The Occasional Pamphlet

[This is a heavily edited transcript of a talk that I gave on January 3, 2013, at a panel on open access at the 87th Annual Meeting of the Linguistic Society of America (LSA, the main scholarly society for linguistics, and publisher of the journal Language), co-sponsored by the Modern Language Association (MLA).]

Thank you for this opportunity to join the others on this panel in talking about open access. I will concentrate in particular on the relationship between open access and the future of scholarly societies. I’m thinking in particular of small to medium scholarly societies, which have small publishing programs that are often central to the solvency of the societies and to their ability to do the important work that they do. In one sense it should be obvious, and I think it’s been made obvious by the previous speakers, that open access meshes well with the missions of scholarly societies. LSA’s mission, for instance, is “to advance the scientific study of language. LSA plays a critical role in supporting and disseminating linguistic scholarship both to professional linguists and to the general public.” [Emphasis added.] So I’ll just assume the societal benefit of open access to researchers and to the general public alike. For the purpose of conversation let’s just take that as given.

Nonetheless, many scholarly societies, and the faculty that support them, are worried that open access – at least as they understand the concept – could exacerbate the serious financial distress that many of those societies are already under, and even undermine their very existence and thereby their ability to carry out this mission. I’ve heard faculty worry that even “green” open access (self-archiving of articles in open-access repositories) could undermine the economics of journal publishing in such a way that their scholarly societies could be endangered.

I want to argue that there is a real threat that many scholarly societies accurately perceive in their publishing programs, but that we must be careful not to misdiagnose this problem. In fact, a general move to open access would be the best outcome for scholarly society publishers. If the entirety of journal publishing magically metamorphosed somehow to an open-access system, scholarly society publishers would be much better off. From a strategic point of view then, the best course of action for scholarly societies and for the faculty and researchers who support them would be to promote a shift to open access as widely and as quickly as possible. Now, the threat that societies perceive is an economic threat, so my remarks will be almost entirely economic in nature; I’m just warning you. My talk certainly will have no linguistic import at all.

Economic properties of the subscription market

Let me turn first to some basic truths about the subscription journal market that I’ve come to realize are important in understanding what the underlying economic issues are.

Journal access is a complementary good

The first is that different journals — viewed as products, as goods being sold — are in economists’ terms complements, not substitutes. Substitute goods are products like Coke and Pepsi. If you have one it decreases the value of the other to you, as they fulfill similar functions. Complements are products like a left shoe and a right shoe – that’s the most extreme case. If you have one it increases the value to you of the other. There are less extreme cases of economic complements – printers and toner cartridges, peanut butter and jelly, pencils and erasers.

What about scholarly journals? Suppose you’re a patron of a library that subscribes to a bundle of, let’s say, Elsevier journals, including the journal Lingua. Does the library subscription to that journal make you more or less interested in reading, say, Language? (We’re holding cost aside. When thinking about complements or substitutes, it’s just about the value to the consumer, not the cost.) Of course, you’re not less inclined to read Language just because the library subscribes to Lingua. In fact you may be more inclined, because some Lingua articles will cite Language articles. You read the Lingua article, you want to to read the Language article it cites. So that would lead you to track down those articles and read them if the library had a subscription. And vice versa: a subscription to Language can increase the value of a subscription to Lingua. So journals are economic complements, not substitutes.

Inefficiency in the subscription market

This has important ramifications. Non-substitutive goods don’t compete against each other and complementary goods in fact support each other in the market. If consumers suddenly buy a lot more Coke, Pepsi is worried. But if peanut butter sales skyrocket, the jelly manufacturers are elated. So the complementary subscription of individual journals means that there’s limited market competition between journals, and limited competition leads to inefficiency in the journal market. (That’s not to say that there isn’t competition between publishers. But as we’ll see, the primary form of that competition is in competing to acquire journals.)

Figure 1: Average journal prices in a range of fields, differentiated by commercial and non-profit publishers. Left is based on prices as dollars per page. Right is based on dollars per citation, to normalize for quality. Data are from Bergstrom and Bergstrom, Journal pricing across disciplines, 2002.

