Digitisation part 2: Roger Deakins’ Contact Sheets

“Deakins’ negatives are not accompanied by an exhaustive set of contact sheets. A digital contact sheet will therefore be made with a scanner before being digitised. This will be a more lengthy process than the Ravilious collection requiring more time allowance.”

The statement above was the shortened version of my project brief for Beaford Arts Hidden Histories, a project funded by the Heritage Lottery Fund and Leader 5 North Devon. Contact sheets do not exist for Roger Deakins work, he did make them but cut them into strips and slotted them into negative bags with numbers removed. Although his contacts need preserving, and ultimately digitising for the archive, they are useless for the digital database without being able to identify unique negative numbers.

Unlike Ravilious, there was no numerical labelling of Deakins’ negative bags, so I started the process of making new ‘digital’ contact sheets with the first of four folders of negative bags. These negatives were housed in their original archival, but translucent, paper negative bags, often with written notes marks and connotations. Under close inspection on a lightbox I could see that negatives were often filed in a random order, sometimes up-side-down or reversed, sometimes the same number appeared twice and on one occasion 3 times. It became clear that a negative bag didn’t necessarily contain one film, one specific shoot, location, person or group. In comparrison to Ravilious’ contacts there appeared a randomness, or chaos, to his negative filing. However on closer inspection a thematic approach eg ‘railway stations’ had been taken. The negatives themselves were generally in very good condition for their age, with only a few which had stains through poor fixing or washing.

The negatives would ultimately need re-bagging in a comparable fashion to James Ravilious, clear plastic archival bags (this would also be necessary for making digital contact sheets through scanning), so in consultation with Beaford and the Devon Archives Conservator, I went about consolidating the negative strips and making new collections based on single films. This was very time consuming because Deakins (like Ravilious) often loaded his own 35mm canisters of film from a bulk roll so the first negative on a roll could be any number from 0 to 40 and each roll could be any length from approximately 12 to 38 frames.

I made the new digital contact sheets through scanning Deakins’ negatives, held within clear file pages, on the project’s Epson 800 Perfection V800 scanner. This time is was set to scan transparencies at 600 dpi, large enough to be able to identify the place and/or subject in a photograph and to see those images in the context of the photoshoot, but small enough to make physical and data base storage practical. It was also within the independent consultant’s recommendation and the scanner’s linear relationship between quality and time: whilst a negative sheet was being scanned there was just enough time to make minor adjustments to cropping, exposure and contrast of the previously made digital contact sheet.

This was the most exciting and rewarding part of the project to date because I was able to see, as a positive image, that which had only existed as a 35mm negative since 1972. I wonder how many great images from yesteryear have been lost, because the 36x24mm negative or contact image had been brushed aside or unnoticed through their size, or poor exposure. Seeing each of Deakins’ images, with corrections made to exposure and contrast, 24cm wide on my computer monitor brought them to life. If only this technology was available when I was shooting film myself!

I had earlier written how Deakins’ negatives themselves were generally in very good condition for their age; however once the digital contacts had been made it became clear that some suffered from poor developing and light leaks (and in the case of the example above, double exposure), reminiscent of Robert Capa’s celebrated D-Day Landing photographs in the example below. After discussing the issue with the Hidden Histories coordinator we decided, rather than attempting the difficult and time consuming task of perfecting these faults through Photoshop, that we should embrace their endearing, nostalgic qualities; only digitally correcting damage to the negatives post processing.

Digitisation Part 3 Making copies of negatives


Digitisation part 1: James Ravilious Contact Sheets

I’ve started an 18 month freelance contract for Beaford Arts. My work is to digitise 10,000, 35mm, black and white negatives, of James Ravilious and Roger Deakins. These date from 1971 to 1989 and cover all aspects of life in rural North Devon.

Bowens Illumitran converted to LED for copying negatives is on the right of the image

The archive of negatives is held in a climate controlled, fireproof, strong room at the Devon Heritage Centre in Exeter. This is where the negatives must be kept and downstairs in a conservators office is where I have my work place seen in the images above.

