Over the past two decades, a growing cadre of US states has legalized the drug Americans commonly refer to as “marijuana.” Although every state once prohibited the substance outright, more than 30 states now allow its use for medical purposes, and at least ten of those states allow use by adults for other purposes as well [1]. But even as more states legalize the drug, proponents of reform have begun to shun the term “marijuana.” As one lobbyist recently insisted, “We don’t use the ‘M’ word. It’s cannabis, cannabis, cannabis” [2], see also [3]. Indeed, the term “cannabis” is fast displacing “marijuana” in this policy space. For example, it appears in the names of recently formed interest groups such as the National Cannabis Industry Association, state-licensed businesses such as Green Man Cannabis, and trade publications like the Cannabis Business Executive.

Although “marijuana” and “cannabis” both refer to the same drug, reformers fear that the word “marijuana” has been indelibly marred by longstanding prohibitions that targeted the drug by that name. Until very recently, almost every state and the federal government referred to the drug as “marijuana” (or “marihuana”) rather than “cannabis” in their official legal codes [1]. As Gettman (2015) argues, “marijuana” was the term favored by prohibition supporters “as they sought to demonize its use and criminalize its consumers” [4]. Indeed, some claim that lawmakers originally adopted the foreign-sounding word “marijuana” “precisely because they wanted to underscore that it was a Latino, particularly Mexican ‘vice’” and thereby boost support among xenophobes for laws prohibiting the drug [5], [6], [7], [8], [9].

To its champions, the term “cannabis” carries no such baggage. “Cannabis” comes from the Latin word “cannabis” and the earlier Greek word “κάνναβις” [10] and is the scientific name given to the plant Cannabis Sativa and its variants from which the drug is produced. The origins of the word “marijuana” are murkier, but it likely derived from a Mexican-Spanish slang word or name [11], [12]. Thus, advocates hope that changing the name can boost public support for their reform proposals and for the people who consume and supply the drug [13], [14], [15], [16]. After all, if “marijuana” conjures up negative associations, then shedding that term for one that is ostensibly more neutral (like “cannabis”) should defuse opposition to their cause. The following call-to-arms from a prominent “cannabis” (née “marijuana”) distributor exemplifies the point:

… By changing the words we use to describe cannabis and herbal medicine, we can help our fellow citizens understand the truth about it, and see through the decades of propaganda.That understanding will convert cannabis opponents into supporters, and bring closer the day when all our prisoners go free, and nobody else is ever again arrested for using or possessing ‘marijuana’. [17]

Advocates are clearly making headway in changing the terms used in public discourse. Trends in the media, state legislation, and internet searches all serve to demonstrate the shift. In the national media, where coverage of the drug has skyrocketed in the wake of recent reforms, “marijuana” is still the more commonly used of the two terms. But since 2014, “cannabis” has been swiftly gaining ground, as shown in Fig 1. Furthermore, while nearly every state once employed the label “marijuana” (or “marihuana”) in its legal code (as noted above), several states that have recently legalized the drug have also redubbed it “cannabis,” as shown in Fig 2. In similar fashion, “cannabis” has also been catching up to “marijuana” as a search term in Google, as shown in Fig 3.

All of this suggests that framing—and particularly, the choice between the terms “marijuana” and “cannabis”—has become part of the political debate surrounding this drug. The framing literature suggests that the words advocates use to discuss public policies can influence how the public thinks about those policies, including the factors individuals will consider when formulating their opinions [18]. In this vein, prior work has demonstrated that even slight changes in how policies are framed or named can affect public support for those policies. For example, Smith documents vast differences in public support for “assistance to the poor” as opposed to “welfare,” finding the latter is more likely to conjure up negative associations with “waste and bureaucracy” (p. 75 in [19]). Similarly, Schuldt et al. find not only that partisan elites selectively employ the terminology of “global warming” versus “climate change,” but also that these frames exacerbate the partisan divide in the general public’s beliefs about whether the phenomenon is occurring [20]. But it is just as important to recognize that alterations in framing and naming do not always make a difference. For example, Rudolph finds no difference in the level of support for cuts to the inheritance tax, whether it is framed as an “estate tax” or a “death tax” [21]. Furthermore, even small changes in terminology (such as “climate change” versus “global warming”) can affect different groups in ways that balance out in the aggregate [22]. Our work adds to this literature by examining how changes in terminology that have been championed by policy advocates may affect public opinion on another very salient political issue.

Frames can also conjure different connections between policies and social groups. Social groups play a critical role in shaping public opinion [23], [24], [25]. In this policy domain, no less than others, advocates have sought to associate their own causes with particular social groups in ways that they believe will bolster their prospects with lawmakers and the public. As discussed earlier, many have accused prohibitionists of exploiting xenophobia by calling the drug by a foreign-sounding name (“marijuana” or “marihuana”). Some have even accused modern-day reformers of co-opting this strategy. While not focusing on any of the frames discussed here, Schlussel argues that reformers have intentionally framed campaigns around themes of white individualism to garner majority support for marijuana legalization [26].

Importantly, however, no one has yet demonstrated that changing the terms of discourse—i.e., calling the drug “cannabis” as opposed to “marijuana”—actually has any effect on public opinion. Broadly available survey data from prominent survey houses provide no way of systematically testing the effects of the different terms. A search of the Roper Center’s iPoll database (the most comprehensive database of US public opinion polling) returned 452 survey questions between 2003–2017 including “marijuana,” but not a single one including “cannabis”. At most, the existing data demonstrate some related framing effects when the drug is labeled as “marijuana”: support for legalization of “medical marijuana” outstrips support for legalization of “recreational marijuana” by a wide margin [27] and legalization of marijuana for “medicinal treatment” outstrips support for legalization for “recreational use” [28].

In this article, we employ a novel experimental survey design to test whether the choice of the term “marijuana” versus the term “cannabis” affects public opinion toward the drug and the policies governing it, both for medical use and for use more generally. While other slang terms for the drug exist, we choose to focus on “cannabis” and “marijuana” because none of these other terms has been similarly championed or vilified by policy advocates. Based on evolving public discourse, as well as the arguments advanced by advocates, we hypothesize that public attitudes toward the drug and liberalization of the policies governing it should be more negative when the drug is referred to as “marijuana” rather than “cannabis.” We call this the name frame hypothesis. These changes may be attributable to the different considerations that people bring to mind when prompted by the different names for the drug. For one thing, the terms “marijuana” and “cannabis” may elicit different stereotypes of the people who use the drug. It is also possible that “cannabis” is a less familiar term for the drug. Calling it “cannabis” may thus induce some confusion among respondents, in which case we would expect to see greater incidence of “Don’t know” responses to survey questions. We call this the unfamiliarity hypothesis. We apply these hypotheses to a broad set of dependent variables in this policy domain, including public opinion on legalization, evaluations of morality, tolerance of different activities involving the drug, perceptions of harm, and stereotypes of users.

After a discussion of Methods and Results, we close with a discussion of the ramifications of our findings for debates over policies toward this drug and for framing in policy discourse more generally. We also consider some possible alternative justifications for changing the name apart from marshalling public support for reforms. For example, one justification is that the term “marijuana” is offensive to some because it is associated with racist campaigns that pushed for prohibition [29].