalization is even higher at 91%. Id.

8. Daniel G. Orenstein & Stanton A. Glantz, The Grassrootsof Grass: Cannabis LegalizationBallot Initiative Campaign Contributionsand Outcomes, 2004-2016, 45 J. HEALTH POL., POL'Y & L. 73, 76-77 (2020) (listing successful measures in eight of eleven states with ballot initiatives between 2012 and 2016); State Medical Marijuana Laws, NAT'L CONF. OF STATE LEGISLATURES (May 17, 2021), https://bit.ly/3yzJ74y (noting 2018 Michigan ballot initiative). Additionally, Vermont and Illinois passed legalization bills via their state legislatures during this period. See id.

9. Arizona Proposition207, MarijuanaLegalizationInitiative (2020), BALLOTPEDIA, https://bit.ly/3ufquQ6 (last visited June 16, 2021) (60% of the vote); Montana I-190, MarijuanaLegalizationInitiative (2020), BALLOTPEDIA, https://bit.ly/3uhOcgh (last visited June 16, 2021) (57% of the vote); New Jersey Public Question 1, MarijuanaLegalization Amendment (2020), BALLOTPEDIA, https://bit.ly/3hIZV3i (lastvisited June 16, 2021) (67% of the vote); South DakotaConstitutionalAmendment A, MarijuanaLegalizationInitiative (2020), BALLOTPEDIA, https://bit.ly/3hTNhhM (last visited June 16, 2021) (54% of the vote). A South Dakota court subsequently found Amendment A unconstitutional, but a final ruling from the South Dakota Supreme Court remains pending as of this writing. See Stephen Groves, South Dakota Supreme Court Weighs Pot Legalization Battle, ASSOCIATED PRESS (Apr. 28, 2021), https://bit.ly/3cM9bAp; Abby Wargo, State Supreme Court'sAmendment a DecisionMay Take Months, Law School Dean Says, RAPID CITY J. (Aug. 7, 2021), https://bit.ly/3AR80cm.

10. In the presidential election, Joe Biden narrowly carried Arizona (49.4%), Montana went to Donald Trump (56.90o), and New Jersey to Biden (57.3%). See 2020 ElectoralCollege Map, PBS NEWS HOUR, https://to.pbs.org/2QKeZCI (last visited June 16, 2021). South Dakota was the only one of the four states with a higher margin of victory at the top of the ticket, voting 61.8% for Trump. See id.

11. See Olivia B. Waxman, The 2020 Election Set a Recordfor Voter Turnout. But Why Is It Normalfor So Many Americans to Sit Out Elections?, TIME (Nov. 5, 2020, 9:24 AM), https://bit.ly/348n9aZ.

12. In addition to Washington, D.C., the eighteen states that have legalized adult use cannabis as of June 2021 are Alaska, Arizona, California, Colorado, Illinois, Maine, Massachusetts, Michigan, Montana, Nevada, New Jersey, New Mexico, New York, Oregon, South Dakota, Vermont, Virginia, and Washington, with South Dakota's law under legal challenge as of this writing. See State MedicalMarijuanaLaws, supranote 8. An additional nineteen states have legalized medical cannabis: Alabama, Arkansas, Connecticut, Delaware, Florida, Hawaii, Louisiana, Maryland, Minnesota, Missouri, North Dakota, New Hampshire, Ohio, Oklahoma, Pennsylvania, Rhode Island, Utah, and West Virginia, with Mississippi's law not counted due to being invalidated by the Mississippi Supreme Court. Id. The adult use jurisdictions have a population of approximately 143.1 million, with medical jurisdictions bring the total to 238.5 million, compared to a total U.S. resident population of approximately 331.5 million. See Resident Populationfor the 50 States, the Districtof Columbia, and Puerto Rico: 2020 Census, U.S. CENSUS BUREAU: POPULATION Div. (Apr. 2020), https://bit.ly/2TEShwZ (providing totals counted here as including states in which statutory or regulatory frameworks are not yet in effect or under legal challenge, but excluding populations of non-state territories other than D.C. due to differences compared to other data sets).

13. See generally Sam Kamin, Legal Cannabisin the U.S.: Not Whether But How?, 50 U.C. DAVIS L. REV. 617, 620-24, 651-58 (2016) (discussing rapid state legalization and distinguishing between various general market models for potential future federal legalization). Implications of legalization under the terms of international drug control treaties are also beyond the scope of this Article, but the experiences of Uruguay and Canada will be instructive. See generally, e.g., Antonia Eliason & Robert Howse, A Higher Authority: Canada'sCannabisLegalizationin the Context ofInternationalLaw, 40 MICH. J. INT'L L. 327, 334-51 (2019) (discussing the relationship between Canada's 2018 legalization and relevant international treaties); Michael Tackeff, Note, Constructing a "Creative Reading": Will U.S. State Cannabis Legislation Threaten the Fate of the InternationalDrug Control Treaties?, 51 VAND. J. TRANSNAT'L L. 247, 265-72, 283-87, 293-95 (2018) (describing challenges posed by federal-state discord on legalization in light of international treaties and possible resolutions); Allison E. Don, Note, Lighten Up: Amending the Single Convention on Narcotic Drugs, 23 MINN. J. INT'L L. 213, 223-33 (2014) (noting backlash from the International Narcotics Control Boards to early state legalization in Washington and Colorado and to Uruguay's national legalization in 2013 and arguing that the U.S. is inviolation of the Single Convention). A partial reclassification by the United Nations' Commission on Narcotic Drugs in December 2020 additionally

indicates potential future changes on this front. See UN Commission Reclassifies Cannabis, Yet Still ConsideredHarmful, UN NEWS (Dec. 2, 2020), https://bit.ly/3wo2fkn.

14. This Article uses "cannabis" (rather than "marijuana") to incorporate reference to the full variety of products now derived from Cannabis sativa, and the term is now commonly used in state legalization laws. "Marijuana," in contrast, typically refers only to the dried flower of the plant that is typically smoked, is a term used almost exclusively in the U.S., and has a troublingly racially motivated history. See, e.g., Lauren Yoshiko, The Difference Between Weed and Cannabis, FORBES (Sept. 30, 2018, 7:18 PM), https://bit.ly/3uiPRkk; Cannabis(Marijuana)andCannabinoids:W~hat You Needto Know, U.S. DEP'T OF HEALTH & HUM. SERVS.: NAT'L INSTS. OF HEALTH, NAT'L CTR. FOR COMPLEMENTARY AND INTEGRATIVE HEALTH, https://bit.ly/3hZLdFp (last visited Dec. 12, 2019); Paul Armentano, Marijuana: A Primer, NORML, https://bit.ly/2SnosjW (last visited Dec. 12, 2019).

15. See Hongying Dai & Kimber P. Richter, A National Survey ofMarijuana Use Among US. Adults with Medical Conditions, 2016-2017, JAMA NETWORK OPEN, Sept. 2019, at 1, 7-8.

16. See, e.g., NAT'L ACADS. OF SCIS., ENG'G, & MED., THE HEALTH EFFECTS OF AND THE CURRENT [hereinafter NASEM Report],

FOR RESEARCH

CANNABINOIDS:

STATE OF

EVIDENCE

(2017)

CANNABIS AND RECOMMENDATIONS https://bit.ly/3oHLbmT.

17. See id. at 127-29 (summarizing report conclusions regarding cannabis and

cannabinoid effectiveness for various health conditions).

18. See, e.g., Michael Vitiello, Marijuana Legalization, RacialDisparity, and the HopeforReform, 23 LEWIS &CLARK L. REV. 789, 797-809 (2019) (tracing the role of both overt and "dog whistle" racism in the history of U.S. cannabis regulation).

19. See, e.g.,

JONATHAN P. CAULKINS ET AL., CONSIDERING MARIJUANA INSIGHTS FOR VERMONT AND OTHER JURISDICTIONS 112-13 (2015), LEGALIZATION: https://bit.ly/2ShpNbZ ("Marijuana is a very different commodity from other regulated goods (even alcohol) and early-adopting states simply cannot use cookie-cutter regulations for alcohol to cover all of the important choices.").

20. See infra Section I.B.

21. See infra Section I.A. 22. See, e.g., Kerry Cork, Recreational Marijuana, Tobacco, & the Shifting Prerogativesof Use, 45 S. ILL. U. L.J. 45, 46 (2020) ("For residents of any state with legalized recreational marijuana -users and nonusers alike -where and when the product can be used continues to be confusing and, at times, controversial."); Jane Steinberg et al., A Tobacco Control Framework for Regulating Public Consumption of Cannabis: MultistateAnalysis andPolicyImplications, 110 AM. J. PUB. HEALTH 203, 203 (2020) ("A key regulatory challenge for cannabis-legal states and municipalities is establishing where residents can legally smoke cannabis or consume aerosolized (i.e., vaporized or 'vaped') cannabis.").

23. See NASEM Report, supranote 16, at 1, 9-12. The drug's negative health effects are the most relevant to this analysis, but cannabis and its constituent cannabinoids also have at least some established therapeutic utility, and the medicinal use of the plant has been common for millennia. See, e.g., id. at 13-14 (reporting conclusions regarding therapeutic effects); GLEN R. HANSON ET AL., DRUGS AND SOCIETY 408 (12th ed. 2015) (noting that the earliest written reference to the plant refers to medicinal use around 2700 B.C.E. China).

24. Uruguay (2013) and Canada (2018) were the first countries to legalize adult use cannabis. See Simon Maybin, Uruguay:The World'sMarijuanaPioneer,BBC NEWS (Apr. 4, 2019), https://bbc.in/3pZyVOY. Uruguay's system is limited to low-potency cannabis available only through government-run retailers, see id, and Canada's law did not take effect until fall 2018, limiting currently available data. Colorado and Washington were the first U.S. states to legalize cannabis, in 2012, but these states, and others that have legalized to date, have all done so under the shadow of federal prohibition. See Kamin, supra note 13, at 623-30.

25. Beyond these highly visible examples, there are many more seemingly innocuous products, such as sugar-sweetened beverages or trans fats, that are demonstrably unhealthy and thus arguably may be proper subjects of regulation or prohibition. See generallyLaura Hoffman, Cigarettesvs. Soda?: The Argumentfor SimilarPublic Health Regulation of Smoking and Obesity, 46 CONN. L. REv. 1889, 1892-1900 (2014) (asserting the viability of some paternalistic regulation of products linked to obesity); James G. Hodge, Jr. & Megan Scanlon, The Legal Anatomy of ProductBans to Protectthe Public'sHealth, 23 ANNALS HEALTH L. 161, 163-79 (2014) (describing underlying principles, justifications,

and pathways for product bans and limitations on governmental authority to take such actions, as well as comparing trans fats, cannabis, and tobacco).

