On June 25 I sent a letter to the Secretary of Health and Human Services. It had nothing to do with lettuce. It was about two outbreaks of infant botulism traced to powdered formula — forty-eight babies in seventeen states in the first one, every one of them hospitalized, and four more infants in the second. The Secretary had convened the formula chief executives around a table twice. As far as the public record shows, not one parent of a hospitalized baby had been in a room with him. Before your next roundtable with industry, I wrote, give the families one.

Seven weeks have gone by. As far as I can tell, it has not happened.

I am asking again, and this time about a parasite. CDC counts 13,895 laboratory-confirmed cases of Cyclospora acquired in this country since May, with at least another ten thousand still under investigation. Add up what the state health departments have published and the floor is above twenty-eight thousand. At least 688 people have been hospitalized. Two are dead in Michigan.

It is easy to hear this as a request for sympathy and it is not one. It is an argument about information. Duration and consequence come out of a case count before it ever reaches a desk of a Secretary of Health and Human Services, and a unit of one is silent about whether it meant a bad weekend or eleven weeks. An hour in a room puts back exactly what the briefing binder takes out.

My best example is a man I sued. Dave Theno was brought into Jack in the Box in 1993 to fix what had killed four children, and he stayed sixteen years. I deposed him and I did not go easy on him. Somewhere in all of that he got to know the family of Lauren Beth Rudolph, who was six and who was the first of the four to die, and for the rest of his life he carried her photograph in his wallet. He would take it out in meetings when he was explaining why he wanted something expensive done. Lauren is who we are protecting, he would say. Lauren, and the children she represents, is who I report to.

Dave changed that company and then changed an industry, and he did not do it because of a regulation or a verdict or a line in a report. He did it because he knew a child’s face. He drowned off Lanai in 2017, swimming with his grandson. I spoke at his memorial, which is not something a plaintiffs’ lawyer expects to be asked to do for a man he cross-examined, and I have missed him every year since. What I would give to have him in one of these rooms now.

In one week in 2009. I sat with a family in South Carolina whose daughter was in kidney failure from E. coli O157:H7 she got from cookie dough. She lived. Then I got on a plane to Ohio.

John Strike was a veteran and a grandfather. On April 11, 2009 he ate a cheeseburger at the VFW post in North Olmsted and three days later he was bleeding internally and headed for kidney failure and most of a month in a hospital bed. What he did not know, and what nobody could have told him, was that he had already passed it to his granddaughter when she visited him in the hospital. Abby Fenstermaker was seven. She went into the hospital on May 11, and her parents took her off life support on May 17. She was their only daughter, and my clients were her parents and the grandfather who was still recovering from the hamburger that killed her.

The Food Safety and Inspection Service announced the recall of ninety-five thousand eight hundred ninety-eight pounds of Valley Meats ground beef on May 21, 2009. Abby had been dead four days.

I have never gotten those two rooms out of my head, and I have never once looked at a case count the same way since. That is the entire reason I am writing this.

I am not the only person who has been through that hour, and the best account of it I have read this year was written by somebody not agreeing with me. In May, Kathleen Gillin published a response to one of my pieces. Twenty years ago, her nine-year-old daughter Rylee nearly died from E. coli O157:H7 in spinach and still lives with what it did. Kathleen now serves as the public member of the California Leafy Greens Marketing Agreement board. A few years after that agreement was formed, its leadership invited families harmed by foodborne illness out to California to meet growers, walk the fields and see the practices. Her words for what happened next were that it changed her perspective. She met the people doing the work — field crews, packing crews, auditors, family farm owners — and came away seeing an industry that had been altered by a tragedy it understood in human terms.

She and I do not agree about the leafy greens record. She thinks I give the industry too little credit for twenty years of reform, and she said so plainly. However, we agree completely about the hour.

Notice the direction that hour ran. The industry brought the families to the growers. Two decades later, nobody has brought the families to the government.

Nobody should write a rule because of one family. The hour is not a substitute for the surveillance data; it is a correction to a known distortion in it, applied to the person who has to decide what the data means and how urgently to act on it. Read the numbers, then meet one of them.

