Table of Contents
Berltolt Brecht scholars, even more than than literary scholars in general, face urgent challenges to their work from information technology and its legal and institutional frameworks. Brecht's work is entangled with the law of print and performance copyright, and with technical issues of Internet distribution, audio/visual rendering, and long-term digital storage. Scholarship takes place inside academic and governmental structures devoted to their own diverse agendas. This essay addresses how scholars and other communities of readers can extract maximum benefit from available technology, while at the same time evading the legal and less-legal snares set to extract maximum profit from information.
To the extent that we academics can survive the political and environmental crises we share with the rest of the world, our distinctive social function, alongside artists and many other allies, is to resist the erosion of social commitments that are not native to capitalism. We transmit culture, elaborate it, promote academic freedom and the free exchange of ideas, and defend social institutions that allow these principles to flourish. The following table lays out the minimal conditions that scholars need to do their work, organized by access to information and the possibility of collaboration.
Table 1. What scholars need
Access means the availability of materials and tools unencumbered by burdensome fees or institutional gate-keeping. Collaboration refers to the essentially social nature of learning and art, which far from being "ivory tower" activities are ideally encounters among diverse artists, performers, scholars, students, and audiences of all social classes and bringing every conceivable intention and expectation to bear in their artistic engagement. The key benefit conferred by the Internet and by digital distribution is the potential to remove practical barriers to access and collaboration.
Compatibility with these needs must be the criterion when evaluating a technology for an academic project. Sound judgment requires an understanding of the tools available, ranging across Web sites, discussion lists, social media, chat platforms, collaborative online office tools, online publishing of all sorts, video conferencing, online library holdings, online video, online teaching, online performance, analysis and visualization of data big and small, and pattern-recognition algorithms. However, these technologies exist within governmental, institutional, and commercial contexts that confer varying degrees of freedom and restriction on scholars. The following diagram illustrates some of these relations.
Both axes in this diagram indicate degrees of control, favoring either the scholar or someone else (e.g., a lawyer, a bureaucrat, a dean, or a corporation). The horizontal axis measures access to information (“openness”). The vertical axis measures control by users over the technology, ranging from freely available “tools” to contractually restricted “services.” Each segment of the gray bar (Non-commercial, NGO, etc.) provides typical distribution modes above the bar and specific real-world example below the bar. At lower left, a maximum degree of both freedoms is obtained by using the open World Wide Web to publish material unencumbered by restrictive copyright. At upper right, the least degree of both freedoms is available to users of commercial Internet services and publishers profiting from traditional restrictive copyright. A scholar with advanced technical skills might choose to build custom tools and services, providing greater freedom. A less technical scholar might contract with an NGO or business to use a ready-made service. The choices are not crudely “freedom good, commercial bad” – all choices involve trade-offs. The important thing is to understand what is being traded off.
The challenges to scholarship that computer technology creates or complicates are too numerous to list exhaustively, but they can be categorized as follows.
The problems caused by copyright are mainly legal, not technological, but digitization of content and potential for easy distribution over the Internet have made the legal issues far more acute. Current legal theories are out of step with current technology, despite all the contorted and ineffectual measures to restrict file-copying, collectively known as “digital rights management.”
Like copyright, physical and bureaucratic restrictions on information flow are not new. Governments, universities, and businesses all have strong motivations to hoard and hide knowledge. Computers, however, provide endless new opportunities to do this for power and profit.
Librarians are painfully aware of how hard it is to choose a file format today that will still be readable in 10 years. The go-go-go business culture of the tech industry is littered with antiquated formts, dead companies, abandoned software, and undocumented APIs that can ruin the day of anyone depending on them.
The technological traps aren’t just accidental. Technology companies scheme routinely, and sometimes illegally, to make it painful or impossible for their customers to jilt them in favor of a competitor.
In the confines of this brief paper, I offer two examples of these threats as they have played out at the intersection of technology and creative arts.
