Collection Care blog

19 February 2018

Digitising books as objects: The invisible made visible

Book conservator Flavio Marzo explores how the experience for users of online library material surrogates could be easily improved by enhancing invisible physical features of books.

Working as a book conservator within digitisation projects has been my job for many years. I started in 2006, only one year after joining the British Library Conservation team here in London after leaving my country, Italy.

The subject of that digitisation project was the digitisation and virtual reunification of the Codex Sinaiticus, possibly one of the most known and valuable manuscripts in the Western world. The Codex was compiled in the IV century AD and is the oldest surviving and most complete version of the Old and New Testament. Many years have passed since that project and digitisation has become a common work stream within public institutions. This is especially evident within libraries which now compete in uploading material from their own collections to make them available for scholars, students and readers across the globe.

Technology has improved immensely since then and a lot of ‘ink’ has been spread across physical and virtual pages about the remit, the limitations and the advantages of what is offered to the public through the surrogates uploaded onto countless web portals. This piece is just another little drop into this ocean of ink to share some considerations built upon experience and from the perspective of a book conservator who sees, because of his professional background, the limitations of this, but also the exciting challenges to overcome them.

Books are physical objects and the pleasure of opening them, turning the pages, looking at (when decorated) the illuminations and their pigments, or at the accretions of the ink strokes, even smelling them, cannot be recreated on the screen of a home desktop. This does not mean that we cannot improve the experience and possibly further close the gap between the real object and the two-dimensional images.

I now work for the British Library/Qatar Foundation Digitisation Project and for the past 5 years, with a team of two conservators, I have been repairing documents (printed and hand written) and Scientific Arabic manuscripts for the team of scholars and photographers who are doing the real magic by gifting the world with the content now available on the Qatar Digital Library website (

I have worked with books all my life since I was a 16 year old apprentice in a Benedictine Monastery. I have to admit that I am not an avid reader but I love books as objects and I get very excited about all the different little features and materials they are made from. How is it possible to please someone like me when offering online surrogates of complex items like books?

Books are recognised as 3D items and a lot of work has been done to migrate the content of those printed and manuscript texts into online, easy to access versions, but very little has been done to capture their physicality as objects.

Photographers, like any other professional, follow strict professional standards defined by general rules and specific project boundaries. Those standards are built to assure that the best possible result is achieved consistently and the meter to measure this result is the quality of the final product i.e. the image to be uploaded. Those images are supposed to reproduce as faithfully as possible the text and the carrier of the content of a book. Very rarely attention is given to the substrate or to the physical features of the object.

Lights for digitisation are carefully positioned to avoid shadows and they help to reduce surface irregularities and anomalies. This is all to the benefit of the written text and/or of the decorations, but with much loss for the lovers of the book as an object!

Here I want to describe some very practical ways to achieve different results and show some ‘behind the scenes’ of items I have been working on and how these very interesting results can be achieved with simple straight-forward techniques that do not require any high-tech equipment.

Raking light

I have mentioned the Codex Sinaiticus and I would like to start with it.

Normal light Raking light

Revelation, 2:7 - 3:5, British Library folio: 326. Left: Normal light. Right: Raking light.

All the available remaining pages of the Codex from the different geographical sites where they are presently held (The British library, The Library of the University of Leipzig, The National Library of Russia in Saint Petersburg, and the Holy Monastery of Saint Catherine’s) were digitised and uploaded onto the purposely created website ( Contrary to common practice, all the pages were imaged both with normal and ‘raking light’. 

When imaging pages of books with normal light the attention is placed primarily to achieve the full readability of the text. Lights are placed and conditioned to radiate evenly over the surface of the page making sure no shadows are created and paying great attention to colours and tones to ensure they are as close as possible to the real appearance of the object reproduced.

What ‘raking light’ does is very different and the resulting image reveals a completely new landscape. Placing the source of light horizontally relative to the page results in an enhanced texture of the substrate which highlights and brings to life all the physical features present on the surface of the pages. These interesting and unique features can relate to the preparation of the writing surface or more generally to the specific material the substrate is made of e.g. papyrus, parchment or paper.

Here are some details of pages of the Codex Sinaiticus taken with raking light.

Codex Sinaiticus ruled Scraping of the surface
In the previous images the source of light is now helping us to appreciate this famous manuscript on a completely different level.

