Pre-Scientific Analyses of Memory

Hermann von Ebbinghaus famously remarked (1908) that "Psychology has a long past, but only a short history".  What's true for psychology in general is just as true for the psychology of memory in particular.

The scientific analysis of memory began in 1885, with Ebbinghaus' monograph, Uber das Gedachtniss (On Memory), which reported the first controlled, quantitative analyses of learning and memory.  Prior to that, Theodule Ribot had published his monograph on Diseases of Memory (1982), but that monograph focused on clinical description rather than controlled analyses.  Afterwards, William James devoted a chapter (16) to memory in the Principles of Psychology, proposing a distinction between primary (short-term) and secondary (long-term) memory, based primarily on his own introspections (James referenced Ebbinghaus, but didn't make much of his work).  

But the philosophical analysis of memory goes back to the very beginnings of Western philosophy.

The Art of Memory

Of course, preliterate societies (in the West or the East) had to rely on memory.  The ancient Greeks and Romans had a system of writing -- pretty much the Greek and Roman alphabets known to us today. But written records were terribly expensive: they had to be reproduced one at a time by hand, or (in medieval times) pages of text had to had to be carved into woodblocks for printing.  Moreover, even in medieval times not everyone could read. The widespread reproduction of written materials was not possible until the invention of moveable type by Johannes Gutenberg in the 15th century.  

Before that time, Western culture relied on oral expression.  Literature and speeches were transmitted orally, recited from memory, as in epic poems such as the Iliad and the Odyssey, the plays of classical theatre (Aeschylus and Sophocles), and orations such as Cicero's.  These cultural practices put enormous strains on memory.  For this reason, there developed a set of techniques for easing the cognitive burdens imposed by memorization.  These were known as artificial memory or mnemotechnics.

The history of artificial memory has been traced by Frances Yates, an historian of the Renaissance, in the Art of Memory (1966).  See also Moonwalking with Einstein: The Art and Science of Remembering Everything (2011) by Jonathan Foer, who hired a mnemonist to teach him the art of memory, and then entered the United States division of the World Memory Championships -- and did remarkably well.

"Art"?  In the ancient and medieval world, the ability to memorize, and to recite texts from memory, was part of what it meant to be a cultured person -- for that matter, part of what it meant to have ethics, or for that matter to be human.  There were few written texts (not to mention computers connected to the Internet) to rely on -- people had to keep it all in their heads.  This all changed with the invention of moveable type, and the proliferation of books and other textual materials, which began a massive cultural shift toward the "externalization" of memory.  The availability of books, in turn, changed how we read.  As Robert Darnton notes in "First Steps Toward a History of Reading", ancient and medieval scholars (or anyone who was literate) used to read a few texts, like the Bible, intensively; now we read extensively, a large number of books and other texts, but usually only once.   Now, we don't have to remember everything anymore.  But we still have to remember where it's recorded. 

The Dialexeis

The first recorded text on memory techniques is the Dialexeis, dating from  400 BCE, which provides both an intuitive theory of memory and the application of these intuitions to mnemotechnics.

"A great and beautiful invention is memory, always useful both for learning and for life."

On attention and memory:

This is the first thing: if you pay attention, the judgment will better perceive the things going through it."

On repetition and memory:

"Secondly, repeat again what you hear, for by often hearing and saying the same things, what you have learned comes complete into your memory."

On memory and knowledge:

"Thirdly, what you hear, place on what you know.  For example, Chrysippus is to be remembered; we place it on gold (chrisos) and horse (hippos).  Another example: we place glow-worm on fire  and shine.  For courage, on Mars and Achilles; for metal-working, on Vulcan; for cowardice, on Epeus.

This last principle provides the basis for all subsequent techniques of artificial memory.

The Invention of Mnemotechnics

The discovery of these techniques is attributed to Simonides, a poet of ancient Greece.  Here is the story as told by Cicero, a Roman orator who lived in the 1st century BCE (and who used these techniques himself).  Quoting from Yates (1966, pp. 1-2), emphasis added:

At a banquet given by a nobleman of Thessaly named Scopas, the poet Simonides of Ceos changed a lyric poem in honour of his host but including a passage in praise of Castor and Pollux.  Scopas meanly told the poet that he would only pay him half the sum agreed upon for the panegyric and that he must obtain the balance from the twin gods to whom he had devoted half the poem.