26. Le., most products are neither wholly "healthy" nor "unhealthy" abstracted from

consideration of consumption patterns and other aspects.

27. Expansion of any public smoking behavior is typically anathema to public health advocates, even with respect to outdoor locations. See, e.g., Outdoors, PUB. HEALTH L. CTR., https://bit.ly/3uayVwg (last visited Dec. 12, 2019). Some scholars have allowed that harmonizing approaches to tobacco and cannabis may be the best path forward. See, e.g., Steinberg et al., supra note 22, at 206 ("One option is to amend language regarding prohibition on outdoor cannabis use while retaining strong smoke-free indoor air rules."); Hannah Holitzki et al., Health Effects ofExposure to Second-and Third-HandMarijuana Smoke: A Systematic Review, 5 CMAJ OPEN E814, E820 (2017) ("Alignment of tobacco and marijuana smoking bylaws, with a coherent policy approach to exposure to smoke of any kind, may result in the most effective public policies."). However, there appear to be no comprehensive arguments for allowing public cannabis use in existing legal or public health academic literature.

I.

PUBLIC USE REGULATORY FRAMEWORKS

28. Socialjustice is itself a key public health objective. See generally, e.g., Lawrence 0. Gostin & Madison Powers, TWhat Does Social Justice Requirefor the Public'sHealth? PublicHealth Ethics andPolicy Imperatives, 25 HEALTH AFFAIS 1053, 1053-57 (2006) (describing the centrality of justice and in particular social justice to public health). However, public health policymaking requires the balancing of competing interests.

29. See 21 U.S.C. § 812. 30. See State Medical MarijuanaLaws, supra note 8. This total does not include

South Dakota's adult use law due to its disputed status as of this writing.

31. See 21 U.S.C. § 812. 32. Most notably, the 2013 "Cole Memo" outlined the Department of Justice's approach to enforcement of the Controlled Substances Act and general reliance on state and local law enforcement to address cannabis activity. The document indicated the Department's intention to focus on eight enforcement priorities, signaling to state authorities that the federal government would not interfere with well-regulated state cannabis programs. See Memorandum from Deputy Att'y Gen. James M. Cole to United States Att'ys (Aug. 29, 2013) (https://bit.ly/2SkKfZx). Attorney General Sessions formally rescinded the Cole Memo and several other related documents in January 2018. See Memorandum from Att'y Gen. Jefferson B. Session III to United States Att'ys (Jan. 4, 2018) (https://bit.ly/3fGUH5p). However, despite initial uncertainty, there were few significant changes in federal policy during the Trump Administration, and state programs continued to operate and proliferate. See Kyle Jaeger, Biden AG PickResates Pledge to Respect State Marijuana Laws, in Writing, MARIJUANA MOMENT (Mar. 1, 2021), https://bit.ly/3iOsD3n.

33. See Daniller, supranote 7. 34. See id. 35. See id. 36. Most notably, in December 2020 the House of Representatives passed the Marijuana Opportunity Reinvestment and Expungement (MORE) Act, a decriminalization bill that attracted 120 cosponsors. See Marijuana Opportunity Reinvestment and Expungement Act, H.R. 3884, 116th Cong. § 3 (2019-2020). An identical bill was

introduced in the Senate by now-Vice President Kamala Harris and had attracted 8 cosponsors as of August 2021. See Marijuana Opportunity Reinvestment and Expungement Act, S. 2227, 116th Cong. § 2 (2019-2020).

37. Advocating for the Responsible Cannabis Industry, NAT'L CANNABIS INDUS.

ASS'N., https://bit.ly/3hN6YIj (last visited July 12, 2021).

38. See sources cited supra note 8. 39. See Daniel G. Orenstein & Stanton A. Glantz, CannabisLegalization in State Legislatures: Public Health Opportunity and Risk, 103 MARQ. L. REV. 1313, 1340-85 (2020) (evaluating existing state adult use cannabis laws against a rubric of public health best practices); see also Rachel A. Barry & Stanton A. Glantz, MarijuanaRegulatory Frameworksin Four US States:An Analysis Against a PublicHealth Standard, 108 AM. J. PUB. HEALTH 914, 914-15 (2018) (applying similar public health standards to the first four state adult use cannabis laws); see also John T. Carnevale et al., A Practical Frameworkfor RegulatingFor-ProfitRecreationalMarijuanain US States:Lessonsfrom Coloradoand Washington, 42 INT'L J. DRUG POL'Y 71, 80 (2017) (applying a related but distinct set of public health standards); see also Rosalie Liccardo Pacula, States Need to Wake Up to Public Health Risks from Cannabis, STAT NEWS (Jan. 21, 2020), https://bit.ly/3vttytD.

40. See U.S. DEP'T OF HEALTH AND HUMAN SERVS., THE HEALTH CONSEQUENCES OF INVOLUNTARY EXPOSURE TO TOBACCO SMOKE: A REPORT OF THE SURGEON GENERAL 11 (2006).

41. See K. Michael Cummings & Robert N. Proctor, The ChangingPublicImage of

Smoking in the United States: 1964-2014, 23 CANCER EPIDEMIOLOGY, BIOMARKERS & PREVENTION 32, 32-33 (2014).

42. See id. at 33. 43. Norma R. Kelly & Felissa L. Cohen, Smoking Policiesin U.S. Hospitals:Current Status, 8 PREVENTIVE MED. 557, 557 (1979). Additionally, 56% of hospitals sold cigarettes on the premises, and smoking was permitted in 20% of labor and delivery areas, 35% of pediatrics areas, 69% of wards, 72% of semi-private rooms, and 79% of private rooms. See id. at 557-60.

44. See 29 Years of Smokefree Skies, AM. NONSMOKERS' RTS. FOUND. (Feb. 14,

2019), https://bit.ly/349GYin.

45. See STANTON A. GLANTZ & EDITH D. BALBACH, TOBACCO WAR: INSIDE THE CALIFORNIA BATTLES 1-18 (2000); Andrew Hyland et al., Smoke-free Air Policies:Past, Present and Future, 21 TOBACCO CONTROL 139, 155-58 (2012).

46. See MODEL ORDINANCE PROHIBITING SMOKING IN ALL WORKPLACES AND PUBLIC (AMS. FOR NONSMOKERS' RTS. 2018)

PLACES (100% SMOKEFREE) §§ 1004-1008 [hereinafter ANR MODEL ORDINANCE].

47. See Tomasz R. Sosnowski & Marcin Odziomek, ParticleSize Dynamics: Toward a Better UnderstandingofElectronic Cigarette Aerosol Interactionswith the Respiratory System, 9 FRONTIERS IN PHYSIOLOGY 853, 853 (2018).

48. See, e.g., Electronic Cigarettes: Secondhand Aerosol Is Unhealthy, AMS. FOR NONSMOKERS' RTS., https://bit.ly/3uaj28Y (last visited July 14, 2020); ElectronicSmoking Devices and SecondhandAerosol, AMS. FOR NONSMOKERS' RTS., https://bit.ly/3wvMeZK (last visited July 14, 2020); ANR MODEL ORDINANCE § 1002(R) (defining "smoking" to include use of electronic smoking devices).

49. World Health Organization [WHO], Framework Convention on Tobacco Control, opened for signature May 21, 2003, 2302 U.N.T.S. 166 [hereinafter WHO, FCTC].

50. WHO, Guidelines for Implementation ofArticle 8: Protection from Exposure to

Tobacco Smoke 1 (2007) [hereinafter WHO, Guidelines], https://bit.ly/3FoKjeN.

51. The agreement has 182 Parties (as of April 2021). See WHO, FCTC, supra note 49. Although the U.S. is not a Party, id., it has adopted many elements of the treaty. See Family Smoking Prevention and Tobacco Control Act of 2009, Pub. L. No. 111-31, 123 Stat. 1776 (codified in scattered sections of 21 U.S.C.). 52. See WHO, Guidelines, supra note 50, at 1. 53. See id. 54. WHO, FCTC, supranote 49, art. 8, cl.2. 55. WHO, Guidelines, supra note 50, at 2. 56. See id. at 4. 57. Id. at 5. 58. See id. at 2.

59. See ANR MODEL ORDINANCE, supranote 46, §§ 1004-1007. 60. See id. § 1008. 61. Id. 62. See id. §§ 1008(C)-(K). 63. Separate limitations apply to some other areas, such as outdoor workplaces. See id. § 1009. Additionally, the closely affiliated American Nonsmokers' Rights Foundation ("ANRF") recommends some specific smokefree outdoor areas as a policy model, but these are limited to beaches, public transit waiting areas, dining and bar patios, and parks. See Municipalities with Smokefree Beach Laws, AM. NONSMOKERS' RTS. FOUND., https://bit.ly/3oKP4XV (last updated Oct. 2, 2017); Municipalities with Smokefree Outdoor Public Transit Waiting Area Laws, AM. NONSMOKERS' RTs. FOUND., https://bit.ly/349tY (last updated Oct. 2, 2017); Municipalities with Smokefree Outdoor DiningandBar Patio Laws, AM. NONSMOKERS' RTS. FOUND., https://bit.ly/2QRe4R2 (last updated Apr. 1, 2021); Municipalitieswith Smokefree ParkLaws, AM. NONSMOKERS' RTS. FOUND., https://bit.ly/3hLc5Zv (last updated Oct. 2, 2017).

64. ANR MODEL ORDINANCE, supra note 46, § 1001. 65. See generally Stella A. Bialous & Stanton A. Glantz, ASHRAE Standard 62: TobaccoIndustry'sInfluence overNationalVentilationStandards, 11 TOBACCO CONTROL 315 (2002) (describing the importance of ASHRAE standards and tobacco industry efforts to influence them).

66. See ASHRAE PositionDocument on Environmental TobaccoSmoke, AM. SOC'Y

HEATING, REFRIGERATING & AIR-CONDITIONING ENG'RS https://bit.ly/3hOl soI.

(Oct.

