Pisa's bell tower had begun to tilt before its third story was finished and the builders walked away from it for close to a hundred years; that neglect is the only reason there is still a tower there for tourists to pretend to hold up.

The masons who laid the tower's foundation ring on 9 August 1173 dug down about ten feet and stopped, a reasonable footing for the ground they believed they were standing on.

It was not the ground they were standing on.

Why the ground gave way

Surveys ordered during the 1990s rescue found three horizons under the Piazza dei Miracoli. The top one is roughly twenty-three feet of estuarine silt and sand laid down under tidal conditions, and it is not even: on the south side of the tower's base it runs softer and more clayey than on the north. Below it sits Horizon B, whose upper member is a soft, sensitive marine clay the Pisan engineers call the Pancone; it runs to about 130 feet down before firm sand appears.

So a marble cylinder now weighing somewhere near 16,000 tons stands on a compressible wedge that is weaker on one flank. It has sunk about ten feet into that wedge over its life. It went south because the south gave first.

The trouble showed early. Roughly five years in, with three orders of arcading up and the fourth going on, the settlement was obvious enough to stop the work.

The design is credited to Bonanno Pisano, whose signature Giulia Ammannati recovered in 2019 from a broken inscription dug up beside the tower in 1838.

The hundred years that saved it

Pisa spent the century that followed at war. Genoa was the enemy at sea; Lucca and Florence were the enemies inland, and no money or attention was left over for a bell tower.

That neglect is why the tower is still there.

Soft clay under a sustained load consolidates: water is squeezed out under the weight, and what remains is denser and stronger. Ninety-four years of standing still let the Pancone do exactly that beneath a half-built tower. John Burland, the Imperial College soil mechanics professor who later directed the excavation that saved it, has said flatly that uninterrupted construction would have brought the tower down long before it reached full height. It is the largest claim anyone makes about the building, and the safety margin computed in 1993 (1.07 against toppling) makes it hard to argue with.

Work resumed in 1272 under Giovanni di Simone, who in six years carried the shaft nearly to the seventh cornice.

Why the tower is banana-shaped

Di Simone's masons could see the lean, and they corrected for it the only way open to them: cutting the marble courses taller on the low southern side, so each new story leaned back toward vertical. The bell chamber, added by Tommaso di Andrea Pisano in the 1360s, sits slightly off the axis beneath it for the same reason.

Roberto Cela, technical director at the Opera della Primaziale Pisana, puts the consequence bluntly: the tower can never be made vertical, because it is not straight.

Did Galileo drop anything off it?

Probably not.

The story that he released two unequal weights from the top between 1589 and 1592, while he held Pisa's mathematics chair, rests on one source. Vincenzo Viviani, his pupil and secretary, wrote it into a biography in 1654, twelve years after Galileo's death, and the book was not printed until 1717. Galileo never mentions it. His argument against Aristotle in Two New Sciences (1638) is reasoning rather than the report of a demonstration, and Simon Stevin had already dropped lead balls from a tower in Delft with Jan Cornets de Groot around 1586. Most historians read Viviani as embroidering; Stillman Drake was the notable holdout.

The founding date has a wrinkle of its own. Pisa began its year on 25 March and ran a year ahead of everyone else, so a document giving 9 August 1173 in modern reckoning reads 1174 in Pisan style. Vasari wrote 1174, and both years still circulate.

Every repair that made it worse

In 1838 Alessandro Della Gherardesca dug a walkway around the base, the catino, so visitors could see the plinth that had sunk into the piazza. The cut went below the water table and had to be pumped continuously. The lean accelerated.

The mid-1930s brought grouting: hundreds of boreholes around the foundation and cement pumped in by the ton. The rate of tilt went up again. Some of that work was never written up. Engineers did not discover the steel tubes tying the foundation to the concrete under the catino until the summer of 1995.

Then the worst month. In September 1995 the committee was cutting away the southern part of the catino to build a ring for ten deep anchors to pull the tower north. The tower responded by tilting south at three to four arc seconds a day. It was arrested by tipping another 400 tons of lead onto the north edge — the anchor scheme was dropped.

How the lean was taken out

The tower closed to the public on 7 January 1990, in the shadow of the Civic Tower of Pavia, which had fallen without warning the year before and killed four people. Its masonry looked a great deal like Pisa's.

The Italian prime minister appointed an international committee under Michele Jamiolkowski of the Politecnico di Torino, the seventeenth commission convened on the tower in its history.

Between May 1993 and January 1994, some 600 tons of lead ingots were stacked on a prestressed concrete ring at the north edge of the plinth. The tower came back 34 arc seconds, then another 54 over the next half year: the first movement in the right direction ever recorded.

The permanent fix was an idea the Italian engineer Fernando Terracina published in 1962 and nobody took up. Drill inclined holes under the high side, withdraw small quantities of soil, let the north edge settle into the cavity, and the whole tower rotates back. The committee rehearsed it on a 23-foot concrete footing out on the piazza and turned that nearly a thousand arc seconds.

A preliminary run from 9 February to 6 June 1999 used twelve holes and took out about 250 cubic feet; the tower rotated north 90 arc seconds and then held.

The full operation began on 21 February 2000 with 41 holes and ended on 19 January 2001. Burland and Carlo Viggiani ran it as an observational method: soil out during the day, the next morning's program written from what the instruments said overnight. Just under 1,300 cubic feet came out, about 77 tons.

The lead came off between September 2000 and January 2001. The recovery was 1,842 arc seconds, half a degree, roughly 17 inches off the overhang and a tenth of the tilt measured in 1993. The tower reopened on 15 December 2001.

What it is doing now

Still going north, very slowly.

The Surveillance Group has read the tower hourly ever since, partly with a pendulum installed in 1935. Nunziante Squeglia of the University of Pisa reported in 2018 that another inch and a half had come out of the lean since the works closed, and that the marble expands in summer, so the tower flexes with the seasons.

Jamiolkowski's committee left two scenarios on the table when it handed the monument back at the end of 2001. In the pessimistic one the tower sits quiet a while, then resumes rotating south at a much reduced rate over centuries. In the optimistic one the rotation has stopped for good. They asked for three to five years of monitoring to tell them apart.

A quarter century of data later it is still creeping the other way, an outcome nobody was willing to commit to print. The line of tourists on the lawn with their arms out, holding it up for the camera, are leaning on something that is very gradually leaning back.

Sources

• M. B. Jamiolkowski, "The leaning tower of Pisa: End of an Odyssey" (International Society for Soil Mechanics and Geotechnical Engineering, 2001), the committee chairman's own account of the 1990–2001 works
• Wikipedia, "Leaning Tower of Pisa" and "Galileo's Leaning Tower of Pisa experiment"
• Grady Hillhouse, "How Engineers Straightened the Leaning Tower of Pisa," Practical Engineering (2023)
• Agence France-Presse, "How engineers are straightening the Leaning Tower of Pisa" (2018), for Nunziante Squeglia and Roberto Cela
• Finestre sull'Arte, "We now know for sure who was the architect who designed the Tower of Pisa" (2019), on Giulia Ammannati's reading of the Bonanno inscription
• Geological Digressions, "Stabilisation of an architectural icon; the Leaning Tower of Pisa," and TheCollector, "The Irregular History of the Leaning Tower of Pisa"