pageok
pageok
pageok
Is the Specter Bill a Good Idea As A Matter of Policy?:
I plan to respond to John Schmidt's constitutional defense of the Specter bill later today, but I wanted to address a question that Schmidt's defense curiously avoids: Is the Specter bill a good idea as a matter of policy?

  I haven't blogged on this question before, mostly because I find the question essentially impossible to answer. Almost no one knows what the facts of the NSA program are, what it does, and whether and how much it works, so I don't know how we're supposed to know whether it's a good idea for Congress to endorse it legislatively. It's like me telling you, "I'm thinking of a government program. Do you want to allow it or prohibit it?" Without knowing more, how are you supposed to answer that question? It's also hard to analyze the effect of the Specter bill's overhauling of the definitions of FISA without knowing how FISA investigations unfold; it's hard to measure the operational impact of a legislative change without knowing more about the operations. And how would Presidents use unregulated authority to conduct domestic wiretapping — what kinds of programs would they want to create that FISA currently blocks?

  Of course, you can always pick a view of the Specter bill based on fear. The more you fear Executive overreaching, the more you'll oppose it, and the more you fear an Al Qaeda attack, the more you'll support it. Or you can pick a view based on politics: if you support Bush, support the bill, and if you oppose Bush, oppose it. But if you want to assess the merits of the Specter bill in a more informed and substantive way, there doesn't seem to be much to go on.
Anderson (mail) (www):
but I wanted to address a question that Schmidt's defense curiously avoids: Is the Specter bill a good idea as a matter of policy?

I take it that the point of the post was to explain why you're not going to address the question? Or am I mistaken in supposing that to "address" a question implies at least exploring a possible answer?
9.19.2006 12:44pm
A.S.:
Um, why do you say in your first sentence that Schmidt's defense "curiously" avoids the question, when the rest of your post seems to say that it is perfectly reasonable to avoid the question?
9.19.2006 12:52pm
OrinKerr:
Anderson,

At least my experience, to "address" a topic does not neccesrily mean to provide an answer to it.

A.S.,

The Schmidt article urges Congress to pass the bill; my post doesn't. It seems strange to me that you would urge Congress to pass the bill without ever discussing whether the bill is sensible as a matter of policy.
9.19.2006 12:57pm
A.S.:
I read the article more as a defense of the elimination of the FISA exclusivity provision than as promoting this (or any particular) NSA program. Indeed, the last line of the article says that deleting the exclusivity provision would be good for this NSA program and other surveillance programs proposed by presidents in a range of circumstances.

To my mind, deleting the exclusivity provision and promoting this NSA program are separate things. OTOH, I guess I understand your point, since the two things are both included in the one bill. But I would argue, in contrast, that if I promote one provision of an omnibus bill, for example, that doesn't automatically make me a proponent of everything else the omnibus bill does.
9.19.2006 1:12pm
Just an Observer:
Specter's bill (S 2453) actually is comprised as several layers. Taken together, it is a sweeping policy change, as far-reaching as FISA was in 1978 except mostly in the opposite direction. Major provisions include:

* The overriding repeal of FISA's binding exclusivity requirement, deferring to the President's discretion to rely on his "constitutional" authority instead. This effectively make everything that remains in the statute optional.

* A provision allowing -- but not requiring -- the executive to apply to the FISA court for its blessing of a general "program" of warrantless surveillance, and having the court approve/disapprove that program and its constitutionality. As far as I can tell, that question would effectively be limited to Fourth Amendment and perhaps First Amendment issues. This is the provision Specter likes to cite as the bill's main feature. This judicial review could not reach the separation-of-powers controversy, because such controversy would no longer exist after the repeal of the exclusivity provision.

* A court-stripping provision that would move third-party challenges to surveillance from the District courts to the FISA courts.

* Far-reaching changes to FISA's definitions section, which drives most of the rest of the act. The scope of communications covered by FISA would be rewritten. Orin and other commenters have warned that the actual language might have unadvertised side effects expanding domestic surveillance loopholes.

* Greater flexibility in obtaining emergency FISA warrants. This provision is similar to that in the Feinstein-Specter bill (S 3001).
9.19.2006 1:25pm
Anderson (mail) (www):
At least my experience, to "address" a topic does not neccesrily mean to provide an answer to it.

Possibly I was reading too much into a question's being addressed, not a topic.

Anyway, the Congress, and by extension the voters, have to come to a decision. And if you're right that there's simply not enough information to make an informed choice, that sounds like a pretty good argument against the Specter bill.
9.19.2006 1:33pm
A.S.:
This effectively make everything that remains in the statute optional.

I don't think that's right. It could just be that it moves the analysis of whether the President has the authority to conduct a program that conflicts with FISA from a Youngstown category 3 to Youngstown category 2. (Of course, a program that complies with FISA would be Youngstown category 1.) Alternatively, it could just be an acknowledgement of reality - that FISA is simply NOT the "exclusive means", because the President has his own powers under Article II that Congress cannot constitutionally extinguish.
9.19.2006 1:34pm
Medis:
I think we have a little more to go on then Orin implies.

For one thing, we know that people like Feinstein and Specter in the Senate, and people like Harman in the House, know the details of this particular program, and we know that they have offered bills specifically designed to address the policy needs that they think underly the program. So, I think one could rationally conclude that insofar as Specter's individual bill does more than those bills, it has to be justified not by the specific details of this particular program and how it addresses specific policy concerns, but rather because the Specter bill outlines the right general approach for dealing with these matters. And we can evaluate that general proposition on more general grounds without knowing the details of any particular program.

Also, supplementing all this is the public testimony by people like Gonzales and Hayden, who also did not discuss the details of the program, but who did outline the policy concerns that supposedly motivated the program. And perhaps not coincidentally, the Specter-Feinstein and Harman bills seemed to address those concerns.

So, this is not quite a complete policy vacuum.
9.19.2006 1:46pm
Just an Observer:
A.S.,

As Dames &Moore established when it adopted Justice Jackson's framework, there is actually a continuum rather than three discrete categories.

What the Specter bill certainly would do would be to remove the situation from Category 3, because FISA no longer would bind the executive at all. So FISA expressly would be made optional by the plain language of Specter's bill.

What category a program of surveillance would fall most closely in would depend, I think, on how the executive proceeded. If it followed Specter's new provision authorizing FISA court approval of an entire program, and the court approved it, then I think that clearly would be at Category 1. If the President simply acted on his "constitutional" authority, since there would no longer be a a law restricting him from doing so, that would be closer to Category 2.