The most lucrative business model in a data economy is one where the company owns everything and the customer owns nothing. That way, the company has no need to produce and sell a product, just a service, which is “licensed,” not sold. Offering a service is almost a good as possessing a monopoly, because only the service provider possesses it, while the customer receives an intangible that is consumed as soon as used, and is safe from the possibility of accumulation or resale. In cloud services, this model is called Software as a Service (SaaS).
For example, Microsoft began as a seller of boxed software—an old-fashioned physical product. It then cut an exclusive licensing deal with IBM for preinstalled operating system software. When cheap clones of IBM hardware grew to dominate the personal computer market, Microsoft software remained the single common component, leading to monopolies in operating system and productivity software (such as Office) that endured through legal and illegal means for 20 years. This ended only when mobile computing, and remotely offered services, began making standalone PCs and the software running them obsolete. Seeing its pre-installed operating system monopoly dwindle, Microsoft doubled down on its productivity business, reinventing itself as a cloud company for “office services.” Office became Office 360, a cloud service. You no longer own anything, and you pay a subscription fee … forever. There’s no longer a need to tout pointless upgrades, because customers either pay the monthly utility bill or they get cut off entirely.
Something similar happened in the music industry. Enticed by lower production costs and promising customers improved audio quality, the record companies embraced digital technology in the 1980s in the form of the Compact Disc. Vinyl nostalgia notwithstanding, that format was hugely popular, and within a decade, every analog music recording with potential for future sales had been remastered to CD. Yet the industry made committed a tactical error: CDs were unencrypted. The ease of perfect duplication that made digital audio technologically attractive also, inevitably, made it convenient for file-sharers. The industry fought file-sharing for years by suing its own customers, but their profits plummeted. Then Apple iTunes introduced license-based, rather than product-based sales, a business model in turn overtaken by product-free streaming services. The result, still evolving, might be called Music as a Service, improving over prehistoric broadcast radio by directly charging money for every listen.
What does this have to do with humanities scholarship? Suppose a project such as a critical edition is entirely unencumbered by copyright. The researchers contract with a software provider to host their data on the web and provide online editing and collaboration tools. Alas, the company goes out of business after five years, and their database structures disappear with them, while their server software is sold to Uber, which sees vague but huge business opportunities. The researchers consult the contract, and learn that they now possess nothing but the original data they provided, and that only because they made a backup. (But now I stray into pure fantasy.)
Enough fantasy; now for a literary nightmare. In the 1970s and 80s, a team of German textual scholars led by Hans Walter Gabler assembled an ambitious new computer-aided edition of James Joyce’s 1922 novel Ulysses. The textual history of Ulysses had been vexed from the start by Joyce’s slow and accretive style of composition. On every typescript or proof returned to him, he crammed the margins full with revisions in a barely legible hand, leading to copy errors at every stage, including some by Joyce himself. The trade editions inherited a famously corrupt original text, and all of them added still more errors of their own. The following galley proof of the the novel's final page gives a sense of the problem.
Gabler and his team attacked this challenge by creating a “synoptic edition”, a genetic text that collated and compared every extant note, manuscript, typescript, galley proof, edition, and errata list that Joyce's authorial hand had touched. After these materials were digitized and appropriately tagged, the team used custom software to generate a continuous authoritative text as well as a “synoptic” text that displayed the revision history. The continuous text, which Gabler touted as containing more than 5000 corrections (Gabler, vii), was published as a new trade edition by Random House in 1986. The image below from the synoptic edition shows the genetic text on the left and the matching continuous text on the right.
