The Illusion of Control: History and Criticism of Interactive Television

Interactive Communication

The idea of television as a two-way communication medium actually predates the invention of television itself. Verity Hunt (2014, 58) notes, for example, that the invention of the telephone inspired various imaginary media that were designed to “do for the eye what the telephone had done for the ear,” such as the “telephonoscope” (Maurier 1878; Robida 1883), the “telectroscope” (Figuier 1878; Twain 1898), the “kinototelephotograph” (Wells 1899), and the “telephot” (Gernsback 1909). Hugo Gernsback (1909, 403) also emphasized that interactivity should be an essential feature of all television systems, as television sets “must be reversible, that is, each instrument must receive as well as transmit.” This early notion of television as a two-way communication medium was not necessarily seen as positive, however, as some writers also speculated that it might be used for nefarious purposes. For example, George Orwell’s (1950, 3) novel 1984 famously depicted a dystopian future in which every home is equipped with a television set that is capable of both transmitting and receiving audio-visual information:

The telescreen received and transmitted simultaneously…. There was of course no way of knowing whether you were being watched at any given moment. How often. . .the Thought Police plugged in on any individual wire was guesswork. It was even conceivable that they watched everybody all the time. But at any rate they could plug in your wire whenever they wanted to. You had to live. . .in the assumption that every sound you made was overheard and. . .every movement scrutinized.

Orwell thus conceived of television as an instrument of propaganda and surveillance, which showed how interactivity could potentially reinforce rather than challenge passive viewer reception. ³

Despite Orwell’s warning, artists often promoted the use of television as a two-way communication medium that would allow users to have more control over program content. For example, Nam June Paik’s Participation TV (1963–1966) was an interactive television installation that displayed bundles of lines, which spread out to form various patterns whenever someone spoke into a microphone. In later versions, the television was also hooked up to a camera, which recorded images of the audience that were then incorporated into the screen display. Unlike Orwell, therefore, Paik conceived of interactive television as a medium of artistic collaboration rather than state surveillance, and he even claimed (qtd. in Lovejoy 1989, 237) that “the best way to safeguard against the world of Orwell is to make this medium interactive so it can represent the spirit of democracy, not dictatorship.” Allen Kaprow’s experimental video art program “Hello” (1969) also enabled interactivity by connecting four remote locations via a closed-circuit television network, as Gene Youngblood (1970, 343) explains: “Kaprow functioned as ‘director’ in the studio control room. If someone at the airport were talking to someone at M.I.T., the picture might suddenly switch and one would be talking to doctors at the hospital.” Douglas Davis also created a series of pieces that explored alternative uses of television as a two-way communication medium, such as “Electronic Hokkadim” (1971), a thirty-minute live broadcast that invited viewers to participate by calling the station, as the wave patterns of their voices would modify the images on the screen. ⁴ Like Enzensberger, he also noted (qtd. in Ross 1975) that “the decision to manufacture radio sets as receivers only was a political decision, not an economic one,” and “the same is true of television; it is a conscious (and subconscious) decision that renders it one-way.” These artists thus promoted a vision of television as a two-way communication medium that would allow viewers to have more control over program content, although their involvement was heavily mediated by the artists themselves.

The idea of television as a two-way communication medium was also the basis of J. G. Ballard’s (1982, 195–205) short story “Intensive Care Unit,” in which all social interaction is mediated by television, including education, work, marriage, and parenting. Despite the fact that the narrator has never physically touched his wife and children, for example, he insists (203–204) that intimacy is only possible through the screen:

True closeness. . .was television closeness—the intimacy of the zoom lens, the throat microphone, the close-up itself. On the television screen there were no body odours or strained breathing, no pupil contractions and facial reflexes, no mutual sizing up of emotions and advantage, no distrust and insecurity. Affection and compassion demanded distance. Only at a distance could one find that true closeness to another human being which, with grace, might transform itself into love.

Television thus produces a sense of intimacy by inhibiting rather than facilitating genuine social interaction, as Jeff Marker (2011, 341) explains: “The voyeuristic distance provided by televisual mediation fundamentally changes human communication from interactions between person and person to exchanges between image and spectator. People no longer interact with each other, but view each other’s voice- and visual-images.” Ballard’s story thus shows how interactive television could still replace genuine social interaction with “para-social interaction,” as viewers effectively transform themselves into programs that are passively consumed by other viewers.

