Virtumundo Ninth Circuit Argument: Impressions

I attended the Virtumundo Ninth Circuit oral arguments in Seattle on Tuesday.  (I came in just as counsel for Gordon was part way through.  You can access the audio file by searching on this page (enter case no.: 07-35487)
.)  Access a copy of the AG’s brief here [pdf], and Virtumundo’s brief here [pdf].

A quick introductory note.  Gordon v. Virtumundo is a case brought by James Gordon, who is described as a plaintiff who ran a “litigation factory” suing spammers.  (Court and media descriptions, not mine.)  The trial court judge granted Virtumundo’s motion for summary judgment against Gordon and imposed attorney’s fees against him to the tune of $111,000.   The trial judge’s decision had two significant points.  First, the court ruled that although Gordon was the provider of an internet access service under CAN-SPAM, he was not adversely affected (because he didn’t suffer any harm different from that of a typical consumer).  Second, the court ruled that Gordon’s claims under Washington’s spam statute were preempted by CAN-SPAM.  The third point was that Gordon had to pay Virtumundo's attorney’s fees.  The court’s ruling on the attorney’s fees issue was (in my opinion) the most significant, but it received little if any attention yesterday.  Overall, the case has potentially significant implications for the many CAN-SPAM cases pending in lower courts in the Ninth Circuit (and other circuits), for the simple reason that it will decide who gets to sue under CAN-SPAM.

The panel was comprised of Judges Gould, Tallman, and Callahan.  Derek Newman argued for Virtumundo (now AdKnowledge) and Tim Walton argued for Gordon.  Also in attendance were Bob Siegel and Doug McKinley, a representative from the Washington AG’s office and a representative from AdKnowledge. 

What Harm Did Gordon Specifically Suffer?  Gordon’s counsel was extensively grilled on this issue.  I thought he had passable answers, and focused the court smartly on the district court's question of whether the harm was “significant”.  He should have come right out and said that there's nothing in the statute that requires significant harm.  Gordon’s counsel obviously agreed that Congress did not grant a general private right of action under CAN-SPAM:  “not everyone who has a computer can sue under CAN-SPAM.”  His argument was that Gordon was not a consumer.  Analytically, of course, he has the problem that Gordon didn't suffer any harm other than typical consumer harm.  The court's discussion on this issue leads to an interesting question - the court focused somewhat on AOL and its anti-spam efforts.  While these efforts have some benefit to AOL, I wonder if the court (or its IT folks for that matter) think these efforts have been enough to curb spam?  Intuitively, I think the answer is no.  We all probably receive roughly the same amount of spam as we did 5 years ago.  (Will plaintiffs like Gordon have any appreciable effect on spam?  Probably not, but you never know.) 

What was the Purpose of Gordon’s Business?  This was the most interesting part of Tuesday's discussion in my opinion and a particularly weak point for Gordon.  The court asked bluntly whether Gordon "set himself up for this?"  I think Judge Gould asked "he [Gordon] needs the spam, doesn't he?"  Gordon’s counsel pretty much conceded the point that Gordon's goal in being an ISP was to sue spammers.  He argued that Gordon set up his business not to offer email accounts as loss leaders, but to offer accounts funded by people who cause spam harm.  I'm not sure there's anything expressly in the statute that speaks to this, but Gordon's position suffers from the weakness that he has admitted that he is not a typical ISP or even a real ISP.  (On the other hand, everyone seems to agree that he satisfied the technical definition of an IAS, so it's unclear as to whether this admission should be dispositive.)

Did Gordon Invite the Harm?  Much of the questioning of Virtumundo's counsel focused on whether Gordon “invited” the spam and how this affected his standing.  Judge Gould asked if there was case law in other contexts where a plaintiff who has technical standing under a statute was precluded from bringing a claim because he or she had brought about that standing through a voluntary act.  This is a good question, but I don’t think this issue was discussed much below and the record certainly was not well developed on this point.  It’s important to note that the issue of whether Gordon availed himself of unsubscribe options is distinct from whether he placed his name on any email lists that Virtumundo utilized.  Finally, even if Gordon put his email address out there, this should not be equated with “inviting” the harm.  It would be different if there was evidence that Gordon actually signed up for mailing lists solely in order to file claims.  I’m not sure even this would matter . . . but I don't think there was any evidence to this effect anyway.  Ultimately, I don’t think the case will turn on this issue.  Often a significant chunk of an appellate argument is spent on an issue that the court does not even mention in its opinion. 

