A simple public choice theory of universal surveillance

Let’s say that everything is known about everybody, or can be known with some effort.  The people who have the most to lose are powerful people who have committed some wrongdoing, or who have done something which can be presented as wrongdoing, whether or not it is.  Derelicts with poor credit ratings should, in relative terms, flourish or at least hold steady at the margin.

It is not obvious that the President, Congress, and Supreme Court should welcome such an arrangement.  Nor should top business elites.  More power is given to the NSA, or to those who can access NSA and related sources, and how many interest groups favor that?

Therein lies a chance for reform.


Always the pessimist

Pessimistically optimist: Powerful people will make NSA behave.

They might but for other reasons than the one mentioned by Tyler Cowen.

The rich and powerful, by virtue of being rich and powerful, can survive their own misdeeds. Indeed, they can afford for their misdeeds to be public and (relatively) well-known and no one even bother to speak up.

the people with most to lose are ethnic groups like white people, not rich people. This surveillance warfare is explicitly targeted at white people. Terrorism is just an excuse.

If Wash DC was serious about terror they would immediately stop 3rd world immigration, & deport all Muslims.

Thinking more about this, I am guardedly pessimistic. There are cases in England where legal actions by powerful people have created a general expansion of privacy (http://en.wikipedia.org/wiki/Privacy_in_English_law) however it could be argued that most people do not have the resources to avail themselves of these laws. There are precedents in computer systems (e.g. hospital records) to give special protection to celebrities, e.g. by creating fake records and tracking access to them. I conjecture that the operator of a surveillance system would find it more attractive to give special privileges to the small number of people with sufficient power to cause trouble than to relax surveillance of the general public. I also suggest that the veil of secrecy which surrounds such things would also favour special treatment of the few over general regulation.

What I was going for is that we know bankers in many big banks have committed crimes, one way or the other. There was the LIBOR fixing scandal. HSBC also seemed to have been criminally negligent when it came to money laundering and embargo enforcing issues.

But we know many other industries have done their level best to avoid or circumvent regulations and, when something bad happened as a result, the people in charge faced no real consequences.

With regards to politics, we have had the case of Blair and GWB. I never believed that Saddam would have more than a couple of barrels of chems left over but I did think (and was wrong to think) that we would 'discover' WMDs anyhow. At the time, it seemed that nothing less could protect the initiators of that war. They HAD to be right. So I thought.

In reality, they were wrong and the only thing they bothered with was "Oooops. Well, anyhow, he was a bad 'un and no point crying over spilled milk".

And you know what? Everybody went along, me included. Stunning.

As I said, I don't expect powerful people to be prosecuted anymore unless they commit some personal crime such as murder or rape...

The President might be quite happy having the dirt on Congress, the Supreme Court and whoever is his opponent for the Presidency. I'll bet he knows a lot about Romney now.

I bet the president knows a lot less about Romney than the NSA does, though.

And no illegal wiretapping was necessary.

It's perhaps interesting that the President's challenger last year had a squeaky clean private life, but was brought down by surreptitious taping of his 47% talk.

"It’s perhaps interesting that the President’s challenger last year had a squeaky clean private life, but was brought down by surreptitious taping of his 47% talk."

It was a combo of that, being a Mormon, and the surreptitious recording of him saying the phrase, "overturn Roe v. Wade."

Potentially creating a political process where (even more than today) power accrues only to the sorts of people who can be ruthlessly careful in the interest of holding power much later in life.

But the secret power elites want defective people to hold these offices. They won't have leverage over anyone that's completely clean. Therefore, having these defective people in office is evidence that there is indeed a covert cabal behind the national security curtain.

I think we should be worried less about the precipice and more about the slippery slope.

Today, the scope is a search for terrorists, with access bounded by the FISA court.

But, after the data is gathered and stored, how soon will we hear: if we could just access this data, we could find the kidnapper of Baby Jane.

Who will say no.

The dam will breach in many small places.

There are so many serial killers at work along the nations highways that the FBI has a special task force. Since some number of the serial killers will be truckers traveling for work, a quick trawl through cell phone and credit card records could potentially narrow down the suspect pool.

+1 Heading down the slippery slope. But, first we will have to analyze all cellphone records to detect the pattern. We can't have a needle without a haystack.

"The people who have the most to lose are powerful people who have committed some wrongdoing."

Um, maybe not...they might even come out ahead. There have been some impressive (almost comic) political rebirths in recent times. Maybe those scandals will be even more ho hum once we've established that everyone 'puts their pants on one leg at a time' (no superhumans above mess ups). Not to say we should leave the equilibrium to chance, but it's not clear to me where this leads.

Exceptions don't make the rule. I'm skeptical everyone wants their mistresses ousted in hopes of political rebirth.

I'd look at cultures such as Japan where population density made it almost impossible to have privacy. What developed was a societal privacy where everyone knew but not one acknowledged. If everyone has a mistress, and everyone knows their names and predilections then no one gets outed.

The losers are those who have a sense of shame. The winners are folks like Anthony Wiener who have none. A brave new world.

The mistress is a relatively minor issue. Bribery, vote fraud, tax fraud and various corruption charges are what a politician would most want to hide. Or anything that can be made to look like the above to "low information" voters will also work.