A little later, a message was brought in to Simonides that two young men were waiting outside who wished to see him.  he rose from the banquet and went out but could find no one.  During his absence the roof of the banqueting hall fell in, crushing Scopas and all the guests to death beneath the ruins; the corpses were so mangled that the relatives who came to take them away for burial were unable to identify them.

But Simonides remembered the places at which they had been sitting at the table and was therefore able to indicate to the relatives which were their dead.  The invisible callers, Castor and Pollux, had handsomely paid for their share in the panegyric by drawing Simonides away from the banquet just before the crash.

And this experience suggested to the poet the principles of the art of memory, of which he is said to have been the inventor.  Noting that it was through his memory of the places at which the guests had been sitting that he had been able to identify the bodies, he realised that orderly arrangement is essential for good memory.

Now quoting from Cicero: 

Simonides inferred that persons desiring to train the faculty of memory must select places and form mental images of the things they wish to remember and store those images in the places, so that the order of the places will preserve the order of the things, and the images of the things will denote the things themselves, and we shall employ the places and images respectively as a wax writing-tablet and the letters written on it.

Cicero then elaborated on this insight.  But Cicero was also clear on the limitations of Simonides' method.  A powerful memory, in his view, was a gift of nature, and mnemotechnics could only serve to make an already-powerful memory stronger.  But, in his view, even the most "dull-witted" person could achieve some benefit from them.

Ad Herennium and the Method of Loci

Simonides' techniques were further elaborated by the anonymous author of Ad Herennium (hereafter, AdH), written about 86 BCE -- in fact, about the time of Cicero and for a long time incorrectly attributed to him.  This advice on rhetoric, addressed to a pupil named Herennius, became an important part of the European cultural tradition.  It, in turn, is based on ancient Greek and Roman texts which are now lost.  Thus, it is the ultimate source for all later work on the arts of memory.

There are two other ancient sources on the art of memory:

AdH begins by likening the art of memory to "inner writing".  In what we now call the Method of Loci (MofL), images are associated with places.  As Cicero put it, in a famous analogy, the places and images are like letters inscribed on a wax tablet.

First, the orator must have "orderly arrangements" of places easily committed to memory, where the images are to be placed.  

Note the underlying theory of memory: storage is not permanent, and forgetting is a function of both decay and displacement (interference).  You are to give every fifth locus some distinguishing feature, as an aid in maintaining order.

There are also rules for the selection of places.

And there are rules for images:

Again, AdH intuits a basic memory principle of distinctiveness.

AdH then gives an example of remembering in a legal case, in which a defendant was accused of poisoning a man to gain his inheritance:

We shall imagine the man in question as lying ill in bed, if we know him personally.  If we do not know him, we shall yet take some one to be our invalid, but not a man of the lowest class, so that he may come to mind at once.  And we shall place the defendant at the bedside, holding in his right hand a cup, in his left, tablets, and on the fourth finger, a ram's testicles.  In this way we can have in memory the man who was poisoned, the witnesses, and the inheritance.

The Art of Memory in Medieval and Renaissance Europe

AdH served as the primary source for medieval training in rhetoric --the techniques were essentially passed from pagan to Christian Rome, and then taken up by Church fathers such as Alcuin (735?-804), Albertus Magnus (1193?-1280), and Thomas Aquinas (1225-1274).  These principles were often falsely attributed to Cicero, who was a pagan hero to early Christians.  Cicero, for his part, identified a good memory with Prudence, one of the cardinal virtues (the others were Justice, Fortitude, and Temperance).  Memory could be trained; therefore, the practice of artificial memory was part of the virtue of Prudence.

Instruction in the art of memory persisted well into the Renaissance, and was taken to China by Matteo Ricci, a 16th-century Jesuit Prince whose story is told by Jonathan Spence, the Yale historian of China, in The Memory Palace of Matteo Ricci (1984).  Ricci lived in china from 1583-1610, working as the tutor to the son of a Chinese scholar-bureaucrat.  He amazed his Chinese hosts with the power of his memory, and tried to teach them how to do the same - -even regaling them with the story of Simonides (transliterated Xi-Mo-Ni-De).  Artificial memory was of obvious relevance to the Chinese, who had to memorize vast numbers of individual ideographs, each representing a different word.  At this time, movable type was just being invented in Europe, but its invention was no use to the Chinese, who were not about to import a system of "barbarian" writing.  But Ricci's training in memory had other purposes: to demonstrate the power of European culture; as part of the practice of the Christian virtue of Prudence; and as a way of conveying the European, Christian worldview.  Ricci's "memory palace" was constructed in the Chinese style, but the loci were decorated with pictures representing Christian themes: "The Apostle in the Waves", "The Road to Emmaus", "The Men of Sodom", and the "Madonna and Child".

