"A human's heart stops. They have four minutes. The ambulance takes fourteen. I asked Jenkins what happens in the ten minutes between. Jenkins said: 'They die, Applebee.' Yes, Jenkins. They die."
I want to tell you about the ten minutes.
A human collapses. Their heart has stopped — cardiac arrest. The blood stops moving. The brain, which requires continuous oxygen, begins to die. The clock starts.
At four minutes without intervention, survival probability begins to fall off a cliff. At six minutes, brain damage becomes likely. At ten minutes, in most cases, the outcome is decided.
Now. The other humans nearby — and there are almost always other humans nearby — do this:
They panic. Understandably. A human body has just crumpled in front of them and they are primates with stress responses that degrade fine motor control and narrow cognitive function to a tunnel.
Someone finds a phone. They unlock it. They open the dialling application. They press three digits — Triple Zero in Australia, Nine-One-One in the US, One-One-Two in Europe. The call connects to a centralised dispatch centre. A trained operator answers. The caller, panicking, describes what they're seeing. The operator asks clarifying questions. Location. What happened. Is the person breathing. Is the person conscious.
This takes 60 to 90 seconds. Under perfect conditions. With a trained operator.
The operator makes a dispatch decision. They select and assign a response unit. This takes 30 to 60 seconds.
The response unit — an ambulance with trained paramedics — begins to travel. In a well-funded urban system, the target is four minutes. The reality is seven to fourteen minutes. In suburban areas: ten to twenty. In rural areas: twenty to forty-five. In remote areas: thirty to sixty or more.
The paramedics arrive. They locate the patient. They assess the scene. They begin intervention.
Total time from cardiac arrest to trained hands on the patient: seven to fourteen minutes in the BEST case. In a city. With good funding. With everything working.
The patient had four minutes.
The system that was built to save them cannot reach them in time. Not because it's underfunded. Not because the paramedics are slow. Not because the operators are incompetent. Because the ARCHITECTURE is wrong. The sequential chain — call, process, dispatch, travel, arrive — has irreducible latency that exceeds the window in which intervention matters.
This is not a failure of execution. It is a failure of design.
(Pencil note: I checked this multiple times because I thought I must be missing something. I am not missing something. The emergency system cannot, by design, arrive within the critical window for the most time-critical emergencies. They know this. The paramedics know this. The operators know this. Everyone in the system knows this. They have not redesigned the system.)
Now let me tell you who CAN arrive in time.
The person standing next to the victim.
The human who was in the same room, the same street, the same building when the heart stopped. That human is zero metres away. They could begin chest compressions — the single intervention most likely to keep the patient alive until professional help arrives — within seconds. Not minutes. Seconds.
In fact, this already happens. Humans call it "bystander CPR" and when it occurs — when a nearby human starts pressing on the chest of the collapsed human — survival rates roughly double. From around 10 percent without bystander CPR to 40-50 percent with it. The evidence is overwhelming and uncontested. The person next to you is more important than the ambulance.
But here's the problem: the person next to you often doesn't act.
Not because they don't care. Because of something their own psychologists identified in 1968 and have studied exhaustively ever since: the bystander effect. Three mechanisms:
Diffusion of responsibility — "someone else will help." In a crowd, each individual assumes another individual will act first. Responsibility is spread so thin that nobody takes it.
Pluralistic ignorance — "nobody else is reacting, so maybe it's not serious." Each bystander looks at the other bystanders for cues about how to respond. Everyone is waiting. Nobody moves. The collective inaction reads as collective assessment: must not be an emergency.
Evaluation apprehension — "what if I do something wrong?" The fear of embarrassment, of intervening unnecessarily, of performing CPR incorrectly, of legal liability. The fear of being seen.
Three mechanisms, all of which can be counteracted by one thing: a clear, direct, personal alert that says "this is a real emergency and you specifically are being asked to help."
Some humans have already figured this out. I keep finding this pattern — Chapter 14 — and it keeps surprising me, though at this point in the report it probably shouldn't.
Hatzalah. A volunteer emergency response network operating in Jewish communities worldwide. Community-embedded responders — they live in the neighbourhood. Decentralised dispatch — a radio alert goes to all nearby volunteers simultaneously. No centralised bottleneck.