So, basically, what you are saying is that off this information secretly being collected by the NSA should be declassified and publicly made available on the internet, for anyone to do whatever datamining they want on. In fact, all cell phone companies should make all metadata public knowledge as a matter of policy, all the time.

One problem ... the people in power might have their own,secret, communications network, which is kept private. I think it's pretty much a given that Congress the President, DOD, CIA, NSA, etc. have secure communications, isn't it? Wouldn't they just start using the secure network all the time?

I think one winner out of all this is going to be Open Source projects: Email Encryption, voice scrambling, Tor, anonymous peer-to-peer etc.

It is easier to verify open source Thunderbird hasn't left a backdoor for NSA than Gmail or Skype.

In the long run the best solution is not to give the NSA a chance, even if they want to. It's iffy relying on legal protections. e.g. strongly encrypt all communication (voice / email etc.) at source. NSA can go ahead and store all my ciphertext. It's really not as hard to do this as it sounds.

Encryption is overrated. Encryption algorithms assume a random data set, which they never are. In addition, they assume there exists a secret key that the NSA doesn't know about. Did you create your key? Or did you use a program on the internet to create the key for you? Open Source doesn't help you here, it's not some sort of magic, even though the phrase is bandied about as if it were.

I still think a running key cipher is probably the best way to go. But hey, you can always encrypt the running key cipher message for very little additional effort.

What are you talking about?

First, modern encryption algorithms do not need random plaintext to be secure. The assumption in analyzing something like AES is that the attacker gets to send you a lot of plaintexts, then look at the resulting ciphertexts, and iterate that process over and over trying to learn something about the cipher's state. (Sometimes the attacker also gets to ask you to decrypt things, or to encrypt/decrypt things with several related keys.)

Second, key management is a problem, particularly key generation. But I don't know anyone who gets their keys from the internet, though it's a big world and someone somewhere surely does any dumb thing you can think of. Normally, you have a cryptographic random number generator (based on a block cipher like AES or a hash function like SHA2), and then your computer tries to collect hard-to-predict data to come up with an unguessable starting seed value for the random number generator. Typically, all that is handled inside the OS and the application program just asks for random numbers from /dev/random or the CAPI and uses them. When that goes wrong, the keys are easily guessable--the most recent example of that was the big story a couple years back of a bunch of RSA public keys on the internet that shared prime factors. This should get better over time, since Intel is putting a hardware RNG into new chips, and hopefully AMD will as well, and maybe eventually the manufacturers making ARM chips for phones and tablets will follow along.

What open source gives you is the ability to see how keys are being generated and managed. That isn't magic, but it is better than not being able to see what's going on.

A running-key cipher has worse key management problems. If you want it to be secure, you need a random key the same length as your data, and then you have to move the random key securely to your destination. If you're using some other source for the running key, your security depends on the source. (Just doing Vignerre from, say, a running text from a book is very weak--you can find programs that can take a long encrypted text and recover both the plaintext and the running key, because English text only has about one bit of unpredictability per letter once it gets going.)

First, modern encryption algorithms do not need random plaintext to be secure.

Huh? Ok, but that's not what I said.

Second, key management is a problem, particularly key generation. But I don’t know anyone who gets their keys from the internet, though it’s a big world and someone somewhere surely does any dumb thing you can think of. Normally, you have a cryptographic random number generator. Typically, all that is handled inside the OS.

Yeah, once again, that's not what I said.

This should get better over time, since Intel is putting a hardware RNG into new chips

And what are the odds that the NSA has the plans for the chips and has had a team working on the issue since before it was public news?

A running-key cipher has worse key management problems. .. then you have to move the random key securely to your destination.

That's the part that makes it secure.

A running-key cipher has worse key management problems. .. then you have to move the random key securely to your destination.

If it makes you feel better you can substitute 'symmetric key algorithm' for running-key cipher. Though really what I had in mind was a one-time pad implementation. The one-time pad is always going to be secure, assuming the attacker doesn't get the source. A 'symmetric key algorithm' is less secure, but still requires (like a one-time pad) the secure transfer of the key from source to destination.

Anything less than this is always subject to cracking by an agency with a budget in the 10's of billions, assuming their interested in the information.

Thank you @John. I didn't know where to start.

@JWatts: You really need to brush up on the state of art in Encryption.

Encryption might have its blind spots, but not for the reasons you are alluding to.

@JWatts: You really need to brush up on the state of art in Encryption.

Encryption might have its blind spots, but not for the reasons you are alluding to.

I don't think you are aware of the blind spots.

High-quality random numbers are difficult to generate. The random number generation functions in most programming language libraries are not suitable for cryptographic use. Even those generators that are suitable for normal cryptographic use, including /dev/random and many hardware random number generators, make some use of cryptographic functions whose security is unproven.


Communication Theory of Secrecy Systems is a paper published in 1949 by Claude Shannon discussing cryptography from the viewpoint of information theory. It is one of the foundational treatments (arguably the foundational treatment) of modern cryptography. It is also a proof that all theoretically unbreakable ciphers must have the same requirements as the one-time pad.


Honestly, I thought it was common knowledge among anyone that knew the subject that the only secure communication key uses a one-time pad.

If basic public key asymmetric encryption were inherently secure we wouldn't have gone from 40 bit encryption, to 128 bit encryption to 256 bit encryption in the last 20 years. It's just an arms race to keep ahead of the crackers. And there's also the case that time and time again, flaws have been found in the implementations of various encryption methods resulting in the crackers not having to rely on brute force attacks.

