Eliding the real
Plastic is everywhere and it would be easy for most of us to imagine that this has always been the case. While the first plastics were developed in the second half of the 19th century, the explosion of new materials and uses really began in the 1940s and 1950s. This boom saw polymers such as polystyrene, polyvinyl chloride, and polyethyle replace existing components and allow for the creation of new ones.
9.2 billion tonnes of plastic are thought to have been produced between 1950 and 2017. More than half of this total has been since 2004. In 2020 alone, 400 million tonnes of plastic were produced.
Plastics made possible lifestyles for many that had previously been available to only the wealthy. When materials, craft and technique matter, costs rise. Plastic, however, was relatively easy to produce, malleable, and ideal for industrial modes of production.
Recent years have offered a better understanding of the harms these materials bring to human and non-human life. Micro-plastics are showing up more and more in the food chain. The long-term effects of plastics on the body are poorly understood, though it is clear that the presence of plastic in the environment hurt ecosystems and human health.
Synthetic Language
We are on the precipice of a new explosion, this time of language rather than of a material substrate. ChatGPT makes it clear that the time of synthetic language has arrived. Just like with plastic in the 1940s and 1950s, we should anticipate a profound reworking of our cultural infrastructure. Within years, we can expect that the amount of artificial speech created will exceed natural speech from living human beings.
At UKAI Projects, one of our research priorities is how this will impact our culture and the arts upon which we rely to make sense of the world.
Our social and cultural world is communicatively constituted. We talk things into being. This involves more than superficial distinctions. Boundaries, meaning, and the mechanics of everyday life are symbolically mediated, and this mediation often rests on language. We are constantly navigating a tension between the adaptations we made in the past and the emerging needs of the present. My worry is that when our responses are increasingly drawn from the same past, we become incapable of improvising an appropriate language that is directed at the present or the future.
To cope with the enormous complexity of being alive, we develop repeatable patterns of action, behaviour, and thought. We reduce the complexity of incidents to a tolerable, manageable level. These patterns take the form of habits, rituals, modeled behaviours, expectations, prejudices, meaning constructs, and worldviews which are shared through language. The process of reduction makes events seem predictable and adapts us to the perceived needs of the situation at that point in time.
The inner life of human collectives (organizations, communities, etc.) are similarly defined by reductions of complexity. Control across members is maintained through shared symbols, hierarchies of value and vision, customs, rituals, role assignments, structural hierarchies, and above all through the objectification of agreements.
Objectification of consensus and difference takes place both through the construction of artificial structures (infrastructure, legal codes, walls, etc.) as well as through internalization (habits, perception and expectation patterns, behaviour, prejudices, etc.). Objectifications allow for permanent distinctions to be made; what is important or unimportant, safe or unsafe, useful or useless, permissible or prohibited, desirable or undesirable, true or false and so on. Making such distinctions means making selections and, from a system-theoretical point of view, can then be understood as information.
Put another way, as the system selects from all the noise an element and assigns it a specific meaning, the specific element becomes defined and formed – it has importance for that system and thereby becomes information for ongoing self-referential operations.
We must always be aware of the tension between the ongoing development of a system and its desire to harden the internal (reified and internalized) structures. Reductions of complexity allow us to move through the world and not be overwhelmed. However, reductions of complexity can also hinder or prevent our survival, adaptability, and learning. Objectifications not only limit the available behaviours that we can deploy to adapt to our environment, but also the information that we can make use of in decision making and other activities.
I tell my daughter the classic version of Little Red Riding Hood and she comes to understand the forest as a scary place and the advice of family as essential. Updates on the story shift the threat to strange men in the shape of wolves. Modern reinterpretations put power in the hands of men with axes or with the heroine herself. The forest doesn’t change, but our understanding of it does as our broader culture evolves. Evolution, however, entails a living and emergent response to the world around us.
What happens when the language we use only looks backward?
Questions to Consider
Large language models (LLMs) are a type of artificial intelligence (AI) technology that use advanced machine learning algorithms to process and generate language that appears natural.
The explosive growth in LLMs suggests a future where this synthetic language can be accessed as easily as running water or bananas from the tropics.
Of course, vast socio-technical structures underpin these systems. Water from a tap or bananas on a shelf only tell part of the story.
Love letters, novels, poems, epics, satire, and other forms draw on a shared reservoir of culture and language but are organized in different ways based on the fundamental structures appropriate to their genre. Creators of prose art draw on the language(s) around them and give them a shape and a unity of intention.
What happens to the poetics of a work when language is being drawn not from engagement with the world but rather from a large language model trained on a vast corpus of human language? What are the risks and opportunities of relying on synthetic language that is not participating in the continuous process of becoming and evolving that defines language running wild in the world?
Much of the debate around LLMs has centred on the processes by which these models are trained and the relative quality of the outputs that they support. However, we must consider a future where the volume of synthetic language produced by machines will quickly outstrip that produced by human beings.
What happens when water is only available from a tap or in a bottle? What happens when the fruits available to us are limited to only those easily delivered to a supermarket?
Verbal art must now contend with the introduction of an intermediary, akin to a chisel for a sculptor or a cello for a musician. What does this intermediary allow? What does it restrict?
The arrival of large language models presents both challenges and opportunities for those exploring verbal art and my hope is that over the coming weeks and months we can begin answering some of these questions and invite conversation and generate outputs to explore this rapidly changing world.
Mostly, we are curious about the degree to which an automated system allows for other values to be expressed, to detach from constraining contexts, or to open up connotations.
Do large language models arrest the perpetual becoming of language or introduce a new space for genres and forms to emerge?
The answer to these question will extend from the choices we make. How much do we rely on patterns of objectification of the past to make sense of the present? How much efficiency are we willing to give up in order to be responsive to the world in its perpetual becoming?