On June 26, 1916, the voters of Cincinnati approved six million dollars to build a subway. The city already owned the trench it would run in. The Miami and Erie Canal had carried boats through the heart of the city since 1827. By 1904, the canal was officially abandoned and defunct.

Two hundred and seventy-four miles of hand-dug waterways, stretching from Toledo on Lake Erie south to Cincinnati on the Ohio River, now sat empty. What remained in Cincinnati was a long open cut running north to south through some of the city’s most crowded neighborhoods. Miles of excavated ground faced with old stone and brick walls, still unfilled, sat in the middle of a city desperately needing a faster way to move its people. The engineers who drew up Cincinnati’s rapid transit plan looked at that canal bed and saw the most expensive part of their work already done.

The ground was open. No streets to tear up, no rock to blast through. You only had to pour concrete walls and a roof, lay rails, and run trains. That was the plan.

The tunnels they built are still there beneath what is now Central Parkway in downtown Cincinnati. The concrete is in good condition. The station rooms remain intact. The tunnel is wide enough for two tracks side by side.

By every measure, the construction was first-rate work. No train has ever run in those tunnels. Not one. Not for a hundred years.

What happened between the vote in 1916 and the silence in that tunnel is the story of a city that arrived at the perfectly correct solution at the perfectly wrong time. A war rewrote the cost estimates before the first shovel went into the ground. An automobile revolution rewrote the politics before the money could be replenished. A depression arrived exactly when the last chance for federal help was taking shape.

All of it hit Cincinnati at a moment when the project was uniquely vulnerable. The records Cincinnati left behind tell a different story than the one usually told. The rapid transit commission reports, the construction contracts, the newspaper coverage of every proposal raised and killed in the decades after the concrete stopped. What those records show is not a city that failed to build a subway.

They show a city that built about a third of a subway in excellent concrete and then ran out of time. To understand why Cincinnati built the subway where it did, you have to start with the Miami and Erie Canal and what its construction required the first time around. The canal was not Cincinnati’s idea alone. It was Ohio’s answer to a question New York had already answered.

In 1825, the same year New York opened the Erie Canal across the state from Buffalo to Albany, the Ohio legislature authorized two canals running north to south through the interior. The Ohio and Erie Canal on the eastern side of the state, from Cleveland to Portsmouth, and the Miami and Erie Canal on the western side, from Toledo to Cincinnati. The effect of the Erie Canal on the towns it passed through was obvious within a single year. Goods moved, grain shipped, cities grew.

Ohio wanted the same thing. The Miami and Erie Canal was not built quickly. Work began in 1825 on the northern section near Toledo, and the route extended south across the flat farmlands of the Miami River valley over the following years. The terrain south of Dayton was harder.

Hills had to be cut through and rivers crossed. The canal reached Cincinnati from the north, in part, in the late 1820s, and the full route from Toledo to Cincinnati was not completed until 1845, twenty years after work began. What the canal did when complete tells you exactly what Ohio expected from it. Cincinnati in the 1820s already had a strong river trade on the Ohio.

Adding the northern canal turned the city into a transfer point on a connected waterway from the Great Lakes to the Mississippi. Goods coming from Lake Erie ports moved south. Manufactured goods and finished products from Cincinnati moved north. Farms along the Miami valley shipped grain and pork in both directions at canal rates, cheaper than wagon hauling over bad roads.

The city already called Porkopolis got far more pork to process. The problem was the railroads. The Little Miami Railroad reached Cincinnati from the east in 1843, while the canal was still being finished in its northern stretches. The Mad River and Lake Erie Railroad completed its line north to Sandusky on Lake Erie in 1846.

Two railroad competitors were operating within a year of the canal’s completion, and the pattern already seen in New York and Pennsylvania was repeating itself. Railroads did not freeze in winter. They did not dry up in drought. They did not need a mule walking beside them at three miles an hour for every foot of the route, and cars could be added or removed at will, whereas a canal boat could never be larger than the lock chambers it had to pass through.

Canal traffic through Cincinnati peaked in the early 1850s and then began a long, slow decline. The state tried to hold on. It repaired locks, dredged channels, and rebuilt sections of towpath through the 1870s and 1880s, spending money on a system the railroads had rendered uneconomic. Floods in the late 1890s destroyed large stretches of the northern reaches, and repair costs exceeded anything toll revenue could justify.