Uhhh, JWatts, you seem to have quite a pickle in your knowledge of cryptography. (PhD in Algebra and Number Theory here, doing applied cryptography in smartphones for living).

For starters, your last assertion mixes up public-key asymmetric encryption with key sizes of symmetric encryption.

Even if both of them are measured in bits, they aren't any more related than, say, size of human waistband and size of a floppy disk (both measured in inches).

Your assertion that anything less than one-time pad can be cracked for a budget of 10s of billions, is very fanciful. You can't solve some hard algorithmic problems by just throwing money on it; either some genius finds a solution, or not. (And no, you can't brute-force a 256-bit keyspace. That in itself would require more energy than what is available in the Solar System).

Last but not least, most of the successful cracks do not target the algorithms themselves, but weaknesses in implementation. One-time pad is quite susceptible to that. There is nothing inherently secure in products that use one-time pad; quite the contrary, as it is easy to mess up the key generation in subtle and unrecognizable ways.

Indeed, the one obvious, natural constituency at least in terms of self-interest would be military families (disproportionately poor and typically disjoint from the academic libertarian establishment) who might have considered the trade of a little spying, drone strikes and so forth for avoiding a multi-thousand casualty $3Tn war on the basis of "Screwball's" intelligence a reasonable bargain. Clearly there are very few think-tankers in that segment of society -- they instead make up a disproportionate fraction of the internet identity community (but of course that obvious self-interest is just coincidence - they're only looking out for natural rights). And then obviously poor military families generally don't have much say in the political process, so I'd say the odds are good for Tyler's hypothesis.

On the other hand, the economic neoliberal corporate free speech judicial movement that has secured myriad rulings for privacy violation on 1st amendment grounds (a subject which, equally unsurprisingly, has received trivial coverage on this blog by comparison to the present 'big brother' revelations) should continue unabated. And then for some reason we're supposed to think that's a better outcome--because clearly a corporation selling our medical records to pharmaceutical companies has the cleansing influence of the invisible hand (because they're making money off it), while when the democratically elected government does it to try to prevent wars and attacks to maintain it's elected power, there's no moderating force from the invisible hand. Or something.

You're presupposing a linear reward/punishment system, but in all things grand and governmental, the power is set in place already; the reckoning lies in how much power one can toss around and assert shamelessly. We commoners are always late to the party.

In other words, it's a game of ego's. The familiar tier of chain of consequence means absolutely nothing to Obama +. The measure of power is written in the depths to which a leader can pound it...

Applied technology giveth and applied technology taketh away: blessed be the name of applied technology.

We're skating past the point of adapting our stupendous digital technologies and our glorious algorithms to fit atavistic human circumstances and anachronistic human institutions: we're now having to hope we can figure out how to accommodate ourselves to our technologies, since our technologies have no incentive in figuring out those things for us. Either our technologies are not yet fast enough to do what's needed, or we are unable to incorporate social and political changes worldwide quickly enough, unable to appropriate the "progress" our technologies offer (or tempt us with).

Who'd've thunk that temporal velocity could become a critical political challenge so soon? Why, we've barely had time to become modern!

I'm a derelict with a poor credit rating, yet I find myself angrier over this than most people. My relative status seems to have dropped a bit, though I can't explain exactly why. One possibility is that I feel relatively defenseless in the legal department and a more intrusive government seems more likely to work against me than for me.

I've found that I'm always angrier for the sole reason that I'm aware. It also doesn't help that I actually look at the "success" rates of the security measures that we are forced to accept on the speculative promise that they will work. Again, simple awareness rather than notion causes irritation. Froggy in the frying pan is busy with his froggy concerns.

Three years a Harvard Law School student had a private email exposed by a romantic rival. Who did the dean of Harvard Law School, Martha Minow, condemn: the victim or the victimizer?


It might be interesting to study the career of former Senator Larry Craig (R-ID) in some detail. Assuming that it was not unknown in some circles before it was known to the public that Senator Craig routinely attempted to have sex in public toilets with male strangers, what uses of Senator Craig's power and influence stand out as unusual?

One way to test Tyler's theory would be to study the careers of Republican politicians who were known to insiders, but not to voters, to be homosexual. Which special interests did closeted gay GOP politicians cater to more than normal? Those special interests would likely be blackmailers.

For example, Tucker Carlson tells a funny story about how he and Bill Press were broadcasting live from the public park across from the Supreme Court building on the night that the Justices decided Bush V. Gore. The bushes in this park, according to Carlson, are a notorious gathering spot for homosexuals looking for quickies. Carlson says that one of the prominent politicians they interviewed on the spot that night about the recount decision wasn't there to be interviewed, he was a notorious closeted promiscuous gay man who just stumbled out of the bushes pulling up his zipper when they turned the camera on him and asked him to commentate. And being a fine politician, he turned on the spiel.

Here's the transcript of the broadcast:


Perhaps this dialogue is relevant:

CARLSON: Governor XXX YYY of ZZZ, who amazingly enough we found right here in the park. And he doesn't have a Carol Press [a reference to Bill Press's wife] ...

PRESS: No, but...

CARLSON: ... We can fix that out of gratitude for joining THE SPIN ROOM.

PRESS: ... We can fix that. Governor, the first question is what are you doing wandering around a park at midnight?

