Category Archives: DACS

Archival archeology: a working definition

As the project for processing and describing the Meyer Schapiro Collection comes to a close, I’ve come to realize how, like an archeological dig, I sifted through strata of documents, papers, images, photographs, and countless other mediums, to make an intellectual framework for Schapiro’s records.

I first came across the linking of the term "archeology" with "archives" in the paper "A Transition of Bits: A Case Study in Preserving the Michael Joyce Digital Papers at the Harry Ransom Center." The authors, discussing the processing of electronic files by hypertext author Michael Joyce, discuss the data recovery and preparation of these files stored on floppy discs from the 1980s as "digital archeology."

As they write, "many of the procedures that were used to extract and identify the electronic records were dictated by the characteristics of the storage media. "


Their "digital archeology process" is a six step procedure as follows:

1. Receive and identify physical media

2. Create a cataloging system for the physical media

3. Copy files from physical media and record metadata

4. Perform initial file processing

5. Create an item-level listing of all recovered files

6. Create working copies of all files and protect the original copies

Because born digital records are easy to create, and countless iterations of a single document can count as potential "records" in and of themselves, the process of understanding where, when, and by whom records are created require a form of authentication.

As the authors of the paper suggest, in regards to irregular or incorrect data in countless digital records, there is a "fundamental assumption in digital archeology: we assume that the date provided by the original user is correct. This is really no different than the analogous case in physical archives, where dates are written on paper as best the creator can recall and are subject to error. The lesson learned here is that although digitally assigned dates may be reliable in most cases, they are not immune to error and must be taken as a best estimate rather than indisputable fact."

A case in point that is quite important to remember. As archivists and collection managers, we deal with records that, at  times, are difficult to ascertain their veracity. With large physical collections, records are dispersed across cabinets, boxes, and files. With digital collections, files and documents can be of the same content, but with various different file names and located in a variety of folders.

Making a cohesive arrangement and description of records requires that you utilize the records as sources of evidence and that, in the end, those evidential exhibits form a larger whole to the records motive. In this sense, extending "digital archeology" to a broader term "archival archeology" suits both practices of analog and digital archival material, but links them in a process that is principal to archival practice: the importance of provenance and the authenticity of records.

In the article "Making the Leap from Parts to Whole: Evidence and Inference in Archival Arrangement and Description," Jennifer Meehan gives a stunning overview on how evidence gathered from the archival collection itself can be used to infer why those records were created, by whom, and for what purposes.

As Meehan articulates, "As archivists we must reason about the records in this way not only because we are removed from past events by both space and time, but because the information about these events obtained from existing sources is always only partial, and therefore incomplete. Gathering contextual information is not sufficient to understand the various contexts of any given collection. To arrive at such an understanding requires the archivist to make a leap of sorts, an inferential leap from what is available in the present texts (the contextual information) to some past event (the specific activity that gave rise to and/or subsequently shaped the records)."

As Meehan suggests of making inferential leaps,"the archivist in effect creates the external and internal relationships of a body of records. Rather than merely identifying these relationships on the basis of gathered information, the archivist for all intents and purposes constructs these relationships on the basis of the inferences drawn from the gathered information."

As a case in point, I’d like to give a quick gloss into how archival archeology plays a part in inferring the "parts to whole" relationship of archival records.

Meyer Schapiro wrote extensively throughout his whole life. He constantly had in mind various projects that could be potential books, but, for one reason or another, never manifested as books. These files were essentially compiled papers, outlines, and research notes that were assembled and often times paginated to reflect a cohesive manuscript.

Schapiro was notorious for revisiting his writings later in his life, reworking, editing, and expanding upon writings from the 1930s to the 1960s. His wife, Lillian Milgram Schapiro, would also aid him in those efforts, and by the 1990s, after he had passed away, she would continue to edit those records (and, to add to the volume of material, she would copy everything in triplicate!).

Indeed, it was quite a challenge, when as an archivist, you would see files such as the images illustrated in this post. But it was through archival archeology, and the process of inference, that I came to understand the meaning of these records and what their motives were.

For instance, I utilized correspondence and editorial notes to understand the context of these records. I also drew on other record types such as reproductions, editorial notes, and other materials found across the entire collection to synthesize what these records actually meant. The following is an arrangement note for the manuscript Schapiro wanted to publish titled " The content of modern art: studies in the painting of the end of the nineteenth century from Manet to Munch" and it gives a sense of how complex archival processing really is, and how archival archeology and inference play a part in contextualing records.