This is exactly what plaintiffs’ lawyers do for a living. It is. I have been putting families in front of juries since 1993, and I am not going to pretend I discovered the effect by accident. That is precisely how I know what an hour does.

And if officials should meet the sick, should they also meet the grower who plowed a crop under this summer, and the farmworker who lost the season with it? Yes. All. I have written about those losses and I meant it. An hour with a family in Michigan and an hour on a ranch in Salinas would leave a person better equipped than either one alone.

What I am asking for is smaller than it sounds, and it is worth saying exactly what it is not. It is not a press availability. It is not a family asked to testify, or to perform, or to be photographed on a stage next to somebody who needs the picture. Private. No cameras. No agenda beyond listening. Stop Foodborne Illness has been doing this work for decades and could arrange it inside a week; the walls of their offices carry photographs of the children who did not survive, which tells you they do not need to be taught how to handle it.

The ask is three hours, total, from three people. An hour for the acting Commissioner of Food and Drugs, who told the country last week that it should feel confident about produce. An hour for the new Director of CDC, sworn in six days ago into the largest Cyclospora outbreak this country has ever recorded. An hour for the Secretary, who has now been asked twice.

I have published fifteen things this government could do about this parasite and four I would do first. Every one of them costs money, or rulemaking, or an act of Congress, or all three. This meeting costs a calendar entry. It is the only item I have ever put on a list that requires nothing from anyone except the willingness to be in a room and hear something you would rather not hear.

It is hard. That is the honest objection nobody says out loud, and it is the real one. It is hard to sit across from a mother and have nothing to offer her but attention. I have done it too often, and it has never once gotten easier.

Do it anyway. It is an hour, and it changes what every number after it means.

Lobbyists – President Ulysses S. Grant is often credited with popularizing or coining the term in the 1860s while smoking and greeting people in the lobby of the Willard Hotel in Washington, D.C., though printed records of the word actually predate his presidency.

Jessica Glenza read the lobbying disclosures and named four organizations that had lobbied on the Food Traceability Rule. I spent yesterday reading further into the same database. What follows is what the filings say, with a link to every one of them, and one finding that has nothing to do with anybody’s money.

Start with the cleanest sentence in the record. The International Foodservice Distributors Association, reporting on the last quarter of 2025 and signed by its chief executive Mark S. Allen Jr. on January 20 of this year, lists its work on the Food Traceability Rule and then describes that work, in its own words, as support for the 30-month compliance date extension and targeted flexibilities. No euphemism, no talk of collaboration. The three people who did it are named on the page: Mala Parker, Kristin Spiridon and Joseph Kalmin.

The National Restaurant Association filed on the same day for the same quarter, signed by Sean Kennedy, its executive vice president of public affairs. Its food page lists the fiscal 2026 agriculture and FDA appropriations bill, H.R. 4121 and S. 2256, for provisions regarding FDA traceability. It then lists H.R. 5371, the fiscal 2026 continuing appropriations, for provisions regarding FDA traceability. It lists the Food Traceability Enhancement Act as draft legislation. H.R. 5371 is the bill that ended the November shutdown, and it carried Section 780, the provision barring the FDA from spending a dollar to administer or enforce the rule before July 20, 2028. The association named the vehicle by number in the quarter it passed. The lobbyists on that page are Matt Walker, Dan Roehl, Sean Kennedy, Laura Abshire and Aaron Frazier.

One aside, because the confusion is everywhere. The delay did not ride on the Big Beautiful Bill. The same restaurant association filing lists the One Big Beautiful Bill Act, Public Law 119-21, on its tax page and H.R. 5371 on its food page. Two different bills, four months apart. The funding prohibition is in Public Law 119-37, the shutdown deal.

Now go back two years. The National Retail Federation’s report for the first quarter of 2024, signed by general counsel Stephanie Martz, says the federation requested language in the fiscal 2024 agriculture appropriations bill to require the FDA to work collaboratively with industry to address concerns with the traceability rule, and that it sought support for H.R. 7563, the Food Traceability Enhancement Act. One lobbyist is named on that page: Scott Vinson.

