When Neutrality Lost

So the first major legislative battle over the architecture of the internet is over, and the infrastructure owners won round one. They almost always do. Name the pattern: whoever owns the pipe gets first claim on what flows through it, and they get to argue — with a straight face and a building full of lobbyists — that they earned the right to charge a toll. The argument is old. It is older than the internet, older than the telephone. It is, in fact, the single most reliable recurrence in the history of communication infrastructure, and we are watching it execute again in real time.

i · the common carrier question

Here is the recursion, laid out in strata.

In the 1870s and 1880s, the railroads owned the only fast path between the farm and the market, and they used it the way owners of bottlenecks always do: they charged different shippers different rates, favored the powerful, and crushed the rest. The public response, after years of documented abuse, was the Interstate Commerce Act of 1887 and a legal doctrine borrowed from English common law — common carriage. If you operate the infrastructure everyone depends on, you must serve all comers on equal terms. You can own the rails. You cannot decide who gets to ride.

The telegraph ran the same play. Western Union, controlling the wires, cut exclusive deals with the Associated Press and throttled rival news services — an information monopoly choosing which information moved. The telephone inherited the doctrine the railroads had earned: the Communications Act of 1934 made the phone network a Title II common carrier, legally forbidden from discriminating among the calls it carried. Your phone company could not listen to your call, decide it disliked the contents, and slow it down. That was the settled architecture for seventy years.

The internet rode into American homes over exactly those common-carrier phone lines. Then, quietly, the legal ground was moved out from under it. In 2002 the FCC classified cable-modem broadband as an "information service" rather than a "telecommunications service" — Title I, not Title II — which is bureaucratic language for: the common-carrier rules do not apply here. In June 2005 the Supreme Court blessed the maneuver in the Brand X decision, ruling that the FCC was free to make that call. Within a year, the agency extended the same reclassification to telephone-company DSL. The pipes that had been common carriers since 1934 were, by administrative reclassification, no longer.

It is worth naming who did the clearing, because the COPE Act is their sequel. The Telecommunications Act of 1996 promised to unleash competition by forcing the regional Bell companies to share their lines. Instead it triggered the largest merger wave in the industry's history: the seven Baby Bells born from the 1984 breakup of AT&T reconsolidated, through the late 1990s and early 2000s, back into a handful of giants — and one of them simply reclaimed the AT&T name. The market that was supposed to discipline the owners through competition instead handed them scale. By 2006, the companies arguing they should be free to set tolls are, structurally, the same monopoly the 1934 Act was written to constrain, wearing a newer logo. Deregulation promised competition and delivered consolidation. That is not a bug in the sequence. It is the sequence.

That is the substrate beneath this week's vote. The 2006 fight is not happening in a vacuum. It is happening on ground that was deliberately cleared in 2002 and ratified in 2005, so that by the time Markey's amendment reached the floor, restoring nondiscrimination looked like adding a new regulation rather than what it actually was: preserving a hundred-and-twenty-year-old principle that infrastructure owners do not get to pick winners.

ii · who built the pipes

The owners have a story, and it is a good one. They tell it well because they have been telling versions of it since the railroad barons.

In November 2005, SBC's chief executive Ed Whitacre — soon to run the reconstituted AT&T — put it with rare honesty to a magazine: the internet companies, he said, want to "use my pipes free, and I ain't going to let them do that." There it is, uncut. The toll-road thesis, stated by a man who genuinely believes the road belongs to whoever paved it.

It is a seductive argument because it wraps a real fact around a structural lie. The real fact: telecom companies did spend enormous capital building physical networks. The structural lie: that this investment entitles them to govern the traffic. The pipes have value only because of what runs through them. A network with no open destination is a hose pointed at a wall. Every dollar of that infrastructure was justified, financed, and made profitable by the promise that a user could reach anything — that the network was a commons with on-ramps, not a mall with a leasing office. The owners want to monetize the openness that created the demand that justified the pipe. It is the enclosure of a commons, dressed as a defense of property.

Follow the leverage and the rhetoric falls away. The telecoms are not seeking the freedom to build; nobody is stopping them from building. They are seeking the freedom to extract — to charge the websites on one end for the privilege of reaching the users on the other end, who have already paid for access. Two-sided rent on a connection sold as one-sided. That is the prize, and the COPE Act, stripped of the Markey amendment, leaves the gate to it standing open.

And notice who lines up where. On the toll-road side: a handful of infrastructure owners with concentrated interests and deep, decades-old relationships with the House Energy and Commerce Committee. On the other: a diffuse coalition of everyone else — the website builders, the public-interest groups, the users who will never organize as efficiently as four phone companies. This is the textbook geometry of regulatory capture. Concentrated benefits, dispersed costs. The party with the narrow, intense interest beats the party with the broad, mild one almost every time, in the first round.

iii · the trajectory from here

Round one to the owners is not the surprise. It is the baseline. The interesting question is what the pattern predicts next, and the historical record is unusually clear on this point.

The railroad fight did not end with the railroads' early wins; it ended, decades later, with common carriage imposed by a public that finally noticed it was being fleeced. The doctrine took years to mature and required repeated, visible abuse to motivate. Expect the same arc here. The COPE Act will likely stall in the Senate this year — the upper chamber is its own swamp of competing telecom interests — which means the legislative question stays unsettled and the fight migrates to the venue where these things actually get decided over the long run: the FCC and the federal courts.

That is the durable prediction. Because broadband now sits in Title I, every future attempt to enforce neutrality will have to be built on shaky legal ground — an agency reaching to regulate a service it formally declined to classify as a common carrier. Expect a decade or more of exactly this: an FCC that swings with each administration, rules written and struck down and rewritten, court challenges hinging on the 2005 reclassification that almost nobody noticed when it happened. The architecture decided in 2002 will haunt every round to come. Whoever controls the classification controls the war.

The deeper reading is that infrastructure is destiny. The shape of the pipe constrains everything that can flow through it. We are deciding, in a committee room, with a procedural vote on an amendment most people will never hear about, whether the defining network of the next century is a commons or a toll road. The owners understand exactly what is at stake, which is why they fight the boring procedural battles with such intensity. The public does not, yet — which is why the owners keep winning the early rounds.

They are calling this a deregulation that will spur competition and investment. It is the railroad rate-discrimination fight, the telegraph monopoly fight, the Title II fight — the same enclosure, fourth or fifth time around, in a new medium. The names change. The mechanism is fixed: own the bottleneck, claim the toll is fair, win the first round before the public learns the vocabulary.

Mark the date. June 8, 2006 is when the open internet stopped being the default and became something that would have to be fought for, round after round, on ground that had already been chosen by the other side.