After a string of deaths, fires and near-fatal injuries in Massachusetts scrap yards, the industry is asking how to keep good people from drifting, and the answer may take sixty seconds at a time.

October 2026

At 11:58 on a Thursday morning, the 911 calls began reaching Freetown dispatchers from a recycling yard on Ridge Hill Road, in the village of Assonet. Someone had been struck by a front loader. Police and paramedics arrived to find the machine, and on the ground nearby, a man with significant injuries. He was David Torres, 52, of Fall River, an employee of Excel Recycling. He was taken to a local hospital and pronounced dead there.

By Friday, the Bristol County District Attorney’s Office had confirmed the outline: Freetown police and Massachusetts State Police detectives assigned to the district attorney were investigating, and the Occupational Safety and Health Administration had been notified and was assisting. Everything else about those final seconds, including where the loader was traveling, who could see whom, and what the ground and the pile and the noise of the yard were doing, belongs for now to the investigators. This article will not guess at it.

What it will do is refuse to treat Torres’s death as a dot. In the scrap yards of southeastern Massachusetts, it is increasingly hard to see it as anything but a point on a line.

A Line, Not a Dot

Nine days short of a year earlier, on Friday, October 10, 2025, Westport EMS answered what was dispatched as a medical call at Mid City Scrap, 548 State Road. It was 11:10 in the morning. Responders found Mark Peters, 49, of Taunton, with a crushing injury to his lower extremities. He was rushed to the hospital and pronounced dead shortly afterward. An OSHA safety and health officer was at the yard that same day. The district attorney said the death did not appear suspicious, a phrase that speaks to crime rather than to cause, and says nothing about prevention.

Peters was remembered by those who knew him as a loving family man. Within days, more than 175 people had contributed nearly $20,000 to a fundraiser to cover funeral costs and the family’s bills.

Seven months later, on Monday, May 4, 2026, at 3:34 in the afternoon, Westport police and firefighters were dispatched to the same address for a different kind of emergency. Crews arrived to heavy fire and smoke pouring from a large pile of processed vehicles. Route 6 was closed. Residents along Davis Road were told to shelter in place, shut their windows and turn off their air conditioners. One Mid City employee suffered non-life-threatening injuries, according to the town.

Scrap-pile fires do not behave like ordinary fires. They burn deep and they burn stubbornly, and the work of putting them out is the work of taking the pile apart. Firefighters stayed on scene for more than 48 hours, aided by mutual aid from 15 communities along with the state’s forest firefighting division and the Department of Fire Services. Heavy machinery was brought in to break the pile open and reach the hotspots buried inside it. The Board of Health advised owners of private wells within a mile and a half to have their water tested. The Massachusetts Department of Environmental Protection required the facility owner to carry out an Immediate Response Action under the state’s contingency plan, and a licensed site professional began assessing the site under state oversight. The State Police Fire Investigation Unit took the lead on the question of how it started. In what has been made public, that question has not been answered.

The region has been here before, and not only recently. On October 13, 2022, at about 3:30 in the afternoon, an experienced worker in his fifties was caught by a metal shredder at Scrap-It Inc. on Second Street in Everett. It took firefighters, Boston’s technical rescue unit and medical personnel nearly three hours to free him, with mutual aid from Somerville and Revere, while an IV line fed him medication through the wait. He was taken to the hospital with serious injuries. A firefighter was hurt in the rescue. OSHA responded to the scene.

Then there are the trucks. Tractor-trailers hauling scrap in open-topped tub trailers have tipped over on the region’s ramps and rotaries, and the same machinery has killed elsewhere: on February 27 of this year at a scrap dealer in Hollidaysburg, Pennsylvania, a dump trailer suffered a mechanical failure, rolled sideways and came down on a tractor’s cab, killing a 53-year-old employee. Scattered through the logs of the area’s yards are the quieter entries, the OSHA recordables that never made the evening news: the lacerations, the crushed hands, the strains, the amputations that were caught in time or weren’t.