In 1904, the state made it official. The Miami and Erie Canal was abandoned. The land was transferred to the counties it ran through, and in Hamilton County, home to Cincinnati, the city found itself owning a long shallow ditch running north to south through its western industrial districts. A cut perhaps forty feet wide in places, lined with old stone walls, unfilled, sitting next to the streets and buildings that had grown up beside it.

That cut ran from Cincinnati’s northern neighborhoods, where the canal entered from farmland above the city, south through the industrial belt along Mill Creek, past the edge of the central business district. Not valuable real estate. The land beside it was working-class and industrial, warehouses and rail yards and small factories. But as a right-of-way, a pre-dug channel running exactly where the city needed something to pass, it was precisely what a transit engineer would have wished for if allowed to design his own luck.

Cincinnati’s population in 1910 was about 363,000. The fourteenth-largest city in the country and growing, but growing into a geography that made managing that growth harder every year. The city sat in a basin, with the Ohio River forming its southern boundary. A ring of steep hills closed the other three sides.

Price Hill to the west, Mount Adams and Mount Lookout and the eastern highlands beyond. Clifton and Avondale hills to the north. Every neighborhood built on those hills over the previous fifty years channeled its residents through the same few gaps and valleys to reach the commercial streets below. There were only so many ways off the heights, and all of them converged on a downtown grid planned when Cincinnati’s population was a fraction of what it had become.

The streetcar lines carrying Cincinnati workers ran on surface streets through those same bottlenecks. The main lines came down from Clifton along Vine Street, from Price Hill through the western gaps, from the eastern neighborhoods via Gilbert Avenue and Reading Road. At rush hour, those lines ran at full capacity and beyond, and they ran slowly. Streetcars shared the street with freight wagons and, by 1910, with early automobiles appearing in increasing numbers.

Surface intersections stopped the lines when freight trains crossed. Streets through the downtown grid were narrow for the traffic demands placed on them. The underlying problem was geological. The hills that gave Cincinnati its character also constrained it.

Adding more streetcar capacity in a hill corridor meant adding more vehicles to a street whose width was already fixed by buildings on either side. Widening the street meant buying and demolishing those buildings at a cost no transit revenue could justify. The Cincinnati Rapid Transit Commission, authorized by city council around 1912, identified the solution in its first detailed reports. The logic was direct.

The canal corridor ran along the western edge of the region where surface congestion was worst. A rapid transit line in that corridor could intercept riders from the northern neighborhoods before they reached the downtown bottleneck, carry them quickly along the canal route, and connect through a new downtown tunnel to an eastern line running through the Norwood Cut. The Norwood Cut was a natural valley between two ridge lines running east from downtown Cincinnati. It led to Norwood, Hyde Park, and Oakley, neighborhoods growing with exactly the middle-class population a rapid transit system most needed to attract.

A line through the Cut could serve those neighborhoods in a way streetcars could not, and could connect to the canal section via a downtown tunnel that would move passengers beneath the worst congestion without touching it. The commission’s engineers drew the full plan: the canal section for the northern and middle part of the line, a new downtown tunnel connecting it to the Mill Creek valley on the west, and a Norwood Cut extension carrying the line east toward the suburban neighborhoods. The full system as planned was roughly sixteen miles. The canal section alone was the part requiring the least new excavation and the most straightforward construction.

Start there. What the commission needed was a vote. The bond measure went before Cincinnati’s voters on June 26, 1916. The price asked was six million dollars, based on the commission’s engineering estimates for the canal section and the early stages of the downtown connection.

The campaign was not quiet. The railroads running passenger service out of Cincinnati toward the suburbs were not enthusiastic about a rapid transit competitor, and they did not stay silent. Some property owners along the canal corridor worried about what a multi-year construction project would do to their buildings, several of which sat close to the old canal walls. The business community split between commercial interests that wanted faster transit and real estate interests unsure what a subway would do to property values on streets already marginal.

But the core argument was hard to refute. Anyone who had waited twenty minutes for a streetcar at a downtown intersection during rush hour, or watched five automobiles stack up behind a freight train on Vine Street, knew the problem. The canal corridor was real. It was city-owned, and it was already in the right place.

