History began in a library. The earliest known collections of writing were found near the Sumerian city of Nippur in modern-day Iraq, and for many scholars these earliest libraries mark the beginning of history, and our emergence from our former state of ‘pre-history’. This odd division implies our provisional existence only became fully realised when it became properly recorded; we came completely, traceably into being with our libraries. How like a group of academics, you might think, to decide that history began when we became able to reference it. 

Still, there is something seductive about this idea, despite its tang of musty scholarship. Unlike the spoken word, the written word can be durable, and the longevity of a text can allow us to communicate across generations or even millennia; the written word can be an historical artefact, as well as the agent that communicates history. Its origins are fixed in time even as it travels through the years to be read by new eyes. Our book repositories make our accumulations of written knowledge navigable and recoverable, preserved for our future use; in this, the library acts as a kind of collective memory, containing our past and allowing us to recall it as we wish. 

Memory and history are closely linked: each describes what we tell ourselves about our past, whether personally or collectively, and both are subjective, although we might wish them to be otherwise. Subjectivity is not a term that comes immediately to mind when thinking of a library. A collection of books contains many voices, despite the library’s reputation for hushed silence, and they are arranged according to a system that seeks to make every one accessible, mapped by its classmark. A library operates according to the reassuringly orderly Ciceronian model of the memory: an organised structure that offers the power of recollection through careful placement. Cicero argued that ‘it is chiefly order that gives distinctiveness to memory’ and he advocated systematising what we want to remember within an imagined location, the famous ‘mind palace’, which allows us to recall what we need by the inward navigation of our orderly collection. Librarians everywhere might nod in approval. A library defines periods of time, demarcates the limits of different areas of endeavour, and stabilises seemingly random material into a system that can be navigated because its points are fixed. We can find what we need because we know where it is placed – as with Cicero’s memories, so with his books. 

This kind of mapping requires a looking-back at what we have accumulated, but it is not simply an examination of the view from the mountaintop: it helps to create that view. A library is not an inert repository, contentedly replete with all our past knowings; it is an active influence on the ways in which our knowledge develops, and on the things that we forget. Several weeks ago I had the dust of hundreds of books itching my nose as I participated in the annual stock-take that marks the regular period of a library’s renewal, the shedding of its skin before it moves into the next year. As my nails collected grime, my clothes and skin collected fragments of disintegrating cellophane covers, and my muscles collected the slow ache of books lifted out and systematically replaced, I was immersed in the sheer physicality of those words, the weight of that accreted knowledge. My movements were all conditioned by and contained within the library’s material structure, but through this work I began to understand the library as an organism. Space was made on the shelves for new arrivals, and the dead wood shed for this new growth was either placed in storage or stacked on a large table to be sold, having exceeded its shelf life. Any book not attractive to the private purchaser was to be given or thrown away: waste material excreted as the body heaved onwards. A library might represent our collective memory, a place for reflection, but this activity brought home to me its ruthless momentum. Nothing is immune from our forgetting, and although it might appear to be organised as a stable structure, the library is always in transition, its shape always shifting. Like memory, a library is selective, and its acquisitions and omissions reflect its identity and purpose. 

Cicero conceived of memory as a solid building but it can be much less reliable than that; it can blur, rebuild, or erase the past. In his essay ‘Invention, Memory, and Place’, Edward Said describes memory as a search for roots, a journey towards its own starting point that frequently veers away from the path already trodden to invent a new and unfamiliar way back. For Said, memory is the past remade and reshaped, and it is deeply involved with the formation of identity. He challenges Cicero’s model of memory as a structure, with its connotations of stability and objectivity. Instead, he argues that memory is always in transition, an inventive force that recovers and represents what has been according to what we wish ourselves to be. Said believes that we seek to find in memory a narrative of ourselves, and a sense of our place in the world.

Both of these models of memory are concerned with location, but Cicero’s memory palace offers the stability of a fixed and accessible structure, while Said presents memory as an ongoing search for home, an attempt to navigate through the world around us. It might seem that a library is immoveably, undeniably rooted, resisting any comparison with Said’s mutable and restless memory. The library in which I spend most time is the University Library in Cambridge, a building that towers over its surroundings, swallowing you into its chilly corridors as you search its crowded shelves and crammed windowsills for the volume you need. This is a copyright library, its collection swelling as new material leaves the country’s presses, and so it seeks to be comprehensive, to house the entirety of a nation’s published material. The readers of this vast archive are primarily students and academics who take a systematic approach to the library’s contents, looking for the connections between information and arguments, appreciating the strict organisation and historical sweep of the material within. Its members stream towards this library like bees towards their hive, their books stored in one building, fixed in an orderly system, the still objects of thousands of searching inquiries. In many respects, this is the closest embodiment of the Ciceronian model of memory that it’s possible for a library to become. 

