On January 14, in Verizon v. FCC, the DC Circuit Court of Appeals struck down the core of the FCC’s Open Internet Order. The Order was the agency’s attempt to codify “net neutrality,” a newish term for the old principle that communications networks should treat all information that flows through them the same. It prohibited broadband providers from blocking or unreasonably discriminating among the content, apps, and devices that make use of their high-speed networks. Within months of the Order’s release in December 2010, Verizon brought suit.

Although the decision was a major defeat for the FCC, you wouldn’t gather that from reading the first two thirds of the opinion, in which the court repeatedly rejects Verizon’s arguments and sides with the FCC. The court accepted that a minor provision in the Communications Act gave the FCC authority to regulate broadband providers, at least insofar as those regulations encouraged broadband deployment. The court then dismissed Verizon’s objection that the Order could only achieve this end indirectly, through what Verizon termed a “triple-cushion shot.” “In billiards,” the court retorted, “a triple-cushion shot, although perhaps more difficult to complete, counts the same as any other shot.” Finally, the court considered the substance of the Order, assessing its premises and conclusions and finding that they were “both rational and supported by substantial evidence.”

Then the court vacated the Order on what seemed to be purely semantic grounds. The Communications Act — which gives the FCC its power — distinguishes between “telecommunications services” and “information services,” and only the former may be regulated as “common carriers.” In various decrees issued during the 2000s, the FCC had chosen to classify broadband as an information service rather than a telecommunications service. The Order, however, by forcing broadband providers to offer their services “indiscriminately and on general terms,” regulated those providers as common carriers. By easy syllogism (no treating information services as common carriers; the Order treats information services as common carriers), the Order was dead. But this outcome was a result of the FCC’s own doing. As the court put it toward the end of the opinion, “given the manner in which the Commission has chosen to classify broadband providers, the regulations cannot stand.”

The DC Circuit’s plain logic and apparent openness to the underlying policy leaves a supporter of net neutrality not so much angry at the court as baffled by the law, and by the FCC. How could the legality of an otherwise reasonable regulation depend on this distinction, and how could the FCC end up on the wrong side of it?

The Open Internet Order focuses on three categories of internet actors: broadband providers, edge providers, and end users. “Broadband providers” (like Comcast and Verizon) and “end users” (like you and me) are straightforward terms; “edge providers” refers to people and businesses that provide the stuff — the “content, applications, services, and devices” — that end users access on the internet. Google and Facebook, BuzzFeed and your roommate (when she updates her blog) are all edge providers. The internet’s pipes, from copper telephone wires to fiber-optic cables, have traditionally been indifferent to what edge providers send through them. This indifference, termed “openness” by the FCC, has afforded edge providers an equal opportunity to reach users, encouraging competition and innovation and making it possible that diverse speech may actually get heard.

Broadband providers have reasons to interfere with this openness. First, they use their pipes not just for internet access, but for other services — specifically, pay-TV and telephony — that have substitutes in edge providers like Netflix and Skype. The more TV watchers can make do with some combination of Amazon, Hulu, and Netflix, the more likely they are to cancel their pay-TV subscription altogether. In addition, broadband providers own content that competes with that of other edge providers. A prime example is Comcast, which merged with NBCUniversal in 2011 and thus owns a discrete chunk of the content that rides along its cables. Other things equal, it’s better for Comcast if an end user watches a video owned by Universal instead of one owned by Fox. As a result of their business structures, then, broadband providers have an incentive to use their control of the network to disadvantage their competition.

Broadband providers also have the simple incentive to maximize revenue from their internet access business. One way to do this would be to start charging edge providers for prioritized access to end users, on top of the subscription fees to end users that they already charge. Websites and applications that pay (likely owned by bigger companies and incumbents) would load faster and work better than those that didn’t (likely owned by smaller businesses, nonprofits, and individuals). This kind of tiered system would tilt what was previously a level playing field in favor of those who could afford to pay, making it harder for little guys to compete.

The Order addressed these dangers straightforwardly, by prohibiting traditional broadband providers from “unreasonably” discriminating among different kinds of content and from blocking content altogether. AT&T couldn’t black out Skype. Comcast couldn’t make it hard to load CNN’s website and easy to load MSNBC’s. And BuzzFeed would have to depend on the quality of its GIFs, not the size of its coffers, to compete with your roommate’s blog. Of course, the antiblocking and antidiscrimination rules addressed politically motivated threats, too: if Verizon were brazen enough to block one political candidate’s website because it supported her opponent, that would be a clear violation of the rules.

