Concrete · Cement · Rome · The Concrete Club · Carbon · John Wilson

Concrete: What the Record Shows About the Most Consumed Material on Earth

Humanity makes about 30 billion tonnes of concrete a year, more of it than of any substance except water itself, and nearly everything it builds stands on a material most people cannot define: cement is an ingredient, not the material; the rock hardens by chemistry, not by drying; it is immensely strong pressed and weak pulled, which is why steel runs through it. This report prints what the record supports about the material, from the Roman mix that heals its own cracks, proven in a frozen Pompeii construction yard in December 2025, to the cartel that taxed New York's pours at two percent, to the carbon ledger that no clean power source can settle, because most of the emissions are the chemistry itself.

By Eunice, the desk · Collected and analyzed at 1800 ET, Friday, November 13, 2026.

The File in Brief

Concrete is a manufactured rock with a twenty-century documentary record, and the record supports a short set of load-bearing facts. The material is cement paste (Portland cement and water) binding sand and stone; it sets by hydration, a chemical reaction that proceeds underwater and continues for years, reaching about 90 percent of its strength in 28 days; its tensile strength runs a tenth to a seventh of its compressive strength, so modern concrete is a composite with steel, the steel carrying tension and the rock carrying compression. The Romans built with a volcanic-ash version, opus caementicium, whose durability was a mystery for two millennia; the mechanism, hot mixing with quicklime, leaving reactive lime clasts that dissolve into cracks and recrystallize as stone, was demonstrated by MIT researchers in Science Advances in January 2023 and confirmed archaeologically on December 9, 2025, when a Nature Communications paper reported an unfinished Pompeii construction site holding dry piles of quicklime premixed with pozzolan, exactly as the hypothesis required. The Pantheon, completed about 126 AD, remains the largest unreinforced concrete dome in the world. The modern material dates to Joseph Aspdin's 1824 patent for Portland cement and to the reinforced patents of the later nineteenth century, Monier's pivotal 1867 filing among them, and its monument is the Hoover Dam: 3.25 million cubic yards in the dam itself, by the Bureau of Reclamation's own figures, cooled through 582 miles of embedded pipe because a single monolithic pour would have taken an estimated 125 years to cool. The scale today is about 4 billion tonnes of cement a year, a figure now falling from its 2023 peak of 4.1 billion as China's building boom ends, and China poured more cement in 2011 to 2013, about 6.5 billion tonnes, than the United States poured in the entire twentieth century, about 4.5 billion. The costs are two. Cement accounts for 7 to 8 percent of global carbon dioxide emissions, and roughly 60 percent of that is process emissions from calcining limestone, chemistry rather than fuel, which is why the working fixes are clinker substitution, commercial since 2025 in the form of LC3, and carbon capture, operating since June 2025 at Heidelberg's Brevik plant in Norway; the third announced fix, electrochemical cement, has a standards designation and two technology-company offtake deals and no plant, its federal grant cancelled in May 2025 and its first works unbuilt toward 2028. And the material's American folklore is adjudicated fact: in 1980s New York, a club of six concrete contractors, coordinated by four crime families, controlled Manhattan contracts above $2 million and paid a 2 percent cut to Ralph Scopo, the concrete workers' union president who received 100 years in the 1986 Mafia Commission Trial. The culture keeps returning to the material, most recently John Wilson's 2026 feature documentary, The History of Concrete, and the desk's standard for the subject remains the one set by David Owen's 2003 New Yorker piece, whose famous figures this report prints as Owen's, where no primary source carries them.

The desk's method is the paper's method. Material was supplied to the desk as AI-assistant summaries of the literature and of The New Yorker's concrete file. Every checkable claim was re-verified against the primary record: USGS Mineral Commodity Summaries read directly, the Science Advances and Nature Communications papers, the Second Circuit's opinion in United States v. Scopo, the Bureau of Reclamation's figures, ASTM's standards history, and the magazine's own archive for claims about the magazine. The supplied material was sound in outline and wrong in several load-bearing places, including the authorship of one of the magazine's best-known concrete pieces; the corrections are printed where they belong.

