John Richardson

[Fleetwood Mac]

There has been no invention on my part. All that I’ve stated and re-stated is what the experts on the subject tell us. A subject that, as far as I can see, none of us are qualified to gainsay. I don’t understand the resistance that I see (not just from you) on this?

The discussion on Phillips in relation to this thread is as to how it impacted on our assessment of the witnesses. If we could say for example that medical knowledge tells us that a doctor estimating a time of death in 1888 was overwhelmingly likely to have been correct then the balance would be tipped in his favour but we can’t say that. We can’t even get close to saying that. Even if we could give a 75% certainty level of accuracy it still would eliminate the witnesses but we just can’t do that.

So I’m not saying that Phillips TOD estimate was the work of an incompetent or that it was a pointless exercise in itself because it wasn’t. It was the work of a man doing the best with the knowledge available at the time. But what I don’t understand is why my suggestion that his TOD is of little or no use to us has caused so much irritation? If we can’t assess it accurately and decide how accurate or inaccurate it was, then basically it’s little better than a guess in this case (and I’m not saying that he was actually guessing of course) In effect, it’s neutral.

'Trying to help you here, Sherlock. Jeff has posted an article that he claims demonstrates Phillips' TOD is unreliable. I've posted an objection. Both are based on empirical evidence. Who knows how it will develop. Do your own research/digging around the internet and post it, you may come up with something nobody has considered and you'll get a lot of thanks.

[DJA]

Summary of the 81 pages so far.

[Herlock Sholmes]

'Trying to help you here, Sherlock. Jeff has posted an article that he claims demonstrates Phillips' TOD is unreliable. I've posted an objection. Both are based on empirical evidence. Who knows how it will develop. Do your own research/digging around the internet and post it, you may come up with something nobody has considered and you'll get a lot of thanks.

I don’t want thanks and my username isn’t Sherlock unless you can’t read. Your objection isn’t relevant unless you can provide us with your credentials on Forensic Medicine. No further research is required. Phillips TOD cannot be relied upon apart from those that are simply too biased against the witnesses to accept it.

[Herlock Sholmes]

Here’s a novel suggestion. Wouldn’t it be a ripping idea for those critical of the content or subject matter of these threads to begin threads of their own, on subjects of their own choosing rather than sniping in from the sidelines?

[Herlock Sholmes]

There’s really no need for further comment by me on this subject. It’s gone on too long.

[JeffHamm]

Jeff's article,and I had seen it previously,is concerned mainly with deaths that had occured from a 5 - 50 hour period,and has a plus or minus error of 38 minutes.
No where does the article relate to Phillips 2 hour base.So where Herlock,is my contradiction? How does a 5-50 hour period prove Phillips to have been incorrect?

Hi harry,

The +-38 minute range of accuracy only really applies if you know things that won't be available in most murder enquiries, like the person's body temperature at the time of their death. Normal temperature varies between people, so if you don't know the starting temperature you can't know how much it has changed exactly, and you have to treat starting temperature as an unknown and deal with a range of values. Obviously neither Dr. P. nor we know what Annie's body temperature was when she was killed, so we have treat it as an "unknown parameter". The error range in this situation increases to +-2.5 hours, the +-38 really refers to the accuracy of the equations (model) when all information is available, showing it's a good model. The pragmatic use of the resulting model, however, will not have all values and some need to be treated as "unknown values", which means considering the distribution of the values from the population (i.e. the distribution that people' body temperatures have, which is something like 96-99 F according to a quick google search, the source of all truth and knowledge ).

So, in our situation, even if we could apply their model to Annie's murder, we would be looking at +-2.5 hours. The current methods have a range of +-3 hours (minimum) out to +-7, according to their article at least, which either reflects the ranges reported in other studies or there are important factors that need to be considered as they influence the reliability.

Regardless, Dr. P has only "unknown parameters" as he took no actual measurements.

And yes, their paper only looks at intervals of 5-50 hours. Since the range of "normal" temperature is 96-99, and the "rule of thumb" is 1 hour per degree of F loss, if you don't know the person's starting body temperature, and just go with 97.5, you're already dealing with +-1.5 hours on that one factor alone). And if you don't measure the internal body temperature, but feel the surface and guesstimate it, that error just increases, and not by just a little bit either.