The bill's language does not say that Congress "cannot constitutionally extinguish" the President's powers, as you suggest, although Specter sometimes seems to say so. The actual language of the bill simply repeals the existing provision and recognizes whatever inherent powers the President does have. The most that could be said about the language is that it does not limit or regulate those powers, not that Congress cannot.

In any event, FISA definitely would be optional.
9.19.2006 2:01pm
Fub:
Orin Kerr wrote:
And how would Presidents use unregulated authority to conduct domestic wiretapping — what kinds of programs would they want to create that FISA currently blocks?
Asked and answered over thirty years ago, as the late great Sen. Sam Ervin said, "th' buggin' an' th' breakin' o' th' Wawhtahgate."

Of course, you can always pick a view of the Specter bill based on fear. The more you fear Executive overreaching, the more you'll oppose it, and the more you fear an Al Qaeda attack, the more you'll support it.
I don't think all fear is irrational, either of Al Qaeda attacks or of an executive's unreviewable power of domestic wiretaps.

How hard is it just to legislatively extend the FISA time limit for filing post hoc warrant affidavits? Surely some longer but still reasonable time limit would give an executive agency the ability to quickly domestically wiretap whatever they believed they had probable cause to tap, without removing an ultimate judicial review of their actions.
9.19.2006 2:33pm
Rich B. (mail):

Or you can pick a view based on politics: if you support Bush, support the bill, and if you oppose Bush, oppose it.


I never understood the rationale behind supporting structural changes to the constitutional balance of power based on who happens to hold those positions at the current time.

If you support Bush, you should support strengthening the executive, even if the next Executive might be Hillary Clinton or Ted Kennedy or John Kerry? It is simply nonsense. These are probably the same people who wanted to use the "nuclear option" to get rid of filibusters, just because they are in a historical era with a Republican President and a non-supermajority Republican Senate. Or who support State's Rights when states want to ban abortions, but Centralized Power when Congress wants to (or vice versa).

The proper test of the balance of power is how much power you'd give to each branch, knowing that it might be controlled by your political opponents -- not assuming it will be controlled by your friends.
9.19.2006 2:45pm
abb3w:
...the more you fear an Al Qaeda attack, the more you'll support it.

That is, presuming you believe that implementing such wiretapping will have affect the ability of Al Qaeda to accomplish such an attack. Of course, this presumes that anyone is rationally comparing the worries about the Overreach of the Executive versus Al Qaeda attack.

Alternatively, someone might support such measures as "Duck-and-Cover" type placebo actions to reduce long-term public anxiety over an unpreventable risk. However, I'd hope that any proposed placebo would have side effects rather closer to a sugar pill than this sounds.
9.19.2006 3:00pm
Just an Observer:
It is important to restate one important point, which I think is part of what Orin is getting at: Although we don't know everything about the de facto NSA surveillance, the Specter bill obviously would do much more than just ratify or authorize that program.

These provisions are truly fundamental in their scope, and would sweep away much of the entire FISA structure that was negotiated in Congress over a period of years. Yet there has been little detailed exploration of these changes. The only time-critical factor is an artificial political deadline laid down by the White House: the coming election in November.
9.19.2006 3:28pm
Jacob T. Levy (mail) (www):
It's like me telling you, "I'm thinking of a government program. Do you want to allow it or prohibit it?" Without knowing more, how are you supposed to answer that question?

Orin,

You should know better than that-- you've spent too much time with libertarians to treat that as a rhetorical question!
9.19.2006 3:55pm
Bruce Hayden (mail) (www):
As I pointed out in the other thread, the NSA TSP isn't going magically away. At best, you can hope to have it slightly tweaked around the edges. And that is why I think that the critical element is oversight. Not however just the FISC passing on the legality of certain programs, but also, the implementation of those programs. I think someone should be there to monitor it on a high level to see that it isn't abused. Maybe the FISC can do that, but I think high level Congressional oversight would be better - for example, a scheme somewhat similar to the PATRIOT Act provisions reporting numbers of this and that, but only to selected members (such as the chairs and co-chairs of the Intelligence committees).

I may trust this Administration in this area, as many here do not. But this Administration will only be in office another 28 or so months. And over time, programs take on a life of their own, and that is my worry, that in another 5 or 10 years, one of these programs grows to include domestic communications and strays from the mandate of connection to known or suspected terrorists. And then we are back to what FISA was presumably aimed at preventing - government surveilance of Americans for personal or venal reasons.

On the other hand, too many constraints would potentially end up with us again at where we are now - with a pressing security need that is not meetable within the current statutory framework. I see the proposed legislation giving the Execcutive the flexibility it needs - but potentially at the loss of some accountability, and, thus, opening a potential for abuse.
9.19.2006 4:10pm
Medis:
Bruce,

It isn't an either/or proposition: you can have both judicial review and congressional oversight. Which just happens to be the framework for surveillance, and indeed governance in general, laid out in our Constitution.

And some of those who actually do know the details of the Program think it can be implemented with ongoing judicial review as well as congressional oversight. And the government has not precisely argued otherwise--they have just implied that they think it would be better if they didn't have to bother with the courts.

So, I see no reason to choose between judicial review and congressional oversight. Let's have both.
9.19.2006 4:25pm
jrose:
I sympathize with Orin because I can't judge what oversight is needed without knowing the details of the program. However even not knowing the details, I can reject Specter's bill because it results in no oversight for this and any prospective foreign-intelligence program.

What kind of oversight does Feinstein-Specter call for? And why did Specter and Graham vote for both bills?
9.19.2006 4:51pm
Tom Holsinger (mail):
Professor Kerr,

You should have seen the fuss when FISA was being enacted. My father was the AA for a Congressman on the House Intelligence Committee at the time, showed me a draft of FISA, and tried to explain it to me. My recollection is that I found the draft to be incomprehensible gibberish, and the only thing I recall from my father's explanation was that a compromise was underway in which the House majority got a prohibition on domestic surveillance while the intelligence community could continue whatever it was they had been doing when one end of a communication was out of the country.
9.19.2006 5:45pm
Cyrus (www):
It's like me telling you, "I'm thinking of a government program. Do you want to allow it or prohibit it?"

Let me change this a little bit: "I'm thinking of a government program that works in complete secrecy, and which your legislature had no part in drafting. Do you want to allow it or prohibit it?"

