pageok
pageok
pageok
New Risen Book Sheds Light on NSA Surveillance Program:
In the many debates we've had here at the VC about the NSA's domestic surveillance program, we've been stymied by the lack of facts about how the program works. James Risen's new book, State of War: State of War : The Secret History of the C.I.A. and the Bush Administration, was released just today, and it has lots of juicy new facts to ponder. Risen's civil libertarian views are front and center, so the tone isn't exactly balanced, but the new facts make it an incredible read.

  Risen's book answers a bunch of our questions about why the program was started, and what it does differently than past NSA surveillance programs. We don't yet have definitive answers to the legal questions, but we have a much sharper picture of the issues. Further, I suspect these disclosures may have an impact on the public opinion; at first blush, at least to me, the new facts seem to present the program in a somewhat more sympathetic light than have some previous reports.

  According to Risen, the key to the new program is a shift in telecommunications technology in recent decades that has made U.S. networks the carriers of lots of international telephone and e-mail traffic.
  In addition to handling telephone calls from, say, Los Angeles to New York, the switches also act as gateways into and out of the United States for international telecommunications. A large volume of purely international telephone calls — calls that do not begin or end in America — also now travel through switches based in the United States. Telephone calls from Asia to Europe, for example, may go through the United States-based switches. This so-called transit traffic has dramatically increased in recent years as the telephone network has become increasingly globalized. Computerized systems determine the most efficient routes for digital "packets" of electronic communications depending on the speed and congestion on the networks, not necessarily on the shortest line between two points. Such random global route selection means that the switches carrying calls from Cleveland to Chicago, for example, may also be carrying calls from Islamabad to Jakarta. In fact, it is now difficult to tell where the domestic telephone system ends and the international network begins.
  In the years before 9/11, the NSA apparently recognized that the remarkable growth in transit traffic was becoming a major issue that had never been addressed by FISA or the other 1970s-era rules and regulations governing the U.S. intelligence community. Now that foreign calls were being routed through switches that were physically on American soil, eavesdropping on those calls might be a violation of the regulations and laws restricting the NSA from spying inside the United States.
  But transit traffic also presented a major opportunity. If the NSA could gain access to the American switches, it could easily monitor millions of foreign telephone calls, and do so much more consistently and effectively than it could overseas, where it had to rely on spy satellites and listening stations to try to vacuum up telecommunications signals as they bounced through the air.
  Reading over this part of Risen's book, it seems that most of the new surveillance program was not about domestic surveillance at all; most of it was about the surveillance of entirely international calls and e-mails that just happened to be routed through U.S. networks in the course of delivery. According to Risen, the program typically monitored about 7,000 individuals overseas at any given time, as compared to about about 500 people who were located in the United States. From an operational perspective, then, the big difference between prior NSA practices and the new program was that the NSA was using a back door into domestic privider switches in the U.S. to monitor communications that were mostly foreign to foreign.

  Okay, so now let's take a look at the legal questions again. Recall that FISA prohibits "electronic surveillance," defined in relevant part as follows by 50 U.S.C. 1801(f):
(1) the acquisition by an electronic, mechanical, or other surveillance device of the contents of any wire or radio communication sent by or intended to be received by a particular, known United States person who is in the United States, if the contents are acquired by intentionally targeting that United States person, under circumstances in which a person has a reasonable expectation of privacy and a warrant would be required for law enforcement purposes;
(2) the acquisition by an electronic, mechanical, or other surveillance device of the contents of any wire communication to or from a person in the United States, without the consent of any party thereto, if such acquisition occurs in the United States. . . .
  As I read this language, monitoring communications both sent to and received by individuals located outside the United States is not prohibited by FISA. This surveillance is technically domestic — it occurs within the United States — but it does not involve monitoring "a person in the United States." So if I'm understanding things correctly — always a big "if" when you post after 2 am — most of the program did not violate FISA.

  Now here's the part I can't quite figure out. If most of the program dealt with international calls, and didn't violate FISA, why would the program be designed so that it also tapped the calls of 500 or so people in the United States? If the communications tapped were wire communications, then tapping those communications inside the United States seems to clearly violate FISA under 1801(f)(2). So it seems like there are two possibilities: either the communications were wire communications and the designers of the program thought that the 500 people tapped in the U.S. were of sufficient importance (potential terrorists cells, etc.) that they didn't want to cabin the program to deal exclusively with foreign-to-foreign calls, or else the communications were electronic communications and perhaps the monitoring didn't violate FISA. (Can anyone else tell based on Risen's excerpt whether the communications were wire or radio? Maybe it's just late, but I'm not sure.)

  Here's another puzzle to ponder. A few passages in the Risen book suggest that the legal concerns offered by those who leaked this story may be different from the legal concerns that I've been focusing on in my posts. For example, according to the book, "[s]everal government officials who know about the NSA operation have come forward to talk about it because they are deeply troubled by it, . . . [t]hey strongly believe that the president's secret order is in violation of the Fourth Amendment of the Constitution, which prohibits unreasonable searches." Other parts of the book seem Fourth Amendment-focused, as well.

  That seems somewhat odd to me, because, as I've explained before, my primary legal concerns are statutory, not constitutional. This raises a couple of different possibilities. For example, it may be that the program doesn't violate FISA after all, and the debate within government has really been about the Fourth Amendment. The concern within some government officials may be that scanning traffic en masse for phone numbers or e-mail addresses of even foreign calls may violate the Fourth Amendment rights of domestic people whose communications are scanned (even only in the passing way that any Internet wiretap must scan all traffic). That is consistent with Risen's claim that "Now that [foreign to] foreign calls were being routed through switches that were physically on American soil, eavesdropping on those calls might be a violation of the regulations and laws restricting the NSA from spying inside the United States." The "regulations and laws" don't seem to be FISA, as I believe FISA is implicated only when the person monitored is in the United States; could those "laws" be the Fourth Amendment? Are there other "regulations" that govern the NSA that might be implicated here? It's hard to tell. Of course, it's also possible that the Fourth Amendment concerns are a bit of a red herring: the claim may be a cover for other motives. Who knows.

  In any event, it's way late, and I'm probably not being very coherent at this hour. More pondering tomorrow. Thanks to Lee Tien for the tip.
BB (mail):
I am not sure that I understand the program any better from this snippet. Were the 7,000 and 500 "monitored" or were their calls "tapped" or "eavesdropped [upon]"? It seems to me that there is a difference; the former is more like (or exactly like) a pen register; the latter is listening and possibly recording content. If the issue is the Fourth Amendment, I think that is no insignificant difference. All in all, I think the media reporting of the facts, such as we have them, sounds entirely plausible. The characterizations, however, seem to overplay the facts fairly routinely.

As to how the 500 could be in the US, I have two examples: International cell phones physically brought into the US (T-Mobile, say) and instances where a non-US person is in the US but his or her call is monitored/tapped from outside (your earlier satellite example, or in some other way). The first example would be an easy-to-understand accident, the latter an attempt to "work around" FISA.
1.4.2006 3:47am
Dave:
The key, from what I've read, is the phrase "at any given time." Risen is right that this is fundamentally about new technology, and that's the part we don't know about yet. I think this is a data mining project. Remember, the reason that NYT thought they could publish the leak without harming security was that they witheld the "technical details."

Here's what Rockafeller said:
"As I reflected on the meeting today, and the future we face, John Poindexter's TIA project sprung to mind, exacerbating my concern regarding the direction the Administration is moving with regard to security, technology, and surveillance."


The reference to TIA, or "Total Information Awareness" is what this article focuses on.

It is entirely possible that the NSA technology at issue here is some kind of high-volume, automated voice recognition and pattern matching system... the NSA could very easily narrow down the amount of phone traffic that they'd have to a relatively small fraction of international calls with some smart filtering. First, they'd only monitor calls where one end of the connection is in a country of interest. Then, they'd only need the ability to do a roving random sample of a few seconds from each call in that already greatly narrowed pool of calls. As Zimmermann describes above, you monitor a few seconds of some fraction of the calls looking for "hits," and then you move on to another fraction. If a particular call generates a hit, then you zero in on it for further real-time analysis and possible human interception. All the calls can be recorded, cached, and further examined later for items that may have been overlooked in the real-time analysis... The "softer trigger" here is a phrase that's on a watch list, or a call with an abnormally high volume of a certain type of vocabulary. The "agility" bit is a reference to the technology's ability to move from call to call, taking small slices. That's also probably what's behind the claim that the technology is less intrusive than a traditional wiretap, because the time slices are very short.


In other words, this isn't picking 500 people and monitoring everything they do. It's monitoring 500 calls, if nothing suspicious shows up, monitor another 500, and so on. All automated. And this might be "legal" if Illinois v. Caballes,, rather than Kyllo v. United States is the relevant precedent.. It also means that a lot of innocent people will be searched and large numbers of people will be considered potential terrorists even if false positives are extremely rare.

This is scary stuff.

Dave
1.4.2006 4:20am
Bruce Hayden (mail) (www):
I an beginning to see why a referral was made, if such is the case, for the leak of this information. Maybe those more involved in this area have known about all that traffic being routed through the U.S., but I sure didn't. Yes, it makes sense, but that dot was connected, at least for me, until now. Ditto for the data mining, at least as a previous poster has posited.

I think that maybe part of the damage may be that some of our putative allies may be somewhat upset by this information. After all, do businessmen in France really expect their phones to be tapped by the NSA when making calls to Japan? I think the assumption by many until now has been that we would not be listening in on that sort of thing.

What is a bit bizarre to me is that there appears to be no legal reasons that would keep the NSA from monitoring that conversation between businessmen in France and Japan, but would if one of them were located in the U.S. I see why this is so, but also see why both France and Japan might not be so understanding. And maybe bringing this to their attention may be one reason for the referral.
1.4.2006 4:44am
Bruce Hayden (mail) (www):
Apparently, the book is in demand already. I just put myself on a waiting list at the local library - I was somewhere around 10th in line. It was actualy fairly slick - I used the Amazon link in Orin's post to get the ISBN, and copy/pasted that into my library's search/request function. All in about a minute.
1.4.2006 4:53am
Splunge (mail):
I can't wait until the technonaives realize that every IP transmission that moves across the Internet has its content scanned, many times, by multiple computers, without the sender's permission or a shred of judicial oversight, ha ha, that being the nature of IP. This amusing business of being shocked, shocked! that the NSA actually spies on people when it can is, I think, partially grounded on a naive fantasy that when you communicate halfway around the world with someone you're using a direct dedicated circuit to your addressee, the privacy of which it is practical to defend by appropriate legal checks. But modern switched-packed technology bears little resemblance to the imagined copper wire direct from ear to ear, and I think it's basically gone way beyond the practical reach of the law.

Just imagine the complexity of trying to set up a legal framework for specifying when government (or a private party) may, and may not, read or store a given datagram, or pass it through a given application, or present a given analysis of it to human eyes. Orin could probably write five law review articles on the topic, each more polysyllabically inconclusive than the last.

Whether the graybeards at the NYT know it or not, the era of "secure line" secure communication has been dead for about as long as the dial telephone, and we are deep into the era of "secure protocol" secure communications.
1.4.2006 6:09am
Diversity Hire:
I can't wait until the legalonaives realize that every IP transmission that moves across the Internet is potentially available to the government for traffic-analysis without a warrant.

I'm one of those naives, who just figured that out by piecing together Professor Kerr's statements, Tom Holsinger's claim, some anonymous source here, and something called Smith vs. Maryland.

As I understand it, not withstanding a magistrate's opion, it probably extends to the location information from cell phones, too. Wow. What's the point of FISA, again?
1.4.2006 6:35am
Frank Drackmann (mail):
I've been listening to cell phone conversations for years, amazing the stuff people will talk about, drug deals , murders for hire, plus the usual mundane extramarital affairs.
1.4.2006 6:39am
subpatre (mail):
Can't resist: <i>I told you so!</i> The NY Times and their sources never alleged FISA violations, a giant red flag.

'Data mining' will still raise concerns about the 4th; legitimate or not I don't know. In the same way as this case, people misunderstand or misrepresent the "librarian clause" of domestic-based foreign intelligence. The intent (actual use) is the government compelling a list of those borrowing particular library materials; not fishing for material that a particular known individual obtained.
1.4.2006 6:57am
colt41 (mail):
It seems that international calls, whether routed through the US or not, are fair game for eavesdropping/tapping/data-mining.

But something more must be afoot -- more than just collecting to/from phone numbers like a trap-and-trace or pen register would -- for Jim Comey, and others, to object.

If the NSA is randomly listening to calls involving a US citizen to see if there's reason to *really listen*, then the 4th Amendment concerns would seem apparent.

Perhaps regs could have been written to make such a listening program an "administrative search" -- think airports, gun dealers, liquor stores -- with no warrant or PC required. See United States v. Edwards, 498 F.2d 496, 500 (CA2 1974) (Friendly, J.).

Or maybe the executive order sought to accomplish the same end as regs permitting an "administrative search" without making the program public, even indirectly.

Regardless, Comey is a law-and-order guy. He joined the SDNY as an AUSA in 1987 and worked his way up to the USAtty in both Richmond and the SDNY. Then from there up to DAG.

Seems to me there's a significant, as yet unknown, reason he objected.

All the rest of what we're saying is really guess-work. And, IMO, it's too soon for "I told you so."
1.4.2006 7:25am
elliottg (mail):
I'm not sure any law was broken, but the government's access to telecommunication switches need some oversight. These are owned by corporations. The government mandates that these corporations provide a methodology for them to listen to any call, but it does not require them to maintain an open channel for law enforcement or the NSA to come in and tap at will. Perhaps the issue was authorization. I think CALEA is the place to look for more answers.
1.4.2006 7:40am
John Lederer (mail):
The executive order:

http://www.fas.org/irp/offdocs/eo12333.htm
1.4.2006 8:19am
John Lederer (mail):
Regarding your question of whether it moves by fiber optic cable or microwave, I recall litigation from about 10 or so years ago when AT&T's longlines facility in New Jersey had a backhoe cut many of its fiber optics causing major telecommunications outages.

AT&T had been charging extra to provide redundancy against outages. The redundancy didn't occur. People who had leased the more expensive redundancy sued.

AT&T testified that it no longer knew or could determine the transit route for its telecommunications. The reason is that router/switches along the journey were making the determination of the next place to send a signal based on dynamic determinations of delays in the system.
1.4.2006 8:32am
AppSocRes (mail):
Splunge has it right on the technology. This is an area where the law has yet to go. When it does, it's worth considering that "data mining" techniques now exist that will prevent almost all "innocuous" communications from ever being heard by a human being while making sure that "suspicious" communications are at least flagged for further study. Is privacy violated when a machine does a statistical analysis of a bit pattern? What about when this analysis may save thousands of lives? The scarey thing to me is that people with a woeful ignorance of these technologies are going to wind up making the law.

It's also amusing to consider that many of the same people expressing shock and outrage about bit-sampling of potential terrorist communications were overjoyed a few years ago when a Florida couple's illegal intercepts of cell phone traffic were used to embarass some Republican politicians. Suggests something about their priorities to me.
1.4.2006 8:37am
John Lederer (mail):
"Now here's the part I can't quite figure out. If most of the program dealt with international calls, and didn't violate FISA, why would the program be designed so that it also tapped the calls of 500 or so people in the United States?"

Supposition:

NSA doesn't monitor all the switches in the US.. It only monitors those through which international traffic must pass, i.e. those that serve the cable landing points. Both international transit traffic and US<->international traffic has to pass through those.

If it does monitor everything in the US it could concentrate on the major "hubs" (IPX's formerly NAPs in the case of the internet), but a significant amount of traffic doesn't pass through these. The trend is to have a system that resembles a "cloud" rather than a spoke and hub system.

It just doesn't seem likely to me that they NSA has so massive a physical structure as to be able to siphon all traffic.

The one other possibility, I can think of, would be to alter the routing tables/algorithms so as to divert international traffic through points that NSA monitors -- but why do that when you can simply focus on the landing points?

Note that if you have the capability to monitor international transit traffic, monitoring the US originated/destined international traffic would likley just involve a software change, with no or minimal hardware, and might be something one could implement in a few days.

I would be very interested to know whether they are just monitoring particular electronic addresses ( just reading the address information from the packet header, and examaning those packets with particular headers) or whether they are attempting to find particular voices saying particular words (which would appear as a somewhat distinctive binary pattern in the digitized traffic).

In other words are we just reading the address on the envelope of all traffic to seperate what we look at or are we looking at the contents of everything. There are suggestions both ways (captured phone numbers, versus a recognized code word ), so perhaps it is a combination.
1.4.2006 9:07am
DK:
Another possibility here is that if the NSA is tapping switches in order to access foreign communications, it may not be technically feasible for them to restrict their tap to foreign-only calls. i.e. their technology may just sniff packets, and only determine whether the packets are US or foreign after examining their headers. If examining the sender/receiver identification in a header is considered a search, then it is literally impossible for the NSA to know which communications are foreign or domestic until after it has searched them.

I recall one of the news stories mentioning that the NSA may have 'accidentally' gotten US-US communications in addition to the "border-search" US/foreign communications. If they are accidentally getting US-US communications, it could be that they are just examining an entire switch, without any technical means to determine who is using the switch until after they read the data
1.4.2006 9:07am
chaboard:
"it may be that the program doesn't violate FISA after all"

But there's still that pesky Gonzalez admission:
"Foreign Intelligence Surveillance Act provides --
requires a court order before engaging in this kind of
surveillance that I've just discussed and the President
announced on Saturday, unless there is somehow -- there
is -- unless otherwise authorized by statute or by
Congress. That's what the law requires. Our position is,
is that the authorization to use force, which was passed
by the Congress in the days following September 11th,
constitutes that other authorization, that other statute
by Congress, to engage in this kind of signals
intelligence."
1.4.2006 9:47am
Medis:
Maybe it is just me, but I don't find it very hard to imagine how you could set up this program to avoid conducting electronic surveillance within the meaning of FISA, and then how you could change that. As John L. points out, that would seem to be just a matter of "software"--how one programs the automated filters.

And based on what the Administration, including Gonzales, has said, that seems to be what happened: citing the authority of the 2001 AUMF, the President ordered the NSA to change the program so that it no longer avoided conducting electronic surveillance within the meaning of FISA. That could be as simple a matter as changing the address filters, if these are wire communications within the meaning of FISA (making 1801(f)(2) the relevant provision). But it could also be a matter of tracking the communications of known, particular US persons, perhaps through something like voice recognition technology (making 1801(f)(1) the relevant provision). Or both.

And I guess what puzzles me is why Orin thinks there is a puzzle here. The Administration has said that it changed the program on the authority of the 2001 AUMF to start targeting the international communications of people in the US, including US citizens, and that FISA would have required warrants for such surveillance if not for the 2001 AUMF. So, his first "possibility" is actually just what the Administration has said it was doing--it started conducting surveillance that would have required FISA warrants on the authority of the 2001 AUMF because they thought it was important to start conducting such surveillance.
1.4.2006 9:48am
Bruce Hayden (mail) (www):
I don't see the NSA monitoring purely domestic switches because both parties are presumably United States Persons in the United States as defined by FISA. If the problem is with the international switches, it is probably because it is not always easy to detect when one party is here and not. And, just maybe, that, as the President said, it is nice to know when Al Qaeda calls someone here.

From the description I have heard so far, my supposition at this point is that a lot of the monitoring is random checking for certain voices, not specifically for certain words, etc. That is because a specific voice can probably be recognized much quicker than a random conversation can be decoded, and it appears that a lot of calls are being listened to for very short periods of time. It is only when one of them triggers something that it is really listened to, and that trigger, I suspect, is either a specific voice or a specific, mostly foreign, phone number.

I should add that somewhere in the last couple of days (it could have been the ECHELON article posted here a couple of days ago) it was pointed out that what is most likely happening is that a number of calls are simultaneously listened to for a very short period of time. Then, another batch. But of those listened to, a very small number trigger actual investigation. And of those, there is a winnowing process where ultimately most are automatically discarded. Only a handful are actually listened to by humans.