Ricci's story offers a clue to why the art of memory persisted after the invention of moveable type.  The loci and images were no longer cues to remembering "things" and "words".  Rather, they were representations of an entire worldview, as in Romberich's Spheres of the Universe of 1533.

Perhaps the most famous example of this type is Dante's Divine Comedy, with its nine circles of Hell, ten heavens of Paradise, and seven terraces of Purgatory.

Dante's Inferno consists of nine circles, each representing a sin, and populated with exemplary sinners.  

Dante's Purgatory consists of seven terraces, each representing one of the Seven Deadly Sins, plus places for the Excommunicate, the Unabsolved, and Negligent Rulers.

Dante's Paradise consists of 10 heavens, each representing a virtue, and populated with exemplary saints.

Another Renaissance example is the Memory Theatre of Cuilio Camillo, a visionary philosopher of the 15th century who proposed building an entire physical theatre, with loci and images representing the entire universe.

Bruno proposed a memory system representing "occult" knowledge, filled with signs of the Zodiac, all of whose images had implications for sorcery and witchcraft.  Bruno's use of the MofL makes clear that the purpose of the structure is no longer to remember "things" and "words".  Rather, the structure is a thing in itself to be remembered, because it represents the nature of the physical and moral universe according to a particular philosophical or religious point of view.

Still, it's apparent that Bruno really did use the MoL for memorizing things.  As a young Dominican friar, he gave a command performance before Pope Pius V, reciting Psalm 86 ("Incline your ear O Lord"), in Hebrew, forwards and backwards.  That was, of course, before the Roman Inquisition burned him at the stake in Rome's Campo de' Fiori -- where, in 1889, radical Italian youth erected a statue in his honor.

The art of memory is understandable in ancient Greece and Rome, where paper was expensive,  printing virtually nonexistent, and teachers and orators had to rely on their memory.  But why was the art of memory still being taught in the high Renaissance, where these conditions no longer held?

One theory is that bureaucrats still relied on memory to handle the masses of documents produced by the new literacy and printing technology.  But, as Anthony Grafton (2008) has pointed out, Renaissance clerks were already developing sophisticated filing systems to manage burgeoning archives.  More likely (following suggestions by Ann Blair and Noel Malcolm), Grafton suggests, it was readers of all types, at all levels of society, who had to rely on human memory to cope with all the printed material around them: "As shelves groaned and notebooks swelled to bursting, memory remained the only thread that could lead one back through paper labyrinths to the facts and data that mattered" (p. 77).

The Art of Memory Today

Despite writing, computer text-editing and databases, and iPods, the art of memory persists with us today, in the form of familiar techniques which help us to remember various things, for example:

As an aside, many if not most mnemonics have sexist or racist undertones -- as well they might, having been developed mostly by 19^th-century American and European male professionals.  Here's a project for some sociologist of memory: do outgroups have other mnemonic devices?

Aleksandr Solzhenitsyn (1918-2008), the Russian author who won the Nobel Prize for Literature for his chronicles of the horrors of totalitarianism in the Soviet Union under Lenin and Stalin, used a kind of mnemonic device while imprisoned in the Ekibastuz labor camp in Kazakhstan, where writing was forbidden.  Employing a sort of rosary made out of beads of chewed bread, similar to those that Catholic prisoners used to say their prayers, Solzhenitsyn assigned long prose passages to each bead -- rehearsing each passage until be could remember it perfectly before proceeding to the next bead.  In this way he memorized some 12,000 lines of prose in what became his groundbreaking novel, A Day in the Life of Ivan Denisovich ("Solzhenitsyn, Literary Giant Who Defied Soviets, Dies at 89" by Michael T. Kaufman, New York Times 08/03/08).