GOV. XXX YYY (R), ZZZ: Tonight, we had a Christmas tree lighting, which is a beautiful 75-foot Colorado blue spruce which came here directly from Pike National Forest in the Rocky Mountains. So I came here to help Speaker Hastert light the tree.

CARLSON: And we nabbed you.

YYY: You did nab me.

So, which special interests did Governor XXX YYY of ZZZ service even more than is usual for Republican governors?

Didn't work for Joe the Plumber.

The most interesting thing about this issue is that the story didn't get broken by a US news source. Government agencies will go too far, will abuse their power. Power corrupts. Vigorous and active limits by the press or other branches of government keep a lid on the shenanigans. After watching what happened in 2008 I think the news media is complicit in the abuse of power. It is all fine and dandy until your email gets read.

'Therein lies a chance for reform.'

Well, you see, there was this person actually elected president with the following resume -

1. Chairman of the party he belonged to

2. Director of a spy agency

3. Father of the son who was elected president, and whose administration cemented the security state

It is almost as if Americans have no knowledge of how long this has been going on, or what sort of choices were already made.

Anyone willing to bet on Jeb Bush making a successful run for president?

If not Jeb, then George P. Bush -- he's diverse!

It is difficult to predict the consequences of government surveillance without knowing more about its details. For instance, an opaque statistical model based on telephone records might find enough targeted criminals to gain credibility while also discriminating against people from high-risk ethnic groups or people with unusual hobbies (e.g. people writing thrillers or fan-fic, or people researching the credibility of thrillers they have read). See e.g. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2208240 "Discrimination in online ad delivery" People wealthy enough to have their telephone access gated by secretaries might be recognized by the system as an entirely different class, and so untouched. This is unfortunate, because the operators of such a system would have a motive for being very selective about the details of its design that they chose to release.

This analysis seems incorrect. There has always been a thriving market for information about important people. One example is the paparazzi. The NSA leak is about a new technology (large scale data mining) that has made it possible to gather and filter enough information about the anonymous masses that it is finally a worthwhile project. Elites who have been spied on for decades will not feel especially vulnerable to this development.

The most likely outcome is a kind of "resolution" to policies of high controversy over such issues as drugs and abortion.

Suppose it turns out it's possible to prove that 40% of the adult population has used some illegal drug in the last year, or last 10 years. You can't lock up 40% of the population and have a functioning society. You can't really refuse to hire 40% of the population and have a healthy society. If it turns out that many of the 1st rate practicitioners of some essential profession fall in this class, what do you do? This is part of the problem with very strict DUI punishment - when it's the town's only Dr. who blows over the limit, locking them up or taking away their license can be a real problem.

Likewise - suppose every woman who had an abortion, and perhaps every man who had fathered the aborted pregnancy, is "marked" with what amounts to a scarlet A. Some employers try to not hire such people. Then it turns out to be a way larger part of the population than previously known. And turns out that society cannot go on with that many people marked "shun, never employ".

Seattle is considering a law making it illegal to ask if a prospect has a criminal record. There are obvious issues with this, but the social function argument has merit - if getting a conviction for anything means you can never get any job at all, then society has just turned you into a pauper, and likely reoffender. For smoking dope? For stealing hub caps? For a very ill advised bar fight?
Employers (rightly) object for all sorts of reasons. There's a private/public trade-off. But will the job losses from employers shunning hiring or using back doors outweigh the job losses from "half the population has a scarlet C or scarlet A or scarlet Fool on their head and cannot get work." ?

The sorts of data mining Tyler posits would likely lead to a vast set of very restrictive rules on how that data could be used, and/or the North Korea of North America. I would hope more likely the former.

It is actually necessary that people and society be able to forget, to move on past things, even horrible things. Without that, society would simply end every time there's a horrible crime, which is basically every day.

But history is full of societies that didn't function very well, and they all eventually collapse.

If getting a conviction means anything, "society" has not just turned you into a pauper without your participation and consent. (Antinomianism and autonomianism have limits, too.)

That assumes that it is possible to live a productive life without breaking some law. Anyone who does anything of any complexity is pretty much guaranteed to have not fulfilled some regulation somewhere along the line. It was interesting last week hearing the reactions to the IRS targetting. There is an assumption that one could actually follow the law. Then we had Apple hauled up before congress to explain why they were doing things that the questioners had put into law.

In labor negotiations a tactic to force movement is to work to rule. Only do everything exactly as agreed upon in the labor agreement. Usually everything grinds to a halt.

I find it quite amusing. These idiots write laws with penalties that are purported to represent societal opprobrium towards certain actions, then makes it illegal to consider this opprobrium when hiring.

Or, it simply assumes ready willingness to accept risk and responsibility for outcomes, with or without apprehension: I routinely drive five to ten miles an hour over most posted speed limits (weather and cellphone drivers permitting), but I'm never tempted to pass a stopped school bus on a two-lane road or street.

We've already got that situation with the growing requirement for background checks for practically any kind of a job. Guys who know that they won't pass a background check won't bother to apply and eventually join the growing underground economy. Which is turning into the real economy.

The solution is very simple: completely arbitrary enforcement. "If I don't like you, you'll go down, because you are one of the 40%". Behave and you'll be fine. There was a famous quote from a prosecutor from different time, different country: "show me the man and I'll tell you why"

One of the dangers of PRISM is not only with legitimate authority but all sorts of outside-the-law operators.

