Spem in alium: Orchestrating preservation

[These remarks were delivered to the Preservation Administration Interest Group at the American Library Association’s 2017 Annual Conference in Chicago, IL. By way of context, it was just announced that I was taking up the position of Director for Preservation at the Library of Congress. I spoke about preservation in libraries, but I believe that these apply as well to preservation in archives, museums, and other collecting institutions.]
I worry what Emily Post might say about lecturing on semiotics at 9 in the morning on a week-end, but Richenda got me a little bit fired up the need for a critical theory of preservation last night, so I’m going to start today with Heidegger’s thoughts on early Romantic period music and food storage:

“Beethoven’s quartets lie on the shelf like potatoes in the cellar. All works have this character of a thing. Yet the work of art, beyond its ‘thingness’ is something else as well. This something else constitutes its artistic core.”

A generation later, Lisa Block de Behar challenged Heidegger’s idea in a way that should perk our interest, playing off Heidegger’s conceit that it is the “something else” of a work that is permanent, and countering that it is the materiality of works that actually allows their persistence.

“It is not a matter of retrieving now the ‘thingness’ that Heidegger so looked down on, but rather of bearing in mind that this material condition he disdained and considered as subsidiary and perishable… is the condition of permanence of the work. The material condition does not prevent the fact that in what we call isogory lies also the true being of the work. ”

The term “isogory” is one she coins earlier, the thing itself and of particular interest here in a meeting of the Preservation and Reformatting section, a thing repeated.

“the work has repeatedly been considered to be something else, an allegory; at this point it must be added that the fact that a work is allegory does not prevent it from being “isogory,” the same thing, that a thing repeated is its essence: the thing itself.”

I cannot help but reflect on our own Functional Requirements for Bibliographic Records, but I fear that this also suggests that our job is a dull binary: preserve Items or, failing that, make new Manifestations through reformatting.

O Freunde, nicht diese Töne!
Sondern laßt uns angenehmere anstimmen,
und freudenvollere.

Oh friends, not these sounds!
Let us instead strike up more pleasing
and more joyful ones!

In the late 16th century,

… there was a song sent into England in 30 parts, which being sung made a heavenly Harmony. The Duke of — we believe it was Norfolk — bearing a great love to Music asked whether none of our Englishmen could set as good a song and Tallis being very skilfull was felt to try whether he would undertake the matter, which he did and made one of 40 parts which was sung in the long gallery at Arundell house, which so far surpassed the other that the Duke, hearing that song, took his chain of Gold from his neck & put that about Tallis’ necke and gave that him

The fact that we know this today is a victory for preservation. Tallis composed the motet in question, Spem in Alium, sometime between 1570 and 1573. The recording I will play in a moment was made in 2006, 430 odd years later. To survive across those centuries, Tallis had to set pen to paper, then bring the work into being through the voices of forty singers for a first performance. This cycle had to repeat some unknown number of times, with the music notation copied and recopied, singers trained and rehearsed, performances given and heard, for something like 50 years before Wateridge write recorded his remarks on the history of the piece.

That cycle has happened innumerable times since, with hand-written music notation giving way to printed and engraved notation, then to digitally typeset scores, with performances given and heard, and eventually captured as sound recordings, like this one:


All of this happened using methods we recognize — the collection and care of durable artifacts; coupled with periodic reformatting — but not a single capital-Preservation professional was involved.

I hope that this helps us to achieve the humility we need in order to reinvent our professional role, but not to lose the courage of our calling. Acknowledging that preservation occurs with or without our input is not a disparagement of preservation work, any more than it is a disparagement of medical doctors to acknowledge that people can heal without their help.

Acknowledging that we are not essential to preservation does not stop preservation from being essential to libraries. But we are now free to examine preservation from a new perspective and become the people responsible for shaping the business of libraries in ways that encourage diversity and sustainable modes of access to the whole of the human record.

This is where a thorough study of renaissance polyphony really comes in handy.

The last five or six years, I have been working with libraries, archives, and museums in areas outside the direct care and repair of their collections. The most striking lesson from that experience is that libraries are polyphonic places now. A preservation approach that is solely rooted in textual thinking risks becoming remote from the actual work of libraries.