It's not a rhetorical question to me. When my government decides to follow rules it won't tell me about, rules it won't even tell my elected representatives about, I'm not okay with that. It's MY government.
9.19.2006 5:51pm
Tom Holsinger (mail):
I also point out that the most likely outcome here is:

Different versions of the bill emerge from the House and Senate, it goes to a conference committee which is stacked with Bush administration supporters, they pass out a "compromise" which gives the Bush administration something it can live with, and that is what both houses then adopt.

This gives everyone what they want. Those that don't like the final version can pretend to their constituencies (financial, etc., in addition to their electorates) that their votes for the Senate version, or the House version, are what counts, those that like the final version can tout it, etc., and the feds get what they want.

This happens over and over with many bills, and IMO it will happen here too.
9.19.2006 5:51pm
Medis:
I have to admit, I find it charming that Tom admits his interpretation of FISA is not based on anything like the text of the statute, but rather on what he remembers his father telling him in a conversation almost 30 years ago. Incidentally, if I recall what my father told me correctly in 1978, FISA also contains a provision that when going to school, children would be forced to walk uphill both ways.

By the way, my bet is on Congress ending up doing nothing at all. I have no particular analysis to support that view, but it has been a successful betting strategy with this Congress, so I am sticking with it.
9.19.2006 6:07pm
Just an Observer:
jrose: What kind of oversight does Feinstein-Specter call for?

The bill requires specific warrants by FISA courts. It explicitly requires that congressional oversight breifings be given to all members of the Senate and House Intelligence committees, not just the leadership.

And why did Specter and Graham vote for both bills?

An excellent question, which I cannot answer. Specter's own vehicle is S 2453, although he nominally is the prime sponsor of S 3001 (Specter-Feinstein) as well.

I can speculate about several reasons: Specter likes to play to both sides of the gallery, and continues to pose as a tough critic of the administration. He also may be sponsoring S 3001 to establish some phony legislative history to sandbag the bill: Since its introduction, the Congressional Record shows this "sponsor" doubting the constitutionality of is core provision, which reiterates FISA's "exclusive means" provision. Specter and Graham may have voted out S 3001 as part of a deal with committee Democrats to cease parliamentary blocking action on S 2453 in committee. Or it may be a chip in larger games being played by both senators.

Actually, there were three bills reported. The third, (S 2455) is sponsored by Sen. DeWine (and cosponsored by Graham). It basically would amend FISA directly to ratify the current NSA program, as publicly described, under congressional but not judicial oversight.

Majority Leader Frist, of course, controls the floor agenda. Presumably he will call up the main Specter bill.

Yet another species of bill is the House GOP version (HR 5825) sponsored by Rep. Heather Wilson, Intelligence Chairman Hoekstra and Judiciary Chairman Sensenbrenner. It is roughly similar to the DeWine bill in that it would directly authorize a renewable program of warrantless surveillance, although details differ. Notably, this bill includes the same provisions that Specter's does to overhaul the basic definitions of FISA's scope.

House Democrats, FWIW, have a bill sponsored by Rep. Harman (HR 5371). It essentially would reiterate that FISA remains the "exclusive means" of surveillance, and require the President to report on how to make the program fit its terms.

There is even a House bill (HR 4976) sponsored by a small bipartisan band, notably including a libertarian Republican faction led by Jeff Flake of Arizona, reaffirming FISA and requiring the executive to report to Congress on everyone who has been surveilled without a warrant.

There were reports last week that the White House tried to substitute the Specter version in the House but met resistance.

In any event, there will be great pressure from the White House for action before the election for two reasons: This is most likely the best hope to gut FISA, long a goal of VP Cheney; political operatives believe forcing a vote now will hurt Democrats at the polls. All of the above suggests that Congress is on the verge of legislating important policy changes in the dark of night and in 11th-hour conferences.
9.19.2006 6:10pm
Chris Bell (mail):
Medis:

You underestimate Congress. They will pass a bill with some grand name like "The Bill to Perfect FISA While Combatting Terrorism". The actual text of the bill, however, will not address FISA or the NSA, but will instead give some sort of tax break to Alaska.
9.19.2006 6:18pm
Just an Observer:
Medis: By the way, my bet is on Congress ending up doing nothing at all. I have no particular analysis to support that view, but it has been a successful betting strategy with this Congress, so I am sticking with it.

I actually think that is the best outcome realistically to hope for in the current Congress. Perhaps the next Congress could actually work out some reasonable compromise on the merits. But I am not so optimistic as to expect either. My cynicism differs subtly from yours.
9.19.2006 6:21pm
Tom Holsinger (mail):
Medis,

You are playing straw man games again.

My interepretation in past threads of FISA as enacted is wholly different in kind from my 28 year-old recollection of a draft. The point I made in past threads about FISA's enactment, which is pertinent here, was the that the objectives of the House Democrats and the intelligence community in enacting it were not mutually exclusive - each got what they wanted. The House Democrats got a prohibition of warrantless domestic surveillance while the intelligence community was allowed to continue its on-going surveillance of foreign communications.

The point I made about reviewing a draft of FISA was that consideration of the text of a proposed statute, out of the context in which it is intended to operate, is likely to result in incomprehension as it did for me in 1978, i.e., I was agreeing with Professor Kerr that:
It's also hard to analyze the effect of the Specter bill's overhauling of the definitions of FISA without knowing how FISA investigations unfold; it's hard to measure the operational impact of a legislative change without knowing more about the operations.

Context is everything here.

So when you say that my interpretation of FISA as enacted is based on my consideration of a draft version 28 years ago,

You are intentionally misleading people. You intentionally lied in stating:
I find it charming that Tom admits his interpretation of FISA is not based on anything like the text of the statute, but rather on what he remembers his father telling him in a conversation almost 30 years ago.

There is no value in debating with a liar. I will not comment on your future posts.
9.19.2006 6:22pm
jrose:
Observer,

If the Feinstein bill requires specific warrants, how is it different from FISA? What kind of oversight does DeWine offer? Flake's bill makes some sense to me at first blush.
9.19.2006 6:23pm
Elliot Reed:
Of course, you can always pick a view of the Specter bill based on fear. The more you fear Executive overreaching, the more you'll oppose it, and the more you fear an Al Qaeda attack, the more you'll support it.
Given this characterization, opposing the bill is a rational strategy. History shows that executive overreaching leads to coups, gulags, and mass murder. Al Qaeda terrorism leads, at most, to a few thousand deaths and destroyed buildings.
9.19.2006 6:35pm
Medis:
Chris,

I look forward to visiting the multimillion dollar "NSA Electronic Surveillance Theme Park" in Juno (slogan: "You'll come for the wiretapping, but stay because you've been detained!").

