> I would go so far as to say that it's barely a court at all. More like a panel of judges. Is it a court if it hears no controversies and effectively operates outside the system of review by higher courts?
It doesn't effectively operate outside of the system of review by higher courts, its just that, since the proceedings are non-adversarial, the only review likely is of decisions that are adverse to the government, which are made unlikely, since there is no constraint on the government's ability to -- even assuming no deliberate misrepresentation -- present information selectively in the initial application.
> But it would only be natural for an entity like FISC, which serves the government alone and has effectively no oversight from higher courts
FISC has a higher court (the Foreign Intelligence Surveillance Court of Review) which exists solely to provide oversight for it; FISCR is itself overseen by the Supreme Court.
What the FISA process is missing isn't oversight by higher courts, which it incorporates already, but an advocatus diaboli.
The FISA process concerns surveillance of foreign nationals who enjoy extremely limited constitutional protections.
People are fond of pointing out that the government rarely loses at FISA hearings. But as it turns out, of course, the government also rarely loses when it files for domestic wiretap authority. Relative to other transactions investigators and prosecutors have with the courts, the process of obtaining authority to wiretap is extraordinarily difficult; prosecutors don't waste time doing wiretap paperwork if they're not sure the request will be granted.
Two different processes, neither of them adversarial, both of them secret, both of which produce an authority to surveil people. My question: why would the process that surveils foreigners without Fourth Amendment rights require a devil's advocate, when the domestic wiretap process has no such mechanism?
I ask in part because I think synthesizing an adversary is unlikely to create a meaningful restraint; a synthetic adversary doesn't have the same incentives as a real one.
> My question: why would the process that surveils foreigners
Right here is the fundamental problem. The lack of effective constraints in the FISA process (and we've already discussed, in this thread, why the domestic process targeting criminal action, despite superficial similarity, has stronger incentives for government restraint than the FISA process) is exactly why we cannot trust that it only surveils foreigners (it is intended to do that, but absent effective constraint there can be no confidence that it is used as intended.)
> I ask in part because I think synthesizing an adversary is unlikely to create a meaningful restraint; a synthetic adversary doesn't have the same incentives as a real one.
Sure, from the perspective of constraint, a synthetic adversary is worse than a real one, and selecting a synthetic adversary in a manner which makes it an effective constraint is tricky, but getting a real adversary into the foreign intelligence surveillance warrant process is, well, difficult to do without serious negative consequences.
I submitted a comment a few minutes ago to this effect, but it's as if it never happened, so here goes again. If it should reappear, please forgive the double-posting.
> The FISA process concerns surveillance of foreign nationals who enjoy extremely limited constitutional protections.
It's unclear, from the public's perspective, that FISA warrants are only being directed against foreign entities.
> People are fond of pointing out that the government rarely loses at FISA hearings. But as it turns out, of course, the government also rarely loses when it files for domestic wiretap authority. Relative to other transactions investigators and prosecutors have with the courts, the process of obtaining authority to wiretap is extraordinarily difficult; prosecutors don't waste time doing wiretap paperwork if they're not sure the request will be granted.
Two different processes, neither of them adversarial, both of them secret, both of which produce an authority to surveil people. My question: why would the process that surveils foreigners without Fourth Amendment rights require a devil's advocate, when the domestic wiretap process has no such mechanism?
I hate to keeping beating this horse. But the argument that domestic warrants are equally non-adversarial keeps coming up. So I guess the horse isn't dead. Domestic warrants are part of an adversarial process, if only after the fact. They are typically used to gather evidence for a criminal prosecution. When that happens, the defendant has a right to challenge the warrant. This serves as a counterbalance. Judges, prosecutors, and police know that a bad warrant will likely be thrown out if the criminal case is brought to trial. Thus, they have a very real incentive to write good, defensible warrants. This means it's to their advantage to keep their warrants narrow in scope. And thus the goal of preserving privacy is served.
> I ask in part because I think synthesizing an adversary is unlikely to create a meaningful restraint; a synthetic adversary doesn't have the same incentives as a real one.
Do you mean that there aren't incentives for the opponent of the warrant, or for the government?
If the former, I can only point out that organizations like the ACLU can be quite zealous. I would much prefer advocacy by the ACLU than no advocacy at all.
If the latter, the restraint would take two forms. First, judges and investigators would be aware that overbroad or otherwise improper warrants can be undone. This would incentivize writing good warrants. Second, the warranted surveillance could be halted upon a successful challenge, assuming said surveillance is ongoing. Both of these are meaningful checks.
It doesn't effectively operate outside of the system of review by higher courts, its just that, since the proceedings are non-adversarial, the only review likely is of decisions that are adverse to the government, which are made unlikely, since there is no constraint on the government's ability to -- even assuming no deliberate misrepresentation -- present information selectively in the initial application.
> But it would only be natural for an entity like FISC, which serves the government alone and has effectively no oversight from higher courts
FISC has a higher court (the Foreign Intelligence Surveillance Court of Review) which exists solely to provide oversight for it; FISCR is itself overseen by the Supreme Court.
What the FISA process is missing isn't oversight by higher courts, which it incorporates already, but an advocatus diaboli.