The Washington Post reported today that Google has filed with the US Foreign Intelligence Surveillance Court in Washington a motion for a declaratory judgment that Google has a First Amendment right to publish aggregated statistics on FISA orders it has received. Good for Google. I hope they prevail.
I have done it as text for you from the Washington Post's embedded copy, because I know you are interested in what happens, and I wanted to point out one small correction. Well, it's probably not small to The Guardian.
Here's Google's filing, as text, first:
If you notice, the filing says that both the Washington Post and The Guardian claimed that the NSA had direct access to their servers. But actually, The Guardian's article said this, and I've emphasized the important parts:
UNITED STATES FOREIGN
INTELLIGENCE SURVEILLANCE COURT
IN RE MOTION FOR DECLARATORY JUDGMENT
OF GOOGLE INC.'S FIRST AMENDMENT RIGHT TO
PUBLISH AGGREGATE INFORMATION ABOUT
Docket No. _____
MOTION FOR DECLARATORY JUDGMENT OF
COMES NOW Movant Google Inc. ("Google") pursuant to 28 U.S.C. §2201 and
Foreign Intelligence Surveillance Court ("FISC") Rule of Procedure 6(d) and respectfully moves
this Court for declaratory judgment, or such other relief as appropriate, that Google may disclose
limited, aggregate statistics regarding Google's receipt of orders issued by this Court, if any,
without violating the Foreign Intelligence Surveillance Act or the FISC Rules of
GOOGLE INC.'S FIRST AMENDMENT RIGHT TO
PUBLISH AGGREGATE INFORMATION ABOUT FISA ORDERS
Google is an electronic services provider that offers a wide variety of products, services,
and online tools such as Gmail and Search to millions of users around the globe. Transparency is
a core value at Google and the company is committed to informing its users and the public about
requests it receives from government agencies around the world for the production of users'
information and/or communications. Google publishes a Transparency Report conveying this
1 Nothing in this Motion is intended to confirm or deny that Google has received any order or orders issued by this Court.
information in aggregate form, available at
In 2013, for the first time, Google
included in its Transparency Report the number of National Security Letters ("NSLs") it receives
and the number of users/accounts specified, within a range and on an annual basis. The Federal
Bureau of Investigation confirmed in writing that Google could so, and to Google's knowledge,
no declassification of any such information was necessary.
On June 6, 2013, The Guardian newspaper published a story mischaracterizing the scope
and nature of Google's receipt of and compliance with foreign intelligence surveillance requests.
In particular, the story falsely alleged that Google provides the government with "direct
access" to its systems, allowing the government unfettered access to the records and
communications of millions of user. The story is available at
The Washington Post
also published a misleading story that day, alleging that the U.S. government is "tapping directly
into" Google's central servers in order to surreptitiously obtain user records and
communications. The story is available at
On June 7, 2013, Google Chief Executive Officer Larry Page and Chief Legal Officer
David Drummond posted a blog entry that responded to these allegations as best it could given
the constraints imposed by the government's position that even general information regarding
Google's receipt of and response to foreign intelligence surveillance orders, if any, cannot be
disclosed. Google clarified that the government does not have direct access to Google's servers,
that Google provides information to the U.S. and other governments only in accordance with the
law, and that Google reviews each government request and complies only if the
requests appear proper and lawful. The blog post is available at
In light of the intense public interest generated by The Guardian's and Post's erroneous
articles, and others that have followed them, Google seeks to increase its transparency with users
and the public regarding its receipt of national security requests, if any. On June 11, 2013,
Google requested that the Department of Justice and the Federal Bureau of Investigation permit
Google to publish the aggregate numbers regarding the receipt of national security requests, as
further described below. The Department of Justice and the FBI have not classified the
aggregate numbers regarding the receipt of national security requests described above.
Nonetheless, the Department of Justice and FBI maintain their position that publication of such
aggregate numbers is unlawful.
Google's reputation and business has been harmed by the false or misleading reports in
the media, and Google's users are concerned by the allegations. Google must respond to such
claims with more than generalities. Moreover, these are matters of significant weight and
importance, and transparency is critical to advancing public debate in a thoughtful and
II. ENTITLEMENT T0 RELIEF
Google is a "communications carrier . . . or other specified person" subject to orders by
this Court to assist the government in conducting electronic surveillance or other foreign
intelligence collection activities pursuant to the Foreign Intelligence Surveillance Act, 50 U.S.C. §§
1801-l88lg ("FISA"). Google seeks to be transparent regarding FISA requests that may be
or have been served upon it, if any, and to respond to false or misleading statements about the
scope of its compelled disclosure under national security authorities.
In particular, Google seeks a declaratory judgment that Google has a right under the First
Amendment to publish, and that no applicable law or regulation prohibits Google from
publishing, two aggregate unclassified numbers: (1) the total number of FISA requests it
receives, if any; and (2) the total number of users or accounts encompassed within such requests.