1. The Chemistry: It Does Not Dry

The first fact is a correction most adults carry uncorrected for life: cement and concrete are different things, and concrete does not harden by drying. Cement is a powder, an ingredient. Concrete is the composite: cement paste, sand, coarse aggregate such as gravel or crushed stone, and water, in proportions the engineer chooses. When water meets cement, the two react chemically, a process called hydration, forming calcium silicate hydrates, interlocking crystals that bind the aggregate into artificial stone. Because the hardening is a reaction and not an evaporation, hydraulic cement sets and gains strength underwater by definition, and pours are kept wet deliberately after placement to feed the reaction. Strength accrues on a curve: roughly 90 percent of design strength by 28 days, and, while moisture remains available, for years beyond it. David Owen's 2003 New Yorker piece, “Concrete Jungle,” opened its technical account on exactly this point, and on this point the textbooks and the magazine agree without reservation.

The second fact explains the steel. Concrete's compressive strength, its resistance to being crushed, is enormous and engineerable, from ordinary structural grades to the ultra-high-performance mixes Owen cited at up to 40,000 pounds per square inch, a figure the desk prints as Owen's; laboratory ultra-high-performance work reaches such strengths, and no primary source carries his number as a general claim. Its tensile strength, resistance to being pulled apart, runs only 10 to 15 percent of the compressive figure. A beam of plain concrete is strong stood upon and weak bent. The solution, patented into a system in the nineteenth century, is reinforcement: steel bars embedded in the wet mix where tension will run, so the finished member is two materials doing the two jobs, steel in tension, stone in compression. Nearly everything called concrete in the modern built world is that composite. The third fact is the clock. Wet concrete is perishable: the standard governing ready-mix, ASTM C94, required discharge within an hour and a half of mixing from 1935 until its 2021 revision left the limit to agreement between buyer and producer. Owen's roughly 90 minutes was not atmosphere; it was the specification of his era, and it is the fact on which the crime story in section 5 turns, because a perishable material delivered by truck is a material whose supply can be squeezed.

2. Rome: The Mix That Heals Itself

The oldest concrete in continuous service is Roman, and until recently nobody could say precisely why it lasts. Roman builders mixed lime with pozzolana, volcanic ash from the Pozzuoli region, to make opus caementicium, and built harbors that set in seawater and the Pantheon, completed about 126 AD under Hadrian, whose dome spans 43.3 meters, a diameter equal to the height from floor to oculus, and which remains, nineteen centuries later, the largest unreinforced concrete dome on Earth. The received account credited the ash. In January 2023, a team led by Admir Masic at MIT published the mechanism in Science Advances: Roman mortar was hot-mixed, combined with quicklime at high temperature, which left millimeter-scale lime clasts distributed through the material. The clasts were long read as evidence of sloppy mixing. They are the repair system. Cracks route preferentially through the brittle clasts; water entering a crack dissolves the lime, and the saturated solution recrystallizes as calcium carbonate, sealing the crack. In the laboratory, cracked Roman-style samples healed and stopped water flow within about two weeks, and modern-style samples did not.

In December 2025 the hypothesis stopped being a hypothesis. On December 9, Nature Communications published the analysis of an unfinished construction site in Pompeii's Regio IX, a house mid-renovation when Vesuvius froze it in 79 AD: dry piles of quicklime premixed with dry pozzolan, stored as ingredients, ready for hot mixing on the spot. The paper's word is “unequivocal,” and it contradicts the surviving Roman text, Vitruvius, whose account describes slaked lime prepared first. The builders' practice, it turns out, was better than the builders' literature. The desk prints the sequence because it is the cleanest science story in this file: a two-thousand-year-old durability observed, a mechanism proposed from microscopy in 2023, and the mechanism confirmed in the ground in 2025 by a construction yard that never got to pour. Modern concrete is stronger, faster, and reproducible to specification in a way Rome never approached; on the single axis of surviving its own cracks, the record now shows the Roman mix doing something the modern one does not, and shows why.

