How Addiction handles disagreements over potentially harmful terminology

Abstract

A peer reviewer of a submitted paper on methadone maintenance states that she avoids the author’s term ‘opioid agonist therapy’ because new patients associate it with agony and become less willing to try the medication. She recommends ‘opioid substitution therapy’ instead. When the author uses this term in a revised submission, a different reviewer says that this term implicitly supports the canard that ‘methadone just substitutes one addiction for another’. Trying to moderate the dispute, an assistant editor proposes ‘medication-assisted treatment’, to which everyone agrees. When the paper is published, a reader writes in, angrily declaring that the author is putting lives at risks by ignoring the evidence that the main benefit of methadone comes from medication rather than it merely being an assist to the ‘real treatment’.

What are authors, editors and reviewers to do when people who are sincerely and laudably interested in avoiding language that harms vulnerable people do not agree on what is harmful and what is not? One approach is for journals to create an extensive listing of terms that will and will not be allowed to appear in papers, monographs and website content. Our journal does enforce a few language rules; for example, referring to urinalysis results indicating drug use as ‘positive’ rather than ‘dirty’ [1] and avoiding the term substance ‘abuse’. We would also, of course, not allow racially or ethnically derogatory language were it ever included in submitted papers, but in living memory it has not been.

However, after internal discussion the editorial team has decided not to attempt to generate a more lengthy list of forbidden terms because Addiction is a global, interdisciplinary journal whose readers and authors have diverse, competing opinions on what language is harmful and what is not. We instead follow four principles.

Some of the earliest activists who use drugs in the Netherlands labelled themselves the ‘junkie-bond’ [2]. Members of 12-Step groups often refer to themselves as ‘addicts’ or ‘alcoholics’ [3]. Some are horrified that people would choose to refer to themselves by terms they regard as stigmatizing, but in articles where someone is speaking about themselves in the first person we leave that decision up to the person concerned. Otherwise, we are in the position of putatively advocating for someone while simultaneously trying to deprive them of their right to speak and to define themselves.

Calling people what they want to be called is a sign of respect and a good practice to avoid harm. At the same time, the diversity within vulnerable populations on preferred labels should not be underestimated. For example, in the mental health field, some individuals want to be called ‘service users’, others endorse ‘patients’, others prefer ‘people who use services’ and still others plump for ‘consumer/survivors’ [4].

Similarly, that a term used to describe a population is embraced in academic circles does not necessarily establish that the term is also embraced among the population so described. In the United States, for example, many professors consider the term ‘Latinx’ a respectful and gender-sensitive replacement for the term Latino/Hispanic. But survey data show that only one-quarter of Latinos in the United States know the term, and seven out of eight of those do not use it [5].

Some claims that particular terms are harmful to vulnerable populations have empirical foundation [6], but for many other assertions that are made about certain terms being harmful there is little or no current evidence one way or the other. We respect the right of everyone to express opinions on proper language, but view opinions without evidence as just that.

When it has been empirically demonstrated that a particular term reduces harm, the best practice is to use it precisely as such. For example, there is some evidence that the term ‘addict’ elicits more hostile reflexive responses than does ‘person with substance use disorder’ [7]. This provides some empirical support for benefits of ‘person-first’ language, but we cannot assume that it extends to all applications. For example, even if the term ‘person with AIDS who uses drugs’ reduces stigma, it remains an empirical question whether PWAWUD produces the same benefit. Similarly, because the valence of terms can change over time, evidence that a term was or was not harmful in the past should not be assumed to apply for all time (e.g. the term ‘relapse’ was introduced decades ago as a less pejorative term for ‘return to use’ and is now sometimes criticized as more pejorative). Finally, we would note there are degrees and types of harm and the same term can affect different groups in different ways, meaning that evidence should inform judgements and discussions but cannot entirely supplant them.

Addiction has long had a commitment to illuminating the history of the field, best exemplified by its publication of historical studies and reviews of classic books. Terms that are seen as harmful today may be present in the text of old laws, scholarly studies, diagnostic manuals, newspapers, diaries, letters and other material that a historian analyzes. In such cases, we want the historical terminology to be presented accurately, not to condone it, but to meet the standard of scholarly integrity. In such cases, authors should make a clear delineation between quotations of prior historical terms and the terms that the submitting author is using.

Across the globe, across disciplines and across different life experiences, preferences related to terminology are remarkably diverse, making simple pronouncements about acceptable and unacceptable language potentially counter-productive. Addiction thus takes a de minimus approach on language rules and endorses instead the principles above to guide civil, mutually educative discussions. In doing so we are assuming that the best of the members of the journal’s family; namely, widely shared desires not to do harm nor to take offense easily. Maintaining such a community ethic requires some humility all round. It may be more emotionally satisfying to say ‘Term X is the only morally acceptable one’ than ‘In my particular country and in my particular worksite, people trained in my particular discipline prefer this term’, but the latter opens up opportunities for dialogue that the former does not. Relying upon the goodwill and thoughtful discussion of complex issues has been a hallmark of the journal for many years and our policies in the domain of potentially harmful language aim to continue that tradition.

Keith Humphreys: Writing—original draft (lead); writing—review and editing (lead). Rob Calder: Writing—original draft (supporting); writing—review and editing (supporting). John Marsden: Writing—original draft (supporting); writing—review and editing (supporting). Ed Day: Writing—original draft (supporting); writing—review and editing (supporting).

None.