Four months ago, Washington State Attorney General Bob Ferguson sued Google, alleging the tech giant has spent years ignoring this state's law on election ad transparency. Shortly afterward, Google announced it would stop selling political ads in Washington state altogether—though it has had some trouble making good on that promise.

Then, last week, Google filed an answer to Ferguson's lawsuit in King County Superior Court. It asked a judge to throw out Ferguson's suit and made a rather surprising claim: Technically, Google never sold any political advertising at all in Washington State during the years in question.

Really?

Attorney General Ferguson has alleged that since January of 2008, candidates and political committees in Washington state "have reported $1.5 million in payments to Google related to political advertising which ran on Google platforms."

But that all depends on what you call "political advertising," at least according to Google's lawyer, Stephanie Jensen of the firm Wilson Sonsini Goodrich & Rosati.

"The state’s claims fail," Google's lawyer argued in an October 12 court filing, "because, among other things, Google did not accept 'political advertising' or 'electioneering communications' as those terms were defined under RCW 42.17A.005 at any time prior to the filing of the complaint."

Ferguson's complaint was filed on June 4, the same day he filed a very similar complaint against Facebook.

On June 7, Google's lawyer notes, Washington state's definition of "political advertising" was officially updated to make it extra clear that election ads promoted via "digital communication" are, in fact, "political advertising."

Does that mean, as Google now argues, that digital communications could not be considered "political advertising" before June 7?

Not so fast.

The Washington State Public Disclosure Commission signaled many months before June 7, and then declared several weeks before June 7, that it viewed online political ads as falling squarely within the definition of "political advertising."

Etched in even firmer stone, though, is this:

Back in 1972, an overwhelming majority of Washington state voters approved a landmark election transparency law via citizens initiative, and when they did so they approved a specific definition of "political advertising" that read as follows:

Since 1972, then, any advertising aimed at influencing Washington state's elections through a "means of mass communication" has qualified, under the state's official, voter-approved definition, as "political advertising."

Is Google trying to argue that it is not a "means of mass communication," and therefore the $1.5 million in political ads Google sold in Washington state between 2008 and 2018 weren't technically political ads?

Or does the company have some other interesting logic that supports this notion?

I asked spokespeople for Google yesterday, but have received no reply.

Also, Google thinks it may be immune from Washington state law anyway

In addition to Google's claim that it didn't technically sell political ads in Washington state, the company—like Facebook—is now arguing that a controversial federal law gives it immunity from Washington state law in any event.

Yesterday, I explained the tech industry's long-running use of this federal law, Section 230 of the Communications Decency Act.

It'll be interesting to see whether Section 230 can be used as a tech company shield in this particular case, but the fact that Google is now trying to use it raises a question.

Why, if Google thinks Section 230 means it doesn't have to comply with Washington state's election transparency law, did the company announce in June that it felt it had to stop running political ads in Washington state because it was not prepared to fully comply with Washington state's election transparency law?

I asked spokespeople for Google this question, too, but have received no reply.