【by You-Sheng Zhang, Dec. 2023】
Many people have had similar experiences: if you stare at a word for a long time, especially if you focus on its strokes or letters closely, it’s easy to lose your primary understanding of it, and you may even be temporarily unable to recognize it. This phenomenon shows that words (or language more broadly) must be related to others within a certain context for meaning to arise.
The term virtual is the same; if discussed in isolation, one might get lost on the path to understanding it. A good method is to put it back into its every day (English) context. In sentences like “You virtually succeeded” or “You’re virtually a savior!” the word virtually conveys a sense of “almost” or “practically”: being a little short of the goal. But this is just the literal meaning. We all know that the intention of the people uttering these sentences is not to make a negative statement, such as “You did not succeed” or “You are not yet a savior”; on the contrary, they see the subject as “akin to” or “nearly” successful or a savior, possessing the potential to succeed or to become a savior.
The contradiction and dynamics of “virtual” are specifically and minutely demonstrated in the examples above. Ontologically, even though the virtual and the real are differentiated, they are not disconnected but have a bridging distance, a progressively overlapping potential; epistemologically, even if the virtual is not the real, it bears a likeness to the real in form or effect and a degree of truth. “Virtual” is a word with such a duality. In any case, the virtual is not a logical error, an experiential falsehood, or a subjective sensory illusion, but a series of actions presenting reality, since these actions are always already within the object it presents, the resulting presentation and reality will interpenetrate and mutually affect one another.
In this regard, “virtual” can be considered within the interactive relationship between symbolic systems and the real world. Throughout human history, language, writing, painting, drama, photography, and film have continuously touched upon this problematic. However, we rarely use “virtual” to describe the fields of art and literature (instead, “fiction” is more common), probably for two reasons.
First, in terms of media and hardware, the virtual in the current context is mainly built on digital signals and internet technology, whose elements might not be derived from the real world but are digitally native, whereas traditional arts and literature are clearly not so, as digital internet did not exist before the twentieth century.
Second, when these artistic and literary fields present reality—the given spatial-temporal realm our bodies, senses and thoughts live in together—it is inevitable that they exist in the form of works external to the human agent, with so-called cover, length, frame, screen, and venue constraints, all of which repeatedly label them as symbolic systems, ultimately differentiating them from the real world.
The virtual we speak of now may include various artistic and literary elements, but in the first kind of its development, it integrates them—for example, a 3D Massive Multiplayer Online Game (MMOG) with a good scenario, dialogue and motion graphics—and, by virtue of robust technology infrastructure, (tries hard to) removes such label, breaking through the aforementioned constraints like frame and screen to be “akin to” reality which makes the agent internal to it. In the second kind, the virtual changes how we enter into and move between symbolic systems by adding a new way of communication: not speaking, commanding, searching, but asking actors “prompt.” After inputting some words, texts are transformed into pictures. The whole process is more like what we heard from stories about pre-modern times: putting a spell. In the third kind, the virtual directly diverts its attention from symbolic systems to social behavior in the real world. The actors hidden behind “wallets” can roam freely in the vast ocean of the internet. This ever-expanding internet ocean known as Web3 allows actors to bypass two major centralized authorities—large capitalist platform companies and modern nation-states—preserving their own privacy and assets.
When the virtual world presents reality, which part of it is its object? Or what does the virtual “simulate”? The three developments mentioned correspond to three kinds of technology that complement each other and are worthy of our in-depth exploration.
Firstly, there is virtual reality (VR), which might be one of the origins of the term virtual first heard by the public, especially in the Chinese context. Technologically, its primary method of representing reality is to break through the previously mentioned confines of a frame or screen—by wearing a 360-degree panoramic head-mounted display, users can get a sense of a specially designed 3D environment, with scenes that change perspective as our bodies move. Immersed in this environment, wearing a headset equates to “entering” the entire virtual world. Although it is not without its problems, Taiwan film director Chen Singing’s VR film “The Man Who Can’t Leave” (2022) shows these characteristics perfectly. However, because a head-mounted display is necessary to experience the virtual world, it is isolated from the real world without interaction between the two. To address this issue, mixed reality (MR, a combination of virtual reality and augmented reality [AR]), has become the new design for interaction between virtual and real. This year Apple Inc. launched Vision Pro, which allows users to switch between the real world and the virtual world at will: typically functioning as a head-mounted display similar to glasses, it adds command patterns or icons in front of our eyes through AR. If activated, it makes users immersed in the visuals presented by VR. All of this is irrespective of the content displayed: the virtual reality created can replicate the reality we are familiar with or have experienced, or it can be all sorts of fiction; in fact, the stronger the technological capability to capture reality, the more it can transcend reality, showing various imaginations as if they were real. In short, the reality that VR tries to present is space, especially the simulation of various physical-visual relations between the actor’s movement or position at a specific location and the environment and objects within it.