JaO,

I really don't know what to expect either. But it seems to me that there may not be 60 votes in the Senate for any particular bill, and maybe not a majority for any particular bill in the House, so betting on nothing is looking OK so far. And yes, that is me being optimistic in my own way.

Tom,

That is certainly your right. That said, I don't really see how anything you just wrote differed in substance from what I said. And the bottomline is that you apparently think we should substitute some "legislative history" that arises from a conversation you had almost 30 years ago for what the text of FISA plainly states. And although you deny it, the text of FISA is actually quite plain about precisely when both parties to the communication must be in the United States.
9.19.2006 6:37pm
Just an Observer:
jrose,

The Feinstein bill grants a modicum of new flexibility to the executive in seeking "emergency" warrants (The 72-hour period for filing the application grows to 7 days, but the executive gets a little more freedom in the first 3 days to abort the surveillance if the justification isn't there after all.) Also, the AG gets to delegate authority to trigger such emergency warrants. FISA's current provision providing a 15-day window of warrantless surveillance in case of declared war would also be triggered by an armed attack or an AUMF resolution. The application process would be more automated. More FISA judges would be authorized. (Maybe the newest judges would get nights and weekends?)

The basic thrust is to satisfy the publicly stated objections of the administration that the current warrant process is too burdensome or inefficient. Feinstein's argument, which you can weigh as you see fit, is that she has been briefed on the actual NSA program today and is sure that it could be accommodated by these modest amendments.

The oversight in DeWine's bill, IMHO, is weak. It relies completely on Intelligence subcommittees, with no judicial involvement. Read the linked text above and see what you think.

Flake's bill (co-sponsored by Democrat Adam Schiff) is the toughest defense of the status quo. Although I admire these dissenters -- especially the Republican faction bucking the leadership and the White House -- I cannot believe there is any chance it would ever pass this Congress.
9.19.2006 6:41pm
Medis:
jrose,

Here are a few things the Specter-Feinstein bill would do:

Expand the emergency provision from 72 hours to 7 days;

Change the emergency provision so that the Attorney General does not need to personally approve such surveillance;

No longer require the target be informed if the emergency warrant is denied;

Change the 15 day window after declarations of war so that it would also follow AUMFs and certain attacks on the U.S.; and

Authorize more personnel to process FISA orders, including at the DOJ, NSA, FBI, and even more FISC judges.

Incidentally, it would also reiterate that FISA is the exclusive means for conducting electronic surveillance, provide that no law can implicitly repeal or modify FISA, and deny funding to electronic surveillance programs not authorized by statute.

Anyway, this bill is designed to deal with the issues raised by Gonzales and Hayden when they argued that FISA was too cumbersome.
9.19.2006 6:51pm
Medis:
Oops--JaO beat me to it. At least we are consistent!
9.19.2006 6:52pm
Bruce Hayden (mail) (www):
JaO

But not negotiated between this president and this congress, but rather by a different president, who is now a vocal opponent of this president and his conduct of his war, and a congress that was worried about very different issues, including, waging a very different type of war against a very different type of opponent and facing the use of very different technology. So, the mention that the act was negotiated is really irrelevant.
9.19.2006 7:09pm
Just an Observer:
Bruce Hayden,

I am not against reopening the policy issues, although I do think it is unwise in the extreme to do it in a rush, without seeking careful bipartisan consideration of the merits. I actually am sympathetic to some of the stated intent of some policy changes, but worry that the sweeping changes in FISA definitions actually would do much more than advertised.

As for the crass repeal of FISA's core provision, included in the Bush/Cheney/Specter bill, I have seen no policy arguments advanced to justify the de facto program that lays the groundwork for that extreme gutting of FISA. This ideologically motivated provision goes far beyond legalizing what is said to be included in the "TSP."

So, giving credit to your assertion that the practical geopolitical situtation is very different, that would imply a need for a serious new negotiation based on such practical considerations, not a demagogued ramming-through of a back-room bill.
9.19.2006 7:30pm
Just an Observer:
Bruce Hayden,

I neglected to mention that the same considerations apply to technological changes. There is nothing inherent about technology that prevents Congress from comprehending it and adapting to it. But doing that right takes more time, more serious legislative work, and less political demogoguery.

jrose,

BTW, It can be confusing to call me "Observer" here. Someone else uses that moniker. I am Just an Observer, or JaO to my friends.
9.19.2006 7:43pm
srp (mail):
The key factual questions to me are these: 1) Is there a mass screening component to the program where zillions of calls are monitored in the hopes of finding a tiny number of hot prospects? 2) Is there a "once removed" or "twice removed" aspect to the intercepts, where if domestic party A calls foreign suspect B, all of A's domestic calls to domestic parties C, D, etc. are then scrutinized? The answers to these questions have a lot to do with what kind of oversight is necessary and practical.
9.19.2006 8:26pm
Medis:
By the way, this President DID renegotiate FISA after 9/11, in the USA-PATRIOT Act and also subsequently.

Of course, this is a rather inconvenient fact for those who like to claim FISA was just an outmoded Cold War statute.
9.19.2006 8:55pm
Howard Gilbert (mail):
The only way to get a serious discussion about the merits of the legislation as a matter of policy is to try to avoid thinking about the current NSA program. This means that the original premise of the question was completely wrong. Specter's bill should be judged on the merits of the changes it makes to FISA text and not how it applied to this particular program that none of us fully understand.

FISA was a response to the Watergate era discovery that the FBI used "national security" as a pretense to bug anti-war and civil rights groups. It was drafted during the Cold War. What problems does a post-9/11 reading of the law identify, and does Specter's bill correct them?

To demonstrate the point in the most neutral manner, consider a problem that has nothing to do with the TSP and is not corrected by Specter. In 1801 (b) (2) an "agent of a foreign power" is any US Person who (A) knowingly, (B) knowingly, (C) knowingly, (D) knowingly, or (E) knowingly engages in certain actions on behalf of terrorists or foreign powers. So if an agent of Al Qaeda hires an architect or engineer to do a study that identifies potential targets for attack, but the person discovered to be doing the work does not realize how his work will be used, the FISA definition inappropriately defines away our ability to target the phone of the unknowing agent in order to discover who hired him.

This is criminal justice thinking inappropriately applied to intelligence work. To be a criminal you must have criminal intent. To be an intelligence source, you may be completely innocent and just happen to come into contact with the bad guys. This kind of mistake, confusing the rules of a criminal justice search warrant with the entirely different set of rules that should apply to a FISA national security intelligence warrant, occurs throughout the discussion.