22, 2010),

67. See ANSI/ASHRAE Standard62.1-2019: Ventilationfor Acceptable IndoorAir Quality, AM. SOC'Y HEATING, REFRIGERATING & AIR-CONDITIONING ENG'RS (2019), https://bit.ly/3yBrdlv. 68. See id. 69. See Overview List-Number ofSmokefree and OtherTobacco-RelatedLaws, AM.

NONSMOKERS' RTS. FOUND. (July 1, 2021), https://bit.ly/3ygrCDI.

70. See id. 71. See, e.g., Lindsay F. Wiley, Tobacco Denormalization, Anti-Healthism, and

Health Justice, 18 MARQ. BENEFITS & SOC. WELFARE L. REv. 203,219-20 (2017).

72. See id. at220-21; UNFILTERED: CONFLICTS OVER TOBACCO POLICY AND PUBLIC HEALTH 23-24 (Ronald Bayer & Eric Feldman eds. 2004) (quoting Stanton A. Glantz, Achieving a Smokefree Society, 76 CIRCULATION 746, 750 (1987 ; see also Scott Burris, DiseaseStigma in U.S. PublicHealthLaw, 30 J. L. MED. & ETHICS 179, 187 (2002) (noting that smoking is no longer viewed as "a glamorous activity" but rather "has been transformed into antisocial self-destruction" and that law "has played a role in this by, for example, forcing smokers who wish to light up inpublic to congregate in special and often undesirable areas, such as outside the doors of smoke-free facilities or in sepulchral basement smoking rooms").

73. See Wiley, supranote 71, at 221. 74. See Heather Morton, Open Container and Consumption Statutes, NAT'L CONE. OF STATE LEGISLATURES (June 9, 2021), https://bit.ly/3fnXjF2; Joe Satran, The Secret History ofthe War on PublicDrinking, HUFFPOST (Dec. 6, 2017), https://bit.ly/3owOjtR. Notably, a minority of states take a far more permissive approach. For example, seven states allow passengers (but not drivers) to consume alcohol in a motor vehicle, and Mississippi has no state law prohibiting consumption or possession in a vehicle (though operating under the influence is prohibited). See Morton, supranote 74.

75. See Morton, supra note 74. 76. See Satran, supranote 74. 77. See id.; Allan E. Korpela, Annotation, Prosecution of Chronic Alcoholic for Drunkenness Offenses, 40 A.L.R.3d 321 § 10 (1971); I.J. Schiffres, Annotation, Validity of Vagrancy Statutes and Ordinances, 25 A.L.R.3d 792 § 15 (1969); Robinson v. California, 370 U.S. 660, 677-78 (1962) (regarding narcotics). Contra Powell v. Texas, 392 U.S. 514, 535-36 (1968).

78. See CTRS. FOR DISEASE CONTROL AND PREVENTION, ALCOHOL USE AND YOUR

HEALTH (2021). 79. See id.

80. See Drunk Driving, NAT'L HIGHWAY

TRAFFIC

SAFETY

ADMIN.,

https://bit.ly/3f1lMyz (last visited May 19, 2021).

81. See Alex Gangitano, CoronavirusBrings Quick Changes to State Alcohol Laws, THE HILL (April 1, 2020, 6:00 AM), https://bit.ly/3bGbNiD; see also Marianna Brady, Coronavirus:How the Pandemic is Relaxing US Drinking Laws, BBC NEWS (May 15, 2020), https://bbc.in/3oxMapG.

82. See, e.g., Ellis Henican, New York Is Rescuing Its Bar Life with Looser Alcohol

Rules, MARKETWATCH (May 29, 2020, 2:53 PM), https://on.mktw.net/2Qvd6JQ.

83. See Gangitano, supra note 81; Brady, supra note 81. 84. Unquestionably, cannabis possession remains illegal under federal law, creating complex questions of federalism that are exacerbated by inconsistent federal policy, but these issues are beyond the scope of this Article.

85. This total excludes South Dakota but includes Connecticut. South Dakota lower courts have found the state's voter initiative unconstitutional, but the state's Supreme Court has yet to issue a final decision on the matter. Connecticut enacted an adult use measure in June 2021. See State MedicalMarijuanaLaws, supra note 8.

86. See id. 87. See Updates: Gov. Murphy Signs Three Bills Ending Cannabis Prohibitionin

N.J., MARIJUANA POL'Y PROJECT (Mar. 3, 2021), https://bit.ly/3f2vRxM.

88. See State MedicalMarijuanaLaws, supranote 8. 89. See id.; see also German Lopez, MarijuanaLegalizationHas Won, Vox (Apr.

20, 2021, 9:20 AM), https://bit.ly/2S5WnNU.

90. See, e.g., Campaign to Regulate Marijuana Like Alcohol, BALLOTPEDIA, https://bit.ly/3owilGX (last visited May 19, 2021) (describing campaign activities of Campaign to Regulate Marijuana Like Alcohol, registered as aballot committee in multiple states and connected to legalization advocacy organization Marijuana Policy Project).

91. See Barry & Glantz, supra note 39, at 1, 4. 92. See Steinberg et al., supranote 22, at 203-04; see also Cork, supranote 22, at 48, 50-51. The three states that passed adult use ballot measures in November 2020 (Arizona, Montana, and South Dakota) also will not permit public use based on the measures' text, though final regulations are not yet operative as of February 2021. See Arizona Proposition 207, the Smart and Safe Arizona Act, ARiz. REV. STAT. § 36-28.2 (2020) (amending ARIZ. REV. STAT. § 36-2851(8)(a) and prohibiting "smok[ing] marijuana in a public place or open space"); MONT. CODE ANN. § 16-12-108(i) (2021) (effective Jan. 1, 2022) (adding prohibition on "consumption of marijuana in a public place, except as allowed by the department"); S.D. CONST. AMEND. A § 2(9) (2020) (retaining prohibition on "[c]onsumption of marijuana in a public place, other than in an area licensed by the department for consumption"). South Dakota's measure is not operative as of this writing, following invalidation by South Dakota courts, but the state's Supreme Court has yet to render a final decision. See State MedicalMarijuanaLaws, supra note 8. New Jersey's 2021 law will allow municipalities to authorize consumption areas but will not allow general public consumption. See N.J. STAT. ANN. 24:61-21(f) (2021).

" Two (New York, Connecticut) permit or will permit at least some

public consumption, and at least one will also allow on-site

consumption (New York);

" Nine allow, or plan to allow, on-site consumption (California, Alaska, Illinois, Michigan, New Mexico, New Jersey, Colorado, Massachusetts, Nevada);

" One has not finalized regulations, but has statutory language that

appears to allow on-site consumption (Montana);

"

Six prohibit on-site consumption (Washington, Oregon, Vermont, Maine, Arizona, Virginia).

93. See Steinberg et al., supranote 22, at 204-06. 94. See supra note 92 (discussing recently enacted laws that prohibit public consumption of cannabis); see also Orenstein & Glantz, supra note 39, at 1381-85 (discussing proposed laws regarding public consumption of cannabis).

95. See N.Y. PUBLIC HEALTH LAW § 1399-q(1)(j) (McKinney 2021) (allowing cannabis consumption at authorized on-site locations provided these locations "may only permit the smoking or vaping of cannabis" and not tobacco products).

96. See id. § 1399-q(1)(c)-(h) (listing other locations where smoking and vaping restrictions are not applicable); id. § 1399-n(8) (defining "smoking" to include both cannabis and tobacco products for purposes of state smokefree air laws).

97. See N.Y. PENAL LAW. § 222.05(1)(c) (McKinney 2021) (providing that "using, smoking, ingesting, or consuming cannabis or concentrated cannabis unless otherwise prohibited by state law" is lawful for persons over age 21). Remaining prohibited consumption locations include locations where smoking is prohibited by the state's smokefree air law or on the grounds of a school. See id. § 222.10(1)-(2).

98. See 2021 Conn. Acts 130 (Spec. Sess.) (amending CONN. GEN. STAT. ANN. § 7-

148(c)(7)(H)(xvi (West 2021).

99. See id. 100. See id. at 96. 101. See Steinberg et al., supranote 22, at 204-06. 102. See id. 103. See David Downs, A Day in the Life ofa CannabisLounge, LEAFLY (Sept. 6,

2018), https://bit.ly/3bGEnMT.

104. See Check If You Can Sell Cannabisat Your Event, CITY AND CNTY. OF SAN FRANCISCO, https://bit.ly/3oACDOn (last visited May 19, 2021); see also Steinberg et al., supranote 22, at 205-06.

105. See ALASKA ADMIN. CODE tit. 3, § 306.370 (2020). 106. See Associated Press, Alaska Pot Shops to be Among 1st in U.S. to Allow

Consumption, NBCNEWS (Jan. 24, 2020), https://nbcnews.to/3fbWhgz. 107. See ALASKA ADMIN. CODE tit. 3, § 306.370(a)(1) (2020). 108. See id. § 306.370(c)(2). 109. See id. § 306.370(a)(1), (a)(2)(C), (b)(3)-(4). 110. See id. § 306.200(a). 111. See Ally Marotti, Illinois'FirsthMarijuanaLounge Gets Approved, HoursAway (Jan. 23, 2020),

'It's Going to Be an Experience.,' CHI. TRIB.

from Chicago. https://bit.ly/3bQAvwK.

112. See 410 ILL. COMP. STAT. ANN. § 705/55-25(3) (West 2019) (authorizing "a unit of local government" to "authorize or permit the on-premises consumption of cannabis at

or in a dispensing organization or retail tobacco store"); see also 410 ILL. COMP. STAT. ANN. § 82/10 (West 2010) (defining "Retail tobacco store").

113. State of Michigan, Adult-Use Marihuana Establishments, Mich. Emergency Rules, Rule 62 (2019) (using the less common spelling "marihuana"); see also 21 U.S.C. § 812(Schedule I)(c)(10) (using the term "marihuana").

114. See State of Michigan, Adult-Use Marihuana Establishments, Mich. Emergency

Rules, Rule 59(7)(a)-(c) (2019).

115. N.M. STAT. ANN. § 26-2C-6(I) (permitting smoking in cannabis consumption areas with required features); N.M. STAT. ANN. § 24-16-12(K) (exempting cannabis consumption areas from the state's Clean Indoor Air Act).

116. See N.J. REV. STAT. §§ 24:6I-33, -42(g) (2021); see also William J. Beneduce, MarijuanaSocialLoungesAre Coming to a New Jersey Town NearYou, NAT'L L.R. (Apr. 5, 2021), https://bit.ly/3udXcS2.

