Another Bad Ruling for PissedConsumer on Trademark and 47 USC 230 Claims–Amerigas v. Opinion Corp.

By Eric Goldman

Amerigas Propane, LP v. Opinion Corp., 2012 WL 2327788 (E.D. Pa. June 19, 2012)

You may recall PissedConsumer, the site that solicits user gripes, SEOs the crud out of them, and then offers the griped business pay-to-play to downgrade the visibility of those reviews. Its litigation docket is heating up, and I’ve recently blogged on two prior rulings (Ascentive and Vo).

In a thoughtful but perhaps overly cautious opinion by Judge Buckwalter (a celebrated name in cyberlaw for his pro-free speech 1990s opinions in the CDA and COPA lawsuits), PissedConsumer couldn’t get a motion to dismiss on Amerigas’ trademark claims, and it couldn’t get dismissal of other claims on 47 USC 230 grounds. The judge signals that Amerigas is likely to lose its claims later, but it gives both parties the opportunity to lavish money on their lawyers before the judge reaches the obvious denouement.

Trademark Claims

PissedConsumer makes a number of references to the trademarks of the businesses it reviews, including using the trademark in as a third level domain (i.e., a subdomain), including the trademark in its metatags and showing keyword-triggered ads. In its motion to dismiss, PissedConsumer took a number of whacks at the trademark claims:

Trademark use. The court says Amerigas properly alleged trademark use because “the Complaint alleges that Defendant uses Plaintiff’s “AMERIGAS” trademark in its website’s text, subdomain name, and metadata in connection with the sale of advertisements to Plaintiff’s competitors” and “for a fee, Defendant will allow businesses to address the complaints posted about them on the website.”

I think this is a correct synthesis of current precedent (especially after Rescuecom), but it misses the broader context that we’re talking about a review site selling ads to support its editorial content. I hammer this point in my Online Word of Mouth paper, where I argue that a more aggressive policing of the trademark use doctrine would create necessary breathing room for online word of mouth. Sadly, post-Rescuecom, we’ve seen little judicial interest in tightening the trademark use doctrine, creating the risk that review websites have to work much harder to fend off trademark attacks.

Nominative Fair Use. Unlike some of my trademark scholar peers, I’m not a big fan of nominative fair use as a protection for review websites because it’s so hard to win on motions to dismiss (and isn’t guaranteed on summary judgment, either). This case illustrates that point nicely. The judge says that it would be “premature” to rule on nominative fair use on a motion to dismiss. The judge wants to see more about the likelihood of consumer confusion; and while it’s likely a “gripe site” (in the court’s words) would prove that it had to use the plaintiff’s trademarks in order to refer to the plaintiff, the court wants to see more about whether PissedConsumer took only what was necessary and accurately reflected the relationship between Amerigas and PissedConsumer.

Likelihood of Consumer Confusion. PissedConsumer won the Ascentive case in part for lack of consumer confusion. However, that was in response to Ascentive’s preliminary injunction motion (when the burden is higher for the TM owner), not on a motion to dismiss. As a result, the judge says that he needs to give the plaintiff the opportunity for discovery. Even so, the judge says the Ascentive opinion was “well-written and very persuasive” and the judge expresses skepticism about Amerigas’ chances of success:

This Court also finds it difficult to believe that a person searching online for Plaintiff’s products would be confused into thinking that a site called “PissedConsumer”—which contains an abundance negative reviews, offers no propane products of its own, and links to the websites of Plaintiff’s competitors—was in some way affiliated with or endorsed by Plaintiff

And later the judge says “Plaintiff may face a tall task in demonstrating that Defendant’s use of its mark is confusing.”

Based on the judge’s statements plus the Ascentive precedent, it seems pretty clear that Amerigas isn’t going to win the trademark claim. This is exactly why a defense or doctrinal limit that succeeds on a motion to dismiss is so important. The judge has basically ordered the parties to spend tens or hundreds of thousands of dollars on discovery, all with the strong possibility of reaching the conclusion that Amerigas isn’t going to win–a conclusion that’s pretty clear right now. I’m sorry to keep rehashing the trademark use debate, but the trademark use doctrine could have given the judge an easy way of saving everyone a lot of time, money and heartache.

Along the way, the judge makes an odd statement: “If the unhappy purchasers of Plaintiff’s products are posting complaints on, Plaintiff’s competitors may be more likely to advertise on Defendant’s website, and Plaintiff may be more inclined to pay for Defendant’s reputation management services.

This is essentially an empirical claim, and I’ve never seen any empirical evidence supporting it (or, for that matter, refuting it). I’d be grateful for any thoughts.

Initial Interest Confusion. I keep trying to declare the initial interest confusion doctrine dead, but judges aren’t cooperating! The judge says that initial interest confusion is still a viable theory in the Third Circuit (he’s correct as a matter of precedent), and thus Amerigas can keep arguing it. Unfortunately, like the likelihood of consumer confusion, the judge does hint that it’s probably going to fail later; the judge says “it may ultimately be difficult to establish initial interest confusion in this case.” Once again, lots of wasted time and money ahead!

Contributory Infringement. In a small piece of good news for PissedConsumer, the judge says Amerigas properly failed to allege that advertisers were actually committing trademark infringement or that PissedConsumer knew of their infringing activity. Thus, Amerigas hadn’t stated a proper claim for contributory infringement.

Along the way, the judge makes the following odd but interesting statement: “Plaintiff has not cited to—and the Court is unable to locate—any authority to support the proposition that use of a trademark in a hyperlink constitutes per se infringement.”

47 USC 230

PissedConsumer asserted Section 230 against claims of unfair competition, tortious interference and unjust enrichment. To some extent, those state law claims just extend the trademark claim, e.g., Amerigas argued that it was unfair competition for PissedConsumer to merchandise the trademark as part of PissedConsumer’s overall scheme. However, the court rejects the Section 230 immunity based on two alleged facts: that PissedConsumer allegedly authored some of the reviews in question, and that PissedConsumer allegedly controlled the ads on its site. The latter fact should be entirely irrelevant for Section 230 purposes, and it’s almost certainly not true as a factual matter if PissedConsumer is a Google AdSense publisher. The former fact is just like the uncited Vo case, where the court similarly accepted the plaintiff’s unsupported claim that PissedConsumer wrote the reviews to survive the motion to dismiss. I explained why that’s not a good result in my blog post on the Vo case.

The judge says that PissedConsumer may reassert the Section 230 immunity later in the case, but a Section 230 immunity post-discovery is way less valuable than a Section 230 immunity on a motion to dismiss.


Judge Buckwalter’s opinion is solidly constructed in the sense that it fairly applies existing law to the alleged facts. As taxpayers, we got our money’s worth from this opinion. Nevertheless, it’s clear Judge Buckwalter and other judges lack inadequate doctrinal tools to kill doomed cases early. In the end, this case is really about Amerigas trying to shut down negative reviews of its business. I don’t like PissedConsumer’s business model any more than you do, and they are not the best “face” for free speech online, but PissedConsumer is a platform for letting consumers have their say, and opinions that expose them to legal risk jeopardize important social values. See my Regulation of Reputational Information essay. In this respect, I could imagine other judges finding more doctrinal flexibilities to address the realpolitik of this situation. I can’t blame Judge Buckwalter for failing to do so, but the result is unfortunate nevertheless.