We can see ample evidence of this kind of inefficiency. One clear form of evidence for inefficiency is wide price disparities. The graph in Figure 1 shows average journal prices in a whole range of fields. The data is from a study by Bergstrom and Bergstrom, and they differentiated the cost of the journals by whether the publisher is commercial or non-profit. The dark blue represents the commercial publishers, the light blue the nonprofit publishers. Notice that the commercial publishers on average charge about five times more for their journals than non-profit publishers, as measured by price per page. Now you might think there is a good explanation for this disparity: perhaps these aren’t comparable products. Perhaps the commercial publishers are selling a much better product, higher quality journals, and it’s therefore more expensive to develop them, and that’s what accounts for the price differential. So we can normalize for that by using a proxy for quality. One widely used proxy for quality, admittedly not a great one but at least widely touted by journals themselves through the ubiquitous “Impact Factor”, is the number of citations the journal receives. The second graph in Figure 1 thus shows price per citation. Measured this way, commercial journals are fifteen times more expensive than non-profit journals from the same field. Now, linguistics was not one of the fields examined in this study. But the same holds true here as well. For example, the subscription rate for LSA’s journal Language, published by a non-profit of course, is $3.31 per citation, whereas Elsevier’s Lingua is $32.30 per citation – almost exactly ten times more expensive.

This kind of price differential is a clear sign of market failure, especially as it has been sustained over decades. You just do not get this kind of price disparity preserved over long periods of time in well functioning markets. Go to the various grocery stores in your neighborhood and see if you can find apples at different grocery stores at a price differential of a factor of ten. It does not happen. Such price disparities are a clear sign of inefficiency.

Journal access is a monopolistic good

Figure 2: Elsevier revenues, profit, and profit margin, 2002–2011. Data are from Mike Taylor, The obscene profits of commercial scholarly publishers, 2012.

The second basic truth is that the good being sold in the subscription market is access, and access is a monopolistic good. The monopoly is enabled by copyright, founded in the government’s ability as codified in Article I Section 8 of the Constitution to provide an exclusive right to the creator of a work for a limited period of time. Subscription publishers acquire exclusive rights to the articles they publish — typically by acquiring copyright, sometimes by acquiring an exclusive license, which is a distinction without a difference — and this allows publishers in theory and in many cases in practice to extract monopoly rents in selling access to the articles. We see evidence of this as well. For example, in Figure 2, I show a graph of the revenues, profits, and most importantly the profit margin, for the publisher Elsevier over the last decade. It’s quite a good business with annual revenues of over $2 billion, but that’s not the big point. The big point is the extraordinary 35–40% profit margins. It’s not just Elsevier. Many large commercial publishers have maintained these kinds of profit margins over a long period of time. An interesting thing to look at is the steady increase in the margins even during the financial crisis starting in 2009 when, for instance, many university endowments and library budgets dropped precipitously. Harvard’s endowment went down by 30% but Elsevier did just fine, and the other large publishers as well. So maintaining those kinds of profit margins again is a sign of the ability to extract monopoly rents.

Journal access is a bundled good

The third basic truth is that pricing is controlled not at the level of the individual journal but at the level of a bundle of journals. The large publishers have portfolios of hundreds to thousands of journals. They can therefore apply prices to a bundle of journals, not a single journal. They can show vastly different prices to different buyers and use the bundles to incentivize buyers, the libraries, to pay larger fees. The upshot of this point, that pricing happens at the bundle level and not the journal level, is that a library can find it extremely difficult to control its expenditures by canceling individual journals because the publisher can just price the smaller bundle at essentially the same cost as the larger bundle.

I’ll tell you a personal story. Some years ago, Harvard was one of the first universities to cancel the “big deal” with Elsevier. I don’t want to pick on Elsevier. They’re not bad people. They’re a wonderful group of folks. Lots of the large publishers of journals work this way and it’s not because they’re evil or anything like that. I just mention the Elsevier case as a convenient story. Harvard was one of the first universities to cancel its “big deal” and went a la carte on the journals. In the School of Engineering and Applied Sciences, my own school, we had been subscribing to around 130 Elsevier journals in engineering and applied sciences as I recall. We took the opportunity to cancel about 100 of these journals, leaving something like 30 journals, hoping to recoup some costs. And we did. The first year we recouped about 20%. The following year the total cost was back where it had been before the cancellations, and it has increased steadily from there. From the library’s point of view, you can’t win by canceling journals, because the product is not the journal, it’s the bundle.

