1. No one really understands what a book is. Not so long ago we thought we did: a book was an easily recognizable object, sheets of paper gathered and bound on one side, usually with glue, sometimes with thread. Some kind of cover to protect the pages, made of cloth or thick paper or shiny card stock or (in exalted circumstances) leather was normal, though if the cover happened to be torn away the book was still recognizably a book.

Alas, we are already in difficulties. By the definition I have just provided, the copy of Books & Culture you are holding is a book. But if I tried to retrieve such a copy by asking "Would you please hand me that book?" I would receive only a puzzled look. Everyone I know above the age of eight or so has a reasonably clear and accurate understanding of the conventions that separate books and magazines, even though it would be perfectly possible to publish a novel in the shape of Books & Culture or Books & Culture in the shape of a mass-market paperback or a leather-and-India-paper Bible.

Moreover, some of you may have stuttered for a moment in your reading when I mentioned "the copy you are holding" because you are not holding anything—you're reading on a computer screen—or because what you are holding is not made of paper but is rather an iPad. And if you are holding an iPad, and you shift from its web browser to the iBooks app or the Kindle app, and commence reading Great Expectations, are you then holding a book? What if you have an iPod or other media player clipped to your shirt and are listening to an audio version of Dickens' novel? Where is the book then?

Ah, our postmodern world, deconstructing the everyday: "All that is solid melts into air," as Marx and Engels wrote in The Communist Manifesto, which, not incidentally, is usually called a book but has also been called a pamphlet—is that a particular kind of book or something different from and smaller than a book?—and in any case was published simultaneously as a bound volume and as a serial in a German-language London newspaper. (If it had only been published in the latter format, would it still be a book, or a pamphlet?)

Given that The Communist Manifesto appeared in 1848, we may discern that our problems with definition don't have anything, really, to do with postmodernism. Overleaping a great deal of history, we might even go back to the prophet Jeremiah: "Jeremiah wrote in a book all the evil that should come upon Babylon, even all these words that are written against Babylon" (51:60). So reads the Authorized Version, and some modern translations say much the same; however, the New International Version and others prefer "Jeremiah had written on a scroll …." The same Hebrew word, sefer, turns up repeatedly in the Old Testament: "This is the book of the generations of Adam" (Gen. 5:1). "Then Hilkiah answered and said to Shaphan the secretary, 'I have found the Book of the Law in the house of the LORD'" (2 Chron. 34:15). "Oh that my words were written! Oh that they were inscribed in a book!" (Job 19:23). "You have kept count of my tossings; put my tears in your bottle. Are they not in your book?" (Ps. 56:8). Thus also we refer to the "books" of the Bible: most of them filled one scroll each.

Matthew has the same thing in mind when he begins his Gospel by identifying it as "the book of the genealogy of Jesus Christ, the son of David, the son of Abraham." The word he uses here is biblos, which will be immediately familiar. Whether you translate sefer and biblos as "book" or "scroll" depends, I suppose, on whether you think their prime meaning involves a particular technology of writing or rather a complex and hard-to-specify concept of what I shall call expressive fullness. In this latter sense a book is a textual document of considerable length—would that we could define "considerable," but we cannot, we dare not—that expresses a long story or complex sequence of thoughts wholly, completely. A short poem may also be complete—perfectly so, in fact—but it does not carry with it the sense of fullness that a book does.

When people say "I want to read [or write] a good book," they want to encounter a writer's mind, or just an exciting story, and they want that encounter to be extensive, to require at least time and minimal attentiveness from its readers, and to achieve a satisfying finish. I expect that people expressed similar desires when the reigning technology was the scroll, or even in pre-literate cultures: great bards like Homer could toss off brief tales like bagatelles, but their fame was built on the mastery of narratives of epic scope. People have continued to desire such big works over these many centuries in which the reigning technology is the bound volume—properly called a codex—and will express them still even if codices become as archaic as scrolls are to us, and screens rule the world of text.

So no one understands what a book is in large part because we operate with very different definitions jostling against one another in our minds; and also because, even if we think only in terms of the technologies of writing, these technologies have changed a great deal throughout the history of writing and are changing at an unprecedentedly rapid pace today.

