Inigo Montoya

It’s good to have a foil. Nicholas Carr seems to be one of mine. He often arrives at dramatic statements I can’t quite stomach, ones that actually get worse the more I think about them. His “the Internet rewires your brain” string of hits never really seemed plausible, especially coming from a fellow blogger and heavy Internet user. Was he trying to modify himself? Perhaps in this modified state, he’s latched on to the lack of “fixity” in digital text.

First, Mr. Carr, you’re late to the party. We’ve been dealing with this for a long time, from learning new citation schemes for Web sites to learning how to cite the Kindle.

Second, I will invoke Inigo Montoya from The Princess Bride in saying that “I do not think that word means what you think it means.”

Here’s a quote from Carr’s recent Wall Street Journal article which efficiently captures his concern about a loss of fixity:

Once digitized, a page of words loses its fixity. It can change every time it’s refreshed on a screen. A book page turns into something like a Web page, able to be revised endlessly after its initial uploading. There’s no technological constraint on perpetual editing, and the cost of altering digital text is basically zero. As electronic books push paper ones aside, movable type seems fated to be replaced by movable text.

It’s actually quite an interesting read, because Carr gives a balanced assessment of the pros and cons of fixity vs. fluidity. But, showman that he is, he seems to want to land on the scariest note he can, an A-flat of doom, despite his knowledgeable and informed summation of the benefits fluidity might provide, including:

  • reference and guide books that never go out of date
  • ease in correcting errors or updating references
  • instruction manuals with current information

To achieve his tone of concern, Carr turns to John Updike, who celebrated the “edges” of books, something Carr interprets as indelibility or permanence. But have books really been that fixed? Even Updike himself would revise stories from magazines as they moved into his books, changing a sentence here or there after the serialization of a few chapters. Between the hardcover and paperback, typos were fixed, and pages reflowed for the new trim size. In at least one case, he completely shifted the focus of major narrative events. The difference between a first-edition hardcover of “Rabbit Redux” and the fourth edition paperback could be significant in the ways Carr is considering — text reflowing, new text introduced, things fixed, things changed.

Currently, there’s a controversy about Mitt Romney’s book and the revision of it prior to his presidential run. Yet, in reporting on these changes, reporters turned to electronic editions of the books — first, the 2010 Kindle edition, then a version posted by a political adversary on Scribd. The reporters also turned to the hardcover edition, but it matched the Kindle edition in every substantive way. Electronic books are fixed in their own ways, and by factors beyond the medium in which they’re published.

So what is fixity then? Obviously, it’s not format fixity, because printed materials have routinely adopted different formats through their lifetimes, moving from hardcover to paperback to large-type editions to braille editions to audio books.

Nor is it textual integrity, as Updike and Romney and others demonstrate. As a book person, I have a few Bibles around the house. They don’t all match — one is written in more modern English, one is King James, and one is for kids. Which is the real Bible? Is there a fixed Bible from which they all derive? (And don’t forget your apocrypha when answering that question.)

Searching on for “revised and expanded” gives me 3,451 results. Are these books now unfixed? Were they ever fixed? How any more so than electronic text? In archives? One could argue that there’s a commercial incentive for changing a book, print or otherwise.

It seems the notion of “fixity” is much more specific than Carr might have in mind — almost to the level of the individual artifact and its reproductions.

Carr is right on one level — it’s probably not great to revise materials without notifying people you’ve done it, even when you have benevolent intentions. But there is a spectrum here, between updating an index’s pagination for the paperback edition to correcting three typos to changing substantive ideas or text. But print didn’t guarantee fixity, and digital doesn’t really provide that much worrisome latitude beyond what already exists. This is something Wikipedia handles really well. In fact, the fixity of print makes changes very hard to detect, while the right deployment of online tracking tools makes changes obvious in some cases, and at the very least discoverable with relative ease.

“Fixity” is a sexy word. Carr has a nose for marketing ideas, there’s no doubt. I just wish he’d market better ones.

Enhanced by Zemanta
Kent Anderson

Kent Anderson

Kent Anderson is the CEO of RedLink and RedLink Network, a past-President of SSP, and the founder of the Scholarly Kitchen. He has worked as Publisher at AAAS/Science, CEO/Publisher of JBJS, Inc., a publishing executive at the Massachusetts Medical Society, Publishing Director of the New England Journal of Medicine, and Director of Medical Journals at the American Academy of Pediatrics. Opinions on social media or blogs are his own.

