Moonwalking with Simonides

The long history of techniques for improving memory.

Posted Apr 23, 2011

Moonwalking with Einstein has its flaws. It can seem frustratingly indifferent to essential distinctions between different kinds of memory (long-term vs short-term, semantic vs autobiographical, recognition vs recall), which limits its value as a guide to the modern science. The reconstructive nature of autobiographical memory gets a mention, but the author's narrative interest is necessarily focused on his attempts to memorize large piles of playing cards, rather than on the question (which we're teased with at both the beginning and the end of the book) of how our memories make us who we are. There are better books to read on the science of memory [1, 2], but still I came away glad of having joined the coolly humorous and always engaging Foer on his madcap adventure in head-cramming.

I'm going to focus in this post on Foer's account of the technique that underlies most of the mental athlete's repertoire: the method of loci. Perhaps for evolutionary reasons, we are generally better at processing spatial data than we are at computing temporal information (what came where in a sequence). The competitive mnemonists, when they want to remember huge piles of stuff, imagine spaces within which they can stow certain prearranged images:

I was storing the images in a memory palace I knew better than any other, the house in Washington, D.C., that I'd lived in since I was four years old... At the front door, I saw my friend Liz vivisecting a pig (two of hearts, two of diamonds, three of hearts). Just inside, the Incredible Hulk rode a stationary bike while a pair of oversize, loopy earrings weighed down his earlobes (three of clubs, seven of diamonds, jack of spades)... (Moonwalking with Einstein, p. 248)

The building of such 'memory palaces' was a medieval preoccupation. Foer's approach draws heavily on Frances Yates' account of the medieval craft of memory, as set out in her 1966 classic, The Art of Memory. Our ancestors, Yates argued, needed ways of storing information in a pretechnological world which couldn't rely on the various modes of 'external memory' (initially printing, but now computers, digital media, and the world-wide web) upon which we now depend. Stashing images inside memory palaces was a way of doing the elaborative encoding [1] necessary for successful remembering.

The medieval scholar Mary Carruthers has criticized Yates' account for being too focused on static, rote remembering. In her extraordinary book, The Craft of Thought: Meditation, rhetoric, and the making of images, 400–1200[3], Carruthers argues that the real purpose of medieval memory techniques was to be recombinative rather than recollective: to create new thoughts rather than to dredge up old ones. In a prescient 2005 Nature essay [4], Carruthers foresaw the growth of interest in the links between episodic memory and future-oriented thinking [5]. Foer mentions Carruthers' work on the link between memory and creativity, but it loses its impact as a result of being embedded in a less than flattering portrait of the memory guru Tony Buzan.

Carruthers also asks us to reconsider the myth of Simonides of Ceos, the supposed inventor of the method of loci. Here's how Foer tells the story:

That proud tradition began, at least according to legend, in the fifth century B.C. with the poet Simonides of Ceos standing in the rubble of the great banquet hall collapse in Thessaly. As the poet closed his eyes and reconstructed the crumbled building in his imagination, he had an extraordinary realization: He remembered where each of the guests at the ill-fated dinner had been sitting. Even though he had made no conscious effort to memorize the layout of the room, it had nevertheless left a durable impression on his memory. From that simple observation, Simonides reputedly invented a technique that would form the basis of what came to be known as the art of memory... Just about anything that could be imagined, he reckoned, could be imprinted upon one's memory, and kept in good order, simply by engaging one's spatial memory in the act of remembering. (Moonwalking with Einstein, pp. 93-94)

The Simonides story is a myth, and no one (least of all Foer) want to treat it as a literal description of events. What Carruthers does, though, is invite us to see the Simonides story as an example of the very technique that it is propounding. Make your images memorable, the mental athletes tell themselves. (Foer tends to make his images particularly obscene, which is another reason for liking him as a narrator.) Carruthers points out that such stories are absurd,

[b]ut that quality of absurdity is exactly what makes them memorable, and therefore valuable to cultures in which people relied on their memory to keep all that they knew, cultures which also recognized the essential role of memory in human cognition. (The Craft of Thought, p. 28)

If Carruthers is right, no one really believed that Simonides saw the roof fall in as he was supposed to. Rather, some ancient (and possibly rather boring) magus came up with the idea of the method of loci and wanted to couch it in a story that was as striking, and thus memorable, as possible. The story of the ill-starred feast was not a myth of accidental origin so much as a deliberate invention, designed to educate new mnemonists in an effective technique for remembering. It is not about poets or feasts or architectural crushings at all: it is about learning the principles of a craft.

Carruthers applies the same reasoning to the weird stories in which the knowledge of medieval alchemists was encoded:

The strangeness of these stories has been attributed solely to desires to keep knowledge among an elite guild. But the impulse to clothe difficult technical knowledge in such tales is also accounted for by the need to remember processes exactly: the alchemists' tales are a variety of techno-babble, but a jargon consciously made more memorable than our own. It is a principle of mnemotechnics that we remember particularly vividly and precisely things that are odd and emotionally striking, rather than those that are commonplace. Sex and violence, strangeness and exaggeration, are especially powerful for mnemonic purposes. (The Craft of Thought, pp. 28-29)

The stories are strange, then, because strange stories stick in the memory. In naming the constellations, the ancients did not really believe that this scattering of stars really looked like a dog, and this one a hunter or a bear. They looked for striking images so that those who needed to know the constellations would remember the patterns:

The purpose of organizing stars into constellation patterns is not 'representation', but to aid human beings, needing to find various stars, to locate them by means of a recognizable pattern retrieved immediately and securely from their own memories. Constellations are mnemotechnical tools. (The Craft of Thought, p. 26)

No one will ever know the truth of how these 'inventory fables' came about. But in Carruthers' view, the ancients were much smarter and deliberate in their myth-making than we usually give them credit for. They recognized that our memories are fallible, and that striking images stick in the mind better than any others. This applies too to the myth of Simonides, a fable that memorably celebrates its own wisdom.

1 Schacter, D. L. (1997). Searching for memory: The brain, the mind, and the past. Basic Books.

2 Schacter, D. L. (2002). The Seven Sins of Memory: How the mind forgets and remembers. Mariner Books.

3 Carruthers, M. (1998). The Craft of Thought: Meditation, rhetoric, and the making of images, 400–1200. Cambridge University Press.

4 Dudai, Y., & Carruthers, M. (2005). The Janus face of Mnemosyne. Nature, 434(7033), 567. doi:10.1038/434567a

5 Schacter, D. L., Addis, D. R., & Buckner, R. L. (2007). Remembering the past to imagine the future: the prospective brain. Nat Rev Neurosci, 8(9), 657-661. doi:10.1038/nrn2213