Are Victim Detection and Forensic Evidence Search Dogs reliable?

[pegasus]

" scent samples consist of sterile gauze pads .... placed on abdominal area ..............
NUMBER OF DOGS USED: Five different dogs
POST-MORTEM INTERVAL RANGE: From 70 minutes to 3 days
NUMBER OF TRIALS COMPLETED: As of July 1997, total of 52 trials completed
PRELIMINARY RESULTS: The shortest post-mortem interval for which we received a correct response was one hour and 25 minutes. However, the post-mortem interval for which we received a consistently correct response from all dogs involved is 2.5 - 3 hours."
http://www.csst.org/cadaver_scent.html

EVRD in bedroom was not testing a gauze pad which had conveniently been in direct contact with anything. He had the presumably more difficult task of working here, according to the handler, with nothing but residual scent. In view of this, is not the minimum of  "2.5 - 3 hours" (which in experiment IMO 5 out of 5 dogs detected), more relevant than the "one hour and 25 minutes" period (which in the experiment IMO 4 out of 5 dogs failed to detect even on a gauze pad) ?

[gilet]

Evidential Value...

Value of records given as or in support of evidence, but not the evidence itself.

Indeed so the minutes of a meeting where a financial document was agreed could be used as evidence in a criminal court discussing the financial document. They would be considered to have the requisite evidential value to be used as supporting evidence. The original financial document being discussed would automatically have sufficient evidential value to be used in that court case.

So in exactly the same way, the dog alerts cannot be used as evidence until they are shown to have sufficient evidential value or reliability to be acceptable as evidence.

Anything which has no evidential value cannot be used as evidence.

As I am currently working in the field of the evidential reliability of certain types of computer generated material I am perfectly able to recognise the difference between the terms "evidential value/reliablity" and "evidence". The level of credence which can be given to the latter is dependent on the former.

If an alert or a document or a computer record has no evidential value/weight/reliability then it cannot be used as evidence. If it has evidential value then that material or alert or record can be used as evidence. In some cases certain material is introduced into a court case but its lack of evidential value must be explained where it is clear that there is none. It is not then possible to call that material by the term evidence or to call it an exhibit in a case. Where the evidential value of material is shown to exist (even if it is flimsy) then that material can be used as evidence or as an exhibit within a case. In certain instances it is introduced but with the proviso that the jury is made aware of the level of evidential reliability which it has.

Grime was explicit that the alerts without forensic corroboration had no evidential value at all and therefore could not be used as evidence. They are merely indications which relate to the matter.

This US case explains the reasoning.

http://www.internationallawoffice.com/newsletters/detail.aspx?g=df74cd39-728b-4d35-9720-b9502c3ca95c

[icabodcrane]

It is entirely possible that the cadaver dog alerted in apartment 5A  because Madeleine McCann  died there,  and he picked up the scent

[gilet]

It is entirely possible that the cadaver dog alerted in apartment 5A  because Madeleine McCann  died there,  and he picked up the scent

Of course it is.

It is equally possible that the cadaver dog's alert is not entirely reliable.

It is equally possible that if the alert is reliable then it had no connection whatsoever to the missing child.

All are equally possible and those who claim otherwise are simply mistaken.

[gilet]

Getting back to the direct terms of reference of this thread.

This is the uniform that Eddie and Keela would have been used to when working with Grime. This is the uniform which would have put them in "working mode". (please see my post above, #537).



Grime was not wearing any uniform in PDL.  Is it not possible that the lack of uniform would have affected the working reliability of the dogs, another handler having indicated this possibility?

On a second issue related to the reliability of the alerts, is the claim by Martin Grime that the alert to cadaver odour by Eddie may not even have originated in the bedroom because of the way scent pools in a closed apartment over time not another serious cause for concern?  Is it even possible that if, as Martin Grime tells us, the alert was possibly from another room then it was a secondary alert to the blood detected in an adjacent room which was shown to have no connection whatsoever to the McCanns.

[icabodcrane]

Of course it is.

It is equally possible that the cadaver dog's alert is not entirely reliable.

It is equally possible that if the alert is reliable then it had no connection whatsoever to the missing child.

All are equally possible and those who claim otherwise are simply mistaken.