117. See COLO. REV. STAT. § 44-10-609 (2021); COLO. REV. STAT. § 44-10-610

(2021); see also Steinberg et al., supra note 22, at 204.

118. See MED Licensee Information, COLO. DEP'T OF REVENUE (Jan. 1, 2021), https://bit.ly/3oGcy0B (listing no licenses for marijuana hospitality businesses among licensee data).

119. See Steinberg et al., supranote 22, at 204.

120. See Dan Adams, Buying LegalMarijuanaisMore Convenient than Ever. Good (Mar. 7, 2021),

Luck Finding Somewhere to Legally Use It., BOSTON GLOBE https://bit.ly/3wwc7bf.

121. See CouncilorSeeks LegalReview ofSummit Lounge, WORCESTER MAG. (Feb.

22, 2018, 5:01 AM), https://bit.ly/2U7aPql.

122. See MONT. CODE ANN. § 16-12-108(1)(i) (2021) (effective Jan. 1, 2022) (as amended by 2021 Mont. Laws ch. 576, § 43) (prohibiting the "consumption of marijuana or marijuana products in a public place, except as allowed by the department" (emphasis added . South Dakota's approach would be similar should the state's Supreme Court revive the voter initiative invalidated by lower courts. See S.D. CONST. AMEND. A § 2(9) (2020) (retaining prohibition on "[c]onsumption of marijuana in a public place, otherthan in an area licensedby the departmentfor consumption" (emphasis added .

123. See Steinberg et al., supranote 22, at 204-05. 124. See id. 125. See VT. STAT. ANN. tit. 18, § 4230a(a)(2)(A) (2020). 126. See ME. REV. STAT. ANN. tit. 28-B, § 1501(2)(A) (2018). 127. Arizona has not finalized enabling regulations following passage of its 2020 adult use ballot initiative, but the statutory language does not appear to allow for on-site consumption, as it explicitly "[d]oes not allow any person to ... [s]moke marijuana in a public place or open space" and contains no language explicitly excluding licensed establishments from this restriction. See ARiZ. REV. STAT. ANN. § 36-2851(8)(a) (2020).

128. See 2021 Va. Acts 159. 129. See 2020 Nev. Legis. Serv. 3 (West). 130. See Colton Lochhead, Nevada Lawmakers Again Take Up MarijuanaLounges. (Feb. 27, 2021, 8:16 AM),

Will They Pass This Time?, LAS VEGAS REv.-J. https://bit.ly/3wSEf9D.

131. See Dan Hernandez,

'The Tribe Has Taken Over': The Native Americans 11, 2019),

(Nov.

Running Las Vegas's Only Cannabis Lounge, GUARDIAN https://bit.ly/3oLqzKu.

132. See A.B. 341, 81st Gen. Assemb., 2021 Sess. (Nev. 2021).

133. See John Walsh & Geoff Ramsey, Uruguay's DrugPolicy:MajorInnovations, (2016),

21ST CENTURY

SEC. AND

INTEL.

FOR

Major Challenges, CTR. https://brook.gs/2Sly26A.

134. See id. at 9. 135. See id. at 10; see also 1 Uruguay National Registry of Laws and Decrees

120/014, tit. 1 Art. 27, 40 (2014).

136. See 1 Uruguay National Registry of Laws and Decrees 120/014, tit. 1 Art. 27,

40 (2014).

137. See Cannabis in the Provinces and Territories, GOv'T OF CANADA (Jan. 29,

2021), https://bit.ly/3zWKTO2.

138. See CannabisLaws: Where You Can Use It, GOv'T OF ONT. (Dec. 13, 2019),

https://bit.ly/3fFflnS.

139. See id. 140. See id. 141. See Alberta Cannabis Framework and Legalization: Consuming Cannabis, GOv'T OF ALTA. [hereinafterAlbertaCannabis],https://bit.ly/3hNbWVo (last visited May 22, 2021).

142. See Factsheet: Public Consumption, GOv'T OF B.C., https://bit.ly/3wu5FBU

(last visited May 22, 2021).

81l

143. See Alberta Cannabis,supranote 141. 144. See Amanda Scriver, Everything You Need to Know About Legal Cannabis in

Canada,THRILLIST (Apr. 20, 2020), https://bit.ly/3yvVzm3.

145. For clarity, a "coffee shop" or "coffeeshop" (koffieshop in Dutch) sells cannabis, while a "coffee house" (koffie huis, koffiebar, or koffiesalon in Dutch) sells coffee. See Ross Scully, How Are 'Coffeeshops' DifferentFrom 'Coffee Shops'?, LEAFLY (Mar. 14, 2017), https://bit.ly/3yyKwZa.

146. See Marianne M. J. van Ooyen-Houben, The Dutch Coffee Shop System, Tensions andBenefits, 25 MICH. ST. INT'L L. REV. 623, 628-29 (2017). Some popular media do capture more of the nuance of the drug's unusual legal status. See PULP FICTION (Miramax Home Entertainment 1994) (Jules, asking Vincent about recent travel to the Netherlands: "Hash is legal there, right?" Vincent: "Yeah, it's legal, but it ain't a hundred percent legal. I mean, you can't walk into a restaurant, roll a joint and start puffing away. I mean, they want you to smoke in your home or certain designated places.").

147. See van Ooyen-Houben, supranote 146, at 629-32. 148. See id. 149. See John Bronsteen, Would "Hamsterdam"Work? DrugDepenalizationin The

Wire andin Real Life, 2018 U. CHI. LEGAL F. 43, 58-60.

150. See CAULKINS ET AL., supra note 19.

151. For example, the alcohol and tobacco models can also allow use in private residences. See JUNEAU, ALASKA, MUN. CODE §36.60.030 (allowing cannabis smoking in private residences and in retail establishments licensed for on-site consumption). Additionally, New York's 2021 adult use law straddles the tobacco and alcohol models, allowing both outdoor public use and on-site consumption sites. See N.Y. PUB. HEALTH § 1399-q.

152. As applied to cannabis, the closest alcohol analogue is bars, which may offer few other products, as compared to restaurants or other businesses that offer many other products or services beyond alcohol and thus often do not restrict entry by age.

153. New York's approach does not meet the standard of the tobacco control model because in addition to allowing outdoor public use in compliance withthe state's smokefree air law the state also plans to license indoor consumption lounges. 154. See NASEM Report, supranote 16, at 9, 51-53. 155. See Dai & Richter, supra note 15.

POTENTIAL RISKS OF PUBLIC CANNABIS USE

156. See Jeffrey M. Jones, U.S. Supportfor Legal MarijuanaSteady in Past Year,

GALLUP (Oct. 23, 2019), https://bit.ly/3oJuJlS; see also Daniller, supranote 7.

157. See Daniel G. Orenstein & Stanton A. Glantz, The Grassroots of Grass:

CannabisLegalization Ballot Initiative Campaign Contributionsand Outcomes, 20042016,45 J. HEALTH POL., POL'Y & L. 73, 74 (2020).

158. Following a run of successful state and local measures to decriminalize cannabis possession in the 1970s, liberalization of cannabis laws appeared likely, but the election of Ronald Reagan as president reversed course and ushered in an era of highly restrictive drug policy. See MARK K. OSBECK & HOWARD BROMBERG, MARIJUANA LAW IN A NUTSHELL 38 (2017).

159. See, e.g., EZEKIAL EDWARDS ET AL., ACLU, THE WAR ON MARIJUANA IN BLACK AND WHITE (2013) [hereinafter ACLU, THE WAR ON MARIJUANA IN BLACK AND WHITE]; Christopher J. Coyne & Abigail R. Hall, Four Decades and Counting: The Continued Failureofthe War on Drugs, CATO INST. (Apr. 12, 2017), https://bit.ly/2T64KJz; Graham Boyd, The Drug War is the New Jim Crow, NACLA REP. ON THE AMS. (July 1, 2001), https://bit.ly/3vdGbbR; James F. Mosher & Karen L. Yanagisako, Public Health, Not Social Warfare: A Public Health Approach to Illegal Drug Policy, 12 J. PUB. HEALTH POL'Y 278, 300-01 (1991).

160. See NASEM Report, supra note 16, at 61-65 (describing trends in prevalence of use since 1970s); see also Christopher Ingraham, 11 Charts That Show Marijuana Has Truly Gone Mainstream, WASH. POST (Apr. 19, 2017), https://wapo.st/3yyXrKE.

161. See NASEM Report, supranote 16, at 395-401 (cataloging research barriers). 162. See id. at 15-21. The report also addresses therapeutic effects of medicinal use for certain conditions, see id. at 13-14, but regarding recreational use the potential negative effects are more relevant.

163. See id. at 19; see also Wayne Hall & Michael Lynskey, Assessing the Public Health Impacts ofLegalizingRecreationalCannabisUse: The US Experience, 19 WORLD PSYCHIATRY 179, 181 (2020) (presenting collected studies and findings).

164. See David Moir et al., A ComparisonofMainstreamandSidestreamMarijuana

and Tobacco Cigarette Smoke Producedunder Two Machine Smoking Conditions, 21 CHEMICAL RES. TOXICOLOGY 494, 494 (2008); see also RAJPAL S. TOMAR ET AL., CAL. ENV'T PROT. AGENCY, EVIDENCE ON THE CARCINOGENICITY OF MARIJUANA SMOKE 77-78 (2009), https://bit.ly/2Sjnh4Y (providing evidence supporting inclusion of cannabis smoke on California's Proposition 65 list of substances known to cause cancer or reproductive toxicity, including similarity to tobacco).

165. See JEFFREY DROPE ET AL., AM. CANCER SOC'Y & VITAL STRATEGIES, THE

TOBACCO ATLAS 24-31 (6th ed. 2018), https://bit.ly/3wvNDj 1.

166. See, e.g., Dai & Richter, supra note 15, at 3 (finding current cannabis use reported by 8.9% of non-Hispanic white respondents, 10.7% of non-Hispanic Black respondents, 6.3% of Hispanic respondents, and 7.3% of respondents of other races or ethnicities). Prevalence of daily use was also similar (4.0% Non-Hispanic white; 5.5% Non-Hispanic Black; 2.4% Hispanic; 2.5% other races/ethnicities). See id. at 2

167. See ACLU, THE WAR ON MARIJUANA IN BLACK AND WHITE, supranote 159, at

4, 9.