View All Posts by Kent Anderson


18 Thoughts on "The Mirage of Fixity — Selling an Idea Before Understanding the Concept"

Once again, Kent, you flaunt your inability to see the forest for the trees. Are you really disputing Elizabeth Eisenstein’s argument, in The Printing Press as an Agent of Change, that one of the major effects of the introduction of the letterpress was the arrival of “typographical fixity” for written works? Of course, revisions and other deliberate and accidental variations existed, and continue to exist, from edition to edition, but that hardly lessens the great historical and aesthetic force of the typographical fixity of the printed page. As I write in my Journal article, “Different editions still had textual variations, introduced either intentionally as revisions or inadvertently through sloppy editing or typesetting, but books still came to be viewed, by writer and reader alike, as immutable objects.” And are you really arguing that, in comparison to the indelibility of ink on a printed page, digital text rendered on screens does not entail a lessening of typographical fixity? That just seems like a pigheaded refusal to acknowledge reality.



I commend you for a balanced approach in the article, but your conclusion is too simple. It feels like you’re once again part of the modern movement of “anxiety journalists,” who seek a way into our free-floating anxieties about change by creating memorable places to hang our nagging concerns. You’re good at it. I just think these sexy and simple notions need a bit of deflating.

What you quote yourself as saying actually confirms this: “books still came to be viewed, by writer and reader alike, as immutable objects.” You’re describing a perception issue, not reality. If I call my friend with a paperback revised edition of a non-fiction book and ask her to turn to a paragraph on page 95, there’s a good chance we won’t be talking about the same passage. “Fixity” is more limited to reproductions of print at a certain point than to print overall, and that’s as much about economics as it is about printing. If there’s no money to update an e-resource, it stays pretty fixed, too. There are old blogs of mine that are a decade old and are the same as they were 10 years ago because I have no incentive to update them. Is that because they’re printed? Nope. And they may outlast the “fixity” of a crap hardcover novel that won’t get a second printing, is remaindered, and will soon be pulped.

My request in this post is that you refine your description of “fixity” because it’s not clear what you’re asserting it to be beyond the fixity of one instance of reproductions. Once you do this, I’d ask you to revisit whether the benefits of that outweigh the benefits you note of increased fluidity. If you’re going to stick to your guns, then I’m not the one being pigheaded.

I’m not arguing that e-text is as “fixed” as print — but it’s not a black-and-white difference either, and probably not driven by medium as much as you frame it to be. I’m critiquing your concept as being rather unclear when you get right down to it. Clarify it, and then let’s see your cost-benefit calculation.


Typographical fixity means, on one level, that when you had a page printed in ink, you were able to trust that the page would maintain its integrity; when you picked it up tomorrow, its contents would be the same as what you saw today. The printing press didn’t create that type of fixity; it continued it (on a much larger scale) from the scribal book. That fixity can no longer be taken for granted with a digital book on a networked device. The page can be altered remotely, in the same way other software is routinely “updated” today. This kind of fixity becomes even less assured if you are reading a digital book that is stored not on a device you own but in the cloud.

A second level of fixity – one introduced with the printing press – was the fixity of content across a large edition of a book. This kind of fixity was impossible with the scribal book, when copies were produced one at a time. This fixity was fairly weak immediately after the invention of the letterpress, when there were few or no controls over what a printer could print. But as copyright restrictions and other laws and business norms emerged (and as print runs expanded in size), the fixity across copies strengthened. This fixity never extended to different editions of the same work, which as I noted in my article could include large and small variations, either deliberate revisions or errors. Nevertheless, fixity within editions, often very large editions of thousands or even hundreds of thousands of copies, became a basic characteristic of publishing. This kind of fixity is not assured with e-books. An author or publisher can change the source file at whim, which means that copies of an “edition” downloaded tomorrow may be different from copies of the same “edition” downloaded today.