The engineers had done serious work. The route plan was specific. The six-million-dollar figure had not come out of nowhere. It was a detailed estimate from a commission that had studied the problem for four years.

On June 26, the vote passed. Cincinnati was going to build a subway. The timing of that vote deserves a pause. The voters who said yes in June 1916 were voting in the final months before America entered a world war.

The world those voters were calculating in, with stable construction costs, available labor, and the reasonable expectation that a major public works project started in 1917 would cost roughly what it had been estimated to cost, was about to change in ways that made the six-million-dollar figure obsolete before the city spent a single dollar of it. The United States entered World War I in April 1917, ten months after the vote. The commission did not stop working. Engineering plans were refined to a level of detail sufficient for construction contracts.

Property along the canal route was acquired through negotiation and eminent domain. Specifications were written for excavation and backfill, concrete placement, tunnel dimensions, and station rooms. None of this required large quantities of steel, concrete, or field labor, and all of it was necessary before actual construction could begin. The commission spent the war years doing the work it could do, preparing for the work it intended to do once labor and materials became available again.

What the commission could not do was hold cost estimates steady. The federal mobilization drained the materials large construction projects needed. The War Industries Board directed steel production toward military use. Cement prices rose, concrete form lumber prices rose, and so did the wages that would be paid to the men doing the digging and pouring, as workers shifted into wartime industries competing for the same pool of men.

By the time of the armistice in November 1918, and Cincinnati’s return to the subway project, the engineers reviewing the original estimates line by line found numbers that no longer matched reality. The canal section alone, estimated in 1916 as the cheapest section per mile because the corridor was already open, would cost far more than the original figures. The commission recalculated. The six-million-dollar bond issue Cincinnati voters had approved in 1916, at 1916 prices, was not enough to complete the original plan in full.

It might be enough for the canal section and the early stages only. The vital section, the one most needed, if the commission started there and trusted that more money would come. That decision, start with what you have, build the most important section, and come back for the rest, is the kind of decision made in large public projects all the time. It is not an unreasonable decision.

What it requires is that the second round of funding arrive before the first section becomes a monument to an unfinished plan. Construction of the Cincinnati subway began in 1920. The method was cut-and-cover. Workers opened the old canal bed section by section, moving from north to south.

The old stone and brick canal walls still standing were removed. The floor of the canal corridor was excavated to the depth the subway design required, deeper than the canal had been, digging down into the earth below the old waterway’s bed. Rebar was placed, concrete forms were built, and walls and floor were poured in sections, each section curing before the next began. The concrete roof was the final step for each section, and once it was poured, the street above, Central Parkway, was completed as a wide boulevard built over the subway.

Photographs from those years show the trench in its various stages. Old canal walls being removed, excavation into dark soil and clay, rebar rising from the formed floor before pouring, wooden forms holding the shape of walls. In some photos you can see the transition between what the canal had been, shallow and open and framed by old masonry, and what it was becoming, a deep engineered concrete structure with the proportions of a real subway. The tunnel when complete was wide enough for two tracks side by side.

The interior clear height was sufficient for the rail cars the commission had specified: electric cars of the type then operating in Boston, Philadelphia, and New York. The walls were thick, designed not only to carry the structural load of the roadway above but to support the potential loads of whatever future development might be built over the line. The commission’s engineers designed this to be a permanent piece of infrastructure. Station rooms were formed into the sides of the tunnel at each planned stop.

Each one was larger than the running tunnel, wider with an arched ceiling above the platform areas. Stair openings were roughed in, rising toward the surface of Central Parkway above. Platform lengths were calculated to fit the train lengths the commission expected to operate. Looking at photographs of those rooms during construction, you see space designed with the expectation that passengers would stand in it.

These dimensions are not temporary. They are serious. The contractors who took on the sections were regional firms with experience in massive concrete work. Their names appear in the commission’s contract records for each tunnel section.

Supervision came from the commission’s own engineers on site, enforcing the specifications written before construction began. By any standard of what construction crews do when given a good design and adequate supervision, the work was done well. The workforce doing the digging and pouring was Cincinnati’s construction labor pool of the early 1920s. Immigrant labor formed a large part of it, Italian and Eastern European workers who had come to the city in the immigration waves that filled American industrial cities in the two decades before the war.