The idea of television as a two-way communication medium remained influential until the end of the century. For example, Frederick Williams et al. (1988, 10) defined interactivity as “the degree to which participants in a communication process have control over, and can exchange roles in, their mutual discourse.” Melvin L. Defleur and Sandra J. Ball-Rokeach (1989, 341) similarly defined interactive media as media “that take on some of the characteristics of interpersonal communication.” However, some critics also argued that this definition was fundamentally misguided. According to Sheizaf Rafaeli (1988, 128), for example, it “could miss the essence of development by diverting it toward a standard that is inappropriate and perhaps unnecessarily limiting.” Instead of promoting the idea that interactivity would encourage active participation by transforming television from a one-way transmission medium into a two-way communication medium, critics gradually began to embrace a new concept of interactive television that was inspired by the industry trend toward interactive systems.

Interactive Systems

The second way interactivity was imagined was based on the ability of viewers to interact with program content by 1) selecting various options, such as accessing background information (text layering), viewing more than one program at the same time (nested screens), and choosing different display features (aspect ratios, subtitles, advertisements, etc.) or 2) providing information to producers through a return channel (casting votes, participating in on-screen quizzes, ordering advertised products, etc.). This new concept of interactivity effectively replaced the idea of televisual flow—a programming structure that consisted of an endless stream of unrelated segments—with the idea of the televisual database—a programming structure that offered a dizzying array of inputs and outputs. The proponents of this concept thus imagined that the television set would no longer simply be used to watch programs; rather, it would become a home terminal that provided access to various goods and services.

This notion of interactive television was first introduced by Alvin Toffler (1970, 282), who famously predicted the “demassification” of television: “Technology, far from restricting our individuality, will multiply our choices—and our freedom—exponentially.” Toffler (1980, 178) also promoted the Japanese television system HI-OVIS (Higashi Ikoma Optical Visual Information System) as the first interactive television service—not because it enabled two-way communication but rather because it allowed viewers to select program content: “Hi-Ovis. . .keeps a bank of video cassettes on everything from music to cooking to education. Viewers can punch in a code number and request the computer to play a particular cassette for them on their screen at whatever hour they wish to see it.” In addition to this video request service, users could also participate in public opinion polls, quiz programs, and charity auctions through televoting. The system also enabled home shopping and home banking, and users could even attend video lectures, ask questions, and take tests from the comfort of their own homes.

Toffler’s other example was Warner-Amex’s Qube system, which was also promoted as the first interactive cable television service. ⁵ Qube provided subscribers with thirty different channels and a wide range of specialized programming for specific target audiences. Each console was also furnished with a hand-held device that allowed subscribers to communicate directly with the station. Like HI-OVIS, therefore, Qube was considered interactive because it allowed viewers to choose from a wide range of options and to send back signals when prompted for feedback, which included feedback on individual programs (a new way of calculating television ratings) as well as consumer surveys. Viewers could also purchase merchandise displayed on the screen, and they could watch several original interactive programs, such as the children’s series Flippo’s Magic Circus, which featured interactive games; the talent show Talent Search, which allowed viewers to rate on-screen performers; and the game show How Do You Like Your Eggs?, in which contestants predicted how viewers would respond to various survey questions. The promotional material for the system (Rogers 1986, 62) boldly announced that “the age of passive viewing is over,” and newspapers similarly declared (Anon 1978) that viewers were now “talking back to television.” It is important to note, however, that viewer participation was largely limited to selecting programming options and responding to polls, the results of which were then sold to advertisers or political parties.

Several interactive videotext services were also introduced in the 1980s, including Indax, Teletext, Viewtron, and Gateway, which similarly offered games, news, home banking, home shopping, ticketing services, and information services (such as access to airline schedules, real estate listings, product information, etc.). The first video-on-demand service, Stargazer, was introduced in the mid-1990s, and it was followed soon after by Warner’s Full-Service Network, which offered a wide range of programs as well as video games and home shopping. Warner’s chairman, Gustave M. Hauser (qtd. in Fraser 1977, 32), famously declared that “we are entering the era of participatory as opposed to passive television.” Unlike the older concept of interactivity, however, Hauser was not proposing that television would facilitate user-user interaction; on the contrary, he was merely referring to user-system interaction, which served the interests of producers by preserving their control over program content and by forcing viewers to pay for the privilege of providing valuable market research data.