2. These reflections should help us come to terms with both the achievement and the limitations of The Oxford Companion to the Book, a beautiful two-volume set edited by Michael Suarez, SJ, and H. R. Woudhuysen. The set is divided into two major sections, which do not quite correspond to the volumes: a set of 48 largely historical essays takes up most of the first volume, while the rest of that volume and the whole of the second yields a dictionary. There you may find entries on printing techniques, instruments used in printing (press stones, mallets), certain famous books (the Codex Sinaiticus, Scouting for Boys), publishers, book collectors, legal terms, and a great many other things. One of my favorite entries, so far, is that for one Samuel Paterson (1728-1802), identified as "an unsuccessful English bookseller (because, he said, he preferred reading his books to selling them)." Surely this entry strikes a blow against our culture's worship of success, and, moreover, gives proper acknowledgement of the fate that has always befallen the greater part of the bookselling trade.

The first term I looked up in this dictionary was "book." Sure enough, there it is—twice. The second definition simply concerns what we mean when we say, for instance, "Paradise Lost is a poem in twelve books"; it's the first one that's interesting:

book (1) A word that has long been used interchangeably and variously to signify any of the many kinds of text that have been circulated in written or printed forms, and the material objects through which those words and images are transmitted. The ancestor of the modern word 'book' is used in both senses in Anglo-Saxon documents. This Oxford Companion to the Book is a book in the abstract, non-corporeal sense (and can be thus described in its Internet manifestation), and also in the physical sense of a three-dimensional object in codex format. Books in that first sense have been manifested in the second one in a wide variety of shapes and sizes from ephemera … to elephant folios, using many different materials.

Just as I have tried to distinguish between the book as a particular technology of writing and the book as a concept involving fullness of expression, the author here (David Pearson, one of the Companion's associate editors) makes a distinction between the physical object and … well, what, exactly? "Any of the many kinds of text that have been circulated in written or printed forms"—but that's not very helpful. Pearson seems to be distinguishing in as purely abstract a way as possible the physical book from whatever goes in a physical book. But this raises the question of what does go into a physical book. "Words and images," apparently. But this does not get us very far.

Nor does this definition enable us to make everyday distinctions such as the one I mentioned earlier between the book and the magazine, or, for that matter, between books and letters: that may have been a lengthy missive Mr. Darcy wrote to Elizabeth Bennett, but no one would call it a book. And indeed the Companion's dictionary contains rather lengthy entries on "magazine, origins of" and "letter." It is at this point that one is forced to conclude that The Oxford Companion to the Book is a seriously misleading title: it should rather be called The Oxford Companion to the Written Word, for this is its true subject. (More accurate still would be The Oxford Companion to the Written Word and the Images That Sometimes Accompany It, but one can reasonably ask only so much from a title.) The concept of the book that I tried to articulate in the opening paragraphs of this essay is, as far as I have been able to discover, not referred to in any way in the whole of this reference.

In one sense this may be defensible. We probably need a Companion to the Written Word, in its glorious diversity, more than a Companion to the Book. But someone needs to try to define and, I think, defend the concept of a book as an ambitious, relatively large-scale work of story or argument. For this concept is under some attack these days, by implicit and explicit means. It's certainly true that there are too many books, and that (in the realm of scholarship especially) many ideas that deserve to be treated at essay-length have been unnaturally expanded to fill the space between covers, but it is manifestly not the case that, as some have quite seriously suggested, books can simply be replaced by blog posts, or by blog-like media in which ideas are presented as they develop, are visibly and publicly revised, and are not finished but rather, at some point, abandoned.

To be sure, there are some major intellectual projects that work wonderfully as ongoing and at least partly communal endeavors, intentionally and intrinsically works-in-progress; and in many circumstances blogs and book projects can be delightfully symbiotic. (I have recently completed a book on reading, and I think the book is much better because I tried out many of its key ideas on my blog, Text Patterns.) But many major works of art and thought have benefitted from the long incubations that the practices of book-writing and the book business require. I have never written a novel, but I can testify that the arduous process of writing my way from the beginning of a sequence of ideas or events to the end, trying to keep the thread in mind at all times, indulging in digressions but striving to control them, achieving new insights or interpretations that require backtracking and rewriting, thinking all the time about how to make this path of thought attractive and meaningful to a reader, finally coming to the best arrangement and presentation of such a path as one can manage, and only then making the work public—this arduous process, I say, is immensely and wonderfully challenging. And it is nearly as challenging and in some ways just as rewarding for me to read the work of someone who has achieved these goals far better than I ever will.

Moreover, this model of what a book is has been intrinsic to the development of Western culture from Plato's Republic to the major monuments of recent scholarship and literature (despite the anticipatory denunciations of Socrates). It is not incidental to the power and influence of works as varied as Augustine's Confessions, Descartes' Meditations on First Philosophy, Mary Wollstonecraft's Vindication of the Rights of Women, Gabriel García Márquez' One Hundred Years of Solitude, and John Rawls' A Theory of Justice that they are all books and not some other kind of thing. These reflections manifestly are not the concern of The Oxford Companion to the Book, but they need to be somebody's concern.

