CLPEX Chat Board

I think that's one of those things that depends on where and who you ask. I might not even get back to the latent print section that they denied depositing it. Lots of reasons. It might be that the object was stolen property or a container in a case where someone else had been established as the crook, and the investigators might not care much why he lied about it. People lie for all sorts of reasons, including fear of non-specific consequences. It might well just not be worth trying to work out why, and the lie might well not move them to mess with him criminally for lying. Some folks are recognized to be just too dumb to tell the truth when it would have cleared things up. If it mattered, for instance, they had value as a witness, I'd want to clear it up, if for no other reason than to avoid the defense implying various things when they saw the results in the report. It would be sometimes hard to proceed against the person if they merely claimed they had no memory of how it might have gotten there but you can show they had the opportunity.

If I get your meaning correctly about elimination, that they were among the people being eliminated but happened to be identified, it would depend on how they came to be checked. If they were, as is often the case, one of the people with legitimate access and the latents are being checked to eliminate those who had innocent contact, it would depend entirely on local policy for verification in those cases. While it probably should be checked, I probably wouldn't have done it, unless the true actor was also identified from other latents there, and I would want to have all the ducks lined up to properly prove up the elimination. My verifications had to go outside to already busy agencies.

If the McKie or Ross cases was handled differently to other cases can only be sheer speculation. What however is evidenced by certain members of Senior SCRO Staff is that there were serious discrepancies with procedures.

We know that it only needed TWO Experts to check an Elimination print in a Serious Case, however in this one case it took FOUR Experts.

We also know the decision to check ALL eliminations to a 16 point standard has no substance, and again Senior staff have exposed this themselves. Also another Expert from SCRO questioned why this happened.

We know important paperwork was NOT completed as per the guidelines at the time, and was only filled in one week later. Despite an SCRO Expert telling Justice 1 that procedures HAD been correctly followed, again it comes to the Senior Management of SCRO to expose that lie.

We also know that SCRO Experts were prejudging the outcome of the elimination checks. Evidenced by Geddes who stated that Macpherson had told him that a Police Officer was going to be in trouble. BEFORE the print had even been second checked!

We also know the SCRO had decided that mark Y7 was vital because of certain dates, despite having the very evidence at hand which clearly contradicted this evidence.

So whether any other cases within SCRO did not follow procedures, and were subject to ad hoc procedures, such as initialling evidence on behalf of colleagues, we can only speculate.

We know that TWO years of SERIOUS CRIMES were checked, and only those of the discredited Experts, it clearly leaves a huge number of unchecked cases that have never been examined.

The Inquiry will be given this contradictory evidence which seriously exposes the SCRO Experts. And even more unbelievable is that evidence comes from their own management!

The issue of the FOUR Expert checks is further clouded because at the time it was checked, Shirley McKie's name WAS on a list of police eliminations.

Therefore IF it had been IDENTIFIED correctly, as her name was already on a list of Police Eliminations, it would be of little significance to SCRO.

But we have again contradictory evidence from one of the Experts involved, namely Hugh Macpherson. He tells a colleague after his original FIRST check that a police officer was going to be in serious trouble, but at Justice 1 he told them it was simply yet another police elimination, and was of little interest to him. But remarkably he became Judge and Jury on a alleged elimination from a list of police eliminations!

When SCRO identified a print as belonging to one of the Crime Scene Examiners did Macpherson deem that this individual was going to be in serious trouble too?

Macpherson could not possibly know the Officer was going to be in trouble UNLESS he had prior knowledge. And this ties in with the Mackay timeline of the elimination being phoned out AFTER one elimination check. And then it is questioned. But rather than admit an error, the expert who is hero worshipped by many in SCRO then gets a further THREE of his colleagues to support him.

Yet more lies and deception. Yet more problem for individuals at the Public Inquiry. I wonder how many experts are going to have to dramatically change their evidence for their day in Court???

And there is worse to come for Macpherson at the Inquiry. He has been involved in far worse. Far worse! But that will keep until he is under oath. I know how much they want this day to "clear" their names. Bad enough doing it without knowing what lurks beneath the surface, waiting to haunt you....

Thanks Gerald, you have basically answered my question, you seem to be saying that denied rogue IDs do sometimes happen but there is no standard procedure for handling them.

According to my thinking (and this is generally accepted theory) in cases with an infinitesimal chance that the accused is lying because they did something wrong (low prior probability) it is more likely (relatively speaking) that a failure to individualize will landing on them, because the chance that they did a crime and are identified by a good ID is low. In cases where there is some chance that the subject committed the crime, a good ID has a good chance of identifying them, relative to a misID. The crime itself is usually the most important evidence of all.