- Jeff

[JeffHamm]

Hi Jeff,

Yes, I can see the link.

Thanks for taking the time to find and post this by the way, because it does add something to the discussion.

What I have been able to understand is the following:

1) This research was published in the Science Advances Journal, which I believe is an American publication concerned with research in any area of science.

2) It is a peer reviewed publication requiring at least two reviews.

One caution here is that this is a multidisciplinary journal as opposed to one specifically associated with anatomy or the field of estimating time of death. When they say 'peer reviewed' it is not clear by whom, i.e. is it someone well versed in the ins and outs of estimating time of death, given it is a multidisciplinary journal it is not necessarily the case. What it does state on the Science Advances Journal website, is that bringing in expertise in the field that is being reviewed is at the editors discretion. Given it's your link/post/article, I'm hoping you can clarify whether or not this has been peer reviewed by someone with expertise in the field of estimating time of death.

As I don't know who the reviewers are I can't speak to their qualifications specifically. What I do know is that reviewers are chosen on the basis of them having expert knowledge pertinent to the submitted article.

3) The pertinent part of the article in your link is this: existing methods can only give a window of several hours. This is what we're discussing in this thread. Based on your link/article, it appears that the people have who have written the article are making that statement. It is not a direct quote from the researchers (whereas they do quote the researchers in other parts of the article).

In the full article they cite sources and so forth. The summary version is just that, a summary of their article that the University makes available as they are usually easier for the general public to understand the main finding because they won't necessarily be experts in a field.

4) Following on from the above point, it is not clear what exactly was published in the Science Advances Journal. Did they simply publish the research or did the Science Advances Journal claim that existing methods can only give a window of several hours also. In the event it's the former, then we do not have a peer reviewed assessment of the claim existing methods can only give a window of several hour. Given it's your link/article to support an argument, I'm hoping you can tell me who exactly is claiming existing methods can only give a window of several hour (as far as I can tell it's the person who has written the article and I don't know who that is) and whether or not the Science Advances Journal repeated this claim when they published the research. Clearly it is an important point because it tells us whether or not that claim was peer reviewed, although it won't tell us whether or not peer reviewed by people with expertise in the field of estimating time of death (because according to the Science Advances Journal, drafting such expertise is at the editor's discretion).

I posted a link to the full article as well, so I'm sure you'll find your answers in that.

What may be of interest, however, is that Dr Phillips agreed with the person making the claim existing methods can only give a window of several hours when assessing Mary's body, between 2am and 8am, but not when assessing Annie's body.

The article you have posted is useful because it begins to lead the discussion where it should go, i.e. looking at the science behind estimating TOD. Having said that, it in no way constitutes 'proof' that existing methods can only give a window of several hours, see the reasons I have mentioned in this post.

As the article cites the error range of the "gold standard" current method, you are free to follow back through the literature to assess that claim. However, the error ranges they present survived the peer review process, and generally just making up stuff like that gets caught and weeded out.

Feel free to add some meat to the bones in terms of the questions I've asked in this post.

As I said though Jeff, thanks for posting this. It's definitely piqued my curiosity and I'm going to start having a look at this to try and determine just how unreliable estimating TOD is and how that is impacted by the time interval from death to assessing the body. When I say determine, quite frankly I don't have the scientific apparatus sat around in my loft to be able to do this myself, but there must be studies out there and they may offer a consensus among experts in that field. I do quite a bit of photography in my spare time and it's the first day we've had in ages where there is sun and little wind combined and so I'll be out all day and not looking for studies. At some point over the weekend I'm going to start looking and it certainly deserves its own thread.

