Throughout the debate over the Stop Online Piracy Act (SOPA) and the Protect IP Act (PIPA), there has been one refrain commonly invoked by the pro-copyright, anti-internet crowd. “There is no first amendment right to infringe someone’s copyright.” Or “copyright and free speech do not conflict.” Leaving aside the obvious fact that the Supreme Court has repeatedly stated that there IS in fact a first amendment right to make unauthorized use of copyrighted material under the fair use doctrine, the idea that copyright and free speech do not conflict is demonstrably false.

At its heart, copyright law is a censorship mechanism. It is a way to grant one person or entity a monopoly on certain speech and deny all but those declared to be the “owners” of that speech the right to use it for their own expression. We believe this is justified because it provides financial incentives for speech, and indeed it seems reasonable that people should be entitled to benefit from their creative works. But we must always remember that, at its core, copyright is nothing less than the government dictating who may speak and who may not. Given this, there will ALWAYS be some degree of tension between copyright law and freedom of speech–between the Copyright Clause of the Constitution, and the First Amendment. To say that there is no conflict between copyright and free speech is simply wrong.

The conflict between copyright and free speech can be clearly seen when copyright laws are abused and used to censor speech that does not infringe anyone’s copyright. While the pro-copyright lobby may claim otherwise, the truth is such abuses are both common and widespread. In light of bills like SOPA and PIPA which propose dramatically expanding the powers of both the federal government and private copyright owners to enforce copyrights, it is important to keep in mind the ways in which copyright is abused under existing laws, and the potential for even greater abuses should these bills pass.

Now that outright DNS blocking and the horrible “private right of action” provisions have been removed from SOPA and PIPA, one of the most dangerous remaining provisions is the section granting “immunity” to search engines, payment processors, and ad-serving networks who “voluntarily” decide to cut-off services to websites which they decide “facilitate” copyright infringement. As many people have pointed out, this will most likely result in a process similar to the existing DMCA takedown process, whereby service providers who are notified by copyright claimants that a particular site “facilitates” copyright law will immediately move to cut-off service to that site for the sake of avoiding liability, regardless of whether that site actually violates copyright law in any way.

This is exactly how every content hosting service in existence, from Google to YouTube to Facebook to file hosting sites like Megaupload and Rapidshare (the ostensible targets of SOPA) currently respond to takedown notices under the DMCA. But the “voluntary” blacklists created by private companies under SOPA would have none of the safeguards included in the legally mandated DMCA takedown process, such as the opportunity for accused infringers to file counter-notices and get their content restored. Under SOPA, while companies are encouraged to cut-off service to sites which are accused of piracy, they have no corresponding obligation to restore service if the allegation turns out to be false. The result will be the establishment of private copyright enforcement regimes administered by corporations with no accountability, no safeguards against abuse, and no mechanism for appeal.

How do I know this? Because this is exactly what has happened on YouTube, which currently runs the largest private copyright enforcement regime in existence under the guise of its “Content ID” program. As a YouTube video creator who frequently employs fair use in order to make unauthorized use of copyrighted content in YouTube videos, and because of my work advocating for a fairer copyright enforcement system on YouTube through my website, FairUseTube.org, I have been in a position to see just how often such private copyright enforcement systems are abused.

Under YouTube’s Content ID system, every video that is uploaded is automatically scanned against a vast database of copyrighted works contributed by YouTube’s Content ID partners. If either the video or audio content matches the digital fingerprint of a sample in YouTube’s database, the system applies the copyright owner’s preselected policy to either block the video outright, allow it to remain up but track its view statistics, or “monetize the video” by taking a cut of the revenue from ads embedded in the video page. While on the surface this seems like a great way for YouTube to allow users to upload videos which use copyrighted content while allowing copyright owners to still make money from their otherwise unauthorized use, the system has two fatal flaws: (1) Content ID matches are notoriously inaccurate and wide-open to abuse, and (2) there is no effective way to appeal mistaken identifications or even blatantly false and fraudulent claims.

First, once an entity is accepted as a partner in YouTube’s Content ID program,YouTube apparently does not require copyright claimants to submit any proof that they own the copyright to works which they upload as reference files. There have been numerous reported cases of unscrupulous companies submitting works that are either in the public domain or are simply not owned by them into the Content ID database.  This allows them to claim ownership of, block, and/or receive ad revenue from, videos which they do not own any copyright interest in whatsoever. Even when there is a legitimate copyright involved, the Content ID system is often unable to tell a copyrighted work from a non-copyrighted one. This problem is especially severe with regard to recordings of classical music, where the music itself is in the public domain, but specific recordings may be subject to copyright. The Content ID system cannot tell one recording of the same classical song from another, and thus people who have legally used recordings of classical symphonies that were either in the public domain or that they have legally licensed from a third-party music library (and in some cases even performed themselves) have found their videos misidentified as containing a copyrighted recording owned by someone else.

