5 ms·

A coin toss with little false negatives might be a good filter for a more annoying or expensive test. I think that would be a fine use of the tool even if I ag

A coin toss with little false negatives might be a good filter for a more annoying or expensive test. I think that would be a fine use of the tool even if I agree that as primary evidence it might be lacking.

jariel 7y ago
Yes, that's just it, proportionality. Dogs might be good to sniff out contraband or whatever to flag searches, but perhaps not as incontrovertible evidence of something.
- notahacker 7y ago
  How often do cases rely on dogs as 'incontrovertible evidence of something though'. It's not like they testify! If the dog's mentioned at all, surely normally it's background context to why the person was searched and the drug found [and if the police are lying about the finding, it's not the dog's fault!]
  - liability 7y ago
    Nothing is the dog's fault, because the dog is a dog. But that's beyond the point.
    - notahacker 7y ago
      The point is that if a law enforcement person lies about finding something due to their dog, the problem is juries [reasonably] trust the law enforcement personnel are telling the truth about what they find afterwards, not that they trust the dog's instincts. Indeed your average jury is almost guaranteed to have someone with enough interest in dogs to know they sometimes react to things other than suspicious substances.
      - jariel 7y ago
        The cop has 1000 ways to compromise justice using any and all tools, so this is a secondary issue. If dogs can indeed sniff out stuff in a manner consistently beyond random, then they are a good first pass tool.
  - unishark 7y ago
    I think the complaint is still that a dog's behavior is used as probable cause to allow a search in the first place.
  - DanBC 7y ago
    In some places they have "fruit of the poisoned tree" -- there are clear rules for searches, and evidence obtained without using those means is not admissible in court. It feels weird that law enforcement get just avoid all of that by training dogs to alert at a handler's signal.
    - alistairSH 7y ago
      I didn't elaborate too much in the first post, but this is the gist of it. Any search that was enabled by a dog should be considered "fruit of the poisoned tree". Between their lack of accuracy and a dog's bias towards its handler, they should not ever serve as a basis for search or arrest. When the justification for initial stop/search comes up in court, the dog's hit is held up as incontrovertible proof that a search was justified. Courts have upheld their use many times, all the way to SCOTUS IIRC. This approach is far to prone to misuse by law enforcement. It's bad enough they can pull you over without cause (most states have loosely defined "irregular driving" laws), they can initiate a search of your car just as baselessly.
      - jariel 7y ago
        ""fruit of the poisoned tree". Between their lack of accuracy and a dog's bias towards its handler, they should not ever serve as a basis for search or arrest." I think this is upside down ... the police are never going to have 'incontrovertible proof' of contraband before they do a full search by definition. If they did need such 'total evidence' to search, they wouldn't need to do the search at all ... they could just make the arrest. And a dogs hint is not presented as evidence of a crime, merely reasonable suspicion for a search, which is what we want the tool for. The point is to provide some degree of a reasonable basis for suspicion, after which more effort is used to determine the material outcome: either you have 1Kg of coke, or an illegal gun ... or you don't. The entire point of the tools are to use something that demarcates reasonable suspicion; dogs, and other things, are viable tools. If you attempt to enter a country, and you are asked where you are staying but can't provide an answer, or if you do not have a return flight - these are indications of 'reasonable suspicion' of further questioning.
- lonelappde 7y ago