(Joyce, Ulysses: A Critical and Synoptic Edition, Vol. 1, pp. 448-349)
The Gabler edition was controversial for its methodology, accuracy, editorial decisions, and motivations. Its detractors suggested, not unreasonably, that financial backing from the Joyce estate and Random House was aimed at establishing lucrative new copyrights extending well into the next century. Certainly Random House stood to corner the worldwide market for the novel, and after a fashion it succeeded—at the cost of trashing Gabler's reputation and harming the entire field of Joyce studies. For the publisher, the 5000 corrections in Gabler's Ulysses were just a marketing bauble. After rancorous debate by textual scholars in the pages of the New York Review of Books (Kidd, "The Scandal of Ulysses"; Staley, "The Continuing Scandal of Ulysses"; Rossman, "The New Ulysses: The Hidden Controversy"; Groden, "Ulysses Update"; etc.), and after WW Norton jumped in to promise another, competing correction of Ulysses, Random House abandoned the Gabler edition and returned to selling reprints of its flawed 1961 edition.
Looking back at that pre-Web era, my saddest reflection is that Gabler's project simply ended once financial support ended. The data and software never saw the light of day. Whatever its flaws, the project was a groundbreaking achievement in digital humanities that could have been refined and built upon. Today, if unencumbered by copyright, it would be an ideal candidate for massively distributed scholarly collaboration of a kind unimaginable in the 1980s. But it is encumbered by copyright. If viewed as “a derivative work made for hire by Hans Walter Gabler for the Estate of James Joyce, the U.S. copyright in the edition would last for 95 years from the year of first publication, or until 1 January 2080” (Saint-Amour, 756).
When I moved out of academic work and into the computer industry, one of the first things I learned was that software has been an "intellectual property" battleground. Unlike artifacts of a natural language, such as a play by Brecht, a computer program has two aspects: the source code, which can be written and read by trained human beings, and machine code, which is read and acted upon by computer hardware. As a result, software lends itself to the creation of artificial scarcity. If you hide the source code from customers, giving them only the "compiled" machine code, you can prevent them from studying how it works, modifying it, and creating new value from it. This constitutes a business model.
Software with more or less freely available source code predates this model. Early programmer culture emerged from military and academic environments that were insulated from market incentives. The commercial computer industry of the 1950s and 60s concerned itself with selling super-expensive hardware to governments and corporations, giving little thought to software itself an object of value. The words, symbols, ideas, and algorithms of software became commoditized only in the 1970s and 80s after computers grew more affordable and interchangeable. Once computer code was understood as something worth locking away, companies like Microsoft turned software sales into big business. And so it came to pass that proprietary software development replaced collaborative development with a regime of patents, trade secrets, and copyright restrictions.
When programmers accustomed to freely acquiring, studying, distributing, and modifying source code found themselves locked out by licensing agreements, some of them, led by MIT researcher Richard Stallman, came up with a creative solution. First of all, and rather insanely, they rewrote from scratch the dominant server software of the day, Unix, producing their own copyrightable code. Second, and rather brilliantly, they hacked the copyright system into what is waggishly referred to as copyleft, and more formally as free or open-source software. Whereas traditional application of copyright prevents non-owners from copying and redistributing materials, copyleft allows anyone to copy, modify, and redistribute, but not to hoard the improvements they make. The primary goal was to protect the freedom of programmers to program collaboratively. The first and classic example of this model is the GNU General Public License, or GPL, under which Stallman and his colleagues licensed their new software. Stallman grounded his copyleft philosophy in the “four freedoms”, which have an obvious affinity to the access and collaboration required for scholarly work.
Richard Stallman's Four Freedoms
A program is free software if the program's users have the four essential freedoms:
Freedom to run the program as you wish, for any purpose.
Freedom to study how the program works, and change it so it does your computing as you wish.
Freedom to redistribute copies so you can help others.
Freedom to distribute copies of your modified versions to others. By doing this you can give the whole community a chance to benefit from your changes.
(Stallman, "What is free software?")
Importantly, the meaning of "free" in the term "free software" has to do with freedom of action, not price. Stallman's freedoms do not preclude the sale of software or of technical support for it, but neither do they enshrine profit as a preeminent value. Money can be made directly from free software only if its informational content is fully disclosed and remains open to unrestricted redistribution and modification.