Misidentification or even fraudulent copyright claims would not be so bad if there was a means to appeal such false-positives and punish users who abuse the system. But in fact there is not. While YouTube maintains a nominal mechanism for “disputing” false or mistaken Content ID matches, this dispute system is a joke. This is because the person who gets to decide whether to accept the user’s dispute is none other than the copyright claimant himself.

When a user files a dispute, the Content ID claimant is given three options: (1) release the claim, (2) have the video taken down via a formal DMCA claim, and (3) reinstate the Content ID claim. While the first option removes the false claim entirely and the second invokes the formal DMCA takedown process under the law (allowing the user to send a DMCA counter-notice and get their video restored that way), the third option reinstates the Content ID match, allowing the claimant to either block or receive all the ad revenue from a video, with no further opportunity for the uploader to appeal. Instead, the user is greeted merely with a message that the copyright owner has “confirmed their claim” to the content. If the user attempts to contact YouTube to further appeal a false claim, they are told that their only option at this point is to convince the claimant to retract their claim.

Through this process, YouTube gives copyright claimants the ability to essentially be the judge in their own cases–giving them sole discretion whether or not to accept a dispute against their copyright claim. As experience has shown, Content ID claimants almost universally choose to “reinstate” their claims (likely through an automated process or merely clicking “select all” in the list of disputes). As a result, the Content ID dispute process is next to useless, and those who attempt to dispute a mistaken identification or claim fair use, will most likely find their video re-blocked through Content ID in a matter of days after they file their dispute, with no further recourse or opportunity for appeal.

It is important to note at this point, that everything YouTube has done is perfectly legal under current law. Nothing today prevents companies from establishing their own private systems of copyright enforcement which go far beyond the process prescribed in the DMCA, and which contain none of its safeguards against abuse. In YouTube’s case, Content ID exists alongside the DMCA process, as the DMCA provides an alternate means by which videos can be removed for copyright infringement. But while YouTube originally intended Content ID to serve as a kind of front-end buffer to the DMCA process, allowing users to have recourse to the DMCA counter-notice process in the event a Content ID dispute was rejected, that option no longer exists. Instead, Content ID has almost completely supplanted the DMCA process as the primary means of copyright enforcement on YouTube, and users who have videos permanently blocked by Content ID have no recourse to DMCA counter-notices.

It is under this context that, when last month Universal Music used the Content ID system to have an original commercial by Megaupload taken down without having any legitimate copyright claim against it, Universal could plausibly claim in court that they could use YouTube’s private system to block the video without being subject to any of the penalties for fraudulent copyright claims under the DMCA. It is this system that has allowed unscrupulous companies like GoDigital to illegally hijack ad-revenue from hundreds of original videos using legally licensed royalty-free tracks from third-party music libraries. And it is this system that has allowed others to claim a monopoly on royalty-free music loops and samples provided with popular software such as iMovie and GarageBand, effectively preventing anyone else from using them in YouTube videos. All of these are documented cases of flagrant abuse, with dozens of complaints on YouTube’s help forums that the company has systematically ignored and failed to act upon. (To read more about these specific cases, click here.)

The reason for that is simple. The current law simply provides no incentive for companies like YouTube to protect their users against false and abusive copyright claims. YouTube would much rather placate major copyright holders like Viacom and Universal Music and avoid expensive lawsuits than stand up for users’ rights by forcing copyright claimants to prove a valid copyright interest in videos alleged to be infringing, or in the very least provide an effective means to appeal false copyright claims.

While Google was one of the most vocal parties involved in yesterday’s protest against SOPA, their own system which they have built on YouTube provides a clear example of exactly what we can expect if SOPA passes. When private service providers are deputized to become enforcers of copyright law with no incentive to defend individual users, they will invariably sacrifice the free speech rights of their users for the sake of avoiding expensive lawsuits. YouTube has shown us that nothing good can come of privatizing copyright enforcement, which only serves to harm freedom of speech online.

Imagine how much worse it would be, if instead of specific content being subject to such arbitrary blocking as YouTube currently employs, entire websites could be cut-off from all financial services and revenue sources by the mere accusations of big media companies. What would happen if, instead of individual YouTube videos getting blocked, YouTube itself was de-listed from search engines and denied its life’s blood in ad revenue, with no opportunity for appeal? That is what would happen under SOPA. Let us take warning from YouTube’s own practices, before YouTube itself finds itself in the cross-hairs.