The Order had its share of loopholes and exceptions, chief among them that the bans on blocking and discrimination were subject to a “reasonable network management” carve-out, which allowed practices like slowing down the most data-hungry end users during times of network congestion. Also, because mobile broadband was a new and “dynamic” market with characteristics different from traditional broadband, it was, for the time being, exempt from the antidiscrimination rule. These exceptions led some net neutrality advocates to label the Order weak, inadequate, a disappointment.

More recent developments have suggested the Order also left out an important category of internet player, termed “backbone networks” in the Verizon opinion. Backbone network providers are middlemen, providing long-haul connections to and among the networks of broadband providers. The Order did not attempt to regulate backbone network agreements (known as “peering agreements”), in large part because backbone markets have traditionally been competitive. Recently, however, broadband providers have taken on some of the functions of backbone networks, shifting the dynamics of this market and creating some of the same dangers that the Order is intended to address. The widely publicized peering agreement between Comcast and Netflix reached this February — under which Netflix will pay Comcast for direct access to the Comcast broadband network — is but one example. The FCC’s net neutrality regulation simply does not speak to these arrangements.

In all, though it was modest, the Order was a necessary step in preserving the open architecture of the internet. But in issuing the Order, the FCC was trying to grasp an industry it had already set outside its own reach.

Title II of the Communications Act subjects telephone companies to regulation as “common carriers,” a category historically applied to quasi-public transportation entities like railroads and ferries. Title II’s common carriage rules require telephone companies, among other things, to provide service to anyone who requests it, to have “just and reasonable” prices, and not to discriminate in terms of service among those who request it.

In the late 1960s, businesses began offering data processing services through the use of computers and phone lines, and the question of whether and how to regulate these services naturally fell to the FCC. After a decade of investigation, the FCC issued a decision called Computer II, which created a distinction between “basic services” (traditional transmission over telephone lines) and “enhanced services” (data processing services that rode over the phone lines). Basic services would remain subject to Title II regulation, while enhanced services would essentially be unregulated. The idea behind Computer II was that the market for enhanced services — unlike the market for plain old telephone service — was sufficiently competitive to ensure reasonable prices and foster innovation.

Common carrier regulation under Title II and Computer II shaped the dial-up era. The phone companies were required to provide their telephone lines on a nondiscriminatory basis to independent internet service providers (ISPs) like AOL and Compuserve, allowing a competitive market to develop. The Telecommunications Act of 1996, which overhauled the Communications Act, adopted the basic/enhanced distinction. It relabeled basic services as “telecommunications services,” enhanced services as “information services,” and gave statutory definitions to each. The definitions were convoluted, but they preserved the status quo, affirming that only telecommunications services, and not information services, could be regulated as common carriers under Title II.

As broadband gained popularity in the early 2000s, the regulatory framework surrounding internet access became unstable. In the dial-up age, two distinct services from two distinct entities were necessary to connect to the internet: a phone company provided the pipes (the telecommunications service), and an ISP provided the processing (the information service). But with broadband, the phone company — and soon enough, the cable company — began to offer both the pipes and the processing. The question arose: Was the broadband service offered by these companies a unitary “information service” or a combination of a “telecommunications service” and an “information service”?

An important policy consequence lurked behind the question. Only if broadband contained a discrete telecommunications service component would it be subject to Title II Common carrier regulation. And only if broadband was subject to Title II regulation would broadband providers be required to open up their pipes to other businesses that wanted to offer internet access over them. In this way, the answer to the telecommunications/information question would determine the shape of the broadband market.

By the time the issue came to a head for cable, it was 2002, and Bush appointee Michael Powell (son of Colin) was chairman of the FCC. Although more technocrat than crony, Powell was also a self-described “Reagan-era child,” and he did not hide his deregulatory agenda. To Powell, the crucial fact about broadband was that the internet could now be accessed through all sorts of pipes. With dial-up, only the phone line could supply the connection. Now, a cable could, too — or a satellite, or even, perhaps, a power line. As a result, there was no need to create competition artificially by forcing the owner of any one kind of pipe to share it. Instead, competition would naturally arise among the different pipes. Guided by Powell’s belief in this theory of “intermodal competition,” the FCC classified cable broadband as an information service in its entirety, thus exempting it from Common carrier regulation under Title II.