It seems to me to be quite sensible to let a denied ID drop if it does not incriminate in the crime under investigation and get on with the job in hand. Maybe some people know this instinctively. I was wondering if there is an explanation for the McKie case being the only one in the history of fingerprinting (please correct if I am wrong) where the crime only exists by inference from a denied ID. I was wondering if something normally prevented such things from escalating to a perjury accusation.

Well, I'm a long way from knowing enough of the case to say much about it. I could see how it could get there, though. If we were reporting properly, we would report something like finding a print from one of the responding officers. It most often happens when someone inadvertently handled a piece of evidence, usually because they were too lazy to glove out to properly handle print evidence, and the AFIS report comes back with a hit from the officer's peace officer licensing record prints. The real issue, if it came up at trial, would be if the officer absolutely denied having any contact, and their other work was important to the case. Their denial would call their other reporting into question. If they held to that story, and the prosecutor had confidence in the identification and was mad enough, it could get out of hand. At least around here, it might come down to the relationship between the agency or agency head and the district attorney. If they're on any kind of decent terms, most DA's wouldn't embarrass the agency, particularly a sheriff, to make a point. Both jobs are elected and therefore political, and the sheriff is often the more politically potent of the two.

Most of the time, when it comes up during the investigation, the officer is talked to and maybe becomes the object of some brief fun. Every place I've worked was poison on lying to a supervisor, and I can tell you that if my last sheriff thought a deputy was lying about, for instance, whether or not they had entered a scene, he would have had the print confirmed, and if it was verified, the deputy would become an ex-deputy.

Of course, perjury here is quite a rare prosecution. It usually has to involve a deliberate and either highly malicious or motivated by material gain provable (by some kind of hard evidence, like video or reliable paper record) lie under oath in trial that goes far beyond the routine sort of lying by defendants and their friends and family. I think most DA's would deal with a police problem by just telling the agency they would not accept any more cases by that officer, which pretty much is the end of his job, at least as an officer, and an assignment as night shift "dorm jailer," which is kind of the local Siberia at sheriff's offices where they let him keep some kind of job, rather than outright termination.

There's quite a lot of latitude in how such a thing can be dealt with in this area, since most of the small police departments' and all of the sheriffs' officers are "at will" employees who can be sent off at the will of the agency head. It's usual in some counties for the sheriff to effectively discharge all employees on the day a new term of office begins, and all who want to stay have to request to stay on. It might be different where it's very hard to fire an officer, but a criminal charge will do it.

Gerald Clough wrote:It most often happens when someone inadvertently handled a piece of evidence, usually because they were too lazy to glove out to properly handle print evidence, and the AFIS report comes back with a hit from the officer's peace officer licensing record prints.

That is interesting. The crime scene print that led to the Shirley McKie case was eliminated using an elimination card, not an AFIS hit. An AFIS error will hit on someone at random so the likelihood of an error falsely identifying a police officer investigating a crime in the same location that their fingerprint was found is tiny. But using elimination cards the fingerprint expert and the verifier both know they are checking people who are expected to have been in the location and an error must hit on someone who has a close connection, so it would appear that it would be very easy for them to have done wrong in the location and be lying about it.

My observation about the McKie case (if you have not been following it) is a simple frequency based point. If fingerprinting produces an error approximately once in every “N” identifications, this is not a problem in a normal crime-led case because the number of identifications in any one case is many orders of magnitude lower than “N”. However, all over the USA or (or anywhere else) the number of elimination IDs over a period of time will eventually exceed “N” so it is asking for trouble to wreck someone’s career or put them on trial for perjury by reacting to a denied ID (or any other forensic finding) if this is the only reason to think they are lying. We would have to be very sure that in the same area and time period there were many police officers entering crime scenes, leaving no trace of themselves and lying about it before we can reasonably confident lying was the most likely explanation for the ID.

Anyone making important decisions based on an evaluation of the evidence should be aware of the fundamental difference when the location is set by denied forensic evidence and the crime is wholly inferred from the the same evidence, compared with the normal situation where the crime is independently established as fact, and forensic identification gives evidence of who the perpetrator was.

But I can see that if the officer was identified using AFIS rather than an elimination card that might be different. Out of “N” identifications using elimination cards there is likely to be an error and that error will identify someone who has a close connection to the location. But an error using AFIS will not select someone who has a close connection to the location. (I wonder if this is reflected in current thinking about probabilities and errors using database searching compared with working with suspects).