What is an IAS?  Judge Callahan asked a good question of line drawing and whether it’s better to have a bright line rule which defines an IAS as anyone who facilitates connectivity, as opposed to a rule which looks to the amount of revenue the IAS makes or what its intentions are.  This is a tricky issue – the case law is fairly clear that there’s no “revenue threshold” for who qualifies as an IAS.  There’s a lot of case law and statutory text to overcome in order to come to the conclusion that simply providing some sort of access is not sufficient to turn someone into an IAS.  Virtumundo did not concede this point, but it’s a tough one to win from their standpoint. 

What Does it Mean For an IAS to Suffer an Adverse Effect?  Virtumundo argued (I thought effectively) that the structure of the statute supports the conclusion that the IAS must suffer harm that is not typical consumer harm in order to satisfy the “adversely affected” threshold.  Its theory roughly is that consumers rely on big ISPs to police their networks.  Only those ISPs that really need to police their networks and whose customers will truly suffer should have standing.  They were hinting at a customer-oriented test and although this doesn’t have any support in the text of the statute, intuitively, this is an easy one to win from their standpoint since the evidence was undisputed that Gordon did not charge anyone any money for his supposed IAS services.  Judge Callahan made a statement which compared rich ISPs and poor ISPs, in the course of which she actually referred to AOL as “rich” (this made me chuckle – although AOL may have a lot of customers, they are typically thought of as a tech dinosaur who is stuck in the doldrums).

Does CAN-SPAM Preempt Washington’s Email Statute?  Virtumundo somewhat surprisingly took the position (agreeing with the Washington AG’s office) that CAN-SPAM does not preempt Washington’s email statute in all circumstances.  This seemed like a shift from their position in the briefing and their position below.  The question really is whether CAN-SPAM sets forth the floor for when a state can impose liability based on misleading statements or subject lines, or whether an email can not violate CAN-SPAM but still violate the Washington statute.  Virtumundo’s position is that the court didn’t really need to decide that issue since here the evidence was clear that the emails did not violate Washington’s standards for misleading emails.  The amicus brief filed by the Washington AG's office took this position.  I'm not sure how this squares exactly with the district court's conclusion.  

The Attorney’s Fees Issue: I didn’t hear any discussion at all of the attorney’s fees issue.  This is too bad, since I thought that this was one of the more important issues in this case.  What’s the appropriate standard where a successful CAN-SPAM defendant seeks fees against the plaintiff?  (Previous post here.)  I guess there are a few other upcoming cases where this issue can be hashed out.


Predictions?  Other thoughts?  It’s always a foolish exercise to try to predict the result based on oral argument, so I will refrain.  I just don’t think the court has much reason to reverse the decision of the court below.  Hopefully the court will provide some guidance as to the above issues, regardless of what it does. 

  • No trackbacks exist for this post.

  • 12/11/2008 2:53 PM Eric Goldman wrote:
    Thanks for writing up the hearing, Venkat. I know these posts take a long time to write, so thanks for sharing. Eric.
    Reply to this
  • 12/11/2008 5:53 PM Bill Silverstein wrote:
    I think that the discussion on "inviting" the damage was wrong. There is no such thing as signing up for spam, as spam is unsolicited. Even if you are signing up for spam, any damages under CAN-SPAM is for DECEPTIVE HEADERS, subject lines, etc.

    If the Defendant did send the e-mails, and it was deceptive, there was no authorization for them to deceive.

    The assumption of risk is another "cute" argument. Just by offering a service or connecting your server to the internet is not an invitation to send to deceptive spam to that server. The same argument could be to anyone who connects a computer to the internet, that by connecting a windows based computer to the internet is an invitation to have it infected and use it as a bot net. Or a movie studio selling movie DVDs are inviting piracy because there are pirates who can copy DVDs.
    Reply to this
Leave a comment

Submitted comments are subject to moderation before being displayed.

 Name (required)

 Email (will not be published) (required)


Your comment is 0 characters limited to 3000 characters.