Which of course brings us to the possibility of ethnic profiling. Might Arabic speaking itself be a trigger, or, more likely, a factor that increases the likelyhood of triggering recording for later listening?
1.4.2006 9:50am
Rockfish (mail):
With respect to data mining, Jack Balkin argues that it's irrelevant whether the communications are intercepted by a computer or a human. The real issues are (1) how long the mined data can be stored; and (2) what restrictions govern the humans who query the database.

1.4.2006 10:01am
John Lederer (mail):
Actually, we have a bit clearer picture if we throw in Pelosi's letter and Hayden's very early briefing of Congress.

The initial question on 9/11, I suspect, was "Who the hell just attacked Washington and New York, and are they about to do it again." Hayden, without any detailed direction from the President ordered NSA to do whatever it could to find out. NSA probably already had some Al Qaeda intelligence. He apprently ordered them to start listening to any calls to/from the U.S. and Al Qaeda. Maybe he did the same for anyone else that might be suspected from the Red Brigade to Iraq, Iran, and Hammas.

Good for him.

Pelosi's only reservation when she was briefed appears to have been "Get the President to order this, don't just do it on your own hook."

He did later get a presidential order (November I think we were told), and the program with, I suspect, some modifications continued. Later when the bureaucracy caught up, the program was suspended and further modified (Comey and the Chief of FISC both objected to some parts) till everyone in the bureaucracy was OK with it and the program reinstituted.

This all seems pretty sensible to me. Faulting Hayden now for his original act would be a little like disciplining the Gunner's Mate on the Arizona who destroyed government property when he broke the lock on the ready ammunition to shoot at the Japanese.

To the degree that anyone has constitutional/legal concerns they ought to focus on the program as it operates today..and say "good job" to Hayden for going ahead in September of 2001.
1.4.2006 10:04am
Rockfish (mail):
1.4.2006 10:05am
Rockfish (mail):
I agree with John Lederer that the issue isn't whether the law was bent or broken in the immediate aftermath of 9/11, but why the surveillance is still being conducted lawlessly today. Recall that when asked why the administration hadn't sought to amend FISA, Alberto Gonzales responded that they didn't think Congress would give them the authority to do whatever it is they're doing. That astonishing response causes me to take the AUMF argument even less seriously than Prof. Kerr. I wouldn't want to be the guy who had to stand up in court and argue that without, knowing it, Congress had accidentally given the administration authority it spewcifically intended the administration not to have.
1.4.2006 10:13am
Dylan Alexander (mail) (www):
"After all, do businessmen in France really expect their phones to be tapped by the NSA when making calls to Japan?"

HAHAHA! They certainly do.

http://www.cyber-rights.org/interception/echelon/
http://news.zdnet.com/2100-9595_22-524698.html
http://news.bbc.co.uk/1/hi/world/europe/820352.stm
1.4.2006 10:26am
John Lederer (mail):
Rockfish,

Recall that the DOJ also said:


"In addition, any legislative change, other than the
AUMF, that the President might have sought specifically to create such an early warning system
would have been public and would have tipped off our enemies concerning our intelligence
limitations and capabilities."
1.4.2006 10:31am
corngrower:
Points

1. Not a single elected member of our congress has put forth a law to restrict what they are bitching about. (the battle has more political value than a solution)

2. Congress was fully informed (thru the appropriate committies) of what was going on.

3. Clinton and Carter did the exact same stuff. (notice the lack of commentary from either) Seems the ex-presidents have an opinion on the rest of the Bush Presidency, just not NSA

4. If a person of the present Bush administration would leek, say, a name of a super secret spy, because, gee, they had grave concerns of legal issues. Would a judge throw a person in jail, and talks of impeachment boil over?

The author never quotes a single person involved. It is always, always, someone privy to a conversation of other people. Think about it. Some low staffer in a meeting coughs up a private converstation and the principals could not figure who the leek is? Duh!
1.4.2006 10:33am
Medis:
John L.,

I also agree that in the immediate aftermath of 9/11, one can and should imply an exception into FISA. And, of course, if the 2001 AUMF is taken as a declaration of war, then there is no need to imply an exception into FISA for the 15 days following its passage (because such an exception is explicit).

As rockfish points out, the real issue is the continuation of this program. And as I see it, there are two plausible dates at which this program was legally required to start getting warrants. The first is 15 days after the 2001 AUMF, when the declaration of war exception in FISA expired. The last is the passage of the USA-PATRIOT Act, when any implied authorization to bypass the pre-9/11 FISA contained in the 2001 AUMF would have been replaced with FISA as amended.
1.4.2006 10:34am
Medis:
John L.,

The argument that the President should be able to amend military law as he sees fit without the participation of Congress in order to keep his activities secret is a very dangerous one.
1.4.2006 10:39am
John Lederer (mail):
Yes it is. But I don't think that is what they did. I think they had alternatives:

(1) we can rely on AUMF or
(2) we can go to Congress and ask for new law


They chose (1). In retrospect, I suspect they wish they had gone to Congress in October of 2001 when they could havbe gotten a law requiring all citizens to whistle "Yankee Doodle" except when their mouths were full. But, as we now know, they were still in the process of figuring out what they wanted/needed -- a process that lasted at least until early 20o4 (Comey'sa objections). At that point they probably could not have prohibited double parking in the NSA parking lot without extensive and revealing hearings.

2001 to 2004 seems like a very long time to me for a bureaucracy to move, but, I suppose in their defense they have been dealing with a dynamic situation.

I also think, fwiw, that we don't need to amend FISA, we pretty much need to rewrite it from square one.
1.4.2006 10:56am
AF:
"For example, it may be that the program doesn't violate FISA after all . . ."

Except that, as noted, the Administration says it did (but for the AUMF).

The response is then: But maybe the Administration has national security reasons for concealing its compliance with FISA.

The more I think about that argument, the more implausible it seems. Here we are on this public site, speculating about the technology that would allow the program to avoid violating FISA. Would our speculations be in any way aided by the Administration's stating: "We neither confirm nor deny that FISA applies to the type of surveillance involved in this program"?
1.4.2006 10:57am
corngrower:
Medis

your just being silly. Commander and Chief of the Military.

Seem like simple words to me.

Military law is the perview of the president. He is the Commander. Sorry to break it to you.

I suppose you belive that either house of congress could overturn a SCOTUS decission?
1.4.2006 11:00am
Rockfish (mail):
John,

They did indeed say that changing the law might disclose sources and methods, but I can't see how that changes anything. After all, we waged a 50-year Cold War against an enemy that posed an existential threat, and despite the need for secrecy and black ops, the Congress enacted FISA in 1978. There is nothing new about the need for secrecy, nor is there anything new about the rule of law. This is just one of many ways in which the 9/11 Changed Everything meme is profoundly unhelpful.

The democratic process isn't the most efficient means of government, but it surely beats the alternatives. The president simply can't be allowed to do anything he wants whenever he decides he's acting as commander-in-chief -- even if he honestly believes it would be a really good idea.

"Experience should teach us to be most on our guard to protect liberty when the government's purposes are beneficent. Men born to freedom are naturally alert to repel invasion of their liberty by evil-minded rulers. The greatest dangers to liberty lurk in insidious encroachment by men of zeal, well-meaning but without understanding." -- Olmstead v. United States (Brandeis dissent).
1.4.2006 11:19am
Medis:
John L.,

I think the crucial question is whether (1) was in fact a legitimate possibility, particularly in light of the USA-PATRIOT Act.

corngrower,

What makes you think being a military commander gives you the power to make military law? Ask any officer in the armed forces if they command gives them the authority to replace the UCMJ with their own rules. Or ask Alexander Hamilton, who explained:

"The President is to be commander-in-chief of the army and navy of the United States. In this respect his authority would be nominally the same with that of the king of Great Britain, but in substance much inferior to it. It would amount to nothing more than the supreme command and direction of the military and naval forces, as first General and admiral of the Confederacy; while that of the British king extends to the declaring of war and to the raising and regulating of fleets and armies -- all which, by the Constitution under consideration, would appertain to the legislature."

And of course that interpretation--that the President and not Congress gets to make military law under the Commander in Chief clause--would make the Constitution nonsensical, because Article I actually specifically delegates the power to make military law to Congress.
1.4.2006 11:21am
Hyacinth (mail):
Orin,

Isn't your analysis mooted by the essential admission of the administration that it violated FISA, but was justified in doing so (for whichever constitutional or statutory reasons it chooses to settle upon)? It seems like the question is no longer whether FISA was violated, but whether or not the violation is legitimate - whether based on Article II, AUMF, or other reasons.
1.4.2006 11:21am
John Lederer (mail):
Bruce Hayden said:

"I don't see the NSA monitoring purely domestic switches because both parties are presumably United States Persons in the United States as defined by FISA. If the problem is with the international switches...."

I don't think there are "domestic" switches as opposed to "international" switches.

There are just switches.

Picture a highway map of the U.S. The road junctions are the switches. For any given road junction we cannot say that it is 100% local traffic or a 100% interstate traffic.

About all we can say is that a junction of a couple of two lane roads in Iowa is likely to be mostly local traffic, and most long distance interstate traffic is likely to pass through one or more big cities like Chicago. The big transportation hubs like Chicago woiuld be the equivalent of IXP's where big internet interchanges are.

However, if we also look at at the port facilities of a town on the coast, we can probably point to a few intersections down near the docks and say "most everything that goes through these intersections is trucks carrying containers to or from abroad". Similarly some of the switches that serve cable landing points are proably nearly pure foreign traffic, and in that sense might be called "international switches".

At or near where the cable comes ashore there usually is a building in which carriers can rent space for switches. Those switches are likely nearly 100% international traffic. At some cable landing points, rather than having the switches located there, the cable is "extended" to a nearby city and the switches serving the cable are located in the same place as all the switches for a big switching interchange.

I have to say that when there was a report of a few domestic calls being intercepted by mistake, my immediate thought was that they picked up the night watchman at the cable landing point ordering a pizza.
1.4.2006 11:26am
corngrower:
The President is not acting as a dictator

The House of representitives controls ALL of the money we give them. IF? The President is out of line? Just one member of congress could offer up a bill to eliminate NSA, or, define the latitudes allowed by NSA.

This is all political. nothing more, nothing less. Stop trying to stuff a square peg (law), into a roundish like, sorta,hole (politics).
1.4.2006 11:29am
ChiMod:
Corngrower--

Military laws are the perview of the President?

Section 8 of the US Constitution:

"The Congress shall have the power to... make Rules for the Government and Regulation of the land and naval Forces"
1.4.2006 11:35am
corngrower:
You win Medis

You baited me into discussing this on your faulty premise, that this is a legal, constitutional debate. when in fact it belongs to our elected, not our judges.
1.4.2006 11:38am
Medis:
corngrower,

You say, "Just one member of congress could offer up a bill to . . . define the latitudes allowed by NSA."

I think the precise problem is that they already did: that is a fair description of FISA as amended by the USA-PATRIOT Act.
1.4.2006 11:38am
Medis:
corngrower,

Incidentally, I agree with you that at least insofar as the President is concerned, ultimately this will be a matter for Congress to decide. And obviously, politics will in fact play a major, and likely decisive, role in what Congress decides to do. Still, if this did get to something like a consideration of impeachment, I strongly suspect the Articles of Impeachment would include the charge that the President failed in his duty to faithfully execute the laws. So, in that sense the legal issues may matter, if the House ends up considering impeachment or the President is actually being tried in the Senate.

And for that matter, there is not a neat distinction between the legal and the political. In other words, while some people might not care, others might take seriously the notion that the President ought to be obeying the law.
1.4.2006 11:44am
htom (mail):
Part of this may just be confusion as to how modern communications are moved. Long, long, ago there were real hardware circuits from point A to point B that could be "tapped". These still exist, at least from your typical residential telephone to the box down at the street corner, and can still be tapped. But most of modern communication is bandwidth limited, digitized, and shipped about as packets of information along with hundreds of other packets of other communications, and you can't really tap ONLY that communication; what you do is tap the stream, and filter out the communication that you want to listen to. So you're tapping lots of conversations that you're not listening to.

This is, in many ways, an invasion of privacy, but in other ways, it is not, it's just the way that it has to be done.
1.4.2006 11:47am
John Lederer (mail):
Just a slight addition to my previous post. Switches are like a complex highway interchange. Each switch has two or more ports that are connections to cables. Whereas a switch can handle a wide mix of traffic, part of its job is to segregate that traffic by destination and send it out the appropaite port.

For transatlantic cables, for instance, many come ashore in New Jersey. New Jersey cable landings mostly "backhaul" to main interchanges in or near NYC ( in other words the main switching is done in the NY area by an "extension" of the submarine cable). That means that in NY there are switches that have ports that connect to the cable landing station. These are almost pure international traffic. If NSA could get the phone company to give me access to those ports, I would have a pretty good place to put a siphon for international traffic.
1.4.2006 11:48am
Medis:
htom,

I think what many of us are suggesting is that in light of those circumstances, what matters is the programming of the filters.
1.4.2006 11:50am
htom (mail):
Medis -- I agree. But political hay can be made by correctly claiming that thousands of innocents are being tapped without warrants, and I wonder if that's what is going on.
1.4.2006 12:04pm
John Lederer (mail):
Orin Kerr said:

"As I read this language, monitoring communications both sent to and received by individuals located outside the United States is not prohibited by FISA. This surveillance is technically domestic — it occurs within the United States — but it does not involve monitoring "a person in the United States." So if I'm understanding things correctly — always a big "if" when you post after 2 am — most of the program did not violate FISA."

Look at 1801(f)(4) -- that, I think, is the crucial FISA provision for a international transit call, if one buys into my argument that we are not dealing with a "wire or radio communication" because we don't meet the common carrier requiremnt in 1801(l).
1.4.2006 12:08pm
Tom Holsinger (mail):
DK, it's not a possibility. It's true except for the part about examining the header being a "search".

"Another possibility here is that if the NSA is tapping switches in order to access foreign communications, it may not be technically feasible for them to restrict their tap to foreign-only calls. i.e. their technology may just sniff packets, and only determine whether the packets are US or foreign after examining their headers. If examining the sender/receiver identification in a header is considered a search, then it is literally impossible for the NSA to know which communications are foreign or domestic until after it has searched them."

This is why Polaris has said it is so important to update FISA. It is just too tempting for the feds to examine only the header. I suspect this is also the reason he resigned from the NSA.
1.4.2006 12:20pm
John Lederer (mail):
by "international transit call" I mean a foreign to foreign call that transits the US.

Presumably the determination that FISA doesn't apply to these calls predated the Bush administration. So why did the Clinton Administration conclude that these were OK:

(1) Not a wire communication (not a common carrier)
(2) Software change to switch not a device
(3) No warrant required because of President's inherent authority.

Suddenly Jamie Gorelick's testiimony to Congress about the President's inherent authority to do warrantless intercepts for reasons of national security seems a little more highlighted.
1.4.2006 1:07pm
child (mail):
I've read a few of the comments on this site alluding to a political motivation behind the current Bush power grab.

But while many defenders of the current administration claim partisan bias in the response to the latest in a string of Bushco scandals, what can these defenders say when confronted with the following observations from their own party? (Long list of quotes to follow)

Why do these Republicans hate America?

* Sens. Chuck Hagel (R-NE) and Olympia Snowe (R-ME), "No president is ever above the law. ... We are a nation of laws. You cannot avoid or dismiss a law."

* Senate Judiciary Committee chairman Arlen Specter (R-PA) "the president's decision to inform a handful of members of Congress was sufficient . . .I think it does not constitute a check and balance,. . .you can't have the administration and a select number of members alter the law. It can't be done.''

* Sen. Richard Lugar (R-IN),"I think we want to see what in the course of time really works best, and the FISA act has worked pretty well from the time of President Carter's day to the current time."

* Sen. Susan Collins (R-ME), surveillance abuses 'extremely troubling. . . warrants further inquiry by Congress"

* Sen. John E. Sununu (R-NH) "it is a little bit of a stretch for the administration to say the surveillance program was authorized by the post-Sept. 11 resolution . . .This is the kind of activity that should be approved in statute."

* Sen. Larry Craig (R-ID) and Rep. C.L.Otter (R-ID) "there is a clearly established process of judicial oversight through the Foreign Intelligence Surveillance Act (FISA) Court to obtain warrants for such wiretapping. . . the White House appears to have circumvented that process.

"The Founders envisioned a nation where people's privacy was respected and the government's business was open,. . . these actions turn that vision on its head. If the government is willing to bend the rules on this issue, how are we supposed to believe it won't abuse the powers granted by the Patriot Act?"

* Sen. Lindsey Graham (R-SC) "I don't know of any legal basis to go around" FISA's requirement the government obtain a warrant to conduct domestic surveillance of Americans . . . We can't become an outcome-based democracy. Even in a time of war, you have to follow the process, because that's what a democracy is all about: a process."

* Sen. John McCain (R-AZ) "I know that the situation has changed since September 11th. So the equation has changed. Why did the president choose not to use FISA? That's a legitimate question."

* Former Rep. Bob Barr (R-GA) "Exactly like Nixon before him, Bush has ordered the National Security Agency (NSA) to conduct electronic snooping on communications of various people, including U.S. citizens. That action is unequivocally contrary to the express and implied requirements of federal law that such surveillance of U.S. persons inside the U.S. (regardless of whether their communications are going abroad) must be preceded by a court order.

Alleged associates of al-Qaeda are today's targets of that breathtaking assertion of presidential power. Tomorrow, it may be your phone calls or e-mails that will be swept up into our electronic infrastructure and secretly kept in a growing file attached to your name. Then everyone you contact could become a suspect, a link in an ever lengthening chain that would ensnare us all in the files of the largest database ever created through unlimited electronic spying that touches every aspect of our lives.


* Bruce Fein, "It's more dangerous than Clinton's lying under oath because it jeopardizes our democratic dispensation and civil liberties for the ages. It would set a precedent that, as [former Supreme Court Justice] Robert Jackson said, would lie around like a loaded gun, able to be used indefinitely for any future occupant."

"President Bush presents a clear and present danger to the rule of law. He cannot be trusted to conduct the war against global terrorism with a decent respect for civil liberties and checks against executive abuses. Congress should swiftly enact a code that would require Mr. Bush to obtain legislative consent for every counterterrorism measure that would materially impair individual freedoms. . . .why is he so carefree about risking the liberties of both the living and those yet to be born by flouting the Constitution's separation of powers and conflating constructive criticism with treason?

Congress should insist the president cease the spying unless or until a proper statute is enacted or face possible impeachment. The Constitution's separation of powers is too important to be discarded in the name of expediency."

*William Safire: "the president can't seize dictatorial power. And a lot of my friends looked at me like I was going batty. But now we see this argument over excessive security, and I'm with the critics on that."

* George F. Will: "the president's decision to authorize the NSA's surveillance without the complicity of a court or Congress was a mistake. Perhaps one caused by this administration's almost metabolic urge to keep Congress unnecessarily distant and hence disgruntled.

* Robert A. Levy, a senior fellow in constitutional studies at the libertarian Cato Institute and a Federalist Society member: "The text of FISA is unambiguous: "A person is guilty of an offense if he intentionally engages in electronic surveillance ... except as authorized by statute." That provision covers communications from or to U.S. citizens or permanent resident aliens in the United States. Moreover, Title III (the Wiretap Act) further provides that "procedures in this chapter and the Foreign Intelligence Surveillance Act of 1978 shall be the exclusive means by which electronic surveillance ... may be conducted."

Accordingly, warrants would be required for law enforcement purposes and, therefore, warrantless surveillance absent an authorizing statute would violate the FISA requirement."

_________
So I guess all these rightwingers hate the President and are conspiracy nuts too? No, it is apparent who is playing partisan here, and who really cares about America.