<arrangement>According to correspondence in file 18 of box 242 of this series, plans to publish this manuscript began in 1937, when Oxford University Press, New York, initiated contract discussions with Schapiro. There was also an agreement between Oxford University Press and the Museum of Modern Art to collaborate on the publication in 1939, but that never materialized. In an announcement dated 1938 found in file 19 of box 242 of this series, Phaidon Press was to publish the manuscript as "The content of modern art: studies in painting from Manet to Munch." That publication was also never realized. There are three variations of the title; the longer version is used since it includes variants from the other two titles. When possible, file titles below were supplied or condensed using Schapiro’s annotations or text to guide users to the contents found in the manuscript. Original file order is maintained to preserve Schapiro’s ordering. As a result, file titles may not include sequential sequencing such as section numbers. Files have research notes and editorial notes interleaved in the typescript sheets).</arrangement>

In another example, Schapiro continually worked on a manuscript titled "Pablo Picasso’s Guernica". The arrangement note is as follows

<arrangement> From a book jacket design in file 5 of box 248 in this series and correspondence in file 9 box 248, this manuscript was to be published by the Museum of Modern Art, New York as "Guernica: studies, postscripts" possibly in the 1970s or 1980s. Schapiro wrote consistently on Picasso’s Guernica, and material found in these files is culled from the 1940s through the 1990s. Files are organized by a pagination scheme devised by Lillian Milgram Schapiro in the 1990s through the 2000s relying on Schapiro’s original annotations on the typescripts. Her notes and lists relating to the reconstruction of the manuscript can be found in file 5 of box 248</arrangement>

Providing the researcher a note on how the archival material was arranged, and the process by which inference and archival archeology played a part in contextualing those records, gives the records a sharper focus and plays a part in authenticating their context.

As Meehan rightly suggests, one should provide information about "the analytical work done during processing, including the rationale for a particular arrangement, the reasoning behind decisions, and the sources of information used in reaching a particular decision. This sort of information would go a long way toward documenting the archival context of the records. And, if made available to the public, this information would enable users to make their own decisions about the possible meaning order(s) of a particular collection."

Describing processes that go into contextualing records can be recorded in order to authenticate records. Following the parameters set out by the InterPARES 1 project "Authenticity Task Force ‘Requirements for Assessing and Maintaining the Authenticity of Electronic Records,’"Michael Forstrom elaborates on this discussion in the article "Managing Electronic Records in Manuscript Collections: A Case Study from the Beinecke Rare Book and Manuscript Library" in the journal American Archivist.

The author provides recommendations on how archival archeology, inference, and description can work in tendem “to guarantee the records’ identity and integrity.” These steps include utilizing elements of Describing Archives: a Content Standards (DACS) and explicating on the whole process of preserving and contextualizing the records.

In the end, the work of documenting and explaining how a set of records came to be arranged, described, and processed is part of a larger whole: one that balances the use of evidence gleaned from archival archeology with an archivist’s inference into their relationship with the creator of records.

This can at times be a journey for the archivist and one which opens the work of archival processing into the sphere of archival documentation: an important reminder on how archival archeology should be utilized as evidence in the process of describing archival material.

On "records," papers," and "collection": a DACS case in point

As the project keeps progressing, one thing has become abundantly clear: the Meyer Schapiro archive encompasses much more than a traditional manuscript collection. While a significant potion of the archive is indeed Schapiro’s “papers,” it also houses other material such as audio-visual documents and a major collection of his own drawings, paintings, and sculptures. These components were given to the Rare Book & Manuscript Library at different times and, as a whole, form an aggregate that is best described as a “collection.”

This is so for several reasons and let me unravel the thread a little to explain why.

As always, definitions are helpful and can be used to provide clarity to the prospect of titling this particular “archival unit.”

Describing Archives: A Content Standard (DACS), is a Society of American Archivists approved descriptive standard for the archival community that “guides archivists and catalogers in creating robust descriptive systems and descriptive records.”

According to rule 2.3.18 in DACS, there are three predominant terms that can be used to signify the nature of an “archival unit” within a title. (These are examples from the DACS manual: “Coalition to Stop Trident Records,” “Mortimer Jerome Adler papers,” and “Semans family papers”).

They are as follows:

1) Records–where the materials being described consist of three or more forms of documents created, assembled, accumulated, and/or maintained and used by a government agency or private organization such as a business or club;

2) Papers–where the materials being described consist of three or more forms of documents created, assembled, accumulated, and/or maintained and used by a person or family;

3) Collection–when describing an intentionally assembled collection.

DACS rule 2.3.18 is applied to the highest level of the archival unit and each of the above terms include materials across media. In other words, regardless if the entire unit contains diverse mediums (traditional papers, electronic records, audio records, art works, etc.), these three terms should be used to describe an entire unit.

When the project began, the archival unit was known as the “Meyer Schapiro papers,” which would have been sufficient, but not entirely accurate.

This is because the different units of the archive were not all part and parcel of the unit itself.

Diverse components have been unified to form what is now the “Meyer Schapiro collection” housed at Columbia University’s Rare Book & Manuscript Library. This includes art works (nearly 3,000 individual items) given at a separate time from the bulk of the manuscript collection. There is also archival material housed at the Art History & Archeology Department that is now part of the “collection” itself.

Because these various units have become “intentionally assembled” into an aggregate to be housed in one repository, the term “collection” is better suited than “papers” as outlined in DACS. While the creator is Meyer Schapiro himself, these different units, each with unique custodial histories, form a broader whole that have now been drawn together as the “Meyer Schapiro collection.”

Thanks to DACS, the archival community now has a standard for making questions such as these easier to navigate and give these broad terms sharper focus. Now, the next step is creating a detailed description of the collection’s components and also providing contextual information on Schapiro to buttress the finding aid using Encoded Archival Description (EAD).

In this respect, my work is just beginning…