Edlin and Rubinfeld, in a Law Review article about possible anti-trust implications of this bundling, say “The immediate effect of [bundled pricing] has been to move competition from individual journals to large bundles of journals. … Creating a large bundle of journals to compete with Elsevier or Kluwer seems almost insurmountable. … There are indications that [bundled pricing] is hindering entry. Librarians … say that they would spend more money for journals from smaller and alternative publishers if they could achieve proportionate savings from reductions. By selling electronic bundles, publishers have erected a strategic barrier to entry at just the time that the electronic publishing possibility has made it increasingly possible for alternative publishers to overcome the existing structural barriers.” The fact that competition is at the level of bundles, not at the level of journals, is very important.

The result: market dysfunction

Figure 3: Scholarly journal expenditures percentage increase 1986–2010 compared to consumer price index. Data from Association for Research Libraries.

When we put all these properties of the journal market together, the end result is market dysfunction and a steady long-term hyperinflation in journal expenditures by libraries. Figure 3 shows a graph of serials expenditures over the last couple of decades, the dark blue line. The light blue line is the consumer price index, a proxy for the ambient rate of inflation. You can see that serials expenditures in research libraries have been going up at something like three times the rate of inflation for decades. Exponential real growth in the cost of journals is an unsustainable state of affairs.

I return to the issue of inefficiency. Why is it that the non-profit publishers are so much more efficient than the commercial publishers? Not in every case of course but on average the difference is really striking. There are a couple of possible reasons. One is that the non-profits tend to be scholarly societies who may be motivated not by profit maximization but by service to the field. I think that’s true to a certain extent. But also the non-profits tend to be small publishers with few journals – maybe one, two, three, five, ten journals. Since bundle size governs market power, non-profits have less ability to grow margins. And scholarly societies rightly complain that they’re being squeezed. From the point of view of libraries, if you have to cancel something you can recoup revenue if you cancel the journals from a small publisher. You can’t recoup revenue if you cancel journals from the large commercial publishers. As a library, what are you going to do? Cancel scholarly society journals, just as the societies have been rightly complaining about.

But notice that the problem that scholarly societies face, a problem that will only increase in a status quo future, is based not on open access but on inherent properties of the subscription market that they participate in. For scholarly societies, the status quo is not a good alternative. Doing nothing is a failing strategy.

Open-access journals as a preferable system

The idea of open-access journals is that they provide access to the articles for free. How can this be a better system for scholarly societies, given that much of the societies’ revenues may come from the publishing program?

Open-access journals don’t charge for access, but that doesn’t mean they eschew revenue entirely. Open-access journals are just selling a different good, and therefore participating in a different market. Instead of selling access to readers (or the readers’ proxy, the libraries), they sell publisher services to the authors (or to the authors’ proxy, their research funders).

In fact there are now over 8,500 open-access journals listed in the Directory of Open Access Journals. Some of them have been mentioned already on this panel: Linguistic Discovery, Semantics and Pragmatics. The majority of existing open-access journals, like those journals, don’t charge author-side article-processing charges (APCs). But in the end APCs seems to me the most reasonable, reliable, scalable, and efficient revenue mechanism for open-access journals. This move from reader-side subscription fees to author-side APCs has dramatic ramifications for the structure of the market that the publisher participates in.

Economic properties of the open-access journal market

The open-access APC market has quite different properties from the subscription market. Recall the basic truths about the subscription market. Journals are complements, not substitutes. There’s limited market competition. The product being sold is a monopolistic good. Pricing is controlled at the bundle level. What are the corresponding properties of the publisher services market, the market that open-access journals participate in? In that market, the purchaser of the good is the author or the author’s proxy, not the reader or reader’s proxy. And from the point of view of an author, two journals are not complements but substitutes. You can publish your article in the Journal of Linguistics or Lingua or better yet in Language. But having published it in one, you have no incentive to publish it in the other. In fact, you’re not allowed to publish in both, making journals perfect substitutes. There is no value to the second journal once you’ve published the article in the first journal, from the point of view of the author trying to get a publication.

So journals compete for authors in a way they don’t for readers, and this competition leads to much greater efficiency. Open-access publishers are highly motivated to provide better services at lower price to compete for authors’ article submissions. We actually see evidence of this competition on both price and quality happening in the market. I won’t go through examples but have written about it previously.

