Excavating the Memory Palace: Arts of Visualization from the Agora to the Computer

Excavating the Memory Palace

Excavating the Memory Palace

Arts of Visualization from the Agora to the Computer

Seth Long

The University of Chicago Press    Chicago and London

The University of Chicago Press, Chicago 60637

The University of Chicago Press, Ltd., London

© 2020 by The University of Chicago

All rights reserved. No part of this book may be used or reproduced in any manner whatsoever without written permission, except in the case of brief quotations in critical articles and reviews. For more information, contact the University of Chicago Press, 1427 E. 60th St., Chicago, IL 60637.

Published 2020

Printed in the United States of America

29 28 27 26 25 24 23 22 21 20    1 2 3 4 5

ISBN-13: 978-0-226-69514-3 (cloth)

ISBN-13: 978-0-226-69528-0 (paper)

ISBN-13: 978-0-226-69531-0 (e-book)

DOI: https://doi.org/10.7208/chicago/9780226695310.001.0001

Library of Congress Cataloging-in-Publication Data

Names: Long, Seth, author.

Title: Excavating the memory palace : arts of visualization from the agora to the computer / Seth Long.

Description: Chicago : University of Chicago Press, 2020. | Includes bibliographical references and index.

Identifiers: LCCN 2020021304 | ISBN 9780226695143 (cloth) | ISBN 9780226695280 (paperback) | ISBN 9780226695310 (ebook)

Subjects: LCSH: Mnemonics—History. | Memory.

Classification: LCC BF381 .L66 2020 | DDC 153.1/4—dc23

LC record available at https://lccn.loc.gov/2020021304

This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper).

For my parents

And to Doug Mitchell, in memoriam

Time moves in one direction, memory in another. We are that strange species that constructs artifacts intended to counter the natural flow of forgetting.

William Gibson, Dead Man Sings

It’s a poor sort of memory that only works backwards, the Queen remarked.

Lewis Carroll, Through the Looking-Glass

Contents

Introduction. Memory vs. Mnemonic

Chapter 1. Arts of Memory in the Agora

Chapter 2. Arts of Memory in the Monastery

Chapter 3. The Memory Palace in Ruins

Chapter 4. The Memory Palace Modernized

Chapter 5. Theory and Practice of a Digital Ars Memoria

Chapter 6. The Social Memory Palace

Acknowledgments

Notes

Bibliography

Index

• Introduction •

Memory vs. Mnemonic

A memory:

My wife asks, What was the name of the bar we hung out in? After classes in college?

I draw a blank. There are a lot of bars in that part of Orange County, California.

The one next to the frozen yogurt place? she says.

Frozen yogurt? Right. Frozen yogurt place. I remember that. Pink chairs and tables, cute Japanese frog mascot, more toppings than yogurt.

The red walls, remember? my wife persists. The walls in the place were velvet red. You were getting into scotch.

Dark red walls. Scotch. Next to the yogurt shop with the pink tables.

My mind becomes less blank: The District, I announce. I have not set foot in this bar in over a decade. I have not thought about it for almost as long.

The District! my wife says. I was thinking about it, she says from across the couch, because I heard this song on the radio the other day. She turns up the volume on her laptop—an old pop tune from the mid-2000s, saccharine and synthesized. Remember I’d dance to it when it came on? She’s hinting that we should dance now, but I don’t respond. My mind is time traveling.

Yogurt shop. Pink tables. The District, red walls, smell of scotch. Sound of a synthesizer in minor key.

I have not thought about these things in a decade, but suddenly I’m there, in the eye of the imagination. I’m not only at the District when this stupid song came out. I’m also in 2006. From the red walls to the scotch to the song and beyond, my mind conjures and connects a thousand details I’d misplaced, a thousand details I’d forgotten I’d misplaced: my German professor’s name, the plot of a short story I was writing, every menu item at the Mexican place across from campus, where I had lunch with my friend’s sister who lent me that book (what was it?—right, Confederacy of Dunces, her copy was missing half the cover). Each detail discloses five more. Where have these memories been hiding?