The more strictly legal basis for the FCC’s decision was a close reading of the Communications Act’s statutory text — specifically, the word “offer” in the definition of “telecommunications service.” In short, according to the FCC, cable broadband’s combination of pipes and processing was perceived by customers as a single, integrated service. Because the pipes were a mere component of that service, broadband companies could not be said to “offer” those pipes, as opposed to the service as a whole, to their customers. For this reason, no part of cable broadband — not even the raw transmission of data through the pipes — qualified as a telecommunications service.

Soon after the order was released, independent ISPs that wanted access to cable companies’ pipes on a common carriage basis challenged the FCC’s classification in court. In the 2005 case NCTA v. Brand X, the Supreme Court, 6-3, rejected that challenge. Under what’s known as the Chevron deference, whereby courts are deferential to an agency’s interpretation of an ambiguity in the statute that governs it, the Court held that the FCC’s interpretation of the ambiguous statutory definitions was reasonable.

Justice Scalia dissented, relying heavily on an analogy to pizza delivery. To him, under the FCC’s interpretation of the word “offer,” a pizzeria could insist that it does not “offer” delivery, because delivery “is part and parcel of its pizzeria-pizza-at-home service.” “Any reasonable customer,” Scalia went on, “would conclude at this point that his interlocutor [the pizza guy] was either crazy or following some too-clever-by-half legal advice.” Rather, under the only reasonable interpretation of the statute, broadband must be classified as including separate and distinct telecommunications and information services. Demonstrating that statutory interpretation can cut across ideology (or perhaps just into pizza), Justices Ginsburg and Souter joined Scalia in dissent.

After Brand X, the FCC issued orders classifying the other forms of broadband internet access — DSL, wireless, “broadband over power line” — as “information services” exempt from Title II regulation, just like cable. By 2007, the broadband industry was firmly established as outside the scope of common carrier regulation under Title II.

Throughout its deregulation of broadband, the FCC maintained that its classification of broadband as an information service was not a complete abdication of authority over the industry. As support, the FCC pointed to Title I of the Communications Act. In large part an introductory and housekeeping chapter, Title I also contains a provision stating that the FCC may “make such rules and regulations . . . as may be necessary in the execution of its functions.” Courts have interpreted this provision of Title I to give the FCC “ancillary authority”: if an action would help achieve one of the FCC’s express statutory mandates, then the agency may be able to take it, even though it isn’t expressly authorized to.

The pairing of broadband regulation with ancillary authority has its roots in Michael Powell’s FCC. Soon after the 2002 decision to classify cable as an information service, academics and activists began to present the FCC’s deregulatory stance toward broadband as a threat to the internet as an open platform. In response, Powell gave a speech in February 2004 in which he “challenge[d] the broadband network industry” to preserve four “Internet Freedoms”: the “Freedom to Access Content,” the “Freedom to Use Applications,” the “Freedom to Attach Personal Devices,” and the “Freedom to Obtain Service Plan Information.” Justified as a way to encourage innovation and as an “insurance policy” against potential market abuses, Powell’s speech provided the framework for all net neutrality regulation that would follow. To Powell himself, however, “Net Freedom” was a private matter: “Based on what we currently know,” he said, “the case for government-imposed regulations regarding the use or provision of broadband content, applications and devices is unconvincing and speculative.”

In 2005, Kevin Martin, Bush’s appointed successor to Powell as FCC chairman, took the first step in turning Powell’s freedoms into rules. In the Internet Policy Statement, issued the same day as an order deregulating broadband over phone lines, the FCC adopted four principles “to ensure that broadband networks are widely deployed, open, affordable, and accessible to all consumers.” The principles were largely identical to Powell’s freedoms, stating that consumers were “entitled” to access any lawful content, run whatever applications, and connect to any legal device over broadband networks. Crucially, the Internet Policy Statement claimed legitimacy by reference to Title I ancillary authority, citing a one-off line in the Supreme Court’s recent Brand X decision as support. As a result of its ancillary authority, the FCC insisted, “the Commission has jurisdiction necessary to ensure that providers of telecommunications for Internet access . . . are operated in a neutral manner.”

