The Pace of the Woods

Following Revolutions in Science — and the seam it could not close

When my wife and I came to West Virginia, there were almost no ticks. I want to be precise about that, because the claim sounds like nostalgia and it isn’t. Our home sits at two thousand feet, in the Appalachian hardwoods that hold the lower and middle slopes before the northern hardwoods and, higher still, the red spruce take over — a hollow that kept its own weather. In those first years a walk in the woods ended without the small ritual of the body-check that every Appalachian household now performs by instinct. The ticks were few. Then they were not.

She is a veterinarian; I spent a career reading landscapes for the federal government. Between us we are two trained observers running our own assays on the same ground, and we registered the change well before the maps did. The deer carried it. The mice carried it. The winters stopped killing what they used to kill. By the time the surveillance literature and the county risk maps caught up, the woods had already told us, years earlier, in a language that didn’t wait for a five-year study to ratify it. Long before Kris Newby’s Bitten: The Secret History of Lyme Disease and Biological Weapons gave the unease a literature to argue over, something was amiss, and we knew it the way you know weather is turning — not from a bulletin but from the air.

I open with the ticks because they are the smallest true version of a problem I watched a room full of Nobel laureates and academy presidents circle all day, and never quite name. The problem is not discovery. The problem is time — the gap between how fast a thing can change and how fast we can establish whether the change is true, safe, or wise.

❧

This week I followed Revolutions in Science: Discovery, Imagination, and the Future — the National Academy of Sciences and Smithsonian's symposium marking the nation's 250th year. I watched remotely, and even at a screen's distance it was as good as a day of talks gets.

Mokyr’s argument is that the Industrial Revolution’s deepest cause was not coal or empire but a fall in what he calls access costs — the price of finding out what is already known. The eighteenth century built the machinery of cheap knowledge: encyclopedias, scientific societies, shared notation, the postal correspondence of the Republic of Letters. But cheap access was not enough on its own. What made the gains irreversible — what kept the West from losing its inventions the way China and Antiquity had lost theirs — was a fourth thing Mokyr names almost in passing. He calls it tightness: the degree to which a piece of knowledge is held by consensus, with confidence. Tightness is trust, quantified. It is what lets a person act on a fact they did not personally verify, because the network of people who did verify it can be relied upon.

Hold that word. The entire day turned on it.

Because if you list what is getting cheaper right now, it is exactly Mokyr’s first three access costs. Does the knowledge exist; who holds it; what does it cost to acquire — artificial intelligence is driving all three toward zero. What it is not obviously making cheaper is the fourth. Verification. Synthetic text, fluent and wrong. Citations that point to nothing. A flood of plausible surfaces with no consensus underneath. We are running Mokyr’s engine faster than ever on three of its four cylinders while the fourth — the one that made the whole thing hold — sputters. Acquisition gets cheaper as verification gets harder. That is not progress in reverse, exactly. But it is not simply progress, either.

❧

The panels, without meaning to, staged the question from every side.

Frances Arnold offered one answer to it, though she didn’t frame it as an answer. Directed evolution doesn’t ask the machine to understand the protein. It breeds variants and lets the assay decide — the enzyme catalyzes the reaction or it does not. The bench adjudicates. That is a kind of tightness you cannot fake, because nature is the referee and nature does not hallucinate. It is Bacon’s old instinct, the one Mokyr traces to the seventeenth century: ask first which techniques work, and only later why. When verification is hard, the assay is a way back to solid ground.

But an assay tells you a thing works. It does not tell you whether it should exist. That is the seam Langer sits astride, whether or not he said so from the chair. Drug delivery is a domain where the bench is brutal and honest — the formulation heals or it kills — and also a domain that cannot move a step without the slow machinery of trials, review, the FDA. He needs both at once: the fast assay that says works, and the slow institution that asks should. The same is true on the panel where Edward Chang described AI-scaled models of neurobehavior, correlated now to genetics. Powerful prediction. Real clinical good. And Ellie Pavlick, one chair over, spends her career on precisely the question that shadows it: when a model predicts beautifully, does it understand, or has it found a convincing surface? Carla Shatz, moderating, has spent hers showing how a brain wires itself through activity against a world that pushes back. The cortex earns its reliability by contact with consequence. It is fair to ask what a system trained without that contact has earned.