168. See EZEKIEL EDWARDS ET AL., ACLU, A TALE OF TwO COUNTRIES: RACIAL

TARGETED ARRESTS https://bit.ly/3fJsECC.

IN THE ERA OF MARIJUANA REFORM 4-6, 28-29

(2020),

169. See JACK REED ET AL., COLO. DIV. OF CRIM. JUST., DEP'T OF PUB. SAFETY, (2018),

OF MARIJUANA

LEGALIZATION

COLORADO

20-21

IMPACTS https://bit.ly/3ubWvsz.

170. See generally Kevin E. Jason, Dismantling the Pillars of White Supremacy: Obstaclesin EliminatingDisparitiesandAchieving RacialJustice, 23 CUNY L. REV. 139, 142-48, (2020) (describing various aspects of white supremacy that underlie existing disparities, including differences in law enforcement approach related to race); see also Lydia Denworth, A Civil Rights ExpertExplains the Social Science ofPolice Racism, SCI. AM. (June 4, 2020), https://bit.ly/2TOqPtG.

171. See RacialDisparitiesin D.C. Policing:DescriptiveEvidencefrom 2013-2017,

ACLU: D.C. (July 31, 2019), https://bit.ly/3wpKIrP.

172. See id.

173. See id. 174. See id. 175. See, e.g., Ashley Southall, 'ICan 'tBreathe':5 YearsAfter Eric Garner'sDeath, an Officer Faces Trial, N.Y. TIMES (May 12, 2019), https://nyti.ms/3ywtOVD; Eric Garner: NY Officer in 'I Can't Breathe' Death Fired, BBC NEWS (Aug. 19, 2019), https://bbc.in/3fbCj CJ.

176. See Mike Baker et al., Three Words. 70 Cases. The TragicHistory of 'ICan't

Breathe.,'N.Y. TIMES (June 29, 2020), https://nyti.ms/3bMxill.

177. See id. 178. See Malum Prohibitum,BLACK'S LAW DICTIONARY (11th ed. 2019) ("An act that is a crime merely because it is prohibited by statute, although the act itself is not necessarily immoral.").

179. See Sabina Morris et al., State CannabisReform Is PuttingSocialJusticeFront

andCenter, BROOKINGS (Apr. 16, 2021), https://brook.gs/3wXOysR.

Sociological literature suggests that differential access to private space, such as homeownership, increased the likelihood that illegal behaviors, such as drug use, were detected. Thus, it is theoretically logical to project that legalizing cannabis use overall without also legalizing to disproportionally affect certain cohorts, such as persons of color and those unable to afford housing. In this regard, cannabis is not equally accessible across racial and socio-economic cohorts . . . . [E]nactment of laws permitting the adult use of cannabis may serve as an impact of some of the opportunity

the disparate

space may

consumption

to address

continue

public

180. See JOINT CTR. FOR HOuS. STUD. OF HARV. U., THE STATE OF THE NATION'S

HOUSING 19 (2017), https://bit.ly/2TcmLWT.

181. See id. 182. See id. 183. See id. 184. See id. 185. See id. at 27-28. 186. See, e.g., Cork, supranote 22, at 64-65.

discriminatory policies disproportionately affected people of color.'

enforcement

actions

that

187. Memorandum from Mass. Cannabis Control Comm'n Staff on Soc. Consumption to Cannabis Control Comm'n, Cannabis Control Comm'n (Oct. 4, 2018) (https://bit.ly/3uiAA2E).

188. See Lahny R. Silva, CollateralDamage:A PublicHousing Consequence ofthe

"War on Drugs," 5 U.C. IRVINE L. REV. 783, 801 (2015).

189. See id. at 791-93, 800-01. 190. See id. at 809. 191. See Anti-Drug Abuse Act of 1988, Pub. L. No. 100-690, 102 Stat 4181; see also Marah A. Curtis et al., Alcohol, Drug, and Criminal History Restrictions in Public Housing, 15 CITYSCAPE: J. POL'Y DEV. &REs. 37, 39 (2013).

192. See, e.g., Veterans Affairs and Hud Appropriations Act, Pub. L. No. 105-276, 112 Stat. 2461; Cranston-Gonzalez National Affordable Housing Act, Pub. L. No. 101625, 104 Stat. 4079; Housing Opportunity Program Extension Act of 1996, Pub. L. No. 104-120, 110 Stat. 834; Curtis et al., supra note 191, at 39.

193. See 24 C.F.R. § 960.204(a)(1) (2021). 194. See id. 195. Id. § 960.204(a)(2).

196. See 24 C.F.R § 982.553(a)(1) (2021). 197. Scott Burris et al., HealthEquity in Housing:Evidence andEvidence Gaps,CTR.

FOR PUB. HEALTH L. REs. 33 (Nov. 2019), https://bit.ly/34iduiO.

198. See id. at 31-32; see generally Curtis et al., supra note 191, at 43-49 (finding that public housing authorities use their significant discretion in a manner that creates barriers amounting to de facto bans on some populations due to suspected or verified alcohol, drug, or criminal history).

199. See 24 C.F.R. § 960.204(b) (2021); 24 C.F.R. § 982.553(a)(ii)(C)(3) (2021). 200. Many health and addiction science experts consider "abuse" a stigmatizing term, including at NIDA, whose own name includes it. See, e.g., Words Matter-Terms to Use and Avoid When Talking About Addiction, NAT'L INST. ON DRUG ABUSE, https://bit.ly/3wx9UNm (last visited May 24, 2021).

201. See JOINT CTR. FOR HoUs. STUD. OF HARV. U., supra note 180, at 33-34. 202. See id. 203. See id. 204. See id.

205. See Hannah Knowles, Homelessness in the U.S. Rose for a Third Year, Driven by a Surge in California, HUD Says, WASH. POST (Dec. 21, 2019, 3:35 PM), https://wapo.st/2TaH077. In addition to California's 16.4% increase (a startling 21,000 people), homelessness also increased by 27% in New Mexico and over 10% in Idaho, West Virginia, Kentucky, and Minnesota. See id.

206. Seegenerally The Bureau of Labor Statistics, The Recession of2007-2009, BLS SPOTLIGHT ON STATISTICS (Feb. 2012), https://bit.ly/3wvag6X (according to the U.S. Bureau of Labor Statistics, the most recent recession began in December 2007 and ended in June 2009, during which the national unemployment rate nearly doubled from 5.0% to 9.5%, a more rapid employment decline than the U.S. experienced in other recessions).

207. See generally Emily A. Benfer et al., Eviction, Health Inequity, andthe Spread of COVID-19: HousingPolicy as a PrimaryPandemic MitigationStrategy, 98 J. URBAN HEALTH 159 (2021) (discussing implications of eviction and housing displacement due to pandemic-related job loss and economic hardship).

208. See Ron S. Hochbaum, Bathroomsas a Homeless Rights Issue, 98 N.C. L. REv. 205, 245 (2020) (discussing how criminalization of homelessness leads to unduly frequent contact with law enforcement and the criminal justice system for persons experiencing homelessness).

209. See, e.g., Megan Testa & Sara G. West, CivilCommitment in the UnitedStates,

7 PSYCHIATRY 30, 34-35 (2010).

210. See generally Jessica Mogk et al., Court-imposedFines as a Feature of the Homelessness-IncarcerationNexus: A Cross-sectionalStudy ofthe RelationshipBetween LegalDebtandDurationofHomelessnessin Seattle, Washington, USA, 42 J. PUB. HEALTH 107 (2020) (finding a strong association between homelessness and legal trouble and concluding that the relationship may be connected); see also Hochbaum, supra note 208, at 244-45 (discussing how citations for violations of antihomeless laws lead to warrants that make individuals ineligible for certain public benefits and impede efforts to obtain and retain employment and housing).

211. See Hochbaum, supranote 208, at 243-45.

212. NAT'L CTR. ON HOMELESSNESS & POVERTY, HOUSING NOT HANDCUFFS: ENDING THE CRIMINALIZATION OF HOMELESSNESS IN U.S. CITIES 21 (2018), https://bit.ly/2VGqgGf. 213. See, e.g., U.S. DEP'T OF HEALTH AND HUMAN SERVS., supra note 40, at 11-16

(summarizing report conclusions).

214. E.g., Geoffrey McKee et al., Protecting the Public from Exposure to Secondhand CannabisSmoke and Vapour Following Legalization, 109 CAN. J. OF PUB. HEALTH 223, 224 (2018); Aurdlie Berthet et al., A Systematic Review ofPassive Exposure to Cannabis,269 FORENSIC SCI. INT'L 97, 110 (2016).

215. See Xiaoyin Wang et al., One Minute of Marijuana Secondhand Smoke Exposure SubstantiallyImpairs Vascular EndothelialFunction, 5 J. AM. HEART ASS'N, July 2016, at 9.

216. See Moir et al., supranote 164, at 501. 217. See Christina Sempio et al., Surface Detection THC Attributable to Vaporizer

Use in the IndoorEnvironment, SCI. REP., Dec. 2019, at 6.

218. E.g., Jolanda Palmisani et al., Evaluation of Second-Hand Exposure to

Electronic Cigarette Vaping Under a Real Scenario:Measurements of UltrafineParticle Number ConcentrationandSize Distributionand Comparisonwith TraditionalTobacco Smoke, TOxicS, Nov. 2019, at 6, 7.

219. See OutbreakofLungInjuryAssociatedwith the Use ofF-Cigarette,or Vaping, Products, CTR. FOR DISEASE CONTROL AND PREVENTION, https://bit.ly/3oMs2Aa (last visited May 25, 2021).

220. See, e.g., Andrew Hyland et al., Smoke-free Air Policies: Past, Present and Future, 21 TOBACCO CONTROL 154, 155 (2011); SecondhandSmoke Worker Health, AM. NONSMOKERS' RTS. FOUND., https://bit.ly/3vl2LiN (last visited May 25, 2021); INST. OF MED. COMM. ON SECONDHAND SMOKE EXPOSURE AND ACUTE CORONARY EVENTS, SECONDHAND SMOKE EXPOSURE AND CARDIOVASCULAR EFFECTS: MAKING SENSE OF THE EVIDENCE 113 (2010).

221. See, e.g., Mathew Swinburne & Kathleen Hoke, State Efforts to Create an

Inclusive MarijuanaIndustry in the Shadow of the Unjust War on Drugs, 15 J. BUS. & TECH. L. 235, 261-75 (2020).