Second, publisher services on the author side are not a monopolistic good. Anyone can provide those services. In fact because the service is a knowledge good, there are exceptionally low barriers to entry. Kai von Fintel and David Beaver can just unilaterally set up Semantics and Pragmatics; maybe they’ll be successful and maybe they won’t. In this case, it turned out pretty well. The low barrier to entry further enhances competition and improves the efficiency of the market.

Finally, pricing is controlled not at the level of the bundle of journals. You don’t care about the bundle of the publisher when you’re an author submitting to a journal. You care about the journal. Actually, pricing is not even at the journal level, but at the level of the individual article. So price competition happens at that level as well, with journals competing for individual articles on price as well as quality.

In summary, the open-access APC market is a more efficient market than the closed-access subscription market for reasons of basic economics. That’s not just my opinion. Claudio Aspesi, senior analyst at Sanford Bernstein studying the finances of publishing companies, has estimated that a transition to open access would lead to Elsevier cutting its margins by 41–89%.

Comparative cost of open-access journals

Let me say something about the overall cost for the two kinds of models. The APCs that open-access journals charge range from $0 to around $3,000. The median turns out to be zero. But for those open-access journals that do charge a fee, the mean is around $1,200, and reasonable sustainable fees seem to be shaking out in the $1,000 to $1,500 range. Let’s call it $1,500. Since article processing fees are essentially the totality of revenue that open-access journals receive, the APC is a reasonable figure for average revenue per article. There are open-access publishers who are profitable in that range, including commercial open-access journals.

What’s the corresponding number for subscription journals? What is their average revenue per article? The Scholarly Publishing Roundtable reported total 2008 revenue for scholarly publishing at $8 billion on 1.5 million articles, the vast bulk of that revenue coming from subscription fees. Average revenue per article for subscription journals is, by that measure, over $5,000 an article. Remember that this averages over all of the journals — the high quality and the low alike.

So what’s happening is that authors one way or another are paying. Either you’re paying an APC to an open-access journal or you’re paying with your copyright to a subscription journal, which the publisher then monetizes, turning it into about $5,000 per article. It turns out that if we moved from a subscription journal world to an open-access world, the institutions of the world would go from paying, on average, $5,000 an article to about $1,500. Let’s suppose the $1,500 estimate is unreasonably low. Let’s suppose that really the average APC would be what the most high-end open-access journal, PLoS Biology, now charges – that’s $2,900; call it $3,000. If every article moved from the subscription model to an open-access APC model at the high end of cost – we would still be saving 40%. And more importantly, we would be better executing the scholarly society mission by providing the broadest possible dissemination.

Scholarly societies as open-access publishers

Who wins in this kind of market – a non-monopolistic, competitive market of substitutes where the processing fees are considerably less than the current cost per article for subscription journals? The publisher who wins in that market is the publisher who can provide the best services, including imprimatur, at the lowest price to the author, that is, the publisher who is most efficient. Scholarly society publishers would have a huge lead in this market, because they are manifestly more efficient than commercial publishers by a large factor. If the scholarly journal market were structured as the open-access journal market rather than the subscription journal market, scholarly society publishers would be the big winners. And scholarly societies are beginning to realize that open access could be a boon not only to their mission – that much should be uncontroversial – but also to their solvency. Perhaps for this reason, some 600 scholarly societies, including the LSA, are already publishing open-access journals.

At the root, the reason that scholarly societies benefit from playing in the open-access APC market rather than the closed-access subscription fee market is the difference in the goods being sold. When the good is a journal bundle, the companies with the biggest bundles, the large commercial publishers, win. When the good is publisher services for an individual article, the publishers that can deliver those services for an individual article most efficiently, the non-profit publishers, win. Sure, there are economies of scale, but empirical evidence shows that the scholarly societies are already far better able to efficiently deliver services despite any scale disadvantage.

