Words matter. Ideas matter. They always have. But, in the current context of continued protests across the US in support of the Black Lives Matter movement against ongoing racial discrimination and police brutality, companies are finally listening en masse and are beginning to take genuine action.
We’ve seen many food brands wake up and reconsider what they’re serving up at the breakfast table, and now tech companies appear to be taking a serious look at longstanding problematic, racially charged, and oppressive terminology in software and programming language.
Earlier this month, GitHub CEO Nat Friedman tweeted that the software development platform company is working to remove the term ‘master’ from its services and replacing it with a more neutral term like ‘main’. The ‘master’ in question has for decades been paired in the software—and wider technology industry—with the more overtly offensive and historically painful term ‘slave’ to describe a relationship between two devices, processes, components, etc.
GitHub is just one of the latest companies to make the move in righting this past terminological wrong. Python, one of the world’s top programming languages, changed ‘master’ to ‘workers’ or ‘helpers’ and ‘master process’ to ‘parent process’ back in 2018. In 2017, the Internet Systems Consortium decided to allow the words ‘primary’ and ‘secondary’ in lieu of the master/slave terminology. Three years prior, in 2014, Drupal made the switch to using ‘primary/replica’ and Django changed the word pairing to ‘leader/follower’.
The long march toward change
Questioning around the appropriateness of this racially insensitive terminology is nothing new. It stretches back much further in time to at least the turn of the twenty-first century.
In 2003, Los Angeles County’s Purchasing and Contract Services sent out a memo in response to a complaint about a “master/slave” label on a videotape machine, requesting that vendors “identify and remove/change any identification or labeling of equipment components that could be interpreted as discriminatory or offensive in nature.” Having received a slew of complaints about the memo, the services’ division manager clarified soon thereafter that it represented “nothing more than a request” and not an ultimatum or policy change. At the time, he further clarified that he did in fact receive positive responses from a number of companies willing to reexamine their labelling terminology.
Similarly, Microsoft, a long-time industry leader in geocultural and geopolitical awareness, began making strides to remove these terms as early as the late 1990s, according to sources. However, in an era where populist racism and white supremacy were not on center stage of American political life, the logic behind these changes was not always met with the same gravitas as appears to be the case at the present moment, and requests sometimes met with resistance by terminologists and engineers.
As an industry that came about and boomed so quickly, it is understandable that the software and information technology fields adopted such terms without thorough consideration of historical contexts or social consequences—after all, the contentious terms had been normalized for decades in other areas of technology. Nonetheless, societal attitudes toward oppressive language shift over time, and employees across all levels of the tech industry— an industry that has traditionally had a problem with white male monoculture—are increasingly considering the impact rather the original intent of these terms.
The bad is black effect
‘Master’ and ‘slave’ is not an isolated case—insensitive terminology related to race, gender, patriarchal structures and violence abounds in the software and information technology fields. A few examples of other terms with racial undertones are ‘whitelist/blacklist’ (allow/block [verb]; allowlist/blocklist [noun]) and ‘white hat/black hat’ (ethical/unethical [hacking]).
While these terms may not have a direct etymological link to race, they are based on the use of a pervasive Western metaphor connoting white as good and black as evil—to wit, the latter terms come from Hollywood cinematic symbolism, the heroes in old westerns usually wearing white hats, the villains black. This “bad is black effect” helps perpetuate subtle racism on, at the very least, a subconscious level, and is not reflective of the values and empathy that currently drive so many companies in the industry. It’s worth noting that black/white metaphors not based on this good/bad duality do not necessarily carry a racially charged value judgment and are therefore innocuous.
Be the change you wish to see in the world
If the obvious argument that it’s the just, righteous thing to do for the sake of inclusion, equality and our own humanity does not suffice to convince, there are practical, technical reasons to revisit these terminology choices.
Firstly, there are many alternative neutral terms that can be better used to describe these concepts, which makes for more all-around clarity and understandability—why stick with an offensive legacy term when there are so many better ways to say the same thing?
Secondly, in the words of scholar Heather Brodie Graves in Masters, slaves, and infant mortality: Language challenges for technical editing, “complex and subtle configurations of sexist, racist, or ethnocentric language use in technical documents can derail or interfere with readers’ ability and desire to comprehend and follow important information.”
Finally, as more organizations move toward new terminology the use of these terms will become increasingly less compatible across the industry.
Where to go from here
Reflect on terminology choices that may promote bias and hostility. What terms do you use that might be insensitive or rooted in metaphors of race, gender, gender-roles, societal structures, or violence? How can more inclusive terminology choices help engender a more diverse workplace and industry?
Nimdzi Insights is currently working with clients to help them manage problematic terminology across all languages and conducting ongoing research in this area. If you want to hear more or would like to contribute to the research effort, please contact us at [email protected].
