I crossed cầu Đuống (the Duong Bridge), I was small enough to be a passenger and old enough to study the structure. What I noticed first was the climb: the road did not meet the bridge at grade but rose to it, tilting up from each bank, leveling across the deck, then tilting back down the far side, so that crossing felt less like continuing a road than like going over something. Then the lattice — riveted steel, painted the dull oxide red that all these bridges wear, the truss members crossing and recrossing in a pattern dense enough to read as ornament if you did not know it was load. And the width: narrower than I expected, the roadway and the rail line sharing a corridor that felt rationed.
It looked like cầu Long Biên (the Long Bien Bridge of Hanoi) scaled down — the same truss idiom, the same red, but compressed, as though the longer bridge had produced a smaller version of the same argument. I said something like this to my uncle. He corrected the lineage before he corrected me: it was my granduncle, not my uncle, who knew the bridge from the start, who had lived in the years when the first structure still stood. The resemblance to Long Bien was not imitation, he said. The two were the same age — built the same year, by the same administration, for the same corridor running northeast out of Hà Nội (Hanoi). And then the sentence I had no place to put: ngày xưa nó biết quay — back then, it could turn.
I had no category for this. The bridge I had just crossed was a fixed thing on fixed piers, a static truss that did exactly one job, which was to stay where it was. That it had once been a different machine — one that pivoted on a central axis to open the river to large vessels — was a capability I could not map onto anything I had seen. A bridge that moved belonged to the same shelf as a building that walked. I filed it, provisionally, as something my granduncle might have misremembered.
He had not. In 1902, the Direction des Travaux Publics de l'Indochine (the Indochina Public Works Department) completed a steel truss crossing over sông Đuống (the Duong River), carrying rail and road on the route toward Bắc Ninh (a province northeast of Hanoi), Lạng Sơn (the border province), and Thái Nguyên (the northern industrial region). The structure rested on five masonry piers driven deep into the riverbed. At its center sat the feature that made it more than a span: nhịp quay (the rotating span), a section engineered to swing on a central pivot, rotating clear of the channel so that river traffic too tall or too wide for the fixed clearance could pass through the gap where the bridge had been. The bridge was designed to open. This was not a flourish. It was the engineering argument for the structure's entire form — the reason for the central pier's particular mass, the reason the deck was divided where it was, the reason the whole assembly read as two bridges meeting at a turntable rather than one continuous line.
A bridge that opens is a bridge in negotiation with the thing it crosses. The fixed bridge declares the river subordinate: traffic above, water below, the hierarchy settled in concrete. The swing bridge concedes that the river has its own traffic, its own scale, its own claim on the channel, and it builds a mechanism to yield — periodically, on schedule, with the deliberate slowness of something heavy turning. To pivot a span is to admit the river was there first.
War ended the argument. American bombs brought the 1902 structure down; the steel that had been taught to turn went into the river as scrap, and the masonry piers were left standing or half-standing in the water, five stumps marking where a bridge had reasoned with a river.
What happened next is the part that the name conceals. In 1981 the crossing was rebuilt — but "rebuilt" overstates the continuity. The new bridge was a wholly new structure, a fixed steel truss assembled from Chinese steel, carrying its deck from bank to bank without any provision to open. The nhịp quay was not restored, not welded shut, not locked and abandoned. It was simply never part of the new design; there was nothing of the rotating mechanism to preserve or suppress, because the structure that replaced it was conceived from the start as a thing that stays still. Of the 1902 bridge, the new one kept only the foundations: three of the original piers carried forward, the other two abandoned, and on those three the entire structure above was new. Everything you could see was 1981. Everything you could touch from the deck was 1981. What remained of 1902 was underwater, load-bearing, and invisible — three masonry stumps doing the one job that had outlasted the bridge they were built for.
This is the seam, and it is almost impossible to see. The piers do not announce their age. The deck does not flag the year it was poured. The traffic crosses without pausing, treating the 1902 foundations and the 1981 truss as one continuous fact, one bridge, one name. The approaches still tilt up from each bank and back down the other side — but that grade, which I had read as a quirk, is itself a ghost: it was sized for a deck that had to clear a swinging span, and the slope was kept even after the reason for it was thrown into the river. You climb to the level of a mechanism that no longer exists.
By mid-2026, the substitution completes itself again. Two new structures will replace the 1981 crossing: a cable-stayed road bridge and a steel-framed railway bridge, the latter built with a truss geometry that quotes the 1902 original without repeating it — an echo engineered on purpose, a reference to a bridge that almost no one alive remembers as anything but the static thing it became. The three-pier structure will be demolished. The 1902 foundations, having outlived two bridges, will finally be retired. And the name cầu Đuống will transfer, intact, to both new crossings, as it transferred in 1981, as if a name were a property of the location rather than the structure — which, it turns out, is exactly what it is.
My granduncle was describing an absence that predated me by decades. When he said the bridge could turn, he was not claiming it still could; the mechanism had been gone since before I was born, gone so completely that there was not even a scar to point at. I was surprised anyway, and I have spent a long time trying to locate why. It was not the lost machine itself. It was the discovery that the bridge I had filed as smaller, lesser, derivative — Long Biên in miniature — had been built in the same year, by the same hands, to answer a harder version of the same question: a river that demanded the center of the span be able to move. What I had read as diminution was lineage. What I had read as one continuous bridge was a sequence of substitutions resting on three stone stumps. The continuity I trusted was not in the steel, which had been replaced, nor in the form, which had been simplified, nor in the function, which had been abandoned. It was in the foundations, and in the name — and the name, it turns out, will outlast the foundations too.
The nhịp quay is not coming back. The bridge rebuilt without it is being demolished. The piers that carried both are being retired. The name will outlast all of them.
what is the only part of the bridge that was never built?
June 2026
Bi is a Hanoi-based writer documenting Vietnam's urban textures and cultural margins. About the author.
Comments
Post a Comment