3. These bleats made, acknowledged, and dismissed, it is time to go on to say that The Oxford Companion to the Book contains many excellent essays, and that there is something of interest on almost every page. Some of the essays compress vast tracts of book history into a few pages with élan and accuracy: I think especially of Craig Kallendorf's "The Ancient Book," Cristina Dondi's "The European Printing Revolution," and Daven Christopher Chamberlain's "Paper." There are some surprise entries: I had never really thought about "The Manuscript after the Coming of Print," but Harold Love shows that there is much to be said about "the process by which the handwritten word perpetuated its ancient traditions in the face of competition from the press." I particularly enjoyed Brian Cummings' essay "The Book as Symbol," though I wish it had been several times longer than it is. (Three pages!)

The Companion contains 32 historical essays, and while these are uniformly valuable, the many subdivisions obscure important historical continuities. For instance, the contiguous chapters that cover the history of the book in Eastern Europe—Hungary; the Czech Republic and Slovakia; Poland; the Baltic States; Russia, Ukraine, and Belarus; and the Balkans—all have to account for what the rise of the Soviet Union did to book culture in these lands, and since there are many significant similarities, the narratives feel repetitious when read one after another. It is perhaps inappropriate to ask that a reference work with many authors would pull together a single historical narrative—the sort of thing that's typically the work of monographs—but more coherence would help the cause here. Because these are general essays, they provide the most bang-for-the-buck to readers who are largely or wholly ignorant of the subject under consideration: I knew next to nothing about "The History of the Book in the Muslim World" before I read Geoffrey Roper's account, and now I know how to pursue the subject further. It's a fascinating one.

The second essay in the Companion, after an overview of writing systems, is Carl Olson's "The Sacred Book." It's one of the longer essays, as it should be, and covers five traditions—Judaism, Christianity, Islam, Hinduism, and Buddhism. Olson's expertise is in the latter two traditions, which probably explains why they cover more than half of the essay. I lack the expertise to judge the accuracy of those sections, but the ones on Judaism and Christianity are disappointing, devoted largely to summarizing the models of textual criticism that reigned half a century ago. (Olson includes Robert Alter's The Art of Biblical Narrative in his bibliography, but takes no notice of the damage Alter's work, and that of many other scholars, has done to the Documentary Hypothesis, which Olson breezily presents as unchallenged orthodoxy.)

Nor does Olson have much to say about the role the Jewish and especially the Christian scriptures had in shaping the Western world's, and ultimately the entire world's, understanding of the very concept of "book." (Yes, I am returning to this point, but what's the point of having a hobbyhorse if you're not going to ride it?) Scholars have known for a long time now that Christians made the shift from the scroll to the codex long before either pagans or Jews, but there has been no general agreement on why they did so. Certainly there are practical, material reasons to prefer the codex: for instance, because you can write on both sides of a sheet, it is less expensive than scrolls, which leave one side unused. But it's not clear why that would be more important to Christians than to others. The first codices seem to have been small notebooks that Roman politicians and intellectuals liked—they were the Blackberrys of that place and time—and these had the advantage of portability, which Christians may have appreciated as they sought to spread the Good News, through the Gospels and collections of Paul's letters, across the Empire via the Roman roads.

But, though historians of the book seem unwilling to credit it, there may also have been strong theological reasons for Christians to prefer the codex, reasons which became stronger as the technology developed in ways that allowed more and more of the Scriptures to be included in a single volume. Binding together "the little books"—that's what ta biblia means—emphasizes their chronological and contextual relations to one another. It shows them as part of one story. It took a while for the church to create the early great codices—the Sinaiticus and the Vaticanus date from the 4th century—but once they came into existence they offered a clear material testimony to the oneness of Scripture. Surely this was a prime reason why the Emperor Constantine, in 331 (some scholars say earlier), wrote to the great bishop and scholar Eusebius of Caesarea and required him to have made and then send to Rome fifty complete Bibles. The oneness of Scripture testifies to the oneness of the faith.

And that oneness of Scripture becomes key to the symbolic meaning of the book that Brian Cummings writes about in the Companion. One of Dante's final visions, in the last canto of the Paradiso, as he completes his journey through the universe, is of creation as "pages scattered through the universe" but "by love into a single volume bound." And in a glorious passage that Cummings does not quote, from his "Meditation XVII," John Donne affirms that

all mankind is of one author, and is one volume; when one man dies, one chapter is not torn out of the book, but translated into a better language; and every chapter must be so translated; God employs several translators; some pieces are translated by age, some by sickness, some by war, some by justice; but God's hand is in every translation, and his hand shall bind up all our scattered leaves again for that library where every book shall lie open to one another.