This is not an exercise in association. Not in the logical sense of that word, anyway (A is to B, as C is to ______? or "What do you think of when you hear the word America?"). Rather, it is sensory association of an explicitly personal nature. It is the optical idiosyncrasy of subjective experience. Would anyone else remember A Confederacy of Dunces by way of a yogurt shop?

The phenomenon should be a familiar one: a chance picture, a song, a smell, and all at once the details of another lifetime rush to the mind’s eye, one after another, details unlocked from unconscious spaces you didn’t know existed.

This is what the ancients called the art of memory.

More accurately, it is the natural process that the art of memory hacks into and harnesses for rhetorical ends. This book provides a history of the art of memory—the memory palace technique, as it is known today—and recovers its precepts for the digital age. I write recovers rather than refigures or retheorizes, because classical mnemonics require only minor upgrades to be relevant to the electronic age. Unique among rhetoric’s five canons, memory is and was concerned not with language alone but also with graphic depiction and systems of visualization. An emphasis on the sensory rather than linguistic dimensions of rhetorical activity makes the fourth canon already compatible with the icons, apps, data visualizations, and other means of graphic communication ubiquitous in the digital media ecology.

Recover is also an apt verb because the fourth canon’s visual precepts have yet to be taken up by contemporary rhetorical theorists in a consistent way, despite the fact that the canon has long since been rediscovered and given a central role in the field. Upon inspection, however, rhetoric’s new memoria retains only a fragile connection to the memoria practiced throughout history. The new fourth canon is not the same as the old—which would not be an issue if the old canon were no longer relevant to the twenty-first century. However, the ancient canon of memory remains deeply relevant to the contemporary moment; it was and is useful for more than memorizing speeches. The canon’s rich history, I suggest, can be a touchstone for novel explorations of memory in our hypervisual, hypermediated culture.

To be sure, across the last thirty years of rhetorical scholarship, one can locate provocative allusions to the fourth canon’s visual mnemonics. For example, Jay Bolter moves in the right direction when he says that the art of memory was designed to address the gap between writing and memory and that the way it does this is by writing, not with letters but with images, visualizing one’s speech and mapping it into a mental structure.¹ Likewise, Sharon Crowley has acknowledged that classical memory practices suggest a rhetoric that is not solely literacy-based.² Richard Young and Patricia Sullivan have considered the connection between spatial and linguistic memory, arguing implicitly that the fourth canon evokes semiotic modes beyond language.³ More recently, the collection Rhetoric, Remembrance, and Visual Form has brought together numerous scholars to explore the intersections among visual culture and practices of memory, the potency and fallibility of images as mediums of the past, and the simultaneously authentic yet manipulated qualities of memory when conjured through visual modes.⁴ In the collection’s introduction, Bradford Vivian and Anne Teresa Demo recognize that memory and imagery were intimately linked in the classical era’s mnemonic practices, contending that the same can be said for memory practices today: "Individual proficiency in the classical art of memory depended on a host of visual exercises. . . . The ars memoriae required orators to conjure elaborate mental images of roomy palaces or public spaces in order to memorize lengthy and complex discourses. Both personal and computational, or artistic and digital, memories find their raison d’être in the visual media upon which they rely. To remember, then as now, is to see.⁵ As suggestive as this sounds, however, Vivian and Demo go on to prove my initial point that the visual precepts of the fourth canon, though recognized, are often treated as a historical curiosity no longer relevant to theory or practice. Vivian and Demo state that mnemonics were simply a metaphorical substitute for cognitive mental operations."⁶ While this is perhaps true for Aristotle’s understanding of mental imagery, it is not true for the ars memoria, the art of memory, whose practitioners took the creation of images literally and seriously. In addition, although Vivian and Demo begin their collection with a nod to visual mnemonics, none of the collection’s authors do the same. They instead explore artifacts such as public memorials, war photographs, and paintings. Collectively, they seem to use the canon of memory not as a starting point for new theories or practices—it is no longer a performative technique, Vivian remarks elsewhere—but as a mode of inquiry.⁷ This is a good description of the way memory has been taken up in rhetorical scholarship. Memory is examined in its natural, social, or psychological sense and used as a bridge to cultural criticism, historical inquiry, and so on. It is adopted as a pretext for analyzing the many ways humans mediate the past with objects, from community cookbooks to national war memorials, so that scholars might pose the critical question Who wants whom to remember what, and why? as Alon Confino puts it.⁸ Few rhetoricians have expressed interest in recovering mnemonic precepts as such or in asking what relevance these precepts have for the present. Of course, I am not suggesting that the current direction of rhetorical memory studies is misguided in any way. There is plenty of room in the field for that continuing project. However, in this book, I take rhetorical memory in a different direction (though one that ultimately finds common ground with the larger body of rhetorical scholarship on memory).