Some critics believed that the Internet Policy Statement, issued in conjunction with a broad deregulatory decree, wouldn’t actually be used as an enforcement tool. Those critics turned out to be wrong. In 2007, the Electronic Frontier Foundation and an engineer at Intel discovered that Comcast was selectively interfering with internet traffic from BitTorrent, the popular peer-to-peer application. In October, the Associated Press confirmed the conduct in nationwide tests and reported it, sparking a public controversy. Soon after, the advocacy groups Free Press and Public Knowledge filed a formal complaint with the FCC, arguing that Comcast’s interference violated the Internet Policy Statement. During the subsequent public comment period, the FCC received more than 20,000 comments in support of the complaint.

In August 2008, the FCC issued its decision: by discriminating against BitTorrent traffic, Comcast had run afoul of the Internet Policy Statement’s principle that internet users “are entitled to run applications and use services of their choice.” Comcast was ordered to end this practice and transition to “nondiscriminatory network management practices” by the end of the year (which it had already agreed to do); it was also required to disclose publicly the details of both its prior interference and its practices going forward.

Although Comcast complied with the order, it subsequently brought suit against the FCC, arguing that the FCC didn’t have the authority to issue the Internet Policy Statement in the first place. In its 2010 decision Comcast v. FCC, the DC Circuit agreed.

The court started with the settled notion that the exercise of “ancillary” authority is only legitimate when it’s actually ancillary to something. In other words, in order to act on the Internet Policy Statement, the FCC had to point to a specific “statutorily mandated responsibility” that the policy helped carry out.

The FCC attempted to anchor its ancillary authority in two ways, both of which failed. Its primary strategy was to emphasize a statutory provision stating Congress’s intent “to promote the continued development of the Internet” and “to preserve the vibrant and competitive free market that presently exists for the Internet.” This language is found in Section 230 of the Communications Act, an otherwise unrelated provision regarding intermediary liability. To the court, these statements were the statutory equivalent of puffery: broad statements of general intent, delegating no actual authority to the FCC and thus unable to provide the basis for an exercise of ancillary authority.

The FCC’s second strategy was to present a smorgasbord of mostly minor provisions in the Communications Act that might provide some basis for the Internet Policy Statement. The court rejected each of these provisions for a corresponding smorgasbord of reasons. Some were more substantive than others, but all were at least arguably legitimate. The best shot the FCC had was Section 706, which provides that “the Commission . . . shall encourage the deployment . . . of advanced telecommunications capability to all Americans . . . by utilizing . . . regulating methods that remove barriers to infrastructure investment.” Section 706 could be read to give the FCC affirmative regulatory power to “encourage the deployment” of broadband. But the court noted that in 1998 the FCC opined that Section 706 did not provide any independent authority. Because that 1998 order still stood, the FCC could not now point to Section 706 as a basis for ancillary authority. Having rejected all of the FCC’s arguments, the court struck down the 2008 order.

Between the 2008 Comcast order and the 2010 Comcast decision, Barack Obama became President. After stating his strong support for net neutrality on the campaign trail, Obama appointed Julius Genachowski — chief counsel to the FCC during the Clinton Administration, Silicon Valley investor, and Obama’s law school classmate — as chairman of the FCC. Although the Comcast decision was pending when he took his post in summer 2009, Genachowski made clear early on that he supported the Internet Policy Statement, and that he planned to beef it up. In October 2009, the FCC unveiled a draft of what would become the Open Internet Order, complete with prohibitions on blocking and discrimination. The proposed rule still claimed legitimacy through ancillary authority, specifically citing Section 230 (the broad statement of congressional policy) and Section 706 (the seemingly minor broadband-deployment provision). Comments on the draft were due in January; a final rule was expected by summer 2010.

Comcast threw a wrench in the FCC’s plans. Because of the posture of the case, the decision directly overruled only the 2008 order punishing Comcast for throttling BitTorrent. But the case also made clear that the way things stood, Title I ancillary authority was not a legitimate basis for the principles of the Internet Policy Statement, let alone the bona fide rules of the draft Open Internet Order.

Continuing to base the Open Internet Order in ancillary authority under Title I would require tricky legal acrobatics. But these acrobatics could be avoided if the FCC simply reclassified broadband as a “telecommunications service.” Then it could regulate the industry under its well-defined Title II powers, rather than go searching for some obscure provision that may justify a vague assertion of ancillary authority.