**Edited "We would have to be very sure that in the same area and time period there were very many times more than "N" police officers entering crime scenes" to "many police officers entering crime scenes" There has to be more improper acts than errors (not more than all identifications)

I don't have much faith at all in any estimate of the frequency of error. Because there are, in my mind, two distinct types of error*, and because of variations in training, experience, and organizational policy and procedures and culture, even a good global frequency would mean little to specific cases and agencies. (Aside from the near impossibility of determining such a frequency, any more than we can ever say how many of the wrong people were convicted of crimes at trial.) I personally believe errors are more likely now, with AFIS, than previously, because we may be presented with candidates whose prints, if not necessarily similar, may be more likely to include similar sets of detail than a random selection. At least we've seen errors that would never have come up without AFIS, although AFIS can in no way be blamed for the errors. And the set of local officer prints kept for elimination purposes, are, in this sense, essentially a random selection, since nothing of the physical nature of the prints caused them to be examined, and I would not have any high expectation of them to be so similar to a latent associated with their scene for that alone to invite error.

I think the difference between an AFIS hit on someone with an unknown connection to the scene and a person with a known connection is that, if an identification was made, it may be less critically checked when the person is known to have been there. I say "may," because there should be no difference, but folks is folks. Or maybe it's more accurate to say one may be extra critical when the person resides somewhere else and has no history and you have no idea what they would be doing around here.

From what I know there are really two issues: the validity of the identification and the way the denial was handled. The latter is not really an outside issue, unless the former was in error. Those familiar with all the workings of the case have their own ideas on both issues.


* The "Oops" error where the examiner just flat missed a big fat feature that didn't match and which depends upon failure to properly verify; and the kind of error that can be any combination of personal hubris, overly imaginative extension of the capabilities of the discipline, poor practice, and organizational culture.

Q. What are the chances of a pitcher hitting a bird in flight during a baseball game?

A. Infinitesimal.

http://almax.wordpress.com/2008/06/19/b ... re-a-bird/

Rare things happen. It doesn’t prove anything unless you specify the location in advance.

Pretty much a sure thing that it's going to happen somewhere, sometime, in any activity that involves fast moving projectiles in the open air. At least one other baseball hit (seagull survived), a number of golf ball strikes, and at least one tennis ball victim. How likely it is in any one place depends largely on the frequency of targets. In the case of the baseball and seagull trying to occupy the same space, a flock of them had been a problem at the ballpark. Apparently the rest of the flock took the lesson and cleared out.

Gerald Clough wrote:Pretty much a sure thing that it's going to happen somewhere, sometime

Rare things are not unlikely if we don’t say where they will happen in advance.

Fingerprint errors are rare and we normally specify in advance the location that we are expecting the criminal to lie about having attended (the crime scene). Should a disputed identification subsequently occur, it is reasonable to be confident that error is the unlikely explanation.

But it is very risky to assume that someone is lying to conceal a previously unknown act of wrongdoing if they deny depositing a fingerprint. The location of that was not specified before the evidence was disputed.

A Swahili doctor who saw the video might suggest that a terrorist organisation in the location of the ball park are training a flock of kamikaze gulls. I’d say that he is almost certainly right – but only if someone had suggested such a thing before the game.

Outsider Says

A Swahili doctor who saw the video might suggest that a terrorist organisation in the location of the ball park are training a flock of kamikaze gulls. I’d say that he is almost certainly right – but only if someone had suggested such a thing before the game.

So, if I suggest right now that the next bird that gets hit with a ball in a ball park (before the game)it is because an alien space craft put a force beam on it and made it travel in the path of the ball, then you would respond with "you are almost certainly right."

Interesting concept. A person can suggest any type of cause before the effect and because it is said before the effect then the cause is "almost certainly right".

Will you be writing more on this suggested (any type) cause before the observed effect takes place?

It seems that OJ Simpson is in trouble with the law again. That reminded me of a statistical matter that emerged from OJ’s murder trial, but only after the trial ended. I have found an article that describes it:

On another case article also gives a good description of the prosecutor’s fallacy:

During the trial, Adams's lawyers warned the jury that the lack of supporting evidence greatly diluted the strength of the DNA evidence. They pointed out that the 1 in 200 million was not the chance of Adams being innocent, but was instead an estimate of how much one's initial belief in his guilt should be amplified by the forensic evidence. And if the chances of guilt suggested by the other evidence were very low, then even multiplying them by a factor of many millions could still leave grounds for reasonable doubt.

I had no idea that the explaining probability is not allowed in UK courts.

the Appeal Court has now confirmed its original ruling. The laws of probability, it has declared, are "a recipe for confusion, misunderstanding and misjudgement". Instead, juries should rely on their "individual common sense and knowledge of the world".

That was 1997, I don’t know if things have changed since then.

Article here:
http://technology.newscientist.com/chan ... 621123.100