222. See, e.g., Tom Schuba, Been Busted for Weed? That Will Actually Help You Land a Job When It's Legalized, CHI. SUN TIMES (Nov. 4, 2019, 6:00 AM), https://bit.ly/3 hRbHIO.

223. See, e.g., IndoorAir Quality (IAQ): Is OutdoorExposure to SecondhandSmoke

Comparable https://bit.ly/3vfbNxW.

Indoors, U.S. ENV'T.

PROT.

AGENCY

(Aug.

11,

2020),

224. See Nicholas Evoy, Comment, SecondhandSmoke as a PrivateNuisance: Lost

in the Fog, 44 REAL EST. L.J. 20, 22 (2015).

225. See generally Alexander Posis et al., Indoor CannabisSmoke and Children's Health, 14 PREVENTIVE MED. REP. 100853 (2019) (assessing how indoor secondhand cannabis smoke affects children's health and finding a large, but not statistically significant, association).

226. See, e.g., Kerry Cork, Toking, Smoking, and Public Health: Lessons from Tobacco Control for Marijuana Regulation, PUB. HEALTH L. CTR. 1 (Jul. 2018), https://bit.ly/3bQFBcm; David M. Homa et al., Vital Signs: Disparitiesin Nonsmokers' Exposureto SecondhandSmoke -UnitedStates, 1999-2012, 64 MORBIDITY & MORTALITY WKLY. REP. 103 (2015).

(Supp.

227. See, e.g., Emily Anne McDonald et al., Traversing the Triangulum: The Intersection of Tobacco, Legalised Marijuana and Electronic Vaporisers in Denver, Colorado, 25 that "[p]articipants consistently viewed secondhand tobacco smoke negatively, commenting that it 'smells bad' and is harmful for bystanders" and quoting cannabis-using participants' statements regarding tobacco that included, "I'm glad that you can't smoke cigarettes in a bar because I personally don't like the smell. I think it leaves a really gross stench.").

1) TOBACCO CONTROL

i96-i102

(stating

(2016)

i96,

228. Seegenerally Evoy, supra note 224; DriftingTobacco Smoke & Legal Solutions

for Business Owners, TOBACCO CONTROL LEGAL CONSORTIUM https://bit.ly/3wtGC 1W.

1,

(2010),

229. See NASEM Report, supranote 16, at 53. 230. See id.

231. See id. at 217, 227-30. 232. See id. at 227. 233. See generally Johannes G. Ramaekers, Driving Under the Influence of Cannabis:An IncreasingPublic Health Concern, 319 J. AM. MED. Ass'N. 1433 (2018) (summarizing several studies of driving under the influence of cannabis); U.S. DEP'T OF TRANSP., NAT'L HIGHWAY TRAFFIC SAFETY ADMIN., MARIJUANA-IMPAIRED DRIVING -A REPORT TO CONGRESS (2017) (providing a comprehensive survey of existing evidence on cannabis intoxication and driving, including detection, measurement, legal provisions, and impairment and providing recommendations); Ole Rogeberg & Rune Elvik, The Effects of CannabisIntoxication on Motor Vehicle CollisionRevisited andRevised, 111 ADDICTION 1348 (2016) (presenting results of meta-analyses, finding a statistically significant low to medium magnitude increase in motor vehicle crash risk associated with acute cannabis intoxication, and discussing comparisons to alcohol and resulting policy implications).

234. See R. Andrew Sewell et al., The Effect ofCannabisComparedwith Alcohol on Driving, 18 AM. J. ADDICTIONS 185 (2009). This is closely related to why roadside testing for cannabis impairment and establishing blood concentration levels for per se or presumptive impairment are significant cannabis policy challenges. See, e.g., 2 KEVIN ZEESE, DRUG TESTING LEGAL MANUAL § 9.17 (2d ed. 2020); Eric M. Langton, Comment, Regulating Cannabis-Using Drivers: Why PerSe Laws Are Scientifically Invalid, 34 W. MICH. U. T.M. COOLEY L. REV. 367, 380-83 (2018); Alicia Wallace, Testing Driversfor CannabisIsHard. Here 's Why, CNN.COM (Jan. 2, 2020), https://cnn.it/2QKH9NZ.

235. See Sewell et al., supra note 234. 236. See id.

237. 45 AM. JUR. 2D Intoxicating Liquors § 27 (2021). 238. 14 C.J.S. Elements ofOffense ofPublicIntoxication § 9 (2021). 239. WORLD HEALTH ORG., THE

ICD-10 CLASSIFICATION OF MENTAL AND

BEHAVIOURAL DISORDERS: DIAGNOSTIC CRITERIA FOR RESEARCH, 1, 2 (1992).

240. See id. at 4. 241. See Sophie Legros & Beniamino Cislaghi, Mapping the Social-Norms Literature: An Overview of Reviews, 15 PERSP. ON PSYCHOL. SCI. 62, 66-70 (2020) (describing debate over definitions of norms and pathways by which they influence behavior).

242. See, e.g., ERIC A. POSNER, LAW AND SOCIAL NORMS 1-8 (2002).

243. See Brian C. Kelly et al., Denormalization,Smoke-free Air Policy, and Tobacco Use Among Young Adults, 211 SOC. SCI. & MED. 70, 70-72 (2018); INSTITUTE OF MED., GROWING UP TOBACCO FREE: PREVENTING NICOTINE ADDICTION IN CHILDREN AND YOUTHS 71-73, 76-77, 87-88 (Barbara S. Lynch & Richard J. Bonnie, eds., 1994).

244. See Ronald Bayer, Stigma and the Ethics of Public Health: Not Can We But

Should We, 67 SOC. SCI. & MED. 463, 470 (2008); see also, Wiley, supranote 71, at 24647.

245. See Lawrence v. Texas, 539 U.S. 558, 567 (2003). 246. See Kelly et al., supra note 243, at 71 ("[S]exual activity between two consenting adults is considered normative adult behavior, and yet acts of public sexual interaction are decidedly non-normative. The stigma is associated with the public performance of the act, rather than the act itself.").

247. See Wiley, supranote 71, at 246-48, 251-53. 248. See, e.g., James F. Childress et al., PublicHealth Ethics: Mappingthe Terrain,

30 J. L. MED. &ETHICS 170, 175-76 (2002).

249. See Gillian L. Schauer et al., Modes of Marijuana Use -Smoking, Vaping, Eating, and Dabbing:Results from the 2016 BRFSS in 12 States, 209 DRUG & ALCOHOL DEPENDENCE 107900 (2020).

250. See, e.g., MarijuanaThreatensSmokefree Protections,AM. NONSMOKERS' RTS. FOUND. (Apr. 18, 2019), https://bit.ly/3wgQt8O; Colorado Tobacco Free Alliance Partners Oppose Senate Bill 17-184, CANCER ACTION NETWORK (Apr. 7, 2017), https://bit.ly/3fCSYOB; The Triangulum: Tobacco, Marijuana and Electronic Smoking Devices (ESD) Position Statement, TOBACCO EDUC. AND RES. OVERSIGHT COMM. (Sept. 29, 2016), https://bit.ly/2T8432p.

251. See HEALTH CANADA, A FRAMEWORK FOR THE LEGALIZATION AND REGULATION

OF CANNABIS IN CANADA 41 (2016), https://bit.ly/34afgJv.

252. See, e.g., Carla Alexia Campbell et al., The Effectiveness ofLimiting Alcohol Outlet Density as a Means of Reducing Excessive Alcohol Consumption and AlcoholrelatedHarms, 37 AM. J. PREVENTIVE MED. 556, 560 (2009); see also Task Force on Community Preventive Services, Recommendations for Reducing Excessive Alcohol Consumption and Alcohol-relatedHarms by Limiting Alcohol Outlet Density, 37 AM. J. PREVENTIVE MED. 570, 570 (2009).

253. See, e.g., Lisa Henriksen et al., Is Adolescent Smoking Related to the Density and Proximity of Tobacco Outlets and Retail Cigarette Advertising Near Schools?, 47 PREVENTIVE MED. 210, 211 (2008); Scott P. Novak et al., Retail Tobacco Outlet Density and Youth CigaretteSmoking: A Propensity-modelingApproach, 96 AM. J. PUB. HEALTH 670, 673 (2006); Laura J. Finan et al., Tobacco OutletDensity andAdolescents' Cigarette Smoking: A Meta-analysis,28 TOBACCO CONTROL 27, 31 (2019).

254. See, e.g., John E. Schneider et al., Tobacco Outlet Density and Demographics atthe TractLevel ofAnalysis in Iowa: ImplicationsforEnvironmentallyBasedPrevention Initiatives, 6 PREVENTIVE SCI. 319, 322 (2005).

255. See, e.g., ACLU, THE WAR ON MARIJUANA IN BLACK AND WHITE, supra note 159; Swinburne & Hoke, supra note 221, at 249-55; Elizabeth Danquah-Brobby, Comment, Prisonfor You. Profitfor Me. Systemic Racism Effectively Bars Blacks from Participationin Newly-Legal Marijuana Industry, 46 U. BALT. L. REV. 523, 524-27 (2017); Memorandum from Mass. Cannabis Control Comm'nto Comm'n Stat., supranote 187.

III. APPLICATION TO CANNABIS PUBLIC USE TYPOLOGY

256. See, e.g., Memorandum from Mass. Cannabis Control Comm'n to Comm'n Staff, supra note 187, at 5-6; Somaz Amiri et al., Availability of Licensed Cannabis Businesses in Relation to Area Deprivation in Washington State: A Spatiotemporal Analysis of CannabisBusiness Presence Between 2014 and 2017, 38 DRUG & ALCOHOL REV. 790, 795-97 (2019) (finding higher likelihood of increased cannabis outlet density in most-deprived areas); Yuyan Shi et al., Availability of Medical and Recreational MarijuanaStores and Neighborhood Characteristicsin Colorado,2016 J. ADDICTION 1, 5-7 (2016) (finding that dispensaries and retailers were more likely to locate in areas with higher proportion of racial and ethnic minority population and lower household income).

257. See J.K. Dineen, SF's 'Green Rush'forNew CannabisStores -and a Growing Opposition, S.F. CHRON. (June 22, 2017, 7:19 AM), https://bit.ly/3oKFkgB; see also Shi et al., supra note 256, at 5-7 (finding that dispensaries and retailers were more likely to locate in areas with higher proportion of racial and ethnic minority population and lower household income).