H.R. 7563 deserves its own paragraph. Introduced on March 6, 2024, it would have amended Section 204(d) of the Food Safety Modernization Act to relieve retail food establishments, restaurants and warehouses of the duty to keep traceability lot codes and pass them down the chain and would have required the FDA to run a series of pilot projects before enforcing any compliance date. In an interview published by the Institute of Food Technologists, a former FDA official who oversaw the agency’s outbreak response said that stripping the lot code requirement essentially guts the rule, because true public health tracebacks stop being possible without it. The Safe Food Coalition opposed the bill in 2024 on exactly the ground it wrote to congressional leadership on this month. The bill never got a vote and died with the 118th Congress.

Here is the part that requires no inference about money. H.R. 7563 was introduced by Scott Franklin of Florida and cosponsored by Sanford Bishop of Georgia and Jimmy Panetta of California. Franklin sat on the House Appropriations Subcommittee on Agriculture, Rural Development, Food and Drug Administration and Related Agencies in the 118th Congress and is its vice chair in the 119th. Bishop is that subcommittee’s ranking member in both and was its chairman in the 117th. Andy Harris of Maryland chairs it. The FDA sits squarely in its jurisdiction. The sponsor and the lead cosponsor of the bill to relax the traceability rule were both sitting in the room where the FDA’s money gets written.

A word on where these men answer to. Franklin represents Florida’s 18th, the Florida Heartland, and lives in Lakeland, which the current map places inside his district. Publix Super Markets is headquartered there, and Publix is one of the four organizations the Guardian found lobbying on the rule. Bishop represents Georgia’s 2nd, the largest district in the state and the one that grows more peanuts than any other in the country. It takes in Blakely, the seat of Early County, where the Peanut Corporation of America ran the plant behind the outbreak that killed nine people, sickened hundreds and sent the company’s chief executive to federal prison. Panetta represents California’s 19th, which runs from south San Jose down the Monterey County coast. 

The bill died in Energy and Commerce. Its provisions did not. In June 2024 the Center for Science in the Public Interest reported that language from H.R. 7563 had surfaced in the House subcommittee’s fiscal 2025 bill. It surfaced again in the fiscal 2026 House bill, H.R. 4121. In November it became law as Section 780. Three bills across two appropriations cycles, carrying the same ask through the one committee where it needed no hearing, no floor debate and no vote of its own.

The Senate did not go along. In Senate Report 119-37, accompanying its own version of the agriculture and FDA bill, the Senate Appropriations Committee directed the opposite of what the House wanted: that the FDA may not use funds to delay enforcement beyond July 20, 2028, and that full compliance and enforcement begin by then. The Congressional Research Service notes that this language carries the same weight as the joint explanatory statement. The Senate subcommittee is chaired by John Hoeven of North Dakota, with Jeanne Shaheen of New Hampshire as ranking member, and its bill came out of committee twenty-seven to nothing. Section 780 is a House product. Among the senators on that subcommittee is Jon Ossoff of Georgia, who has been writing letters about this outbreak all summer. He is well placed to ask why his own committee’s language did not survive.

Section 780 did more than defund. It also told the FDA to work with industry on lot-level approaches, and on February 19 of this year the agency announced that the resulting stakeholder engagements would be organized by the Partnership for Food Traceability, an industry membership nonprofit founded in 2024. A client alert from Covington & Burling records the detail the agency notice leaves out: the first session, held on March 6, was open only to the partnership’s members, with later sessions open to the public. The partnership says on its own site that it will not advocate specific policy positions with the FDA as part of this process, and I have no reason to doubt that. The structure is the story. Congress ordered a federal agency to consult industry about a rule industry had asked it to postpone, the consultation is convened by industry, and the first meeting was closed to everyone else.

The industry conversation that produced the delay argument is on the record too. The Reagan-Udall Foundation ran three roundtables for the FDA in 2024 with more than two dozen participants. The Guardian named Chipotle, the Kellogg Company, Kwik Trip, the National Restaurant Association, the National Grocers Association and McDonald’s. Food Dive’s coverage of the same report adds General Mills, FMI, Subway, the International Fresh Produce Association and United Natural Foods. The published summary has industry describing significant increases in labor, equipment and space, with the costs that come with them.