Other parts of Specter's bill seem to move in the right direction. The elements of "probable cause" move away from the incorrect "probably cause to believe that the target of the investigation will be a bad guy himself" toward the more appropriate "probable cause that the program will yield valuable information about a potential attack on the US even if the particular phone tapped is just the Dry Cleaner of the terrorist who will be picking up the suit he plans to wear over the bomb."

Imagine a world in which the President never authoried the TSP, or alternatively one in which the NYT never found out about it. Imagine that Congress is finally getting around to a long overdue attempt to update FISA to the requirements of a real domestic anti-terrorist system. Do you think that the current law is just fine as it is? Are the only changes needed to quantities (3 days become 7, 5 guys become 10, ...) Or did Congress in 1976 fail to anticipate a few things like the internet, disposable cell phones, fiber optics, suicide bombers, and such. Maybe there are parts of FISA that are based on fundamentally wrong conceptualizations based on analogy to criminal justice practices.

I find it inconcievable that any objective analysis will find that the old law didn't overlook important things. I also suspect that the things that really should be done will turn out to have nothing to do with, either in support or opposition to, the current NSA program. They probably won't have much to do with Specter's bill, which instead seems to be a political compromise. Still the bill has some good things, replacing a lot of "proof of guilt" thinking with "useful information about the enemy" criteria. Congress should be able to do better, but partisanship has poisoned the discussion.
9.19.2006 9:00pm
Medis:
srp,

According to Gonzales and Hayden, in this program at least, only international communications are involved, and there is some sort of reasonable basis to believe that one of the parties to the communication is a terrorist, member of a terrorist organization, or affiliated with a terrorist organization. That would seem to rule out your (1) and (2).

Of course, maybe there are other programs which tie into this one.
9.19.2006 9:02pm
Medis:
Howard,

As an aside, once again, I note that FISA has been modified since 9/11, including in the USA-PATRIOT Act. Also, I actually fail to see why unwitting "agents" are a problem unique to agents of terrorist groups, and wouldn't apply equally well to unwitting agents of, say, the Soviet Union.

I have to say, I also think that your hypothetical is a bit far-fetched (we know someone from Al Qaeda contacted them, just not who from Al Qaeda in particular?). And why isn't the FBI say, approaching and asking the unwitting agent to help out? It seems to me that is the right way to handle an American citizen who has been placed in this position.
9.19.2006 9:17pm
Bruce Hayden (mail) (www):
JaO, I may have misspoke above, but we have been cross-wise on these issues for months now. And I have tried to conform to your request that I use something like JaO to distinguish you from "Observer".

Medis, my preference would be for both judicial and Congressional oversight, but with the later constrained to the chairs and co-chairs of the two committees, for security reasons. If given the choice, I would pick Congressional oversight, given that Congress is more accountable than the Judiciary.

But I don't buy that upping the 3 day to a 7 day period for Emergency Orders would address the core problems. The delegation is nice, if it goes far enough. If it merely expands from 2 to, say, 4 people who can authorize emergency FISA warrants, it also doesn't address the core problems.

Part of what is different between when FISA was enacted and now is that instead of tapping a handful of Soviet and Chinese agents, we are now talking a large program that listens just a little bit to a very large number of calls. It is that that is really the core problem, and Feinstein's proposal is just a bandaid for shoe horning the new system into the old paradigm.

I think that the way I would handle it would be to give the NSA, etc. a pass on listening in to conversations with foreign targets, but absolutely require a FISA or Title III warrant for targetting anyone here. But the information from the unwarranted survieled conversations could be used for probable cause when getting a FISA or Title III warrant. Most of this could be done by merging 1801(f)(1) and (f)(2), in recognition that we can't realistically surveil international communications outside this country any more.
9.19.2006 9:35pm
Medis:
Bruce,

I see no reason why the full Intelligence Committees can't be briefed. If you can't trust them, you can't trust the chairs and co-chairs either.

As for the Specter-Feinstein bill--well, they and the other members of the Committee who voted out the bill seem to think it would work, and they actually know the details of the program. So, I'm not sure what good it does us to speculate to the contrary.
9.19.2006 10:16pm
Mark Field (mail):
Bruce's last post brings us back full circle to Prof. Kerr's post: how can we judge the policy without knowing the details? I've said before that I'm willing to discuss amendments to FISA, and I don't think anyone is committed to the exact language of that statute. The objection is to the Administration proceeding without amendment, not to amending per se.
9.19.2006 10:36pm
Tom Holsinger (mail):
The context Professor Kerr refers to is far more important today than in 1978. In 1978 the chief domestic threat posed from abroad was from a discrete, and quite separate, number of foreign spy rings operating in the U.S., mostly from a single source (the USSR and its allies) and the technical means of surveillance was pretty well understood by a fair number of people outside the intelligence community.

Today we face a far greater number of foreign threats to our domestic peace, from a much greater number of sources, the threats are inter-related, and there are constant on-going operations by the enemy and our counter-intelligence agencies. And the technical means of surveillance are more diffuse and harder for outsiders to understand. The USAF enlisted man who worked for the NSA and posted here on the NSA/FISA threads came mighty close to violating his non-disclosure agreement.

If I found a draft of FISA incomprehensible in 1978 on an abstract basis, when I understood the technology involved, drafts of the "Specter" legislation are likely to be even less understandable.
9.19.2006 11:28pm
A.S.:
By the way, this President DID renegotiate FISA after 9/11, in the USA-PATRIOT Act and also subsequently.

Of course, this is a rather inconvenient fact for those who like to claim FISA was just an outmoded Cold War statute.


One wonders why the left continues to raise this clearly irrelevant point. Obviously, a request to amend FISA to legalize the program would have revealed the super-secret program. I know the left is doing everything it can to reveal every secret terrorism-fighting program we have, but a lot of us don't think that's wise.
9.19.2006 11:29pm
Mark Field (mail):

Obviously, a request to amend FISA to legalize the program would have revealed the super-secret program.


This is not at all obvious. If Feinstein is right -- and she's been briefed on the "program" -- then her proposed changes would solve the Administration's problems. Those suggested changes in no way would have revealed anything important, or even anything at all, had they been made in October 2001.
9.19.2006 11:37pm
Howard Gilbert (mail):
Consider a real case. Zacarias Moussaoui pled guilty and was sentenced as an Al Qaeda conspirator. He was arrested a month before 9/11 at a school where he wanted to learn to fly a plane but did not want to learn how to take off or land. The FBI agents could not look at his laptop without a warrant. Some have criticized them for not trying hard enough, but based on what they knew, they really did not have probable cause. It isn't against the law to pay cash to learn to fly a plane without landing, so no criminal law violation was involved. Nor was there probable cause to believe that he was an agent of a foreign power and apply for a FISA warrant.

