A part of me died when I first heard the word “unalive” on social media about a year ago. The influencer in the video was telling a story about a conflict where someone died. Thanks to the context the video provided, it didn’t take long for me to realize that the word this person was really saying was “dead.” Many more instances of this substitutive diction have become common on the internet, like substituting the word “pedophiles” for “PDF files” or “suicide” for “sewer slide.” Insane to think about, right?

I have learned this is an example of “algospeak,” a pragmatic approach social media users take to avoid being banned for mentioning words platforms deem taboo by deliberately censoring themselves. Adam Aleksic, linguist and content creator, coined the term in his book of the same name, in which he suggests that social media is enabling younger generations to revolutionize the way we speak English. While I agree that we are witnessing constant changes in our diction, we must also acknowledge that such self-censorship comes at a price. In other words, we need to challenge the algorithm’s limits and say things for what they really are.

Using substitutive words for the sake of bypassing the algorithm only complicates discussion of subjects that are already difficult to approach. For instance, First Light, a non-profit organization that supports survivors of sexual assault, had their account suspended by Meta for using sensitive language in their posts. They released a statement acknowledging the trend of algospeaking, stating that “this censorship is making a subject that has been traditionally deemed inappropriate to discuss even more difficult to discuss on what should be a widely reaching, easily accessible platform.” If a non-profit for sexual assault survivors uses words like “grape” to suggest sexual assault like “rape,” that might raise some eyebrows if they don’t understand why such self-censorship is being exercised.

We have a higher chance of misunderstanding sexual assault victims by being unaware of algospeaking terminology. Posters use the word “mascara,” for example, to mention a partner or penis. In 2023, Actress Julia Fox was involved in a controversial misunderstanding on TikTok, not realizing that “mascara” meant “penis” in algospeaking terms when she commented that she didn’t “feel bad” for TikTok user Conor Whipple’s post of him opening up about his experience with sexual assault. Without appropriate context or prior knowledge, it can be a slippery slope when deciding how to go about responding to algospeaking terms.

Those who aren’t aware of these self-censoring practices wouldn’t know what I meant if I said “grape,” rendering any following information about such important topics equally illegible. As someone who’s aspiring to go into education, I often stumble across Instagram reels from teachers and administrators. In one reel, a user gave suggestions on what to do if a student watched “corn” in class, recalling a prior scenario they experienced. I was confused at first, because my mind automatically went to the rock band, Korn. It wasn’t until they mentioned their meeting with an administrator that I realized they meant “porn.” Some people who work in education might have understood their message based on context, but not me.

When we refuse to say what we really mean, we run the risk of downplaying the seriousness of sensitive topics. Take the topic of self-harm, for instance. In terms of “algospeaking,” users refer to “suicide” as “sewer slide.” In a more specific context, they say someone would “kermit sewer slide.” How seriously would you take such a sensitive subject at face value if those were the words you’d hear it described in? Sometimes heavy topics require us to face the music and confront such subject matter with language we won’t always be comfortable with.

It is pertinent to acknowledge that some of these words can be triggering for those who have had traumatic experiences, which would call for some instances of censorship. It is necessary to consider vulnerable audiences who are curious to learn more about certain subjects that are tough for them to confront directly. Even in these instances, I think algospeaking especially pushes its boundaries. It should be acceptable to use other words like “assault” or the acronym “SA” when mentioning sexual assault without defaulting to terms like “grape.” This practice especially becomes useful when providing trigger warnings. If, for example, someone sees a trigger warning mentioning terms like “assault” or “violence,” we’ll have better context of the topic and will be more inclined not to engage with the content.

While self-censoring may be useful to avoid facing consequences on social media, such as account bans or platforms removing posts, I have learned that it is much more nuanced than that. It is already hard enough to bypass the algorithm in general without some creative strategies, like using memes. Once algospeaking terms become more widely known, I’m curious to see how long it would take for platforms to catch on and flag them as well. What would we say then if the implications of terms like “grape” catch up to “rape?” All in all, there is no guarantee that algospeaking will keep you safe from the algorithm now or in the future.

Our voices are diminishing day by day on social media due to policies that dissuade us from saying how we truly feel. I am aware that what I am asking for is a tall order. Bypassing algorithms in the digital age has become virtually impossible unless we censor ourselves and exercise code-switching. I am up to the task of using other means of expression if platforms like Reddit or Instagram silence the language I intended to use.

In order to “unalive” algospeaking, we must first defy the algorithms that have sparked this linguistic revolution to begin with. At the end of the day, it goes without saying: Just say it how it is.