  Absolutely not OK! K9 units are OK used to fabricate "reasonable suspicion" for a search when there is none. It's a component of overpolicing.
  - jariel 7y ago
    This is a little bit of a secondary issue. Every tool can be misapplied or misused. If the 'dogs' can provide a scientifically credible means for 'flagging' that's significantly above random chance, but still below the threshold 'beyond a shadow of a doubt' they can absolutely be used. If a police officer smells the odor of marijuana in the car, or alcohol on the breath, or, for example, sees a crack pipe or 'empty beer' cans - those things are all examples of reasonable suspicion and it's the whole point. That those powers can be abused is again a secondary issue. Finally, is the obvious issue that there's very little harm done otherwise, the 'false flags' from either dogs or 'empty beer' cans that result in searching doesn't compromise the individuals or the community at large: a 'blow test' or a 'legal search' results in more accurate presentation of justice, not less.
    - alistairSH 7y ago
      If the 'dogs' can provide a scientifically credible means for 'flagging' that's significantly above random chance That's the problem. They cannot. " narcotics-detecting dogs and their handlers aren’t very good at discerning the presence of illegal drugs. Multiple analyses of drug-dog alerts have consistently shown alarmingly high error rates — with some close to and exceeding 50 percent. In effect, some of these K-9 units are worse than a coin flip." https://www.washingtonpost.com/opinions/2019/02/05/supreme-courts-alternative-facts-about-drug-sniffing-dogs/ https://www.washingtonpost.com/opinions/2019/02/05/supreme-c...
- LanceH 7y ago
  There is nothing scientific about it. They don't present the dogs when something isn't found. They don't provide a record of how frequently something was/wasn't found when a dog alerted. They just classify it differently when the dog was wrong, "trace presence of drugs was indicated by the dog". Instead of, "the dog was wrong" and is wrong x% of the time.
kaonashi 7y ago
If it were truly neutral, dogs respond to and enable the handler's biases.
matheusmoreira 7y ago
> A coin toss with little false negatives might be a good filter for a more annoying or expensive test. Highly sensitive tests are great for this. They're usually cheap and quick. They have a low false positive rate so a negative result is likely to be correct. Since they detect anything that even remotely looks like the targeted substance, they have a substantial false positive rate and a positive result must always be confirmed with specific tests. However, police forces seem to treat a positive result as sufficient cause for an arrest. https://www.ncbi.nlm.nih.gov/m/pubmed/12423006/ https://www.ncbi.nlm.nih.gov/m/pubmed/12423006/ Without confirmation, false positive rates varied from 2 to 4 percent.
- jdsully 7y ago
  > police forces seem to treat a positive result as sufficient cause for an arrest My impression from watching too much COPS and LivePD is that police don't really see arresting people as that serious, "They'll be out in few hours". For the people they arrest most often it's probably not that big a deal either. But if it's your first ever then your lifestyle is under serious and permanent threat. EDIT: Removed "middle class". Getting arrested is terrible no matter what your economic or class status is.
  - Espressosaurus 7y ago
    What do you think happens when you're trying to get to work at a shitty retail job and you get arrested? If anything, it's worse for the working class folks because they don't have a buffer, their friends and family likely don't have one either, and they are considered infinitely disposable by their employers.
    - jdsully 7y ago
      It's not so much about class but whether or not your already a felon. Most of the arrests you see on those shows are people out on parole or with massive rap sheets. Going from no criminal record to your first is a huge change of circumstance vs getting another incremental charge. I shouldn't have used the word middle class as it distracts from the core dynamic at play.