There is possibility of insider trading from rogue elements of the NSA for private use or with black operations not even known by top management.
(I discount the idea that NSA is not saving all the content.)
Plus whatever other criminal activities one can use simply by knowing who is communicating with whom.

Oh, NSA Personnel will prevent that by great screening. Yes, just as with Snowden — he is the living proof of the danger of the system — who knows who can be prevented from misusing the information.

It is not obvious that the President, Congress, and Supreme Court should welcome such an arrangement.

It is not obvious that they'll have any choice in the matter.

"I suspect that, you know, on — on a list of people who might be targeted, you know, so that somebody could read their emails or — or listen to their phone calls, I’d probably be pretty high on that list. So it’s not as if I don’t have a personal interest in making sure my privacy is protected." - President Obama.

Although "Obama is an asshole" is still a stronger theory than "Obama is already being blackmailed", I'm not sure what better evidence for the latter theory we could expect to find than the above quote. Maybe if he was blinking "SOS" in Morse code while he said it?

"Maybe if he was blinking “SOS” in Morse code while he said it?"

Great line.

I'd wager $1000 that Obama doesn't know Morse Code for SOS.

I'd wager that his teleprompter does.

Of course not. The joke is the reference to former Alabama Republican Senator Jeremiah Denton who as a POW in Hanoi blinked the message "T-O-R-T-U-R-E" while participating in a propaganda film:


I know the story well from my days teaching Code of Conduct while in the JAG Corps. It was an object lesson in continuing to resist by any means.

I learned Morse Code and Semaphore as a Boy Scout. I don't think I could blink di di dit dah and talk at the same time.

There is an additional twist -- by putting all information about every American in one spot, including Mr. Clapper of NSA, etc etc we increase our national vulnerability. All "the adversary" (whoever and whenever they are) has to do is hack one site -- NSA -- rather than dozens and dozens.

Seems like a system designed to weaken us.

Of course no one could ever hack NSA through (forget technology) suborning, bribing, blackmailing etc etc a NSA employee.

It always has to be about the rich and the powerful, eh?

"Let’s say that everything is known about everybody"

From this you proceed to analyze the political implications, but what matters more to most people are the personal implications. Let's say that everything is *suddenly* known about everybody. The immediate response is immense embarrassment on the part of almost everybody. Who is most embarrassed? The powerful politician who frequents expensive call girls or the unemployed masturbater who frequents RedTube?

Who tends to be more paranoid: the weak or the strong?

There are meaningful long-term concerns about the growth of a police-state infrastructure, such as the likelihood that politics would become ever more evil and brutal, but the short term effects are that low-status infovores become even more low-status. The assault on free speech isn't direct, but it is real or felt, due to creative ambiguity, as if it might be real.

I happen to be in a protracted battle with the IRS. I'm hoping they will ultimately be reasonable, but I don't know. Should I worry that they might know that I sometimes comment on iSteve, which is officially a "hate blog"? Should I worry that on Twitter I engage in the sort of dark humor that would get me arrested at an American airport?

The security state makes me feel a lot less secure. I'm not concerned about terrorists, but the government terrifies me based upon my direct experiences with it in court.

Hate to quote the SS, but I fear a who, whom? situation where I'm on the receiving end because I fit the bourgeois white male profile, but without the actual social power to defend myself in court.

Yes, I'm paranoid. My point is that the surveillance state just might be the hardest on us paranoids.

I can't resist the urge to stretch this thought experiment.

So suddenly we all know everything about everyone. We immediately begin rounding up all the murderers, rapists, and other violent felons. It would probably take years and hundreds of billions of dollars to incarcerate them all. Or rather, since the fear of injustice has melted away, we would probably resort to summary executions. The bodies stack up.

Some people would take justice into their hands knowing they will suffer consequences. Lots of people would get fired. Lots of people beaten for various transgressions. Relatively minor acts of retribution are tolerated because there aren't enough resources to handle them.

Wars are declared over atrocities that were previously unknown or unproven.

Markets become a lot more efficient...immediately.

In other words, total chaos for a very long time.

Let's face it - much of human existenc is built around lies, and civilization demands that 1) we accept most of these lies as long as 2) the lies are not too big.

There is plenty of room for good stuff. A lot of people will atone for sins or parties will realize they are equally guilty and settle their differences.

Reminds me of a proverb I heard in law school: "Life isnt fair, but would you really want to live in a world where everyone got exactly what they deserved?"

The quest for efficiency is precisely why command economies end up as police states; they require perfect knowledge for efficiency.

What really gets you in trouble these days is making well-documented statements in public of what everybody already more or less worries is true: e.g., Jason Richwine got Richwined not for anything he said in secret but for his Harvard Ph.D. dissertation.

Steve, you should win an award for most determination to insert your pet topics into a completely different topic.

+1. He's not even very subtle about it.

I've read way too many comments on this on the internet today and it seems that the average person buys into Tyler's theory: this will hurt the strong and help the weak. In other words, the surveillance state is popular with populists. So contra Tyler's conclusion: how much chance for reform when the median voter loves the surveillance state?

That's an idea I hope is wrong. We voted Obama into office because we thought he would end the surveillance state. I really hope the thesis that Obama has lost touch with his constituency on this issue is the correct one. I'm going with the theory of executive overreach, that Obama has no values, and that Snowden is, yes, a hero.