National Security means securing the nation, and not only in the physical sense. A significant part of our national character is freedom from the sort of authoritarian intrusions some are so willing to let the current President get away with.

So if you really care about securing our nation, you cannot pick and choose the parts you want to defend. Our personal liberty is essential to our identity. We can defend our borders and our personal freedoms at the same time.


1.4.2006 1:09pm
John Lederer (mail):
Rockfish,

re Balkin on Data Mining, I think he makes an error by focusing on length of data retention as the key, making the argument that as storage costs approach zero, evrything will be stored.

Just some back of the envelope scribbles-- according to a 2004 FCC report, there was a total of about 150 GB/s in the capacity of the submarine cables landing in the US in 2002. I am sure there is proably twice that now. Even when costs are near zero, if you are pumping in 100 Gigabytes a second at 33% utilization of the cables, you are talking some real storage needs.

I think they filter before storing (except perhaps for some temporary buffering), and the key thing to focus on for legal constraint is the filters.
1.4.2006 1:22pm
John Lederer (mail):
Child,

I think it safe to say that very few of the commentators know what the actual facts are, and the NYT seems to have played a little of the "..a dark sinister figure, camouflaged to blend into the shadows, furtively slunk by on padded feet that made no sound" when talking about the cat.
1.4.2006 1:28pm
Jack John (mail):
It is funny to see how all these people who assumed the President was breaking into people's closets and stealing their panties modify their arguments so that they don't look like complete fools.

Hi, Medis.
1.4.2006 1:43pm
Medis:
Jack John,

I refer you to this 2:01 pm post on the "Comey Tried to Limit NSA Surveillance Program" discussion yesterday. Not to toot my own horn, but I think I got pretty close!

"The basic story--as in fact told by Gonzales--is that the President simply authorized the NSA to conduct surveillance without warrants that would have required warrants under FISA. That surveillance need not involve any new technology or capability whatsoever--it could be new only in the sense that the NSA is doing things it was not ALLOWED to do before, not things it was INCAPABLE of doing before.

For example, suppose this is just ECHELON-based. As I understand it, the legal regime post-FISA(1978) and pre-2001 AUMF was likely instantiated (assuming it was instantiated) just by rules for programming the ECHELON "dictionaries". So, the legal regime governing the use of ECHELON could be changed just by changing those rules for programming the ECHELON dictionaries--eg, by allowing the NSA to track the international communications of a known, particular US person without getting a warrant. In such a case, what the NSA was actually doing would have changed without the underlying technology or capabilities of the system changing."

Of course, in this case it is not ECHELON-based, but rather based on tapping these US-based switches. Nonetheless, the basic point is the same: apparently we had developed this capability in the years before 9/11, and what we changed was simply the programming of the filters.
1.4.2006 1:58pm
Jack John (mail):
Medis, I said hello to you! I'm not sure why you responded to my earlier, separate, and distinct comment directed at cognitively dissonant fools....
1.4.2006 2:01pm
Medis:
John L.,

Maybe I am missing something, but on my read, 1801(f)(4) only applies if it is NOT radio or wire communications. So, if this program IS considered to involve wire communications, the applicable provision is 1802(f)(2), which requires that the communication be "to or from a person in the United States." So, as I read FISA, wire communications that are not to or from a person in the United States are covered neither by 1801(f)(2) nor 1801(f)(4). And that would be a very simple explanation of why this program could operate without FISA warrants when the communications in question were merely "in-transit" through the United States, but not sent from or to the United States.
1.4.2006 2:03pm
Medis:
Jack John,

Honestly, I was just itching for an opportunity to show off my prognosticatory skills.
1.4.2006 2:05pm
corngrower:
Child;

Cute, you quote responses of elected persons that disagree with the President. Now what? Do any single one of the persons you quote offer a bill in either house? Just asking. See you have an opinion, and, I have an opinion, but the only people you quote (at length) are the only people that have the power to introduce a bill that can be voted on by my servants.

Why?

Because not a single person that is elected would even offer up a second to any bill that would correct the imagined infractions.

Why?

It is political suicide.

Anyone, anywhere, can opine all they wish. But until at least one of these spinless SOB's offer a solution...This is a political rather that judicial problem. AS I have stated before. The Battle is a much larger asset than any solution not yet offered. (Stop hiding under ther robes of judges)

(which wouldn't be so bad if the Senate had the guts to put judges up to a vote)
1.4.2006 2:09pm
Just an Observer:
John Lederer,

I commend the research you have done on the technical details of the FISA definitions vis-a-vis modern technology.

Your post here raised several questions for me.

1) Is it your conclusion that since many (most?) modern submarine cables are not operated as common carriers, this would exempt their tapping from the definition of 1801(f)(2) because such cables do not fit the definition of "wire communication?"

That would seem to provide a significant loophole for unregulated tapping of such cables, provided that some other definition under 1802(f) is not applicable. My own rough understanding is that while it is difficult or impossible to "tap" any fiber-optic cable, even above ground, such an effect could be easily accomplished by inserting a device into a fiber-optic splice with the connivance of its non-common-carrier owner.

2) Examining 1801(f)(4) as you suggest, in the case of U.S. citizens, would not their communications still be covered because they have a reasonable expectation of privacy and intercepting their communicatons would require warrants for law-enforcement purposes?

3) You say that a software change in a switch is not a device. Does the definition of "device" exclude software? Have you found authorities on this point?

I'd appreciate your further thoughts.
1.4.2006 2:10pm
Stephen_Ve (mail) (www):
I agree that "monitoring communications both sent to and received by individuals located outside the United States is not prohibited by FISA." But your conclusion that most of the program did not violate FISA seems to overlook the contrary descriptions of the NSA surveillance program made by Attorney General Gonzales in the December 19, 2005 Press Briefing including the following:

The President has authorized a program to engage in electronic surveillance of a particular kind, and this would be the intercepts of contents of communications where one party to the communication is outside the United States.


What we're trying to do is learn of communications, back and forth, from within the United States to overseas with members of al Qaeda. And that's what this program is about.

The Attorney General clearly stated the NSA surveillance program involves intercepting communications between a party in the United States and a party outside the United States. If the program actually involved intercepting communications sent to and received by individuals located outside the United States, there is no doubt the Attorney General would have said so.
1.4.2006 2:13pm
KMAJ (mail):
Even with the Risen book, it seems there is still much we do not know. We all hear about warrantless searches, but that does not mean all searches were warrantless. I think that would be apparent from the article about FISA's unprecedented second-guessing and modifying warrants, that warrnats have been applied for. We have all heard that FISA has only denied four warrants (in 2003 and 2004 they rejected or deferred 6 requests - their first outright rejections ever) in its history, but the unprecedented number of modifications (prior to 2001, only 2 were modified, since 2001, 179 were modified) can only have one of two implications, either the warrants applied for have been ineptly handled on a massive scale or FISA has become an impediment to security by not adjusting to technological changes and applying them to their rulings. It raises the constitutional question of where, during a time of war, the line is drawn for executive authority to 'faithfully protect and defend' ?


If purely 'domestic' calls were monitored without a warrant, how were these calls flagged for monitoring ? For example, when terrorism figures are captured, specifically high level ones like Zubaydeh, where computers and cell phones are also captured, how long is their capture kept quiet to follow up on whatever information is gleaned from them and their possessions ? It would seem to make sense to keep it quiet for as long as possible so one could follow up on the information before they shut it down knowing that it is compromised. Zubaydeh, allegedly, had quite a long list of phone numbers inside and outside the US. Applying for a warrant increases the probability of a leak and the loss of valuable intelligence. How would FISA apply to the surveillance of these numbers for primarily intelligence purposes, which may reveal criminal complicity ?

On a side note, Judge Robertson's resignation from the FISC was specualted as a protest resignation, without any personal verification or statement to that effect from him. I proffer the question, in the light that some warrants had been applied for, and that he would most likely be aware of warrantless searches, could his resignation possibly be in recognition that he may be exposed as one of the leakers ?
1.4.2006 2:15pm
Rockfish (mail):
John,

You may well be right about the capacity issues. I have no basis for an opinion on that.

I'm not necessarily concerned about the information being collected -- we really don't know that -- but with the fact that it's being collected outside the law, with no independent oversight. There would presumably be filters to determine what information the system collects, and there should be rules to determine what they are. There should also be rules to govern who may query the database for which kinds of information in what circumstances. Finally, there should be judicial oversight to insure that these rules are followed.
1.4.2006 2:22pm
Medis:
KMAJ,

One of your questions is easy to answer. FISA applies to surveillance done primarily for intelligence purposes because that is exactly what FISA is about.
1.4.2006 2:40pm
KMAJ (mail):
Medis,

Mea culpa for not adding the descriptive 'foreign' to intelligence. It seems some of the debate centers on a blurring of foreign and domestic and the ability to legally separate the two with technological advances in communications.
1.4.2006 2:47pm
John Lederer (mail):
1) Is it your conclusion that since many (most?) modern submarine cables are not operated as common carriers, this would exempt their tapping from the definition of 1801(f)(2) because such cables do not fit the definition of "wire communication?"

I think it certain for some traffic, debatable for others.

The certain:
If I send an email to Usama (I am on charter.net a cable TV company) I don't deal with a common carrier. The submarine cable is not a common carrier, I don't think a common carrier ever enters the picture, unless in Afghanistan. We have a brand new Supreme Court decision confirming the FCC's conclusion that Charter is not a common carrier.

Therefore it not a wire communication and is not covered by (f)(1) or (f)(2). f(3) is out on its own terms, f(4) is the only possibility.

The debatable:

If my internet access provider was, say, SBC which is a common carrier, the submarine cable is still private and not a common carrier, but it could be argued that SBC "furnished" the submarine cable and also argued that what matters is what the communication was from my end, not at the point of tapping.






That would seem to provide a significant loophole for unregulated tapping of such cables, provided that some other definition under 1802(f) is not applicable. My own rough understanding is that while it is difficult or impossible to "tap" any fiber-optic cable, even above ground, such an effect could be easily accomplished by inserting a device into a fiber-optic splice with the connivance of its non-common-carrier owner.

Yes.

There have been periodic murmurs that the "repeaters" and "amplifiers" used on cables emit rf radiation that can be detected, possibly from quite far away (airplane? satellite?) but as the technology improves the need for these repeaters and amplifiers declines. NSA seems to spend money on technologies likely of interest in tapping fiber optics.


There also is some leakage of light from the optical strand, particularly where it curves, so one could imagine bending a cable and dtecting the light that leaks from it without cutting the fiber, but part of the reason there is less need for amplifiers and repeaters is that the optics of the cable get better and better with less and less leakage.






Splicing is far simpler than it used to be with new tools. The difficulty is doing so on a live cable without interrupting the communications carried. Of course doing it in deep water adds some complications tool (cables are buried in the sea floor in shallow water to avoid being snagged by trawlers).

2) Examining 1801(f)(4) as you suggest, in the case of U.S. citizens, would not their communications still be covered because they have a reasonable expectation of privacy and intercepting their communicatons would require warrants for law-enforcement purposes?

That is why the focus on what "device" means. You have to meet four conditions to violate (f)(4)
(a)not radio or wire
I think this is likely met at least for some traffic (my email to Usama using Charter)
--use a "device" as described
Is a software program change a "device"? I don't know. Put this down as 50/50
--no reasonable expectation of privacy
I don't know--but since submarine cables have always been surveilled in England (Zimmerman telegram) does one have a "reasonable expectation"?
--not require a warrant
I don't know, but 18 USC 2511(2)(f) seems to suggest that if it is a foreign or international communication, and is not electronic surveillance under FISA, it is legal to intercept (which makes things a bit circular)







3) You say that a software change in a switch is not a device. Does the definition of "device" exclude software? Have you found authorities on this point?

There is no "device" definition that I found in the statute, I have not researched it. (I am at home and don't have Lexis here).
1.4.2006 3:22pm
Medis:
KMAJ,

I'm not entirely sure I follow you. People have been talking about "foreign" versus "domestic" communications, and one of the reasons that conversation may have been a bit confused is that FISA itself defines no such terms.

But FISA does in fact define "foreign intelligence information":

"'Foreign intelligence information' means—
(1) information that relates to, and if concerning a United States person is necessary to, the ability of the United States to protect against—
(A) actual or potential attack or other grave hostile acts of a foreign power or an agent of a foreign power;
(B) sabotage or international terrorism by a foreign power or an agent of a foreign power; or
(C) clandestine intelligence activities by an intelligence service or network of a foreign power or by an agent of a foreign power; or
(2) information with respect to a foreign power or foreign territory that relates to, and if concerning a United States person is necessary to—
(A) the national defense or the security of the United States; or
(B) the conduct of the foreign affairs of the United States."

The whole point of FISA is to regulate how the government goes about getting foreign intelligence information. And, of course, it does allow the government to seek out foreign intelligence information which may be found within "domestic" communications (if that means communications from one person in the US to another in the US)--it just specifies procedures for doing so.
1.4.2006 3:34pm
AJStrata (mail) (www):
Mr. Kerr, With all due respect, communications systems are highly complex and not something to grasp or understand from some reporters vague interpretations. Risen explains very little. What the question should be is he deliberately holding back obvious clarifying information or is he,like most, struggling to grasp the systems and methods involved. An understanding some of us have taken decades to understand in detail.

Your first assessment is there is new technology

<i>"According to Risen, the key to the new program is a shift in telecommunications technology in recent decades that has made U.S. networks the carriers of lots of international telephone and e-mail traffic."</i>

New technology? Or more cost effective technology? What about information flowing through one of our satellites? Is that 'American soil'? The truth is we monitor foreign communications whenever and whereever we can. That it the purpose of SIG and COM INT. The only partial legal barrier is when an American is involved or one end of the communication (not one part of the path) is terminated in the US. 60% of the world's internet traffic goes through Northern Virginia. Does that make Al Qaeda email between Pakistan and Iraq now US communication?

No more than a radio wave passing through our space from one side of the country to the other. So focus on the end points - the termination points. In communications architecture there are levels which clearly perscribe the end points vs gateways, routers and switches.

Your next conclusion is spot on

<i>"As I read this language, monitoring communications both sent to and received by individuals located outside the United States is not prohibited by FISA. This surveillance is technically domestic — it occurs within the United States — but it does not involve monitoring "a person in the United States." So if I'm understanding things correctly — always a big "if" when you post after 2 am — most of the program did not violate FISA."</i>

Warrantless surveillance is a case where the 'target' is someone like a terrorists overseas who, for all intents and purposes, is treated as if they warrant exists! So let's take that perpsective on extrapolate. When a crime boss is being monitored and he calls the neighbors to borrow sugar - was monitoring the call illegal did it require a warrant? Of course not! Half the time a warrant is to discover another player in a criminal act. Half the time is to see if the target says something incriminating.

So if a legal warranted surveillance incidentally captures conversations iniated and controlled by the target (i.e., the target is selecting who get's contacted and monitored) then the same is true for legal warrantless surveillance.

Now to your question: How did it happen 500 people (WAG on Risen's part) are 'monitored'. First off lets start at the very beginning. NSA surfs the communications trying to detect a possible terrorist communication. This is all international. They are looking for contacts between known elements overseas. They want to find a Hamburg cell talking to a London cell. The goal is ti establish an area of focused monitoring.

Note that an area of focused monitoring can come from intel in the field like Sheikh Mohamed's computer. Or CIA intelligence leads. Or listening in on known terrorists who tip off the identity of other terrorists. Data sifting or mining is only one way to establish a lead.

Once they establish the lead, the monitor connections as well as content (though content is challenging given the propensity to use code words and phrases). See Andrew McCarhy's point on what is legally available, even here in the US

"As a matter of law, all telephone/email information is not constitutionally or statutorily equal. The interception of content (what you say to me and I say to you in a call or an email) requires court approval based on probable cause in most (but, importantly, not all) instances. But other telecommunication data — e.g., the fact that my phone number called your phone number at such-and-such at time on such-and-such a date, and that we spoke for seven minutes — does not."

Content can be dodgy and unclear. Contact points and patterns of calling are crystal clear.

Once a lead is established to the US, it is possible and probable NSA could monitor to make sure they are not listening on hotel reservation lines. Once they determine it is a high probability target, then they can take it to FISA using the intel as part of the probable cause argument.

And this seems to be Robertson's biggest problem with NSA and FISA- NSA intel being used for probable cause and 'tainting' the FISA process.

If the entire faux issue is how intel promulgates to law enforcement without using the intel, then this is a lot of partisan hot air which will burst easily.
1.4.2006 3:58pm
corngrower:
Medis

I thought the Supreme Court of the United States, should follow the law. If that is your standard, explain the ruling banning executions of convicted crimminals that were committed when they were 16 or 17 years old? Please help me follow that 'law'. Seems only SCOTUS, has no laws to follow.
1.4.2006 4:01pm
Medis:
corngrower,

Although I think that is an interesting topic, it strikes me as a bit tangential. Suffice it to say that I agree with what I take to be your premise--the Supreme Court has the authority to interpret and apply the law, which includes the authority to interpret and apply the Constitution, but it also has the duty to faithfully interpret the Constitution. Of course, whether the Court is doing thatin a particular case--faithfully interpreting the Constitution--is often a matter of some controversy.
1.4.2006 4:21pm
KMAJ (mail):
Medis,

Sealed Case seems to imply that there are areas within which FISA cannot restrict or rule on executive branch authority in attaining foreign intelligence. The specifics I wonder about in technological advances in communications that blur the difference would include pre-paid cell phones and phone cards which could be used to disguise foreign communications and make them appear domestic. It is an area that the FISA description seems to be lacking any clarity or definitive application.

If or when, more likely when, Padilla is brought before the Supreme Court, we might get a clearer picture, of how the many arguments, from AUMF, Article II and Fourth Amendment, separation of powers, etc, come into play. Even then, unless the NSA program is directly brought in a suit before them, we will be left with unanswered questions that will continue to be fodder for debate.
1.4.2006 4:33pm
AJStrata (mail) (www):
Regarding FISA and it's application to Foreign intelligence, it is not all encompassing. In the ruling by the FISA Review Court in October 2002 it clearly states the President has auhority to allow warrentless searches. It is based on Truong, which also held there were aveues outside FISA.

FISA was meant to keep domestic intel and law enforcement reasonably apart. It was never intended to control NSA, DIA, CIA, etc.

Again, a US person may be monitored and not be a target and therefore cause the need for a warrant. If AQ Terrorist A overseas is being monitored, and we detect and capture is regular communications with Person B in the US - then that is legal because person B in the US is making contact with the person under invesigation. Same with warranted wiretaps. The people the target is in communication with are legally monitored since they are on the 'wire' authorized for tapping. If a legal warrantless 'tap' is the same as legal warranted tap, then those who fall into the 'tap' are fair game under the same legal umbrella.

We don't need to over think this either. If a law exists that would allow a modern day Atta in Hamburg to call a modern day Mihdar in San Diego, and we cannot monitor that call immediately, and all subsequent communications between these two, then the law is broken and toss it out.

One other thing surprises me, and that would be anyone who thinks Nancy Pelosi and Jay Rockefeller grasp or understand a fraction of this issue.

So, if the entire issue boils down to a far left FISC judge denying warrants to track Al Qaeda in the US because the probable cause was developed from NSA snooping on Al Qaeda overseas (my strong bet), then this whole debate is sort of useless - isn't it? Robertson's supporters clearly stated Robertson felt NSA was tainting FISA.

There is only one way for NSA to taint FISA - and that is if FISA was the path used to gain proper warrants from intel developed by the NSA!

I'll say it again. The wall has always been flowing intel to law enforcement, and vise versa. This is the rub with many on the left - on 4th amendment grounds! Because intel is against the 4th Amendment! Wouldn't that 'taint' FISA, to have intel data used to prove probable cause?