Their names appear in the commission’s accident records, though those records are sparse, and in occasional newspaper items when one of them was injured seriously enough to merit a mention. The detailed record of who these men were, where they slept, what they were paid, what this work cost them in health and years, is the kind of record that rarely survived that period intact. The engineering records are excellent. The human records are thin.

Between 1920 and 1925, construction crews completed about two and a half miles of tunnel. The section ran from roughly Hopple Street at the northern end, where the canal corridor narrows toward the edge of downtown, south beneath Central Parkway to the vicinity of Liberty Street. Along that distance, several station rooms were formed and roughed in. The concrete was poured on the cold-weather schedule the construction season permitted, joint by joint, section by section, the same way road crews poured concrete on Route 66 across the country in those same years.

By 1925, the tunnel under Central Parkway was as complete as it was going to get. In 1925, the last concrete was poured. The six-million-dollar bond issue was spent in full. The existing portion, two and a half miles of excellent tunnel with station rooms built and concrete walls and roof in good condition, represented about a third of the full system the commission had designed.

The downtown connection, which was supposed to take the line west to meet the Mill Creek valley and complete the circuit through the central business district, had not been started. The Norwood extension, which was supposed to carry the line east through the ridge valley to the neighborhoods generating much of Cincinnati’s ridership, had not been started. What existed was a fragment, the best-built part, the most important part to start with, but without any working connection at either end. To run trains on what had been built, Cincinnati needed to go back to its voters for more money.

The commission prepared a proposal. The new estimate for completing the system was far larger than the 1916 figure for the entire project had been. The combination of rising construction costs and the fact that the original bond had been exhausted without completing the original scope meant the commission laid out the route, the costs, and the expected ridership, and asked voters to fund the next stage. The voters said no.

That vote deserves a pause. The tunnel under Central Parkway was real. The concrete was actually there. The station rooms had been formed.

Everyone with business in the neighborhoods along that route passed over it, and many had watched the construction from the street. But the part of the system that would have made those tunnels carry passengers, the connections and the eastern extension and operating equipment, existed only on drawings. Those drawings asked for money at a moment when transit politics had changed. Between 1916 and the mid-1920s, Ohio’s automobile registrations nearly doubled.

The share of Cincinnati’s population that owned cars, and was therefore less personally dependent on the public transit system, increased every year. The original subway voters had made rational calculations in 1916 about a transit problem they experienced daily in crowded streetcars. Some of those voters had since found a different solution to the problem. The coalition that passed the 1916 bonds was no longer the same coalition in 1925.

The commission tried again in later years. Proposals appeared in the late 1920s with revised estimates and adjusted route plans, attempts to find a version of the project that could attract a majority. None succeeded. There were discussions about using early New Deal money once it became available after 1933.

Arguments that completing the subway was exactly the kind of sustainable, labor-intensive infrastructure project those programs were created to fund. The money went elsewhere. Then in October 1929, the stock market collapsed, and conversations about large public spending became very difficult everywhere. What the Depression did to Cincinnati’s subway tunnel was what it did to unfinished public projects across the country.

It made completion politically and economically impossible, and it turned the structure from a construction site waiting for continuation into something else: a piece of completed infrastructure with no function. The city found uses for it. The tunnel was dry. It held a constant temperature year-round, cool in summer, not freezing in winter.

It was structurally sound and accessible from several points along Central Parkway. Parts of it were used for storage. Municipal records from the 1930s reference the tunnel in the context of various city departments finding space for equipment and supplies. That is not what a structure built to move electric trains swiftly through a dense city was designed to do, but it was better than letting the concrete deteriorate from disuse.

The argument for completing the subway did not disappear during the Depression. If anything, the logic of using available federal money for labor-intensive public works made the transit argument stronger, not weaker. The commission and its successors made the case repeatedly. The WPA and the PWA funded roads, bridges, and public buildings across the country.

They could have funded a subway extension in Cincinnati. Part of what defeated that argument was exactly what was funding the roads. The federal highway program of the New Deal era was producing paved roads at a pace that changed the practical transportation options available to American families. A family in a Cincinnati suburb that relied entirely on streetcars in 1916 had different calculations in 1935, when the roads leading to those suburbs were paved and a used car was financially within reach.

A subway serving the canal corridor became a less pressing need for residents adapting to automobile use. The commission filed more reports. The city’s population kept growing. The tunnel remained unused.

