Is single context recording the only way?

[Marc Berger]

Not having any experience or knowledge of having ever dug a face down canvas on the floor my first effort would be to remove the canvas and label it 1. I would then take a picture and make a plan of the next context (2) that I considered to be on "top" and so proceed. I imagine that I would start to notice the direction in which the "paint" had been applied and this would lead to interesting theories in post ex as to whether the picture had been painted by a left handed person with their head pointing to the west and the feet to the south whilst laying on their back staring at the ceiling. Any way at some point I would come to a floor (3 ), the density of which was such as to stop the "painting" moving any closer to the centre of the earth. I don't see how other than by the device of cloud stratigraphy I could get out of the matrix 1,2,3. or suggest that 2 was applied upside down (which might make a 2 look like a 5).

Came across this but haven't got much beyond the title. http://www.assemblage.group.shef.ac.uk/4/4rxt.html

[Wax]

Chemical analysis of the different paint types might tell you which were the earliest or tell you that the floor paint is very different to that on the front or the back of the canvas. You might have problems phasing things from their dirrect physical relationships to each other so would look for scientific dating evidence. You might even identify three different artists from the way they used the paint.

Of course your interpretation might get it wrong but as you will have produced a detailed record of your analysis someone else might be able to reinterpret.

If a canvas was laid in wet paint it would leave a faint depression. The way the paint on the front and back of a painting relates to the canvas and its stretchers would tell you which side was painted first. So yes physical releationships are clues to temporal releationships.

[Marc Berger]

but Mr Wooldridge agreed that "always thought the Harris Matrix showed the order in which things happened (chronological)." We need to get 2 above 1 but connected to 3 if 3 is older than 1. If we do this but have an arrow on the edge between 2 and 1 pointing to 1 we can show chronology and upside down deposition. Let us hope that they did not use an old tin of paint that predated the construction of the canvas and the floor.

[P Prentice]

its just interpretation - no need to get your knickers in a twist

[Wax]

If twisted Knickers were found in relationship with the painted canvas and the floor I would start suspecting the whole set up was an installation by a well known lady artist and start looking for the evidence. (Though of course there might be several other stories there)

This painting analogy is working well for me and is great fun.

The paint made before the floor wass constructed or the canvas made was still used on the floor after the floor was constructed and before the painting was placed on the floor. The dating of the floor paint has given us an interesting side story and added to the over all narrative. BAJR is right its about narrative, PP is also right it is just an interpretation

[Crocodile]

I was kinda presuming that when you start an excavation that gravity informed you which way was down and that this would then orientate the stratigraphic terms lowest and bottom.

The original face of the surface that you are excavating or examining orientates which way is "down".

This is all covered in Principles, I think you can still download it for free.

[Crocodile]

How does it work if you had to clean the picture but it was face down on the floor.

I would pick the picture up and turn it so its painted face was up. If you were to remove deposits from the back of the painting then this represents a different face and the deposits on this face do not have any direct stratigraphic relationship with the paint on the other side.

let's say the face down painting is instead a collapsed painted ceiling. The collapse of the ceiling transforms it into a new layer which is later than the surface it fell onto even if the painting and ceiling were originally earlier than the lower layer.

[barkingdigger]

Surely a painting in a frame on the floor would be lifted as a "find", and not as a context? :face-stir:

Unless, of course, the frame and canvas had rotted away - then the paint fragments would be either a "layer" in their own right or inclusions within a deposit. You'd be hard-pressed to notice there were multiple coats of paint to differentiate! Still, it's a dead cat in a box, or strapped to buttered toast, or something, so we can get all abstract...

If I KNEW the paint coats had different dates, and related to the canvas, etc then I'd eventually draw up the matrix with the "newest" event (or cost?) at the top. Doesn't matter about physical orientation.

[Crocodile]

All I was referring to was the fact that the smoke deposits on a ceiling form downwards and as does the sequence of rafter-plaster, paint, smoke AND that the deposited layers of plaster, layers of paint, layers of smoke also form downwards which is counter to superposition. Now you can throw these deposits into Harris matrix but I am unaware of a convention that expresses both these characteristic's in Harris matrix other than the use of interpretation which I imagine would result in a matrix where the flow (? how about "edge") lines must come out of the bottom of the rafter context and turn upwards to the plaster then the paint and then the smoke, possibly then equating to the embers before proceeding down below rafters to find the floor context.

The floor and the ceiling are seperate surfaces with no direct stratigraphic relationship, I don't believe there is any convention for extending the matrix downward through the space of the room as if it were a layer in the matrix.

Are you winding us up or do I have to go back and read Harris again?

[GnomeKing]

(a good thread - time to post)
remember that Harris is not ~God (:0) and derives from older traditions in geology - where the canvas has not only been flipped, torn, stuck back together, flipped again, folded, had parts of another completely differnt picture punched through it, but then baked in magma for a few million years....

the key assumption is that fragmneted units retain enough integrity to be objectively/rationaly 're-assembled' into a sequence (or at least sufficently enough to propose a hypothesis, or to plan a transect/bore)....but this is not always the case, especialy in the lithosphere-biosphere interaction zone (where archaeologists most commonly reside)

one thing Harris-type Matrix is very bad at is pedology, drift geomorphology, concepts of equifinality, and generally all things Soil.

I can not begin/be bothered to count the number of times i have seen spurious 'stratigraphic' relationships made because "there HAS to be a sequence" - when infact the evidence is equivocal at best.

It is difficult at first to operate in the context of equifinal data - it is a strange and somewhat parralel world, when it must accpeted as a matter of methodological fact more than one process/event/factor could lead to evidence/data that apperas identical to us today....but, after all, how certian could we ever be, and, fundementally, about what?