Thankyou for acknowledging that it is entirely possible that the cadaver dog  barked in apartment 5A because Madeleine had died there and he picked up the scent

I think it is important  that the most obvious reason for a cadaver  dog alerting in an apartment where a missing child was last seen alive  is not  lost amongst the less logical propositions

[gilet]

Thankyou for acknowledging that it is entirely possible that the cadaver dog  barked in apartment 5A because Madeleine had died there and he picked up the scent

I think it is important  that the most obvious reason for a cadaver  dog alerting in an apartment where a missing child was last seen alive  is not  lost amongst the less logical propositions

You will find that from my earliest posting on this forum I have always acknowledged that fact.

However I disagree with the weight you place on this option with respect to other options. With nothing to support the alert and with serious questions about the reliability in general of the dog alerts then other options as I listed above, have, in my view, equal weight.

[icabodcrane]

You will find that from my earliest posting on this forum I have always acknowledged that fact.

However I disagree with the weight you place on this option with respect to other options. With nothing to support the alert and with serious questions about the reliability in general of the dog alerts then other options as I listed above, have, in my view, equal weight.

But gilet,  what  supports  the  'option'  that the cadaver dog was alerting to the dead body of the missing child is right there in the phrase,   'missing child' 

She is  missing  ...  disappeared without  trace  ...  not seen alive in the seven years since a  cadaver dog barked in that  apartment

THAT  is what supports the  'option'   that it was the scent of death that the cadaver dog alerted to

[pathfinder73]

" scent samples consist of sterile gauze pads .... placed on abdominal area ..............
NUMBER OF DOGS USED: Five different dogs
POST-MORTEM INTERVAL RANGE: From 70 minutes to 3 days
NUMBER OF TRIALS COMPLETED: As of July 1997, total of 52 trials completed
PRELIMINARY RESULTS: The shortest post-mortem interval for which we received a correct response was one hour and 25 minutes. However, the post-mortem interval for which we received a consistently correct response from all dogs involved is 2.5 - 3 hours."
http://www.csst.org/cadaver_scent.html

EVRD in bedroom was not testing a gauze pad which had conveniently been in direct contact with anything. He had the presumably more difficult task of working here, according to the handler, with nothing but residual scent. In view of this, is not the minimum of  "2.5 - 3 hours" (which in experiment IMO 5 out of 5 dogs detected), more relevant than the "one hour and 25 minutes" period (which in the experiment IMO 4 out of 5 dogs failed to detect even on a gauze pad) ?

Cadaver was in 5A for at least 2 hours IMO. The routine change holds the answer.

[gilet]

But gilet,  what  supports  the  'option'  that the cadaver dog was alerting to the dead body of the missing child is right there in the phrase,   'missing child' 

She is  missing  ...  disappeared without  trace  ...  not seen alive in the seven years since a  cadaver dog barked in that  apartment

THAT  is what supports the  'option'   that it was the scent of death that the cadaver dog alerted to

It is one option. That is all.

The other options are equally valid.

You claim the dog alerts support your option.
Equally the lack of any evidence denies your option.

You prefer to place reliance on the dog alerts. I have very serious questions about their reliability in this case in particular.

[gilet]

Cadaver was in 5A for at least 2 hours IMO. The routine change holds the answer.

You are perfectly entitled to hold your opinion. I don't see things as starkly as you appear to.

When we are offered some evidence that the dog indeed alerted to a cadaver then I may look again at your claims.

But in the meantime can you please tell me why the alert in the apartment (if it indeed related to cadaver scent) could not simply have been due to cadaver scent brought into the property on some contaminated object/clothing etc which could have lain there for much longer than a few hours?

[gilet]

To get to the nub of this thread and before anyone challenges me to do so, I will set out the reasons why I am not convinced of the reliability of the dog alerts in this particular case.

If I was asked to challenge the dog work in PDL these are some of the considerations I would reflect on and bring to the fore.

1.   There is absolutely nothing but the dog alerts. Not a single piece of evidence supports those alerts. Not a speck of Madeleine’s blood, not a single indication in any form.

2.   There was (as far as I am aware from the files) not the slightest attempt by the PJ to look at any of the other possible explanations for the dog alerts. We know that the Amaral team immediately took them as meaning that there was evidence of the death of Madeleine, even suggesting that the dogs had found things which we later discovered they never found. Is it not perfectly possible that (allowing the dog alerts to stand) that the cadaver scent in the apartment could have been introduced by some other means than an actual cadaver?

3.   The dogs were worked for very long periods of time (into the early hours of morning) in very concentrated bursts. I would wish to ask other dog handlers as well as Grime what the effect of such long hours on a dog’s reliability might be?