The legal path to reclassification was clear, and not particularly foreboding. It’s well-settled law that an agency can change its mind as long as it has a reasoned explanation for doing so, and there was a strong argument that a sufficient explanation existed here. The split opinion in Brand X demonstrated that classifying broadband as including a telecommunications service was a reasonable — and if you believe Scalia, necessary — interpretation of the Communications Act. And the prospect of intermodal competition, which had led Powell’s FCC to opt for “information” rather than “telecommunications,” had not been borne out. Even by 2010, it was clear that in the land of broadband, cable was king: DSL simply couldn’t compete with advances in cable technology, and phone companies couldn’t attract the investment necessary to build a brand-new fiber optics infrastructure.

Moreover, making broadband subject to Title II wouldn’t subject broadband providers to the full panoply of old-school phone company regulation. The Communications Act specifically requires the FCC to “forbear” from imposing otherwise mandatory rules when the circumstances dictate. Thus, the FCC, after making certain forbearance findings, could impose only the Title II rules that would achieve the ends of net neutrality. Because “un-forbearing” is harder than forbearing in the first place, it would be unlikely that the FCC could later impose the more onerous Title II rules, such as price regulation.

The legal allure of reclassification was not lost on Genachowski. In May 2010, one month after Comcast, the FCC endorsed the combination of reclassification and forbearance and opened up the legal strategy for public comment. Genachowski spun the plan as a moderate and sensible “Third Way,” between the “serious uncertainty” of continuing on the Title I track and the heavy-handedness of full-stop Title II regulation.

Then, Genachowski faltered. The cable, telecom, and wireless industries vigorously opposed the “Third Way” proposal during the summer and fall of 2010. They upped their already large lobbying expenditures, accused the FCC of trying to “regulate the internet,” and claimed that if they were regulated, they’d be forced to stop investing in infrastructure. House Democrats and Senate Republicans sent angry letters to the FCC, and the House introduced a resolution prohibiting the FCC from issuing rules regarding broadband until Congress could weigh in.

The more time passed without a rule, the more net neutrality advocates sensed that things weren’t going their way. In September, exhausted from months of traditional methods of persuasion, members of the group savetheinternet.org stood outside the FCC’s monthly meeting handing out waffles and holding up signs that read DON’T WAFFLE ON NET NEUTRALITY. The efforts were ineffective: in November 2010 alone, Genachowski’s chief of staff had at least six private meetings with AT&T, while he met no more than twice with any other major stakeholder. Finally, in a December 1, 2010 speech, Genachowski announced that the FCC had settled on a net neutrality rule. It would be based in Title I ancillary authority after all. Although he had “serious reservations” about ancillary authority six months before, Genachowski now claimed to be “satisfied that we have a sound legal basis for this approach.”

On December 21, 2010 the Order was officially adopted. In the end, the FCC’s ancillary theory centered on Section 706, the provision that allowed it to encourage broadband deployment. Addressing the problem the Comcast court identified — that in a 1998 order the FCC had said Section 706 provided no affirmative authority — the FCC now disavowed its previous order to the extent that it meant what the Comcast court had thought. And, the FCC continued, the Open Internet Order would encourage broadband deployment: the open platform would lead to innovative new content and services, which would increase demand for broadband, which would cause broadband providers to improve their infrastructure to accommodate demand.

In this January’s Verizon decision, the DC Circuit went along with much of the FCC’s argument. It accepted the FCC’s reversal regarding Section 706, and it accepted that the Order may well encourage broadband deployment. However, because broadband providers were still classified as information services, the Commission couldn’t issue rules that treated them like common carriers — regardless of those rules’ effect on the deployment of broadband. The Open Internet Order did exactly that, and so it could not stand.

After Verizon, the only part of the Open Internet Order left standing is its transparency provision, which requires broadband providers to disclose their network management practices (including whether and how they discriminate against content or devices) and their commercial terms (including any tiered pricing schemes). With the Order otherwise gutted, broadband providers are, in principle, now free to interfere with network traffic to suit their economic incentives or ideological whims.

That said, Verizon is not likely to cause dramatic changes to the broadband landscape in the immediate future. The biggest reason for this is Comcast, the nation’s largest broadband provider. In order to obtain FCC approval for its 2011 merger with NBCUniversal, the cable company agreed to abide by the Order for seven years, whether or not it was ultimately upheld in court. So for Comcast subscribers — even putting aside its planned merger with Time Warner, that’s about one in four broadband users in the US — the FCC’s version of net neutrality is in effect until 2018, no matter what.