Doing a quick google search, not entirely rigorous science I know, but I've come across some sites that indicate the 95% range of normal human body temperature is between 96 and 99 F. ToD is roughly calculated as 1F loss of temperature per hour. If you don't know what the person's normal temperature is, though, how can you calculate loss? You have to use a guess (say 97.5) which means you've already got a +-1.5 hour error range. And if you don't actually measure the current temperature of the body, but touch the surface and estimate it, then you have another source of error, increasing that +-1.5 to something larger. And given our ability to estimate temperature is, I'm sure you will agree, pretty inaccurate, then if one's estimate of the current temperature can be out by +-1 degree (just to pretend we're that good at it), we're now dealing with +-2.5 hours (and we've not even considered the influence of the environmental factors, like temperature, which would change that rule of thumb rate, etc).

But, if you can find an article on short duration intervals, that would be great. Particularly if you can find one where ToD is estimated by touch.

- Jeff

[DJA]

I again refer to Dr Biggs who says

, if the victim is a malnourished, slight, alcoholic female then rigor mortis may be less pronounced than might be expected, and so detection of rigor mortis in such an individual may indicate a longer time has elapsed since death

www.trevormarriott.co.uk

In fact,a malnourished person would have a lower level of ATP, therefore rigor mortis would be evident earlier.

[Trevor Marriott]

I’ve said no such thing Trevor. Why make things up?

Ive never said that estimating TOD’s is frowned up or even hinted at it. What I’ve said is that we cannot know if Phillips was right in his estimate or wrong in his estimate. We know that the methods used by him were unreliable because modern experts tell us this.

So if we cant know if he was right or wrong and we can’t even suggest a likeliness of him being right or wrong then what use is the information that we have from him? In terms of deciding whether the witnesses should or shouldn’t be overruled in favour of him…..absolutely none. This is why I say that Phillips is pretty much neutral. We have to focus on the witnesses (accepting of course that the witnesses could have been wrong)

What about the balance of probabilty then if you cant decide? because one minute you are stamping your feet saying Phillips was wrong, now you want to sit on the fence.

and you ignore that fact thats the witnesses were in conflict with each other, so how can you put them in front of Dr Phillips? and the main one you place on your pedestal is Richardson, which version of his testimony as reported in the various newspapers is the correct version?

https://www.trevormarriott.co.uk

[JeffHamm]

Interestingly, what I'm finding is that internal measures of temperature will initially rise by 0.5C (about 1F), and don't tend to start falling for a while(an hour or so), but surface temperature starts falling more or less within minutes following an exponential decrease (as does internal eventually), requiring some complicated calculations involving environmental temperature, surface area exposed, and so forth. The implications, of course, is that internal temperature will be uninformative for the first hour or so (because it doesn't change), and surface temperature requires a fair number of values to be measured and entered, etc, and even then the estimates are imprecise. (But I want to read some more on this before asserting that too strongly - this is my impression from a few abstracts I've seen, but I've yet to grab the articles to delve into it and make sure the above is correct).

Basically, though, from the literature on this it boils down to the fact that estimating the ToD based upon temperature is possible (it's not a random guess) but it is also associated with very large error ranges. So even if Dr. P. had taken all the measurements required, it's just not possible to distringuish between his estimation and a ToD based upon the witnesses. His statement and the witnesses, therefore, are in agreement because his assessment must be considered in conjunction with the error ranges associated with the method (and his touch based method cannot be more accurate than a measured based one, and given the lower quality data touch based assessment gives, the error will be larger, that's not an opinion, it's how measurement error works).

Again, there's nothing inconsistent between his stated opinion and the witnesses. His statement, coupled with it's error range, would be considered "correct" if she was killed around 5:25ish - he's within the margin of error. So really, there's no conflict in the testimony to resolve here.

- Jeff

[DJA]

Interestingly, what I'm finding is that internal measures of temperature will initially rise
- Jeff

Yep.
It's one of those threads

[Herlock Sholmes]

What about the balance of probabilty then if you cant decide? because one minute you are stamping your feet saying Phillips was wrong, now you want to sit on the fence.

and you ignore that fact thats the witnesses were in conflict with each other, so how can you put them in front of Dr Phillips? and the main one you place on your pedestal is Richardson, which version of his testimony as reported in the various newspapers is the correct version?

www.trevormarriott.co.uk

What’s the point of responding Trevor when you just keep making things up? You just can’t be reading posts properly, I can’t think of another explanation. I’ve never said that I’m certain that Phillips was wrong (and I certainly haven’t stamped my feet) Yes I favour the witnesses over the Doctor in this case but not because I can state that he must have been wrong. Please try and understand this Trevor.

