15 year old ghost mystery possibly solved

[Soapy Sam]

Stray Cat- great animation, funny thing is it had the opposite effect on me, I thought they were similar but not the same image until I saw your comparison (and Orthoptera's too), now I'm 100% convinced it is the same.

I agree. The original shots did not impress me as very similar, but both Orthoptera's images and Stray Cat's blink comparator converted me. I agree it's the same image.

 

[Careyp74]

Hi Steve, my point is that I shouldn't have to raise doubt. As sceptics, the doubt should be there to start with. The evidence only points to a similarity, not a conclusive certainty.

Given what I know about shadows and contrast and old photos, I have enough evidence to say that I feel they are the same. That is what everyone else is doing, they are not flipping a coin here, they are making decisions.

Otherwise I could show two photos of remarkably similar looking people and claim they were the same person.

OK so do it then, and prove us wrong.

But knowing that there have been millions upon millions of photos taken since the invention of photography and some of those photos will look remarkably similar to each other, I can not be certain that the girl in the postcard has been used to fake the photo of the ghost in the fire.

It isn't about being remarkably similar. They are the same, you just won't admit it. Show us the differences, your skewed lines aren't doing it.

 

[cj.23]

Um, for some years ASSAP have been explaining this as an example of paredoilia, a falling beam causing the image. It now looks like simple negative fraud. IT seems one sceptical explanation was in fact completely bogus -- and the real explanation simpler?

cj x

 

[Stray Cat]

Show us the differences, your skewed lines aren't doing it.

Tell you what, you show me the EXACT matches.
You are the one making a claim here... burden of proof and all that.

Oh, hang on, you can't, because although they look similar, they are not an exact match and your vague red circles aren't doing it.

 

[fls]

[qimg]http://i246.photobucket.com/albums/gg117/ThePsychoClown/GF1.jpg[/qimg][qimg]http://i246.photobucket.com/albums/gg117/ThePsychoClown/G1b.jpg[/qimg]
[qimg]http://i246.photobucket.com/albums/gg117/ThePsychoClown/GFamend.jpg[/qimg][qimg]http://i246.photobucket.com/albums/gg117/ThePsychoClown/G2b.jpg[/qimg]
[qimg]http://i246.photobucket.com/albums/gg117/ThePsychoClown/G3.jpg[/qimg]

Remarkably similar, yes.
Exactly the same, no.

This reminds me of putting markers on echocardiograms for measurement. Are those lines something you placed based on your visual assessment?

Linda

 

[Soapy Sam]

Hi Steve, my point is that I shouldn't have to raise doubt. As sceptics, the doubt should be there to start with. The evidence only points to a similarity, not a conclusive certainty.

Otherwise I could show two photos of remarkably similar looking people and claim they were the same person.

Built upon a preponderence of evidence (and complete lack of contrary evidence) collected over years of research I can fairly confidently say that this photo is a fake. But knowing that there have been millions upon millions of photos taken since the invention of photography and some of those photos will look remarkably similar to each other, I can not be certain that the girl in the postcard has been used to fake the photo of the ghost in the fire.

Stray Cat- You are of course correct that the two images are not the same. If they were the same , there would only be one image.

It's significant (IMO) in this case that at least three people who initially did not agree that the pictures showed the same person have changed their minds based (in part) on your comparison.
If you handed me two printout digital photos taken 1/10th second apart of the same scene , I might probably say they were the same , which would be wrong. But in this specific case,where one photo is of a girl in a wholly unremarkable context, while the other shows a girl looking astonishingly like the first girl, but in a situation utterly incredible, it seems that the assumption "these are the same image" becomes overwhelmingly likely.

This is not just a photo comparison; it's a comparison of an unremarkable photo with an impossible event. I think once we take that into account, our "certainty" becomes more acceptable.

 

[Stray Cat]

This reminds me of putting markers on echocardiograms for measurement. Are those lines something you placed based on your visual assessment?

Linda

Some them yes, and on some of them I did actually measure pixel brightness... But I was smoking at the time.

 

[Stray Cat]

But in this specific case,where one photo is of a girl in a wholly unremarkable context, while the other shows a girl looking astonishingly like the first girl, but in a situation utterly incredible, it seems that the assumption "these are the same image" becomes overwhelmingly likely.

There are situations not a million miles away from this example where similarities would be claimed a mere coincidence with the same level of certainty. But I agree with the words assumption and likely.

This is not just a photo comparison; it's a comparison of an unremarkable photo with an impossible event. I think once we take that into account, our "certainty" becomes more acceptable.

So your judgement is based upon the impossibility of the event... Well so is mine to reach the conclusion that the photo is faked. But regardless of the event, the technical detail of a how a photo is faked doesn't rest upon the need to believe in or be sceptical of ghosts.