And so did Tony Judt, the political historian who was struck by an aggressive variant of amyotrophic lateral sclerosis ("Lou Gherig's Disease"), which -- within about a year of diagnosis -- rendered him quadriplegic, paralyzed from the waist down, and requiring an assistive breathing device.

By last February, Judt could no longer move his hands.  "I thought it would be catastrophic,", he recalls matter-of-factly.  How would he write? He discovered that a lifetime of lecturing -- often without notes and in complete sentences and full paragraphs -- had trained him to think out loud.  He can now, "with a bit of mental preparation," dictate "an essay or an intellectually thoughtful e-mail."  Unable to jot down ideas on a yellow pad, Judt has taught himself elaborate memorization schemes of the sort described by the Yale historian Jonathan D. Spence in his 1984 book, The Memory Palace of Matteo Ricci.  Like Ricci, a 16th century Jesuit missionary to China, Judt imagines structures in his head where he can store his thoughts and ideas.  The basic principle: Picture entering a large house; turn left and there is a room with shelves and tables; leave a memory on each surface until the [room] fills.  Now head down the hall into another room.  To retrieve your memories, to reconstruct a lecture or recall the content and structure of an article, you re-enter the building and follow the same path, which should trigger the ideas you left behind.

"It works," Judt says.  In fact, he tells me, his mental acuity has grown stronger over the past year.  He compares his situation to that of a blind person with uniquely sensitive ears, or of a deaf person with extraordinary eyesight.  "I knew it to be theoretically true that when you are deprived of everything else, the thing you are not deprived of gets better," he says.  "But it has been very odd to experience that in practice."  After a moment, he goes on: "I'm a 61-year-old guy, I'm not as sharp as I was when I was 51.  But the things I could do last year I can do better this year."  ("The Trials of Tony Judt", by E.R. Goldstein, Chronicle of Higher Education, 01/15/2010).

The MofL also plays a role in The Guest Lecture, a novel by Martin Riker (2023).  In the novel, Abigail, the protagonist and narrator, rehearses the lecture she is going to give the next day on the economist John Maynard Keynes.  She has associated each of her points with a room in her house, and she rehearses by mentally walking through it -- accompanied by a sort of spirit of Keynes herself, who tries to keep her focused when she gets distracted by her personal and professional difficulties.

Evaluating Mnemotechnology

These sorts of mnemonic devices do work, but there is something odd about them -- which is that they don't really circumvent the need to engage in rote memorizing.  Even Cicero didn't think that the art was good for remembering "words", and medical students still have to memorize the bones of the upper limb, stages of cell division, and upper cranial nerves.  As Roediger has shown in an empirical analysis of the MofL, the ancient art of memory is really only good for preserving things in order.  And besides, there is no "art" for remembering the "places" to which "images" have been consigned: these have to be committed to memory by rote.  And it's not at all clear that the art is particularly efficient.  Consider once more Cicero's example of the court case, where the mnemonic consumes more words than the "thing" to be remembered.

Or consider this case, from Giordano Bruno, as paraphrased by , and described by Joan Acocella in a review of Ingrid Rowland's Giordana Bruno: Philosopher/Heretic ("The Forbidden World", New Yorker, 08/25/08).

Closer to Bruno's time, a Catalan mystic named Ramon Llull had refined the method [of artificial memory], imagining memory as a system of concentric wheels.  Bruno adopted Llull's schema and enlarged it.  Rowland bravely tries to summarize the methods he developed.  One involves storing words by their syllables, she says: "the first syllable as an 'agent' who is a mythological figure (the Egyptian Apis bull, Apollo, the witch Circe); the second syllable as an action (sailing, on the carpet, broken); the third syllable as an adjective (ignored, blind, at leisure); the fourth as an associated object (shell, serpent, fetters); the fifth as a 'circumstance' (a woman dressed in pearls, a man riding a sea monster)".  Given thee rules, Bruno described to his readers how, for example, one would remember the word numero, "number".  In Rowland's paraphrase:

The "NU"... is the Apis bull, "ME" is "on the carpet", "RO" is "neglected"....  Bruno, clearly influenced by Ramon Llull, advises envisioning these stored sets of syllables and their imagery on concentric wheels, each with thirty compartments corresponding to the various combinations of letters.  The outermost wheel in the system stores the agents (or first syllables of words), the second wheel stores the actions (or second syllables), the third wheel stores adjectives (or third syllables), and so on inward to the fifth wheel.  A single sentence thus becomes a pageant of mythological characters set n strange places, engaged in strange actions in strange company.