This is fairly simple if you stand back and look at the players, and their responses.
1.4.2006 5:17pm
Medis:
KMAJ,

Actually, the dicta in Sealed Case implies more than that. The now infamous line is: "We take for granted that the President does have that authority and, assuming that is so, FISA could not encroach on the President's constitutional power."

If any sort of limit or procedural requirement "encroaches" on the President's power, then all of FISA is out, and so for that matter would be the entire US Code whenever the President wants to order someone to violate it.

Which is why it is a little hard to understand what to make of this dicta--without further qualification, it makes little sense.
1.4.2006 5:50pm
KMAJ (mail):
Medis,

It might be my legal naivete in viewing it as a common sense reading. I could not conceive of dicta intentionally implying such a wide scope of authority, but one that is rather narrow in construct vis a vis security, in light of the realities 9/11 brought to bear. As such, without any legal expertise to bear, I would see the dicta as more of a restriction on FISA and not an outright disavowal, as well as a concession to Article II executive powers when the nation is threatened.
1.4.2006 6:04pm
Medis:
KMAJ,

Of course, that is one of the problems with dicta--without an actual set of facts to which a legal proposition is being applied, it is often hard to know what that proposition really means. And this proposition is particularly vague, since there is neither an explicit nor implied definition of what they meant by "encroach".

So, if I was a judge on the FISC, I would not be sure what, if anything, I should do with that dicta.
1.4.2006 6:22pm
KMAJ (mail):
Medis,

Could it possibly be intended as a warning to the FISC to be very careful where they tread ?
1.4.2006 6:28pm
KMAJ (mail):
Medis,

An addendum, since Sealed Case, FISC modified 173 warrants and rejected or deferred 6, I wonder if there is any correlation. Prior to that, there had been no more than 8 modifcations and no rejections. They had turned down 4, but I believe those were for application errors. Another interesting item would be how many of those modifications, rejections and deferrals were Robertson's.
1.4.2006 6:34pm
Just an Observer:
KMAJ: "An addendum, since Sealed Case, FISC modified 173 warrants and rejected or deferred 6, I wonder if there is any correlation. Prior to that, there had been no more than 8 modifcations and no rejections."

Post hoc ergo propter hoc?

Sealed Case was only one event in the time period, which generally spanned the period of more aggressive surveillance activity across the board. (Notably, it also is roughly the same time the Bush adminstration began bypassing the FISC court entirely.) It would not be at all surprising if, when they did seek court orders, they pushed the envelope as far as they dared.
1.4.2006 6:53pm
KMAJ (mail):
JaO,

I did not mean to infer Sealed Case would be the only reason, only wondering about any correlation. The NSA program was initiated in 2001, Sealed Case was 2002, the modifications, rejections and deferrals were in 2003 and 2004. It certainly could be because of pushing the envelope, but I would think that there is something(s) more substantive to result in the huge increase in numbers. Logic would require that an over 2,000% jump in modifications would raise some red flags that something is out of balance.
1.4.2006 7:08pm
Medis:
KMAJ,

Although increased modifications would indicate less deference to the government, so if the dicta was sending a message to the FISC, I'm not sure that message was well-received.
1.4.2006 8:08pm
Just an Observer:
KMAJ,

I wasn't just referring to the NSA program, but rather to the whole aggressive legal stance the Bush administration adopted immediately after 9/11.

You seem to infer that the shift in statistics reflected some change within the FISC. I suggest it is more plausibly a function of a change in government goals and methods.
1.4.2006 10:08pm
KMAJ (mail):
JaO,

I would not presume to argue one thing right and another wrong, there could be many plausible reasons or combinations thereof. It would be plausible to consider that FISC kept operating in a pre-9/11 mode while the government was reevaluating, thus both were not operating on the same page. The courts are notorious for changing slowly, which in and of itself is not a bad thing, but it can create loggerheads and problems when dealing with things that need quick action and decisions. I think that is one of the reasons the Founders laid out for executive authority, that both the legislative and judicial branches were cumbersome and ineffective in that role. This could simply be a confluence of those differences.
1.4.2006 10:44pm
Charlie (Colorado) (mail):

But something more must be afoot -- more than just collecting to/from phone numbers like a trap-and-trace or pen register would -- for Jim Comey, and others, to object.


Colt, given that Comey raised objections that were eventualy resolved, I don't think that's a justifiable inference. It could just as easily be that he objected until he understood that point.
1.5.2006 1:27am
Charlie (Colorado) (mail):

I agree with John Lederer that the issue isn't whether the law was bent or broken in the immediate aftermath of 9/11, but why the surveillance is still being conducted lawlessly today.


Begging the question again: we don't even know IF the surveillance is being conducted "lawlessly." Certainly the Administration maintains otherwise.
1.5.2006 1:31am
Charlie (Colorado) (mail):

Except that, as noted, the Administration says it did (but for the AUMF).


Except they didn't say that. They specifically stated reason why they think FISA didn't apply. If you sell a 14 year old a bottle of Martinelli Sparkling Cider, that doesn't mean you violated legal drinking age laws, no matter how much the bottle looks like champagne.
1.5.2006 1:36am
Charlie (Colorado) (mail):

The Attorney General clearly stated the NSA surveillance program involves intercepting communications between a party in the United States and a party outside the United States. If the program actually involved intercepting communications sent to and received by individuals located outside the United States, there is no doubt the Attorney General would have said so.


I don't think the legality of intercepting comms with BOTH endpoints outside the US was ever controversial, or required any sort of NCA decision. It's the comms with a US endpoint that are controversial.
1.5.2006 1:41am
Charlie (Colorado) (mail):

Even with the Risen book, it seems there is still much we do not know.



From your mouth to God's ears.
1.5.2006 1:42am
Charlie (Colorado) (mail):

Actually, the dicta in Sealed Case implies more than that. The now infamous line is: "We take for granted that the President does have that authority and, assuming that is so, FISA could not encroach on the President's constitutional power."

If any sort of limit or procedural requirement "encroaches" on the President's power, then all of FISA is out, and so for that matter would be the entire US Code whenever the President wants to order someone to violate it.


Medis, I don't have the case in front of me (and I'm not going to read it again right now!) but as I recall, the sentence preceding that make this explicit as a statement about the president's ability to collect foreign intelligence.

You know, I wonder if this isn't where the whole "chinese wall" thing comes from? The recognition that it's permitted for NSA and CIA to collect, but that FISA says it can't go to FBI et al when it's a "US person".
1.5.2006 1:48am
Robert Romano (www):
AJ Strata wrote:


And this seems to be Robertson's biggest problem with NSA and FISA- NSA intel being used for probable cause and 'tainting' the FISA process... If the entire faux issue is how intel promulgates to law enforcement without using the intel, then this is a lot of partisan hot air which will burst easily.


This is exactly the point I raised on another blog last week, on Dec. 29th:


...[L]et's assume that these sorts of intercepts led to an eventual request for a warrant for broad-based monitoring of the terrorist suspect. Clearly, with the resignation of the Judge from the FISA court, he was saying that this possibility is distinctly the reason for his resignation. And now terror suspects are trying to get any evidence which may have orignated from the intercepts thrown out of court. But besides the intercepts, how else could the government know that the person was involved with the bad guys? How else would probable cause that the person was linked to the terrorists be established? Are the critics suggesting that the President must prevent future terrorist attacks on the United States, but that he is not supposed to use those technologies which make that possible?

I thought we were removing the wall which separated intelligence from law enforcement...


My question is: if warrants can only issue upon probable cause constitutionally, how is an intelligence agency supposed to establish that without using intelligence-gathering techniques? I think it would be appropriate, when probable cause is established, that a warrant could be obtained where necessary, but not beforehand.

Finally, can intelligence be used in law enforcement or not? If so, then how does an intelligence agency establish probable cause without using these sorts of techniques?
1.5.2006 1:50am
Robert Romano (www):
To ask the question a different way, if intelligence can't be used in law enforcement, then why is anybody expecting the President to gather intelligence based on probable cause and with written warrants?

How many thousands of warrants would they have to request on a daily basis anyway, if every intercept, whether or not it is analyzed, is a search?

Are folks suggesting that government must demand a warrant based on less than probable cause? They can only get a warrant for broad-based monitoring based on probable cause. If they cannot use intelligence-gathering techniques to establish probable cause, then I do not see how the AUMF with language that "The President is authorized to use all necessary and appropriate force... in order to prevent any future acts of international terrorism against the United States by such nations, organizations or persons..." would have any weight at all.

Who is to decide what is necessary? Congress? The Courts? Or the President? My contention that this is a political question with which the President has discretion.
1.5.2006 2:09am
Medis:
Charlie,

Well, of course that sentence refers to FISA, so it is only directly talking about foreign intelligence surveillance (the subject matter of FISA). My point was simply that if there is some general rule against "encroaching" on the President's "constitutional power", and if any legal limit or procedural requirement counts as "encroaching", then the entire US Code potentially "encroaches" on the President's "constitutional power".

Of course that is not what the Supreme Court has said on this subject, so on such an interpretation that dicta must be wrong. Which doesn't necessarily mean the dicta is wrong--it just means the dicta is unclear, because it doesn't explain how to avoid such absurdities. In other words, as I noted, the basic problem is that this dicta contains no definition of "encroach", and without knowing what that verb is supposed to mean, it is difficult to know what that dicta was supposed to mean.

Robert,

FISA requires a court order only for certain forms of electronic surveillance, and the relevant probable cause standard is actually very different from the criminal standard. So, FISA warrants aren't really like criminal investigation warrants.

As for how the government can establish probable cause for a FISA warrant: to begin with, not all intelligence gathering is electronic surveillance. To give a simple example, if we capture some member of Al Qaeda with his address book, and in that address book it says, "John Smith, 101 Birch Street, Anytown, USA (one of our US operatives)," we can take that evidence to the FISC in order to establish probable cause for surveillng John Smith.

Moreover, we can also use lawfully-obtained electronic surveillance to establish probable cause. So, if we start monitoring John Smith, and he says in a phone conversation, "I'm going bowling tomorrow with Bob White, my fellow Al Qaeda operative," we can use that evidence to establish probable cause for surveilling Bob White. And so on.

As for why we started requiring court orders for certain forms of foreign intelligence surveillance--basically, it was because the government had been abusing this power in the 60s and 70s, so in 1978 we instituted a system of limited judicial review, which in turn allowed limited congressional review.

And incidentally, that seems like a pretty sensible answer to your final question: when it comes to deciding what are necessary and proper government actions, all three branches of government should have a role to play. Choosing just one would be a dangerous concentration of power, something the Founders intended our Constitution to avoid.
1.5.2006 7:38am
corngrower:
Medis

you refuse to address my point,

your opinion is that the courts are the final thing in all matters. But via my 5th grade government class I learned the the United States of America is governed by 3 co-equal branches of government. Your contention is that you disagree with the executive branches usurpation of power. Violated a law, written by congress. That is your opinion. Fine. So? You quote the FISA law chapter and verse. YOU interpet the language the way you want. The only quotes I have not seen, are from the person that wrote the bill. The person that wrote the bill, the people that voted to enact the bill, the person that signed the bill into law, are strangly SILENT! GEE! Why is that? You will the judiciary to solve a problem(in your mind)that needs to be solved by those agrievied.

I will fully explain my privious post, you seem to not be able to grasp my thought. You put all your faith in the law. You quote the law chapter and verse. You belive Judges should by the final arbiter of conflict and use the laws enacted and follow the Constitution. I do not trust judges. I gave one example of why I feel that way. What I wanted from you is Quote the portion of the law, mesh the law with the constitution and give those two things to me to prove that, in fact, the five smartest lawyers in the world overturned the will of the people. Going back to my 5th grade schooling I cant find how SCOTUS can overturn a state law that allows the execution of a convicted criminal.

So pony up your vast and deep knowledge of all that is legal and explain to me what happened in this one case, and then square that with your opinion that the courts are the solution to your current squawk.
1.5.2006 10:40am
Medis:
corngrower,

The reason I "refuse to address" your point is that you are attributing to me a viewpoint that I don't hold and then asking me to defend it.

For example, you say: "your opinion is that the courts are the final thing in all matters." But, of course, that is not my opinion. I do think that the courts possess the judicial power of the United States, which means they do have the final say on how to interpret the law, including the Constitution. But Congress possesses the legislative power of the United States, and the President possesses the executive power of the United States, and it is through those other, non-judicial, powers that the other branches participate in "all matters" of government.

You also say: "You will the judiciary to solve a problem(in your mind)that needs to be solved by those agrievied. . . . You belive Judges should by the final arbiter of conflict and use the laws enacted and follow the Constitution." Again, that is not my view, because I have advocated a role for all three branches. It is true that insofar as these issues do arise in court cases, the courts will have jurisdiction to address them. But the President also has the duty to faithfully execute the laws, and that includes not just bringing criminal charges, but also making sure that he only orders people to do lawful things. Finally, if the President refuses to obey this constitutional duty, then the remedy is not going to be in the courts--it will be in Congress through the impeachment and removal process.

Finally, you say: "So pony up your vast and deep knowledge of all that is legal and explain to me what happened in this one case, and then square that with your opinion that the courts are the solution to your current squawk." First, I have never claimed to have an encyclopedic knowledge of "all that is legal". And although I do find the Supreme Court's struggles with the death penalty in recent decades interesting, I claim no expertise in that area. And while I would be happy to discuss that matter with you elsewhere, it is not particularly relevant to the discussion we are having here.

Moreover, you are again staring with a faulty premise: that I believe the courts are infallible. Just the opposite is true: I believe that judges are human, and therefore prone to error, and sometimes even bad behavior. I also think legislators are human, and so are Presidents, and thus all these people are prone to error and sometimes even bad behavior.

That, of course, is a large part of why our Constitution is designed to prevent the concentration of power into the hands of any one branch. And that is why, contrary to what you have claimed, that I do not put my trust in any one branch, including the judiciary. Rather, I am advocating an ongoing role for all three branches precisely because I trust none of them on their own.

Now, if you want to argue that I SHOULD instead place all my trust in one branch--in this case the executive branch-- then I would be happy to have that debate with you. But there is really no point in having that discussion if you insist on starting with false claims about what I actually believe.
1.5.2006 12:00pm
Just an Observer:
John Lederer,

Thanks for your response to my questions yesterday on definitional issues raised by the common-carrier status, or lack thereof, of the fiber-optic cable providers.

Given what has been reported about the contours of the NSA program, I am more interested in what kind of taps might be installed within the United States with the connivance of the cable operators than by a more surreptitious method.

Following up, it seems that the common-carrier loophole might obtain not only with submarine cables, but also domestic cables.

So if the major-telecom cables coming into the switch from the domestic side are owned by a non-common-carrier operator, either that domestic operator or the submarine operator on the other side of the switch might be able to host the NSA's tap. Is that what your research teaches you?

On a related matter, I found it interesting to see how the definition of "wire communication" in FISA compared to its domestic criminal-wiretapping analog in Title 18.

The FISA definition 50 USC 1801(l):

(l) "Wire communication" means any communication while it is being carried by a wire, cable, or other like connection furnished or operated by any person engaged as a common carrier in providing or operating such facilities for the transmission of interstate or foreign communications.


The domestic definition 18 USC 2501(1):

(1) "wire communication" means any aural transfer made in whole or in part through the use of facilities for the transmission of communications by the aid of wire, cable, or other like connection between the point of origin and the point of reception (including the use of such connection in a switching station) furnished or operated by any person engaged in providing or operating such facilities for the transmission of interstate or foreign communications or communications affecting interstate or foreign commerce;


Interestingly, no common-carrier requirement applies in regulating purely domestic wiretapping. (The domestic definition does seem to apply only to "aural" communications, not data.)
1.5.2006 12:13pm
corngrower:
Medis

I return to MY original point. IF someone in the legislative branch has a bitch with how the executive branch is using a law....FIX IT. How do I say this in a more clear and concise manner. BUT there is a big if. Does a single person elected to congress have a problem with the law???? See? If one single elected person to congress offered a law and narrow NSA reach into our lives, I have not a single problem. But. The only people squawking are people that disagree with the current executive branch.

And so.... that makes it a political, rather than a legal problem. So stop interpting the law to your own slant and get busy in an attempt to find even a single politician with a spine that is willing to offer up a bill.
1.5.2006 12:48pm
Medis:
corngrower,

The precise problem is that there already is such a law.
1.5.2006 12:58pm
Robert Romano (www):
Medis wrote:


FISA requires a court order only for certain forms of electronic surveillance, and the relevant probable cause standard is actually very different from the criminal standard. So, FISA warrants aren't really like criminal investigation warrants.


Well, being a bit ignorant of the nuance you're pointing towards between one probable cause standard for criminal investigations and another for FISA warrants - just an undergraduate here - how can there be a different standard?

Either there is probable cause or there is not, from what I can tell. Obviously, instead of probable cause that a crime has been committed, we're looking at probable cause that U.S. citizen in question is communicating, coordinating, or connected with the terrorists.

From what I'm reading here, the chief complaint against the NSA program is that it is being conducted without warrants. In other words, that the surveillance is reasonable only if it is done via court order. Now, the FISA allows for the 72 hour after the fact warrants, and this too would have to be based on probable cause.

That would mean, if - and this is the big if - the NSA program were subject to the restrictions of the FISA, that no intercepts can occur without there being probable cause. That would also mean that every single intercept would have to be reviewed to establish probable cause if it involved a U.S. citizen, correct?

However, based on what I've read about the program - and I can't possibly know the details - it appears that in some circumstances, there will be less than probable cause to ask for a warrant. Now, if it were the government's goal to establish probable cause via these sorts of techniques, knowing that not all communications would be relevant to establishing probable cause, then why would the government have to apply for a warrant based on less than probable cause?

Or in other words, does the government need to ask for a warrant for every intercept which involves a U.S. citizen, or not? If so, then I could see what folks are complaining about.

Shouldn't the government just disregard irrelevant information? And along those lines, when it wants to apply for broad monitoring of the individual in question, assuming it's a U.S. citizen, that they bring the relevant intelligence which establishes probable cause when requesting a warrant would be required?

Why would the government need a warrant to conduct surveillance which may not necessarily result in probable cause? In other words, if the resultant relevant information occurs through weeding through irrelevant information, why should the government have to apply for a warrant for communications that are irrelevant to the task of detecting and preventing terrorist acts?

If one of the ways probable cause is being established is through the program, and not necessarily all of the communications intercepted in the program are relevant towards those ends, then would it not make sense that not all intercepts would be conducted with a warrant?


As for how the government can establish probable cause for a FISA warrant: to begin with, not all intelligence gathering is electronic surveillance. To give a simple example, if we capture some member of Al Qaeda with his address book, and in that address book it says, "John Smith, 101 Birch Street, Anytown, USA (one of our US operatives)," we can take that evidence to the FISC in order to establish probable cause for surveillng John Smith...


Okay, so one way could be through obtaining information via a known terrorist operative.


Moreover, we can also use lawfully-obtained electronic surveillance to establish probable cause. So, if we start monitoring John Smith, and he says in a phone conversation, "I'm going bowling tomorrow with Bob White, my fellow Al Qaeda operative," we can use that evidence to establish probable cause for surveilling Bob White. And so on.


And another way would be to obtain information from a known terrorist operative.

So, exactly how is the government supposed to gather information on unknown terrorist operatives that are acting independently of known terrorist operatives without a program like this, which filters relevant information to do with terrorism from irrelevant information?

Intelligence is not necessarily gathered in the first place based on probable cause. Signals intelligence seems to have entirely less to do with probable cause in the first place since it would be filtering relevant information from irrelevant information.

So, I go back to the question: assuming the goal is to obtain intelligence on unknown terrorist operatives acting independently of known terrorist operatives, how can that be done without this sort of program in place?
1.5.2006 8:08pm
Medis:
Robert,

Sorry--that is what I meant by a different probable cause standard: the government has to satisy the court there is probable cause for something different.