4.   I would be concerned that the dogs were not sensing a “working mode” because of the way in which, having been used to working with a handler in uniform, they were now working with him in civilian clothes. At least one handler I have cited says this does have an effect. I would wish to ask other handlers and Grime what their thoughts on this issue were.

5.   I would be concerned specifically about the late night alert in the garden area of Apartment 5A. We know that Grime said that this was not a strong alert and that further investigation was necessary. However I am not aware of the results of this further investigation or even whether anyone bothered to do it. I would wish to learn of those results.

6.   With regard to the car. I would ask others more expert than myself what the effect of the mishandling (in UK terms) of the car by the PJ might be and why it was simply driven to the test by a PJ officer in civvies.

7.   With regard to the clothing. I would challenge very strongly the contention that it can be said that three separate alerts occurred when all three items alerted to came from one box of unprotected items where cross-contamination would almost inevitably have occurred.

8.   I would seek very careful elaboration as to why the clothing was not alerted to in situ in the villa or if the clothing had already been removed why the residual scent was not alerted to.

9.   I would wish to know why the cuddle cat was not alerted to (rather it was played with by Eddie) before it was hidden in the cupboard. I would want to ask for the opinions of other experienced handlers as to the meaning of this anomaly.

10.   I would wish to know why the alert to the toy was so non-specific. It seemed that the alert was more generally to the kitchen area than the toy. I would want to ask for the opinions of other experienced handlers as to this clear anomaly.

11.   I would reflect on the fact that the videos which are available came not from the PJ directly but through the hands of a person whose credentials are unknown and that the videos do appear to have been edited.

12.   I would wish to ask Grime as to why he was so clear that the dogs alerts are not evidential and whether (as we are told by Serendipity) he was forced to add this to the report.

Those are just a few of the problems I see regarding the reliability of the dog alerts in PDL. Till they are all satisfactorily explained, I will continue to see the alerts as simply indications which may or may not have a real bearing on the case.

[icabodcrane]

It is one option. That is all.

The other options are equally valid.

You claim the dog alerts support your option.
Equally the lack of any evidence denies your option.

You prefer to place reliance on the dog alerts. I have very serious questions about their reliability in this case in particular.

No 

I say that it is the fact that the child has never been seen alive again that supports the 'option'   (  that the cadaver dog barked  in apartment 5a because he picked up the scent of her dead body ) 

Had the child been found alive,  or proof of her having left the apartment alive been found,   then   'other options'  might be seen as valid

She has not

The most  obvious explanation for the cadaver dog's alert remains,  therefore,  that he picked up the scent of the dead body of the missing child

[gilet]

No 

I say that it is the fact that the child has never been seen alive again that supports the 'option'   (  that the cadaver dog barked  in apartment 5a because he picked up the scent of her dead body ) 

Had the child been found alive,  or proof of her having left the apartment alive been found,   then   'other options'  might be seen as valid

She has not

The most  obvious explanation for the cadaver dog's alert remains,  therefore,  that he picked up the scent of the dead body of the missing child

You are perfectly entitled to your opinion. It is, however, based on dog alerts which may or may not have the slightest bearing on the disappearance.

Till you provide me with one tiny fragment of evidence that the alerts relate to Madeleine or even that the alerts are actually reliable then I believe that I am also perfectly entitled to my own opinion which clearly differs from yours.

Your entire reasoning is based on alerts which have never been shown to relate to the missing child or even to cadaver scent at all. Even Grime goes no further than saying that the alerts are "suggestive".

It appears that those "suggestive" alerts are enough to sway you.

I prefer to remain more open-minded and await actual evidence.

Just to add. Your third paragraph seems to suggest that no other option but the one you support is even valid. That is simply untrue.

[pathfinder73]

You are perfectly entitled to hold your opinion. I don't see things as starkly as you appear to.

When we are offered some evidence that the dog indeed alerted to a cadaver then I may look again at your claims.

But in the meantime can you please tell me why the alert in the apartment (if it indeed related to cadaver scent) could not simply have been due to cadaver scent brought into the property on some contaminated object/clothing etc which could have lain there for much longer than a few hours?

IMO the cadaver was hidden inside the wardrobe on the shelf where Eddie sniffed and then barked very loud and continually. He will bark where he finds the strongest scent source. As soon as he got to 5A Grime recognised his change in behaviour due to cadaver scent.