The problem for open access: the transition

Now, all that sounds great, but I don’t want to be too positive. As I said at the outset, there is a real worry that society publishers should have about the open-access APC market. But it’s not that they’d be at a competitive disadvantage in that market; I think that they’d have a huge advantage. And it’s important to remember that they’re already at a huge disadvantage in the subscription journal market; status quo is a failing strategy. Rather the problem is this. Open-access journals are at a disadvantage in their competition for authors against subscription journals. That is, the problem arises across the two markets. When the only kind of journals are open-access journals, scholarly societies have the upper hand. When there are both kinds of journals in the market, both subscription journals and open-access journals, the open-access journals are at a competitive disadvantage because (from the author’s point of view) publishing is free in a subscription journal. (Of course, it’s not really free; it’s just that the research libraries of the world are underwriting the very high $5,000 cost per article.) By contrast, in an open-access APC journal, the author personally could be out let’s say $1,200 or $1,500 or whatever. This is a problem not just for scholarly societies but for all publishers exploring the possibility of going fee-based open access.

To make a transition possible, what we, as supporters of scholarly societies, should be working on is placing open-access journals on a level playing field with subscription journals. There’s a principle at stake here, and the principle is this: Dissemination of research results is an inherent part of the research process. This is something that publishers themselves are frequently pointing out — that they are part of the research process. Consequently, the funders of that research should underwrite dissemination of the results. Who are the funders of the research? In science, technology, and medicine, public and private funding agencies are the primary research funders. By this principle then, the funding agencies giving the grants in those areas would be on the hook to pay the $1,000 or $1,200 or $1,500 or $2,900 publication fees. Most funding agencies already will pay for publication costs for open-access journals (though not in an ideal way, which I’ve written about in the past). What about fields where there aren’t funding agencies handing out large grants? In the humanities and social sciences, universities are the de facto primary research funders. Faculty members in universities are doing research in those fields as part of their employment as researchers. As the primary research funders in the humanities and social sciences, in linguistics in particular, the universities that employ us should be on the hook to disseminate the research results that their researchers generate.

This is the principle behind an effort called the Compact for Open-Access Publishing Equity. It was set up by a group of five universities initially — Cornell, Dartmouth, Harvard, MIT, and Berkeley — to place the open-access revenue model on a more level playing field with the subscription model. Since then another dozen or so institutions have signed on. The Compact says that these universities commit to providing a mechanism for underwriting reasonable publication fees for articles written by their faculty and published in fee-based open-access journals. From the point of view of these signatory institutions, and the many other institutions that don’t happen to be signatories but have similar funding policies, if you structure your journal as an open-access journal charging a publication fee, you don’t need to worry that the authors will be personally out of pocket to pay those fees; the university will pay on their behalf.

Next steps

Given that the open-access publication fee market would be preferable from the point of view of scholarly societies and their members, what should scholarly societies be doing from the strategic point of view? What is in the best interest of us as supporters of scholarly societies? Happily the best interest of scholarly society publishers is the best interest of the scholars themselves, namely as rapid a transition to open access as possible. So scholarly societies should be doing what they can to speed that transition, and I’m glad to say that the LSA and the MLA are working in that direction. I wish all scholarly societies would do so as well.

Of course, the ideal action for a scholarly society is to convert all of its journals to open access. By doing so, they help set expectations among authors that we don’t restrict access to articles, thereby hastening the day that closed-access journals find it impossible to compete for authors.

But some scholarly societies may still find it too worrisome to take such a bold move, not because they disagree with my conclusion that they fare better in an open access world, but because they fear not making it through the transition to that world. I’m sympathetic to that worry. Still, there are important actions that societies can take short of converting all of their journals to open access, actions that will still greatly contribute to changing the expectations of scholars that research results should be and are being made accessible. Scholarly societies can:

• Experiment with open access for at least some of their journals (as LSA is doing with Semantics and Pragmatics), thereby gaining exactly the experience with open-access publishing that will be invaluable in the future.
• Along the same lines, commit to open access for any new journals.
• For the legacy non-open-access journals, provide delayed open-access to articles, making them available with a broad license after, say, six months or one year. The LSA has already taken this important step. Once conditions are right, the delay can simply be dropped.
• Explicitly allow self-archiving of articles published in their journals, the green open access that I alluded to at the start of the talk. Doing so sends a strong signal that the society supports open access. At the same time, there is “no persuasive evidence that increased access threatens the sustainability of traditional subscription-supported journals, or their ability to fund rigorous peer review.” TheLSA does this, and the MLA recently announced that they are modifying their publication agreement along these lines, and even allowing distribution of the final published version after one year.
• Recognize, accommodate, and promote university and funder open-access policies. Accommodation requires only the addition of a single sentence to a publisher’s publication agreement. The pertinent sentence taken from the Science Commons addenda is this:
  

  Where applicable, Publisher acknowledges that Author’s assignment of copyright or Author’s grant of exclusive rights in the Publication Agreement is subject to Author’s prior grant of a non-exclusive copyright license to Author’s employing institution and/or to a funding entity that financially supported the research reflected in the Article as part of an agreement between Author or Author’s employing institution and such funding entity, such as an agency of the United States government.