These interactive television systems prefigured the rise of internet television in the late 1990s. In 1998, for example, SimplyTV made 20 television shows available on the internet and announced that it would make a thousand additional shows available the following year. The web design firm Virtual Melanin also produced a series of music videos, live performances, and “magazine” format television shows, which were streamed on the website Cafe Los Negroes—a “virtual hangout” for communities of color. The most prolific and long-lasting internet television provider was Pseudo Programs, ⁶ which began streaming original content in 1997 and gradually developed into a multi-channel network that featured both prerecorded and live television shows. The most popular channels targeted niche audiences by focusing on hip hop, video games, professional wrestling, computer hacking, and spoken-word poetry, and these channels produced roughly 240 hours of programming each month. Like HI-OVIS and Qube, therefore, Pseudo Programs offered a wide range of content for smaller audiences whose interests were not being served by the major networks, and most of these programs incorporated audience interaction through an online chat function, which allowed producers to respond directly to the needs and desires of viewers. As one commentator (Anon 2000, 7) explained, “they imagined a world where programming wasn’t just spoon-fed to the viewing public but instead involved the viewers’ active participation.” Other early internet television providers included Digital Entertainment Network, which released thirty interactive programs aimed at various subcultures (Kaplan and Huffstutter 1999), and WireBreak Entertainment, which released four interactive shows aimed at younger viewers, including “a customizable news program,” “an interactive video-comedy,” and “a girls’ gab-fest that responds to e-mail questions” (DeSantis 1999).

While many of these companies were short-lived (Virtual Melanin, Pseudo Programs, and Digital Entertainment Network all filed for bankruptcy following the dot-com crash of 2000), the idea of television as a database gradually became the industry standard, and it was accepted by many critics as liberating and empowering. For example, Richard Schwier (1987, 172) reinforced the definition of interaction as “a reciprocal dialogue between the user and the system,” and Nikhil Hutheesing (1993, 244) similarly defined interactive television as a “mechanism for delivering image, text and sound data in which the user interacts with the database.” However, this notion of interactivity was also criticized by Erkki Huhtamo (1995, 86-87), who described it as “a strategy for persuasion in a manner reminiscent of traditional television commercials,” as it “merely facilitates selling faster than mail order phone numbers on a TV screen or shop addresses in a newspaper advert.” Jennifer Stromer-Galley (2000, 118) also pointed out that this concept of interactivity was only made possible through the integration of television and digital networks, which allowed users to interact with a technological interface “without ever directly communicating with another person.” Peter D’Agostino (2012, 510) similarly noted that these interactive systems “limit feedback to the mere illusion of participation,” as viewers are unable to interact with one another. The irony of this development, in other words, is that it rejected the idea of television as a two-way communication medium at the very moment when this idea had finally become feasible through the development of digital networks.

Interactive Narratives

The third way of imagining interactive television was based on the ability of viewers not only to choose but also to change program content. This idea was already prevalent in the early period of television, as Ray Bradbury’s (1953, 18) novel Fahrenheit 451 provided one of the earliest depictions of a television program in which viewers play an active role, as the protagonist’s wife Helen explains:

They write the script with one part missing…. The home-maker, that’s me, is the missing part. When it comes time for the missing lines, they all look at me. . .and I say the lines: Here, for instance, the man says, “What do you think of this whole idea, Helen?” And he looks at me sitting here centre stage, see? And I say, I say—” She paused and ran her finger under a line on the script. “I think that’s fine!” And then they go on with the play until he says, “Do you agree to that, Helen?” and I say, “I sure do!”

This program is clearly designed to transform passive viewers into active participants by allowing them to become directly involved in the narrative, yet Bradbury emphasizes that this interactivity is fundamentally illusory, as the viewer’s contributions have no impact on the story or the characters. In other words, interactivity is not conceived as a feedback channel that allows for mass customization; rather, it is merely a gimmick that serves to keep viewers watching a program that was scripted and filmed in advance.