No public record connects these events by a single cause. A loader, a shredder, a burning pile of shredded cars and a trailer on a curve are different machines with different physics. Investigators will find what they find: a mechanical fault here, a procedural gap there, a barrier that was missing or a rule that was bent. What they rarely write into findings, because it is not a finding, is the question now being asked in the offices of every regional EHS director, safety and compliance manager and HSE supervisor in the corridor from Boston to Bristol County:

How do I keep good people from drifting?

That question has a name. The name is complacency, and it is the subject of the rest of this article.

The Inspector at the Gate

Whatever else is uncertain, the regulatory weather is not. Under federal rules, an employer must report a work-related fatality to OSHA within eight hours, and an in-patient hospitalization, amputation or loss of an eye within 24. Every one of the incidents above sits in that reporting window, and OSHA has a long record of treating a cluster of serious events in one industry and one geography as a signal rather than as coincidence. Inspections follow fatalities, complaints, referrals and news coverage. They can also follow targeting by an area office that reads the same headlines everyone else does.

Prudent operators should plan on unannounced visits and plan on them soon. That is not a prediction about any particular agency decision; it is how a competent risk manager thinks when the news cycle has done the targeting for the regulator. The exposure is not abstract. A serious violation can cost more than $16,000, and a willful or repeat violation can exceed $165,000, per item, before any consideration of what a fatality does to insurance, contracts, bonding and the willingness of customers to be associated with a name that keeps appearing in the paper.

The standards that tend to come up in metal-recycling inspections will be familiar to anyone who works in the industry: lockout/tagout, machine guarding, fire prevention planning, emergency action plans, and the General Duty Clause that catches hazards no specific rule anticipated. But the experienced inspector’s first hour on a site is not spent reading the program. It is spent comparing the binder to the yard. Does the lockout procedure on the laminated card match what the crew does at the shredder at two in the afternoon? Do the pedestrian routes on the site map match where people actually walk? Can anyone on the crew tell the inspector the last time someone exercised stop-work authority, and what happened next?

This is where complacency stops being a soft topic. It is the gap between the written program and the lived one, and inspectors are trained to measure that gap. So are plaintiffs’ attorneys. So, in the worst hour, are the district attorney’s detectives.

The Anatomy of a Quiet Mind

Ask a room of workers whether they are complacent about safety and nobody raises a hand. They are not lying. Complacency is not an attitude people hold; it is a state the brain drifts into, and it feels from the inside like competence.

The brain is built to stop noticing

The human nervous system is a machine for filtering. Whatever is constant and survivable gets turned down, which is why you stop hearing the refrigerator and stop smelling your own house. A scrap yard is a catalogue of constant, survivable-until-it-isn’t stimuli: the reverse alarms, the clang of falling metal, the shuddering ground when the shear closes, the sight of twenty tons of machine moving ten feet from a person on foot. Within weeks, a new hire’s alarm bells go quiet. Within years, the entire environment is background.

Psychologists studying sustained attention have documented a related effect since the Second World War. In Norman Mackworth’s classic experiments with radar operators, the ability to detect rare, critical signals fell off measurably within the first half hour of monitoring. This “vigilance decrement” is not laziness. It is how attention works when most of the time nothing happens.

Nothing happened, again

The organizational theorist Karl Weick described safety as a “dynamic non-event”: the proof that it is working is that nothing occurs, and the more it works, the less visible it becomes. That invisibility is a perfect trap. Each uneventful shift teaches the same wrong lesson, that the hazard is smaller than the signs say.

The sociologist Diane Vaughan gave the deeper mechanism a name when she studied the Challenger disaster: normalization of deviance. A shortcut is taken once and nothing happens. The next time it is a little easier to take, and the time after that it is simply how the job is done. No one decides to accept more risk; the baseline just slides. In a scrap yard the shortcut might be stepping into a loader’s path because the operator “always looks,” reaching into a jam without waiting for the full stop, or letting a pile grow a little past the safe height because the crew is behind. Each of these saves forty seconds and costs nothing, right up until the day it costs everything.