Im saying that there is significant doubt about the level of accuracy (the reliability) of his estimation. I haven’t criticised him as a doctor or questioned his competence. But we have no way of assessing his estimate because we cannot recreate those conditions with a modern day specialist in place to make an modern day estimate. And so, in effect, Phillips is a might have been, might not have b

I haven’t ‘ignored’ the fact that the witnesses conflict with each other. Firstly, Richardson and Cadosch don’t conflict, and Long conflicts by a small matter of 15 minutes. Unlike you, I don’t dismiss witnesses just because they aren’t perfect. What I do, and what I’d hope that we should all do, is analyse what we have, assess and weigh up the pro’s and con’s and suggest an interpretation/assessment. I might also say that’s it’s a little strange that all of the witnesses regarding TOD (only 3 I know) were all mistaken or liars.

If you accuse me of simply accepting Richardson at face value (which I clearly don’t) why shouldn’t I accuse you of simply accepting Phillips at face value? Why can’t you ever accept that other people are capable of making assessments too? Yes we are all capable of being wrong but why can’t you at least accept that just because someone disagrees with your viewpoint it’s not because they’ve ignored evidence, or that they’ve taken something at face value, or that they are somehow rigidly attached to a particular version of events. It’s a tactic of yours that a bit like the one employed by someone like HR who rather than debate fully seek to dismiss by labelling. Someone disagrees with Trevor and it just can’t be because they’ve assessed the evidence and come to a different conclusion can it? Because Trevor can’t be wrong can he?

[DJA]

Given that Dr Phillips described the knife that was used on Chapman would not be found in the ORDINARY post mortem case,leads me to believe
he knew the killer,one of his peers who many years earlier passed by his house on the way to and from work at the London Hospital.

Same person might have links to the Black Swan at 23 Hanbury Street.
Crkey,where's the other DJA when I need her

Personally,Chapman's TB of her brain would most likely be the reason for her low body temperature.

[Trevor Marriott]

What’s the point of responding Trevor when you just keep making things up? You just can’t be reading posts properly, I can’t think of another explanation. I’ve never said that I’m certain that Phillips was wrong (and I certainly haven’t stamped my feet) Yes I favour the witnesses over the Doctor in this case but not because I can state that he must have been wrong. Please try and understand this Trevor.

Im saying that there is significant doubt about the level of accuracy (the reliability) of his estimation. I haven’t criticised him as a doctor or questioned his competence. But we have no way of assessing his estimate because we cannot recreate those conditions with a modern day specialist in place to make an modern day estimate. And so, in effect, Phillips is a might have been, might not have b

I haven’t ‘ignored’ the fact that the witnesses conflict with each other. Firstly, Richardson and Cadosch don’t conflict, and Long conflicts by a small matter of 15 minutes. Unlike you, I don’t dismiss witnesses just because they aren’t perfect. What I do, and what I’d hope that we should all do, is analyse what we have, assess and weigh up the pro’s and con’s and suggest an interpretation/assessment. I might also say that’s it’s a little strange that all of the witnesses regarding TOD (only 3 I know) were all mistaken or liars.

If you accuse me of simply accepting Richardson at face value (which I clearly don’t) why shouldn’t I accuse you of simply accepting Phillips at face value? Why can’t you ever accept that other people are capable of making assessments too? Yes we are all capable of being wrong but why can’t you at least accept that just because someone disagrees with your viewpoint it’s not because they’ve ignored evidence, or that they’ve taken something at face value, or that they are somehow rigidly attached to a particular version of events. It’s a tactic of yours that a bit like the one employed by someone like HR who rather than debate fully seek to dismiss by labelling. Someone disagrees with Trevor and it just can’t be because they’ve assessed the evidence and come to a different conclusion can it? Because Trevor can’t be wrong can he?

But you are never wrong are you? Mr Perfect

https://www.trevormarriott.co.uk

[DJA]

TB or not TB,that is the question