Computer Virus
[14 Jul 2020. Part of an ongoing essay on computation, capitalism, and Covid-19.]
In 1982, in a small suburb outside of Pittsburgh, the first known computer virus was seen in the wild. Spread by floppy disk, ‘Elk Cloner’, as the virus called itself, broke out in a community of early computer gamers centered around the local high school. But the virus soon began to infect people outside the community, reaching teachers’ instructional computers and eventually escaping to circulate in the Pittsburgh area for years. News of the outbreak soon reached Scientific American and Time Magazine, becoming the first ‘virus’ to be nationally reported.
The virus was simple and clever. If an infected disk were inserted into an Apple II computer (then popular with gamers for its color screen), the virus would copy itself to the machine’s core memory, or ‘boot sector’. Then, it did nothing. Incubating and infectious, it lay in wait for other disks to be inserted; then, it would silently copy itself onto them, too. If one of these newly infected disks were inserted into another computer, the process would begin anew inside a new ‘host’ machine. Only on the fiftieth reboot of an infected computer would the virus reveal itself, causing its one and only ‘symptom’—blanking the screen and printing the following few lines:
Elk Cloner:
The program with a personalityIt will get on all your disks
It will infiltrate your chips
Yes, it’s Cloner!It will stick to you like glue
It will modify RAM too
Send in the Cloner!
It was all just a prank—and a poetic one at that. Programmed by a 15-year-old high schooler, Elk Cloner was Richard Skrenta’s masterpiece of petty revenge. A well-known trickster in the local video game community, Richard Skrenta had already acquired a reputation for sneaking disruptive code onto the floppy disk copies of games which others received back from him. Friends began to stop sharing games with him, and Skrenta—perhaps an early example of that perverse blend of Ken Kesey and Ayn Rand so distinctive of Silicon Valley and early computing culture—plotted revenge. But how, he asked himself, could he continue to get his code onto his friends’ computers, without having direct, tangible access to their disk drives?
The eventual form of that solution in a ‘virus’, spreading from computer to computer in the open air through floppy disks, was unprecedented. It alarmed experts in the nascent computer security industry, who until then were familiar primarily with ‘worms,’ ‘Trojan horses,’ and ‘logic bombs’: non-replicating programs which infiltrated a single computer or mainframe, typically wiping its data or clogging its memory. On April 15th, 1985, for instance, a ‘logic bomb’ detonated in the IBM mainframe of the Los Angeles Department of Water and Power (DWP), freezing access to its files and grinding work to a halt. When the DWP took the unusual step of calling the police, which had established a cybersecurity division only a month earlier, twenty experts arrived to repair the sabotage, taking a week to bring the system back online; the suspect, likely an employee, was never apprehended.
A rising problem for industry, cybercrime afflicted about a quarter of Fortune 500 companies in the 1980s, bleeding about $750 million in damages. Before the rise of the Internet, which connected the various, generally exclusively academic or military networks like ARPAnet, MILnet, R&Dnet, Usenet, and Bitnet, ‘hacking’ computers remotely via digital networks was rare, confined to dialing by modem directly into an institution’s private network. ‘Worms’ and ‘bombs’ placed in the computers from the inside were much more common. In 1982, for instance, the same year as the Elk Cloner virus, a logic bomb was discovered in the computer systems of Collins Foods International Inc., operator of more than 200 Sizzler and Kentucky Fried Chicken restaurants across the U.S. and Australia. Although never detonated, the bomb was discovered buried in the firm’s code by another programmer and reported to the police; Dennis Williams, a 33-year-old disgruntled programmer at the firm, was later arrested and convicted of new statutes against cybercrime. In the trial, the L.A. Deputy District Attorney charged Williams of planting the logic bomb because he was ‘dissatisfied with the company and in particular, one employee, a supervisor.’ Such attacks were in fact often made by disgruntled programmers fed up with poor working conditions—auguring the first wave of casualized employment in the computer industry, about which more below.