My starting point is a suggestion made by Collin Gifford Brooke in Lingua Fracta. Memory, Brooke states, "is the one canon whose status as practice is in need of rehabilitation."⁹ This notion of mnemonic practice was central to the canon historically and must remain so if it is to be useful today. Memory as practice—in particular, a visual practice—is a common theme in the history of rhetoric’s memory arts. Ancient, medieval, and early modern rhetoricians made a clear distinction between natural memory and artificial memory, the latter denoting mnemonic techniques, methods, and habits through which one’s memoryscape could be harnessed for rhetorical action. It was the job of the rhetorician to develop and teach these precepts for mnemonic practice.

As I have suggested, the most valuable insight to be gained from the canon of memory is its emphasis on the visual. Stewart Whittemore defines rhetorical memory as a practice of data management, and as he rightly notes, the idea of memory qua information management is an ancient one.¹⁰ While Whittemore’s book Rhetorical Memory emphasizes the organizational aspects of memory, I want to emphasize the optical, sensory ones—which are related to but not synonymous with the notion of organization. From pre-Socratic Greece until the early modern period, the art of memory espoused the principle that words and information could best be recalled when re-mediated into a visual form possessing emotional resonance with the rhetor. Perceptions received by the ears or by reflection can be most easily retained, states Cicero in De oratore, if they are also conveyed to our minds by the mediation of our eyes.¹¹ Converting information into images—housed in memory palaces in the mind or on the page—turned thinking into a form of seeing.¹² It translated the semiotics of language and thought into a more palpable semiotics of sight, dimension, and affect. Importantly, this association between memory and the oculus imaginationis—the eye of the imagination—was central to the long-standing link between memory and invention. Visualizing information made it easier to recall, browse through, and integrate when inventing new discourse or speaking extemporaneously.

Visualizing information to aid invention is an ancient practice, to be sure, but one that should resonate with anyone who has ever used a computer. Though no longer occurring in the mind, the work of transforming and accessing information with visual modes is as commonplace today as it was in the past. The terabytes of information stored on the world’s servers would be useless without the graphical user interfaces developed to aid access to them. Data have multiplied, but the techniques used to manage access are the same today as they were for Cicero: make it visual, make it spatial. The history of rhetoric’s fourth canon can thus provide a wealth of insight into the rhetorical dimensions of graphical data representation, which is omnipresent in and fundamental to the new media age. Indeed, the memory palace can be understood as one of history’s first interfaces. It was a medium, recognized as such, connecting the knower with the thing known (or, to put it in digital terms, connecting the user with some form of underlying program logic, which is, of course, just another semivisual layer connecting the user to the metal). Words and information are abstract. By incarnating them in image, they become not real but more accessible to the human who would use them to coordinate activity. The history of rhetoric’s ars memoria can thus be read as a history of information visualization systems, from emotionally charged mental phantasms to mundane pictures in print to full-scale memory theaters—and, eventually, to file folders, icons, and network maps. Across their varied guises, the images of the ars memoria have acted as visual portals between information and the users seeking to integrate information into a new assemblage.