The war years and the decades that followed found a use for the Cincinnati tunnel that its builders never imagined. The city’s civil defense planners designated the tunnel as a potential emergency shelter. Deep enough, structurally sound, and large enough to hold a large number of people in the event of an enemy attack. The tunnels were stocked with supplies, marked, and documented in the city’s civil defense plans.

No attack came. The Cincinnati tunnel kept its stocks untouched through the war years, and when the fighting stopped, the city returned to asking what should become of a two-and-a-half-mile subway route in excellent condition. The postwar period brought new transit studies, as it did to most American cities. Cincinnati’s population was spreading into the suburbs, and the downtown congestion that had prompted the original subway proposal was changing its character rather than diminishing.

Streetcar bottlenecks decreased. Automobile traffic pouring into a street grid not designed for that volume increased. The transit argument updated. A rapid transit line could bring suburban commuters downtown without adding to the automobile congestion that had become the dominant problem.

The Norwood Cut route still existed. The density of the neighborhoods around it was real. The existing tunnel still pointed in the right direction. Proposal after proposal went into study and official reports, then came out without producing a single foot of construction.

The Interstate Highway Act of 1956 ended that round of proposals more decisively than anything else. The law funded 90 percent of the cost of building a national network of limited-access highways from federal revenue, and Cincinnati’s planners and politicians, like their counterparts in every other American city, turned their attention to routing highways instead. The I-75 corridor along Mill Creek and the I-71 corridor to the northeast were planned, studied, debated, and built. Highway construction consumed the institutional attention and political capital that might have gone to extending the subway.

When the highways opened, they reinforced the automobile’s position as the primary mode of transport for the suburban residents who were the presumed ridership base of a subway extension. Among the proposals that surfaced and faded in the 1970s, the Southwest Ohio Regional Transit Authority produced a rail transit study that put the canal corridor at the heart of a comprehensive plan. Use the existing tunnel for the middle section, extend east through the Norwood Cut, extend north toward the University of Cincinnati neighborhoods of Clifton. The study was serious engineering work.

Federal transit money available through the Urban Mass Transportation Act was theoretically available for such a project. What was required was a local matching share, and raising that share required the approval of the same Hamilton County voters who had been saying no to transit expansions for fifty years. The proposal was not completed. In the late 1990s, Cincinnati watched other mid-sized American cities build light rail systems.

Portland opened its MAX lines. Denver was expanding from its downtown. Dallas was growing a network that attracted actual ridership. The Cincinnati regional rail project, and Hamilton County’s, gathered enough momentum to reach the environmental review stage.

Engineering studies were commissioned. Route lines were drawn. The project reached a point where the people working on it felt the weight of a real possibility. On March 19, 2002, Hamilton County voters rejected the sales tax increase that would have funded regional transit improvements.

The margin was not large. Opponents made their argument about costs and whether projected ridership justified the investment. Supporters made their argument about congestion and about a region watching other cities invest in transit while it stayed in place. The opponents won.

The 2002 vote was the last time Cincinnati came close to building on what had been started in 1920. Since then, the conversation has continued in a quieter register. Every new planning cycle runs its study and runs into the same questions. Where would the system go beyond the existing tunnel?

Who would pay for it? Is the density along the canal corridor and the Norwood Cut enough to justify operating costs once built? Without reaching a different answer. The tunnel beneath Central Parkway is about a hundred years old.

By all accounts, it is in good structural condition. The concrete walls have held. The bore is intact. The station rooms remain, the platform areas, the arched ceilings, the rough timber stair shafts, just as the construction crews left them.

The city has opened the tunnel for public tours on a few occasions. People who have walked through the bore describe a space that seems strangely complete for something that never fulfilled its purpose. The station rooms have the dimensions of subway stations because they were designed as subway stations by engineers who expected them to be used as subway stations. The ceiling height is appropriate for what was planned.

The wall surfaces are smooth where the concrete forms left them. No deterioration suggesting neglect, no evidence of failure in the structure itself. The tunnel reads, by most accounts, as something waiting rather than something that gave up. There have been proposals to convert parts of it to other uses.

Wine storage, emergency shelter, space for a data center, bicycle paths. Some of these ideas were studied. None came to full fruition. The tunnel remains, for the most part, as it has been since 1925, a piece of completed infrastructure lying in darkness beneath a street, holding all the air that has settled in it since the construction crews left.