A cybersecurity concept of the ‘virus’, attacking not by direct sabotage but through indirect contamination, did not therefore fully consolidate until as late as 1987, when the first conference on ‘artificial life’ was held at the Los Alamos National Laboratory—the site of the infamous Manhattan Project and of a number of early computing advances, including that of self-replicating programs or ‘automata’. As Jussi Paraka writes in Digital Contagions, self-replicating programs therefore have a longer history (discussed further below), with ‘more widespread roots in … modernity than the security-oriented notion a computer virus implies.’ But in the 1980s, the closer, more explicit context for understanding the emergence of a computer ‘virus’ was not in fact these military-academic roots, but rather the far more immediate crisis of the ongoing AIDS epidemic. In 1981, a year before the first computer virus was discovered, the HIV virus began to emerge in the gay communities of New York, Los Angeles, and San Francisco. ‘Last week there were rumors that an exotic new disease had hit the gay community in New York,’ wrote Dr. Lawrence Mass, infamously, on May 18th in the New York Native. San Francisco, whose Chronicle was praised for its accurate reporting on computer viruses and worms, was home, at the same time, to the very different kind of viral epidemic of HIV.
Computer viruses therefore arose historically alongside biological viruses, becoming legible only against the backdrop of the spread of HIV. ‘Computer AIDS’, ‘digital hygiene’, ‘safe hex’ were all phrases used to understand the possible dangers of a new computer epidemic. One 1990 article commented on the metaphorical conjunction:
Like the AIDS virus, the computer ‘virus’ attacked the ‘immune systems’ of the computers it invaded. Like AIDS, the computer ‘virus’ spread through exchanges between between individuals. Instead of body fluids being exchanged through sex, it was software exchanged by electronic mail. … In both cases, the entire society proved to be at risk because of the actions of a subculture.
Both epidemics occurred on the fringes of normative society, initially ‘invisible’ through their unfamiliarity (computer culture) or social repression (HIV). Both involved intimate and, to many, frightening forms of vulnerability—disk copying, homosexual sex—which, formerly secluded in a corner of the social landscape, suddenly seemed to threaten the rest of society through their interconnectedness with the totality of social exchanges and relations.
A virus must therefore be understood as primarily a social relationship. Transcending the medium of its transmission, whether data transfer or sexual contact, the emergence of any virus reveals the network of social relations that is its true ‘body’ and the true medium through which it moves. In April of this year, as Boris Johnson lay ill with Covid-19, he felt compelled to correct the infamous assertion of Margaret Thatcher: ‘there really is such a thing as society’, he said from a video message sent from self-isolation. Even for Johnson, a Tory invested in the primacy of individual liberty, a virus makes plain the networks of transmission and circulation out of which ‘society’—for Thatcher, an imaginary abstraction—is, in fact, terrifyingly real.
The arrival of Covid-19 has made this more apparent than ever, just as the prevalence of 5G conspiracy theories—some of which claim that 5G radio waves directly transmit the virus—show that the social imaginary of the virus continues to blend biological and computerized notions of disease. Covid-19 also shows that quasi-racialized fears of globalization (centering on Huawei, a Chinese telecommunications company and the manufacturer of 5G cellphone towers) underlie both imaginaries, with the U.S. President Donald Trump continuing to call the coronavirus the ‘China virus’. Finally, as will be explored below, the initial formalization of self-replicating ‘automata’ in 1953 preceded by only a few months the discovery of DNA, which itself was explicitly understood in relation to computer code. A biological virus, in this sense, was itself a rogue program, a set of copying instructions designed through natural selection to replicate itself efficiently. More than simply a convenient metaphor or analogy, then, the conceptual and cultural overlap between computer and biological viruses like Covid-19 reveals something about the sociology of viruses, computers, and, as I will argue, the current computational regime of capital accumulation, which neither alone could show.