 

[Skeptical Greg]

Some them yes, and on some of them I did actually measure pixel brightness... But I was smoking at the time.

How can pixel brightness, of different copies of an image, be meaningfully used for matching purposes ?

 

[fls]

Some them yes, and on some of them I did actually measure pixel brightness... But I was smoking at the time.

Why would you think that it is more likely that the photos are slightly off than your accuracy is slightly off?

Edit: For example, the line you placed under her nose looks to be right at the line of contrast between black and light in one photo, but you have placed it below that contrast point in the second photo.

Linda

 

[Geezer]

How can pixel brightness, of different copies of an image, be meaningfully used for matching purposes ?

As I see it only as a general indicator with very low accuracy (i.e. oh that pixel there is brighter than the surrounding pixels and so are/are not the pixel at the same coordinates on that other copy)

 

[Careyp74]

Um, for some years ASSAP have been explaining this as an example of paredoilia, a falling beam causing the image. It now looks like simple negative fraud. IT seems one sceptical explanation was in fact completely bogus -- and the real explanation simpler?

cj x

I had never seen this photo before now, but I couldn't possibly imagine blaming something else for it, the pic is definitely showing too much detail for it not to be an actual person.

Do you have a link to this claim?

And yes, it is simpler. That was my idea.

 

[richardm]

I had never seen this photo before now, but I couldn't possibly imagine blaming something else for it, the pic is definitely showing too much detail for it not to be an actual person.

Do you have a link to this claim?

Can't find a link, but it has been claimed. The fire brigade have a video of the fire, and a beam was filmed falling in about the right spot. It was guessed that it was the coruscating end of the burning beam, since for some reason the solution "The photographer is lying" had been discounted. It really does look like an actual person, but pareidolia can be like that. However it seems evident that in this case it was a deliberate fake. Really interesting to see it laid to rest so convincingly (convincing unless you're Stray Cat, apparently )

 

[Stray Cat]

How can pixel brightness, of different copies of an image, be meaningfully used for matching purposes ?

It can't... which is why that's not what I did.

I took the photos as being independent of each other and measured pixel brightness to follow lines and highlights within each photo.

 

[Lucky]

Um, for some years ASSAP have been explaining this as an example of paredoilia, a falling beam causing the image. It now looks like simple negative fraud. IT seems one sceptical explanation was in fact completely bogus -- and the real explanation simpler?

cj x

I had never seen this photo before now, but I couldn't possibly imagine blaming something else for it, the pic is definitely showing too much detail for it not to be an actual person.

Do you have a link to this claim?

And yes, it is simpler. That was my idea.

I noticed from the article that 'photographic experts' explained it as 'a trick of the light caused by the fire', but this explanation is so far-fetched that I assumed the reporter had invented these 'experts'. Such a convincing 'ghost' is orders of magnitude more likely to be a fake than an accident.

Does anyone know whether the photographer was able to produce the negative, in which case it would have to be a double exposure (or, I suppose, he could have combined the images in some other way and then photographed the result).

 

[Stray Cat]

Why would you think that it is more likely that the photos are slightly off than your accuracy is slightly off?

I'm sure my accuracy is off. I neither have the time nor inclination to spend hours on every little detail proving something where the burden of proof is not really mine to bear. The illustration is a guide to the differences in the two pictures, that's all.

 

[JcR]

Candle girl 2. Since she burned down all of Wem.

ETA: Toned down fire ghost

 

[P.J. Denyer]

Stray Cat-If it isn't the same image then the question becomes 'why are they so similar?'


Option 1 It was a different shot taken within moments of the postcard shot, but given the age of the original photo I would say that the chances of a camera of that period taking multiple pictures quickly enough to be so similar and then to have two different but almost indistingushable shots survive for this long is pretty remote.

Option 2 would be a a coincidental similarity of face, clothing, pose and expression, given that even the folds of the clothing match this seems quite far fetched.

Option 3 is that the photographer deliberately copied the older photograph with an remarkably similar child model and managed to replicate every fold of the clothing perfectly.

Option 4 is that it's a real ghost in both shots.

Frankly I think that the chances of any of these four options being correct is vanishingly small compared to the probability that they are the same image but one copy has been through several more photographic processes since it was first taken (something that can be taken as read in any option but 4 anyway).

 

[richardm]

Stray Cat-If it isn't the same image then the question becomes 'why are they so similar?'

Right. The differences SC highlighted are minuscule and debateable; the similarities are huge and unmistakeable.

 

[chillzero]

StrayCat,
Any particular reason why your blue line along the side of the girl's face does not meet the actual change in tone? You seem to have drawn it up the side of her face but leaving a line of shadow, whereas the red line follows the line of shadow.