“..James Ravilious’ work includes a complete collection of contact sheets made by the artist. These will be digitised by the Digitiser using a flatbed scanner. Most contact sheets will be straightforward scans, however, some have overlapping images and exposure issues to resolve in digitisation.”

“All Ravilious contact sheets will be digitised to a high enough quality to enable negative selection to take place and research and background data to be gathered without repeated handling of the original contact sheets. ‘Good’ and ‘Best’ images will be marked up on digital contact sheets by the Digitiser, referencing the original database, to ensure easy reference and prevent repeat selection. Two sets of digital contact sheets will be made one for negative selection, and one left unmarked to reveal James’s original markings only, for dissemination purposes. Tests suggest scanning contact sheets at 600dpi enables high quality images suitable for on-screen viewing purposes.”

James Ravilious started his commission from Beaford Arts to “show north Devon people to themselves, in 1973, continuing the Beaford Archive started a year earlier through Roger Deakins (who became the great Hollywood cinematographer). The first two months of my contract have been spent scanning all 2306 of Ravilious’ contact sheets on an Epson Perfection V800 Scanner. This was the first stage of the process to archive the contact sheets digitally so that in the short term a curator can select approximately 9000 of his images for digitisation and in the long term the digitised contact sheets can be themselves searchable items in a database. Ravilious worked right from the start with an archive in mind; archivally processing his negatives as best he could (developing, fixing and washing his negatives in an often difficult, imperfect working environment) and numbering, filing and storing his negative bags and contact sheets. He also annotated his contact sheets on their face and rear indicating his prefered best images, description of place, name of subject etc.

A small percentage of the contact sheets (and negatives) had been archively rehoused in separate new folders prior to my start which speeded up the digitising process. However once I got to view the contact sheet and negatives together in the same folder I noticed some negative strips and many ‘end of film’ negatives which didn’t make it onto the contact sheets. I started to make digital copies of these to add to the digitised contact strips but then abandoned this extra time-consuming work, marking negative bags with ‘post-it’ notes for later in the life of the project.

There were 2 or 3 sets of negatives, and occasional single frames, which had yellowed through under fixing or inadequate washing at the original time of processing the film. These films are not archival and will not last in the long term and will need re-fixing and washing in the coming weeks.

Although the negatives were all black and white I scanned them in colour to preserve the highlighted annotations, often in red pen. However, the drawback of the colour scan was often a slight colour shift towards green or blue, so a selective colour desaturation was applied in Photoshop afterwards. There were also handwritten annotations on the rear of Ravilious’ contact sheets which will need digitising and archiving later in the project.

Today, photographing digitally, one gets used to getting a ‘correct’ exposure. Most people now shoot everything automatically and today’s technology enables us to achieve excellent exposures for most of our photographs. Even professionals, myself included, who use a camera manually, are used to checking our pictures at the time of shooting so that any adjustments to exposure can be made there and then. These same professionals will benefit from incredible advances in sensitivity to light when taking pictures, in comparison to the 100-400 iso films used by Ravilious in the 1970’s & 80’s; and also shoot in RAW, benefiting from its greater exposure tolerance. So it shouldn’t have come as a surprise to find James Ravilious’ contact sheets having both inconsistent exposure, one to the next, and having under and over exposure within a single film. But it did, because the last time I photographed on film was last century, and my last monochrome film was probably at the time Ravilious finished his work for Beaford Arts.

Therefore, digitising James Ravilious’ contact sheets, preserving them archivally for eternity, was a greater task than it seemed. Each sheet needed slight adjustments to exposure at the scanning stage through tweaking the levels of the histogram. Each digitised sheet needed adjustments for some of its images. These adjustments were not to perfect the images, or ‘make good’ his exposure, but to enable a reader of the digital contact sheet to be able to see what the image was, and in the short term to enable the project curator to identify images (negative) to be digitised.

I feel quite privileged to be one of probably only a handful of people to have seen 9,000+ of James Ravilious? images, albeit small and with little detail, and to have gained an insight into his working practice through seeing 2,300+ of his contact sheets made in chronological order through his time at Beaford.

Digitisation part 2: Roger Deakins’ Contact Sheets