Strange indeed.  How much easier just to remember to say numero.

In the final analysis, we have to agree with one of Matteo Ricci's pupils, who noted that "It takes a powerful memory to use a technique like that".  (Or words to this effect.  I haven't been able to find the quote in Spence's book, but my source is the poet Deborah Larsen, author of Stitching Porcelain, a book of poems inspired by Ricci's story.)

The art of memory does illustrate some important principles which we'll discuss later in this course:

But in the final analysis, Yates is probably right: It's unimaginable to us that anyone ever used these techniques.  And, in fact, it may be that nobody ever did, really, use them -- which explains why the instruction manuals are so vague.  They're of no use to us now, in the age of palmtop computers, except perhaps as vehicles for ostentatious display at cocktail parties (a point of criticism raised in the time of Plato and Aristotle).

Now, the art of memory is best studied not so much as a technology for artificial memory, but rather for the insight that they offer into the worlds of those who devised this technology in the first place.

The Philosophy of Memory

Setting aside the art of memory, ancient philosophers actually had a great deal to say about memory -- philosophical analyses that form a backdrop to the scientific research that began to emerge in the late 19^th century.

There are three key figures in this story, who together exemplify the Golden Age of Greek philosophy -- whose guiding principle, laid down by Socrates (470-399 BCE), was that "the unexamined life is not worth living".  

Together, Socrates, Plato, and Aristotle set the agenda for all subsequent Western philosophy.

Plato (and Socrates)

Plato's theory of knowledge is based on his notion of the Forms, ideal objects that exist outside space and time.  For Plato, these ideas are the only reality.  The particular objects we experience are only shadows of the Forms, which share the essence of the Forms to a greater or lesser degree.  The theory of Forms has implications for learning and memory.  Valid knowledge does not come through the senses, because sense perception can only give us knowledge of appearances, and appearances may be misleading.  

In Plato's Allegory of the Cave (presented in the Republic), people are likened to prisoners in a cave, who experience only the shadows thrown on the walls; the reality behind the shadows is quite different.

So how do we get knowledge of the Forms?  Plato holds that this knowledge is innate: in a sense, we have always known them -- we just didn't know that we knew them.  His proof is in the Meno, an early "Socratic" dialog, where Socrates elicits knowledge of geometry from an ignorant slave boy.  The boy has never been taught any geometry, and in his inquiry Socrates carefully avoids explaining anything to him -- he merely asks questions.  Therefore, the boy must have known the answers all along.  For Plato (and for Socrates), learning is anamnesis, or remembering; the role of the teacher is to get the pupil to "remember" what he has always known.  This is the core of the metaphor of education, from e duco (L., "to lead out of").

Implicit in Plato's theory of knowledge is the later concept of metamemory, or knowledge about memory.  Obviously, we are not simultaneously aware of everything that we know.  But we do have some sense of what we know, and what we do not.  This gives rise to the sense of a "feeling of knowing", which in turn motivates search through memory for available knowledge.  

Modern experimental psychology offers a number of demonstrations of metamemory:

Returning to Plato, of course there is still a role for experience, and thus for sense perception.    People have experiences, which they subsequently remember, and they are led by experience to awareness of things they've always known.  This knowledge is inscribed in memory as well, and is available to subsequent reincarnations (Plato had a theory of reincarnation).  In any event, for Plato the primary problem of knowledge is not how to get it in memory (because it's always been there); the problem is how to get it out.

Plato thus offers us three metaphors for memory, which are precursors to the stage theory described earlier.

In his dialogue Phaedrus, Plato actually criticized the invention of writing, because it made memory unnecessary, and would induce forgetfulness.

Aristotle

Aristotle differed from Plato on the validity of sense-perception.  For Aristotle, the only world we can know is the one we actually live in, and this is the world we should be thinking about.  Because there is no way to validate assertions about any other world, we might just as well not talk about it.  For all practical purposes, just as the empiricists were to argue roughly 2000 years later, knowledge comes to us from experience -- either through sense-perception, or through reflection on our sensory experiences.