The history of the art of memory is also a history of memory. Although I make a distinction at the outset between natural memory and artificial mnemonics, a dynamic relationship exists between the two, as it exists in all cases between humans and their artificial, secondary knowledge. Mnemonics mirror memory. Artificial memory techniques tap into natural memory processes and thus reflect, at one remove, how our minds convert a trillion flickers of continuous sensory impression into semistable visual models we call memories. For ancient rhetors, the memory palace was a site of purposeful visual curation; memory itself, whatever else it may be, can also be understood in those terms. Without reproducing a naïve Lockean theory of mind as tabula rasa—contra Locke, the human system emerges prewired to attune to some things but not others—we can nevertheless assert that human brains do curate the world’s interminable data streams into sensory-driven models. Expectedly, humans do consciously what their brains do automatically: they reduce complex information to lucid visual depictions.

The impulse to capture knowledge of intangible things by converting the intangible to tangible, visible maps or models is a universal one. The impulse is synonymous with the workings of memory itself. However, the human ability to conjure secondary knowledge—in the form of visual models of intangible information—is rife with ethical complications. When the intangible is made tangible or the nebulous is made visible, it can be harnessed, controlled, objectified, commodified, and used for unethical ends. Should image making or at least certain images be condemned? The dialectical forces of image creation and image destruction are as universal as the powers of remembering and forgetting. Looming behind the imagery of memory—in natural and artificial form—is the countering force of iconoclasm. From Plato to Oliver Cromwell to contemporary software critics, iconoclasts have condemned those who convert intangible data into visual, tangible knowledge. Images become idols. Idols lead to the worship of false gods. But where idols exist, prophets arise to smash them. The history of ars memoria is thus a history not only of visualization but also of its opposite: iconoclasm.

A detailed history of rhetoric’s memory arts—written for rhetoricians or new media scholars as opposed to historians—is lacking in the field. This book in part is intended to fill that gap. More essentially, however, a discussion of the mnemonic tradition’s digital relevance becomes possible only when grounded in a comprehensive history. This book is thus arranged in chronological order. It begins in chapter 1 with the art of memory’s classical origin. I provide an overview of its earliest treatments in Dissoi logoi, Plato, and Aristotle before moving on to its treatment by the Roman rhetoricians Cicero and Quintilian. Chapter 1 argues that the fourth canon, from its earliest conception, exploited the psychological principle of sensory-emotional association to produce an art of info management through info visualization. The chapter explores further the art of memory’s close connection with inventio in ancient Mediterranean rhetoric. This memory-invention link is reiterated throughout the book, supporting the argument that a digital ars memoria should similarly adopt the classical outlook that data visualization is an art of invention rather than a method for data mastery.

This book also examines, as a secondary concern, how the fourth canon’s precepts have adapted to the social and technological ecologies into which they have migrated. Chapter 2 follows the art of memory in the Middle Ages as it moves from the secular, public agora to the more interior scribal culture of the Christian monastery. Working from the foundational scholarship of Mary Carruthers and Frances Yates, I locate the influence of classical mnemonics on medieval figures such as Alcuin of York, Hugh of Saint Victor, and Thomas Bradwardine, exploring how the visual character of the ars memoria did and did not change in its Christianized form. I also draw attention to imageless memory systems that circulated alongside the visual art of memory, allowing me to examine an important distinction between information management and information visualization—between spreadsheets and data visualizations, as we might say today. Chapter 2 explores how this difference manifested itself in medieval mnemonics and why it matters.

Chapter 3 tracks the art of memory’s evolution into the early modern period, during which time the mnemonic arts witnessed a renewed interest across Europe—except in England. Assessing, confirming, and enlarging a thesis forwarded by Frances Yates, I contend that a unique confluence of iconoclasm and Ramism (an imageless memory system) contributed to the marginalization and loss of visual mnemonics in Reformation England. Several English memory treatises—including John Willis’s Mnemonica—are read against this iconoclastic backdrop. I argue that an iconoclastic tendency robbed English rhetoric not only of memoria’s imagery but also ipso facto of the ancient link between memoria and inventio.