The third level of fixity involves what you term a “perception issue”: as the publication of a book went from being a vague, ongoing process to an event – a date on a publishing calendar – the sense of a book as a final, finished creation strengthened, particularly in the mind of an author but also in the minds of editors, proofreaders, and book designers. This sense of fixity, and finality, was, I believe, essential to the emergence of literary culture in its current form. That doesn’t mean that a particular author might not revise a book for subsequent editions, but it does mean that each of the subsequent editions was in itself a final, fixed creation – at best, a work of art aimed at posterity as well as the present day. Because it lacks the necessity and the fixity of a print run, e-publishing once again can become an ongoing process rather than an event, which is likely to once again change the perceptions of writers and their collaborators. (Perception can be extremely important when people are acting on it, as is often – always? – the case. I wouldn’t underestimate its effects.)

Those are the three levels of fixity that I was writing about in my brief essay. (There may be other levels.) At each level, e-publishing substantially reduces fixity when compared to print publishing.

I would like to hear your argument that “e-text is as fixed as print.”


Well, that took you a while.

So, what you call the first level of fixity is no different than the level of fixity found in a handwritten manuscript or letter or Post-it Note? In that realm, then, the pencil and eraser are threats to fixity? You don’t need typography to fix something to paper, or even paper to fix something in written form (headstones and cuneiform do this OK). What you call the second level of fixity is reproduction, which isn’t anything unique to paper but can be accomplished today in many ways — movies, audio, etc. Reproduction with fidelity does help different cultures enlarge, but literary culture emerged before textual reproduction technology — in fact, it had to, in order for there to be enough incentive for reproductions. As for what you call the third level, revisions aren’t trivial things in most cases, and technology isn’t the main impediment. Authors and publishers have to be incentivized to make changes and revisions, and once they decide to, the mechanisms of production kick in to allow it. With printed books, a new edition might be published, or a revised edition. The old books might be pulped if the author is prominent enough and the prospects of new sales are high enough. When Jonathan Franzen’s book was printed from the wrong computer file, thousands of copies were pulped and it was essentially erased from the face of the Earth. How “fixed” is that? I’ll bet the publisher hated fixity that day. Fixity can be assured with e-books — just turn off your wireless or save the file or PDF, and it’s fixed. You can leave the network pretty easily. And the benefits of updates, as you wrote, often outweigh the benefits of fixity, which were in a lot of situations detrimental to utility (maps, directories, reference works).

After seeing you force “levels of fixity” like this, I remain convinced you’re over-marketing a thought as a “big idea.” It’s not a big idea, it’s not scary, and it’s not worth so many electrons. On the bright side, we’ll probably end up in the Library of Congress with this exchange — they have hundreds of terabytes of Internet information archived, in addition to millions of tweets.

“In that realm, then, the pencil and eraser are threats to fixity?” That’s specious. Yes, I suppose some nefarious individual could break into my home some night, pull a book off my shelf, erase a few lines (difficult, by the way, with printer’s ink) and pencil in a revision. But, frankly, that seems like a stretch, and I’m pretty sure I’d be able to tell that the page had been fiddled with. With an e-book, as I wrote, the fixity of a printed page can no longer be taken for granted. A page of digital type can be altered remotely, in the same way other software is routinely “updated” today. This kind of fixity becomes even less assured if you are reading a digital book that is stored not on a device you own but in the cloud.

“You don’t need typography to fix something to paper.” Also specious. I’m not arguing that ink is the only way to fix text. You can certainly use a chisel and a stone tablet (good luck commercializing that). My point is that e-text lacks that kind of fixity.

“What you call the second level of fixity is reproduction, which isn’t anything unique to paper.” Also specious. As just noted, I’m not arguing that ink-on-a-page is the only way to achieve fixity; I’m arguing that the shift from ink-on-a-page to e-text reduces fixity. Also: different modes of reproduction have different degrees of fixity. Print reproduction has a high degree of fixity within an edition; scribal copying and e-distribution have lower levels of fixity.

“literary culture emerged before textual reproduction technology.” True, but as I wrote, literary culture “in its current form.”

“Authors and publishers have to be incentivized to make changes and revisions, and once they decide to, the mechanisms of production kick in to allow it.” Correct, and because the costs of altering e-text are much, much lower than the costs of publishing a new print edition, the incentives required to spur changes can also be much lower. It’s pretty simple economics.

“When Jonathan Franzen’s book was printed from the wrong computer file, thousands of copies were pulped and it was essentially erased from the face of the Earth. How ‘fixed’ is that?” Specious. Mistakes happen, and they get corrected. Big whoop.

“I’ll bet the publisher hated fixity that day.” I’ll bet you’re right. As I noted in my piece, there are costs as well as benefits to typographical fixity.