jariel 7y ago

Yes, that's just it, proportionality. Dogs might be good to sniff out contraband or whatever to flag searches, but perhaps not as incontrovertible evidence of something.

notahacker 7y ago
How often do cases rely on dogs as 'incontrovertible evidence of something though'. It's not like they testify! If the dog's mentioned at all, surely normally it's background context to why the person was searched and the drug found [and if the police are lying about the finding, it's not the dog's fault!]
- liability 7y ago
  Nothing is the dog's fault, because the dog is a dog. But that's beyond the point.
  - notahacker 7y ago
    The point is that if a law enforcement person lies about finding something due to their dog, the problem is juries [reasonably] trust the law enforcement personnel are telling the truth about what they find afterwards, not that they trust the dog's instincts. Indeed your average jury is almost guaranteed to have someone with enough interest in dogs to know they sometimes react to things other than suspicious substances.
    - jariel 7y ago
      The cop has 1000 ways to compromise justice using any and all tools, so this is a secondary issue. If dogs can indeed sniff out stuff in a manner consistently beyond random, then they are a good first pass tool.
- unishark 7y ago
  I think the complaint is still that a dog's behavior is used as probable cause to allow a search in the first place.
- DanBC 7y ago
  In some places they have "fruit of the poisoned tree" -- there are clear rules for searches, and evidence obtained without using those means is not admissible in court. It feels weird that law enforcement get just avoid all of that by training dogs to alert at a handler's signal.
  - alistairSH 7y ago
    I didn't elaborate too much in the first post, but this is the gist of it. Any search that was enabled by a dog should be considered "fruit of the poisoned tree". Between their lack of accuracy and a dog's bias towards its handler, they should not ever serve as a basis for search or arrest. When the justification for initial stop/search comes up in court, the dog's hit is held up as incontrovertible proof that a search was justified. Courts have upheld their use many times, all the way to SCOTUS IIRC. This approach is far to prone to misuse by law enforcement. It's bad enough they can pull you over without cause (most states have loosely defined "irregular driving" laws), they can initiate a search of your car just as baselessly.
    - jariel 7y ago
      ""fruit of the poisoned tree". Between their lack of accuracy and a dog's bias towards its handler, they should not ever serve as a basis for search or arrest." I think this is upside down ... the police are never going to have 'incontrovertible proof' of contraband before they do a full search by definition. If they did need such 'total evidence' to search, they wouldn't need to do the search at all ... they could just make the arrest. And a dogs hint is not presented as evidence of a crime, merely reasonable suspicion for a search, which is what we want the tool for. The point is to provide some degree of a reasonable basis for suspicion, after which more effort is used to determine the material outcome: either you have 1Kg of coke, or an illegal gun ... or you don't. The entire point of the tools are to use something that demarcates reasonable suspicion; dogs, and other things, are viable tools. If you attempt to enter a country, and you are asked where you are staying but can't provide an answer, or if you do not have a return flight - these are indications of 'reasonable suspicion' of further questioning.
lonelappde 7y ago
Absolutely not OK! K9 units are OK used to fabricate "reasonable suspicion" for a search when there is none. It's a component of overpolicing.
- jariel 7y ago
  This is a little bit of a secondary issue. Every tool can be misapplied or misused. If the 'dogs' can provide a scientifically credible means for 'flagging' that's significantly above random chance, but still below the threshold 'beyond a shadow of a doubt' they can absolutely be used. If a police officer smells the odor of marijuana in the car, or alcohol on the breath, or, for example, sees a crack pipe or 'empty beer' cans - those things are all examples of reasonable suspicion and it's the whole point. That those powers can be abused is again a secondary issue. Finally, is the obvious issue that there's very little harm done otherwise, the 'false flags' from either dogs or 'empty beer' cans that result in searching doesn't compromise the individuals or the community at large: a 'blow test' or a 'legal search' results in more accurate presentation of justice, not less.
  - alistairSH 7y ago
    If the 'dogs' can provide a scientifically credible means for 'flagging' that's significantly above random chance That's the problem. They cannot. " narcotics-detecting dogs and their handlers aren’t very good at discerning the presence of illegal drugs. Multiple analyses of drug-dog alerts have consistently shown alarmingly high error rates — with some close to and exceeding 50 percent. In effect, some of these K-9 units are worse than a coin flip." https://www.washingtonpost.com/opinions/2019/02/05/supreme-courts-alternative-facts-about-drug-sniffing-dogs/ https://www.washingtonpost.com/opinions/2019/02/05/supreme-c...
LanceH 7y ago
There is nothing scientific about it. They don't present the dogs when something isn't found. They don't provide a record of how frequently something was/wasn't found when a dog alerted. They just classify it differently when the dog was wrong, "trace presence of drugs was indicated by the dog". Instead of, "the dog was wrong" and is wrong x% of the time.

kaonashi 7y ago

If it were truly neutral, dogs respond to and enable the handler's biases.

matheusmoreira 7y ago

> A coin toss with little false negatives might be a good filter for a more annoying or expensive test. Highly sensitive tests are great for this. They're usually cheap and quick. They have a low false positive rate so a negative result is likely to be correct. Since they detect anything that even remotely looks like the targeted substance, they have a substantial false positive rate and a positive result must always be confirmed with specific tests. However, police forces seem to treat a positive result as sufficient cause for an arrest. https://www.ncbi.nlm.nih.gov/m/pubmed/12423006/ https://www.ncbi.nlm.nih.gov/m/pubmed/12423006/ Without confirmation, false positive rates varied from 2 to 4 percent.

jdsully 7y ago
> police forces seem to treat a positive result as sufficient cause for an arrest My impression from watching too much COPS and LivePD is that police don't really see arresting people as that serious, "They'll be out in few hours". For the people they arrest most often it's probably not that big a deal either. But if it's your first ever then your lifestyle is under serious and permanent threat. EDIT: Removed "middle class". Getting arrested is terrible no matter what your economic or class status is.
- Espressosaurus 7y ago
  What do you think happens when you're trying to get to work at a shitty retail job and you get arrested? If anything, it's worse for the working class folks because they don't have a buffer, their friends and family likely don't have one either, and they are considered infinitely disposable by their employers.
  - jdsully 7y ago
    It's not so much about class but whether or not your already a felon. Most of the arrests you see on those shows are people out on parole or with massive rap sheets. Going from no criminal record to your first is a huge change of circumstance vs getting another incremental charge. I shouldn't have used the word middle class as it distracts from the core dynamic at play.