I like this musical example because it puts the whole chain of creative activity on display: here, the textual materials are essential ingredients: invaluable for the study and analysis of the work, useful for its rehearsal and performance, but rarely used by people listening to that performance and indeed, only used as an aid to memory for the performers themselves. We can tell this same story about the computable datasets, assistive technologies, or varieties of interactive spaces that libraries now provide: these activities relate to and build upon the collected records of the library, but they are additive and distinct in themselves, just as the performance of a play or a motet is a thing unto itself.

When I think about preservation lately, I try to start with this understanding of the library, and then reverse and expand on usual formulation of “preservation and access” to arrive at “access enabled by preservation.” This is a change from a textual sense of our work to a performative sense, a shift from static or timeless ideas of preservation, to preservation as an outcome of doing the work of the library in a particular way. Our job is not to preserve things, but to refine the way the mechanism of the library operates, so that it generates preservation outcomes more efficiently and more consistently.

Music theory is a nice way of describing what this means, as well, and of overcoming one of our more fearsome professional worries: the risk of duplicative activity. Music reminds us that there is a virtue in repetition, and that when mixed with art, repetition transforms from simple duplication to informative reiteration and insightful variation. Indeed, one of the most challenging aspects of a work for forty voices is making them independent from one another yet coherent as a whole, and repetition is a key part of how this is possible. Tallis’ motet is a monument to duplicated effort, filled with imitation and repetition, call and response, and statement and variation.

The other key to marrying independence in parts with coherence at scale can be understood through another analogy to music composition. We talk about Tallis having composed this motet, but we should recognize that composition does not mean creation from first principles. Tallis was deploying a set of accepted practices for how to make music. There is still a great deal of novelty and invention within that framework, but the framework is crucial to allowing forty independent voices to create coherent music. Preservation has an existing and relevant practice here, as well, in setting standards for our work, but once again I hope we will continue to align our standards with the service orientation of our libraries:

Does the format of the information or the condition of the media create impediments to library services?
Can changes to format or condition improve discoverability or usability, or make costs more predictable?

The difference between duplicative effort and meaningful repetition comes from the use-case. Digitizing a page for reading is different than digitization to allow computation on the data in the table printed on that page. A conservation treatment to enable that digitization may different than the treatment to enable exhibition, or the repairs made to enable a book to be dropped in a backpack and read later without the loss of any pages. As preservation develops in the contemporary library, we need to shift our focus from the act to the outcome. In this motet, many voices sing the same series of pitches, but they sing them in different contexts and different times, to achieve different purposes.

Duplicated or re-iterated effort is worth thinking about in the emerging work of shared print archives, as well, which we expect to lead to the eventual draw-down of collections so that we rely on a cooperatively managed set of specifically selected copies of last resort. In that context, reformatting multiple instances of a title or engaging in conservation work on each remaining copy starts to have a value proposition that is distinctly different than working on that title without the context of a nationally valuable preservation commitment.

This cooperative change in library collection management should mark a dramatic shift in our entire approach to preservation. In fact, the word order of the Latin text that Tallis chose for this motet makes for a nice joke that’s right on point for this:

Spem in alium nunquam habui: Faith in another… I have never had.

We are going to soon find ourselves in new relationships to one another. We are coming to rely on a set of commonly held resources, and we must have be able to put faith in one another to see that those collections are kept available, yet there is only one collection we can be sure of.

I suspect that there is a rule of thumb that will start to emerge from this seeming paradox, and that an optimal number of copies will somehow correlate to the number of copies that is worthy of duplicative effort. What that number will be or how those copies will be distributed, I am not quite sure. Probably not forty, but eight and five both seem possible.

Let me conclude with one last musical allusion. I this there is general agreement in our profession already that it is a mistake to think of our work as primarily being a fixit shop or security warden, but I encounter a lot of uncertainty about what we do beyond those roles. I want to encourage you all to think of preservation as the function of the library is responsible for sustainable operations. Think of preservation as a metronome or a pitch pipe, a tool that helps keep all of the parts of the library in tune and in sync. Better yet, think of preservation work as orchestration, the process of choosing the best set of instruments to perform the work we want the library to accomplish.

In straightforward terms, it is all too easy for preservation to become one of the groups in the library that, in trying to speak for the collections, ends up saying “no” all the time. The collections deserve better than that. They have more to say. They are operatic in their scope and potential. I hope that we can find a more harmonious part to play, and that as we develop a preservation practice for the 21st century, we can become the group in the library that says, “Yes, and here’s how we’re going to do it.”