“Fixity can be assured with e-books — just turn off your wireless or save the file or PDF, and it’s fixed. You can leave the network pretty easily.” Are you kidding me? That’s your argument that “e-text is as fixed as print”? Sorry, you’re going to have to do better than that.


Obviously, you’re in love with this idea, so enjoy your courtship. I’m just pointing out that I think she’s pretty superficial and will ultimately break your heart.


Nice dodge.

I take your response to mean that you have no argument to make in support of your statement that “e-text is as fixed as print.” I’m disappointed, but not surprised.


I was trying to gently and wittily move on. You’re like a dog with a bone when it comes to what is essentially an ephemeral thought — that Eisenstein’s “fixity” is gone in the digital age. So, let’s engage once more. But I am getting tired of this, and may just move on without another word. That does not constitute surrender, just time management.

You claim that fixity is diminished in the digital age, yet offer no evidence of this loss of fixity, only fearful speculation. In fact, you accidentally offer evidence of print’s lack of fixity when you cite Updike as a proponent of “edges”, failing to realize that he manipulated his books after publication, recasting major characters based on second thoughts. You fail to outline exactly what you think “fixity” means — format integrity? textual fidelity? inability for outside forces to affect it? You do acknowledge that it’s fixed to an edition, much as a digital book is. If I download your e-book, and you update it, the one on my device doesn’t update.

It’s as fixed as we want to make it. If you want your books to remain fixed, don’t update them. I’ve not corrected typos in my e-books for this very reason — they are part of that edition.

You write: “Once digitized, a page of words loses its fixity.” OK, try changing these digitized words:

“Nicholas Carr can’t change these digital words.”

When you change them, which you certainly will be able to do because a page of digital text ABSOLUTELY loses its fixity because its digital per your argument, let me know how you did it. Otherwise, I’ll assume digitized words can be just as fixed as print.

So, I’m thinking about a model some publishers are using now (including my former press Penn State) where a book available free online is offered for optional purchase as a POD edition. Since no inventory exists and copies are printed one at a time (usually), what is to prevent the online version from being changed so frequently that, in theory at least, every single copy of the POD edition would be a different “edition”? There is a temporary “fixity” for each corresponding online/POD “edition,” but continuous change is a more apt description of this process than “fixity,” and it applies equally to both online and print “editions.”


I guess I can understand why you become eager to move on when your thoughts are challenged, but to bore you further here’s my response to your latest comment on my latest comment (before this one, of course):

You write: “You claim that fixity is diminished in the digital age, yet offer no evidence of this loss of fixity.”

1. All the major e-book publishing platforms – Kindle, Nook, iBooks – allow continuous editing of e-book source files. This kind of continuous editing was impossible with print books. If you don’t see that as a diminishment of typographical fixity, I’m not entirely sure what you would see as a diminishment of typographical fixity.

2. E-books downloaded to devices can be retroactively edited, and owners of the books need not be alerted to the retroactive edits. For many books, such as guidebooks and textbooks, this will be a feature many people will value. But it does represent a diminishment of typographical fixity. Retroactive changes to printed books were impossible (unless you count errata sheets). Amazon has pledged to request the permission of owners before making retroactive edits (spurred by controversy over the remote deletion of two Orwell books from Kindles a couple of years ago). That’s good, but it underscores the fact that retroactive editing is possible (and in fact easy) and provides little assurance that companies (and governments) in the future will always request permission.

3. Printed books have a great deal of longevity (ie, fixity through time) – even though, admittedly, that longevity has decreased in recent years due to the use of crappier paper and crappier bindings. It seems likely, based on historical precedent, that e-books in (often proprietary) digital formats will have less longevity, given how gadgets and file formats go out of date and companies go bust. I hope this problem is fixed, but it doesn’t seem a top concern of e-publishers, or of e-book buyers, for that matter. So for the time being it represent a loss of fixity.

All of these factors entail a diminishment of the typographical fixity we take for granted in printed books. I have nowhere argued that printed books have absolute fixity or that e-books have an absolute lack of fixity (when you make that charge, in all caps, you’re simply trying to distort my argument so you can dismiss it); what I am arguing is that text in e-books is less fixed than text printed on paper in ink. There are advantages to the greater malleability of e-text, but there are also disadvantages. We are a long way from knowing how they will play out.