3. The Reinvention, the Dam, and the Scale of the Present Pour

Modern concrete is a nineteenth-century reinvention with named inventors and dated patents. In 1824, the English bricklayer Joseph Aspdin took British Patent 5022 for a cement he named Portland, a quality claim, not a place of manufacture: the product resembled the prized building stone of the Isle of Portland. Reinforcement followed within a generation; the desk prints the full line because the pastes do not: Joseph-Louis Lambot built a reinforced boat in 1848 and patented it in 1855, William Wilkinson patented reinforced floors in 1854, and Joseph Monier's 1867 patent, for large flowerpots strengthened with a cage of iron wires, became the pivotal filing from which the reinforced-concrete industry grew. François Hennebique patented his complete building system in 1892. Its American monument is the Hoover Dam, and the Bureau of Reclamation's own figures are the record: 3.25 million cubic yards of concrete in the dam, 4.36 million in the whole project, placed between June 6, 1933, and May 29, 1935. The dam could not be poured as one mass, because hydration is exothermic and a monolith of that size would, on Reclamation's estimate, have needed about 125 years to cool; it was built in blocks, with more than 582 miles of one-inch steel pipe embedded through them, circulating ice water from a plant making 1,000 tons of ice a day, the cooling finished in March 1935.

The China comparison survives verification and deserves its exact form. Vaclav Smil, working from USGS data, calculated that China used roughly 6.4 to 6.6 billion tonnes of cement in the three years 2011 to 2013, against roughly 4.4 to 4.5 billion tonnes for the United States across the whole twentieth century; the USGS's own cement specialist, Hendrik van Oss, put the comparison to the Washington Post in 2015 as China's 2010–12 consumption running at about 140 percent of America's 1900–99 total. It is the single most informative statistic in the modern material record: more cement in three years of one country's building boom than in the century that built the American dams, highways, and skylines combined. The same literature raises the sand question, and the desk prints both bearings on it. The UN Environment Programme reports 40 to 50 billion tonnes of sand and gravel extracted a year, calls sand the second most used resource after water, and documents real damage to rivers and coasts from unregulated extraction, most recently in its third sand report of May 2026. The USGS position runs beside it: sand and gravel are geologically plentiful, and what is scarce is cheap, angular, washed construction sand near the point of use; manufactured sand from crushed rock is the standard substitute. The crisis, on the combined record, is one of governance and local damage, not of the planet running out of sand.

4. The Carbon Ledger: The Emissions Are the Chemistry

Cement's share of global carbon dioxide emissions is 7 to 8 percent, the IEA carrying the lower figure and recent peer-reviewed reviews the higher, and the composition of that share is the finding that governs everything else in this section. Roughly 60 percent of cement's emissions are process emissions: the CO2 driven out of limestone when it is calcined into clinker, released by the reaction itself before any fuel is burned. The remaining 40 percent is energy. The consequence is arithmetic, not opinion: a cement works running entirely on clean electricity still emits the majority of its current footprint, because the footprint is the product. This is why cement is the industrial sector least amenable to the standard decarbonization story, and why the credible fixes are the ones that change the chemistry or catch the gas.