I am not reporting alarm. The day was, mostly, wonder — and it should have been. But across every panel the same shape kept surfacing: discovery has gone supersonic, and the apparatus that establishes whether a discovery can be trusted still moves at the pace of a convened committee. Two clocks. A widening gap between them.

 \The Widening Gap — two clocks, 1775–2025. A schematic, not a measurement: the green line is the rate of biological change, nearly flat across 250 years — generational time, selection, the slow wiring of a brain against a world that pushes back. The blue line is technological change — slow through the early Industrial Revolution, inflecting hard in the late nineteenth century, near-vertical in the digital and AI era. The shaded wedge between them is the verification gap: the distance by which our power to generate has outrun our means to confirm.

❧

Then Fineberg, in the closing keynote, named it.

His closing slides did, in the academy’s own vocabulary, what I had been turning over in mine. He listed the day’s tensions, and two of the seven were these: the speed of technology against the pace of institutions and trust; the urgency of the moment against the long-term nature of science. There it was, from the closing keynote — the gap between the two clocks, set down as a defining tension of the age rather than a footnote to it.

And he offered the thing that holds the seam. He called it corrigibility — the shared structure of science and democracy. Both build in mechanisms for revision: replication, peer review, falsification on one side; amendment, election, judicial review on the other. Both rest on the premise that no single authority is infallible. Both treat dissent as a feature, not a flaw. Corrigibility is what tightness looks like when you watch it work over time. It is not a wall against error. It is a machine for correcting error faster than error accumulates.

Which is exactly why one number on his slides should stop us. Public confidence in scientists fell from 87 percent in April 2020 to 77 percent in January 2026. That is the erosion, measured. And his fifth and final commitment was the one that matters most for what comes next: earn trust — by science understanding the public first, not lecturing but learning. Stand in the seam. Honour the tension rather than rushing to close it.

❧

North and South America as they would appear from space, 35,000 km (22,000 miles) above Earth — the “Blue Marble,” Western Hemisphere. A true-color composite combining MODIS land-surface data from NASA’s Terra satellite (collected over 16 days) with cloud data from NOAA’s GOES satellite. Image created by Reto Stöckli, Nazmi El Saleous, and Marit Jentoft-Nilsen, NASA/Goddard Space Flight Center. Public domain. Even our truest image of the one planet is a composite — trustworthy because its making is documented, and it travels with its papers.

So here is where the woods come back, because they teach the rest.

For most of what is accelerating, the answer is not regulation in the heavy sense — it is something more like a governor, in Watt’s mechanical meaning: a coupling that lets a system sense its own speed and hold itself stable. Most discovery is reversible-and-fast. There, the work is to let curiosity run and require that claims travel with their provenance — who verified this, against what, with what confidence — so the network can adjudicate at the speed claims now move. That is tightness as something built in, not bolted on. My wife’s caseload, the birder’s field notes, the naturalist’s eye registering a changed forest: that distributed signal ought to feed the system at the speed it is generated, not wait years for a study to confirm what the ground already says.

But some change is not reversible. A released capability, a self-replicating construct, an edited germline — these do not decelerate when you cut the fuel. For that narrow class, the governor is not enough; there the slow committee is not a bug but the safety feature, and the honest word is gate. The hard, unglamorous work — the work I spent a career doing inside a federal agency, deciding which calls could be distributed and which had to be held — is drawing the line between the two. Most of the current argument fails because it reaches for one instrument and applies it to both.

The ticks are the lived proof of the first failure: a governor uncoupled, a signal that ran ahead of the institutions meant to register it. And they point, without my having to litigate who did what, toward the second: the class of risk where reversibility fails and a gate is not optional. Lived signal first, declared principle second. The woods earned the argument.

Fineberg ended on a single word. Who stands to gain from science? — Humanity. And then the condition, which is the whole point: if we preserve openness, encourage generosity, strengthen institutions, renew our commitment, and never cease to wonder. Strengthen institutions. He said it last because it is load-bearing. The Enlightenment did not just build the engine of discovery. It built the steering wheels — the corrigible institutions that kept the engine from running into the trees. We are very good, just now, at the throttle. The question the day left open, the one the woods have been asking us for years, is whether we will keep the brakes in working order while we accelerate.