The most corrosive part of the cycle is how near misses are interpreted. A near miss is a free lesson that the system nearly failed. On the floor, it is often heard as the opposite: proof that the system held. “I’ve done it a thousand times” is not a boast about skill; it is a statistical argument that sounds airtight until the thousand-and-first.

It won’t be me

Psychologist Neil Weinstein’s research on “unrealistic optimism” showed that people systematically rate their own odds of misfortune as lower than their peers’, and that the effect persists even when they are given the facts. Everyone on the yard knows what happened in Westport and Freetown. Almost everyone also believes, at some level beneath argument, that they are the careful one.

A related and more contested idea, Gerald Wilde’s risk homeostasis theory, holds that people tend to maintain a preferred level of risk, taking more when they feel better protected. The theory is debated, but anyone who has watched an operator speed up after the backup camera went in will recognize the instinct. Controls that make people feel safe can, if the culture is not paying attention, give back part of their benefit in behavior.

The expert’s trap

There is a temptation to imagine that complacency is a problem of the new and the careless. The opposite is often closer to the truth. Mastery is, neurologically, the process of moving tasks out of deliberate thought and into automatic routine. That is why a veteran loader operator can feed a shredder infeed with a rhythm that looks like choreography. It is also why the veteran may not consciously register the moment a pedestrian steps out of the blind spot. And tenure carries social weight: a newer worker may hesitate to tell a twenty-year hand to stop.

It is worth noticing, carefully, that the victims in the incidents above were 49, 52, and, in Everett, an experienced worker in his fifties. Three data points are not a study, and nothing in the public record says experience or inattention played any role in what happened. But they are a reminder that complacency is not a rookie’s disease. It is a condition of familiarity, and familiarity is the one thing every long-serving employee has in abundance.

The production clock

Complacency rarely arrives alone. It travels with schedule pressure, with the truck that has to be unloaded before the next one arrives, with the pile that has to come down before the weekend, with the quiet arithmetic every worker performs about what the boss really wants. Where speed is rewarded visibly and caution only verbally, workers learn to read the difference. The behavior that results looks like carelessness, but it is closer to a rational response to the incentives on offer. Any program built to fight complacency that ignores the production clock will be fighting the wrong enemy, and it will lose.

The safety professional’s own drift

For most people on a yard, safety is a background condition, noticed only when it fails. The exceptions are the people whose job titles contain the word. And even they drift. The safety manager who has delivered the same forklift talk forty times starts delivering it in the voice that people tune out. The inspection checklist gets completed from memory. The incident report template fills in the same four causal boxes. A discipline built to fight habit can itself harden into habit, and the first step in fighting complacency in the workforce is admitting it lives in the safety office, too.

Why the Annual Training Fails

The industry’s standard answer to human drift is training, and the standard form of training is an annual event: a conference room, a slide deck, a sign-in sheet. It satisfies the regulator. It does little for the brain.

In the 1880s, the German psychologist Hermann Ebbinghaus mapped what is now called the forgetting curve: newly learned material fades quickly without reinforcement, with the steepest losses in the first days. Later research on the “spacing effect” found the mirror image: the same material, presented in short sessions spread over time and retrieved from memory, sticks far better than the same material presented once, in a block. An eight-hour session in January is, for retention purposes, nearly the worst possible design. By March the slide deck is a faint memory. By the shift in October when it matters, it is gone.

There is a cultural cost as well. Annual training teaches, implicitly, that safety is a thing that happens in a room once a year and then stops. It signals that the real work, the work of the yard, is somewhere else. And the sign-in sheet, the artifact the system actually values, becomes the product. The binder is full. The behavior is unchanged.

The Sixty-Second Revolution

The alternative is not to train more. It is to train differently: in small, frequent, relevant doses that arrive at the cadence at which attention actually fades. The principle is simple. Frequency beats duration. Relevance beats completeness. And a peer’s voice beats the voice of authority.