After pitting iconoclasm and visual mnemonics against each other, I further highlight, in chapter 3, a tendency toward digital iconoclasm in the work of Wendy Hui Kyong Chun, Friedrich Kittler, and other new media scholars and philosophers of technology. I connect Puritan iconoclasm with the contemporary distrust of visually mediated information: then, as now, visual knowledge was criticized for its role in sustaining, circulating, and justifying unethical power structures.

Iconoclastic dogma led to the marginalization of the memory arts in England by the early 1600s. However, by the Enlightenment in the eighteenth century, the mnemonic tradition found itself marginalized in most European contexts. The tradition, surprisingly, then underwent a resurgence in the nineteenth century. Chapter 4 offers one of the first contemporary overviews of the memory palace’s Enlightenment and modern-era trajectory, examining works on memory by well-known figures (such as Gottfried Leibniz) and more obscure ones (such as Marius D’Assigny). During its period of Victorian resurgence, the memory arts grew in popularity but became detached not only from rhetorical invention but also from the rhetorical arts generally. The art of memory migrated into the orbit of faculty psychology, then later into the handbooks of train-your-memory teachers, a progression that ultimately diminished the importance of mnemonics within Western traditions. The art of memory, among other mnemonic techniques, was bowdlerized into a tool for rote memorization, disconnected from any epistemological or imaginative concerns. Far removed from Cicero’s exalted treatment, the art of memory met an ignominious end, which can be read as a proxy for the marginalization of rhetoric itself in an increasingly scientific age.

Chapter 5 arrives in the present, grappling with technological concerns introduced in previous chapters. I frame contemporary data-visualization technologies as a renewed manifestation of the visual mnemonic tradition, exploring how digital mnemonics might or might not be informed by the memory palace technique. Walking a line between the elusive goal of complete data mastery and the familiar iconoclastic censure of artificial memory in all its guises (from writing to software), I suggest that we return to the classical conception of mnemonic technique as an aid—not to flawless recall but to rhetorical invention and knowledge construction. Responding to the work of digital humanists such as Stephen Ramsay, Wendy Hui Kyong Chun, and Johanna Drucker, I examine how digital tools and techniques can function as aids to inquiry—that is, as starting points for knowledge construction rather than as idols that speak transparently for themselves. Indeed, I argue that data visualizations are valuable for their ability to picture not knowledge but the process of constructing knowledge from partial information and biased data—which is precisely what human minds do when creating memories. As a mnemonic device, data visualization can thus illuminate the workings of natural memory qua dynamic interface. In the chapter, I explore the theoretical and social implications of this artificial/natural memory binary in the digital age.

Whereas chapter 5 examines the scientific imagery associated with the term data visualization—networks, interactive maps, graphs, and the like—the final chapter returns to the principle of sensory-emotional association harnessed by the classical art. It investigates the life of personal artificial memories circulating on Facebook, Instagram, Twitter, and other apps in the social mediascape. Working from an idea first developed by Jeff Pruchnic and Kim Lacey, I read social media as a digital memory palace, created and arranged with associative images of our own design. However, unlike the imagines or images of the interior memory palace, mnemonic images constructed online are networked, searchable, public. The externalization and socialization of the memory palace raises problems and offers potentials as the (digital) art of memory migrates once more into a new technological ecology.

This book explores an abstract thing, the art of memory, in various historical eras. I undertake many close readings of primary and secondary material, but at certain points I use computational methods to gain a perspective on the concept that is either much larger or more microscopic than the individual text. In each case, texts, their features, and/or their bibliographic metadata stand in as imperfect but still-revealing proxies for discourse about the fourth canon in various periods and contexts. Some of these methods are straightforward. For example, in chapter 3, I graph the publication sites of over four hundred memory treatises, the graphs acting as evidence for larger claims regarding early modern memory practices. Other methods, however, will be unfamiliar to the humanities and thus do require an explanation, which I provide in endnotes, in-text explanations, and references.