It is this image of absolute coherence, of all truth bound firmly together in a single vision, that has for dozens of generations constituted the very ideal of the book. Augustine had it in mind when he wrote The City of God, and Joyce when he wrote Ulysses. It is an image worth holding in our minds.

4. In the essay entitled "The Electronic Book," Eileen Gardiner and Ronald G. Musto introduce their topic thusly: "An electronic book (also e-book, ebook, digital book) is a text- and image-based publication in digital form produced on, published by, and readable on computers or other digital devices …. The e-book is a young medium and its definition is a work in progress, emerging from the history of the print book and evolving technology. In this context it is less useful to consider the book as object—particularly as commercial object—than to view it as cultural practice, with the e-book as one manifestation of this practice." Note here the recognition that book-as-object must often be distinguished from what I have called book-as-concept and they call book-as-practice. Changes in technology force these distinctions upon us.

Note also that this essay makes no mention of Kindles or Nooks or other e-readers, which did not exist when it was written. This absence signifies, because it is increasingly common for reference books to be "published" only in digitally accessible form because reference books tend to be uncomfortably large and because they need to be updated more often than, given the constraints of print publication, is economically or practically feasible. Consider that greatest of all reference books, the Oxford English Dictionary: its first edition took 68 years to complete; the second took a further 32 years; and the third edition is currently in progress and is not expected to be completed until 2037, thanks primarily to explosive growth in world English and the internet's provision of near-total access to it. No wonder Oxford University Press plans for this third edition to have only a digital existence—but given that decision, does it even make sense to think of a "completion date"? Long before the final sections are completed, the earliest-revised onces will be out of date. Surely from now on making the OED will be like painting the Golden Gate Bridge, a continuous process.

But what about my Oxford Companion to the Book? The volumes sitting proudly on my shelf will never mention the Kindle; and errors that reviewers have pointed out will never be corrected. But the online version—mentioned in the Companion's definition of "book," cited earlier—can be corrected at any time. Indeed, some such corrections may already have been made. Moreover, I can read it from anywhere I have a computer, or even a smartphone. Much as I love the crisp, glossy paper and lovely binding of my two-volume set, complete with its slipcover and long silk bookmarks, it's hard not to see the digital version as constituting an improvement in many ways. Even a simple word search on the book's website connects the Companion's vast treasures of information in ways the book cannot, or at least does not. I loved browsing through its substantial dictionary, but was often unsure how to know what is important, or how to find something unless I already knew what or who I was looking for. It pleased me to learn that the Rittenhouse Mills in Pennsylvania was the first paper mill in the United States (founded in 1690), but stumbling across the item was all I could do: it wasn't cross-referenced from the general entry on paper mills. There are cross-references in the dictionary, but they are incomplete and inconsistent, as is almost always the case in works of this kind: cross-referencing is an art, not a science, and anyway a book that cross-references everything quickly becomes unreadable and unusable. Digital reference books largely solve these problems.

Moreover, many of the terms defined in the Companion could really use accompanying illustrations. There are some, but not enough. For instance, it's hard to visualize the various forms of script (miniscule, majuscule, secretary hand) without images. Various printing techniques—lithography, intaglio, relief—could also benefit from images. In a codex the inclusion of many such images would be prohibitively expensive, whereas an online version could accommodate as many of them as are needed. So, given our culture's long association of "the book" with "the codex," it may be historically insensitive or churlish or both for me to prefer the Companion in its online form. Yet, I must admit, prefer it I do—at least, usually.

But enough of these complaints. After all is said, The Oxford Companion to the Book is a delightful companion indeed; it is not and does not claim to be an exhaustive encyclopedia. But I can't help thinking, in this age of digital abundance, that it could become just such a complete reference, with gaps filled and illustrations provided and references crossed. Perhaps I have too much in mind that perfect Book that binds everything together with perfect knowledge and overflowing love. Then again, perhaps it is not possible to have that Book too much in mind.

Alan Jacobs is professor of English at Wheaton College. He is the author most recently of Original Sin: A Cultural History (HarperOne) and Looking Before and After: Testimony and the Christian Life (Eerdmans).

Copyright © 2010 by the author or Christianity Today/Books & Culture magazine.
Click here for reprint information on Books & Culture.