The men who built it. I would give a great deal to know who was on the last crew that poured concrete in that tunnel in 1925. None of the records I found mention them. They were construction workers, men who showed up and did hard physical labor in a hole under a street and then went home.

They built a structure they expected to hand over to transit workers to serve the city’s passengers for generations. They did not know in 1925 that they were finishing something that would never be used for its intended purpose. They thought they were building a subway. There is no marker on Central Parkway at the point where the work stopped, no monument to the construction workers.

The names of the contractors who supervised the work appear in the commission’s contract records. But those records are not visible to anyone driving over the tunnel today. The workers and supervisors who poured that concrete are long gone, and the tunnel they left behind is the only evidence the work was done. The Jericho Gap district in the Texas Panhandle, the last dirt stretch on Route 66, got a bypass and then a celebration in Oldham County when the final mile of the highway was fully paved in 1938.

The workers who closed that gap got the recognition that comes from completing something. The concrete workers in Cincinnati in 1925 never got that. The celebration was supposed to come when the first train ran. The first train never ran, so the celebration never came.

And the men who did that concrete work went on through the following years without recognition that what they built had made a difference. The question people usually ask about the Cincinnati subway is whether it will ever be completed. That is certainly the wrong question. The answer is no, probably, not because of anything specific to Cincinnati but because of the cost of building new subway tunnel in the United States in the twenty-first century.

Costs have reached numbers that make financing subway construction effectively impossible except in the largest, highest-density cities. Cincinnati’s density is not New York’s. The moment when completing what was built would have required a relatively modest additional investment was the late 1920s. That moment is gone.

The right question is: what is this tunnel? In 1916, Cincinnati’s engineers and voters identified a real problem. A city whose geography created traffic bottlenecks with no obvious surface solution. They arrived at a real solution.

The solution was correct. The canal corridor was the right place for a rapid transit line. Cut-and-cover was the right construction technique for that corridor. The station designs fit the city’s actual needs.

The Norwood Cut extension would have served the right neighborhoods. Every part of the plan that was actually carried out was built correctly. What defeated it was not poor engineering. It was a collision between wartime inflation and a good-faith cost estimate, followed by a second collision with the political consequences of the automobile’s spread, then a third with the Great Depression, and a fourth with the Interstate era.

Each of those collisions arrived at the moment when the project was most vulnerable. If the war had ended two years earlier, if the automobile’s political dominance had been slower to form, if New Deal money had been directed differently, if the 2002 referendum had gone the other way. But none of those things happened. And so Cincinnati has a tunnel that is simultaneously a demonstration of competence and a symbol of bad timing.

The engineers did everything right. The voters who approved the funding were not wrong to do so. The construction workers who poured the concrete delivered first-rate work. The city that resulted from all that effort has an excellent tunnel and no trains to put in it.

The Miami and Erie Canal opened in full in 1845, lasted about fifty years under heavy use, a further decade of decline, and was officially abandoned in 1904. Tunnel construction began in its bed in 1920 and was completed, to the extent it was completed, in 1925. The tunnel is now approaching its centenary. There are not many places in America where you can stand above a major piece of infrastructure built to professional specifications by expert crews, and know that it has never once served the purpose for which it was designed.

The Brooklyn Bridge has carried traffic since 1883. Hoover Dam has generated power since 1936. The Golden Gate Bridge has carried cars since 1937. Most things that get built get used.

The Cincinnati tunnel belongs to the minority. Built by people who expected trains to come, the trains never came. The concrete is still there, quiet and dry beneath Central Parkway. If you stand at the intersection of Central Parkway and Race Street on a Tuesday morning and watch the cars stack up at the light, you are standing above a tunnel built specifically to carry people through that exact congestion.

It is fifty feet beneath you. It has been there for a hundred years, and it has never carried a single passenger. That is how Cincinnati built the first subway no one ever rode. A canal made obsolete by railroads became, in its abandonment, the perfect corridor for a faster transit system.

A city suffering a real congestion problem voted the money to build it. A world war turned every cost estimate into a document from another age before the digging began. Then came the automobile, the Great Depression, the Interstates, and ballot proposals that came close enough to succeed to prove the conversation was serious, and stayed far enough away to keep the trains out for a hundred years.