None of which means all of industry asked for this. Part of industry asked for this. Kroger told its suppliers to be ready ahead of the deadline. De Ann Davis of the Western Growers Association told the Guardian that growers have been collecting this shipment data voluntarily for more than a decade and that her organization supports the rule. FMI gave the Guardian a serious answer, and it deserves to be stated rather than waved at: the rule is extraordinarily complex, it demands unprecedented coordination among growers, manufacturers, distributors, wholesalers, retailers and foodservice operators, and its member companies have four years of work invested in it. That is not frivolous. Traceability only functions when everyone in a chain is doing it, and companies that were genuinely ready did find themselves waiting on partners who were not. The associations that lobbied for delay represent the ready and the unready alike. They spoke for the unready or unwilling.

A sequence of dates to finish on. The FDA held its public meeting on lot-level traceability on June 15 of this year, with written comments due July 15. The CDC now puts the first illness in this outbreak on June 14. The meeting on how to make the rule easier to live with convened the day after the first person got sick, and the comment period closed the day before the FDA announced it was investigating Cyclospora in lettuce. Nobody in that room knew. That is the whole point of a traceability rule, and the reason fifteen and a half years of delay is not a scheduling matter.

Traceability draws a line. It says what everything else is. Without the line, fear prices the entire category and the whole category pays, which is why the associations that asked for the delay were arguing against their own members’ interests as much as against anybody’s health. An epidemiologist in Michigan asked three times where the lettuce came from. The records that would have answered her exist, in some form, in warehouses and distribution centers across the country. Congress has told the FDA it may not spend a dollar to go get them until July of 2028.

That can be undone, and two sets of people are placed to undo it. The first is the Senate. Its own appropriators wrote language directing that enforcement begin by July 2028 and not a day later, and that language did not survive. Senator Hoeven, Senator Shaheen and Senator Ossoff, whose letters this summer are already part of the record, are entitled to an answer about how the opposite provision came out of conference, and so is everyone who got sick. Nobody has asked in public. The second is the part of the industry that never wanted the delay. Western Growers says its members have been collecting this data voluntarily for more than a decade. Kroger told its suppliers to be ready. If the companies and associations that are already compliant asked Congress to unwind Section 780, the argument that industry cannot manage the rule would collapse, because it would be industry saying otherwise. Eleven consumer, public health and labor organizations have already asked. They should not be asking by themselves.

Three reporting teams have now been through this outbreak’s paperwork, and between them they have documented three separate delays with three separate causes. Daniela Sirtori at Bloomberg News filed a public records request in Michigan and published on August 3. Jessica Nix and Kristina Peterson, also at Bloomberg, reconstructed the week of the recall and published on July 21. Christina Jewett and Julie Creswell at the New York Times obtained the same Michigan correspondence under the state’s open records law and published Saturday morning. Put them end to end and you have a calendar.

The first delay: twelve days to get a supplier list.

On July 1, Michigan’s rapid response team asked Yum Brands for Taco Bell’s supply chain records. On July 2, the state health department held a call with the company; in a follow-up email, officials asked to be connected with Yum’s supply chain quality assurance team and its communications team. On July 3, the state asked for the records again. On July 6, the rapid response team asked a third time — Lauren Edwards, an epidemiologist on that team, wrote that given the pace of the outbreak and the number of people already sick, the information was needed in time to preserve public health. Later that day, some records arrived. On July 8, FDA asked for more. On July 13, the agency had everything it needed. On July 15 it named a Taylor Farms plant in Guanajuato.

The Food Traceability Rule, FSMA 204, would have required those records within twenty-four hours of the request. It was set to take effect in January of this year. Last year the administration pushed it to the middle of 2028, on the reasoning that companies needed more time to get ready, and Congress then barred FDA from spending its appropriation to enforce the rule at all while urging a rewrite with more flexibility in it. The provision written to make exactly this request answerable in a day is four years out and unenforceable in the meantime. I wrote about that yesterday.

Records in hand on July 2 does not mean a recall on July 3. Traceback takes real time after the paper arrives — you work backward through distributors and processors and growing regions, and you check what you find before you name a company. The honest claim is narrower and much harder to argue with. Eleven of those twelve days sat in precisely the place the rule was written to eliminate, and they were spent by a state epidemiologist sending the same request three times.