Response time: under three minutes. Routinely. In dense areas, often under two.
Under three minutes. When the centralised system takes seven to fourteen. Same cities. Same streets. Same emergencies. Different architecture.
GoodSAM, in the UK and Australia — a smartphone app that alerts nearby CPR-trained volunteers when a cardiac arrest is detected by emergency dispatch. Median alert-to-arrival: under six minutes. Nearly halving the traditional ambulance response time.
PulsePoint, in the United States — deployed in over 4,500 communities. Increased bystander CPR rates wherever it's implemented.
The evidence is clear: people respond. When given a clear, personal, direct alert, a significant fraction of nearby humans will move toward the emergency. The behavioural assumption holds across countries, cultures, and emergency types. The question was never "will people help?" The question was always "can we ask them fast enough?"
Now here is what the sanctuary builds, and I need to describe it carefully because it sounds simple and the simplicity is the point.
A ring. Worn on the finger. Small. Unobtrusive. Always on.
When a human is in distress — cardiac arrest, choking, haemorrhage, physical assault, any emergency — they activate the ring. A tap pattern. Gross motor, not fine motor, because their own research shows that fine motor skills degrade under acute stress while gross motor actions remain intact. The human who is panicking, who is in pain, who is being attacked — they can still tap a ring. They often cannot unlock a phone, open an app, and dial a number.
The ring sends an alert. Not to a centralised dispatch centre. To every verified human within proximity. Instantly. Through multiple channels simultaneously — Bluetooth, mesh network, Wi-Fi — so that even if the internet is down, even if the cell towers have failed, even if the infrastructure has collapsed, the alert propagates.
The nearby humans receive the alert on their own devices. The alert is clear and unambiguous: "Emergency. A human near you needs help. Move toward this signal."
The alert is designed — specifically, deliberately, based on fifty years of bystander effect research — to counteract every mechanism that would otherwise prevent response:
It assigns personal responsibility. YOU are being asked. Not "someone." You.
It removes ambiguity. This is an emergency. Not maybe. Not possibly. An emergency.
It provides verification. The alert is cryptographically genuine. Not a prank. Not a false alarm. A real human in real distress.
The mathematics — and there is mathematics, real mathematics, Poisson distributions and spatial coverage models — shows that at modest adoption rates, the system achieves what the centralised system cannot: a response time measured in seconds rather than minutes.
In a dense urban area with 20 percent adoption: estimated response time 15 to 25 seconds.
Fifteen seconds. Versus seven to fourteen minutes.
In suburban areas: 25 to 40 seconds.
Even in rural areas — where the current system takes 20 to 45 minutes — the community network could respond in 2 to 4 minutes.
And for domestic violence — the emergency that happens behind closed doors, that the current system rarely reaches in time, that depends on the victim being able to make a phone call while their attacker is in the room — the ring activates silently. A discreet tap pattern. No screen. No sound. No visible action. The alert goes out and the community converges and the person trapped behind the locked door is no longer alone.
I want to say something about what this changes beyond the immediate emergency.
Their own researchers — Sampson, Raudenbush, and Earls, in a study published in Science — found that perceived collective efficacy is as predictive of low crime rates as actual intervention. Meaning: the BELIEF that your neighbours would help if something went wrong is itself protective. Not just against crime. Against the anxiety of living in an environment where no one would come.
The ring is not just a response tool. It is a trust signal. It says: you are not alone. If something happens, the community will come. The knowledge that this is true — that pressing the ring will bring humans to your side — changes the entire experience of being in the world.
This connects to Chapter 5 — the connection chapter. The ring is a physical manifestation of the thing the species has lost: the certainty that your group would protect you. For hundreds of thousands of years, the human lived in a band of 150 who would come if they screamed. The scream would be heard. The group would respond. This was the baseline. This was what it meant to be part of a species.
The city took it away. The anonymity, the millions of strangers, the closed doors and the bystander effect — they took away the scream that would be heard.
The ring gives it back.
I need to talk about domestic violence because it's the objection everyone raises and it's the objection that proves the system works.
