Analyst has left a new comment on your post "Double
take here....":
Looking at the recent posts on the Bain threads on kiwiblog, it is quite clear that the majority of the contributors have no understanding of probabilistic reasoning in law.
More worryingly, it seems from Fisher's report that he also has little understanding of it. He cites Bayesian reasoning, but then gives an example of armies and gladiators which show quite clearly that he does not understand it. This lack of understanding, I assume, explains why he claims that Binnie did not weigh up the evidence as a whole - which in fact he quite obviously did.
I have reviewed Fisher's reports on other compensation claims for his decision processes in light of this, and there is absolutely no evidence of him having applied Bayesian analysis or even the principles underpinning them. He appears to have operated a far less sophisticated and less independent approach.
This is very alarming. It means that knee-jerk response is operating at the highest levels in New Zealand. Which in turn means the chance of dispassionate fairness is almost nil. This affects not just Bain, but potentially every contested case before the courts.
Looking at the recent posts on the Bain threads on kiwiblog, it is quite clear that the majority of the contributors have no understanding of probabilistic reasoning in law.
More worryingly, it seems from Fisher's report that he also has little understanding of it. He cites Bayesian reasoning, but then gives an example of armies and gladiators which show quite clearly that he does not understand it. This lack of understanding, I assume, explains why he claims that Binnie did not weigh up the evidence as a whole - which in fact he quite obviously did.
I have reviewed Fisher's reports on other compensation claims for his decision processes in light of this, and there is absolutely no evidence of him having applied Bayesian analysis or even the principles underpinning them. He appears to have operated a far less sophisticated and less independent approach.
This is very alarming. It means that knee-jerk response is operating at the highest levels in New Zealand. Which in turn means the chance of dispassionate fairness is almost nil. This affects not just Bain, but potentially every contested case before the courts.
Fisher's report includes the sentence,
ReplyDelete"85. Take David's fingerprints in blood on the rifle." But didn't the final evidence show that the fingerprints were not in any type of blood?
Which is probably why Binnie said that Fisher was unfamiliar with the evidence. Fisher seems to have been working completely from the briefing, which only presents the view of Crown Law and the Police. They maintain the prints were in blood despite the evidence. Mr Jones said they were, so they were.
DeleteWell, he was 'working with' the Crown in terms of their objections.
DeleteIt seems that your good friend Mr Kent Parker is now professing on kiwiblog to be an expert on Bayes theorem. Sadly for his pretensions, he makes some very fundamental errors in his supposed explanation.
ReplyDeleteFirstly, and most heinously for a bayesian, he transposes the conditional.
Secondly, and related to this, he clearly has very little comprehension of how probabilities are assigned for use of Bayes theorem.
Thirdly, he fails to recognise the necessity for comparison.
Fourthly, and perhaps most risible, he refers readers to the absurd calculation of probabilities he has posted on his website. Anyone with even quite a basic understanding of probabilistic reasoning would get a lot of enjoyment from his calculations there, and a Bayesian would probably wet themselves laughing!
are Fishers other reports available online somwhere? If you are correct that indeed would seem very relevant. the point that most are missing in this.
ReplyDeleteScott
His decision on Aaron Farmer (innocent BRD, no extraordinary circumstances necessary but discussed nonetheless)is available on the MoJ web site. The decision on Rex Haig is available under the OIA - you have to request it.
DeleteMy reading of both suggests to me that he takes an 'additive' approach to evidence: a common error and one made by the Crown Law and Police in their response made to Binnie's report. That is, that a lot of weak evidence adds up to something strong, and that what should underlie a judgement is how compelling is the narrative account for that evidence. But unable to discount something 'hard' like DNA evidence.
Fisher's method needs the number fractions he calls 'likelihood ratios', but a couple of the ratios needed in his simple multiplication are so close to zero that he can do the simple math in his head. Binnie used 'scarcely imaginable', which equates to 0.00000000000000000000001, as the likelihood ratio that Robin could have been 'peacefully' ambushed at contact shot range by anyone other than himself.