258. Legalization itself may affect use rates regardless of public consumption laws, but data is so far mixed on this point. Compare, e.g., Bin Yu et al., MarijuanaLegalization andHistoricalTrends in MarijuanaUse Among US Residents Aged 12-25: Results from the 1979-2016NationalSurvey on Drug Use andHealth, 20 BMC PUB. HEALTH 1, 1, 4-10 (2020) (finding a relationship between medical cannabis laws and prevalence trends but no statistically significant relationship between adult use cannabis laws and prevalence), with Rosanna Smart & Rosalie Liccardo Pacula, Early Evidence of the Impact of Cannabis Legalizationon CannabisUse, CannabisUse Disorder, andthe Use ofOther Substances: Findingsfrom State Policy Evaluations,45 AM. J. DRUG & ALCOHOL ABUSE 644, 647-50 (2019) (reviewing existing literature and concluding that medical cannabis laws are generally associated with negative or insignificant changes in prevalence among adolescents and increased cannabis use by adults over age 25, as well as concluding that adult use cannabis laws have produced mixed findings for youth use and insignificant effects for adults).

259. See U.S. ENV'T PROT. AGENCY, REDUCING EXPOSURE TO SECONDHAND SMOKE:

U.S. PROGRESS SINCE EPA'S 1993 LANDMARK REPORT (2018).

260. See James Tsai et al., Exposure to Secondhand Smoke Among Nonsmokers -

United States, 1988-2014, 67 MORBIDITY AND MORTALITY WKLY. REP. 1342, 1344-46 (2018).

261. See GillianL. Schauer et al., Self-ReportedExposureto, PerceptionsAbout, and AttitudesAbout PublicMarijuanaSmoking Among USAdults, 2018, 115 ADDICTION 1320, 1322-23 (2020).

262. See Dai & Richter, supra note 15, at 3. 263. See discussion supra Section II.E.3. 264. See StevenW. Bender, The Colorsof Cannabis:Race andMarijuana,50 U.C.

DAVIS L. REv. 689, 703 (2016).

265. See Chad Cotti et al., Alcohol-Impaired Motor Vehicle Crash Risk and the

Location ofAlcohol Purchase, 108 SOc. SCI. & MED. 201, 207-08 (2014).

266. A delivery-only retail model would further reduce cannabis outlets, but this

approach is beyond the scope of this Article.

267. See Press Release, Mintel Group, 28% of Younger MillennialsDrinkat Home (June 5, 2018),

to Go Out, MINTEL

Because It Takes Too Much Effort https://bit.ly/3ufzFA5.

268. See Alexandre Tanzi & Steve Matthews, AmericansAre BuyingMore Alcohol

to Drink at Home, BLOOMBERG (Apr. 3, 2020, 10:27 AM), https://bloom.bg/3fLiykj.

269. See Dirk J. Korf, Dutch Coffee Shops and Trends in Cannabis Use, 27

ADDICTIVE BEHAVS. 851, 853-54 (2002).

270. See Robert J. MacCoun, What Can We Learn from the Dutch Cannabis

Coffeeshop System?, 106 ADDICTION 1899, 1900-04, 1906-08 (2011).

271. For comparison, the price of alcohol at a bar is far higher than that at a liquor store or other outlet for off-site consumption. See, e.g., Will Gordon, Why the Hell Does Your DrinkCost So Much?, DEADSPIN (Nov. 1, 2013, 2:06 PM), https://bit.ly/3hR9qgT (noting mark-ups at a Boston bar ranging from 300% to 1,206% depending on the type of alcohol).

272. See Pot Tourists Can Smoke It Where They Buy It in San Francisco,Fox40

(Mar. 15, 2018, 12:13 PM), https://bit.ly/3hQkuuF.

273. See discussion supra Section I.A. 274. See, e.g., Annie Lowrey, Don't Blame Econ 101 for the Plight of Essential Workers, ATLANTIC (May 13, 2020), https://bit.ly/2VH05zd (arguing that factors including lack of unionization, firm consolidation, and the relationship between demographics and economic and political power explain the persistence of low wages and poor working conditions for many working in "essential" jobs better than simple conceptions of supply and demand).

275. See, e.g., Xisca Sureda et al., Secondhand Tobacco Smoke Exposure in Open and Semi-Open Settings: A Systematic Review, 121 ENV'T. HEALTH PERSPS. 766, 767-71 (2013).

276. See, e.g., Emily Brennan et al., Secondhand Smoke Drift: Examining the Influence ofIndoor Smoking Bans on Indoor and Outdoor Air Quality at Pubs and Bars, 12 NICOTINE & TOBACCO RES. 271, 273-75 (2010).

277. See State of Michigan, Dep't of Licensing & Regul. Affs., MarijuanaRegulatory Agency Adult-Use MarihuanaEstablishments, Mich. Emergency Rules, Rule 59(7)(a)-(c) (2019).

278. See ASHRAE PositionDocument on Environmental Tobacco Smoke, supra note 66; see also ANSIASHRAE Standard 62.1-2019: Ventilation for Acceptable Indoor Air Quality, supra note 67.

279. See Memorandum from Mass. Cannabis Control Comm'n to Comm'n Staff,

supra note 187.

280. See Cotti et al., supranote 265, at 207-08. 281. See generally supranotes 46-48, and accompanying text.

282. See Samantha Goodman et al., Prevalenceand Forms of CannabisUse in Legal

vs. IllegalRecreationalCannabisMarkets, 76 INT'L J. DRUG. POL'Y 1, 4 (2020).

283. See id. at 2. 284. See Michelle Rotermann, What Has ChangedSince CannabisWas Legalized?,

31 HEALTH REPS. 11, 11-14 (2020).

285. See id. at 12; see also Dana E. Lowry & Daniel J. Corsi, Trends andCorrelates of Cannabis Use in Canada:A Repeated Cross-SectionalAnalysis of National Surveys from 2004 to 2017, 8 CMAJOPEN E487, E488-91 (2020).

286. See Hannah Thibedeau, CannabisEdibles Available for Sale Legally in Mid-

December, CBC NEWS (June 14, 2019, 5:00 AM), https://bit.ly/3fNghlg.

287. See generally, e.g., Ming Wen et al., SpatialDisparitiesin the Distributionof

Parks and Green Spaces in the USA, 45 ANNALS BEHAV. MED. 18, 24-25 (2013).

288. With respect to housing, some accommodation maybe appropriate for medicinal

cannabis use by persons with disabilities or those lacking reasonable access to appropriate

outdoor spaces. This may include allowing limited indoor consumption, increasing the to medical needs, or facilitating availability of non-inhaled alternatives appropriate improved access to outdoor spaces, among other options. However, such accommodations need not be made for recreational use, as no protected interests are implicated. See generally Hudson B. Kingston, There is No ConstitutionalRight to Smoke or Toke, PUB. HEALTH L. CTR. 4, 6-16 (2019), https://bit.ly/3yz26fX.

289. See Stanton A. Glantz et al., Marijuana, Secondhand Smoke, and Social

Acceptability, 178 JAMA INTERNAL MED. 13, 13-14 (2018) (citing U.S. DEP'T HEALTH &

HUMAN SERVS., THE HEALTH CONSEQUENCES OF SMOKING-50 YEARS OF PROGRESS: A

REPORT OF THE SURGEON GENERAL 26-29 (2014 .

290. See U.S. ENV'T. PROT. AGENCY, supranote 259. 291. See Tsai et al., supranote 260, at 1344-46. 292. See Rotermann, supranote 284, at 11-14. 293. See supranotes 138-143. The fourth, Quebec, does not allow any public use. See Gouvernement du Quebec, The Legislation on Cannabis, https://bit.ly/34aqTsE (last updated Mar. 11, 2020).

294. See Hall & Lynskey, supranote 163, at 182-83 (assessing existing literature).

295. See supra Section II.F.2. 296. See Ravin v. Alaska, 537 P.2d 494, 510 (Alaska 1975) (approvingly noting the argument that "marijuana usually produces passivity and inactivity, in contrast to alcohol, which increases aggressiveness" in striking down the state's prohibition on cannabis possession for personal use in the home under state constitutional protection of the right to privacy).

297. See Navitha Jayakumar et al., Co-use and Mixing Tobacco With Cannabis

Among OntarioAdults, 23 NICOTINE & TOBACCO RES. 171, 173 (2019).

298. See Glantz et al., supranote 289, at 31. 299. See Map ofCurrentCigarette Use Among Adults, CTRS. FOR DISEASE CONTROL

& PREVENTION, https://bit.ly/3wffVOn (last updated Sep. 14, 2020).

300. See CaliforniaClean IndoorAir Law FrequentlyAskedQuestions (FAQ), CAL.

DEP'T OF PUB. HEALTH (Sept. 26, 2016), https://bit.ly/3477dWH.

301. See S.F., CAL., HEALTH CODE art. 19L § 19L.3(a) (2013). 302. See S.F., CAL., HEALTH CODE art. 191 § 1009.81(a) (2006). 303. See generally S.F., CAL., HEALTH CODE art. 19-19C, 19E-19F, 191, 19L. 304. See S.F., CAL., HEALTH CODE art. 19L § 1009.22(e) (2018).

305. Principles of Responsible Cannabis Use, NORML

(Feb.

3,

1996),

https://bit.ly/3yr3dOu.

306. See Adult-Use: Core Attributes of Adult Access Regulations, NORML,

https://bit.ly/3olasNt (last visited May 23, 2021).

307. See generally GLANTZ & BALBACH, supra note 45, at 1-18 (discussing early

tobacco control efforts relating to California's failed Proposition 5 in 1978).

308. See Press Release, Matthew L. Myers, President, Campaign for Tobacco-Free Kids, As U.S. Celebrates 25 Years of Smoke-Free Airlines, It's Time to Make All Workplaces and Public Places Smoke-Free (Feb. 23, 2015), https://bit.ly/3yzFNgo.

309. See Michael P. Eriksen & Rebecca L. Cerak, The Diffusion and ImpactofClean Indoor Air Laws, 29 ANN. REV. PUB. HEALTH 171, 174-76 (2008); see generally Preemption:The Biggest Challenge to Tobacco Control, PUB. HEALTH L. CTR.: TOBACCO CONTROL LEGAL CONSORTIUM (2014), https://bit.ly/2Tct5xn.