Why did you expect that of Obama? How much in Obama's life story suggests that he had much objection to the surveillance state other than that he wasn't in charge of it? Did you bother reading Obama's memoir all the way to the end?

The reality is less that there are all sorts of secrets out there that we don't have access to, and more that there is a vast amount of information already available to citizens, but it has to be researched and interpreted, and most people can't be bothered to do that well. For example, Obama's "Dreams from My Father: A Story of Race and Inheritance" is written in a flowery, snooze-inducing style that almost nobody in journalism finished. That's all it took.

Nah, Snowded should be put on a suicide watch protocol designed to cause him to go insane and suicidal. That's how a good government would roll.

We already can know a lot about the rich and famous, but what matters is less what's known than the socially approved attitude we're supposed to hold toward that person.

For example, there's been plenty of information online for over a decade documenting that Morris Dees of the Southern Poverty Law Center is a pretty hilariously rotten guy. Here's Charlotte Allen's recent Weekly Standard article amusingly synthesizing some, although hardly all, of what's known about good old Morris:


But, who cares about the truth? Morris is widely believed to be a saint and a scholar because we are constantly told what a great guy he is.

What matters more is who is holding the megaphone.

Consider the career of George W. Bush: In 1999, leading Republicans gave their blessing to George W. Bush, a nonentity and a jerk, even though many of them knew he wasn't up to the Presidency, having had the displeasure of coming in contact with him over the decades through his father. Many GOP insiders knew Dubya's mother (!) didn't think much of him.

Or consider John F. Kerry. In 2004, I went through Kerry's and Bush's publicly available military officer test scores and demonstrated that Kerry probably wasn't even as smart as Bush. That took a lot of analysis, but you could have guessed that just from listening to Kerry's speeches. Not surprisingly, Kerry turned out to be a mediocre candidate for the Democrats.

Consider the career of George W. Bush: In 1999, leading Republicans gave their blessing to George W. Bush, a nonentity..

What? That's just a ridiculous statement. George W. Bush was hardly a nonentity in 1999. He was a highly popular governor of the second biggest state in the country.

Throughout Bush's first term, national attention focused on him as a potential future presidential candidate. Following his re-election, speculation soared. Within a year, he decided to seek the 2000 Republican presidential nomination.


For those who want more surveillance, why don't you go ahead and volunteer: http://andreasmoser.wordpress.com/2013/05/02/you-want-more-surveillance-you-can-have-it/ ?

"The people who have the most to lose are powerful people who have committed some wrongdoing, or who have done something which can be presented as wrongdoing, whether or not it is."

One of the reasons why government's powers, including surveillance powers, seem to perpetually expand and increase is that most people seem to have some sort of behavioral bias that causes them to underestimate the probability that those powers will be used against them. Both Democrats and Republicans salivate about all the "great" things that they will be able to do when they seize the powers of government, while giving little thought about the approximately 50% probability that they will be out of power after the next election cycle. By the time they are no longer in control, it is too late to limit government's powers.

The fact that the NSA was given these surveillance powers to begin with is evidence that the governing elites have not believed that granting those powers will come back to haunt them.

Not often I think Cowen is catastrophically wrong. This is one of those times.

He said there is a 'hope.' Like Obama said there is a hope. More likely we just get the kind of sociopaths who get off on it- which we've already gotten. A good PhD project would be why management selects for sociopathy.

Because managers select managers, and the sociopaths drove out everyone else.

Seriously, self confidence is probably the most attractive trait in any social endeavor, and sociopaths possess an abundance of that, whether deserved or not. They tend to be quite smart, ambitious, cunning, able to play the game well. In most cases, they survive even their hubris; it's usually in fiction where they have their comeuppance.

Research has diminished the conventional wisdom that intelligence and criminal success are correlated; selection bias at play. Nevertheless, I saw first hand as a prosecutor that smart criminals are very difficult to prosecute. They are methodical in their planning.

So...on Obama...how 'bout now? Opportunist? Yet? Well, we've got a few years still.

Whenever I get into a tiff with someone, I tend to tag them as a narcissist or whatever it is because they will always come back with some petty unrelated thing when I would be sure to steer clear of additional friction. Then I know which bucket to put those people in. Has Obama steered a wide berth around what a good liberal would? Would a good liberal be trying to skate the razor's edge on spying, bad governance, corruption?

Can we at least spitcan this Holder clown? I don't even know if he has anything to do with this.

On MR E. Barandarian had a monopoly on the word "clown".

Wait, wasn't his a specific type of clown?

Anyway, google fisa court and top headline is

"FISA Court Has Rejected .0003 Percent Of All Government Surveillance Requests"

Does Obama and Holder think that FISA court is a real court? Clowns.

Someone will win and someone will loose for sure. Today's wealthy and powerful will be the most insulated seems like a good prior given any change, however, even though some small faction will get screwed.

The group that stands to loose the most is the non-powerful who try to organize, and oppose the consensus of the powerful. Given that globalization and spying are jointly distributed with a high positive rho, and are strongly favored by the elite, I expect lower upper class and middle class Americans who oppose either to loose out big time.

"The people who have the most to lose are powerful people who have committed some wrongdoing, or who have done something which can be presented as wrongdoing, whether or not it is."