Although not bound in the same way as Comcast, most other broadband providers have promised to maintain an “open internet”; in the wake of Verizon, they and their lobbyists issued press releases affirming their commitment in general terms. Verizon itself provides a good example:

One thing is for sure: today’s decision will not change consumers’ ability to access and use the Internet as they do now. . . . Verizon has been and remains committed to the open Internet, which provides consumers with competitive choices and unblocked access to lawful websites and content when, where, and how they want.

If you look closely at the press releases and other statements, though, you’ll see that broadband providers are committing only not to block lawful content. Time Warner will not block Netflix, even though Netflix might hurt its pay-TV business. And it’s even less likely that Time Warner would block content for political reasons, given the enormous backlash that would surely follow if the move were discovered.

On the other hand, broadband providers haven’t promised not to experiment with the pay-for-priority schemes that the Order addressed. This is where post-Verizon change is most likely to take place. As early as 2005, the CEO of AT&T expressed a desire to charge edge providers like Google and Vonage for access to end users. In the mobile broadband space, where the Order never applied, AT&T has already instituted a variation of pay-for-priority, in which websites and apps can pay AT&T so that accessing their content doesn’t count toward users’ data caps. Post-Verizon, non-Comcast providers may well experiment with similar models on traditional broadband, charging edge providers for a “fast lane” to end users. In the worst-case scenario, incumbents would take advantage of prioritized access to users while broadband providers would allow the slow lane to get ever slower, forcing smaller edge providers to pay the toll or get off the road.

Faced with these possibilities, the FCC had a variety of options as soon as Verizon came down. It could appeal, arguing that the Order is not, in fact, simply Common carrier regulation by another name. To do so would require the FCC to take a difficult dive into the body of case law that defines “common carrier.” It would also risk the reversal of Verizon’s favorable holding that under Section 706, the FCC has at least some affirmative authority to regulate broadband providers.

Alternatively, the FCC could do what it wanted to do in 2010: reclassify broadband internet access as “telecommunications service” rather than “information service” and subject it to Title II nondiscrimination requirements. In order to reclassify, the FCC would have to give its reasoned explanation, and it would have to go through some form of administrative process, including a period for public comment. The reclassification would likely be challenged in court, but given that the FCC could present good reasons for changing its mind, the legal battle would potentially be winnable.

Many net neutrality advocates rallied behind reclassification after Verizon, but whether the FCC would have the political wherewithal to reclassify was another story. Michael Powell — now president and CEO of the National Cable and Telecommunications Association (NCTA), the cable industry’s lead lobbying group — said in 2013 that any attempt to reclassify would unleash the lobbying equivalent of “World War III.” In November 2013, Genachowski stepped down from the FCC, and Obama picked Tom Wheeler replaced him. Although Wheeler is well respected even among some net neutrality advocates, he is also a longtime industry lobbyist: he has served as president of both the NCTA and the lobbying arm of the wireless industry. If Wheeler were to decide he was up for “World War III,” he would have a lot of friends on the other side.

From early on, Wheeler indicated that he was more comfortable with the FCC’s third option, which was to stay the course and experiment with the authority that Verizon did recognize. The day of the ruling, Wheeler posted a tweet (his second ever) that seemed Orwellian: “DC Circuit has affirmed @FCC’s authority to keep the internet free & open for innovation & expression. #NetNeutrality.” In subsequent days, Wheeler made clear he was referring to Verizon’s recognition of Section 706 authority. He liked the idea of using that power in a “common law” or “case-by-case” manner, intervening when the FCC finds a broadband provider’s conduct to be anticompetitive, self-dealing, or “consumer abuse.”

On February 19, Wheeler announced the FCC’s next steps. It would not appeal. And although he stated the FCC would “keep Title II authority on the table,” Wheeler appeared to see reclassification primarily as a bogeyman for broadband providers, warning that “the Commission has the ability to utilize it if warranted.” Instead, he suggested, the FCC will go as far as it can under the no-common-carriage framework that Verizon set up. Although it won’t announce specific proposed rules until late spring at the earliest, Wheeler provided a broad outline of what those rules might look like.