Goog1e's publication would disclose numbers as part of the regular Transparency Report
publication cycle for National Security Letters, which covers data over calendar year time
periods. There would be two new categories to cover requests made under FISA (50 U.S.C. §§
1801-1881g):(a) total requests received and (b) total users/accounts at issue. Each of these
entries will be reported as a range, rather than an actual number. That range would be the same
as used by Google in its reporting of NSLs currently, in increments of one thousand, starting
with zero. As with the NSL reporting, Google would have a Frequently Asked Questions
section that would describe the statutory FISA authorities themselves.
To be clear, Google would not state, either in the published statistics or the FAQ, which
FISA authorities the government has actually invoked to compel production of data from
This Court has the power to declare that aggregate data about such orders for all
providers is protected by the First Amendment and is not classified or subject to any other legal
limitation on disclosure. The Court may, pursuant to a rule or inherent in its own authority,
make clear that providers can publish aggregate numbers of orders received and the aggregate
number of affected users under such orders.
Accordingly, Google respectfully requests that
this Court issue a declaratory judgment
indicating that Google may lawfully disclose such information.
Pursuant to FISC Rule of Procedure 7(i), Google certifies that the following responsible
employees for relevant matters hold security clearances: John Kent Walker Jr., General Counsel
(FBI-Secret), and Richard Paul Salgado, Legal Director (FBI-Top Secret). These clearances
were granted for the purpose of handling classified
legal process. Google's undersigned counsel
does not hold a security clearance.
DATED: June 18, 2013
Perkins Coie LLP
[address, phone, fax, email]
Attorneys for Google Inc.
CERTIFICATE OF SERVICE
I hereby certify this 18th day of June, 2013, that the foregoing document was served via hand
delivery on the following:
Litigation Security Group
United States Department of Justice
Perkins Coie LLP
[address, phone, fax, email]
Attorneys for Google Inc.
The National Security Agency has obtained direct access to the systems of Google, Facebook, Apple and other US internet giants, according to a top secret document obtained by the Guardian.
You see the difference? The Guardian's Glenn Greenwald is an attorney, so the wording was pretty carefully crafted, as lawyers are wont to do, to say that while the companies denied it, the document -- not The Guardian -- claimed that they had direct access. And he showed a graphic of the slide that does say that.
The NSA access is part of a previously undisclosed program called Prism, which allows officials to collect material including search history, the content of emails, file transfers and live chats, the document says.
The Guardian has verified the authenticity of the document, a 41-slide PowerPoint presentation – classified as top secret with no distribution to foreign allies – which was apparently used to train intelligence operatives on the capabilities of the program. The document claims "collection directly from the servers" of major US service providers.
Although the presentation claims the program is run with the assistance of the companies, all those who responded to a Guardian request for comment on Thursday denied knowledge of any such program.
The Washington Post made a stronger claim, which it then corrected to say what The Guardian said, that the document made that claim. You can still see the original claim on this page. Readers might not notice the difference, but legally, there is a huge difference.
It's like in Declarations, which are sworn to, if there is anything in there that you think is true but you can't say so by personal knowledge, you might say, "on information and belief" such and so happened. Maybe it did, and maybe it didn't, but you believe it enough to state it. For example, SCO said this about IBM and Novell in its Second Amended Complaint back in 2004:
199. Specifically, commencing on or about May 2003, Novell began falsely claiming that Novell, not SCO, owned the copyrights relating to UNIX System V. On information and belief, IBM had induced or otherwise caused Novell to take the position that Novell owned the copyrights -- a position that is flatly contradicted by the Asset Purchase Agreement. Since that time, Novell has improperly registered the same copyrights that it sold to SCO and that SCO had previously registered. In that instance, it was SCO's belief that was simply wrong, factually and legally, and that's why the jury told it that it was wrong. But you can't say SCO lied about it, at least not from this statement in the Second Amended Complaint, because SCO didn't claim IBM did this by personal knowledge; rather it said it was a belief it held, and relying on its understanding of the Asset Purchase Agreement, it believed Novell had violated it and that it was following an IBM-encouraged strategy of denial. It turned out Novell was right, but the way SCO phrased it, it wasn't a flat out lie, not provably.
Or sometimes you might read a news article that says that So and So said today that such and such a company would be doing XYZ. The news entity isn't saying it knows. It's saying So and So said it's so.
So legally, you can say things very carefully, and that's sort of like what The Guardian article did -- it didn't say it was so, but rather that the document, which it had verified as being authentic, said it was so.
I don't know why there is a difference between the document and what the Internet companies are saying. My point is simply that The Guardian was careful in its language, even if some readers missed the distinguishing line it drew.