Should we finally five years later get around to changing FISA to allow this search? To do so we have to admit that FISA is asking for the wrong thing. Before you can complain that the intellegence services failed to connect the dots, you have to let them gather the dots to connect. If an investigation is clearly a legitimate national security intelligence investigation, if the target is not as in the '70s some domestic political enemy, and if it is reasonable to believe that the investigation may (just may) produce information about a terrorist attack on the US, shouldn't the executive be allowed to proceed.

Prior to FISA, the courts recognized broad executive authority to conduct warrantless searches for national security. FISA changed that. The purpose was to prevent executive abuse directed to domestic enemies. In the Moussaoui case FISA clearly did not accomplish any of its intended goals and may seriously have damaged the security of the US. That sounds to me like the law needs to be changed to more precisely conform to its objectives.
9.19.2006 11:53pm
Mark Field (mail):
Howard, the search of Moussaoui's computer is not controlled by FISA. It's a simple 4th A/probable cause case. No one will ever know if a court would have issued a warrant because the FBI never asked for one.
9.20.2006 12:19am
Just an Observer:
I think there is some merit to Howard Gilbert's suggestion that in analyzing such legislation, we do so without regard to the controversial NSA surveillance program. But I think it is useful to consider it both ways. Several of the proposals on the table, such as Specter's, actually go far beyond just legalizing that particular surveillance, but the language would affect that surveillance or much of it.

A good example of such a provision is the general redefinition of "electronic communication" that determines the scope of FISA, contained in the Specter bill in the Senate and the Wilson bill in the House. AFAIK, this change responds to a longstanding wishlist item for the NSA to adapt to technological change. The stated intent of the administration witnesses endorsing this change is to make FISA "technology neutral."

In redefining "electronic surveillance," these bills replace the four-pronged definition in FISA with a simpler and more general definition. They drop the provision in FISA today that brings intercepts into its scope if the intercept occurs within the United States, and drops distinctions made between wire and radio communications. The definitions are based only on who and where the target is.

As Gen. Hayden said in recent Senate testimony, when FISA was originally written most international calls were over the air (satcom), and most domestic calls were on a wire. But now, he said, the reverse generalization is closer to the truth.

It seems to be an open secret that intercepting satcom communications is deemed not to take place in the United States, so 30 years ago this was considered to be fair game for the NSA's primary mission of intercepting foreign communications. So long as the target was abroad, the intercept apparently did not fall under FISA at all. But now most international traffic is on fiber cables, and even much foreign-to-foreign traffic happens to pass through this country. The NSA wants to be able to harvest this "windfall."

But, according to the NYT, the NSA domestic surveillance program does include large-scale intercepts of traffic at U.S.-based communications switches. This has not been confirmed expressly, but Hayden and Gen. Alexander came pretty close to saying so in their Senate testimony.

I am sympathetic to the stated purpose of making FISA "technology neutral," which at first blush seems to restore something like the original intent of Congress from 30 years ago. But the devil is in the details, and the actual language in these bills may open up other domestic loopholes. Orin explored some of that here.

As I mentioned, these changes are very general and not directly related to the TSP specifics. But if we believe the NYT, many or most of the calls involved fall within FISA's scope only because they are intercepted here; the target presumably is abroad. Feinstein essentially confirmed this in a dialogue with Hayden, saying that she knew "those numbers" and a small fraction of the overall program actually targets U.S. persons.

Overall, it seems to me that Congress really has not analyzed this pig in a poke very thoroughly. Rushing to legislate out of ignorance is a bad idea, motivated purely by politics.
9.20.2006 12:27am
Medis:
A.S.,

You write: "One wonders why the left continues to raise this clearly irrelevant point."

First, it is relevant to dispell the idea that FISA was simply a 1978, Cold War era, law. By the time the President issued his order, FISA was actually a 2001, post-9/11, law.

Second, in case you haven't noticed, it isn't just people on "the left" who object to the President's actions in this case. I know this is not your way of thinking, but it is not the case that conservative = diehard Bush supporter, and left = everyone else.

Third, as we have discussed, I also know that you aren't a big fan of the fact we have a constitution which provides for a republican form of government and the rule of law, but some people do find that "relevant".

Howard,

My understanding of the Moussaoui case is that all of the officials who have reviewed the evidence available to the FBI at the time have concluded that they could have gotten a FISA order under the existing standard. Rather, what happened is that the SSA (Supervisory Special Agent) and FBI lawyers screwed up: the SSA did not give the lawyers all the facts collected by the agents, and the lawyers also did not correctly interpret FISA before pronouncing a lack of probable cause for a FISA order. Moreover, even more information was available in the FBI computers (the "Phoenix Memo"), but no one performed a search.
9.20.2006 10:12am
jrose:
Howard,

You raise good points that may (or may not) reveal possible shortcomings in the Feinstein bill. It would be nice to have meaningful hearings to vet the various proposals against both general principles (e.g., the warrant standard and the technology neutral principle) and specific programs (I don't think it is sufficient to apply only the general principles). However, don't you agree that whatever modifications of FISA make sense, getting rid of the exclusivity clause and thus effectively removing almost all oversight (save Fourth Amendment issues in front of the FISC) is not advised?
9.20.2006 11:52am
Just an Observer:
This is more about politics and process than policy, but I note that the Specter bill now has been sent to Intelligence, where the DeWine bill may be favored. From today's Washington Post:

Frist also surprised senators yesterday on the warrantless wiretapping issue, sending surveillance legislation already approved by the Senate Judiciary Committee to the intelligence committee for further review. With one week left to consider the bill on the Senate floor, Sen. Olympia J. Snowe (R-Maine), an intelligence committee member, said passage before the election would be "extremely ambitious."

The intelligence committee is considered hostile to legislation worked out between Senate Judiciary Committee Chairman Arlen Specter (R-Pa.) and the White House. That bill would allow but not order the administration to submit its warrantless surveillance program to a secret national security court for constitutional review. The program involves monitoring overseas phone calls and e-mails of some Americans when one party is suspected of links to terrorism.

Three Republicans on the intelligence committee -- Snowe, Sen. Mike DeWine (Ohio) and Sen. Chuck Hagel (Neb.) -- have co-authored competing legislation that would give Congress considerably more oversight of the program.