As for your other questions: I only gave two examples, each of which did in fact involve another operative, but any sort of lawfully-obtained information might establish probable cause. For example, someone could witness the operative doing something suspicious (eg, casing the Brooklyn Bridge), and alert the authorities. They could conduct surveillance that did not require a warrant (eg, trail the guy in public, gather public records, go through his publicly-accessible trash, and so on), and acquire enough information to establish probable cause for a FISA warrant and to commence electronic surveillance.

However, I think you are basically asking a different question--if this sort of electronic surveillance program would be a particularly effective (although not necessarily the only) way of gathering information that could sort the suspicious from the innocent, who could then be targeted for ever-increasing levels of surveillance, then why isn't it legal to use this form of surveillance for this initial (pre-probable cause) sorting?

And the answer is basically that we are worried about how the government could abuse this sort of power, as well as about invasions of privacy (perhaps even if we never know it happened).

So, this is a typical policy-balancing issue, and that is why Congress is mostly in charge, although there is also a general policy decision on this issue written into the Constitution (and this is not an entirely new issue--just being able to open and read people's letter mail would have been an effective way to establish probable cause as well).

As for why we don't let the sorting occur first and then just discard irrelevant information: actually, various people think that as long as no human being is involved in such an initial process, and the filters are properly programmed, such sorting should not require a warrant. Still, others think the same potential for abuse exists and some still have concerns about invasions of privacy--even by computers.

So, that is the basic setup: it is a complicated tradeoff between competing values. Hence the need for law.
1.5.2006 9:10pm
Robert Romano (www):

However, I think you are basically asking a different question--if this sort of electronic surveillance program would be a particularly effective (although not necessarily the only) way of gathering information that could sort the suspicious from the innocent, who could then be targeted for ever-increasing levels of surveillance, then why isn't it legal to use this form of surveillance for this initial (pre-probable cause) sorting?


Well, as it pertains to this particular case, with communications of an international nature. Either going overseas or being received domestically from overseas. But yes, that's the question.


And the answer is basically that we are worried about how the government could abuse this sort of power, as well as about invasions of privacy (perhaps even if we never know it happened).


I'm more worried about the gaps in intelligence we would have if we did not pursue this sort of program.
1.6.2006 12:36am
Medis:
Robert,

You are, of course, entitled to prioritize that concern. But we live in a democracy, and all of us are supposed to get a say in how these issues are resolved. In part, we have had our say through the 4th Amendment. But for more specific issues, we have elected representatives to Congress, and their job is to represent our views and interests, debate these issues, and ultimately decide the question.
1.6.2006 10:29am
Medis:
Robert,

Sorry, I forgot one point. You, of course, limited your views to this particular factual circumstance, as you are entitled to do. But I would note that some people (including a number of well-respected thinkers on this subject) see no principled reason for limiting such warrantless surveillance to international communications. Indeed, some suggest that the entire notion of "privacy" is overrated, and we would be much better off if we generally had much less ability to keep secrets from each other.

I don't note that to suggest you are on some sort of slippery slope. Rather, I note that only to suggest that insofar as you too might want to preserve any sort of privacy in communications--or any sort of privacy at all--you are also trying to draw the correct balance between security and these other concerns.
1.6.2006 10:35am
corngrower:
Medis

You are so confusing to me. You state we already have a law, yet, you interpet in a different manner than others.
Then you state we are a representative democracy and the 'people' should decide thru their elected representatives.

That has been my contention from the start. Got a beef with whats going on? Introduce a bill. But, I will go you one better, name a single elected person of congress that has railed about the problem that 'you' see, and tell me what bill is introduced to clarify the meaning of the law in question.
1.6.2006 10:49am
Medis:
corngrower,

As an aside, someone recently posted a list of comments by Republican Senators expressing their concerns about this program. So you might find it interesting to search for that.

Anyway, I really don't understand your point. When someone breaks the law, the proper response is not to pass another law. That makes no sense, in fact--if they were willing to violate one law, then another law won't help.

Rather, the proper response is to indict and try the person who broke the law. Of course, if you are talking about the President, it would have to be Congress doing that, through the impeachment and removal process.

But obviously that is not something Congress will do lightly (well, at least not this Congress with this President). Prior to initiating the impeachment process, Congress would want to investigate. Indeed, that is basically normal procedure: before you indict someone you have to investigate them.

And we already know that Congress is going to investigate this matter. If Congress determines that the President has not broken the law, that will be the end of the matter (at least as far as impeachment is concerned). But if Congress determines that the President has indeed broken the law, then they will have to consider impeachment (although even then they may not conclude these are "high crimes").

But somehow you seem to think that I think that Congress should leap right into impeaching the President without any sort of investigation. And I definitely do not think anything of the sort.

And frankly, I hope the President has not in fact violated the law, and thus that there will be no grounds for impeachment. Unfortunately, from what the Administration has said so far, I can't rule out that possibility. But if further investigation reveals that the President did not violate the law, I will be happy.
1.6.2006 11:40am
corngrower:
Medis

Stop the legal double speak,

I. Have. Not. Talked of impeachment.

You keep returning to the same dry well.

Maybe a more simple anology will explain.

If I thought my neighbor was selling drugs to the kids in the neighborhood, I would not write letters to the editor, I would not go the local TV and radio stations. I would not form commitees and hold public meetings.

I would walk into the police station and ask the chief to stop the activity. I would provide evidence of the crime committed.

Now, assume that the infraction of my neighbor is not so clear and simple as selling drugs to children?? maybe the grass is to long over there. There is a code that adressess this, but the local officials cannot agree as to if, any, or all of the code really applies in this case..

My solution?

Go to the city council and advocate a clarification of the code (law). I convince one of the persons I elected to present a clear deffintion of the existing code. If I fail at that, maybe I'm wrong, or, maybe what I percieve as an infraction is not, according to the majority of the community.
All I ask is for, is, you to get at least one elected congress person to present a bill, that shows this is not just a political cat fight.

And, yes, I have seen the statements by Republican Senators voicing their concerns. SO? Republicans are concerned over actions of a Republican Adminstration. I guess that means that 'concerns' are valid, but, not valid enough to introduce a bill.
1.6.2006 12:15pm
Medis:
corngrower,

OK, so what if the Mayor lives next door to you, and he is watering his lawn late at night when there is a no lawn-watering ordinance in effect. You ask him why he is doing it, and he claims the law is unclear.

Do you just take his word for it?
1.6.2006 12:22pm
corngrower:
Gee I don't know?

What if the neighbor on the other side does not have an issue with this activity? What if the code states that watering you lawn at an time not conducive to sleep? what if I sue the Mayor in court? the court says. Well we cant determine the meaning of watering. But, we see in international law that green grass is good, and, brown grass is bad, we have no choice but to overturn the the law in question. and rule that no infraction has taken place.

I then go and find at least ONE !! person to write a clarification to the law that states 'no lawns can be watered between the the hours of 6:00 AM and 10:00 PM.

Now you are stuck with the quandry of the fact. That, the city council and the mayor and 90 % of the town does like to do what the ordinence would prevent. So, no 2cnd and no clarification of the ordinence. But! You got One person that was elected by the people to proffer a bill to settle a dispute between the people and the admistration, (and failed because thats what the people wanted)

Huh?
1.6.2006 1:21pm
Medis:
corngrower,

You say:

"What if the neighbor on the other side does not have an issue with this activity?"

That doesn't seem relevant. Maybe he likes the Mayor, and doesn't want to get him into trouble. But that has nothing to do with the law.

"What if the code states that watering you lawn at an time not conducive to sleep?"

Right, first you look at the code to see if the Mayor's watering does in fact fall within some exception. That is what we have been doing here--looking at what the law says.

"what if I sue the Mayor in court?"

That wouldn't be a bad idea, but in this case the Mayor has immunity from court suits.

"the court says. Well we cant determine the meaning of watering. But, we see in international law that green grass is good, and, brown grass is bad, we have no choice but to overturn the the law in question. and rule that no infraction has taken place."

Right, if a case did go to court, we would have to look at what it said. But that hasn't happened, and may never happen because the Mayor has immunity.

__________

But here is the good news: although the Mayor cannot be sued by a private citizen, the Mayor can be removed from office by the City Council.

But, of course, the City Council won't just do that because you raised the issue. First, they will want to investigate. Then, they will look at the law and apply it to the Mayor's case.

So, you end up waiting to see what the City Council does.

__________

Now, suppose after investigation, the City Council decides the law was unclear, and lets the Mayor off the hook. At that point, they might consider proposing a change in the law. But that point comes AFTER they have investigated, and AFTER they have applied the existing law to the Mayor's actions.

So, that is what is wrong with your argument that someone should be proposing a new law: you are jumping ahead to the end of the process, but we are just at the very beginning--the investigation phase.
1.6.2006 1:48pm
corngrower:
Or?

Perhaps? Maybe? It is a bad law that no one wants to inforce. That even not even a single person is willing to do anything but give lip service to, and squeel about some imagined 'crime'. Why spend the time to investigate something there is considerable debate about, when in a few minutes a law could be drafted and into committe in a matter days?? Why? Because....The battle is political capital the the politicians refuse to give up.

Dont respond until you consider the truth of what I have just said
1.6.2006 4:03pm
Robert Romano (www):
Medis wrote:


You are, of course, entitled to prioritize that concern. But we live in a democracy, and all of us are supposed to get a say in how these issues are resolved. In part, we have had our say through the 4th Amendment. But for more specific issues, we have elected representatives to Congress, and their job is to represent our views and interests, debate these issues, and ultimately decide the question.


While I agree that we are a representative democracy (I like to call it a constitutional federal republic), I disagree that this is a political question with which Congress necessarily has discretion in light of the AUMF. Congress cannot, on one hand, authorize the President to do everything necessary and appropriate to prevent terrorist attacks, give him the discretion to determine what is necessary, and then quibble about what is necessary.

The President has inherent war-making powers to authorize this program for the purposes of gathering intelligence when he deems it necessary and appropriate. If the AUMF, or any AUMF, does not serve as an exception to the FISA, then in my opinion, the FISA is unconstitutional, because then the FISA would be limiting powers already in the Constitution. We only have one executive for a reason. The President has to issue the orders, and in my opinion, the restrictions of the FISA cannot apply in the face of the AUMF because it limits what is "necessary" to what is allowed under the FISA when it's supposed to be up to the President to determine what is necessary.

To touch on that point, we're supposed to be looking for targets. Am I to believe the President could target a terrorist in the U.S. with a smart bomb no questions asked but that he would need a warrant to intercept the terrorists international communications? Where's the proportionality here?


I would note that some people (including a number of well-respected thinkers on this subject) see no principled reason for limiting such warrantless surveillance to international communications.


Well, I could think of a few principled reasons for and against. There must be a balance between security and privacy, in my opinion, favoring security. This is the standard I would use, or codify (though it probably needs tweaking yet):


Under this principle, the truth should be known to all except when it A) poses a threat to the peace and security of society and to the lives and liberties of the people, in which case the governing body ... identifies and documents all information [pursuant to the danger] so as to make it reasonably available to certain positions and offices open to all, and made publicly available [so long as this itself does not produce a new danger] when the certainty of the threat is fully realized and proven based on a preponderance of the evidence, or B) infringes upon individual liberties as articulated in the Federal Constitution. In other words, the truth should be known to our civil government when there is a substantial danger to society and to the lives of the citizens in order to best protect the citizenry, and the truth should be hidden from all when it violates personal liberties, in particular, one's reasonable expectation of the freedom of privacy from intrusive means and measures that would bring about harmful consequences... Condition A, or the overall peace and security of society, overrides condition B, or the reasonable expectation of privacy from intrusive measures. Thus, it is more just to live in a society where the government can be reasonably expected to deal with threats posed to all than in one where the rights of privacy are applied under all circumstances.


This is purely a philosophical argument, though I wanted it to be workable in the context of our constitutional protections. My view is that if we applied the foregoing principle, we can prevent attacks. But if we twist it around, with privacy trumping security, we cannot conceivably prevent attacks. Therefore, I look at the Constitution and do not automatically assume that there is an impenetrable seal of privacy. I do not believe in such a seal.

I also do not believe that gathering intelligence on targets in a war ought to be subject to the same standards as gathering evidence of crimes which have already been committed. In the former, we're seeking out information on potential and probable acts of war against our homeland. I include "potential" attacks because the government needs to know about those, too, to ascertain if they are probable. In gathering intelligence, we ought to start with potentialities, and move up from there, prioritizing our resources to probable dangers.

As for the issue of international vs. purely domestic intercepts, I would favor there being no distinction, but not because I do not see no principled reason not to do so. I do see principled reasons not to do so. However, those, in my opinion, are outweighed by other principled reasons. There are going to be conflicts between powers and rights. My view is that when we're talking about a danger to peace and security from foreign or domestic causes, there is no reasonable expectation of privacy. For our enemies, there must be no protection.

(If somebody doesn't even accept the proposition that we're at war, don't even bother with the notion of enemies as described above.)

I would hate to set up a system whereby all the terrorists had to do was recruit U.S. citizens, or make certain that their communications were purely domestically based in order to ensure the secrecy of those communications. A danger I think that comes out of revealing this program is that the terrorists may be switching over to phones with U.S. area codes in hopes of avoiding detection.

Let me know what I could do to tweak the principle. And try rationalizing what a system we would have if the protections of privacy were applied under all circumstances, or if they were to trump dangers to the peace and security of society. I understand then it becomes less about what the law is (which I know less about, so I hope you understand), and more about what folks think is reasonable.
1.6.2006 4:29pm
Robert Romano (www):

[I]f this sort of electronic surveillance program would be a particularly effective ... way of gathering information that could sort the suspicious from the innocent, who could then be targeted for ever-increasing levels of surveillance, then why isn't it legal to use this form of surveillance for this initial (pre-probable cause) sorting?

In determining what is "necessary and appropriate force", whether or not the program is more effective than other means should be a standard that I think is considered, and whether it is proportional to the danger. If using signals intelligence is more effective than traditional law enforcement techniques, then I think it's allowed for under the AUMF. In determining what is necessary, if it were up to me, I'd use those techniques that are the most effective. Is the program proportional to the danger involved? If the program is meant to cope with gaps in our intelligence which existed prior to 9/11, then I think its allowed for under the AUMF. In determining what is appropriate, I'd limit the use of those effective techniques to identifying potential and probable dangers to peace and security.

So, if the program is the most effective as is practicable to prevent terrorist attacks, and is limited to identifying terrorist dangers, then it is, necessary and appropriate under the AUMF.
1.6.2006 4:56pm
Robert Romano (www):

I don't note that to suggest you are on some sort of slippery slope. Rather, I note that only to suggest that insofar as you too might want to preserve any sort of privacy in communications--or any sort of privacy at all--you are also trying to draw the correct balance between security and these other concerns.


Thanks. I appreciate that. Let me know how I did.
1.6.2006 5:10pm
Noah Klein (mail):
Robert R.,

I think in your past three posts ignore the second that you quote: "appropriate." I have always assumed and I believe case law backs me up that "appropriate" means abiding by the law. The Congress is granted many powers in Article I related to war including making the regulations governing the military forces of the U.S. It therefore does not seem to me to be unconstitutional for Congress to provide the framework that the executive must conduct to surveil U.S. citizens in the U.S.

Second, I do not think the framework provided by FISA is very restrictive or extreme considering our recent past, where the chief executive conducted warrantless surveilance on U.S. citizens, including anit-war activists, Martin Luther King, the Kennedys and a variety of political opponents of various administrations. These administrations asserted that the surveillance was needed for national security. The FISC does not require probable cause of a crime, but only probable cause of being an agent of a foreign power. I think this is a pretty low standard and that is why the FISC has rarely rejected or modified a application.

Finally, I wanted to make sure that you are not saying that it is legal or constitutional for a President to bomb a U.S. Citizen in the U.S. I hope that is not what you are saying, because that is pretty extreme by any standards.
1.6.2006 8:14pm
Medis:
corngrower,

I have carefully considered your comments. And I still think that before Congress did anything else, it would want to investigate what is actually happening first. Which makes a lot of sense--don't you think?

Robert,

If I understand it correctly, one problem with your principle is that any application of what you call Condition B will somewhat diminish Condition A. In other words, anything less than full access to all information, all the time, makes it slightly harder for the government to protect us, and therefore Condition B will always be overridden. The only way to prevent this is to somehow provide for when Condition A does not override Condition B--in other words, to balance them, as opposed to making one always override the other.

Another problem is that you are entirely missing an important factor: the government's potential abuse of its power. What you are laying out is an idealized sense of what the government could do, but governments are made of human beings, who are prone to error, and sometimes corruption, and sometimes crime. So, your institutional design cannot assume that everyone in the government will use their power wisely and morally.

Finally, when you say that FISA could not constitutionally prevent the President from using his war powers to do whatever he deems necessary, you are no longer talking about "inherent" powers--you are talking about "exclusive" powers, meaning you are saying that the President can ignore the legislative power of Congress as applied to war.

But as Noah points out, the Constitution specifically enumerates an extensive role for Congress in passing laws to regulate war and the military in general. So your thesis that the President has not just inherent, but also exclusive, war powers is not supported by the Constitution. And I might note this view goes against a great deal the Supreme Court has said, including as recently as in Hamdi--a case interpreting the 2001 AUMF.

Of course, that does not mean Congress can do whatever it wants in war. For example, it cannot issue specific military orders to military units, nor can it place someone above the President in the command hierarchy of the armed forces, nor remove the President from the command hierarchy of the armed forces. But it can pass military laws and laws of war--and like any commander in the military, the President is bound to obey those laws when he gives his orders.
1.6.2006 8:39pm
Medis:
corngrower,

FYI, if you are interested, someone posted a memorandum from the Congressional Research Service on the "GW Panel" thread. As you will see, Congress is in fact moving ahead with assessing the President's claims.
1.6.2006 9:33pm
Robert Romano (www):
Noah Klein wrote:


Finally, I wanted to make sure that you are not saying that it is legal or constitutional for a President to bomb a U.S. Citizen in the U.S. I hope that is not what you are saying, because that is pretty extreme by any standards.


Provided that the citizen was an enemy combatant, that is precisely what I am suggesting. Was not Rigoberto Apilzar, the man who was shot by air marshals in Miami, posing a clear and present danger when he claimed to have a bomb? Now, it turns out he did not have a bomb after all. However, I do believe that the deadly force which was used was necessary considering the circumstances.

To create a hypothetical circumstance, if a couple of U.S. citizens were carrying dynamite strapped to themselves as they were about to board a train in Penn Station, I hope they would be shot on sight by the National Guard, before they had an opportunity to detonate the dynamite. Another circumstance, if a couple of U.S. citizens were flying a small airplane towards a civilian target, and they were not veering off-course, I hope that we would shoot them out of the sky. Now, wouldn't the acts to eliminate these dangers be legal under the AUMF? I believe that America is a theater in the global war on terror, and that the AUMF should have similar effect as it has elsewhere. If it's okay to drop a bomb on an enemy target in Afghanistan or elsewhere, it should be okay to drop a bomb on an enemy target here in America. Correct?
1.6.2006 9:51pm
Medis:
Robert,

But there are already laws in place for when armed officers can use deadly force. Basically, the standard is that they can use deadly force when necessary to protect a person from imminent danger of loss of life or serious bodily injury.

To make a parallel argument, you would have to argue that the 2001 AUMF actually changed all those laws--as in, that the President could order some US citizen killed when no such danger was imminent.
1.6.2006 10:12pm
Robert Romano (www):
Medis wrote:


If I understand it correctly, one problem with your principle is that any application of what you call Condition B will somewhat diminish Condition A. In other words, anything less than full access to all information, all the time, makes it slightly harder for the government to protect us, and therefore Condition B will always be overridden. The only way to prevent this is to somehow provide for when Condition A does not override Condition B--in other words, to balance them, as opposed to making one always override the other.