Sure, but they also have by far the most ability to control whether or not knowable embarrassing facts are distributed widely. By, for example, blackmailing people who find those knowable facts by threatening to blackmail them. After all, just because the data is all out there doesn't mean people are going to look for it. Or look *at* it, unless it's thrust in front of them.

The people who have the most to lose may be the powerful, but the people who will *actually* lose the most are the average people who have embarrassing things they would prefer to hide, but have no means or leverage to prevent their average enemies from spreading those embarrassing things far and wide.

I'm not as optimistic as you. The dispersed costs/concentrated benefits problem comes into mind. Big business probably has a relatively low prior that detrimental data will surface (especially with metadata operations), that if it does it will be used against them, and that if it is used against them profits will hurt. Americans as a whole obviously don't care about security.

On the other hand, the military-intelligence-industrial-complex has a near certain Bayesian prior that they will benefit from the security state. Better put, they have a certain prior that they will be destroyed without it. So they have an incentive to lobby hard to keep it alive. If big business does manage to lobby against the state, it will be only against the last two links in the chain ("that if it does it will be used against them, and that if it is used against them profits will hurt. Americans as a whole obviously don't care about security").

Americans will still be monitored. Oh, and, from some thoughts on this: http://bit.ly/11CCN8v
"By the way, we know that Americans don’t trust their government. A recent study showed, however, that people respond to surveys with “cheap talk”, as a means to display political affiliation. What if our indifference towards surveillance is a revealed preference that we do, indeed, trust our government?"

We may hate our government, but perhaps we trust it?

If there is hope, it lies with Proles.

Sorry, Tyler, you're missing the point entirely. The danger of the surveillance state is not to any particular person, though of course if any particular person were to become "interesting" to the state, they would have no recourse at all. The danger is to free society as a whole. The 1st Amendment guarantees the right to peaceably assemble - but in our current system, we can't do that without the state knowing. As long as we keep to ourselves, or assemble to discuss state-sanctioned topics, it's no problem, but if we discuss something the state doesn't like, we become "interesting" - and it's a small step from "interesting" to "terrorist."

Those who fear the emergence of a police state overestimate the competence of our government. Consider how much information there is to be known, and question how well you can catalog and retrieve it. They might be successful at torpedoing the careers of a few rising stars, but the media can do an adequate job plucking fruit from the grapevine. It did not take long to uncover improper IRS activity perpetrated with confidential information. The victim usually knows something is wrong and is inclined to complain about it. And people have causes of action in federal court for violations of law by federal agencies.

Information is only as good as your ability to use it. Aside from the ability to use information, there also has to be a willingness. Would someone at the NSA take an interest in destroying TC or prior approval or Andrew' or Willitts just because of something we wrote here?

I don't fear government because indifference is as prevalent as incompetence and cowardice.

"It is not obvious that the President, Congress, and Supreme Court should welcome such an arrangement."

Did you notice how President Obama immediately adopted most of his predecessor's policies, even the ones he'd loudly argued against as a senator and on the campaign trail, as soon as he got into office? Wasn't that weird? It was like somebody whispered something in his ear right after inauguration that instantly changed his mind about them all. I wonder what could have done that.

The same thing that changed candidate Bill Clinton's mind about SDI - New information that blew his campaign rhetoric out of the water.

Either that or threats by the Illuminati.

"Either that or threats by the Illuminati."

Good point-- the idea that a powerful politician could be blackmailed in this day and age is obviously a ludicrous conspiracy theory. It's not like we know of any *actual* groups that have a strong interest in the course of US policy, have an entrenched culture of playing fast-and-loose with the law, and are capable of extensive surveillance of all US citizens.

Police states don't require enormous competence. North Korea manages to be one, despite being unable to keep their population fed or heat their buildings. The Soviet Union was not known for the endless competence and dilligence of its public sector, yet they managed a pretty good police state for decades.

At any rate, I don't think a police state is anyone's first concern with this stuff. Long before you get to locking people up, you get a scary concentration of power in the hands of the people with access to the data. Those very people have institutional interests--they want their jobs to be easier, they don't want their programs cut, they don't want public debate about their programs. Wouldn't it be kind-of amazing if everyone working with all that data was so high-minded that they never used the data to protect their funding or silence their critics? Similarly, partisans do some pretty amazing stuff to gain or keep power. Wouldn't it be pretty striking if *everyone* in a Democratic or Republican administration was too high-minded to peek at that data and see if they could learn something useful?

Now, you might imagine that they would refrain from doing that because they would fear going to jail. But I don't think that's much protection, because in general, when illegal surveillance, kidnapping, murder, and torture has come out in the war on terror, the only people who have gone to jail were the whistleblowers, or very low-level people who overstepped their authority. (Just following orders *is* a defense, as it turns out, and powerful people who give the orders are above legal consequences. You know, like in Russia or China or Pakistan.)

You might think that at least they would not abuse that power because of the public outcry if they got caught. But again, they have been caught violating the law and spying on people, and there were few consequences. And the war on whistleblowers and spying and prosecuting journalists who publish leaked documents is pretty explicitly aimed at making that kind of public disclosure a lot more rare.