Set the shelf life next to the calendar. FDA says the recalled iceberg was distributed from June 29 through July 16. Bagged lettuce runs about nineteen days. Roughly half of that distribution window ran while the question of where the lettuce came from was still open — not because nobody asked, but because asking was the only thing anyone was entitled to do.

Donald Prater, FDA’s acting deputy commissioner for food, told the Times the agency responded quickly with the information it had, and that even with the twenty-four-hour rule in force it would not have made a major difference to this response. That is the government’s position on the record, and it deserves a direct answer rather than an eye-roll. Here is mine – July 1, July 3, July 6, July 8, July 13. If the rule would have changed nothing, it is fair to ask which of those five dates would have stayed where it is.

The second delay: fourteen days from the first conversation about warning people to the day people were warned.

This one is not a rule problem, and no regulation I know of would have fixed it. The July 2 follow-up email asked Yum to discuss any intervention or notification that could be taken that week to bring future case numbers down. Sarah Lyon-Callo, Michigan’s state epidemiologist, wrote that it was reassuring the company had received no complaints and had no sick staff, and that Michigan’s epidemiologic interviews were nonetheless giving a strong signal. By that point Taco Bell had already fielded a media inquiry about some of its Michigan locations. On July 4 the state told businesses and commercial kitchens handling raw produce to take extra precautions with lettuce and leafy greens. On July 7, local news in Detroit reported signs said to have been posted at some Michigan Taco Bells warning of an ingredient recall — at a time when no recall had been announced by CDC, by the state, or by the company.

The public heard on July 16. The trade got twelve days’ notice. Some customers apparently got a sign in a window. Nobody has to answer for that under any regulation, which is exactly why somebody should have to explain it out loud, and they will under oath at some point.

The third delay: sixty hours, and this is the one nobody has written about.

Nix and Peterson reconstructed the week itself. On Wednesday, July 15, CDC and FDA told Taylor Farms they believed the culprit was iceberg lettuce the company supplied to Taco Bell locations in Indiana, Kentucky, Michigan, Ohio and West Virginia. The administration gave the company a deadline to initiate a voluntary recall by noon the next day. At that point the parasite had been detected in thirty-four states.

The deadline came and went.

Then the White House got involved. A pair of White House officials convened a call with FDA and CDC, asked the health agencies to walk through the specific evidence tying the outbreak to the lettuce, and wanted to review drafts of the coming press releases. Within hours they agreed the patient surveys painted a clear picture and approved the statements. The agency updates posted after 10 p.m. Eastern, hours behind schedule. They named Taco Bell. They did not name Taylor Farms, and they did not say whether there would be a recall. A US health official told Bloomberg the back-and-forth kept the public in the dark an additional sixty hours.

Read that paragraph twice, because the part that matters is easy to skim past. The evidence held. The White House looked at the epidemiology and agreed with the epidemiologists, which is more than I expected and which I will say plainly. What came out of the process was not the finding. It was the name.

I have been writing about that particular omission for most of thirty years. It has a history and a vocabulary — Restaurant Chain A, Grocery Store B, the unnamed processor. The usual defense is that the agencies withhold a name because the evidence is not firm enough yet. Here the evidence was firm enough that the White House signed off on it the same evening, and the name still came out 60 hours later. That is not a scientific judgment about certainty. That is an editorial decision about a company.

What is not in dispute is what a US official told Bloomberg about the company’s conduct through that week: it challenged the science privately and obfuscated it publicly, and the result was customer confusion. Susan Mayne, who ran what was then FDA’s Center for Food Safety and Applied Nutrition, told Bloomberg the recall notice was the most obtuse she recalls ever seeing, and that she had not encountered that much misinformation about an outbreak in her time at the agency. She apportioned the blame in both directions, including at FDA’s own weekend communications, which is why the criticism lands. She said essentially the same thing to the Times three and a half weeks later.

None of the three delays belongs at the feet of the people who ran the investigation. Nobody on Michigan’s rapid response team voted to move 204 to 2028. Nobody working a traceback wrote the appropriations language forbidding its enforcement, and nobody at CDC asked to have their press release reviewed. The people who come out of this best are the ones with the least authority — two state epidemiologists who kept asking, and the CORE analysts and traceback teams who found a processing plant in central Mexico with no positive product sample and no whole genome sequencing to help them, because for this parasite that tool does not exist.

