4 ms·

One problem with that line of thought is that law is so complex that just about anybody could easily be tripped up by an agent educated on the nuances of the la

One problem with that line of thought is that law is so complex that just about anybody could easily be tripped up by an agent educated on the nuances of the law. While most agents are honest, truth-seeking individuals, clearly some harbor political bias and/or are being used for ideological ends. “Show me the man and I’ll show you the crime” - Lavrentiy Beria, Chief of Police under Stalin A system that is highly secretive and whose only accountability is to pretend to investigate itself inevitably will be abused more than an open one with transparent checks and balances. One has only to see that FISA has only turned down 85 out of 40k requests to understand this is a rubber-stamp process. (source: https://en.wikipedia.org/wiki/United_States_Foreign_Intelligence_Surveillance_Court https://en.wikipedia.org/wiki/United_States_Foreign_Intellig...)

mehrdadn 7y ago
> One has only to see that FISA has only turned down 85 out of 40k requests to understand this is a rubber-stamp process. I don't think that logic follows? That statistic could just mean that they're actually doing most things legally and properly, which is the outcome you'd want, right? I'd be worried if a significant percentage of their requests were turned down. Also, as a comparison point: apparently several federal judicial districts have a 100% conviction rate, and overall, federal conviction rates are pretty high. Do you see these statistics as implying the whole thing is just a rubber-stamp process altogether? [1] [1] http://justicedenied.org/issue/issue_67/federal_courts_jd67.pdf http://justicedenied.org/issue/issue_67/federal_courts_jd67....
- _jal 7y ago
  > Do you see these statistics as implying the whole thing is just a rubber-stamp process altogether? That's overstating my (not the GP) view, but I do believe the operational mechanisms in federal law enforcement push heavily towards conviction once accused. Let me ask you a question: Do you believe there have been zero false convictions in those districts?
  - mehrdadn 7y ago
    > That's overstating my (not the GP) view, but I do believe the operational mechanisms in federal law enforcement push heavily towards conviction once accused. Possibly, but that's a separate point from the one I'm making. > Do you believe there have been zero false convictions in those districts? Seems unlikely. But that's not the baseline.
    - _jal 7y ago
      > Seems unlikely. But that's not the baseline. Ok, what is, and on who's authority? And just to be sure I understand, you're judging mechanisms designed to come to factual conclusions by evaluating how many of their decisions lean one way, without reference to or concern with their factual basis?
      - mehrdadn 7y ago
        >> Seems unlikely. But that's not the baseline. > Ok, what is I don't have a number. > and on who's authority? By common sense. The only way to get zero error in the absence of perfect information is to avoid convictions entirely. I'm not saying I like this trade-off; I'm just saying this is reality. > And just to be sure I understand, you're judging mechanisms designed to come to factual conclusions by evaluating how many of their decisions lean one way, without reference to or concern with their factual basis? No, that's the complete opposite of what I was saying, which was that you have to look a the factual bases and can't just judge by the statistics. Hence my reply earlier: https://news.ycombinator.com/item?id=21773010 https://news.ycombinator.com/item?id=21773010
- cartoonworld 7y ago
  >I don't think that logic follows? That statistic could just mean that they're actually doing most things legally and properly, which is the outcome you'd want, right? I'd be worried if a significant percentage of their requests were turned down. I think you're correct that the logic employed doesn't follow. It is useless to look at the statistics alone and conclude one way or the other that FISA is or is not a rubber stamp. However, I think you are doing a category error: Criminal court proceedings are adversarial, FISA court is not. Because there is no defense, public oversight, or judicial review. FISA cases are sealed, and secret. There is no risk of jury nullification or new caselaw because these cases are not argued by an adversary. Outside entities subpoenad in this fashion are barred from speaking of it (Under 99% conviction federal criminal statutes) Nobody is risking re-election here and the judges are appointed. The adversary is a strawman in FISA. The public is functionally shut out as well. It was designed for this purpose, the lack of accountability is a critical feature. In order to employ the statistics properly you require context. In my high school sociology class the instructor posed this statistical fallacy: Drownings and grass growth rates are positively correlated. A naive observer born and living on Mars might ban grass to prevent drownings. This missing the mark because correlation != causation, and the true correlation is more people swim in the summer when grass grows faster. There is not swimming on mars. >Also, as a comparison point: apparently several federal judicial districts have a 100% conviction rate [...] Do you see these statistics as implying the whole thing is just a rubber-stamp process altogether? Comparing to high criminal conviction rates (often in the high 90%), the prosecutors and attornies general use the conviction statistics for their re-election campaigns. Cases which would result in "Jury Nullification" or some other modification of caselaw or jurisprudence are abandoned, keeping the rates artificially high. Refer to "Stingray" caselaw. This effect is partially caused by the adversarial court system being helmed by elected officials. This is to say, that the regular criminal courts are an adversarial public system in which the accused get lawyers. Those lawyers sometimes win in court, other times the AG or prosecutor drops the charges when it looks like they will lose due to the legal arguments employed by the adversary. Dropped or unajudicated charges aren't counted in the statistics. Losing cases are not tried. In order to maintain your claim here, there is some missing data: 1: Who is the adversary in FISA? 2: When does the public review the court proceedings? Given the answers to 1 and 2 being "Nobody" and "Never", the statistics tell a much different tale than the criminal conviction statistics. Now the logic may follow.