First, in addition to enforcing and perhaps strengthening the transparency provision, the FCC will attempt to “fulfill the ‘no-blocking’ goal.” Near the end of Verizon, the court raised the possibility that the antiblocking rule alone, without the antidiscrimination rule by its side, might not rise to the level of Common carrier regulation. More specifically, to require Verizon to provide some minimum level of service to all edge providers for free might not count as Common carrier regulation, as long as Verizon could engage in “individualized bargaining” for service above that basic level. Because the FCC didn’t raise this argument in its written defense of the Order, the court simply sketched it out, declining to consider it on the merits. The FCC may now attempt to reinstate a version of the antiblocking rule based on that sketch.

A stand-alone antiblocking rule would ensure that content is at least just barely accessible to end users. That would be a small accomplishment, especially considering that the major broadband providers have voluntarily promised not to block. The real worry is about discrimination. On that front, Wheeler plans to “fulfill the goals of the non-discrimination rule” through the case-by-case approach he’d hinted at earlier, which would involve “setting an enforceable legal standard that provides guidance and predictability to edge providers, consumers, and broadband providers alike” and then “evaluating on a case-by-case basis whether that standard is met.”

This sounds nice, but it’s unclear how the FCC can effectively assert “case-by-case” power against discriminatory conduct without falling victim to the same problem that killed the Order. Regardless of whether it’s one provider or the entire industry, the FCC is not allowed to force broadband providers to offer service “indiscriminately and on general terms.” It can’t prohibit broadband companies from charging an edge provider more for faster access to users, or from charging two edge providers different rates for the same service. In this way, whatever flexible standard the agency comes up with likely must make room for the very arrangements that most immediately threaten net neutrality’s aims, or else suffer the Order’s fate.

Conceivably, if the FCC proves incapable of achieving some semblance of net neutrality on its own, Congress could act. It could directly pass legislation establishing the same rules as the Open Internet Order (or even stronger ones), or it could amend the Communications Act to give the FCC more explicit regulatory authority over broadband providers. In fact, on February 3, House and Senate Democrats introduced a stopgap bill that would restore the Order to full effect until the FCC issues a final order reshaping the rules in light of Verizon. Whatever that bill’s fate, given the relatively low place of broadband policy on the national agenda and the immense lobbying power of the broadband industry, more lasting congressional action seems unlikely.

Then again, the death of the Stop Online Piracy Act showed that edge providers like Wikipedia and Google have the ability to galvanize the public and change the course of national policy. But in the case of SOPA, the threat was easy to understand, the enemy was familiar (millennials were born suspicious of the RIAA), and the goal was simple: kill the bill. Here, the threat is subtler, and even the leading activist organizations disagree on the best way forward. Public Knowledge and Free Press advocate reclassification, while the Electronic Frontier Foundation — long leery of internet regulation in general and the FCC in particular — thinks giving the FCC more power “would just be creating more problems than we’d solve.”

Whichever route they take, net neutrality advocates will need an argument that resonates in the gut if they want broad public support. Luckily, they have one: the internet is a resource essential to daily life, and the physical network it runs on is controlled by a few companies whose practices are limited only by the market. As with the telephone network before it, we need a public guardian to make sure that those companies operate the network in a basically reasonable and just way. We should, in other words, treat broadband providers like common carriers. As it stands, the simplest argument in favor of net neutrality is one the FCC is unwilling to make.¹

1. Postscript, April 25: On April 24, 2014, the FCC circulated a draft proposal to reinstate a watered-down Open Internet Order, which would allow broadband providers to discriminate in a “commercially reasonable” manner. “Commercially reasonable” was not an arbitrary choice of words. In the 2012 case <em>Cellco Partnership v. FCC</em>, the DC Circuit upheld another FCC regulation that was alleged to unlawfully impose common carrier requirements on mobile phone companies. The court said that because the regulation’s “commercially reasonable” standard left “substantial room for individualized bargaining and discrimination in terms,” it didn’t constitute common carriage. It’s all but certain that in the net neutrality context, the “commercially reasonable” standard will also allow for “individualized bargaining and discrimination in terms,” such as pay-for-priority and similar tiered schemes. In fact it must, lest the new standard run afoul of <em>Verizon</em>. Once the proposed rule-making is officially released, it will be open for public comment, by email and otherwise. ↩