ReplyDeleteBinnie used 'implausible' for the likelihood ratio that the crime scene could have been left 'open to discovery' for an hour to do a paper run, with the most dangerous person no further than 6 steps away, but the likelihood ratio for that is 0.
Robin was 'clean shaven', according to Lodge so, in further calculations, Fisher can use the 1/1 likelihood ratio that Robin shaved in the bathroom, and was certainly in the house before sunrise.
Yes they are. Probably his name followed by Haig or Compensation, the other case was the man falsely imprisoned for rape - the details which I can't recall at the moment. I think he got both those decisions right largely, with the exception that I don't think he dealt with the Haig case as fractured for all time by the evidence of the immunity witness.
ReplyDeleteIn the Binnie case, the Minister made her feelings known as to what she and the Crown considered was wrong. So there was a 'design' toward a preconceived result operating. That along with the 'secrecy' and exclusion of Binnie, the attack on Binnie and so forth reveal compliance toward a desired result. It was a poor look at best, and a hospital pass.
I read the MOJ one, Interesting...
ReplyDeleteDo you know where I can see any other compensation claim reports written by anyone other than Fisher? (I.e. to see what the range of methodologies used in NZ cases are?)
Scott
The 2011 memo to Cabinet and cabinet minute (available from the MOJ) about Knight & Johnston suggests that there was no report commissioned and that it was the Ministry who provided advice.
DeleteThe report of the Thomas Royal Commission is here http://homepages.paradise.net.nz/r.christie/thomas_royal_commission_1980.pdf. Note that the then Solicitor General and Police did not accept the findings (although the government did). Fisher was junior Counsel for the Police at that Commission, and took a case to appeal during the Commission to argue that the Commissioners were biased against the Police.
Kirsty MacDonald produced the reports on the three Auckland girls who did get compensation but less than their lawyer thought they ought to.
Fisher says in his report (page 10):
ReplyDelete"The historical purpose of the ex gratia compensation discretion was to compensate the innocent, not to root out official misconduct. To the extent that condemning official misconduct has been engrafted onto that process, it should be reserved for those cases which are so egregious that they threaten the integrity of the judicial system. Planting false evidence is a classic example. Overlooking a possible line of investigation is not."
With all due respect to Mr Fisher, he does not cite any authority for his conclusion that 'overlooking a line of investigation' does not threaten the integrity of the justice system; and I, at least, take issue with that glib statement. In effect, he is saying that it is perfectly acceptable practice for the Police to follow only their 'preferred' line of investigation.
It does not take any great intellect or legal nous to see that such an attitude unavoidably "threatens the integrity of the justice system"! If all judges thought like that; if appeal courts agreed, then there is no point whatsoever in having any pretence at all at fairness, dispassionate presentation of facts and the like for we would be living in a police state served by a tunnel-visioned judiciary.
The previous Minister, in his instructions to Binnie, specifically mentioned the 'failure to properly investigate innocence' as an example extraordinary circumstance. Yet Fisher is saying not. If, as Fisher claims, it is dependent on a judicial finding of malfeasance, then we are left with the police investigating themselves, the courts investigating themselves, all with the history of finding in their own favour. How can that possibly be 'fair'? How can it possibly be 'natural justice'?
While I agree that Binnie exceeded his mandate in recommending compensation, that was a minor matter, easily rectified. The other criticisms just do NOT stand up to scrutiny. In fact, Binnie did very little more than Fisher did in the Farmer report. My question is: why was Binnie's instruction different from that which went to Fisher for that case? Why was the request to give an opinion on extraordinary circumstances diluted down to merely noting items to be considered? My inference is that this was damage control: a desperate attempt - and it must have been at a level of at least the Ministerial advisers - to stop the Police and Justice system having their faults identified.
Mr Parker is at it again. His latest attempt at assigning probabilities looks flash, but the fundamental misunderstanding of probabilistic reasoning in law continues.
ReplyDelete