It was Marcia McNutt who opened the day, in her role as president of the National Academy of Sciences, and it is worth ending on her — because the abstraction “strengthen institutions” only means something when you can put a life to it. I have followed her work since her years leading the U.S. Geological Survey, and what I have watched across that span is a particular and increasingly rare thing: a serious scientist who chose the harder vocation of public service and stayed in it. She has steered the Academy through tough and shifting weather, the kind that tests whether an institution is a building or a commitment. Institutions do not strengthen themselves. They are held — through changing times and changing winds — by people who decided the holding was worth a career. That is the work the woods cannot do for us, and it is the work this anniversary should honor most.

❧

Tapadh leat to everyone who built that remarkable day — and to the reader who has stood in a seam long enough to know it is a country of its own.

 

A note on how this was made

This essay is a collaboration. The standing is Sam Lammie’s — a career spent reading landscapes for the U.S. Forest Service, and a life lived at two thousand feet in the West Virginia hardwoods, where the ticks and the weather and the woods do their own teaching. The argument, the lived material, and the final judgment are his.

Claude (Anthropic) was the thinking partner across a long day and a longer night: drawing out the structure of the day’s symposium, pressing on the weak joints of the argument, drafting and redrafting prose, and making the figures. The “Three Velocities” diagram has a third hand in its lineage — its three-curve decomposition began in a graphic generated by ChatGPT (OpenAI), which Claude redrew in this blog’s own visual register, stripping the false precision and restoring the honest caveat that the shape, not the numbers, is the claim. The header image is NASA’s public-domain “Blue Marble,” credited above.

It takes a village to think clearly now — human standing, machine help, and the discipline to say plainly which is which. A piece about verification and provenance should carry its own. This one does: Sam’s hand, Claude’s, an openly credited assist from a third tool, and a NASA composite that wears its making on its sleeve. Where this work used a machine, it says so.

— Sam Lammie & Claude, The Musical Stone, 2026

 

A Working Whole

 On the charge that the environmental industry “captured” the Forest Service — and what the charge gets wrong

A response to Franklin Otis Carroll’s “How the Forest Service Didn’t Fail Us” — Facebook, June, 2026.

Carroll’s argument — that the Forest Service was captured by an environmental industry, and that cutting its partnerships is the first step toward restoring it — is not wrong about everything, and the honest place to start is with what it gets right. The multiple-use mandate has been hollowed out. Active management did decline. Roads fell into disrepair, fuels built up, and in a great many western communities the agency lost the trust of the people who live closest to the trees. I spent thirty-five years in federal service and the majority as a Forest Service employee — fifteen as GIS Coordinator on the Monongahela, a stretch with line authority as Acting District Ranger on the Cheat–Potomac, and a last chapter as geospatial program manager for the Northern Region, responsible for the spatial record across twenty-five million acres in Montana, Idaho, and the Dakotas. 

But the diagnosis is wrong, and it is wrong in ways that matter.

Begin with the exhibit the essay leads with: the decision to sever forty-three partnerships and claw back roughly four million dollars from groups said to be operating “in direct opposition” to the mission. The essay names that correctly as an Interior Department action — and then asks it to carry weight it cannot bear. The groups Interior cut were flagged for climate advocacy, equity, and immigration support; the named examples were Conservation International and a cultural-landscape preservation foundation. None of it touched timber, fuels, or fire — the actual subject of everything that follows. And in a related round of terminations the same department pulled a set of conservation grants, money funding native-species and habitat restoration by groups like the Institute for Applied Ecology, that a federal court in Oregon has since ordered restored, finding the cancellations a purge by political viewpoint rather than any audit of performance. So the opening “audit” does rhetorical work, not evidentiary work: it signals that the tide has turned against the activists without establishing a single thing about how the Forest Service manages fire.

The capture story is the load-bearing claim, and it has the history backwards. The multiple-use mandate did not bend because lawyers “moved in.” It bent because Congress rewrote the agency’s orders, in daylight, over two decades — the Wilderness Act of 1964, the National Environmental Policy Act of 1969, the Endangered Species Act of 1973, the National Forest Management Act of 1976 — much of it signed by Republican presidents.

And here I can speak from the ground this argument is standing on. The National Forest Management Act, the modern charter of all national forest planning, exists because of a clearcutting lawsuit on the Monongahela — the forest where I spent fifteen years. A federal court read the 1897 Organic Act as written, found that clearcutting violated its plain terms, and effectively halted the timber program until Congress acted. The author would call that capture. I call it the system Pinchot built doing precisely what it was designed to do: when the agency drifts from its founding law, the courts and Congress pull it back. That is not an industry mining the government. That is the rule of law, working as intended, on the very ground this piece claims to defend.

