MJK1 Sun/Shadow Study

I wanted to attempt to discover the time of day that MJK1 (the external of Miller’s Court) was photographed. In the original photograph there is evidence of sunlight and shadows cast across various parts of the scene. We know (or at least we think we know) the date that MJK1 was taken – 9th November 1888. It was taken at some point during the day that Mary Kelly’s body was discovered in her room at number 13 Miller’s Court.

Lots of things happened at Miller’s court during that tragic day and I thought it would be useful to work out what time of the day MJK1 was photographed. Pin-pointing a time might help researchers in their work.

So I’ve attempted to answer this question using a CGI reconstruction of the scene along with modern and historical maps,  a solar positioning calculator and 3D sun-mapping software.

The first thing I needed to do was accurately plot the position of Miller’s Court onto a modern map and establish latitude/longitude and other coordinates. I used Goad’s plans and various other contemporary maps of Dorset Street and overlaid them onto a Google map which also provided a handy sun position overlay:

5 map overlay of Miller's Court
5 map overlay of Miller’s Court

I obtained the global coordinates for the exact position of Miller’s Court and put them into the 3D sun position software along with the date – 9th Nov 1888.

Armed with this data, I constructed and positioned the 3D model:

Now, I wanted to test the sun position calculations and the model’s position against something that could demonstrate the simulation’s accuracy. I decided to build another 3D model of the front of Miller’s court based on Leonard Matters’ 1928 Miller’s Court photographs:

I set the date to June 20th 1928 – Leonard Matters took the above photographs in LATE June 1928. I arbitrarily picked the 20th but any of the last 10 days of June 1928 would set the sun in an almost exact same position.

I ran the program and acquired these time lapses for the hours between 9am and 5pm:

And these frames (frame 14) demonstrate that Matters probably took his photographs at about 11:45 on a morning in late June 1928: (click to enlarge)


Note the shadowing is very close to the actual photographs. I endeavored to create as close a match as possible when constructing the 3D models. In Matters 2 note the shadows from the buildings on the opposite side of Dorset street. In frame 44 they cover the same amount of pavement as in the actual photograph. Note The shadowing of the chair in both photographs is accurately demonstrated in my 3D’s.

I was happy that the simulation provides a good representation of how the sun/shadow would cast in a real world situation. However, note the time lapse of No.13. Note that NO sunlight reaches No.13. The sun in June is at it’s highest throughout the year yet No.13 never gets any direct sunlight. This didn’t bode well for my next test which was a simulation of the 9th November 1888:

As you can see, the sun is so low in the sky that it never gets high enough to cast light onto the front of Miller’s Court, let alone reaching the back of the court. See the sun position images (above) and you can see how low the sun is.

Frame 54 (above) shows Dorset Street and No.13 at 1pm on 9th Nov 1888. Only ambient light/shadow will be cast. Miller’s Court, and in particular No.13, is surrounded by high imposing buildings. The light of the day can only penetrate into the court from the top down. Shadows might change slightly as the sun moves across the sky due to light reflecting off the surrounding buildings. But in November 1888 the day was overcast. Details of the weather on the 8th and 9th Nov 1888 acquired from casebook.org state:

“8th Nov – Dull cold day: rain at night.

9th Nov – Wet till 11A.M. then overcast.”

So not only was the sun very low in the sky but also the cloud and rain would result in a very dark, drab environment. Miller’s Court that day would be dark enough to perhaps require the gas lamp to burn through the day? Which leads me on to ask the question: What light source is casting the HARD shadow from the archway above No.13’s door?

Look at frame 54 (above) and note where I think my shadows (circled in red) match with the actual photograph of No.13. I believe that the shadowing in my 3D models does actually match. Note the shadows cast by the ‘rungs’ of the BIG window, Note the shadows cast by the sills and the archway above the door to No.13. Those shadows seen in the actual photo are, I would suggest, simply ambient light/shadows cast from the bleak daylight above and down into the court. And as such it’s not possible to ascertain the time of day that MJK1 was taken (after all that!).

But the question remains (to me at least); what is causing that hard shadow to be cast from the archway over the door to No.13?

Was the photographer using a ‘flash’ or flood light to illuminate the court enough to get a photograph? Is the gas lamp – situated opposite the door to No.13 – still burning? Or perhaps the photograph has, over the years, been adjusted in terms of light/dark contrast so as to make the once soft shadows more harsh and black (and the light colours more washed-out)?

In an attempt to answer one of those questions, I added a gas-lamp to my 3D model of the court, positioned on the opposite wall over-looking the door to No.13. When this lamp is burning my simulation suggests that the shadows over the archway, as well as the shadows within the door to No.13 are lost. If the lamp was burning at the time photo was taken then there would be lots of light and less shadow cast into the arch and doorway to No.13.

So in conclusion, the question I set out to answer – what time of the day was MJK1 taken? – has categorically NOT been answered! I feel that my simulation, the data, maps, positions and models are accurate. I was pleased to see the result of the Leonard Matters simulation but disappointing that MJK1 refused to yield an answer.