This problem can also be seen in the earliest experiments with interactive narratives on children’s television. In the 1950s, for example, the U.S. children’s show Winky Dink and You (1953–1957) sold a “Winky Dink Television Game Kit” that included a thin sheet of plastic that adhered to television screens and an erasable crayon that could be used to draw on the sheet. The program then featured a cartoon character who encountered a series of problems, and the host invited viewers to “be a part of the program” by completing connect-the-dot drawings of various objects that would help the character, such as a bridge that would allow him to cross a chasm (Carey and O’Hara 1995, 220). Viewers were thus invited to participate in the program by making objects appear on the screen. Like the program depicted in Bradbury’s novel, however, it only created the illusion of control, as the contributions of viewers had no impact on the story, which was always scripted in advance.

The idea of interactive television narratives remained largely speculative until the rise of digital media, which seemed to offer new technological possibilities. Janet Murray (1997, 126) was one of the first critics to discuss this new concept of interactive television (originally known as “cyberdrama”), which would allegedly give viewers a greater sense of autonomy and agency by allowing them “to take meaningful action and see the results of [their] decisions and choices.” She also argued (1997, 170) that this new form of interactive television would make viewers more deeply aware of the consequences of the characters’ decisions, as they would “assimilate them as personal experiences.” While Murray’s vision of interactive television was not feasible at the time, as television programs were still limited by broadcast schedules (which required a standardized format, frequency, and duration) and serial structures (which required the advance scripting of story arcs), it appealed to critics precisely because it surpassed the limitations of earlier notions of interactivity. For example, Jens F. Jensen (1999, 59) argued for a new definition of interactivity as a measure of television’s ability not to provide interactive menus or feedback channels but rather “to let the user exert an influence on the content.” Instead of simply interacting with the system, in other words, critics began to imagine a new form of interactive television that would allow viewers to modify programs in real time.

Sheila C. Murphy (2011, 102) argues that this idea of interactive television became popular following the rise of television-based video game consoles like the Magnavox Odyssey (1972), the Atari Video Computer System (1977), Intellivision (1980), and ColecoVision (1982), which introduced a “model of interactive television [that] upsets one’s expectations about passive versus active viewership, as video games entail interactive, embodied participation.” However, it did not become feasible until the rise of televoting. For example, the Finnish show Akvaario (Aquarium, 2000) allowed viewers to control the actions of a fictional character by using their telephones to vote for various narrative options. All of the options were filmed in advance, and a computer edited the program in real time by selecting scenes that corresponded to the majority opinion. This early experiment thus conceived of interactive television as a collective and collaborative experience that generated a unique narrative based on the average tastes of the viewing public. ⁷ Sydän kierroksella (Accidental Lovers, 2006–2007) was another Finnish show that similarly allowed viewers to determine how a relationship between two characters developed by sending advice and opinions via text messages. A computer analyzed the messages by identifying certain keywords, which indicated whether the romance was heating up or cooling down, and scenes were then selected based on the characters’ emotional states. The producers also scripted various endings, depending on whether one, both, or neither of the characters fell in love, and they broadcast several different versions of the program in order to allow viewers to experiment with different permutations. Like Akvaario, therefore, Sydän kierroksella was presented as a collective collaboration that was made possible through televoting, although participation was reportedly low and it was impossible to employ this technique in a serial narrative due to the tremendous number of prerecorded scenes. ⁸