It is worth adding that the agency had overshot. In the postwar boom it became, in the scholars’ phrase, a regional employment “lunch bucket” — a timber-first operation wearing a multiple-use motto while timber took the major use. The correction came from a changing society and the laws it passed, not from a conspiracy in Ventura.

And the agency did not become the Park Service. This is the essay’s sharpest mischaracterization — the claim that the Forest Service traded its working charter for a preservation model unsuited to lands meant to be worked. It did no such thing. It still sells timber, runs grazing allotments, permits minerals, builds and maintains recreation, and manages for wildlife and water — the same multiple-use mandate it has always carried, and one fundamentally different from the Park Service’s preservation charge. What changed is not the mandate but the balance within it, and the arrival of ecosystem health as a frame. That is evolution, not surrender. Multiple use was never a fixed ratio. It was always a negotiation, and the negotiation matured.

Even wilderness — the part the essay quietly files under “museum” — is a form of use, and one of the oldest. I have believed in wilderness since I was old enough to carry a gun, and I have never felt the contradiction the author assumes. Wilderness on a national forest is not a national park: it still carries hunting, fishing, packing, and outfitting, the hook-and-bullet tradition that built American conservation in the first place. Stand on a Bitterroot trail in the fall and watch a hunter bring a quartered elk out on his back, and then try to tell that hunter wilderness is the enemy of working land. The people who use that country hardest are very often its fiercest defenders. The author has mistaken restraint for abandonment — and the two are not the same thing.

And multiple use is not a slogan. It is a payroll. I have worked, in one capacity or another, across most of the agency’s program areas — timber and aquatics, recreation and wildlife, and the dozen-odd others the public rarely hears named: hydrology, soils, range, fuels, heritage, lands, minerals, engineering, the geospatial work I spent a career on. A national forest is run by something close to twenty professional disciplines, not by foresters alone and certainly not by lawyers. The essay’s picture — the foresters let go, the briefcases moving in — erases that entire workforce, the people who actually deliver the “multiple” in multiple use.

Ponderosa Fuel Treatment - North of Flagstaff

And it leaves out the half of the founding equation that matters most. The author lists watershed once, in a passing string of uses, and then forgets it. But the national forests were never created only to grow timber. The Organic Act of 1897 — the same law the Monongahela case turned on — gave the reserves two coequal purposes: to furnish a continuous supply of timber, and to secure favorable conditions of water flows. Water was there at the creation. It is not a soft amenity bolted on later by activists; it is why a great many of these forests exist at all. They are the largest single source of water in the country — roughly a fifth of the nation’s supply — and by the agency’s own accounting some 180 million people in tens of thousands of communities depend on national-forest watersheds to catch and filter what comes out of their taps. Manage these lands as nothing but a timber base and you have not restored their mission. You have forgotten half of it.

None of this is to pretend NEPA litigation is costless. It isn’t. Environmental analysis consumes something like a fifth of the money available to manage the national forests and close to a third of the agency’s field capacity, and the threat of suit does delay treatments that need to happen before the fire season. I have prepared those documents and lived inside that process; I will not romanticize it. But the picture of an agency paralyzed and forever losing is a caricature. Across the suits filed against it over a representative stretch of years, the Forest Service won more than it lost. NEPA’s burden is real — and it is a problem of process and capacity, not evidence that the agency was conquered.

On fire, the essay wants suppression and the collapse of active management to be rival explanations, with climate change cast as the activists’ alibi. Start with the history it skips. Fire suppression was not an environmentalist invention — it was the Forest Service’s own founding doctrine. The agency was forged in the Great Fire of 1910, the Big Burn that took some three million acres and dozens of lives across Idaho and Montana, in the country I now live in. Out of that trauma came a policy of total fire exclusion, hardened into the 1935 “10 a.m. policy” and sold to the public for a generation by Smokey Bear. That was the agency talking to itself, decades before NEPA existed or the modern environmental movement was born. And when the science finally turned — when ecologists began arguing that these forests need fire — it was largely conservationists, not commodity interests, who pushed to put fire back on the ground. The author has the arrow backwards. Suppression is the original sin here, and it is the agency’s, not the activists’.