mehrdadn 7y ago

> One has only to see that FISA has only turned down 85 out of 40k requests to understand this is a rubber-stamp process. I don't think that logic follows? That statistic could just mean that they're actually doing most things legally and properly, which is the outcome you'd want, right? I'd be worried if a significant percentage of their requests were turned down. Also, as a comparison point: apparently several federal judicial districts have a 100% conviction rate, and overall, federal conviction rates are pretty high. Do you see these statistics as implying the whole thing is just a rubber-stamp process altogether? [1] [1] http://justicedenied.org/issue/issue_67/federal_courts_jd67.pdf http://justicedenied.org/issue/issue_67/federal_courts_jd67....

_jal 7y ago
> Do you see these statistics as implying the whole thing is just a rubber-stamp process altogether? That's overstating my (not the GP) view, but I do believe the operational mechanisms in federal law enforcement push heavily towards conviction once accused. Let me ask you a question: Do you believe there have been zero false convictions in those districts?
- mehrdadn 7y ago
  > That's overstating my (not the GP) view, but I do believe the operational mechanisms in federal law enforcement push heavily towards conviction once accused. Possibly, but that's a separate point from the one I'm making. > Do you believe there have been zero false convictions in those districts? Seems unlikely. But that's not the baseline.
  - _jal 7y ago
    > Seems unlikely. But that's not the baseline. Ok, what is, and on who's authority? And just to be sure I understand, you're judging mechanisms designed to come to factual conclusions by evaluating how many of their decisions lean one way, without reference to or concern with their factual basis?
    - mehrdadn 7y ago
      >> Seems unlikely. But that's not the baseline. > Ok, what is I don't have a number. > and on who's authority? By common sense. The only way to get zero error in the absence of perfect information is to avoid convictions entirely. I'm not saying I like this trade-off; I'm just saying this is reality. > And just to be sure I understand, you're judging mechanisms designed to come to factual conclusions by evaluating how many of their decisions lean one way, without reference to or concern with their factual basis? No, that's the complete opposite of what I was saying, which was that you have to look a the factual bases and can't just judge by the statistics. Hence my reply earlier: https://news.ycombinator.com/item?id=21773010 https://news.ycombinator.com/item?id=21773010
cartoonworld 7y ago
>I don't think that logic follows? That statistic could just mean that they're actually doing most things legally and properly, which is the outcome you'd want, right? I'd be worried if a significant percentage of their requests were turned down. I think you're correct that the logic employed doesn't follow. It is useless to look at the statistics alone and conclude one way or the other that FISA is or is not a rubber stamp. However, I think you are doing a category error: Criminal court proceedings are adversarial, FISA court is not. Because there is no defense, public oversight, or judicial review. FISA cases are sealed, and secret. There is no risk of jury nullification or new caselaw because these cases are not argued by an adversary. Outside entities subpoenad in this fashion are barred from speaking of it (Under 99% conviction federal criminal statutes) Nobody is risking re-election here and the judges are appointed. The adversary is a strawman in FISA. The public is functionally shut out as well. It was designed for this purpose, the lack of accountability is a critical feature. In order to employ the statistics properly you require context. In my high school sociology class the instructor posed this statistical fallacy: Drownings and grass growth rates are positively correlated. A naive observer born and living on Mars might ban grass to prevent drownings. This missing the mark because correlation != causation, and the true correlation is more people swim in the summer when grass grows faster. There is not swimming on mars. >Also, as a comparison point: apparently several federal judicial districts have a 100% conviction rate [...] Do you see these statistics as implying the whole thing is just a rubber-stamp process altogether? Comparing to high criminal conviction rates (often in the high 90%), the prosecutors and attornies general use the conviction statistics for their re-election campaigns. Cases which would result in "Jury Nullification" or some other modification of caselaw or jurisprudence are abandoned, keeping the rates artificially high. Refer to "Stingray" caselaw. This effect is partially caused by the adversarial court system being helmed by elected officials. This is to say, that the regular criminal courts are an adversarial public system in which the accused get lawyers. Those lawyers sometimes win in court, other times the AG or prosecutor drops the charges when it looks like they will lose due to the legal arguments employed by the adversary. Dropped or unajudicated charges aren't counted in the statistics. Losing cases are not tried. In order to maintain your claim here, there is some missing data: 1: Who is the adversary in FISA? 2: When does the public review the court proceedings? Given the answers to 1 and 2 being "Nobody" and "Never", the statistics tell a much different tale than the criminal conviction statistics. Now the logic may follow.