You seem determined to reject my observations, which is certainly your prerogative. But, honestly, I think you’re shooting the message in hopes of wounding the messenger.


You keep saying it’s “print” that created fixity and typography, but it’s the economics of mass paper duplication and the related expense that created barriers to revision. Yet revisions in print would and could and do occur when the economic incentives are sufficient. You revised your e-book file because it was no-cost to do so. If Amazon had imposed a $15K charge for the change, your book would remain “fixed.” So I disagree with you as to the reason books remain fixed. Again, Franzen’s book was destroyed because the wrong file was put into the mass paper duplication process. It was economically sensible to pulp those books, that entire edition. Those words were not fixed per se. The could not withstand the economics of the publisher. They were disposable, and disposed of. I’m sure that file was deleted as well. But, perhaps not. Perhaps the digital file persists still, more fixed than the printed edition could have been — because of economics.

In my original post, I gave you credit for a lot of good observations. My objection was with your apparent desire to stir up fears about a loss of fixity, and then misattributing fixity to print and metal type when it’s more an economic phenomenon. If e-books cost thousands of dollars to revise, they’d be pretty fixed.

It’s not printing that determined fixity, it’s print economics, and we rationalized that into what you call “literary culture.” Now, the economic barriers to revising, repairing, and redacting files are gone, but that doesn’t mean digital is less fixed AS A MEDIUM. It merely means the economic incentives in publishing have changed.

Beyond that, fixity remains a choice. If we have a literary culture that treasures editions and fixity, it will stay. I won’t edit your comments even though I could. That’s editorial integrity. Where was that in your dissection of fixity? Was Updike lacking in editorial integrity because he couldn’t leave well enough alone? It was a source of controversy at the time.

Again, I’m not sure you’ve explored the concept fully, which is why I critiqued it. But let’s move on. I don’t mind having my ideas challenged, and I love a nice back and forth, but I kind of know when the point of diminishing returns has been reached, and for me, this is it. We’re talking past each other at this point.

Talk to all those scholar-editors who have struggled to create “authoritative” texts for critical editions from multiple variants about the “fixity” of the printed text. I agree that the change from print to digital is a matter of degree, not conceptual difference.

I think this speaks to the shifting paradigm in how traditionally readers ‘took delivery’ on information instead of simply ‘accessing’ information. eBooks are more and more becoming the medium of access and not storage. Same goes with music. When I listen to a song on Pandora, I am simply playing a binary version that is stored somewhere in the cloud that millions of other people are accessing. Fixity, to me, is a necessary bi-product of physical information distribution.

Having once lived in a country where human rights are not always respected, I can appreciate Nick’s concern about fixity. There are certainly people out there who would have a desire, on occasion, to alter a historical record to fit a narrative. Nick is right that once a book has been published, it cannot be taken back – the record is out there for all to see.

However, I am not so sure that digital records are any less safe. Once a version of an electronic document is out there it can be saved and its integrity protected. Just this morning NPR broadcast a report of how someone in authority tried to alter a picture of a leading dissident in Russia (to put him in a bad light). Within hours of the altered photo appearing on official websites others released the original image in it unaltered state.

New technologies present both new opportunities and problems for the honest and nefarious alike. I think Nick is right in that electronic formats make it easier to alter records (if that is what he means by fixity). I think the larger issue is where and how electronic records are being stored. Books offered security because “the original version” was stored in thousands of geographically dispersed locations – so if one copy was destroyed there were many more copies elsewhere. The works of Thomas Mann destroyed during the book burnings in Nazi Germany were preserved because copies of the original were on book shelves in Paris, London and New York. But in the electronic age how many copies are being stored, where and by whom? How is the data protected and the integrity of files preserved (electronic files do degrade over time)? It seems to me that at least in this context, electronic formats may by less secure than the printed word. But isn’t that the purpose of LOCKSS; to resolve this issue?

Carr’s essay turns an idle thought into a somewhat clever ending. This thought is rooted in Nostalgia, not scholarly concern. Updike (the subject of my dissertation) was a romantic about books and the printing process. He had a similar notion about fixity when he wrote about the damage brought about by the introduction of film. It’s hard to argue against a nostalgic personality, except to remind them that sometimes memory is selective and things can improve when they change. As long as earlier versions are available (at least for historians) I see few problems with fluidity. We need more of it, not less, in scholarly publishing.

Comments are closed.