  (My guess is that it should be possible to generate an English version of such a sentence as well.)

• Support pro-open-access legislation such as the Federal Research Public Access Act. At the least, scholarly societies should disavow anti-open-access statements made on their behalf by publishing consortia, as the MLA did in its statement opposing the Research Works Act.
• Leverage the society’s membership to push for open-access underwriting by funding agencies and by universities such as envisioned by the Compact for Open-Access Publishing Equity.

To the extent that we can get a transition to a primarily open-access publishing system to happen, scholarly societies, their members, and the general public will all be much better off, which is a happy confluence of interest. Thank you very much.

[I am pleased to present a guest post from my friend Ann Velenchik, professor of economics at Wellesley College, director of their writing program, and expert monologist. This post is reproduced from her private blog, which I am privileged to have access to, in which she has chronicled her experience with her leukemia diagnosis and treatment over the last three years.]

…one of my idols… “Oprah Winfrey speaks at the launch of the Born This Way Foundation” image by flickr user HarvardEducation.

January 18, 2013 — Despite living with Bicycle Boy, I take no interest in competitive cycling. We were in Paris for the end of the Tour de France in 2008, and while Tom, Becca and Nate all stood on benches to see the riders circle the Arc de Triomphe, I was happily drinking an orangina at a table far from the crowds. Tom has assured me, for years, that Armstrong has clearly been doping, and I frankly didn’t think much more about it.

But I obviously couldn’t escape the news that Oprah would be interviewing Armstrong on TV last night and, because Oprah is one of my idols, I did a little web surfing this morning to find out what was said. And I found something that made me so angry that I had to respond.

In a part of the conversation about why he had started doping, and how he justified it to himself, Armstrong lay some of the blame on the “fighting spirit” he developed during his “battle” with testicular cancer from October 1996 to February 1997.

“That process turned me into a person — it was truly win at all costs,” Armstrong said. “When I was diagnosed, I said, ‘I will do anything I need to do to survive,’ and that’s good. And I took that attitude, that ruthless and relentless and win-at-all-costs attitude into cycling, and that’s bad.”

Let’s leave aside the fact that there’s evidence that he started doping before he got cancer, and that it’s possible that taking a lot of testosterone might have made that cancer worse. Let’s just talk about the idea that the attitude that helped him “win” the cancer battle justifies, or even explains, what the evidence indicates he has done since.

I think it is highly possible that cancer diagnosis and treatment in the prime of his life was a deep and abiding trauma that warped his moral compass. As I have said before, I don’t think cancer is a blessing in disguise, and I don’t think all the lessons we learn there are good ones, let alone worth the price. So I am not even pissed off that he has the audacity to use his status as a cancer patient to explain the appalling way he has treated people.

What pisses me off is his description of the attitude he brought to treatment itself. When he says “I will do anything I need to do to survive…ruthless and relentless and win-at-all-costs…,” as though lying and cheating and doing terrible things to other people were part of the cancer process, that’s when my head starts to explode.

Because, here’s the thing. There isn’t much in cancer treatment that requires lying or cheating, that requires you to sue for libel the people who are actually telling the truth, or that allows you to threaten and bully and defame other people. Yes, there’s a lot of win-at-all costs to be found there, but those aren’t costs you get to impose on other people.

Lance Armstrong misspoke. Cancer treatment isn’t about being willing to do anything to anyone in order to win. It’s about being willing to endure anything onesself. Here’s my guess. Lance Armstrong is a very bad guy who was a bad guy before he got cancer and perhaps a worse one afterward. He doped because he was getting away with it and getting richer and more famous every minute. He lied and intimidated and threatened and bullied because, as I heard one person say, when he got cornered his strategy was to double down. And maybe his experience as a cancer patient was part of the list of things that made him so broken. But that’s about him, not about cancer.