Horizontal and vertical lines, holes pierced through the page, and scratch marks now appear clearly. They are traced on the surface of the pages for a purpose; those are features related to the page preparation that happened before the text was traced onto it.

The ‘bounding lines’ (vertical) and the ‘writing lines’ (horizontal) are impressed with a blind (not too sharp) tool onto the parchment sheets. The holes, highlighted with red circles in the second image, are used as a reference. This is known as pricking holes for the ruling of the page to provide the scribes with a guide for writing.

The scratches visible on the surface of the page are most likely the marks left by the pumice stone. The pumice stone was commonly used to prepare the surface of the abraded parchment sheet to make it more absorbent and therefore improve the grasp between the grease substrate and the writing ink.

Thanks to this lighting system it is also possible to see the direction of the indentation of those lines and holes. This information can help codicologists, even from the comfort of their homes, to understand from which side of the folio they were traced and pierced and so recreate the step by step process of the creation of an ancient manuscript.       

Letter from Emir Letter from Emir raking light
In this image we see the images of a letter sent by the Emir of Bagdad in 1899 to Lord Curzon when he was appointed Viceroy of India, first taken with normal and then with ‘raking light’. In the first image the letter is just a sheet of paper beautifully arranged and decorated with writing. In the second image the light tells us a completely different story; it shows us the use of this letter, the way it was folded and the number of folds it had.

How incredible that it is possible to see all these different insights by just slightly moving a lamp!

Transmitted light

Another technique to read paper from a different perspective is using ‘transmitted light’. 


Simply by placing the same sheet of paper onto a light table (i.e. illuminating from below) it is possible to bring a completely new scenario to life. In this image for example we can clearly see the watermark impressed onto the sheet of paper of the previous letter, detail impossible to be seen only looking at the image taken under normal light.

Paper can be hand or machine made, and sheets can bear chain and wire lines or possess watermarks or not. These details can be of great interest to scholars and add valuable information to the understanding of documents in relation to their use and circulation.     

Laid and wove paper Transmitted light India Office Records

Here are some more examples of sheets of laid and wove paper taken from different files from the India Office Record material, some showing again the characteristic chain and wire lines (except the last one which is actually a sheet of machine made wove paper) and some very distinctive water marks highlighted and made visible thanks to the used of transmitted light. 

Visualization of the physical collation of manuscripts

Books are made of folios and pages and those folios are ‘bound’ together. How the bindings are made is one of the real wonders of books. The variances are numberless and the materials and details of execution not only delight nerds like myself, but more importantly they inform researchers about the history of those books, giving insights into the objects that open doors to sometime unexpected cultural landscapes through links between different craftsmanship and cultures.

To describe a book structure is a very delicate and laborious process, but one that conservators are trained to do and that they automatically do many times when conserving those books as they record the treatments being carried out.

A lot of work has been done during recent years to create tools able to easily make this complex information sharable with the wider audiences. One I wish to mention here is VisColl, developed by Dot Porter at the Schoenberg Institute for Manuscript Studies at the University of Pennsylvania ( in collaboration with Alberto Campagnolo at the Library of Congress, a friend and colleague.

Book structure digitisation

In this image we have, additional to the images of the digitised pages, diagrams (on the left) of the structure of the section where those pages are located and, highlighted in white, the specific pages shown on the screen.

Those diagrams, surely more easily understandable than many wordy descriptions, can help researchers to step into a completely new level of understanding for the manuscript, providing vital information about the history of those items, the way they were put together and possibly evidences of late alterations or even forgeries which may have occurred throughout the centuries.

Digitisation has opened new ways to look and make use of books and, I believe, the improvement of understanding of physical features is the next step that should be consistently and widely taken to enhance the online user experience.

One of the issues digitisation has brought to the attention of conservators and professionals involved in the care and preservation of library material is the fact that by enhancing the ‘fame’ of objects we can cause an increase in how much those same objects are requested for access.

To justify restriction in handling objects, which for the most part are very fragile and extremely valuable, we need to improve the online metadata and the amount of information available with the surrogates. Those presented here are just some examples in how, quite easily, this can be done.

Obviously the smell will stay within the walls of the libraries, but those are pleasures to be experienced in situ, and alone (almost..!) at the table of the reading rooms. No surrogate can replace that for the lovers of books.