Condition A is satisfied if the truth does not pose a clear and present danger to peace and security. If, for instance, you are talking about picking your kid up from little league practice, that would be protected under your reasonable expectation to privacy. The only circumstances when Condition A would be invoked is when we're talking about such a danger. And when it is invoked, the government has a responsibility to identify and document all information pursuant to the danger.

In other words, the government would be acting within the principle only when it is gathering information on dangers to society. So, Condition A will always override Condition B, but Condition A will not always and more than likely not apply. The priority rule I'm asserting here is covered by John Rawls in A Theory of Justice. One possibility he considers in dealing with the priority problem...


"is that we may be able to find principles which can be put in what I shall call a serial or lexical order. (The correct term is "lexicographical," but it is too cumbersome.) This is an order which requires us to satisfy the first principle in the ordering before we can move on to the second, the second before we consider the third, and so on. A principle does not come into play until those previous to it are either fully met or do not apply. A serial ordering avoids, then, having to balance principles at all; those earlier in the ordering have an absolute weight, so to speak, with respect to later ones, and hold without exception. We can regard such a ranking as analogous to a sequence of constrained maximum principles. For we can suppose that any principle in the order is to be maximized subject to the condition that the preceding principles are fully satisfied... Certainly the concept of a lexical, or serial, order does not offhand seem very promising. Indeed it appears to offend our sense of moderation and good judgment. Moreover, it presupposes that the principles in the order be of a rather special kind. For example, unless the earlier principles have but a limited application and establish definite requirements which can be fulfilled, later principles will never come into play." (Rawls, 37-8)


So, I made the principle pursuant to Rawls' priority rule. There is his quote from his book, so you have a better idea of how the rule I was using works.
1.6.2006 10:19pm
Robert Romano (www):

To make a parallel argument, you would have to argue that the 2001 AUMF actually changed all those laws--as in, that the President could order some US citizen killed when no such danger was imminent.


Medis,

To make a parallel argument, wouldn't all I need to argue is that the AUMF resulted in a state of war, and that the President can issue the same lawful orders on the battlefield as he can with any state of war?
1.6.2006 10:34pm
Medis:
Robert,

Rawls had some wise things to say about lexical ordering at the end of that quote. How does the government know I am only talking about picking up my kid, as opposed to plotting to harm someone? There is only one way to know for sure: to monitor my conversation. So, because Condition A "does not have but a limited application" and does not "establish definite requirements", Condition B "will never come into play".

On the use of deadly force: sure, that would in fact change the current rules. For example, in the battlefield, a military commander can order his forces to bomb an enemy barracks while they sleep. So, the parallel would be to say that on the authority of the 2001 AUMF, the President can order the military to blow up a citizen's house while they sleep, merely because the President believes that citizen may be an agent of a foreign power. But the law would say that is illegal, because a sleeping citizen is not an imminent danger, so the government should try to arrest this citizen instead.
1.6.2006 10:47pm
Robert Romano (www):

I think in your past three posts ignore the second that you quote: "appropriate." I have always assumed and I believe case law backs me up that "appropriate" means abiding by the law.


Noah,

Well, that would be the first I've read on that. I assumed appropriate would mean proper, or suitable for a particular circumstance, or from Hamdi and to apply it to military force, a fundamental incident to waging war. So the standard I wrote on would be applicable:


Is the program proportional to the danger involved? If the program is meant to cope with gaps in our intelligence which existed prior to 9/11, then I think its allowed for under the AUMF. In determining what is appropriate, I'd limit the use of those effective techniques to identifying potential and probable dangers to peace and security.


So, in other words, is the force being directed against terrorists, or is the intelligence being gathered from terrorists' communications? My view is that if intelligence which can result in a bombing on the battlefield is not subject to a court order, then neither is this communications interception program, which uses the most effective means (i.e. signals intelligence) and is appropriately directed at terrorists and dangers of the like kind. It's a military program, and it has a military function against transnational terrorists.
1.6.2006 10:52pm
Robert Romano (www):

On the use of deadly force: sure, that would in fact change the current rules.


Well, are we not at a state of war? The authorization of military force does not distinguish between any state, organization, or person. So, we acknowledge that it is applicable to persons living inside the United States. There was no exception in the authorization to say that it applies everywhere except the United States. It's a broad authorization.
1.6.2006 10:59pm
Medis:
Robert,

Its application is not the issue. Rather, the question is what Congress meant by "necessary and appropriate force". And the subquestion is whether Congress intended that language to implicitly repeal all federal laws as the President deems necessary. As a matter of Congressional intent, I think you will find that a tough sell.

And FISA is actually in a special category. It is not a general law, like the laws for the use of deadly force. Rather, it is specifically about foreign intelligence information, which is defined in 1801(e) to include information that is related to actual or potential attacks, sabotage, or international terrorism by foreign powers or agents of foreign powers. Moreover, 1811 makes it clear that FISA is suspended for 15 days after a declaration of war, but not longer.

So, FISA is specifically designed to deal with exactly this sort of situation, including during times of war. In that sense, it is more like the Uniform Code of Military Justice, or any other law specifically regarding the armed forces. And I think you will find the idea that the AUMF implicitly repealed all those laws as the President deems necessary a particularly tough sell.
1.6.2006 11:13pm
Robert Romano (www):

Rawls had some wise things to say about lexical ordering at the end of that quote. How does the government know I am only talking about picking up my kid, as opposed to plotting to harm someone? There is only one way to know for sure: to monitor my conversation. So, because Condition A "does not have but a limited application" and does not "establish definite requirements", Condition B "will never come into play".


Medis,

By limited application, it only applies to dangers against the foreign arms and influence, and dangers of the like kind arising from domestic causes, to paraphrase The Federalist No. 3. The definite requirements for its satisfaction or it being fulfilled is that only information pursuant to the danger can be collected and that anything beyond that would violate the principle. As a technical matter, it could be satisfied via signals intelligence which can somehow sort through information, and would only produce information on such dangers as described.

In short form, the principle is that the government has a responsibility to gather intelligence on such dangers to peace and security.
1.6.2006 11:19pm
Medis:
Robert,

But your principle is not limited to signals intelligence, because all information is potentially relevant to such dangers. For example, any room in any house could currently be in use for terrorist plots.

So, why not have a camera in every room of every house of the United States? As with the sigint, the feed from the cameras could go to a computer first, which would sort out images deemed suspicious. Those could then be monitored by an actual human being.

But this computer sorting may not be perfectly effective. So, maybe it would be best to also have a large team of human beings randomly checking the camera feeds.

And I don't see anything in your principle that rules this out--it seems to me it would be more effective than any more limited system.
1.6.2006 11:41pm
Robert Romano (www):

Its application is not the issue.


Of course it is a relevant issue. If it's okay to use military force against any nation, organization, or person involved with transnational terrorism, that does not create any exception whatsoever.


Rather, the question is what Congress meant by "necessary and appropriate force".


Well fortunately for the sake of this argument, Congress included additional language in the preamble pursuant to what is "necessary and appropriate" in terms of applicability, as pointed out in AAG William Moschella's letter to Congress:


The AUMF clearly contemplates action within the United States, Jee also id. pmbl. (the attacks of September 11 "render it both necessary and appropriate that the United States exercise its rights to self-defense and to protect United States citizens both at home and abroad").
The AUMF cannot be read as limited to authorizing the use of force against Afghanistan, as some have argued. Indeed, those who directly "committed" the attacks of September 11 resided in the United States for months before those attacks. The reality of the September 11 plot demonstrates that the authorization of force covers activities both on foreign soil and in America.



And the subquestion is whether Congress intended that language to implicitly repeal all federal laws as the President deems necessary.


Who's arguing that laws are being repealed as the President deems necessary? The laws, like the FISA, either applies under the AUMF, or it does not. This is not an arbitrary decision, it is one which is limited by the AUMF. And when it comes to the AUMF, what is "necessary and appropriate" is also limited by the AUMF against nations, organizations, and persons involved with transnational terrorism. It can't be used against non-terrorists.
1.6.2006 11:47pm
Medis:
Robert,

The AUMF does not say "military force". Moreover, the armed forces of the United States are still subject to law, so the mere fact the military was using force in the US would not mean they could violate US law. And that passage from Moschella's letter simply states what I stated--the AUMF is not limited to people in Afghanistan.

Again, the question is one of Congressional intent--did they intend the 2001 AUMF to void the laws applicable to foreign intelligence surveillance, deadly force, the use of the military in the United States, and any other law as it arises? Again, I think that is going to be a tough sell.

As for FISA--as I noted, it specifically applies to surveillance for this purpose, and during times of war. So to say the 2001 AUMF authorized the President to ignore FISA is to say that the 2001 AUMF implicitly repealed FISA.
1.6.2006 11:56pm
Robert Romano (www):
Medis,

To go to an earlier quote you made:


As for why we don't let the sorting occur first and then just discard irrelevant information: actually, various people think that as long as no human being is involved in such an initial process, and the filters are properly programmed, such sorting should not require a warrant.


You can include me with those various people. The initial process includes information that consists of 0's and 1's. To apply it with the principle, if all the computer is doing is sorting through 0's and 1's, it is not violating yours or anybody's privacy. The information that those 0's and 1's contain is not revealed if they do not constitute a danger.


But your principle is not limited to signals intelligence, because all information is potentially relevant to such dangers.


No it is not. Just because you think all information is potentially relevant does not make it relevant. Under the principle, the government only gets information relevant to the specific danger. It has to be relevant under the principle.

You're making an argument for the unscrupulous use of the principle, and how it may potentially be abused, but please do not mischaracterize it. It was specifically enumerated. If you think it is possible to abuse, that's fine, but that's not what the principle is encouraging. It's only supposed to apply to those dangers that I describe.
1.7.2006 12:02am
Robert Romano (www):

As for FISA--as I noted, it specifically applies to surveillance for this purpose, and during times of war.


Which is specifically why I've noted before that if it does not provide exception for the AUMF or any AUMF, that it is unconstitutional, because it attempts to tell the President who he can and cannot wage war against, and under what circumstances, even if the person in question is waging war against the United States.
1.7.2006 12:10am
Robert Romano (www):
Medis,

In other words, the FISA nor the 4th Amendment cannot be an absolute protection inside the United States against the same force as is incident in waging war outside the United States. Otherwise, the Congress, as it pertains to the FISA, is parcing where it is okay and not okay to wage war when it is broadly defined in the AUMF. It states any nation, organization or person involved with transnational terrorism.

Either the President can conduct signals intelligence against any nation, organization or person involved with transnational terrorism under the AUMF, or if he cannot pursuant to the FISA, then the FISA is unconstitutional, because it necessarily means that America cannot be a theater in any war.
1.7.2006 12:17am
Robert Romano (www):

The AUMF does not say "military force".


Sure it does. Check out the title:


SECTION 1. SHORT TITLE.

This joint resolution may be cited as the `Authorization for Use of Military Force'.


It's right in there. It's intended to be military force, if we're arguing over whether Congress intended the force to be of a military nature.
1.7.2006 12:30am
Medis:
Robert,

On the principle: I guess I don't understand. You agree that many of the signals that come in will not be relevant (in fact, likely they will mostly be irrelevant, with a very little bit of relevant information within). But I understand why that is OK on your principle--the government is searching for relevant information, so it is OK if it looks at a lot of stuff that might be irrelevant as long as some of it might be relevant.

So, I'm not sure on your principle why the cameras in rooms idea is out. Again, most of it will be irrelevant, but a little of it might be relevant, so as long as the government is searching for the relevant information, which can't it look through all these cameras?

I'm also not sure how you get a distinction between computers and people out of your principle. Computers are better at some things, and humans are better at other things. So, a mixed sorting system with both computers and humans is likely to be the most effective sorting system. So why not use both, given your principle?

If you want to draw these lines--sigint versus cameras, or computers versus humans--I might suggest that you do in fact see Condition B as actually balancing Condition A. And, of course, there is the other factor that I mentioned--abuses of this power--which as yet has no place in your principle.

On the constitutionality of FISA: Again, I'm not sure I understand. FISA doesn't say the President cannot order electronic surveillance during war. It just establishes certain procedures he would have to follow. That is what all military laws do: set up certain rules and regulations. And the Constitution specifically says Congress can do that.

Unfortunately, I am heading off now--but for an extended discussion of these issues, I would direct you to the memo from the Congressional Research Service, linked in the Data Mining thread above.
1.7.2006 12:33am
Medis:
Robert,

Sorry, I have time to respond to your additional post.

Despite the title, the 2001 AUMF is not plausibly just about military force. The specific provision is not limited to military force, and surely the President can use non-military resources. And in any event, even insofar as it allows using military resources in the United States, it does not say those military resources can be used in violation of the law. Indeed, consider the UCMJ--would you suggest it does not apply to the military because of the AUMF? So why is FISA out?

But now I really will go.
1.7.2006 12:38am
Noah Klein (mail):
Robert,

It looks like I am going to pick up where Medis leaves off. I do not understand your principle. Is it your opinion that the executive is able to override every law or constitutional provision which checks the executive no longer applies during wartime?

The U.S. cannot bomb a person's home. The FBI or police can raid, if reasonably sure that no innocents would be harmed. That was the whole issue with Ruby Ridge and Heaven's Gate people. The FBI did not ensure that innocents would not be harmed and raided the house and FBI agents and AFT agents were fired. They can shoot down a plane, but only with explicit orders from the president. This is what we learned on 9/11.

I am sorry if I misrepresented your opinion.

Noah
1.7.2006 1:05am
Robert Romano (www):

On the principle: I guess I don't understand. You agree that many of the signals that come in will not be relevant (in fact, likely they will mostly be irrelevant, with a very little bit of relevant information within). But I understand why that is OK on your principle--the government is searching for relevant information, so it is OK if it looks at a lot of stuff that might be irrelevant as long as some of it might be relevant.


No. It would only be okay to look at the relevant information. If there could be a computerized sorting process to parce that information which is relevant from that which is not, nobody would be looking at any irrelevant information whatsoever. That's the distinction I was making. It would be akin to an officer on patrol that disregards all non-crimes that he witnesses throughout the course of the patrol, or a soldier on patrol that disregards all non-enemies that he encounters.


So, I'm not sure on your principle why the cameras in rooms idea is out.


For the same reason that the principle is limited in its application to only those dangers to peace and security. You would not start monitoring a household, for instance, under the principle, unless what was going on inside was a danger. However, I suppose you could do things like set up radiation detectors, and chemical and biological agent detectors, and bomb detectors, outside of buildings and on streets to locate dangerous materials. And once a red flag goes up, that could result in monitoring of that building, for instance. Under the principle, there has to be a reason to believe that there is a danger, or Condition A would not be invoked. So, the detectors would act in a similar fashion that the computer acts, finding information relevant to a danger.

However, I see your point. If it's okay to use a computer to sort through communications, or to set up detectors outside of locations, then why not sort through camera videos inside of homes similarly in accordance with the principle? I suppose that if a computer could do all that, we wouldn't even need 911 then, right? It would spot all sorts of suspicious activities, crimes, and dangers in real time, throw up red flags, and alert the authorities. Under the principle, it would work only if the authorities (actual persons with the job of protecting society from dangers) were alerted in the event of dangers being detected.


If you want to draw these lines--sigint versus cameras, or computers versus humans--I might suggest that you do in fact see Condition B as actually balancing Condition A.


Well, there is a balance which I've drawn, but not a perfect one. Under my principle, the interests of security would override the reasonable expectation of privacy, but not eliminate it.

But then again, I'm not perfect either. How to better draft an abstract principle related to security, if not to weight the interests of security versus the expectation of privacy? I doubt there could be a principle where both are co-equal, since what if there are conflicts between the two?

I'm open to suggestions, it would certainly forward my philosophical inquiry, and I could modify the principle with more constraints.
1.7.2006 3:00am
Robert Romano (www):

On the constitutionality of FISA: Again, I'm not sure I understand. FISA doesn't say the President cannot order electronic surveillance during war. It just establishes certain procedures he would have to follow. That is what all military laws do: set up certain rules and regulations. And the Constitution specifically says Congress can do that.


My gripe with the FISA is that if an AUMF is not an exception to it, even one as broad as the one passed in 2001, then it eliminates the United States as being a theater in war, even if there are individuals, organizations, or nations waging war in the United States. That would mean what is permissible on the battlefield in foreign countries is not permissible here at home, or what is permissible in gathering intelligence on targets abroad is impermissible here at home. It sets up a huge distinction between the homefront and the battle overseas, even if the battle is being waged here at home.

This, in my opinion, sets up at least two different standards for fighting the war, depending on the theater. It would afford the enemy more protections here in the United States than it affords them in their homelands.

There can only be one set of rules of war, and they must apply across the board. My view is that if it's an incident to fighting war, it would be permissible regardless of the theater.

Can a defensive war be waged inside the United States or not in your view? If so, do the rules of war apply the same or differently as they would apply elsewhere? If differently, and if they serve to lessen the powers the government has in its own homeland, how is that not encroaching on the government's ability to wage war inside its own homeland? Why two different standards?

How can the President have the authority to do all that is necessary and appropriate to wage war in foreign countries, but not in the United States? It would limit the sets of orders he could give for actions inside the United States. Wouldn't that make the United States a safe haven for our enemies?
1.7.2006 3:27am
Noah Klein (mail):
Robert,

You are basically advocating putting cameras in people's homes as long as the information is not viewed by a person, but instead is viewed and sorted through by the government. You don't see this as a little extreme. Your principle would completely eliminate a reasonable expectation of privacy. A person would have no privacy either in their home or car or in their electronic communications. As I read this principle, it is a clear violation of the Fourth Amendment. How can a person be secure in their persons or papers or other effects, if the government is always watching you.

You must remember Robert that it will be the executive that is setting the standards for the computer to sort through the video and vocal recordings.

Let me give you an example. Let's keep this in the area of national security and say that the administration determines that any visual evidence of violence and any vocal expression of the words "bomb" "Al Qaeda" and "infidels" would result in the review by a human. Or for that matter any google or other search engine search of Al-Jazzera or bomb-making websites would result in a review by a human. In terms of the visual recording of violence in the home, this would result in a fight between two brothers being recorded would be reviewed by a human. In terms of the vocal recording, this could often result in the capture and review of many conversations that have nothing to do with terrorism. Finally, as to the recording of searches of those two websites can have a variety of legitimate research purposes. Yet for all those surveillances, I can think of legitimate reasons that the government would want to surveil and review instances where those specific circumstances arises.

I haven't even yet gotten into the possibility of abuses. Abuses was why the entire FISA system was established in the first place. Leaving to the executive sole authority to conduct intelligence on U.S. citizens can expose Americans to rampant abuses of power. As the memo provided by the CRS showed, Every administration from FDR to Nixon violated the Fourth Amendment rights of people who were not agents of a foreign power.

I think you need to walk back a little from that ledge on which you are.

Noah
1.7.2006 3:33am
Noah Klein (mail):
Robert:

There are two different standards specifically because it is the homeland that is the theater of war. The American government is supposed to respect the rights of U.S. citizens. Therefore, in a war where we are fighting no armies, but instead a loose association of terrorist cells, the government cannot assume all authority to treat all its citizens, or even those it thinks are involved in terrorism as they would treat foreigners in a foreign country.

In a defensive war against any army or a more structured enemy, it is a different. As in the Revolutionary War, the War of 1812 and the Civil War, the property of citizens was seized, their actions were surveiled and citizens fighting to destroy the government of the United States were killed according to statutes passed by Congress.

The enemy in this case has a natural advantage because of its ability to hide among innocent civilians. Yet this advantage should not be countered by destroying the values that created this country.

Noah
1.7.2006 3:42am
Robert Romano (www):

The U.S. cannot bomb a person's home.