Government zealotry in prosecuting brilliant people is a repeating theme. It gave rise to one of the great intellectual tragedies of the 20th century, the death of Alan Turing after his appalling treatment by the British government. Sadly, we have just been presented with another case. Aaron Swartz committed suicide at his apartment in New York this week in the face of an overreaching prosecution of his JSTOR download action. I never met him, but I understand from those who knew him well that he was a brilliant, committed person who only acted intending to do good in the world. I’m on the record disagreeing with the particulars of the open access tactic for which he was being prosecuted, on the basis that it was counterproductive. But I empathize with the gut instinct that led to his effort. I hope that it will inspire us all to redouble our efforts to eliminate the needless restraints on the distribution and use of scholarship as Swartz himself was trying to achieve.

…our little tiff in the late 18th century…“NYC – Metropolitan Museum of Art: Washington Crossing the Delaware” image by flickr user wallyg. Used by permission.

I’m shortly off to give a talk at the annual meeting of the Linguistic Society of America (on why open access is better for scholarly societies, which I’ll be blogging about soon), but in the meantime, a linguistically related post about punctuation.

Careful readers of this blog (are there any careful readers of this blog? are there any readers at all?) will note that I generally eschew the peculiarly American convention of moving punctuation within a closing quotation mark. Examples from The Occasional Pamphlet abound: here, here, here, here, here, here, here, and here. And that’s just from 2012. It’s surprising how often this punctuation convention comes into play.

Instead, I use the convention that only the stuff being quoted is put within the quotation marks. This is sometimes called the “British” convention, despite the fact that other nationalities use it as well, presumably to emphasize the American/British dualism extant from our little tiff in the late 18th century. I use the “British” convention because the “American” convention is, in technical terms, stupid.

The story goes that punctuation appearing within the quotation mark is more aesthetically pleasing than punctuation outside the quotation mark. But even if that were true, clarity trumps beauty. Moving the punctuation means that when you see a quoted string with some final punctuation, you don’t know if that punctuation is or is not intended to be part of the thing being quoted; it is systematically ambiguous.

Apparently, my view is highly controversial. For example, when working with MIT Press on my book on the Turing test, my copy editor (who, by the way, was wonderful, and amazingly patient) moved all my punctuation around to satisfy the American convention. I moved them all back. She moved them again. We got into a long discussion of the matter; it seems she had never confronted an author who felt strongly about punctuation before. (I presume she had never copy-edited Geoff Pullum, from whom more later.) As a compromise, we left the punctuation the way I liked it—mostly—but she made me add the following prefatory editorial note:

Throughout the text, the American convention of moving punctuation within closing quotation marks (whether or not the punctuation is part of what is being referred to) is dropped in favor of the more logical and consistent convention of placing only the quoted material within the marks.

I would now go on to explain why the “British” convention is better than the “stupid” convention, except that Geoff Pullum has done so much better a job, far better than I ever could. Here is an excerpt from his essay “Punctuation and human freedom” published in Natural Language and Linguistic Theory and reproduced in his book The Great Eskimo Vocabulary Hoax. I recommend the entire essay to you.

I want you to first consider the string ‘the string’ and the string ‘the string.’, noting that it takes ten keystrokes to type the string in the first set of quotes, and eleven to type the string in the second pair. Imagine you wanted to quote me on the latter point. You might want to say (1).

(1) Pullum notes that it takes eleven keystrokes to type the string ‘the string.’

No problem there; (1) is true (and grammatical if we add a final period). But now suppose you want to say this:

(2) Pullum notes that it takes ten keystrokes to type the string ‘the string’.

You won’t be able to publish it. Your copy-editor will change it before the first proof stage to (3), which is false (though regarded by copy-editors as grammatical):

(3) Pullum notes that it takes ten keystrokes to type the string ‘the string.’

Why? Because the copy-editor will insist that when a sentence ends with a quotation, the closing quotation mark must follow the punctuation mark.

I say this must stop. Linguists have a duty to the public to use their expertise in arguing for changes to the fabric of society when its interests are threatened. And we have such a situation here.

What say we all switch over to the logical quotation punctuation approach and save the fabric of society, shall we?