It is worth saying what the three delays are and are not. They do not simply add up; they overlap, and one of them is a judgment rather than a clock. But they are not the same failure wearing three hats either. One is a rule that was postponed and then defunded. One is a company deciding when to tell its customers. One is a White House asking to see the drafts. Fix any one of them and the other two are still there. 

For context on the numbers: FSMA set a target of 19,200 foreign food inspections a year by 2016, and in 2025 FDA made roughly 1,100, with significant problems cited at about 80 of them. The agency has something like 440 people to inspect more than 300,000 registered food facilities here and abroad. Its last visit to the Doctor Mora plant was in 2019, as CNN reported and the Times has since confirmed. The plant shut down on July 18; inspectors reached it this week.

The Times puts the national toll at more than 24,500 confirmed or suspected cases and two deaths, which is CDC’s figure and the right one for a newspaper to cite. My own tally, built only from what the fifty state health departments publish about their own residents, stood at 30,523 last Friday night. Both are floors. Neither is the number of people who got sick.

Everyone writing about this summer’s Cyclospora outbreak has been reaching for the same four letters and a number — FSMA 204 — and most readers have no particular reason to know what that is or why it keeps coming up. It is worth laying out plainly, because the history is not a story about a rule being delayed once. It is a story about a rule that Congress ordered, that an agency did not write, that a federal judge eventually forced into existence, and that has now been postponed twice more, the second time by Congress itself in a bill that ended a government shutdown.

Start with what the rule does. Section 204 of the Food Safety Modernization Act directs FDA to identify the foods most often tied to foodborne illness and to require anyone handling them to keep standardized records — who they got it from, what they did to it, who they sent it to, tied to specific lots. Those foods go on something called the Food Traceability List, and leafy greens are on it. The point of the standardization is speed. When an outbreak starts, investigators can ask a restaurant chain for a spreadsheet and get an answer in a day instead of assembling a supply chain out of whatever each company happens to keep in whatever format it happens to keep it. The rule requires those records to be produced within twenty-four hours of a request, or a longer period the agency agrees to.

That is the whole idea. Now the calendar.

Congress passed FSMA with bipartisan support and it was signed in January 2011. Section 204 gave FDA a deadline: propose the recordkeeping requirements no later than January 2013. FDA did not. Years went by. The Center for Food Safety sued the agency over the missed deadline, and a federal court entered an order setting new dates — propose by September 2020, finalize by November 2022. FDA met the court’s scheduleand published the Food Traceability Final Rule on November 21, 2022, nearly a decade past the deadline Congress wrote. It took effect in January 2023, with a compliance date of January 20, 2026, giving industry three years to get ready.

In March 2025 the administration announced it would push that compliance date back thirty months, to July 20, 2028. The proposal published in the Federal Register that August and the comment period closed in September. FDA’s stated reasoning is worth repeating fairly, because it is not frivolous: traceability only works if everyone in a supply chain is doing it, and even companies that were ready said their readiness depended on partners who were not. The agency said it would spend the extra time on technical assistance and cross-sector work, and that it had no intention of changing what the rule requires.

Then Congress went further than the agency had. In November 2025, the bill that ended the longest government shutdown in American history — H.R. 5371, the Continuing Appropriations, Agriculture, Legislative Branch, Military Construction and Veterans Affairs, and Extensions Act of 2026 — carried a provision at Section 780 of Division B. It prohibits FDA from spending appropriated funds to administer or enforce the Food Traceability Rule before July 20, 2028, and directs the agency to work with industry on lot-level approaches. The same section also restricted enforcement money for the Produce Safety Rule and the pre-harvest agricultural water requirements for certain commodities. FDA has said it intends to comply with the directive and held a public meeting on lot-level traceability in June of this year in response to it.

That distinction matters and it is the one most often lost. A delay is an agency deciding the industry needs more time. A funding prohibition is Congress removing the agency’s ability to act even if it changes its mind. The rule is on the books. It is law. FDA simply may not spend a dollar enforcing it until the summer of 2028.

