When we talk to each other face to face, the words we choose are only one of an enormous toolkit of resources we have to communicate the nuances of our messages. When I call someone a piece of shit, you can tell how I really feel by everything else that goes with it. How loud am I being? Am I smiling? Am I shrugging? Am I talking fast or slow? Nonverbal cues (e.g., gesture, facial expression, eye gaze, posture) and paralinguistic cues (e.g., sighs, laughter, pitch, speed, volume, breathiness) make it really easy for the same exact words to communicate a million different things. If I say “he’s a piece of shit,” you can infer from all this other stuff whether I’m really upset, whether I’m expressing empathy but not investment, whether I’m being entirely sarcastic, whether I’m just having fun swearing…whatever. AND if you’re not sure what I mean, all it takes is a slightly confused expression from you and I can immediately clarify.

When you take the conversation online, we lose all that. When I write “he’s a piece of shit,” in my head it still comes with the million flavors of nuance it could have in conversation. When you read it, you get none of them. Everything comes off literal and straightforward. This is the problem that things like emoji and \s are attempting to solve, but nothing will ever really replace all the context of conversation.

I’m not saying it’s good or bad. Maybe we need to learn to use a wider range of “linguistic colors” to be more effective communicators online. And maybe there’s an element of cultural reproduction too: nobody starts out meaning to sound a extreme s they do but then the internet just starts to feel like an extreme place so we expect that that’s how we should talk in this context. I’m not sure about how we SHOULD talk online, but I do believe the cause of what you’re describing isn’t malicious, lazy, or otherwise ill-intended. I think it’s just things lost in translation.