Other types of library drift from this comforting Roman order, with a more quixotic appeal that lies closer to Said’s vision. If you aren’t a student, you’re most likely to see the word ‘library’ prefixed by ‘iTunes’, a kind of technologically-powered seventeenth-century notion of the library as a private collection, serving one’s own needs and a reflection of one’s own taste. Said’s suggestion that memory is a necessary part of identity creation seems to find its fullest expression in such libraries. Their content is not comprehensive but a fulfilment of personal desires, and the organisation of the material is an important and individuating aspect of its consumption: instead of the easily navigated and commonly understood system of the reference library, iTunes allows its users to invent their own playlists, while the Shuffle option abandons order completely to embrace random juxtapositions and the attraction of the unexpected. Such a library is portable; organised digitally and contained in a device that can be carried in the pocket, it travels with its owner. This library is on the move.

Walking through a small town or village in the British countryside, you might come across a collection of books housed by a disused red phonebox. This quaint phenomenon almost enacts a reversal of the contemporary gadgetry of iTunes: while telephones have shrunk, themselves becoming portable and individually owned, a library occupies the shell left behind. These tiny red collections do not aim for comprehensiveness, however, nor are they systematised and orderly. A mish-mash of donated books, they mostly consist of popular literature such as Dick Francis, Danielle Steele and the ubiquitous E. L. James, and they reflect a desire for shared reading that doesn’t fulfil the specialised needs of academic researchers or the practical requirements of students. The material demonstrates not the personal taste of an individual, but the disorder of the communal. These libraries indicate the joy of books as an amusement, a dog-eared copy of a crime novel you’ve never seen before, containing a stranger’s shopping list and stained by someone else’s coffee, which you read on your sofa while curled under a blanket; the unexpected discovery of a few happy hours. The sheer pleasure in books as objects, and in reading as a pastime, is told to us by these libraries.

Time passed, and not time past: such collections are not archives to be navigated systematically, but repositories for a chance moment. The red phonebox elicits nostalgia, a quaint preservation of the past also seen in the resolute physicality of the books themselves, staunchly persistent in the age of the e-reader. However, these collections are opportunities for present enjoyment. Their initial discovery is always a welcome surprise and their charm lies partly in the fact that they’re disguised as something else; the decision to visit a library building does not exist here. Reading as happenstance is perhaps most celebrated by a practice that appears to do away altogether with the library, both as building and as collection: book swapping. This phenomenon goes beyond a simple exchange of favourite novels between friends. Organised by websites such as, it requires you to leave a volume of your choice in a public place – on a park bench, in a train, on the table of a waiting room – with a label inside encouraging the recipient to pass the book on when they’ve read it. This indulgence of chance appears to be at the farthest remove from a carefully organised repository, purposefully navigated. The book is a piece of detritus, happened upon rather than housed, and it is passed from person to person, not visited in its designated place on the library shelves. 

Nevertheless, the book is mapped. Its initial location is logged online by its first owner, and when the book is finally let go by its next recipient they’re asked to register its next temporary resting-place on the website. Each reader can therefore follow the volume’s progress as it journeys from owner to owner, and the pleasure of the activity is not simply in its chance elements, but in its tracking. The book becomes a route on a map, one journey among many marked on the website: itself a kind of library, its stock in constant motion.

Location and legibility are connected. Both Cicero and Said’s models of memory involve mapping of different kinds: Cicero’s fixed structure allows us to recall by travelling through its defined points, while Said suggests that we use memory to help us carve out our route through the world. Memory is a process of meaning-making, of understanding the connections between things. So it is with reading. Libraries enact this process on a grand scale, and they bear witness to its attraction. They hold our books in relation to each other, whether in the massive system of the research library or the chance meanderings of book swapping, and they draw us in to their networks, housed on wooden shelving, in a country phonebox, or on a website. Through its various mutations the library persists, remaking itself along familiar lines.