The idea of customizable interactive narratives has become more prevalent in recent years due to the development of streaming services like Netflix, which is the current leader in the production of interactive television. As Amanda Lotz (2018, 118) points out, Netflix originally assumed control of the rental film market, but “the capacity to stream high-quality, longform video lent itself at least as much, if not more, to television distribution,” and “by 2016, the split in viewing was 70 percent television, 30 percent film.” Inspired by “Choose Your Own Adventure” novels, Netflix developed interactive episodes for six of their children’s series—Buddy Thunderstruck: The Maybe Pile (2017), Puss in Book: Trapped in an Epic Tale (2017), Stretch Armstrong: The Breakout (2018), Minecraft: Story Mode (2018), Captain Underpants Epic Choice-O-Rama (2020), and Carmen Sandiego: To Steal or Not to Steal (2020)—as well as three of their adult series—Black Mirror: Bandersnatch (2018), You vs. Wild (2019), and Unbreakable Kimmy Schmidt: Kimmy vs. The Reverend (2020). All of these programs feature “decision points” that allow viewers to choose between various narrative options, which is explicitly designed to transform passive viewers into active participants. Netflix executives have also echoed Murray’s claim that interactive television creates a stronger sense of character identification. As Todd Yellin (qtd. in Streitfeld 2018, 14), Netflix’s vice president of product innovation, explains: “If bad things happen, you’ll feel even more crestfallen, because you were responsible, [and] if the character is victorious, you’ll feel even more uplifted because you made that choice.” Netflix executive Carla Engelbrecht Fisher (qtd. in Newton 2017) also argues that interactivity is a natural extension of their existing platform because “Netflix is an interactive device ecosystem” that is “not beholden to terrestrial television.” In other words, Netflix viewers already have a considerable degree of control over what they watch, and executives assume that they also want more control over program content. And while Netflix has not released any ratings information for these interactive episodes, they are reportedly developing interactive programming for a wide range of genres, including horror and romance, which suggests that these experiments have been at least partially successful (if they have not succeeded in attracting new subscribers, they have at least succeeded in attracting media attention).

The idea of interactive narratives thus appears to resolve some of the problems with earlier notions of interactivity by offering viewers more control over program content, yet this concept of interactivity has also been criticized for several reasons. First, critics argue that it tends to produce confusion, frustration, and disorientation—a criticism that dates back to the early reception of hypertext fiction. For example, Jeff Conklin (1987, 19) noted that the freedom granted to the user often makes it difficult to navigate the narrative map, which makes it “distressingly easy for a user to become lost or disoriented.” Sharon McDonald and Rosemary J. Stevenson (1996, 62) similarly argued that disorientation is the “major limiting factor” of interactive narratives, and William P. Eveland and Sharon Dunwoody (2001) argued that this disorientation was due to the fact that interactive narratives require considerably more effort and concentration. Nitzan Ben-Shaul (2004, 155) also argued that interactivity leads to cognitive overload and exhaustion, as “few, if any, real humans are capable of effectively dealing with many of the cognitive tasks demanding split attention, due to the cognitive load that these efforts have on our limited working memory.” Second, critics also argue that this concept of interactivity interferes with the suturing process. For example, Peter Vorderer et al. (2001, 346) argued that “instead of purely witnessing the events in order to develop sympathy, empathy and suspense toward the hero, [viewers] have to struggle with new technology and making decisions about the ongoing narrative.” Instead of facilitating identification, therefore, interactivity actually makes it more difficult for viewers to empathize with characters. Third, critics also argue that this concept of interactivity does not represent a genuine form of creative control, as viewers are always limited to various scripted alternatives. As Lev Manovich (2001, 61) pointed out, viewers “are asked to mistake the structure of somebody else’s mind for [their] own.” Instead of allowing viewers to generate their own content, in other words, interactive narratives merely present viewers with false choices that potentially function as a form of ideological interpellation by instilling desires that are artificial rather than authentic.

A related concern is that Netflix reportedly keeps a record of the decisions made by viewers, which could be used for a wide range of purposes without the viewers’ consent. For example, Yellin (qtd. in Streitfeld 2018, 14) notes that this information could be used to create more entertaining programs, and Jesse Damiani (2019) speculates that it could also be sold to advertisers, as it often indicates the viewers’ “product preferences” and “musical taste.” Like the Qube system, in other words, Netflix’s interactive platform offers a more effective means of gathering market research data, which suggests that interactive narratives are essentially no different from interactive menus. Some computer scientists (Mitra, Vairam, and Chandrachoodan 2019) have also expressed their concern that this information might be sensitive, as it could reflect a viewer’s interest in narcotics or propensity for violence, and it is now possible for hackers to obtain this information due to “recent advancements in the domain of encrypted network traffic analysis.” These concerns clearly echo Orwell’s fear that interactive television could be used as an instrument of surveillance, although the fear is not only that state authorities could use this information for the purpose of criminal profiling but also that private citizens could access this information for the purpose of blackmail or public humiliation.

