A lack of US cryptanalytic security before Midway?

[Steven M. Bellovin]

On 7 Sep 2006 15:33:15 -0000, John Levine <johnl at iecc.com> wrote:

> >The conventional wisdom is that the successful US cryptanalytic efforts
> >against Japanese naval codes was a closely-held secret.
> 
> Has the conventional wisdom forgotten that it was reported in the
> Chicago Tribune in 1942?
> 
> See, for example, http://www.newseum.org/warstories/essay/secrecy.htm
> 
> Fortunately, the Navy Department had enough sense not to make a public
> stink, and the Japanese evidently didn't read the Chicago paper.
> 
The URL you cite does not support your claim.  It speaks of the successful
cryptanalysis of JN-25 as "one of the closest kept secrets of World War
II".  It also notes that the reporter learned of some data just from
seeing a piece of paper in a senior officer's quarters, rather than
knowning about the real source of the data, and that the Trib's headline --
"NAVY HAD WORD OF JAP PLAN TO STRIKE AT SEA" -- was not in fact justified
by what the reporter had seen and written. In other words, there was not a
factual leak of the real secret, though admittedly Japanese
counter-intelligence would likely have drawn the proper conclusion had they
seen the story.

I should note that if Kernan's account is correct, the danger to American
SIGINT efforts were far greater than were realized.  Three downed American
airmen were rescued by Japanese ships; they were then interrogated and
executed.  None of them (again, according to Kernan) had had proper
training on what they should or should not disclose.  If, indeed, the fact
of cryptanalysis was common knowledge, it was lucky indeed that the proper
questions weren't asked -- or if they were asked, they weren't answered,
even though at least one of them did give away more information than he
should have.


		--Steven M. Bellovin, http://www.cs.columbia.edu/~smb

---------------------------------------------------------------------
The Cryptography Mailing List
Unsubscribe by sending "unsubscribe cryptography" to majordomo at metzdowd.com