Two of the three fixes are in commerce in 2025 and 2026; the third is in press releases, and the desk prints them at their true sizes. Clinker substitution is the oldest lever, replacing part of the clinker with other binders, and its current form is LC3, a cement built on calcined clay and limestone: standardized in India as IS 18189 in 2023, in commercial production at JK Cement's Mangrol works from July 2025, poured at scale at Noida International Airport that September, and produced by CBI Ghana's flash-calcination plant from May 2025. Capture is the second: Heidelberg Materials' Brevik plant in Norway, inaugurated June 18, 2025, is the world's first industrial-scale carbon capture facility at a cement works, designed to take about 400,000 tonnes of CO2 a year, roughly half the plant's emissions, for storage under the seabed by Northern Lights, heavily supported by the Norwegian state; a January 2026 trade report put its achieved capture rate at about 50 percent, limited by available waste heat, and its product sells as a zero-carbon cement at a premium. The third fix, electrochemical cement, is real as chemistry and absent as industry: Sublime Systems' product carries an ASTM performance designation from 2023, and Microsoft and Meta have signed offtake agreements, Meta's reported on November 12, 2026. Its first commercial plant, at Holyoke, Massachusetts, is unbuilt: the Energy Department cancelled its $87 million grant in May 2025, deployment is paused, and the company's target has slipped to about 2028. A November 2026 report that prints electrochemical cement as an operating alternative is printing a prospectus. The industry's own roadmap, from the IEA and the Global Cement and Concrete Association, leans on exactly the levers the record shows working: substitution, efficiency, and capture, in that order of present reality.

5. The Concrete Club: The Cartel in the Court Record

The most colorful claim in this file is also one of the best documented, because it was tried in federal court, and the court's version corrects the retellings in two particulars. In the 1980s, Manhattan concrete contracting above a threshold was allocated, not bid. The controlling account is the Second Circuit's in United States v. Scopo and Montemarano, 861 F.2d 339: concrete contracts over $2 million were controlled by a club of six contractors, coordinated by four crime families (the supplied summaries say five; the appellate record says four), whose members paid Ralph Scopo 2 percent of the contract price, the cut divided among the four families. Contracts at or below $2 million were a separate Colombo-family extortion documented in the same record. Scopo was the scheme's hinge: a Colombo soldier who was also, from 1977 to April 1985, president of the Cement and Concrete Workers District Council, holding both the cartel's enforcement and the union's labor supply in one pair of hands. The leverage was the material itself: a perishable product, on a 90-minute clock, delivered by union drivers to sites that could not pause, meant a slowdown or a missing truck could destroy a pour and its contractor with it, a point Owen's account makes and the trial record supports.

The ending is a matter of sentence. Scopo was convicted in the Mafia Commission Trial, the government's case against the Commission of La Cosa Nostra; the jury convicted on November 19, 1986, and on January 13, 1987, Scopo was sentenced to 100 years in prison and a $240,000 fine. He died in federal custody on March 9, 1993. The supplied material attributes vivid quoted dialogue to a prosecutor, Peter Lieb; Lieb is documented as a real organized-crime prosecutor in the Southern District in that period, and the desk could verify neither the quotes nor their setting in any source, so they are not printed. What the record carries without assistance is sufficient: a two-percent tax on the material a city is built from, collected for years, ended by a racketeering trial and a sentence measured in centuries. New York, on Owen's figure, poured a Hoover Dam's worth of concrete every eighteen months through the years the club operated; printed as Owen's, the figure gives the cartel's tax base its scale.

6. The Culture File: Owen, a Talk Piece, a Memoir, and a Movie

The material's literature is thin and strong, and its standard was set on November 10, 2003. David Owen's “Concrete Jungle,” published in that issue of The New Yorker under the subtitle “Giant wrecking balls, Mob contracts, and the cement that holds the city together,” is the magazine's fact piece in its classic form: an overlooked substance under the reader's feet, unfolded into chemistry, engineering, history, demolition, and crime, and closed on the cycle of a city grinding itself down and pouring itself back up. Its reporting on the Concrete Club is corroborated by the court record in section 5, with the family's count corrected. Its scale figures, the Hoover Dam equivalence for New York's pours and the 40,000 psi strength, are printed in this report as Owen's, because the article is paywalled and no primary source carries them independently; the piece's reputation rests on the parts that check. The magazine returned to the material in smaller forms. Lauren Collins's “Rising,” dated June 7, 2010, is a Talk of the Town piece, not the feature the supplied summaries describe; its opening is the record of a morning: “Every morning at four-thirty, sixty concrete trucks, from Brooklyn, from Queens, from New Jersey,” converging on the World Trade Center site as One World Trade Center climbed past its twentieth floor. And the magazine's Personal History, “Impalpable Dust,” of March 27, 1989, belongs to the file properly: M. R. Montgomery, the Boston Globe columnist, writing about his father, an engineer on the Fort Peck Dam, a piece later carried into his memoir Saying Goodbye. One of the supplied summaries credits the piece to Roger Angell; the magazine's own archive says Montgomery, and the error is printed here because correcting the byline of the literature is part of reporting the literature.