Even if it were an enemy target in a war? What about the Civil War? Would actions taken there have been legal or constitutional?
1.7.2006 4:00am
Robert Romano (www):
Noah,

Well, if the principle does not work, in your opinion, how would you construct an abstract principle related to security?
1.7.2006 4:19am
Noah Klein (mail):
Robert:

It is not very hard for me to form this principle, because it already exists in our government, but here it is:

A) In terms of foreign activities related to the war-making, the executive is free of many constraints, except those that were agreed to as part of a treaty, such as the Convention Against Torture, or circumscribed by Congress. Congress' actions cannot though replace the president as the chief executive authority in terms of military affairs.
Thus under this condition, the president has almost entirely free to conduct himself and the foreign policy of the United States as he sees fit.

B) In terms of national security in the United States, the president should act under the statutory and Constitutional limits on his authority. This means that he cannot overrule the statutes of Congress nor go outside the bounds that are placed on the government by various limiting provisions, such as the Bill of Rights. Therefore, the president cannot take away citizens weapons or search homes, persons or electronic communications without probable cause that they are an agent of a foreign power. Furthermore, the president cannot place troops in citizen's homes or reject their due process rights. He must also abide by all other relevant constitutional standards.
Some of these standards can be eased in times of invasion and rebellion, but those are the only exceptions.

Robert, please understand that the Founders of our country were not wholly unaware of the situation that we currently confront. I am not saying that they would have any comprehension of the technology or methods used by terrorists. It is unlikely, even a forward thinker like Ben Franklin, would have imagined jetliners and skyscrapers. Much less would they think of the use of jetliners as a weapon against skyscrapers.

Yet they led a nation through a bitter fight with a Congress that had no prescribed rules, like a constitution. Furthermore, during this war, as John Adams famously put it one third of the nation supported independence, one third opposed it and one third was neutral. They dealt with the seizure of property and territory by a foreign power, the constant attacks on innocents and military by unorganized and non-uniformed Native Americans and Loyalists and the threat of spies everywhere. Yet even during this period of great turmoil where the independence of the country and our government was threatened, unlike any threat we face from terrorism, they developed and followed Articles of War, which prescribed how the military and government can fight this war. And they respected Americans' liberty. Were there abuses? Of course. The military and especially the militia lacked discipline. Yet the government prescribed rules to protect the public.

After the war, the had to deal with internal security threats like the Shay's and Whiskey rebellions. They had to deal with impressment of their citizens into the British navy. They had to deal with the kidnapping of their citizens by North African nations. And also constant attacks by Native Americans. Yet even with these problems, they wrote a constitution designed to protect civil liberties. These were smart, dedicated and hard-working people who attempted to design a system that would balance the protection of the country with the protection of civil liberties they felt necessary to ensure a strong Republic.

The threat we face today is serious and we need to update our methods to defend our country and our understanding of the balance civil liberties and security. We did this when we passed the Patriot Act, the Homeland Security Department and a variety of other laws to defend against and attack terrorism. We will probably have to do this again. And I expect the executive to act with all energy possible within law. To the extent of even skirting. Yet I cannot advocate the executive dismissal of the laws passed by the Congress or the Constitutional limitations on the his/her power.
1.7.2006 4:51am
Medis:
Robert,

On your principle:

As suggested above, I think a lexical approach simply won't work here. I might introduce another bit of terminology: in this case, we have orthogonal values in play, including safety on one dimension and privacy on another (and I would put abuses of power on a third dimension). In such a case, we need a principle that explains how to weigh increases along one of these orthogonal dimensions of value against decreases on another dimension of value. A lexical approach basically says that any marginal increase along the higher-priority dimension outweighs any marginal decrease along the lower-priority dimension.

As Rawls and I both pointed out, such an approach essentially weights the lower-priority dimension by a factor of zero in our calculations. As our consideration of various hypothetical tradeoffs points our, that approach does not accord with our actual valuation--meaning we are not willing to give up all of our privacy (or all of our protection from abuses of power) in the name of any marginal increases in safety.

So, what you would need your principle to include is a multi-dimensional function which explains how to weight (with non-zero factors) the different dimensions. That would allow you to assess whether a certain marginal increase in safety was worth the resultant marginal decrease in privacy (and the resultant marginal decrease in protection from abuses of power). And I might note that is basically what the 4th Amendment requires.

In theory, that is not too complicated. The difficult part is in the details--how to measure and assign weights to the various dimensions such that they can be balanced against each other. And that is what you need a deliberative body, like Congress, to do.

On the battlefield versus the home-front: Of course it is reasonable for different rules to operate in these two areas during war, even if there are enemy agents trying to attack us on the home-front. And the reason that is obvious is that the rules of war are in part designed to ensure that the armed forces of the United States ultimately work to the benefit of the people of the United States, rather than to their harm.

In other words, that difference in rules just reflects the fact that the whole point of fighting wars is to benefit US citizens.

Indeed, we have had some interesting historical discussions here about the theories on this issue that existed at the time we framed our basic constitutional structure. What became clear in those discussions is that military law does not just have as its goal maximizing the power of the military, although that was one of its goals. Again, another goal was making sure that such power was turned to the advantage of the people of the United States, and not against their advantage.

I might suggest that you are again overlooking the same important dimension--concerns about abuses of power. Following the basic framework above, if military power is one dimension, then preventing abuses of that power (uses of power contrary to the interests of United States citizens) is an orthogonal dimension. And the difference in rules when it comes to the use of the armed forces within the United States is simply a reflection of the fact that we do not assign a zero weight to the second of these dimensions.
1.7.2006 11:16am
Robert Romano (www):
Noah,

I agree with you to a large extent, on many of your points. In fact, the structure of principles I inevitably formulated had several of the principles you articulated above security. I'll get to those in due course. In a practical sense, I would not argue for my principle of security as being the only one to consider. I'd put several above it.

I was hoping that you could construct a principle that is more abstract in nature. What I meant by abstract principle was to remove those other principles you might place above security and to soley think about security and privacy, and the proper balance between the two in an abstract situation. So, what I'm asking you to do is disregard any idea of history of the Constitution, etc. and just think about it from the point of view of balancing these two ideas. One either is a greater principle than the other, or they are equal (it gives you three choices in my view).

I'm working with the proposition that there is no right to endanger the peace and security of all of society, and that in this position of abstraction (Rawls would call it the original position), that nobody would agree that there is. From there, I hypothesize that if there is such a danger, that the government would have a responsibility and an obligation to know about it. And in consideration of privacy, I would also hypothesize that nobody has the right to carry forth their plot of endangering society in secret or perfect privacy.

Since there is no right to be protected for that circumstance, I rationalized the principle around that consideration.

So, in this abstract situation, your two concerns would be security and privacy. I reckon there are three choice you would have to make. Either:

1) Privacy overrides Security

2) Privacy is Co-Equal with Security

3) Security overrides Privacy

If there is another choice you can think of, too, that'd be great. But the point is to not think historically, by philosophically. Look at it from the point of view that a government has yet to be formed, and you're thinking merely about a rational formulation of principles of justice for adjudicating disputes, and the first topic that came up was security vs. privacy. Sorry if I was not clear on that.

I'll get to the plurality of principles that I ultimately decided on, for I did consider the problems you raised.
1.7.2006 11:33am
Robert Romano (www):

In a defensive war against any army or a more structured enemy, it is a different. As in the Revolutionary War, the War of 1812 and the Civil War, the property of citizens was seized, their actions were surveiled and citizens fighting to destroy the government of the United States were killed according to statutes passed by Congress.


Well, then it appears then we would be in agreement that in the case of a defensive war, the President would have far more discretion.

Our disagreement then appears to be over whether or not this is a war in which America is a theater (i.e. is it a defensive war?), and whether what is necessary and appropriate overseas should apply to the homeland.

I would say we are as much in a defensive war as any, though the factor that we are in an assymeticral war would mean that the enemy does not gather conventional forces together to fight armies. Despite that, I would still apply rules for a defensive war, and have but one set of rules of the war.

I'd premise it on the notion that the AUMF resulted in a state of war, and that Congress cannot take away from the powers of the executive to fight that war based on the theather, considering that the AUMF does not enumerate a specific theater.
1.7.2006 2:08pm
Medis:
Robert,

As it turns out, we don't have to guess what the Framers of the Constitution thought about the interaction of the armed forces and the citizens of the United States during times when the United States was itself a theater of war. The following is one of the Articles of War passed by the Continental Congress during the Revolutionary War, and readopted by the First Congress for the armed forces of the United States:

"Whenever any officer or soldier shall be accused of a capital crime, or of having used violence, or committed any offense against the persons or property of the good people of any of the United American States, such as is punishable by the known laws of the land, the commanding officer and officers of every regiment, troop, or party, to which the person or persons so accused shall belong, are hereby required, upon application duly made by or in behalf of the party or parties injured, to use his utmost endeavors to deliver over such accused person or persons to the civil magistrate; and likewise to be aiding and assisting to the officers of justice in apprehending and securing the person or persons so accused, in order to bring them to a trial. If any commanding officer or officers shall wilfully neglect or shall refuse, upon the application aforesaid, to deliver over such accused person or persons to the civil magistrates, or to be aiding and assisting to the officers of justice in apprehending such person or persons, the officer or officers so offending shall be cashiered."

So, the idea that the citizens of the United States cannot just be treated by the armed forces like the citizens of foreign countries, even when the war is going on inside the United States, is hardly new.
1.7.2006 2:44pm
Noah Klein (mail):
Robert:

I am sorry for my lengthy response in light of your clarification of the question. I agree with your basic framing of the issue as a choice of one either privacy or security or they are coequal. Thus let me give you my answer as if I were one who were going to set up a government and needed to strike the correct balance.

From a purely philosophical and idyllic point of view, I would also say that security and privacy must be co-equal. Each are extremely important values and each are necessary to the maintenance of a happy civil society. No person can enjoy his/her freedom if they are dead and no person who is secure from bodily injury can enjoy his/her security if they are oppressed.

Moving away from the idyllic, it is necessary for a person to use the historical examples of a society that either went to far one way or the other or had a balance to judge whether one system is better than another. In this examination, I will not be entirely thorough, because that would take a long time and I will exempt the U.S. because we are trying to determine which way it should be.

Security overrides privacy

Rome: Privacy was never a very high priority to the Romans. Even when it was a Republic knowledge of the actions of many of its citizens were known throughout Italy. Yet the Republic did respect the basic idea of a person having a right to property and being secure in that property. When Caesar rose to power, security of the state became more important than the property rights of its citizens. Subsequent emperors would thus often murder citizens and loot their poroperty. Eventually this system where the security of the state was most importnat fell due to the corruption of its rulers and citizens.

Soviet Union: I am going to use this example not to say this is eventually what we will become, but as example of a state where they recognized no individual rights and recognized only the security of the state. Because of the Communist belief that the state is more important than the individual, millions were murdered and property and privacy were not respected. The ultimate failing behind the Communist system led to their eventual defeat, yet also the state's dismissal of the rights of the individual made the citizens of the Soviet Union even more resentful of their government.

Privacy and Security co-equal

Britain: The idea behind privacy is more of a American creation, yet it is rooted in the Anglo-Saxon system of the right to property. Thus the idea of property rights have always been of real value to the British state. Yet even with this value they have cameras throughout the streets of London. They do not have the same warrant requirement as the U.S. does. They also have acts like the Official Secrets Act. Finally, no protection enjoyed by the subjects of the British Crown is an absolute protection because it can be overturned by an Act of Parliament. With this system, the British state has lasted over a thousand years. It is a system that has gone to extremes at times including two civil wars and variety of rebellions from the British and its colonies, but it still remains strong.

Privacy overriding Security

I cannot think of any examples of a state where privacy overrode security. The only state where this may be the case is the U.S., which I said I would not use as an example.

The history of the world is a history of the state overriding and oppressing its citizens. The state will always seek more power over its citizens. The old adage that power corrupts and absolute power corrupts absolutely is well-founded. The state's most persuasive appeal to its citizens is national security. This is because of the threat to G-d's gift to all people, which is life. Yet G-d also bequeaths to his children freedom. And the state's actions in defense of the first gift can and sometimes do destroys the gift of freedom. It has always been my opinion that what is the point to life if a person does not have the freedom to act as one see fits.

Thus in this non-idyllic world where the state and its rulers can be corrupted and can use national security threats to circumscribe the necessary freedomes of its citizens, I choose privacy over security. The state will always try to aggrandize its power and it will be most persuasive in times of war. Remember what Benjamin Franklin said "Those who would give up freedom for a little temporary safety deserve neither freedom nor safety." I have always that those who give their freedom will most likely never get it back nor will they be safe, because how can you be safe under an oppressive government.

Please don't misunderstand my position as saying that the government can't and shouldn't work to defend the nation nor that privacy is so much more important than and always trumps security. What I am saying is that there should always be a check on the actions of one authority by another or by the people. Thus secret wiretaps are legitimate, if they can be checked by the judiciary. The draft, which is the loss of freedom to help defend the state, is legitimate if passed by Congress. The detaining of Americans is legitimate if it is possible for them to have some due process. The scanning of public buildings and so on for radiological or chemical weapons is legitimate. The searching of people and bags before entering public buildings or using public transport is legitimate. And exceptions to the warrant requirements are legitimate if prescribed by the legislature or determined by the judiciary. Yet there are certain levels or places of privacy that we should never let the government intrude, because once we give up that we give up everything.

Finally, as to question about asymetrical warfare, I would assert by its very nature asymetrical warfare cannot be defensive. It may include the U.S. because of its global nature and we are definately responding to an attack on our homeland. Yet the U.S. is most certainly the stronger power and the war against terrorism will mostly be fought in other countries and thus it is hard to call this a defensive asymetrical war. Also, even if you could define it as such the absence of a military or militia organization makes fighting this war as you would any invasion from a foreign power a danger to liberties and values of this nationa.

Hope I answered your question,

Noah
1.7.2006 4:13pm
Robert Romano (www):

As suggested above, I think a lexical approach simply won't work here.


Medis,

Well to be fair to Rawls, he was thinking in terms of strict-compliance theory. In other words, in order to avoid dealing at all with corrective justice, he simply assumed, for the sake of argument, that everybody living in society would be compliant with the rules of society. This freed up his considerations so he could think of constructing a society in which everything in society and all of its values, are fairly distributed. This paves the way for his notions of distributive justice (this happens to encompass much of Nozick's critique, who was arguing against distributive justice).

So, one possibility we might consider is that a lexical ordering of principles of justice may not work in corrective justice as it does in distributive justice. However, I tend to doubt it. Let's look at the three rights identified by Locke: life, liberty, and property, and see how these rights have been prioritized.

In our own criminal legal system, the greatest crime to be committed is treason, and then murder and the like. Treason is a capital crime. Murder carries with it a greater penalty than theft, and could be a capital crime. So, the right to life is more important than the right to property, or carries with it a greater prioritized protection than that of property.

When an improper search is carried forth by the government, and the government attempts to use this in a court room, the evidence is thrown out, but no crime has been committed which results in prosecution. Or if free speech is suppressed on a State college campus, a court can rule against that action, and provides the framework for the college to act constitutionally in the future. Therefore, the right to life is still more important than the right to liberty, or carries with it a greater prioritized protection than liberty.

Also, on that latter point. One might point out that the protections enumerated are against the government itself. So, nobody can be deprived of life, liberty, or property by the government without due process of law. However, neither the government nor a citizen can deprive a person of that. The protection applies across the board. It doesn't merely restrict the actions of government, but also of the people.

So, even if we consider an example of one person depriving another of liberty, such as in kidnapping, this still does not carry with it nearly the penalty that murder does. Therefore, the deprivation of liberty is not as serious as the deprivation of life.

As far as the question of the priority rule is dealt with by Rawls, here is what he wrote:


"We must recognize the possibility that there is no way to get beyond a plurality of principles... The assignment of weights is an essential and not a minor part of a conception of justice. If we cannot explain how these weights are to be determined by reasonable ethical criteria, the means of rational discussion have come to an end. An intuitionist conception of justice is, one might say, but half a conception. We should do what we can to formulate explicit principles for the priority problem, even though the dependence on intuition cannot be eliminated entirely." (37)


If that is true, and the priority problem is real, then there cannot be co-equal principles. When there's a conflict between two claims, you have to prioritize. I don't think you escaped prioritizing whatsoever. In fact, you seemed to encourage it. But let's look at how you dealt with it:


I might introduce another bit of terminology: in this case, we have orthogonal values in play, including safety on one dimension and privacy on another (and I would put abuses of power on a third dimension). In such a case, we need a principle that explains how to weigh increases along one of these orthogonal dimensions of value against decreases on another dimension of value. A lexical approach basically says that any marginal increase along the higher-priority dimension outweighs any marginal decrease along the lower-priority dimension.


How can you do it without a lexical approach of weighting maximum constrained principles, though? If you're talking about weighing one against and at the cost of the other, aren't you still dealing with a lexical approach?


So, what you would need your principle to include is a multi-dimensional function which explains how to weight (with non-zero factors) the different dimensions.


But from reading your posts, it seems you've already established as your first principle the prevention of governmental abuses into a system of security. And along those lines, an invasion of privacy is an abuse. Therefore, privacy would have to trump security, under your lexicon. You're saying it's okay to go ahead and look at something private, but you must have probable cause that there is a danger, and it must be reviewed by a court, even if it's for security reasons.

A probable cause principle, though, favors privacy. As it relates to security, you're saying that no intelligence of a danger to security can be gathered with less than probable cause. Therefore, if the government gathers intelligence without probable cause on the danger, it's supposed to be reviewed, and when it's established in the court that there is no probable cause, then the government must delete the intelligence. It may turn out to merely be a potential danger in the court.

So, I'd say you're still using a lexical approach, but you are not dealing with co-equal principles whatsoever. Your first principle appears to be the prevention of governmental abuses against privacy which is satisfied if less than probable cause is submitted, in which case the matter must be dropped by the government.

However, I would say that gathering intelligence on dangers is not like gathering evidence on crimes which have already been committed. Maintaining security is not necessarily like prosecuting crimes. This is the rub. I'm talking about fighting a war, with targets. You're talking about enforcing the law, with criminals. And with the war on terror, we're talking about in many cases, if this is a law enforcement operation, the prosecution of crimes which have not been committed yet! We're talking about preventing acts of terror.

A probable cause principle cannot apply to intelligence on dangers which have not fully materialized. Even if an attack can be proven to be probable, or a link to terrorism probable, you'll still be dealing with lots of information that does not amount to probable cause. And chances are, it seems that you'd be adding bits and pieces of information together to establish that sort of probable cause.

But, I digress. I don't think you've dealt with the priority problem whatsoever. You've merely asserted a lexical method for dealing with it.
1.7.2006 9:54pm
Robert Romano (www):

From a purely philosophical and idyllic point of view, I would also say that security and privacy must be co-equal.


Noah,

Nice try, but have you thought about weighting one against the other at all? What if there's a conflict between the government's claims for security and an individual's claims for a reasonable expectation of privacy?


Each are extremely important values and each are necessary to the maintenance of a happy civil society.


I agreee with that. I believe they are both priorities. I just prioritize one over the other.


No person can enjoy his/her freedom if they are dead and no person who is secure from bodily injury can enjoy his/her security if they are oppressed.


I agree with that as well. But you've barely scratched the surface here. You would need to disprove that assigning weights has nothing to do with moral reasoning. Or in the least, that there is no conflict whatsoever between security and privacy. But I don't think anybody who's been talking about this thinks that.

Clearly the prevailing wisdom here is that too much security violates privacy. How can that be if they are co-equal. If they are co-equal, security would not interfere with privacy, and everyone is mistaken that security in any way violates privacy.

So try again.
1.7.2006 10:05pm
Robert Romano (www):

Thus in this non-idyllic world where the state and its rulers can be corrupted and can use national security threats to circumscribe the necessary freedomes of its citizens, I choose privacy over security.


Noah,

That's easily turned on its head. In a non-idyllic world where there are enemies to the freedom and independence we enjoy as a nation, and where evil guides their actions to take our lives and can use our good will to destroy us and our lives, I choose security to take priority over privacy.

