14 March 2006
There was a piece in the NYT Magazine on Sunday on the surveillance program, titled â€śCan Network Theory Thwart Terrorists?â€? by Patrick Radden Keefe
It offers some basic ideas in applying network analysis to detecting terrorists, although it contains, I believe, a couple of mistakes. First, it states that â€śNetwork academics caution that the field is still in its infancy and should not be regarded as a panacea.â€? Not a panacea, certainly true; but in its infancy? Certainly notâ€”the modern field can trace its roots to the 1930s with the work of Moreno, Newcomb, and others. However, this is really just a quibble, in that the analysis of very large full network data (e.g., see Nathan Eagleâ€™s posting on his work on the phone data of an entire country) is in its infancy, due to the lack of tools and data until recently.
A more serious apparent error is in the following statement:
â€śThe use of such network-based analysis may explain the administrationâ€™s decision, shortly after 9/11, to circumvent the Foreign Surveillance Court. The court grants warrants on a case-by-case basis, authorizing comprehensive surveillance of specific individuals. The NSA program, which enjoys backdoor access to Americaâ€™s major communication switches, appears to do just the opposite: the surveillance is typically much less intrusive than what a FISA warrant would permit, but it involves vast numbers of people.
In some ways, this is much less alarming than old-fashioned wiretapping. A computer that monitors the metadata of your phone calls and e-mail to see if you talk to terrorists will learn less about you than government agent listening in to the words you speak.â€?
From what we have seen in the news reporting (excerpted in previous postings), however, the NSA program involves more than analysis of metadata. It also involves recording and content analysis through computer algorithms of the conversations of large numbers of individuals, with snippets selected out for human analysis. Indeed, I believe this was the whole basis of the controversyâ€”the stories on the metadata came after the wiretapping story. While I wouldnâ€™t say that metadata pose a non-intrusion, this is a much bigger deal. Indeed, as I believe I have said in previous postings, I would consider a digital recording of my conversation, content coded, to be a greater intrusion on my privacy than having two FBI agents listen to my conversations, with paper transcripts going into a file somewhere. Those bits do not decay like human memories, or easily lost, like files stored in the bowels of an enormous bureaucracy.
Posted by David Lazer at March 14, 2006 9:21 PM