The copyleft software model was so successful that today, 30 years after the GNU GPL was introduced, it protects the main software powering the Internet (GNU Linux) as well as the world's most popular mobile operating system (the Linux offshoot Android). This success derives from the paradox of free software's profitability. Commercial software giants such as Google, Amazon, Microsoft, and IBM contribute funding and expertise to the ongoing development of Linux because broad collaboration ultimately spares more costs than it consumes.
While the GPL is intended to protect computer programs, the reinterpretation of copyright through the lens of the four freedoms has been generalized to cover a much wider range of intellectual property scenarios. Most notably, the Creative Commons project provides a comprehensive framework of licenses ranging from traditionally restrictive copyright to completely unrestricted distribution. The image below is a representation of the various Creative Commons licensing icons, and the accompanying text is Wikimedia's licensing statement for the image, itself licensed under Creative Commons. This particular license confers rights on the user that are approximately equivalent to those provided by the GPL.
(Wikimedia, File: Creative commons license spectrum.svg)
Applied to academic publishing, copyleft licenses can protect new material that scholars create, including new editions of primary literature that has fallen into the public domain. As the Brecht copyrights approach expiration, the Brecht estate and its allied publishers will attempt to create new editions with new copyrights that will guarantee profit and control for another century. They require scholarly labor to achieve this. I urge Brechtian to drive a hard bargain with publishers and universities. Don’t contribute to projects that lock away intellectual work. Use copyleft to build a distributed, high-participation, computer-aided future of open scholarship.
The following table maps the previously described threats to their potential mitigations.
Table 2. Mitigating the threats
Gabler, Hans Walter. "Foreward." 1984. In Joyce, James. Ulysses: A Critical and Synoptic Edition. Pp. vii-viii.
Groden, Michael, reply by John Kidd. Ulysses Update." New York Review of Books. February 2, 1989. Vol. 36, No. 1. URL: https://www.nybooks.com/articles/1989/02/02/ulysses-update/. Retrieved 25 March 2020.
Joyce, James. Ulysses: A Critical and Synoptic Edition. Ed. Hans Walter Gabler et al. New York and London: Garland Publishing. 1984. Three volumes.
Joyce, James. Ulysses Placards [galley proofs]. 1921. MS Eng 160.4. Houghton Library, Harvard University. Digital material. URL: https://iiif.lib.harvard.edu/manifests/view/ids:44067970. Retrieved 25 March 2020.
Kidd, John. "The Scandal of Ulysses." New York Review of Books. June 30, 1988. Vol. 35, No. 11. URL: https://www.nybooks.com/articles/1988/06/30/the-scandal-of-ulysses-2/. Retrieved 25 March 2020.
Rossman, Charles. "The New Ulysses: The Hidden Controversy." New York Review of Books. December 8, 1988. Vol. 35, No. 19. URL: https://www.nybooks.com/articles/1988/12/08/the-new-ulysses-the-hidden-controversy/. Retrieved 25 March 2020.
Saint-Amour, Paul K. et. al. "James Joyce: Copyright, Fair Use, and Permissions: Frequently Asked Questions." "James Joyce Quarterly. Vol. 44, No. 4 (Summer, 2007). Pp. 753-784. Accessed via JSTOR. URL: https://www.jstor.org/stable/25571082. Retrieved 25 March 2020.
Smithies, James. The Digital Humanities and the Digital Modern. Palgrave Macmillan. 2017.
Staley, Thomas F., John O'Hanlon, and Hans Walter Gabler, reply by John Kidd. "The Continuing Scandal of ‘Ulysses’: An Exchange." New York Review of Books. September 29, 1988. Vol. 35, No. 14. URL: https://www.nybooks.com/articles/1988/09/29/the-continuing-scandal-of-ulysses-an-exchange/. Retrieved 25 March 2020.
Stallman, Richard M. "What is free software?" URL: https://www.gnu.org/philosophy/free-sw.en.html. Retrieved 25 March 2020.
Wikimedia. "File:Creative_commons_license_spectrum.svg example." URL: https://commons.wikimedia.org/wiki/File:Creative_commons_license_spectrum.svg.