Not to eliminate privacy. Rather, that the government has a role and responsibility to gather all information on potential dangers to determine which ones are the most likely to be perpetrated. You're waiting for an attack to be probable, it appears, and I'd say you're waiting too long to gather intelligence.
1.7.2006 10:16pm
Robert Romano (www):
As for the plurality of principles I decided upon, here's what I came up. They'll look very familiar:


1) Truth

2) Justice, Natural Law, and Natural Rights

2a) Right of Life

2b) Right of Equal Liberty

2c) Rights of Property and the Pursuit of Happiness

3) Freedom and Independence (Sovereignty)

4) Constitution, Limited Government, and the Rule of Law

5) Rights are Protected by these Institutions

5a) Security Protects Life

5b) Politics Protects Equal Liberty

5c) Capital Economy Protects Property and the Pursuit of Happiness
1.7.2006 10:35pm
Medis:
Robert,

No, a weighting system is not a lexical system, as in fact Rawls makes clear.

Consider, for example an action that would increase safety but decrease privacy. With a lexical system, we already know what we should do. If safety is prioritized, then we should do this action. If privacy is prioritized, then we should not do this action. In either case, no more information, and no further determinations, are needed.

But on a weighting system, we do not yet know what to do. We first have to measure exactly how much safety is being increased, and how much privacy is being decreased. We then must multiply the increase in safety by the relevant weighting factor, and multiply the decrease in privacy by the relevant weighting factor. Finally, we must determine the net effect--is the weighted increase in safety outweighed by the weighted decrease in privacy, or vice-versa? So, we need more information and we need to make some crucial determinations.

And that is the difference between lexical and weighting systems.

Applied to the questions before us, suppose this surveillance program did in fact increase safety but did in fact decrease privacy. Given your proposed principle, we would already know what to do--we should conduct the program because you prioritize safety.

But given the proposed principle I have sketched, we don't know what to do yet. Again, we would have to actually measure and weight the effects before making a judgment one way or the other.
1.8.2006 12:53am
Robert Romano (www):
Medis,

Ah, I got you now. So, on a case by case basis, rather than a constructed constrained principle to deal with all matters, you want to come up with a formula and weigh each circumstance on its merits based on the outcome of the formula. So, you're saying that in some circumstances security would be favored, and in others privacy, based on the outcome of the formula. And we don't know without constructing the principle.

So what you have going there is a balancing principle, as opposed to asserting priorities with absolute weight. If you could work that out, I'll be honest, it would be a significant challenge to Rawls' theory on priorities - which is no small task. If you pulled that off, you'd be quoting in philosophical journals the world over.

Do you suppose that every value will have the same weighting formula? Or would you have to construct a different weighting formula for each value?
1.8.2006 1:47am
Noah Klein (mail):
Robert:

I know you addressed your point about probable cause to Medis, but if I could I would like to answer. Due to the danger that terrorism or espionage causes to the state, FISA specifically applies a different probable cause standard than applies in criminal cases. In intelligence, all that is needed is probable cause that the target is an agent of foreign power vs. in criminal probable cause that a crime was committed. Thus FISA specifically addresses the issue that allowing terrorism or espionage is a greater threat than allowing a normal crime.

As to the points that you addressed to me, you say:

"That's easily turned on its head. In a non-idyllic world where there are enemies to the freedom and independence we enjoy as a nation, and where evil guides their actions to take our lives and can use our good will to destroy us and our lives, I choose security to take priority over privacy."

You are right. You asked me which I prioritize. I told you that in the real world I would prioritize privacy over security. The purpose behind this is that the natural desire of governments is to aggrandize their power. Placing limits on government thus allows the citizens to ensure that government fails to intrude on the G-d-given rights a person has. This does not mean that the government, especially the executive should ignore security. In fact, the first duty of any executive is to ensure the security of his/her nation. It means that the people through their representatives needs to ensure that the executive does not overstep his bounds.

I, like you and Locke, believe that life is a more important right that our civil liberties. I just fear that once government treads on our civil liberties it is a short step to tread on our lives. I tried to point this out in the historical examples above. Government controls the all the legitimate use of force. This is a natural part of society. Society yields all its legitimate use of force to their government. Even government thus can tread on one of our natural rights it is a short skip and jump to the next rights.

Noah
1.8.2006 2:48am
Medis:
Robert,

Unfortunately, there will be no publications in journals for what I have said so far. That is because it is already well-discussed, both specifically as a challenge to Rawls and otherwise. In fact, all I am doing is describing the way utility theory models preferences when there are orthogonal values--which is nothing new.

I should note, by the way, that this is just a simplified model. In the real world, it as yet would be impossible to actually derive a formula like the one I described (one which was mathematically precise and comprehensive). Moreover, people would disagree on the details of this model. So, in the real world, people basically intuit the right balance in various cases, often with the help of heuristics, and then debate with each other over the correct result, again often with the help of heuristics. All this is discussed in public choice theory.

Anyway, on weightings: it actually depends on how you are measuring the orthogonal values. In fact, it is a little arbitrary to separate measurement and weighting--safety and privacy do not automatically come in the same "units", and so when we define how a given change in safety or privacy affects distance from the origin on their respective dimensions, we have already begun the process of "weighting" these competing values.

As I described it, the next step would be to assign a factor by which to multiply these changes, and that would not have to be the same for each dimensions. But we could construct a system with identical factors simply by changing our definition of how given changes in fact are reflected by changes in the distance from the origin.

All this is just to say that since safety and privacy do not come naturally in the same "units", it doesn't make much sense to ask whether they are "weighted" the same way, or one more than the other. What we are actually doing is just modeling which tradeoffs (more saftey and less privacy, or more privacy and less safety) we are willing to make, and the "weightings" that appear in the model are more or less arbitrary in the sense that they depend on our system of measurement.
1.8.2006 9:28am
Robert Romano (www):

All this is just to say that since safety and privacy do not come naturally in the same "units", it doesn't make much sense to ask whether they are "weighted" the same way, or one more than the other.


Medis,

Well, the reason I asked was because it would seem that either all values would be weighted the same way, or one would naturally have to be given more weight than the other, and then you'd still be asserting priorities in some fashion. Even if you were using different units to weigh each one, for instance, 1,000,000 pounds of security is worth 1 ounce of privacy, clearly that 1 ounce of privacy is way more important on the scale, for if it's only 999,999 pounds of security, it's not worth the 1 ounce of privacy.

So, if they cannot be weighted in the same way, or one more than the other, then how can they be weighted at all?
1.8.2006 4:15pm
Robert Romano (www):

You are right. You asked me which I prioritize. I told you that in the real world I would prioritize privacy over security.


Noah,

So, do you believe that there is a right to privacy, which our enemies possess, to plot their designs against us?

The reason I ask that is because it appears that you are really making no claims about the privacy of the bad guys, rather, you are asserting your own right to privacy.

So, let's relate this back to the question, and get out of the abstract situation, since you've asserted your priority. According to the President, "Before we intercept these communications, the government must have information that establishes a clear link to these terrorist networks..."

Now, I'm not asking about the legality of this pursuant to the FISA, I'm going to merely ask this: If this is true, and only folks with clear links to terrorist networks are in actuality targeted, what privacy have you really lost?

If all that is lost here is the privacy of terrorists and their contacts to conduct international communications, then my next question is, are you more concerned about their loss of privacy since privacy is your asserted priority? If so, then why not also be concerned about them potentially being convicted of crimes that may not necessarily have been committed yet? Why not also challenge the President's power to declare even U.S. citizens to be enemy combatants? Why not just assert that intelligence cannot be used whatsoever in law enforcement, and advocate for "the wall"?

I think you're being inconsistent. Your prioritization of privacy seems violated by all of the above. If you're opting for privacy above security, shouldn't you dissent against the very idea of fighting a war here in the homeland?

I think putting privacy above security would be national suicide.

The purpose behind this is that the natural desire of governments is to aggrandize their power.
1.8.2006 4:39pm
Robert Romano (www):
Noah, sorry, I accidentally clicked on "post" in the middle of writing. Disregard that last sentence in my last post, that was yours. Back to your quote:


The purpose behind this is that the natural desire of governments is to aggrandize their power.


Well, that's fine, and certainly history is replete with examples of governments overstepping their bounds. Your fears are justified in that sense.


I, like you and Locke, believe that life is a more important right that our civil liberties.


Well, if you agree that life carries more weight than liberty, or privacy, for that matter, then why the heck would you value the protection of privacy more than the protection of life? Well, you do explain at least part of your rationale:


I just fear that once government treads on our civil liberties it is a short step to tread on our lives. I tried to point this out in the historical examples above. Government controls the all the legitimate use of force.


So, you're saying that while life is more important than liberty as rights, that the best way to protect life (from the government) is to prioritize liberty?

What about protecting your life against the ones who really mean you and I harm, our enemies? What do we gain by tying the government's hands in this regard?


This is a natural part of society. Society yields all its legitimate use of force to their government. Even government thus can tread on one of our natural rights it is a short skip and jump to the next rights.


I'd be more concerned about our enemies. Any greater premiums we make on our privacy rights is going to open up advantages for our enemies to exploit. That you or anyone would want to give them any given distance, which they would surely take and murder again, baffles me. Ideally, you've advocated for privacy and security being co-equal, even.

In my view, even if that were your position, you're offering more protection to your enemies to conduct their plots in private.

What if we take the President at his word, and the only folks subject to this program are actually linked to terrorists? If you're placing a higher premium on privacy, and the program were ended, would we not lose the ability to react quickly to intelligence? Why would we rationally place any procedural hurdle when seconds could count?
1.8.2006 4:59pm
Noah Klein (mail):
Robert,

That is one heck of post with a lot questions. I will do my best to answer them and then pose a couple to you.

As to your first point, I do believe that our enemies' have the same natural rights as us. BUT and this is a big but we don't have to respect the rights of foreigners. Foreigners did not yield their legitimate use of force to our government. They thus do not have any rights, except of course to life, that our government needs to respect. If we are talking about Americans, this is a different story. Americans do have rights that must be respected by our government. This does not mean though that our government must sit on its hands. On the contrary, it should act vigorously to protect our nation using all legal means.

If may ask you question, do murders have rights that need to be respected by our government? Murders flout our laws and murderers deprive people of the greatest freedom given to a person. Thus shouldn't we deprive people suspected of murdering others of any expecttation of privacy? We also know that people who commit violent crimes are more likely to murder than others. Should these after paying for their crimes lose any expectation of privacy? If the government is to respect the privacy of those that follow the law, than all people in society need an expectation of privacy.

If this is true, and only folks with clear links to terrorist networks are in actuality targeted, what privacy have you really lost?

I have lost no privacy in the circumstance described above. I have not participated or even contemplated committing terrorism against this country. But you question requires a faith that not is the government telling the truth, but also that they have not failed to make a mistake. This is the whole purpose behind the warrant requirement in law enforcement and the probable cause in this instance that the person has links to terrorism. The idea is that to invade the privacy of a citizen of this nation, which each person who has not been convicted of a crime or has served there time equally holds, an independent authority must be appealed to so that the executive does not go beyond its authority.

Would you be fine with the administration mistakenly surveiling you, because they thought you had links to terrorism?

If all that is lost here is the privacy of terrorists and their contacts to conduct international communications, then my next question is, are you more concerned about their loss of privacy since privacy is your asserted priority? If so, then why not also be concerned about them potentially being convicted of crimes that may not necessarily have been committed yet? Why not also challenge the President's power to declare even U.S. citizens to be enemy combatants? Why not just assert that intelligence cannot be used whatsoever in law enforcement, and advocate for "the wall"?

Privacy is more important than security, but it is not the only important thing. I did not say it above but privacy is only slightly more important than security. This is why there is a lesser standard when intelligence on Americans is for national security vs. criminal prosecution. I am concerned about people being convicted of crimes, they haven't committed. Remember material or connection to a terrorist group is now a crime. Why should a person be convicted for a crime they haven't committed? Do you believe that terrorists should be convicted of crimes they haven't committed? What national security purpose would this serve? I also am concerned about the administration designating people enemy combatants and claiming that these people have no due process rights to demonstrate that they are not. Do you believe that Americans should be jailed indefinately on the executive's say so. The wall is a different matter. Since I know that this intelligence requires a court order, I do not believe in a wall, because some independent authority has justified the action of the executive.

Merely because I advocate respecting the rights of people who have not yet been proven to even be criminals let alone terrorists, does not mean that we should allow our enemies to recruit U.S. citizens and employ them to harm this country. It merely means that we must recognize that we are dealing with fellow Americans and not some organized enemy on a battlefield.

I do not see why we should treat this as if it were a battlefield against an organized enemy. In such a case, the enemy wears a uniform or carries a weapon and is distinguishable from the innocent. Here we cannot say that. If we treat everyone as if they are an enemy then how are we run our country in the manner that our values require?

While respecting our privacy by going to a court to demonstrate probable cause that their target is a threat, the government still has many means to protect our nation. Merely because we ask them to respect our rights does not mean that they should allow us physical harm. Honestly, what is asked in this instance to protect privacy? We ask that the government justify itself to an independent authority. Actually an authority that almost always grants what is asked. This justification is done in secret, without an adversary to challenge the government and can be done retroactively. This does not seem like a hurdle that is so insurmountable that to abide by it threatens the security of our nation. Think about it. Do you think the founders of our nation, if told that the government can secretly tap your communications and sneak into house and search without telling the target, would think that this is legitimate for the small government that they established? Let me be so presumptous, as to answer for them. NO. Yet the threat is one that this is necessary, but checks are still needed to ensure the enemy does not become the government.


I'd be more concerned about our enemies.

I am very concerned about our enemy. I see the holes in the nation's security at its ports, airports, public places, chemical plants and so on, but securing our nation from our enemies and granting our government carte blanche to violate whatever laws and procedures it see fir are two very different things. As I pointed out above, our chemical and nuclear facilities are woefully insecure. Should the government seize these facilities and make the appropriate changes? Congress has ever since 9/11 failed to appropriate Homeland Security money based on threats, but instead has based the appropriations to give an equal amount to each states. Should the president ignore Congress's role has controller of the appropriations and give the money as it sees fit?

Finally, I would like to remind you that I am advocating for the protection of U.S. citizens from encroachment by our government on our rights. Tyrants have always used national security to trampel upon the privacy, the property and the lives of its citizens. Bush is not a tyrant, but he is advocating some scary ideas.

Noah
1.8.2006 7:33pm
Noah Klein (mail):
Robert,

But you question requires a faith that not is the government telling the truth, but also that they have not failed to make a mistake

This sentence is supposed to read: But your question requires a faith that not only is the government telling the truth, but also that they have not made a mistake.
1.8.2006 7:38pm
Medis:
Robert,

You say: "Even if you were using different units to weigh each one, for instance, 1,000,000 pounds of security is worth 1 ounce of privacy, clearly that 1 ounce of privacy is way more important on the scale, for if it's only 999,999 pounds of security, it's not worth the 1 ounce of privacy."

But this hypothetical relies on the fact that there is a fixed, defined, conversion between pounds and ounces. And that is possible because they are both units used for measuring the same thing (weight).

Now consider intead the proposition that one "inch" of security is worth one "ounce" of privacy. Which is being weighted more than the other?

And that is really the situation we are in, because security and privacy are different kinds of thing.
1.8.2006 11:25pm
Robert Romano (www):

If [I] may ask you question, do murder[er]s have rights that need to be respected by our government?


Noah,

Of course. I was clearly making a distinction between criminal law, which seeks to prosecute crimes which have already been committed, and in which searches pursuant to investigating those crimes require in most cases to my knowledge a warrant, and fighting a war, which in part seeks to identify targets to be captured or killed, and that the requirements of the former necessarily cannot apply to the latter. And thus there is a necessary distinction between the rights of our enemies in a war-time situation, and the rights of criminals as afforded due process under our Constitution.


If the government is to respect the privacy of those that follow the law, than all people in society need an expectation of privacy.


I don't think our enemies in a war-time situation have any reasonable expectation to privacy. That's part of the exception I create, and one which favors security over the privacy of those who would plot our destruction. If there's no distinction for you, that's fine. I just strongly disagree.


Would you be fine with the administration mistakenly surveiling you, because they thought you had links to terrorism?


Assuming, which I do, that there is a process in place for making certain that resources are being directed in the manner that effectuates that prosecution of the war effort in an appropriate manner, and seeks to target terrorists, if I were being surveiled, I doubt it would have anything to do with my links to terrorism. It could have to do with keeping me safe from terrorism, for all I know. In that case, I'd be perfectly okay with it, since the subject of the surveillance could have to do with preventing terrorists from making contact, in any way, with me. And would I need to know about that? I doubt it, unless it were essential to necessarily prosecuting the war.

I've already established that I would not require a court order to gather intelligence overseas, and I would not require one to gather intelligence here at home either. I want the executive to have a wide grant of authority in gathering intelligence on potential dangers, and to share intelligence with law enforcement, knowing that the means with which they come into information is radically different from law enforcement. In my view, this is necessary. There must be a different standard applied.

If the President is to have any inherent authority to conduct warrantless surveillance in gathering foreign intelligence information, even here at home, and warrants can only issue upon probable cause constitutionally, that must mean that the President has the power to conduct surveillance based on less than probable cause. He has a wider grant in gathering intelligence on clear and present dangers than law enforcement has in the prosecution of crimes which have already been committed.


Privacy is more important than security, but it is not the only important thing. I did not say it above but privacy is only slightly more important than security.


Well, security against dangers both foreign and domestic is more important than privacy, but security is similarly not the only important thing. I would state that security against these sorts of dangers is a great deal more important than privacy. I don't mind taking my shoes off when I get onto a plane, for instance. Necessarily, we are putting appropriate safeguards in place.


I see the holes in the nation's security at its ports, airports, public places, chemical plants and so on, but securing our nation from our enemies and granting our government carte blanche to violate whatever laws and procedures it see fir are two very different things.


I've granted no carte blanche to the President to violate any law. My consistent position is that he has upheld the laws and our Constitution. Your mischaracterization of my position, and frankly this redundant claim, is getting beyond being redundant, it is being redundantly dishonest, or at least mistaken. Find me a quote of any supporter of the President's position where they are affirmatively placing security above the rule of law. I certainly have not, and I resent the assertion that that is my position.

If you are saying that that is the effect of my position, that is fine. I've clearly explained what I think the effects of your position are. I don't mistake effect for intent, though. And if I've mistaken your intent, I apologize in advance.

In any event, I think we may have exhausted our debate. We're both firmly entrenched in our positions. This will have to be a circumstance where we agree to disagree over both what is legal and what is morally acceptable.

I thank everyone for an opportunity to have a debate about security and privacy. It's an important debate, and one which I am confident will favor security. :)
1.9.2006 4:21pm
Noah Klein (mail):
Robert,

I don't know if your going to come back to this thread and read this, but I wanted to answer some of your points. First, I noticed that you didn't answer my questions about whether the president can make certain decisions to seize some industries that pose dangers to the U.S. and to appropriate money as he sees fit rather than as Congress decides.

One thing I don't understand is how you don't seem to believe that all Americans should have rights. You say you don't mind the president surveiling you even though you have no ties to terrorism. What if he used that information to embarass you or to force you to no longer speak out against the administration?

You say you don't mind him violating the rights of our enemies, but some of these "enemies" are Americans. First, we have to trust that the government is actually pursuing terrorists instead of political enemies as it has in the past. Second, we have to assume that they did not make a mistake. You say you don't mind the president dropping a bomb on U.S. person with possible links to terrorism. What about collateral damage? What if he's wrong? When we fight foreign countries we don't mind as much these possibilites, but that is to defend our citizens. Now you say we can do this to kill our citizens to protect other citizens. This seems illogical to me.

Noah
1.9.2006 9:45pm