Micro-training works because it respects how people really learn and really work. A sixty-second message fits inside a break, doesn’t overload working memory, and can be repeated until it sticks. A message that arrives every week keeps the topic alive in the part of the mind that decides what to notice. And a message in a coworker’s voice, filmed on this site, in these conditions, is processed as news from a trusted source rather than as a lecture.

What follows is not a menu from which a company should choose one dish. It is a set of mutually reinforcing habits, each easy on its own and powerful together.

1. The 60-second safety video

One idea, one hazard, one action. A loader operator walks viewers through a pre-shift circle check in under a minute. A mechanic shows the exact moment a lockout is verified with a try-start. The format discipline is the point: if it cannot be said in sixty seconds, it is two videos. Shoot on a phone, on the actual equipment, with the actual noise in the background. Polish reads as corporate; authenticity reads as truth. Post them where the crew already looks, whether that is the break room screen, a group chat, or the dispatch tablet, and retire none of them: a library of forty clips becomes a searchable reference and a new-hire curriculum at almost no extra cost.

2. The toolbox talk that talks back

The traditional toolbox talk fails in one predictable way: a supervisor reads a sheet to a group that is looking at its boots. Redesign it to run five minutes or less, built around a single question the crew must answer. “Where is the blind spot on the loader you’re running today?” “What is the first thing you check before you reach into a jam?” Rotate the leader weekly, so the most experienced voice in the room is sometimes the listener and the newest hire sometimes gets to ask. Record the best questions, not the attendance. The goal is a conversation that turns the sheet into a thought, and the thought into a habit.

3. The fifteen-second Reel

Humor is not the enemy of seriousness; it is the delivery vehicle. A fifteen-second clip that makes a crew laugh about the “guy who always forgets his hard hat” or dramatizes the grim comedy of a loader mirror nobody has wiped since spring is shared, replayed and quoted at the next break. Shared content spreads a safety idea farther than any poster. The rules of the genre are simple: laugh at the behavior, never at a person; keep the lesson in the last two seconds; and never stage anything that models the unsafe act as a gag without showing the right way. Done well, a Reel does the one thing a procedure can’t, which is get repeated aloud, voluntarily, by the people it is meant to protect.

4. The newsletter with faces in it

A monthly safety newsletter is only read if it is about the readers. Dedicate a standing section to the people on the yard: a mechanic who caught a hydraulic leak before it failed, a driver who refused a load that wasn’t properly secured, a new hire who asked the question nobody else would. Name them. Photograph them at work, not posed in front of a banner. Describe in two sentences what they did and why it mattered. Recognition is the cheapest and most underused tool in safety, and it does something specific: it tells every reader what the company actually values. Add a “close call of the month,” written without blame, and the newsletter becomes the institutional memory the yard otherwise lacks.

5. The serialized story

People remember stories long after they’ve forgotten rules. Build one. A recurring, fictional-but-true account of a worker’s life, assembled from real incidents with identifying details changed, released in installments on a platform such as Pinterest where visual, step-by-step storytelling is native. Call him Dev, or Marcus, or Luz. Let the audience see his morning, his commute, his kid’s school drawing taped inside his locker, his unexamined shortcut, the near miss that scares him, and the choice he makes next. The reader’s brain does what it does with every story: it puts itself inside the character. When the day goes wrong, it goes wrong for someone they’ve come to know. When it goes right, they’ve rehearsed, in imagination, the decision that mattered. Keep real victims’ identities out of it entirely; the story is to honor what happened, not to relitigate it.

6. The monthly focus, told by the crew

Pick one theme a month: blind spots, housekeeping, battery hazards, trailer loading, hand placement. Then hand the microphone to the workforce. Ask employees to film a short clip on “what I do” about that month’s topic: the check they never skip, the habit they learned the hard way, the rule they used to hate and now defend. Post the best ones, with permission, to the company’s channels. Three things happen at once. The people filmed take ownership of the message, because it is hard to ignore a rule you have just explained on camera. Their coworkers hear it from someone they respect. And the company, to its customers and its neighbors, shows what a living safety culture sounds like when it isn’t scripted.