310. See, e.g., Brian A. King et al., Attitudes Toward Smoke-Free Workplaces, Restaurants, andBars, Casinos, and Clubs among US. Adults: Findingsfrom the 20092010 NationalAdult Tobacco Survey, 15 NICOTINE & TOBACCO RES. 1464, 1466 (2013) (reporting that 81.6% of U.S. adults believe workplaces should be smokefree, 74.9% saying the same for restaurants, and 50.0% saying the same for bars, casinos, and clubs).

311. For example, the WHO has argued for a precautionary approach, proposing that users of ENDS (electronic nicotine delivery systems) "should be legally requested not to use ENDS indoors, especially where smoking is banned" and that this should be the legal norm "until exhaled vapour is proven to be not harmful to bystanders and reasonable evidence exists that smoke-free policy enforcement is not undermined." Report by World Health Organization [WHO], Electronic Nicotine Delivery Systems, at 11, WHO Doc. FCTC/COP/6/10 (July 21, 2014). The WHO bases this approach on the premise that "the reasonable expectation of bystanders is not a diminished risk in comparison to exposure to

second-hand smoke but no risk increase from any product in the air they breathe[.]" Id. In contrast, some have questioned the effectiveness and ethical propriety of smoking restrictions due to potential stigmatization and argued that applying such restrictions to ENDS requires greater consideration. See Kristin Voight, Smoking Norms and the RegulationofE-Cigarettes, 105 AM. J. PUB. HEALTH 1967, 1968-70 (2015).

312. See Overall Tobacco Trends, AM. LUNG Ass'N, https://bit.ly/3oO3mqQ (last

visited Aug. 11, 2020).

313. See id. The proportion smoking fewer than 15 cigarettes per day increased from

32% to 66% over the same period. See id.

314. See Smokefree Policies Reduce Smoking, CDC: SMOKING & TOBACCO USE,

https://bit.ly/3yBkFQs (last visited Aug. 11, 2020).

315. See Glantz et al., supranote 289, at 13-14. 316. See, e.g., McDonald et al., supra note 227, at i100-01; Carla J. Berg et al., Perceived Harm, Addictiveness, and Social Acceptability of Tobacco Products and MarijuanaAmong Young Adults: Marijuana,Hookah, andElectronic Cigarettes Win, 50 SUBSTANCE USE & MISUSE 79, 86-89 (2015).

Figure 1: Risk Matrix

Private Property

Alcohol Model

Tobacco Model

Unrestricted

Frquency

. Enforcement

* Housmg

" Prevalence and

* Enforcement * Housing

Prevalence and " Prevalence and

r Prevalence and

Risks to Consumers

.7Housig

:

* Housig

-Enforcement

. Enforcement

Gener Expowsur

Unlikely or Limited

Others

Unlikely or Liuned

+ Home Exposure

p

Genal Exposure

* Woe

r xpre

. Impaired Driving s Public Disturbance a Public Disturbance s General Expos-ure

* Home Exposure

x

. lmpaired Driving

* Home Exposure

" Public Disturbance . Puble Dis.iirbancc

rHome

Exposre

. Worker Exposure

Worker Exposure

Unlikely or Limited

. Normalization * Outlet Density

________

" Normalization *Outlet Density

.Nrmalization

* Normalization * Otlet Desity

* Oudet Density Desiy_______Dnst

IV. AN ARGUMENT FOR OPEN PUBLIC USE IN LIMITED OUTDOOR

LOCATIONS

317. See, e.g., Childress et al., supra note 248, at 171-72. 318. See id. 319. See id. at 173.

320. See generally Kingston, supra note 288 (arguing that neither rights to free speech, privacy, nor due process encompass a general "right to smoke" either tobacco or cannabis that can overcome a government restriction rationally related to the government's interest in protecting public health). As former U.S. Surgeon General Dr. C. Everett Koop explained, "[T]he right of smokers to smoke ends where their behavior affects the health and well-being of others; furthermore, it is the smokers' responsibility to ensure that they do not expose nonsmokers to the potential harmful effects of tobacco smoke." The Health ConsequencesofInvoluntaryExposure to Tobacco Smoke, DEP'T OF HEALTH AND HUMAN SERVS., supra note 40, at 6.

321. Continuing inequalities in the arrest and prosecution of persons of color under remaining cannabis prohibitions in legalizing states (e.g., public consumption, youth possession) is a point raised not only by criminal justice reform advocates, but also by staunch legalization opponents. Compare Bender, supra note 264, at 690-703 (arguing from the position of advocating reform), with Kevin S. Sabet, MarijuanaandLegalization Impacts, 23 BERKELEY J. CRIM. L. 84, 92-93 (2018) (arguing from a position opposing legalization).

322. See, e.g., Cigarette Smoking and Tobacco Use Among People of Low Socioeconomic Status, CDC: SMOKING & TOBACCO USE, https://bit.ly/2QSeTJi (last updated Nov. 25, 2019); Tobacco Disparities:Evidence Supports Policy Change, PUB. HEALTH AND TOBACCO POL'Y CTR., https://bit.ly/34ecGL1 (last updated June 2020).

323. See sources cited supra note 322. 324. See, e.g., McDonald et al., supranote 227, at i99. 325. See OutbreakofLung Injury Associatedwith E-cigarette Use, or Vaping, CDC:

SMOKING & TOBACCO USE, https://bit.ly/3oPfgAS (last updated Feb. 25, 2020).

326. See, e.g., Daniel G. Orenstein & Stanton A. Glantz, PublicHealth Language for Recreational Cannabis Laws, UCSF CTR. FOR TOBACCO CONTROL RES. AND EDUC. 33 (2018), https://bit.ly/2T8YlgM.

327. See 2021 Conn. Acts 130 (Spec. Sess.) (amending CONN. GEN. STAT. ANN. § 7-

148(c)(7)(H)(xvi (West 2021).

328. See id. (adding authority over property "under the control of the municipality"); see also Mike Massaro, With CannabisLegalization Come RestrictionsForAll Smokers, NBCCONNECTICUT (June 22, 2021, 6:53 PM), https://bit.ly/2Sqwb0K.

329. See, e.g., McDonald et al., supranote 227, at i99. 330. Some products can accommodate both cannabis and tobacco products, as well.

See, e.g., id. at i99.

331. See id; see also Julie Jargon, Vaping Moves from the Bathroom to the Classroom, WALL ST. J. (Aug. 20, 2019), https://on.wsj.com/2QR7qdx. However, some visible emissions and often odor are still produced, so use is not easily confused with another product. Alcohol arguably presents some similar enforcement challenges due to ease of concealment. See, e.g., Justin Peters, The Slate Guide to Crime: How to Drinkin Public, SLATE (Mar. 12, 2013), https://bit.ly/2T73HJp. However, desire for furtive alcohol use is likely diminished by the wide availability of public locations where consumption is legal, such as bars.

332. To realize the equity benefits and other advantages of the tobacco control model, location restrictions must also be reasonable and enforced with appropriate attention to equity, proportionality, and other key considerations. These issues are part of a broader societal reckoning over law enforcement tactics beyond the scope of this Article, but smoking and vaping restrictions are among the many illustrations of how enforcement of minor rules can produce shocking results. See, e.g., David K. Li & Biance Britton, Viral Video Shows MarylandPolice Use Taser on Teen to Enforce Vaping Ban, NBC NEWS (June 15, 2021, 10:32 AM), https://nbcnews.to/3jeyBKO. 333. See, e.g., Am. Lung Ass'n, supra note 312.

334. See, e.g., McDonald et al., supranote 227, at i99-il01. 335. See id. at i99-il01. 336. See, e.g., Kerry Cork, supranote 226, at 8; Rachel A. Barry & Stanton A. Glantz, A PublicHealth Frameworkfor LegalizedRetail Marijuana Based on the US Experience: Avoiding a New TobaccoIndustry, PLOS MED., Sep. 2016, at 1, 4 (comparing laws in the first four adult use states against a public health standard based on tobacco control); Rosalie L. Pacula, et al., Developing Public Health Regulations for Marijuana: Lessons from Alcohol and Tobacco, 104 AM. J. PUB. HEALTH 1021, 1022-25 (2014) (presenting several reconmmendations based on tobacco control, as well as alcohol control); Regulating CommerciallyLegalizedMarijuanaas a PublicHealth Priority(PolicyNo. 201410), AM. PUB. HEALTH Ass'N (Nov. 18, 2014), https://bit.ly/3vl6FID.

337. The importance of this principle has sadly been highlighted by negative public response to U.S. public health authorities during the COVID-19 epidemic. See, e.g., Colleen Barry et al., Trustin Science andCOTID-19, JOHNS HOPKINs BLOOMBERG SCH. OF PUB. HEALTH (June 17, 2020), https://bit.ly/34g7uXh.

338. See, e.g., Glantz et al., supra note 289, at 13-14. 339. See Berg et al., supra note 316, at 86-89.

340. Seegenerally Electronic Cigarettes:An Overview ofKey Issues, CAMPAIGN FOR TOBACCO-FREE KIDS (Mar. 4, 2021), https://bit.ly/3fNtspM (explaining e-cigarettes, trends in youth use, and public health concerns, including flavorings).

341. See, e.g., Alex Seitz-Wald & Lauren Egan, Could Trump's Re-Election Go Up in a Puff of 'Smoke'? Vapers Say Watch Out, NBC NEWS (Nov. 10, 2019, 3:16 PM), https://nbcnews.to/3fgXjYG.

342. See U.S. 100% Smokefree Laws in Non-Hospitality Workplaces AND RestaurantsAND Bars, AMs. NONSMOKERS' RTS. FOUND., https://bit.ly/3uiY4EX (last updated Apr. 1, 2021).

343. See Sam Kamin, Legal Cannabisin the U.S.: Not Whether But How?, 50 U.C. DAVIS L. REV. 617, 630 (2016). Notably, the Rohrabacher-Farr Amendment restricts the Department of Justice from expending funds to interfere with state medical cannabis programs but does not apply to recreational cannabis programs. See Consolidated Appropriations Act of 2021, Pub. L. No. 116-260, § 531, 134 Stat. 1182, 1282-83 (2020); see also FederalPolicy, MARIJUANA POL'Y PROJECT, https://bit.ly/3AlsRFk (last updated May 28, 2021) (explaining the history of the Rohrabacher-Farr Amendment).

CONCLUSION