Meanwhile, Wilson has agreed to alter her bill in the House to counter objections about its triggering mechanism, which I always have considered to be a red-herring of a controversy anyway. Here is the AP story on that.

If there is not gridlock, it may be that the final bills on the floors of both houses will look more like DeWine's and Wilson's than Specter's.
9.20.2006 1:12pm
Medis:
I thinking tying the surveillance issue to the detainee issue may have been a bad idea. But in any event, my generalized bet on the Do-Nothing Congress looks a little better all the time.
9.20.2006 2:10pm
Bruce Hayden (mail) (www):
Medis,

I don't oppose some full Intelligence Committees oversight, just not full oversight by the entire committee because of its propensity to leak. I would be fine with the full committees getting the sort of summary information that is provided by the PATRIOT Act. But I don't think it would be in the best interests of our national defense to disclose full details of ongoing programs to the full committees, so that they can just show up the next day, above the fold, on the first page of the NYT.

As I indicated before, the types of things I would like to have reported to, for example, the full committees, would be how many intercepts were made (broken down by foreign, international, and domestic (and there shouldn't be any of the later); how many warrants were used for each of those classifications; breaking down the international calls surveieled, in how many cases, was the target within the U.S., and if so, of those, how many were under FISA or Titlte III warrant; how many of each category are passed to the FBI, and then how many of those are used for prosecution; etc. One purpose of this would be to determine whether there really was a problem or not with, for example, the TSP. Also, knowing that any unwarranted surveilance of domestic targets would be questioned (and esp. any unwarranted purely domestic surveilance), such would hopefully not occur.

JaO,

The reason, of course, that SATCOM surveilance could be considered international acquisition is that until recently (i.e. Iridium), they depended upon satellites in geosynchronous orbit - which are, by necessity, located over the equator.

The problem today is that most international electronic communications utilize fiber optics, and for a number of technological reasons, it is highly, highly, difficult (even for the NSA) to tap such except at the switches. And if you look at a fiber map of world, only about 1/4 to 1/3 of such coming into this country are coming from ECHELON participant countries. Much of the rest is coming from such NSA friendly locations as Hong Kong (which at one point would have been fine - but now is part of the PRC). Anyone interested in more details can find it on the bottom half of: FISA - problems with the law.
9.20.2006 2:51pm
Just an Observer:
I am not really sure what the White House would consider acceptable. It wants a bill out of this Congress, I think, because the next one is not likely to be as accommodating. It also wants a floor vote before the election that can be spun as pro- and anti-terrorist.

The most extreme provision of Specter's bill, repealing the "exclusive means" requirement in deference to the President's "constitutional" authority, obviously is dear to the Bush-Cheney true-believers, and hubris might compel them to insist upon it. But the White House might settle for something like the DeWine or Wilson bills, which would preserve the congressional prerogative to keep FISA binding but directly authorize warrantless surveillance within that framework.

I think the DeWine bill would be harder for Senate Democrats to filibuster. I assume that somehow they will get the opportunity to vote on the Feinstein approach as a substitute amendment, which presumably would fail on a party-line vote but give them a rationale for opposing the majority's vehicle.

On the twinned issue of detainee trial and treatment, I think the White House likely will give in to the Warner-McCain-Graham faction, and the Democrats will fall in behind. On FISA, there is some disarray in Republican ranks, but there is no similar dramatic opposition to Bush. Democrats would have to take the lead in any final showdown, and I am not at all sure they have the guts.

Notably, the White House has supported all three GOP bills -- Specter's, DeWine's and Wilson's -- and has refused to commit to a single vehicle by sending up its own language as it did in the detainee-tribunal bill. I think the White House strategy has been to treat the three FISA bills like a Chinese menu, and put together a final version from their various provisions, likely in an 11th-hour conference committee.

It remains to be seen how well this strategy will work.
9.20.2006 2:51pm
Medis:
Bruce,

I think the full Committees should get what you want the chairs and co-chairs to get, and the full Congress should get what you want the full Committees to get.

And there is an important reason why. Oversight isn't really oversight if the person doing the overseeing is powerless to act. And the Committee chairs and co-chairs can't really do much without the full Committee, so only keeping them informed provides no meaningful oversight.

Which, of course, was much the point of Bush's "Gang of Eight" order. It allowed him to claim he was keeping people in Congress informed of his activities, but in such a way that nothing could actually be done about his activities.

JaO,

I also don't know what the Administration is after in either of these cases, which makes it hard for me to predict outcomes. There certainly are a lot of potentially competing interests: getting the precedents they want on executive power, getting to conduct certain programs, getting political ammunication, avoiding judicial review, getting immunity for things that they and others have already done, and so on.

But we shall see.
9.20.2006 3:22pm
srp (mail):
If the disputed program only targets calls with at least one international party, then I don't see how FISA can/should apply/be constitutional. There can't be a reasonable expectation for privacy for conversations to foreign countries when those countries give no one privacy rights from electronic surveillance, and to the extent that intercepts are pursued as part of ongoing military operations, the Article II barrier should prohibit the other branches from intervening on a case-by-case basis.
9.20.2006 6:59pm
Just an Observer:
srp,

Your theory that there is no reasonable expectation of privacy in international calls is often speculated about, but no one ever can cite a case that establishes that. If you have such an on-point citation (not an argument inferring something from other facts, but a real case that actually establishes what you claim) that would add something to the discussion.

It would be especially important to know that before the Specter or Wilson bills pass. Because if you are right, under their revised FISA definitions, all of our international communications -- personal and business -- could be intercepted by the NSA without a warrant. FISA would not even apply, even if our phones are tapped inside this country.

I'm sure that citizens are interested to know that before such sweeping changes are made to the law.
9.20.2006 8:25pm
Medis:
I'll also note military law in this country, including the UCMJ, is filled with provisions that constrain military operations.

So, if the entire system of military law is unconstitutional, that is another thing I would like to know.
9.21.2006 11:15am
Just an Observer:
Even aside from the assertion srp makes, I think there are important questions about "reasonable expectation of privacy" as it applies to FISA and the proposed amendments to FISA.

Orin has suggested a different theory, that the border-exception doctrine might arguably be construed to mean there is no REP for international calls, although he notes that this question has not been addressed by precedent. Similarly, he has noted that there are unsettled questions right now about how REP applies to email, wireless phones and cell phones.

One point Orin has made in other posts is that whatever exceptions might exist to the REP doctrine, the privacy exposure could be magnified broadly by the streamlined, "technology-neutral" redefinitions proposed to FISA in the Specter bill. (The same redefinitions also occur in the Wilson bill.) That is because the most significant criteria remaining in these stripped-down definitions is hard-wired to the "reasonable expectation of privacy."