Some of Netflix’s interactive programs have also addressed these concerns within their narratives by demonstrating the illusory nature of interactivity. In the case of Unbreakable Kimmy Schmidt: Kimmy vs. The Reverend, for example, most of the “decision points” have a correct choice, and viewers who make an incorrect choice are often told that their decisions contradict the characters’ motives or morals; the show then rewinds and continues as if the correct choice had been made. Perhaps the most obvious subversion of interaction occurs during the title sequence, as viewers who choose to “skip ahead” (a standard feature on all Netflix programs) are scolded and then forced to watch an extended version of the theme song. The show thus effectively deactivates one of the only inherently interactive elements of the platform, which serves to remind viewers that producers always have the power to limit and control their choices. Black Mirror: Bandersnatch ⁹ similarly critiques the illusory nature of interactivity by requiring viewers to make certain choices, such as when the protagonist (a young computer programmer) decides to work on his computer game adaptation of a “Choose Your Own Adventure” novel at home (which gradually drives him insane), when he decides to stop taking his medication (which exacerbates his mental illness), and when he decides to murder his father (which represents the culmination of his mental breakdown). At each of these points, the protagonist also attempts to resist the choices required by the narrative, and he even complains that he does not seem to be in control of his own actions. In one scene, for example, he receives an electronic message from the Netflix viewer that states, “I make decisions for you.” Within the narrative, this message is represented as the intrusion of the viewer into the protagonist’s world, yet this message is actually generated by the narrative itself rather than the viewer. Instead of giving the viewer a sense of control, therefore, it actually reveals the degree to which the viewer is also being controlled—a criticism that seems to apply to other interactive media as well, including novels and games. ¹⁰

Bandersnatch also contains several entirely meaningless “decision points” that have no apparent impact on the narrative, such as choosing between breakfast cereals or product purchases. These decisions represent an even more obvious example of false choices, as they are entirely arbitrary (i.e., the choice is unimportant) and absolutely required (i.e., the narrative cannot continue until a choice has been made). In other words, their sole function is to interpellate the viewer into a capitalist way of life by encouraging them to equate their own individual freedom with the freedom to choose between various consumer goods, and the only choice that would be truly meaningful in this situation—that is, the choice not to consume at all—is the only choice that is unavailable. These choices also appear to function as a form of advertising, as they influence the kinds of products that appear throughout the program. This not only replicates the kinds of targeted advertising that were made possible by the Qube system, but it also represents a new form of product placement that is well-suited to streaming services, which charge subscription rates in exchange for the promise of commercial-free viewing. While a viewer’s choices may not serve to construct a unique narrative, they may be used to construct a unique profile or “data double,” which once again invokes Orwell’s vision of interactive television as a surveillance system.

Conclusion

Marshall McLuhan (1964, 8) famously argued that the content of every medium is always that of an earlier medium, and a similar phenomenon can be seen in the history of interactive television, as each new concept of interactivity was based on the interactivity of an earlier medium. For example, the concept of television as a two-way communication medium was based on the telephone, as television was often imagined as a kind of videophone that would enable face-to-face conversations between viewers. The idea of television as a database interface was similarly based on vending and voting machines, as the use of interactive menus allowed viewers to access to a wide range of products and services while simultaneously expressing their own personal preferences. The idea of television as a medium that facilitated interactive narratives was also based on hypertexts and video games, which were similarly celebrated for offering users a greater degree of control over media content. These competing notions of interactive television were also criticized for the same reasons that earlier media were criticized, as user-user interaction facilitated new forms of surveillance (through the use of feedback channels), user-system interaction introduced new forms of commercial manipulation (through promotional mechanisms and marketing research tools), and user-content interaction failed to decenter the role of the producer (by disorienting viewers and limiting their choices to a fixed number of options that were all scripted in advance). Despite their obvious differences, therefore, all of these competing notions of interactive television failed to address the fears of early critics and even inspired new fears, as the promise of interpersonal communication was replaced by fears of social alienation, the promise of public participation was replaced by fears of social control, and the promise of individual autonomy was replaced by fears of social conformity. While each of these approaches was initially celebrated for transforming passive viewers into active participants, they were also criticized for providing a more effective method of surveillance, a more effective form of ideological interpellation, and a more effective way to keep viewers watching. This continuing struggle over the meaning and value of interactive television thus reflects the dynamics of media panics more generally, as new media technologies often promise to solve the problems of the past while simultaneously provoking new fears of the future.