§ The Viaduct That Changed the Argument
The Garabit viaduct, completed in 1884 across the Truyère gorge in the Cantal, was the structure that made Eiffel's reputation as a builder of tall, slender ironwork in exposed terrain. Its central arch spans 165 metres and carries the track deck at 122 metres above the river — figures that were, at the time of opening, unmatched in Europe for a railway bridge. Eiffel had not originated the project: his predecessor on the commission, the engineer Léon Boyer, had established the basic concept. But the execution, including the system of temporary falsework used to support the arch during erection without ever touching the water below, was entirely Eiffel's. The method demanded precise prefabrication of components at his Levallois-Perret workshops, numbered and delivered in sequence, then assembled by riveting on site. It was exactly the logic — factory-made, field-assembled, tolerance-controlled — that would produce the tower five years later.
§ The Iron Inside the Copper
In 1881, Eiffel was brought in to resolve a structural problem on a project conceived by others. The sculptor Frédéric Auguste Bartholdi had been working with the engineer Eugène-Emmanuel Viollet-le-Duc on what would become the Statue of Liberty, but Viollet-le-Duc died in 1879 before the interior structure was finalised. Eiffel replaced him and redesigned the armature entirely. His solution — a central wrought-iron pylon from which the copper skin hung on a secondary framework of flat bars — was not a conventional approach for a sculptural monument. It was an engineering structure dressed in art, built to flex: the skin was not rigidly fixed but allowed to move independently of the pylon, accommodating thermal expansion and wind load without cracking or distorting Bartholdi's form. The statue was assembled in Paris, shipped to New York and raised on its pedestal in 1886, three years before the tower commission was formally awarded.
§ Wrought Iron, Not Cast
Running through both projects, and through dozens of railway bridges Eiffel's firm constructed across France, Portugal and elsewhere in the 1860s and 1870s, was a consistent material commitment: wrought iron, worked in tension and compression through riveted lattice members rather than cast in bulk. Cast iron, dominant in earlier British railway construction, was strong in compression but brittle under tensile or shock loading. Wrought iron, refined by puddling and then by rolling, was tougher and more predictable, its behaviour under load well-understood by the time Eiffel's workshops were producing bridge components at scale. The Levallois-Perret facility, opened in the 1860s, was among the most capable fabrication yards in France, with drawing offices and workshop floor integrated so that engineers' calculations translated directly into drilled and punched components. By the mid-1880s the firm had delivered more than a hundred significant structures.
Key structures and dates
From the record| Garabit viaduct, Cantal | completed 1884; 165-metre arch, 122 metres above the Truyère |
| Statue of Liberty armature | Eiffel's redesigned interior pylon, assembled in Paris; statue raised in New York 1886 |
| Levallois-Perret workshops | opened 1860s; integrated fabrication and drawing office; source of prefabricated components for all major Eiffel commissions |
| Eiffel Tower | 18,038 prefabricated components; first sketches by Koechlin and Nouguier, 1884; construction began 1887 |
§ The Commission Was a Conclusion, Not a Bet
When the Comité des Expositions de Paris began planning the 1889 Exposition Universelle, it was Eiffel's own engineers — principally Maurice Koechlin and Émile Nouguier — who drew the first sketches of a 300-metre lattice tower in 1884, before any competition had been announced. Eiffel bought the rights to the design, refined it with architect Stephen Sauvestre, and submitted it knowing exactly what his workshops could deliver and at what tolerance. The tower's 18,038 components were prefabricated at Levallois-Perret, each one drilled to within a tenth of a millimetre; on site, only riveting was done, with a workforce that rarely exceeded 300 at any one time. The method was the same as the Garabit falsework problem in reverse: every joint designed to close correctly, first time, in the field.
The Garabit viaduct and the Liberty armature are not footnotes to the tower's story. They are its prior art — the accumulated proof that Eiffel's firm could build tall, light and true in wrought iron at a scale no rival had matched.