Secondly, there is artificial intelligence (AI), which “virtualizes” not space, but information or various messages. Over the past year, the world has witnessed significant advancements in image simulation and large language models (LLMs); unlike the developments of the past decade, this stage of progress (or should I say process) has allowed the general public to operate in a relatively intuitive manner—inputting natural language instead of programming code. In terms of image simulation, as mentioned earlier, the virtual images created can be of real people or scenes, or they can be imaginations that have never existed or occurred.[1] In commercial applications, this is sufficient to avoid the costs and disputes of portrait rights but also suitable for creating fake news and conducting cognitive warfare in politics. The virtual world is directly disturbing the real world. In terms of LLMs, they have been of great assistance to general communication and administrative work; from language translation to drafting letters, users can utilize the results with slight modifications after asking the right questions. Interestingly, under proper guidance, LLMs can also engage in fictional creation with literary significance; of course, whether such fiction possesses “creativity” or is merely “imitation” due to statistical regularity of data or the logic of “(de)compression” – Ted Chiang presented a very interesting perspective on this in an article called “ChatGPT Is a Blurry JPEG of the Web” for The New Yorker – remains to be observed. Hong Kong writer Dung Kai-cheung once explained at Nowhere Bookstore in Taipei how he had ChatGPT write something “similar” to Kafka’s famous novel Metamorphosis. But is that creativity or creative?
No matter whether VR or AI, the reality they attempt to present belongs to appearances or phenomena of experience, hence emphasizing the user’s experience. At this level, the reality virtualized, although never seen before, is not invisible. Is the so-called reality limited only to the visible level of appearance? Now, let us take a look at another type of virtual technology.
Third, is the decentralized network (DN). This concept can be traced back to the design of P2P (peer-to-peer) networks, which initially had little to do with the user’s experience but instead emphasized the significance of behaviors. Over the past decade or so, with blockchain as a main tool, it has systematically virtualized identities, currencies, and properties, even extending to the groups formed by multiple actors; in short, it presents the essential level or underlying mechanisms of reality: social relations. These virtual identities, currencies, and properties are not merely game treasures developed by some enterprise (although they can be combined with these) but are determined by a multitude of computer nodes across the globe based on mutual agreements and specific cryptography, characterized by a high degree of cooperation. In this virtual—or “hashed”[2]—world, identities are no longer given by states, currencies are no longer issued by central banks, the amount and value of properties are public and transparent, and money flows can be easily traced; as long as the community reaches a consensus, encrypted identities can be established, encrypted currencies issued, and encrypted assets owned, hence this world has many subsystems. More importantly, the private keys that permit the operation of assets are stored entirely on the encrypted client side, not retained on a single company or its single server.[3] Those data or assets in the form of tokens could be put into decentralized storages like InterPlanetary File System (IPFS) which makes them not easy to lead to Single Point of Failure (SPOF).
For VR and AI, simulating a world that has never been seen is just an option. In the view of the DN, however, building a world structurally distinct from the current state seems inevitable. This is because the realms of exchange and circulation in the current stage of capitalism have been monopolized by centralized platforms. The encrypted meaning of the virtual world, through the issuance and collection of tokens on the blockchain, allows producers and consumers or creators of content and their supporters to establish direct connections, bypassing the intermediation and exploitation of platforms. They can even form decentralized autonomous organizations (DAOs) with such tokens, possessing a group identity. For example, through the use of multi-signature wallets, multiple participants can collectively create a group-level account, thereby forming an organization that is distinct from corporations and governments, with decision-making processes that are as bottom-up as possible. Because it is somewhat at odds with the existing ownership system, and due to its openness that gives opportunities to speculators, this leads to distorted media reporting, and thus the impact of DN is not as immediate as that of AI.
These three technologies are shaping a given spatial-temporal domain: Nodes on the DN encode all transmitted information around the clock, putting data on the blockchain; AI databases never cease to absorb materials, stamp them, and record all inquiries and commands; VR continues to endow characters and scenes with increasingly lifelike forms, dynamics, and spatial sensations, as long as we wear specific equipment, we can enter in a new world in the process. The three kinds of virtual technologies are drastically changing the real world, from the images perceived through the senses to the language used in communication to the underlying social relations that support these actions.
The so-called virtual is not just a passive presentation of reality because the best way to present reality is to participate in it, approach it, and change it in the process. Therefore, let’s temporarily stop discussing what is virtual. Shift your gaze away from this word and even its context, then step sideways into any of the three technologies above. Perhaps you will get totally different ideas about the virtual.
AUTHOR
You-Sheng Zhang, Performing Arts Critic and Co-Founder of Volume DAO, Taiwan
NOTES
[1] From 2020 to the present, AI has progressed quite rapidly. In 2021, many digital artists were creating with Generative Adversarial Networks (GAN)—the art NFT (Non-Fungible Token) marketplace objket.com has witnessed this phenomenon—and by the end of 2022, the Diffusion Model had become prominent, with Stable Diffusion and Midjourney being among the main ones.
[2] This mathematical algorithm takes an input (or message) and returns a fixed-size string of bytes. The output is typically a digest unique to each input. Hash functions are designed to be a one-way function, which is computationally infeasible to reverse. In other words, it should be extremely difficult to recreate the original input given only the hash output.
[3] I have not distinguished between “encrypted” and “crypto.” In the context of Chinese, both can be translated the same way. However, the former is closer to the technical implementation of cryptography, while the latter emphasizes the decentralized nature of blockchain. But among blockchain practitioners, sometimes crypto is also equated with cryptocurrency. Furthermore, I have deliberately replaced “wallet” with “identity.” Decentralized Identity (DID) is one of the important issues now. Although wallets cannot currently correspond one-to-one with real owners or individual humans, their goal indeed is to become the identity of a real person.