Tyson Yunkaporta, in *Sand Talk: how indigenous thinking can save the world*, offers a perspective to rethink what safety means: > ...there is no such thing as safety in Aboriginal worldviews. We have no word for it in our languages. Safety provided by an invisible hierarchy is complete anathema to our way of being. There is no agency in safety, which places a person in a passive role at the mercy of authorities who may or may not intervene when needed. So we have no word for safety or risk... however we have plenty of words for protection. > > Protection has two protocols: > The first is to look out for yourself. > The second is to look out for the people around you. > > This is such a wonderful way to live: knowing that you have the power to defend yourself and the ones you love while also being intensely aware that at any given moment there are dozens of people who are watching your back as you watch theirs." protection, as Yunkaporta describes it, requires agency. it requires mutual care and responsibility: watching your own back and the backs of others. safety, on the other hand, is passive. it relies on external systems or hierarchies, which often fail to show up when it really matters. this distinction made me think about how we can build conditions for protection in daytoday contexts, especially when it comes to navigating harm or conflict. one tool i’ve found useful for this is a simple conversational framework: oops, ouch, whoa. - oops: if you say something that lands wrong -- maybe it sounded fine in your head, but it didn’t come out right – you say “oops.” - ouch: if someone says something that stings or feels off, you say “ouch” to name the harm without escalating. - whoa: if the conversation is moving too fast, you’re confused, or you need clarification, you say “whoa” to pause and slow things down. i think of these as a kind of protocol of protection. it's a way of pausing between stimulus and response, and signalling at a meta-level that something in the way we're relating needs to be addressed. it also normalizes that mistakes will happen (oops), harm is repairable (ouch), and it’s okay to not have all the answers (woah).