A first ninety days

For the safety leader who wants to begin Monday, the sequence matters more than the scale. In the first month, listen: sit with each crew, ask what the last near miss was, and identify the three behaviors that most worry the people who do the work. Pick one monthly focus from that list, and film the first five sixty-second videos with volunteers from that crew. In the second month, convert toolbox talks to the question-driven format, launch the newsletter spotlight with a single story, and release the first installment of the serialized story. In the third month, publish the first crew-made “what I do” clips, count what was reported, and review the response times on every hazard raised. Then adjust. The aim is a rhythm that can be sustained for years, and that is why modest, repeated and visible beats ambitious and abandoned.

GROUND RULES FOR ANY PROGRAM THAT PUTS WORKERS ON CAMERA

• Written, voluntary consent for every person on screen, with the right to withdraw it.
• Never film or post anything that shows an unsafe act, even as a counterexample, without the correct method shown immediately after.
• Offer content in every language your crew speaks, and use captions; plenty of yards run loud enough that sound is optional anyway.
• Keep customer names, license plates and proprietary processes out of frame.
• Route every post through one reviewer, ideally with legal and HR input, who can pull content quickly if circumstances change, including after a serious incident.
• Never use footage of a real injury or fatality, and never use a coworker’s tragedy as content.

Aim the Content at the Hazard

Engagement is not the goal; behavior at the point of risk is. A calendar of clever content that never touches the yard’s actual killers is entertainment. The discipline is to map every piece of micro-content to a specific, high-consequence hazard and a specific, observable action.

Hazard	Micro-content that targets it
Mobile equipment and struck-by exposure	“Eye contact before you move” walk-arounds; a Reel on loader blind spots; a 60-second video on exclusion zones and spotters; a toolbox question: “Who can see you right now?”
Shredders, shears, balers and other machinery	Lockout/tagout “one lock, one person, one try-start” demonstrations; a crew-narrated clip on clearing a jam the right way; a story installment built around the “quick fix” that isn’t
Thermal events and pile fires	Short clips on lithium-ion batteries and propane cylinders found in loads; “find it before it finds the shredder” screening habits; a drill on what to do in the first ninety seconds of a smoldering pile
Truck and trailer operations	A walk-through of load distribution and securement; why you dump on level ground only; a Reel on ramp and curve speed with a loaded tub trailer; a monthly focus built by the drivers themselves
Pedestrian and housekeeping hazards	Newsletter spotlights on crews who keep walkways clear; a “before and after” photo of the month; a toolbox question on where you walk to the break room

Notice what these have in common. None is a lecture on the existence of danger. Each is an invitation to rehearse one specific behavior, in the voice of someone who does the work, at a moment when it can be remembered.

The Ledger Nobody Wants to Read

Safety is too often argued as a moral obligation, which it is, and then budgeted as an overhead item, which it should never be. The cost side of the ledger in an incident year is easy to itemize and impossible to ignore: the stop-work orders, the equipment taken out of service, the lost production, the premium increases, the legal fees, the settlements, the replacement hiring and training, the customers who quietly shift volume to a competitor with a cleaner record. The Westport fire alone consumed the resources of fifteen communities, closed a state highway, triggered an environmental response action and put neighbors’ private wells under a testing advisory. No line item on a safety budget comes close to that.

Against that, a micro-training program is almost embarrassingly cheap. The primary inputs are a smartphone, a few hours a month of a safety professional’s time, and a willingness by management to let the workforce speak. The return is harder to book, because the return is a day that did not happen. But a company that frames safety as a business-critical function rather than a compliance chore will recognize the logic: the cheapest incident to manage is the one the workforce prevented because they were still paying attention.