There is one more piece of Section 204 that has quietly gone nowhere. The same section directed the Secretary of Health and Human Services to build a system inside FDA capable of receiving traceability information and improving the agency’s capacity to trace food. A Congressional Research Service report on the rule, citing a Government Accountability Office finding from this January, says no such internal product tracing system has been established. FDA says it has been developing one since 2022. That is fifteen years after Congress asked for it.

Who wanted the delay is a fair question and it deserves a careful answer rather than a satisfying one. The organizations that pushed hardest and most publicly for postponement were trade associations — the Food Industry Association, the National Grocers Association, the Global Cold Chain Alliance — and the International Fresh Produce Association publicly claimed the flexibilities in H.R. 5371 as a win for its members. That is ordinary, documented, above-board lobbying by groups whose members include nearly every food company in the country. It is not a secret, and nobody has pretended otherwise.

There is an argument for this rule that the industry has never made loudly enough on its own behalf, and it is an argument about its own money. A traceability system that works does more than find the contaminated lot faster. It draws a line around it. It says which company, which facility, which days of production, and in saying that it also says what everything else is. Without the line there is nothing to reassure anyone with, and fear does the only thing it can do with a category it cannot sort. It prices the whole thing.

Watch what happened this summer. FDA and the company identified iceberg lettuce from one facility in central Mexico. Everything else took the hit anyway. Fresh lettuce unit sales fell nine percent in the week ending July 18, and dollar sales of prepackaged salads fell fourteen percent over the four weeks ending July 25 against the same period a year earlier. Sweetgreen, which does not use iceberg lettuce and has been told by nobody that it is connected to any of this, lost roughly six hundred basis points of comparable sales in July and cut its full-year guidance to a decline of seven to eight percent. In the Salinas Valley, Larry Cox of Coastline Family Farms chopped three hundred thousand pounds of romaine hearts back into the soil. California romaine, not Mexican iceberg, with no evidence of contamination of any kind. The harvest work his crews were counting on went with it, and his sales are down twenty to thirty percent.

The clearest illustration is on the buying side. Sysco, the largest food distributor in the country, stopped sourcing iceberg from Mexico altogether and moved to domestic growers. Its chief executive said in the same breath that Taylor Farms is a high quality, high integrity shop. Both of those things were true at once, and that is the entire point. When a buyer cannot show a customer where a particular case came from, the only risk management left is to walk away from the category.

Somebody will answer that the line was drawn here, because the recall did name a company. It was drawn twelve days after Michigan first started asking, and it was drawn around a supplier rather than around the product in front of a shopper. By then the public advice was necessarily categorical. Buy whole heads. Skip the bagged salads. That was the right advice on what was known at the time, and it is not a criticism of anyone who gave it. It is a description of what is left when the records are not standardized and the answer takes twelve days instead of one. Categorical advice prices a category.

None of this is my observation. The peer-reviewed study of the 2018 romaine outbreak put the total societal loss at $276 million to $343 million, and its authors concluded that the episode demonstrated the economic benefit of industry-wide food safety standards and improved traceability. That was published in an agricultural economics journal. The case for Section 204 was made in the industry’s own literature. The rule the trade associations asked Congress to postpone is the same rule that would have drawn a line around one facility in Guanajuato and left everybody else’s lettuce alone.

The argument is live again right now. Eleven consumer and food safety organizations wrote to congressional leaders this month asking that the traceability rule not be delayed further and pointing at this outbreak as the reason. Whether that goes anywhere is a question about appropriations language, which is where this rule has spent most of its life.

Here is the sentence I would want a member of Congress to sit with. Congress ordered this rule in January 2011 and set a deadline of January 2013. The agency missed it, a nonprofit had to sue to get the rule written at all, and a federal judge set the schedule that finally produced it in late 2022. Compliance was to begin in January 2026. Enforcement will now begin, at the earliest, in July 2028 — and it will begin then only because Congress said the money cannot be spent before that date. Fifteen and a half years from the deadline to the enforcement. In the meantime, this country recorded the largest Cyclospora outbreak in its history, and a state epidemiologist in Michigan spent the first two weeks of July asking a restaurant company, three times, where the lettuce came from – More on that tomorrow.