Making and Involuntary Memory

The Oklahoma Center for the Humanities at The University of Tulsa is pleased to continue its series of blog posts from the fellows of the 2018-19 academic year. The fellows are engaged in wide-ranging creative and research activities surrounding the idea of memory – a topic the center will explore through lectures, symposia, public debates, readings, writing, films and exhibitions.

This post was composed by Dr. Jeffrey Drouin, an associate professor of English at TU.

Making and Involuntary Memory

Shortly after 9/11, while living in Brooklyn with an old friend, I was surprised to hear a woman singing in his room. “Jonathan didn’t tell me he had a guest,” I thought, as I walked in to see several black boxes with glowing glass bottles on them. Strangely, there was no one else in the room, yet my ears told me that a real person was there, standing exactly in one spot and singing about science like an impassioned schoolteacher. I had walked into 1954. Mabel Mercer was performing Cole Porter’s “Experiment” with piano accompaniment in a studio room that had somehow been overlaid with this room by means of a seamless space-time fold. I was there, able to hear every nuance of breath, the wetness of her voice, the acoustics of a space that was considerably larger than the “real” one we inhabited, and most importantly the emotion that guided her interpretation of Porter’s classic tune.

It turned out that my friend Alex had loaned Jonathan a turntable, some records, and a system of thermionic vacuum tube amplifiers. It changed my life, leaving an impression that, many years later, led me to learn the basics of vacuum tube electronics in a search for that lost sound – that lost time. I built many different circuits but eventually poured all of my scholarly effort into documenting and building the amplifiers I’d heard back in 2001, in the hope that duplicating them would result in the same uncanny effect.

Only after three years of building amplifiers did I realize that I was engaged in a project very similar to that of one of the authors whom I study, the French novelist Marcel Proust. His À la recherche du temps perdu (In Search of Lost Time, or Remembrance of Things Past), a sprawling masterpiece comprising roughly 4,400 pages in seven volumes, seeks to unlock the mystery of involuntary memory, to make the past present again as a form of being. Strangely, Proust’s narrator ascribes the locus of involuntary memory to objects rather than the mind.

“[T]here is much to be said for the Celtic belief that the souls of those whom we have lost are held captive in some inferior being, in an animal, in a plant, in some inanimate object, and thus effectively lost to us until the day (which to many never comes) when we happen to pass by the tree or to obtain possession of the object which forms their prison…. And so it is with our own past. It is a labour in vain to attempt to recapture it. All the efforts of our intellect must prove futile.” (I.59)

Trying to achieve this type of memory deliberately is impossible, since its power depends on chance, yet try we must. Hence the title: Recherche, the Search. It can become a lifelong attempt to uncover the mysteries of a form of memory that only occurs when we stop trying, when some object provides a sensory impression that opens the boundary between past and present. The memories we recall by acts of will tend to appear as images of past experiences held in the mind’s eye, while involuntary memories are a form of resurrection, the re-vivification of a past event, a flashback – a chance moment in which some sensory channel makes the past present and alive again.

Releasing the genie from the thermionic bottles is a goal not limited to old amplifiers, however. What analog design brings to literature and the resurrection of recorded music, digital design in the humanities offers by way of a different yet somewhat similar kind of making.

Digital Humanities is a somewhat new field that uses computing applications to ask and answer humanistic questions. During graduate school, in order to facilitate recall for a meditative essay on the Recherche’s recurring use of church architecture as a motif, I made a spreadsheet containing all of the relevant passages (totaling about 190,000 words) and related information such as keywords and contextual notes. With the help of a friend, I later turned it into a database, paired the passages with church images, wrapped it in a blog, and put it all online as Ecclesiastical Proust Archive. Later on, I was exposed to computational techniques such as text mining, which processes natural language texts to reveal linguistic and stylistic patterns; topic modeling, which discovers patterns of terms that recur together; and network graphing, which shows connections among a set of structured data such as a spreadsheet.

The magic in these algorithmic techniques lies in visualizing patterns in the text that would not be uncovered in the act of serial reading: they reveal a latent reality by means of a different way of seeing. Take network graphing of the church motif spreadsheet, for instance. If we eliminate the temporal dimension to make all connections simultaneously present, what shape would the motif take? What patterns would appear among the passages and their keywords? How might it change our understanding of the novel? Without any particular question or reasoned possible outcome, I let the algorithm do its work to unlock the specter I’d been hunting. This may have been a naïve use of tools, but the first graph I generated with Gephi resembled the explosion of a firework in all the uncanniness with which the apparition of Mabel Mercer appeared to us all those years ago. Later iterations resemble a rose window. The church motif was centrifugal, radiating from the center rather than appearing as a periodic repetition along a chronological line, an artifact of the layout algorithm used in that instance.  Patterns revealed through text mining and topic modeling of the whole text lead to similarly unpredictable insights. In a way, digital techniques are akin to involuntary memory in that they expose hidden realities that we hadn’t thought to seek and raise questions we hadn’t thought to ask.

Proust’s narrative, like my exploration of audio design and Digital Humanities, works by creating the conditions through which involuntary memory can work upon us. All of this is made possible by the fact that the digital medium has shifted the archive from its institutional home in a building that stores official records, to a space in which anyone with sufficient means and technical ability can compile their own material, then create something new with it. Throughout much of the Recherche, objects such as gothic cathedrals and Norman churches play a kind of pagan role as the repositories of involuntary memory, of memory that we do not directly control, of collective, historical memory whose art forms act across generations and transform the subjectivities of individual lives wrought in time. Narratives of national, local, and religious memory are stored in stained glass, statuary, ornamentation; they are reproduced in performances of the liturgy, weddings, funerals, and even the simple gathering of the townsfolk once a week. They stand as archives of the primary material out of which the individual psyche and the social body are made, sacramentally transforming the deep history of Proust’s France into the forms of experience undergone by the narrator and his contemporaries. Today, prosthetic memory devices store the digital data of the past and, though we might purposely initiate a search or an algorithmic analysis, they remind us that sometimes we don’t get to choose the ways in which memory works, or works through us.

Work Cited

Proust, Marcel. In Search of Lost Time, Volume I: Swann’s Way. Trans. C.K. Scot Moncrieff, Terence Kilmartin, and D.J. Enright. New York: Modern Library, 1992.