The existing FISA definitions also rely on this term of art, but also include a separate paragraph that brings intercepts into FISA's scope if they occur within the United States, regardless of REP.

I have wondered myself if there are classified OLC opinions that push the envelope of these unsettled questions about REP. For example, might OLC secretly have promulgated an opinion that there is no REP in international calls? So long as DOJ lawyers never push that envelope with regard to Title III investigations -- and the DOJ handbook constrains them from doing so because of the statutory limitations of Title III itself -- there probably never would be a case on the books that would knock down such a secret internal opinion, because the question never would be raised in non-secret courts.

Given the creative propensities of this OLC to push the limits of the law, especially when no judicial review is available, such a scenario does not seem far-fetched to me. I even wonder if such secret legal opinions might form part of the body of state secrets that the government asserts in various civil cases about the "TSP" to comprise a defense that is too secret to reveal.

Notably, the administration still refuses even to disclose to the Intelligence committees its legal opinions justifying the program.
9.21.2006 12:42pm
Medis:
JaO,

And if I may add my own two cents, my basic theory of the Fourth Amendment is that while it is not always the case that warrants must be obtained in advance, the Fourth Amendment essentially requires that all searches and seizures must be subject to review for reasonableness by a neutral tribunal at some point, even if it is after the fact.

In that sense, I have my own version of the problem with seeing FISA through the lens of more traditional Fourth Amendment issues. In most traditional cases, the people with privacy interests at stake will eventually know what has happened, and they will have some sort of ability to challenge what has happened in a court--either through an evidentiary hearing, or a 1983 action, or a Bivens action, or so on.

But the precise problem with intelligence-related surveillance is that the government will often want to keep it secret, and the affected parties may never know what has been done. So, there is a distinct danger of the government being involved in unreasonable activites which violate the Fourth Amendment, and yet there could be no mechanism to address this danger.

Accordingly, I think that any version of FISA which does not provide for reasonableness review by a neutral tribunal at some point violates the Fourth Amendment. In other words, I don't think that our Constitution and the Fourth Amendment would tolerate the same party both conducting the surveillance and being the final arbiter of its reasonableness of that conduct, and therefore it cannot be the case that the only "court" the Executive Branch ever has to face is the lawyers in the OLC and maybe a select few members of Congress.
9.21.2006 1:13pm
srp (mail):
I do not have a case citation on foreign calls and REP. Many threads ago, we had people who claimed to work with NSA, who, while constrained by security restrictions, asserted or strongly hinted that even such benign places as Canada routinely intercept calls without warrants. My superficial understanding is that French investigative magistrates can bug anything they want to without outside review. So as a matter of logic--not caselaw, which is not the issue we non-lawyers are concerned with--it's hard to see how anyone calling France can have a reasonable expectation of privacy.

And Medis, I have explained before that the UCMJ does not involve prior restraint on tactical or operational decisions, nor does it place non-executive-branch personnel into the chain of command. Hence, the UCMJ is not the same kind of problem as FISA. The Congress can pass laws prohibiting soldiers from bayonetting enemies in the neck, but it can't set up a bayonetting court that decides on a case-by-case basis, in advance, whether a particular bayonetting is OK, under even a minimalist interpretation of Article II. The same is true for bombing and communications intercepts--it's okay for Congress to ban them or regulate them in general, but not to approve or disapprove particular actions in advance, because that violates the unity of command that underpins the Commander-in-Chief role.
9.21.2006 4:07pm
Just an Observer:
srp,

Thanks for acknowledging your "superficial understanding." I do think there are good reasons your particular theory is wrong, because "reasonable expectation of privacy" does not depend on such a lay interpretation of the term. (Prof. Kerr, a Fourth Amendment scholar, explains that better than I do.)

As I noted, there actually are even more credible and arguable theories that -- if sustained -- would mean international calls don't enjoy a "reasonable expectation of privacy." But those theories also are unsettled law.

I actually am much more concerned that some such theory might be consided "right," in the sense that executive branch lawyers secretly adopt it and instruct officials secretly to act on it, even though the theory has never been established in court.

Consequently, if Congress enacts the proposed new FISA definitions that crucially depend on when a "reasonable expectation of privacy" obtains, Congress cannot know the actual effects the the law it passes. In the context of the policy debate, which is what this thread addresses, that is rather important.

Congress could rewrite the definition so it is more similar to that governing Title III wiretaps, without tying it to "reasonable expectation of privacy" at all. But that is not what the proposed language does.

To rush such fundamental changes into law without fully understanding their tangible implications, just to meet a partisan political deadline, is a horribly poor way to legislate.

The public mostly has been led to believe the legislation has to do with authorizing the limited surveillance the President says is occurring de facto today -- which has, BTW, been ruled to be unlawful by a federal judge. But the legislation in question actually goes far beyond that, and might authorize warrantless wiretaps of all our international calls and even some purely domestic calls and email, depending on how the key phrase is interpreted. None of which has anything to do with terrorists.

I seriously doubt that many people know that. I doubt that most members of Congress know that is even in question, let alone know the answer. I think Congress and the public should know in advance if that is what the bills would do.
9.21.2006 5:21pm
srp (mail):
JaO: Note the limited context of my admission about understanding--it had to do with French magistrates and in any case goes unrebutted by you. I'm pretty sure I'm right about that.

I agree that we are operating pretty far in the dark here on the NSA program and that legislating on it is tricky. The problem, of course, is that we don't want the enemy to learn the details of what we are doing. I would suggest trying to rule out specific practices that seem too abusive or too prone to abuse, regardless of what is actually known to be in practice right now. Then apply congressional oversight and make violators subject to the UCMJ.

In this policy discussion about what practices seem to abusive, there is in my mind a need to engage in some sort of cost-benefit analysis. Different people will value things differently, a political compromise seems inevitable.

Personally, I take the view that snooping on me or anyone else to try to prevent Americans from getting blown up, poisoned, or decapitated falls into the narrow range of government activities to which my tax dollars and citizenly allegance are appropriately directed. Subsidizing other people's pharmaceutical purchases? Not so much.

So I would be mostly concerned with the use of NSA surveillance data to blackmail political opponents or suss out their strategy, to help members of the security establishment carry out personal vendettas, to favor or harm particular businesses, and similar sorts of abuses. And I would favor methods of oversight (NOT mass rubber-stamp warrants) to contain these specific abuses. Other people will have other priorities, which is why compromise will be required.
9.22.2006 6:17pm