I adopt these computational, corpus-based approaches because a link exists between them and the classical art of memory. Recall what I said about the art’s visual precepts: the creation of mnemonic imagery and memory palaces converted the semiotics of language and thought into the semiotics of sight and dimension. Memory, it was believed, operates most keenly when occupied with visual forms. The images of the classical art of memory are thus similar to the data visualizations of computational analysis. Graphs, charts, networks, even word lists—these objects distill into a single image a greater totality of information than working memory alone could readily synthesize. They are memory aids, and creating them constitutes a mnemonic practice. Like the art of memory, computational methods can be used to visualize words and knowledge to aid invention. By incorporating visualizations into the book, I provide a present-day equivalent of historical memory practices, demonstrating their utility as arts of invention for the twenty-first century.

• Chapter 1 •

Arts of Memory in the Agora

At the twilight of the Roman Republic, circa 55 BCE, the orator and statesman Marcus Tullius Cicero wrote his De oratore, a treatise on rhetoric and its moral role in political leadership. In the text, Cicero works his way through the five rhetorical canons—the first three being invention, arrangement, and style—arriving in the second book at memoria, the canon of memory. About the fourth canon, Cicero states, As Simonides wisely observed, the things best pictured by our minds are those that have been conveyed and imprinted on them by one of the senses. Now the keenest of all our senses is the sense of sight. Therefore, things perceived by our hearing or during our thought processes can be most easily grasped by the mind, if they are also conveyed to our minds through the mediation of the eyes.¹ For Cicero, as for Quintilian after him and the Greek sophists before him, the practice of memoria is an eminently visual one. It is the practice of converting words and abstract information—things perceived by our hearing or during our thought processes—into visual form. The mind, Cicero states, can grasp visual, sensory things better than it can grasp the abstractions of linguistically mediated information. To remember, the orator should thus convert information into vivid imagery within the oculus imaginationis: the mind’s eye, not just the mind, must be engaged if information is to be capably recalled, integrated, and applied. From its earliest elucidation, the art of memory operated on this basic premise.

In 1995, in the midst of the digital revolution, computer engineer James A. Wise and his colleagues published an article entitled Visualizing the Non-Visual: Spatial Analysis and Interaction with Information from Text Documents. Graphical user interfaces had been part of the computer landscape for over a decade at this point, but Wise et al.’s article offered an early attempt to apply graphical thinking to text-based documents, which, by the mid-1990s, had begun to multiply into gigabyte territory. The need to read and assess large amounts of text, Wise et al. state, puts a severe upper limit on the amount of text information that can be processed. Working memory, in other words, is not capable of comprehending mega- and gigabytes of textual information; text at these scales cannot be read in the traditional sense. The solution, Wise et al. argue, is to overcome constraints to working memory through content abstraction and spatialization, transforming documents into a new visual representation that communicates by image instead of prose. Comprehending information contained within thousands or millions of texts, Wise et al. continue, can be facilitated by converting the raw textual data into a spatial representation which may then be accessed and explored by visual processes alone. Re-mediating texts into navigable images allows users to deploy their powers of visual perception to explore patterns emerging across multiple documents.² The best way to access and comprehend proliferating textual databases, in short, according to Wise et al., is to visualize them.

Juxtaposing these two texts—one an ancient rhetorical treatise, the other a computer engineering article—reveals that across two millennia both the orators of the ancient world and the scientists of the digital one arrived at an identical conclusion about mnemonic access: visualization is key. The most efficient way for humans to access and utilize information is to deploy semiotic modes other than the written or spoken word. A more visual medium is required. Put another way, what these texts demonstrate is that, from the agora to the computer, the art of memory has been an art of visualization—of converting information into imagery, vague thought into lucid sight. The technology has changed, but the concept has remained the same. Telling the story of how we got from the memory palace of the mind’s eye to the one on our screens is the objective of this book.

Memories. You’re talking about memories.

What, precisely, is the ars memoria, the art of memory, as conceptualized and practiced throughout history? Taking a cue from Cicero, I think the best way to introduce it is by way of moving pictures.

In an early scene in Ridley Scott’s Blade Runner, a woman named Rachael enters a boardroom in the Tyrell Corporation building, where she is asked by Rick Deckard and Eldon Tyrell