Stop Calling It Eco-Anxiety. Most of It Is Grief, and Grief Isn't a Disorder.

A sixteen-year-old in one of the studies I read this week described watching footage of bleached coral and feeling something she couldn't name — not fear exactly, more like arriving late to a funeral. Her school counselor's intake form had one box for it: anxiety. She checked it, because it was the only option, and then spent four sessions doing breathing exercises for a feeling that was never actually about the future.
That mismatch is the whole problem with how we talk about climate emotions in 2026. A meta-analysis out of Simon Fraser University, reviewing 48 studies for the Journal of Environmental Psychology, found that what gets bundled under "eco-anxiety" is really a spectrum with distinct psychological profiles: grief for ecosystems already lost, anger at institutional indifference, a specific kind of powerlessness researchers call eco-paralysis, and — less discussed but present in the same data — collective hope. Each of those has a different shape, a different cause, and a different thing that actually helps.
We kept the one label that pathologizes all of it as a malfunction to be managed. That was a choice, and it was the wrong one.
Anxiety implies a threat that hasn't happened yet. Grief means something already ended.
This distinction isn't academic hair-splitting. Clinically, anxiety is future-oriented — a threat-detection system firing about something that might occur. The standard response to anxiety is reassurance and regulation: breathe, ground yourself, the threat probably won't materialize the way you fear.
Grief is not future-oriented. Grief is about something that has already happened — a species, a coastline, a version of the world that existed and now doesn't. A 2026 Frontiers in Psychology paper on climate emotions documented young people explicitly describing "mourning for ecosystems," language that has nothing to do with anticipating a bad outcome and everything to do with metabolizing one that's already locked in. You cannot breathe your way out of grief. You can only go through it, the way you go through any other loss — which is precisely why the reassurance-based interventions built for anxiety keep failing this population. They're treating a funeral like a fire drill.
Anger shows up in the same research as its own distinct category, aimed specifically at institutional indifference rather than at the climate itself — anger that adults who had the power to act didn't. That's not dysregulation. That's an accurate read of the situation. Telling a teenager to "manage" appropriately-placed anger at a real institutional failure is asking them to distrust their own judgment, which is a worse outcome than the anger itself.
The clinical label changes what "getting better" is allowed to look like
Here's where the stakes get concrete. If you label a feeling "anxiety," the treatment goal becomes reducing the feeling — get the patient calmer, less activated, back to baseline. If you correctly label the same feeling "grief" or "moral anger," the treatment goal changes entirely: not reduction, but integration. Grief work aims at a person who can hold the loss and still function, not a person who stops feeling it. Anger work aims at a person who can channel the anger into something — organizing, advocacy, changed behavior — not a person who's talked out of being angry.
A youth counselor I'd trust on this would tell you the difference shows up immediately in the room. Kids handed anxiety-management tools for climate grief report feeling more alone afterward, not less — because the tool implicitly told them the feeling itself was the problem. Kids given permission to grieve, and shown that grief can coexist with continued engagement in the world, report something the research calls "constructive hope" — not naive optimism that everything will be fine, but a functional belief that their actions still matter despite the loss. That's a categorically different outcome than "calmer."
I've watched something similar play out in how workplaces mishandle grief more broadly — institutions default to whatever process is administratively convenient (three bereavement days, a breathing app) rather than the process that matches what the person is actually experiencing. Climate grief is getting the same treatment at a generational scale: convenient label, wrong intervention, applied to millions of teenagers simultaneously.
The kids protesting outside city hall are not the anxious ones
The uncomfortable implication is that the young people most visibly "activated" by climate change — protesting, organizing, showing up angry to city council meetings — are frequently the ones metabolizing their grief and anger correctly, not the ones failing to regulate. Meanwhile the quieter eco-paralysis cases, the ones who've gone flat and disengaged, are often the harder clinical picture, because paralysis is what happens when neither grief nor anger gets anywhere to go.
If your entire diagnostic framework treats loud engagement as the symptom and quiet disengagement as the healthy baseline, you will consistently intervene on the wrong kids for the wrong reasons. You'll medicate the organizer and miss the kid who's gone numb.
What actually helps looks less like therapy and more like a funeral
Every therapeutic approach that has shown real traction in this research shares a structure closer to mourning ritual than clinical treatment: naming the specific loss, sitting with people who share it, and converting some portion of the feeling into action that has nothing to do with making the feeling go away. None of that requires a diagnosis. It requires accurately identifying what a person is actually feeling before deciding what "better" should look like.
The sixteen-year-old with the coral reef footage didn't need four sessions of breathing exercises. She needed someone to tell her that grieving something you never got to see is still grief, and that the appropriate response to a real loss was never supposed to be calm.
What would it change if the next intake form had four boxes instead of one?