Aristotle's theory of memory is expressed in De Memoria et Reminiscentia, which classifies memory as a special state of perception or conception (thinking), conditioned by a lapse of time.  Thus, what we remember are past experiences of perceiving or thinking.  This implies that we must be able to perceive time in order to remember.  Those animals which can't perceive time can't remember either.  And the mechanism by which we perceive time is identical with the mechanism by which we remember.  The implication of all this is a distinction between remembering and knowing: we remember the past (as in retrieval from episodic memory), but we know in the present (as in retrieval from semantic or procedural memory).

For Aristotle, memory belongs to the faculty of sense-perception: What we remember are prior presentations to ours senses; Memories, like percepts, are re-presentations of experience.

This view lies at the root of the familiar distinction between "lower" and "higher" mental processes.  The "lower" mental processes -- sensation and perception, learning and memory -- are tightly bound to sensory experience.  By contrast, the "higher" mental processes -- thinking, problem-solving, and language -- are expressions of pure intellect, freed from sensory experience.  "Lower" animals can have memory, just as perception is universal; but they lack a capacity for intellect.  Of course, we now understand that the distinction between "lower" and "higher" mental processes is bogus.  This is because we now understand that perception and memory critically depend on judgment, reasoning, problem-solving, and decision-making.  

Just as perception is universal, so lower animals can have memory -- but they lack a capacity for intellect.  But Aristotle also makes a distinction between memory and recollection.  Recollection is the act of bring memories to mind -- or, in modern terms, gaining access to memories available in storage.  Based on a principle of association, Aristotle believed that recollection began with a cue that is associated with that which one wants to remember.  The difference between recollection and relearning is that recollection involves an act of one's own will, whereas relearning is imposed from outside.  

While nonhuman animals have memory, only man has the capacity for recollection.  Recollection has the qualities of an inference, beginning with an inference about the past, and proceeding as a kind of investigation.  Thus, while memory classifies as a "lower" mental process, recollection is a "higher" mental process.  Here Aristotle anticipates the later views of Bartlett, that judgment, inference, and decision-making are crucial to remembering.

Together, Plato and Aristotle set the terms of a great tension in Western philosophy.

Augustine

Within early Christian Europe, the Platonic tradition was preserved by St. Augustine (354-430 CE), an early Church Father who lived in North Africa.  Before his conversion to Christianity, Augustine had been a pagan teacher of rhetoric, so he was already aware of the ancient art of memory.  After his conversion, he attempted a fusion of Christianity with Platonism.  In the process, he promoted Plato's view of memory.

For Augustine, memory retains knowledge derived from sense experience -- images of things that one has encountered.  But memory also retains knowledge of abstract concepts, or the things themselves.  Here, perhaps, we have an early version of the distinction between episodic and semantic memory.

Augustine also offers an extended meditation on some paradoxes of memory:

Scholastic Thought

After the Sack of Rome in the 5th century CE, the fall of the Roman Empire brought with it the destruction of libraries and other institutions of learning.  Greek philosophy, including the philosophy of Plato and Aristotle, was mostly lost -- and the "Dark Ages" ensued.  However, Greek philosophy was preserved by Arab and Irish scholars, who were both geographically immune from the attacks of the Vandals.  Later contact between Christians and Muslims set the stage for the rediscovery of Greek philosophy.

And with that rediscovery, this time, it was Aristotle's turn to be fused with Christianity, in the Scholastic thought of Albertus Magnus (1200-1280) and Thomas Aquinas (1224?-1274).  And with this fusion both the ancient art of memory and Aristotle's theory of memory were revived.

The Renaissance and the Enlightenment

Moving right along, the Renaissance of Greek learning in the 15th century placed less emphasis on knowledge through revelation from God, and greater emphasis on empirical and critical inquiry, as Renaissance thinkers sought to understand themselves and their place in the universe.  This project is epitomized by Rene Descartes (1596-1650), whose philosophy began from a position of extreme skepticism.  Descartes set aside the authority of both the Scholastics and the Greeks, and attempted to think in an entirely new way.  In the course of this, Descartes developed a new approach to epistemology -- the theory of knowledge.

Still, in many ways Enlightenment thought recapitulated the old dispute between Plato and Aristotle.

This page last modified 11/06/2023.