None of which lets the fuels off the hook. A century of exclusion and the decline of thinning and harvest both loaded the stands — the author is right about that — and a warming, drying climate then stretched the season and dried the fuels further. Ask the researchers at the Missoula Fire Sciences Laboratory, or the smokejumpers based up the valley from me, whether the season runs longer than it used to. Climate is not an alibi; it is a multiplier.

And there is a fact the essay leaves out altogether, because it complicates everything: the forests do not, in the main, light themselves. More than eighty percent of wildfires in this country are started by people. When you narrow to the fires that actually threaten homes, the share climbs higher still. In one large study of a quarter-century of ignitions, humans caused ninety-seven percent of the wildfires in the wildland-urban interface and accounted for ninety-two percent of the homes threatened. What burns down a community is rarely an un-thinned stand that a logger was kept out of by a lawsuit. It is a human ignition meeting an interface we keep building deeper into fire-prone woods. You could thin every acre the essay wants thinned and still lose the town if the ignition source and the building pattern go unaddressed. That is not an argument against thinning. It is an argument against the single-cause story — the very thing the essay accuses the other side of telling.

Which brings us to the owl, offered as the emblem of single-species folly, with the barred-owl removal as the movement’s bloodthirsty endgame — up to 450,000 birds, branded “wildlife ethnic cleansing.” Set aside that the phrase loans the vocabulary of atrocity to a wildlife program and cheapens both the atrocity and the genuine, painful dilemma it describes. The deeper trouble is whose program it is. The barred-owl strategy is a federal intervention, run by the Fish and Wildlife Service — Interior again — and it is opposed across the spectrum: animal-welfare groups suing under NEPA, bipartisan members of Congress, and a good many of the very environmentalists this piece indicts. The author is against intervention when it is a court-ordered chainsaw and for it when it is a shotgun. Strip that inconsistency away and the real objection was never single-species fixation. It is environmental groups as a category.

The essay also reaches for science when it helps — the fire-suppression story is “not settled,” merely “convenient” — and goes quiet about what the administration it applauds is doing to the science itself. The Chief of the Forest Service told a House appropriations subcommittee, in plain words, that research and development is zeroed out in the budget: roughly three hundred million dollars driven to nothing. The reorganization moves to close better than half of the agency’s research stations — among them the Seattle smoke lab whose work helped build the public wildfire-smoke maps that millions of families now check before they let their children outside. The long-term experimental forests, some carrying datasets decades deep, would be snuffed out with them. The Missoula Fire Sciences Laboratory I pointed to a moment ago survives — spared, it appears, because it serves the favored emphasis on fire and timber — but the broader science that would actually tell you how to build a forest that resists catastrophic fire is the part on the block. You cannot run science-based forestry once you have defunded the science, and members of both parties have said so. Zeroing out the research arm of the nation’s forest agency is not reform. It is a disgrace.

And here is the contradiction the essay never resolves. It wants active management — thinning, open roads, woods workers, fuels brought down. But the partnerships it cheers Interior for cutting are how active management actually gets done now. The grants a court just restored funded habitat restoration. Across the fence in Agriculture, the same work runs through collaborative forest landscape restoration, stewardship contracting, Good Neighbor Authority, shared-stewardship agreements with states and, yes, with conservation groups. You do not rebuild active management by firing the partners who carry it out. You rebuild it by funding and staffing the field. The essay mourns the lost forester while applauding the policy that loses more of them. The foresters I watched leave did not leave because of anyone’s briefs. They left because of budgets, hiring freezes, reductions in force, mills consolidating, and the old-growth that once propped up the cut running out — and now they leave because of the very dismantling this argument calls restoration.

The strongest line in Carroll’s piece is its insistence that the forest is a living, working whole, not a museum. I agree without reservation. But so is the agency, and so is the body of law that governs it — an ecotone of competing uses held in tension, which is the only place a multiple-use mandate can actually live. You do not restore a working whole by amputation. Defend multiple use by rebuilding capacity, not by purging the people who do the work.

I want what the author says he wants: forests that are worked and healthy, and rural communities that trust the agency again. That is reached by putting skilled hands back on the ground — not by firing the briefcases and discovering, too late, that some of them were holding the saws.

 

A note on sources.

 USDA Forest Service. 2007. Forest Service Research Image Library. Fort Collins, CO: Forest Service Research Data Archive. https://www.fs.usda.gov/rds/imagedb

This piece was produced in collaboration with Claude, an Anthropic AI assistant. The argument, the sources, and the editorial judgments are the author’s.