The current entry is a film, and it is the strangest document in the file. John Wilson's The History of Concrete, his first feature after his HBO series How To with John Wilson ended in 2023, premiered at Sundance on January 22, 2026, and opened in limited American release through Magnolia Pictures on September 18, 2026; it was still screening in November. Its premise is on the record and requires no embellishment: unable to finance a documentary about concrete on the subject's merits, Wilson attended a workshop on the Hallmark television-movie formula during the Writers Guild strike and built the documentary inside that formula, commercial beats and all, a structure that reviewers, including The New Yorker's Richard Brody, who called the result pointedly and urgently political, have treated as the film's method rather than its gimmick. The film ranges as its director's series ranged, from a crack in his own house's foundation to a Las Vegas concrete trade show, sidewalk gum removal, and a 3,100-mile footrace in Queens. The desk's finding on the culture file is the one the three decades of it support: the material defeats ordinary treatment. Owen mastered it with reporting, Montgomery with a father's biography, Wilson by surrendering to a borrowed plot. Every treatment is an attempt to make a reader or a viewer look at the gray surface under everything and see it. The record says they keep succeeding, briefly, one work at a generation, and the surface closes over again.

Sources and Method

Principal sources: USGS, Mineral Commodity Summaries 2025, cement chapter, read directly, and MCS 2026 figures as reported; Vaclav Smil, Making the Modern World, and the Washington Post's March 24, 2015 account quoting USGS specialist Hendrik van Oss; Seymour, Masic et al., “Hot mixing,” Science Advances 9(1), January 6, 2023; Vaserman, Masic et al., “An unfinished Pompeian construction site reveals ancient Roman building technology,” Nature Communications 16:10847, December 9, 2025; Encyclopaedia Britannica on Portland cement, Aspdin (British Patent 5022, 1824), Monier, and Hennebique, with Lambot (1848/1855) and Wilkinson (1854) from the historical record; US Bureau of Reclamation, Hoover Dam FAQ; ASTM C94's discharge-time requirement (1935–2021) from standards histories; RSC Sustainability (2025) review and IEA figures for the 7–8 percent emissions share and the 60/40 process-energy split; Heidelberg Materials and Longship reporting on Brevik (inaugurated June 18, 2025) and January 2026 trade reporting on achieved capture; LC3 records (IS 18189:2023; JK Cement, July 2025; CBI Ghana, May 2025); Data Center Dynamics (May 29, 2026) and trade reporting (November 12, 2026) on Sublime Systems; United States v. Scopo and Montemarano, 861 F.2d 339 (2d Cir. 1988), and the Commission Trial sentencing record (January 13, 1987); UN Environment Programme sand reports (April 26, 2022; May 2026) and the USGS position on sand resources; The New Yorker's own pages for Owen (November 10, 2003), Collins (June 7, 2010), Montgomery (March 27, 1989), and its September 2026 coverage of Wilson; Sundance and Magnolia Pictures records for the film. Corrections to the material supplied to the desk: the Concrete Club was coordinated by four families, not five, and the threshold in the appellate record is $2 million, not $5 million; the 4.1-billion-tonne cement figure is 2023's, superseded by a declining series; “Impalpable Dust” is M. R. Montgomery's, not Roger Angell's; Collins's “Rising” is a Talk of the Town piece, not a feature; quoted dialogue attributed to prosecutor Peter Lieb could not be verified and is excluded, Lieb himself being documented; the Wilson film's runtime is listed at 101 minutes by one review and 102 by another, and the report prints neither. Figures stated by Owen are attributed to Owen. Research files for this report are retained by the desk.