The Journal Issue and the Record Album: Two Fundamentally Irrational Information Products
by Rick Anderson (Associate Director for Scholarly Resources & Collections, Marriott Library, University of Utah; Phone: 801-721-1687) email@example.com
Over the past few years I’ve become more and more convinced that the scholarly information world has a lot to learn from the music industry. Not so much from what the latter is doing either right or wrong, but from what has happened to it over the past 100 years, how it has happened, and why.
From the early decades of the 20th century until the 1950s, “buying a record” generally meant buying a shellac disc that contained only a bite-sized portion of music: a popular song, a single performance of a jazz composition, a brief piece of light classical music. Each disc could hold about three minutes of recorded sound. If you wanted to listen to something longer (an entire symphony, for example), you had to buy an “album” — a package of multiple records that you played in sequence.
In the 1950s, technological advances made possible the advent of the vinyl “long-playing record,” or LP. LPs were two-sided, and could hold twenty or twenty-five minutes of music on each side. They quickly changed the way musicians made music and the way record labels marketed it: having two chunks of twenty or more uninterrupted minutes to work with opened up all kinds of new expressive possibilities, and also made it possible to put together programs of ten or twelve songs and sell them as a one-disc package. “Singles” (smaller records containing one song on each side) declined sharply in popularity over the next couple of decades, and increasingly, when people talked about “buying a record,” they came to mean buying an album — which was no longer a physical “album” of separate discs, but rather an album of songs on two sides of a single disc.
The significance of this development to the future trajectory of the music business can hardly be overstated. For the first time, the way most people gained access to a song that they had heard and enjoyed on the radio was not to pay $1 for a recording of the song itself, but to pay $7 for a collection of ten or twelve songs that included the one they wanted. Record labels were thrilled; selling music quickly became far more profitable than it had ever been. Record buyers realized benefits as well, since a twelve-song album cost considerably less than twelve singles would have cost. But record buyers also assumed more risk than they had before: the likelihood that they would like all twelve of an album’s songs as much as the one heard on the radio was low. Every music lover has had the experience of being deeply disappointed by an album that was purchased on the strength of a great single. When compact discs took over from vinyl LPs in the late 1980s, the possible length of an album had increased (from about 45 minutes to 80), but the fundamental, albumbased marketing model remained virtually unchanged.
Now let’s consider the scholarly journal. For centuries, journals were printed publications and were therefore subject to all the physical limitations of print. Since paper is expensive and heavy and hard to distribute, journal articles had to be gathered into batches before they could be printed and sent out to subscribers — selling articles individually wasn’t feasible. This meant that the only way for researchers to get access to the articles they wanted was to buy articles they didn’t want. It was kind of like buying albums — only the journal was even more of a gamble. Subscribing to a physics journal because you had a research interest in physics wasn’t like buying a country album because you liked a particular song on it; rather, it was like asking a country music label to send you every album it released because you liked country music in general. The problem with such an arrangement would be obvious: while every album would probably have one or more songs you did like, each would also have songs you didn’t like and wouldn’t have paid for if you could have picked them out one by one. The same was, and remains, true for journals: very few people read every article in every issue of the journals they subscribe to. Instead, their subscriptions act as a kind of security blanket — a guarantee of access to some of what you want, secured by the simultaneous purchase of what you don’t.
Obviously, the physical culture surrounding the acquisition of both music and journal content has changed radically in recent years, and for a single reason: the Internet, which has liberated both kinds of information from the constrictions of physical format, thus making it possible for both songs and articles to be sold in the way that makes the most sense: by the piece. What’s interesting, though, is how completely the music marketplace has changed in response to this development, and how little the fundamental structure of the scholarly information marketplace has changed. In the music realm, we have moved very quickly back to the model that prevailed between the 1920s and the 1950s, when the basic sales unit was the song. Yet even though scholarly journals have moved aggressively out of the print environment and into the online one, we — publishers and librarians alike — still behave as if the basic unit of scholarly research were either the journal title or the journal issue. Yes, it’s possible to buy articles individually, but publishers strongly discourage such purchasing by pricing them at punitive levels.
And why wouldn’t they? After all, journal publishers currently make much of their money by selling you what you don’t want — which is not a criticism of journal publishers, only of the traditional, print-based model of article distribution that still prevails. The problem for publishers is that if libraries and end users were to buy only what they wanted, many of them would go out of business. That’s exactly what happened in the music industry, as record label executives knew it would. The music industry didn’t change dramatically because visionary label executives saw change coming and responded with innovation; it happened completely against the will of those executives, who were dragged kicking and screaming (and suing) into this new reality. They knew that many if not most of them would go out of business if they went back to selling songs at $1 apiece rather than CDs at $18 apiece, and that’s exactly what has happened.
This, of course, begs a question: what forced the record labels into accommodation with the new reality, and why has nothing forced journal publishers into a similar adjustment? I think there are two main explanations.
First, there are a lot more music listeners than journal readers, which means that piracy has a greater impact and is therefore a much greater threat in the music marketplace. When record labels tried to force their customers into continuing to buy physical CDs, the customers simply resorted to a strategy that combined active resistance (large-scale piracy) with passive resistance (simply not buying CDs anymore). In the world of scholarship, audiences are smaller and acts of piracy unlikely to cause as much damage. A band like U2 has tens of millions of fans who, given the opportunity, will happily upload unauthorized copies of U2 songs to the Internet, where anyone who wishes to can download them for free; when they do so, the impact on U2’s record label is significant. Relatively few people are interested in histology articles, and those who are interested in them are very frequently served by libraries and do not have to pay for them directly anyway. Satisfied customers who are small in number don’t pose much of a threat to the established order of publishing. None of this is to say that piracy of scholarly information doesn’t matter or is without effect — only that its economic impact is very small potatoes compared to the impact of music piracy.
Second: libraries, and the institutions they serve, are still measured and evaluated according to criteria left over from the print era. Suppose that a major research library were to negotiate reasonable per-article pricing from the five major science publishers, and thereby make everything from those publishers’ journals available to its patrons without any actual subscriptions (or a Big Deal whereby all articles are purchased ahead of time). How would the library report its ARL statistics? How would the university explain the situation to accreditors, who want to simply count subscriptions and bound volumes so they can go home? No library wants to be a pioneer in this area — there’s no percentage in being known as the university that finally challenged the accreditors. Much better to wait for some other university to do that.
Will we ever get to a rationally-structured scholarly information economy? I don’t know — there are many factors at work beyond those mentioned above. The fact that scholarly articles tend to be written because authors need to write them (for promotion and tenure) rather than because anyone demonstrably wishes to read them, for example, surely figures into the equation somehow as well. One thing I can say with confidence, though: an economic model that relies on people being forced to buy what they don’t want is going to fail sooner or later. The current scholarly communication model has lasted a good long time, despite its fundamental irrationality — but cracks are starting to show.
Leah was appointed Executive Director of the Charleston Conference in 2017, and has served in various roles with the Charleston Information Group, LLC, since 2004. Prior to working for the conference, she was Assistant Director of Graduate Admissions for the College of Charleston for four years. She lives in a small town near Columbia, SC, with her husband and two kids where they raise a menagerie of farm animals.