From: David Sternlight <david@sternlight.com>
Date: Sat, 27 Jul 1996 07:06:53 +0800
To: cypherpunks@toad.com
Subject: Re: Defeating "Perp Profile" Analyses Of Written Materials
In-Reply-To: <01I7GBFL287694F9CD@delphi.com>
Message-ID: <v03007805ae1ed6b7c9a0@[192.187.162.15]>
MIME-Version: 1.0
Content-Type: text/plain


At 10:08 PM -0700 7/25/96, JonWienk@ix.netcom.com wrote:
>On Wed, 24 Jul 1996, JMKELSEY@delphi.com wrote:
>>I wouldn't count on even heavily-chained anonymous remailer messages
>>to protect my identity from moderately wealthy and determined
>>attackers, if I did many anonymous posts.  Writing style and topic
>>alone may narrow the suspect list down to a manageable number.
>
>There is an easy way to defeat psycholinguistic analysis techniques used by
>LEA's to profile perps.  Buy a translation program, (such as Globalink's
>Spanish
>Assistant) use the program to translate the text to Spanish, (or any other
>language) and then use the program to translate the foreign language text
>back
>to English.  The baselines of word choice, grammatical structure, etc.
>will be
>shifted to reflect the biases of the program rather than the biases of the
>writer.  As an example, I will use the entire text of this message as a
>demonstration.

You are using two code books for double encoding. This is the kind of
problem analysts solve while brushing their teeth. All the analyst has to
do is determine your translation programs (easy to do since they have such
obvious anomalies) and create reverse code books. The malapropisms are so
obvious that there should be little difficulty aggregating the longest
phrases that make up one codebook entry.

David