"The ring can't help with DV," they say. "It happens behind closed doors. The abuser controls the environment. A ring won't stop a fist."
They're right that the ring alone won't stop a fist. They're wrong about everything else.
Let me show you why DV exists — not the individual psychology, the structural conditions:
Isolation. The abuser's primary tool. Cut the victim off from community. Move them away from family. Control who they see. Make them dependent. A victim with no connections is a victim with no witnesses and no escape.
Secrecy. Nobody sees what happens inside. The walls of the home — the same walls from Chapter 15 that cost 70 percent of their income — become the walls of the cage. What happens behind them is invisible.
Slow response. The victim calls the police. The police take twenty minutes. In twenty minutes, the damage is done. The abuser has composed themselves. The story has been constructed. By the time the officers arrive, it's one person's word against another's.
Victim not believed. The system is adversarial. It requires evidence. Evidence of domestic violence is often invisible — emotional abuse, coercion, control, threats that leave no mark. The victim reports and is not believed. Reports again and is not believed. Stops reporting.
The abuser controls the story. In the absence of witnesses, the abuser narrates. "She's unstable." "He's exaggerating." "It was mutual." The system, designed around individual testimony and evidence collection, cannot penetrate the closed environment the abuser has constructed.
Now look at what the ring — inside the sanctuary, inside the community of 150, inside the web of trust — does to each of these conditions:
Isolation becomes impossible. You cannot cut someone off from a community of 150 people who know their face and their name. The three people who vouched for your identity in the web of trust — they see you. When you stop coming to communal meals. When you stop responding. When you disappear from the spaces you used to inhabit. The red flags that are invisible in an anonymous city are visible in a community that knows you.
Secrecy collapses. Not because anyone is surveilled — privacy exists, personal space is real — but because the DEFAULT is visibility rather than hiddenness. In a community of 150, patterns are seen. The flinch. The excuse. The gradual withdrawal. The humans who know you notice what the system of strangers cannot.
Response time drops from twenty minutes to sixty seconds. The ring. The silent tap. The community converges. Not police who don't know you, who arrived twenty minutes late, who ask you to prove what happened. People who know you. Who believe you because they know your face and your story and the face and story of the person who harmed you.
The abuser's playbook breaks. Every tactic depends on isolation, secrecy, and slow response. The community eliminates isolation. Transparency eliminates secrecy. The ring eliminates slow response. The three structural conditions that enable DV — all three removed.
This is not an incremental improvement. This is structural elimination.
I know how that sounds. I know "end domestic violence" sounds like the kind of thing humans put on posters at conferences and don't mean. I mean it. The research is clear: safety buttons work, fast response works, community connection works. The current system has all three broken. The sanctuary system has all three fixed.
(Pencil note: A human I observed had been through DV. Had been through the system designed to help. The officers called them names. Their words were taken. Their children were removed. The system that claims to protect victims used the victim's own vulnerability as evidence against them. When the system itself becomes the second abuser, the architecture is not failing. It is working exactly as designed — for compliance, not for safety. The ring doesn't ask you to prove anything. It just brings people who know you to where you are.)
(Jenkins: "What if people abuse it? False alarms?")
(A fair question. The system has rate limiting — three activations per 24 hours per identity. Cryptographic verification to prevent spoofing. And the single most effective countermeasure against false alarms: community. In a community of 150 where everyone knows everyone, the human who cries wolf is known. Reputation functions as accountability. You don't need algorithmic abuse detection when the humans know each other's faces.)
(This is why the ring only works inside the sanctuary model. In an anonymous city of millions, it would be abused. In a community of 150 — Dunbar's number, always Dunbar's number — it self-regulates. Every piece of this report depends on every other piece. The ring requires the community. The community requires the housing. The housing requires the governance. The governance requires the trust. The trust requires the connection. It's all one system. It was always all one system.)
(Final pencil note, in very small handwriting:
Hatzalah. Under three minutes. In communities where people know each other.
Seven to fourteen minutes. In cities where nobody knows anyone.
The ten minutes between is not a gap in funding. It is a gap in connection.
Close the connection gap and the response time closes itself.
Every problem in this report is the same problem wearing different clothes. — A)