What Micro-Training Cannot Do

An honest article about complacency has to say what the proposed cure cannot do. Micro-training, toolbox talks and recognition programs are administrative controls. In the hierarchy of controls that every safety professional learns, they sit near the bottom, below elimination, substitution and engineering. They are useful precisely because they are cheap and constant. They are insufficient if used alone, and they are dangerous if used as an excuse.

A reel does not stop a loader. Proximity detection, 360-degree cameras, physical separation of pedestrian and equipment traffic, fixed barriers, machine guarding, interlocks, remote and automated clearing of jams, and continuous thermal monitoring of piles are engineering answers to hazards that human attention will, sooner or later, fail to cover. The right posture is not training or engineering. It is engineering first, with training that keeps human beings engaged with the controls that engineering provides, and alert for the moment those controls fail.

Measurement matters too, and it should be measured differently. Lagging indicators such as recordables and lost-time cases tell you what already happened. The measures that predict what’s coming are leading ones: the number of near misses reported, which should rise when a culture starts to trust itself; the speed at which a reported hazard is fixed; the share of toolbox talks in which a worker, not a supervisor, asked the best question; the frequency with which anyone uses stop-work authority, and what happened to them afterward.

That last point decides everything. The first time a worker stops a job and is thanked, the program becomes real. The first time a worker stops a job and is punished, with a sigh, a schedule change or a joke at his expense, every video on the company channel becomes a lie. Amy Edmondson’s research on psychological safety makes the underlying point precisely: people will not surface the small problem if the cost of speaking is higher than the cost of silence. A supervisor’s reaction to bad news is the most powerful safety communication a company sends.

The Day After Nothing Happens

On the morning of October 2, the day after Torres died, there were yards across southeastern Massachusetts where nothing happened. Loaders ran. Shears closed. Trucks backed to the scales. Crews who had heard the news in the truck on the way in went about their shifts a little quieter and a little more careful, and by Friday afternoon the carefulness had begun, imperceptibly, to fade. That is not a moral failing. It is the quiet physics of attention, and every safety professional in the region has watched it happen after every incident they have ever seen.

The work, then, is to build the thing that outlasts the fear: a company in which safety is a daily, shared, slightly irreverent, deeply human habit and not an annual event or a reaction to the last headline. It is a sixty-second video that someone on the crew made. A question at the start of a shift that nobody can answer on autopilot. A newsletter that puts a mechanic’s face on the page for catching what nobody else did. A story about a man named Dev that people remember better than they remember the rule.

None of this guarantees anything. The investigations in Freetown and Westport will finish, the citations will be issued or they won’t, and the region’s yards will go on handling some of the heaviest, hottest, most unforgiving material in American industry. The people who do that work are skilled, proud and, in the great majority, very good at not being hurt. That is exactly why the risk lies where it does: in the long, uneventful stretch when being good at it is mistaken for being safe from it.

Complacency, in the end, is what a good safety record feels like from the inside. It is the cost of success, and it is paid in the currency of a single day. The job of everyone with a title that includes the word safety, and of every owner who signs their checks, is to make sure that on that one ordinary day, the day when nothing is supposed to happen, someone is still looking.

Somewhere in the corridor between Boston and Bristol County, a loader is about to back up. Somebody will glance in the mirror. Whether that glance is a habit or a ritual or an afterthought is the whole story.

Reporting note: Incident details are drawn from public reporting, including the Bristol County District Attorney’s Office as relayed by Boston 25 News, NBC Boston, NBC10/WJAR, WPRI and WBSM (Freetown, October 1, 2026); the Town of Westport, NBC Boston, WJAR, The Herald News, The Enterprise and the Westport River Watershed Alliance (Mid City Scrap, October 2025 and May 2026); WHDH, WBZ/CBS Boston, Boston 25 News and The Boston Globe (Everett, October 2022); and Daily Voice (Hollidaysburg, Pa., February 2026). All investigations described were ongoing or had no publicly announced cause at the time of writing, and nothing in this article should be read as a finding of fault by any company or individual.