You might at least hope that the partisan use of this data would be kept to a minimum because there will be both Republicans and Democrats in postions of power within these agencies and companies. That's possible. But it probably means that the mainstream of the two parties is protected, but probably not anyone else. The GOP turned off the microphones of people who got off message at their last two conventions--it sure seems like a bipartisan agreement not to spy on John Boehner or Nancy Pelosi might not extend to annoying gadflies like Ron Paul or Alan Grayson. Certainly that kind of protection wouldn't extend to the sort of marginal people involved in Occupy or antiwar protests.

This is mostly troll. The assumption is that transparency will ever be applied to the communications of those who have the greatest power over it. The falseness of that assumption brings the entire thread to a halt. There are no cases in which the rulemakers will have anything like the universal transparency the less-powerful will be forced to endure.

I can't get rid of or skip past the French language FBI warning to get to my screaming kid's Barney video, while before most people even know what data mining is the government is already using it to make our lives worse. That's really all I need to know about Public Choice.

The fallacy is giving the executive branch responsibility for stopping terror attacks and then expecting them not to use any legal tool they can to do it. Suppose there are terror attacks that could have been spotted by snooping on Facebook and they aren't stopped because the government isn't snooping on Facebook. A hue and cry will occur about this. The only way to control it is to have very specific laws about what us allowed and what is not. Violators of these laws inside government should be strictly punished even if their intentions are good. There should be confidential access by whistle blowers to Congressional oversight committees, if such a thing does not already exist.

The problem is that nothing could be stopped and the evidence is that nothing has been stopped. The only things that have been "stopped" are when the FBI goes in, recruits some disaffected or mentally unstable youths, entraps them and then stops them. They then publicize this "success" with the added bonus that the public thinks that terrorist cells are more prevalent than they actually are. What in your model you attribute to fear on the part of the Executive is more about being caught with their pants down is really the government looking to justify these powers. Has anyone lost their jobs over not preventing the actual terrorist attacks?

The Grand Strategy of the Bush-Obama Administrations has been:

Invade the World
Invite the World

Thus, obvious examples of refugee system abuse, like the Tsarnaevs and Todashev, don't register on the conventional wisdom as reasons to keep out dubious foreigners, they register as proving the government needs to spy on Americans more.

I bet I could show mathematically that actual police work beats spying. Didn't everyone agree that what we failed at was human intelligence?

The Light of Other Days is a decent effort by Stephen Baxter to flesh out an idea of Arthur C. Clarke's: the destruction of all secrecy and privacy after wormhole technology which allows viewing of people and events at any point in time or space.

Tyler, very much appreciate your coverage on this subject. The conceptual analyses are much better than what I've been seeing anywhere except for Ygesias' posts.

Given that there are roughly one million people with top-secret access, and that Bradley Manning and Edward Snowden seem to have been both very junior and lacking in rare talent, is it even conceivable that the Russians and Chinese have been unable to secure all of the Prism information over our shoulders? The system seems inconceivably vulnerable to spying.

There's another country besides Russian and China with a long history of successful spying on the U.S., a country whose military-intelligence complex dominates the American market for telephone software.

Trying to solve your puzzle. Who? Israel? UK? India? Germany? What's the answer?

I think a lot of people are missing the point on this. In today's world, almost everyone is breaking laws constantly, without knowing it, and constantly saying things (in emails etc) that taken out of context, could look bad. The people with access to this information now have a tool to blackmail anyone they choose. If you don't play ball, just make an example of you. This has got to have an increase in power for the people with access to to information. Maybe there are enough checks and balances in the US system to take care of this, but there sure as hell isn't in the other countries.

Another point - now this data is known to be available, isn't it subject to discovery in any legal action? Many people use private emails and unregistered cell phones to get around the discovery process. Why wouldn't a federal prosecutor in an insider information case say, not want and be entitled to access to this data? Once that data is available for general legal matters, how secure could it really be?

Maybe some, but not as much as you'd think.

Requests must be calculated to lead to the discovery of ADMISSIBLE evidence. There is a balancing test between the probative value of the information and the likelihood of undue prejudice. Knowing of the existence of information makes it harder to thwart discovery, but it doesn't change the rules of evidence. In criminal matters, there are still fourth amendment concerns. Once information is publicly available, it does obviate many privileges. Don't underestimate the ability of the legal profession to obscure obvious truths in a deluge of highfalutin self-deception. You'd be amazed at how often truth is of little value.

How far are courts generally willing to go to impose burdens on third parties for electronic discovery? If I'm suing my wife would the courts be willing to burden the NSA (an unrelated party to the dispute) to trawl its Petabyte database for details of my wife's phone calls to her lover? Gratis?

In any case, since NSA is not the primary source of the tracking data, would it be an admissible defense that the data could be more easily provided by the primary source? (e.g. the phone company)

Finally, there's the distinction between data and analysis: NSA's main strength may be its ability to internally data-mine & cross-correlate diverse data-sources (through NSA's fancy algorithms / hardware) which are not very meaningful when viewed individually. Even if courts can compel third parties to release raw-data, can they compel them to analyse it and produce the more useful versions?

I don't know the answers. Maybe Willits, our resident lawyer, knows! :)

Universal Web Designers has a broad portfolio of clients, and this includes organisations in both the private and public sectors, that sell B2B or B2C, offering products or services.

Universal Website Designhas a broad portfolio of clients, and this includes organisations in both the private and public sectors, that stsnzhou B2B or B2C, offering products or services.

Comments for this post are closed