Fields of Meaning: Towards a Speculative Theory of Spatial Intelligence

A Thought Experiment

Consider your movement through the world each day. You leave your home and walk to your favorite café. But when exactly did you enter the café's sphere of influence? Was it when you walked through the door? When you first caught the scent of coffee? Or was it earlier - when you first thought about going there, when its pull began to shape your movement through the city?

The truth is, we don't move through space like pins on a map, hopping from discrete point to discrete point. We flow through fields of meaning¹ that shape our movements in subtle but powerful ways. That café creates a field of influence that spreads out into the surrounding streets, blending with the bookstore next door, the park across the street, the office buildings nearby. Each location radiates meaning into the space around it, and we navigate through these overlapping fields of significance every day, usually without conscious awareness.

This isn't just metaphor. Just as physics evolved from thinking about discrete particles to understanding continuous fields and waves, we need to evolve our understanding of human spatial behavior. The dominant artificial intelligence architectures today - transformer models that power large language models - work with discrete tokens, like words in a vocabulary. But human spatial intelligence operates in the continuous realm of fields and waves of meaning.

Writing Your Spatial Story

Your spatial patterns are a form of autobiography written in the language of movement and place. Just as a writer carefully selects each word, or a painter deliberately chooses each brushstroke, your daily movements through space compose the story of your aspirations and values. The city becomes your canvas, and your movements through it create a living artwork of intention and purpose.

Think of it as choreography of purpose - every location you frequent, every path you travel, is part of an intricate dance that expresses who you are and who you're becoming. Your regular visits to a local café aren't just about coffee; they're brushstrokes in a larger portrait of someone cultivating creativity and community. Your gym routine isn't merely exercise; it's a recurring motif in the story of your commitment to personal excellence.

Just as a signature uniquely identifies you, your spatial patterns create a distinctive "spatial signature" - a unique choreography that reflects your personal algorithm for turning aspirations into reality. This signature becomes a kind of fourth-dimensional sculpture - three dimensions of space plus time - shaped by your choices and goals. Each person's pattern is unique, like a fingerprint of intention, revealing how they've chosen to write their story in the geography of their life.

Cultural Dialects of Space

These spatial stories aren't written in a universal language. Different cultures create distinct spatial dialects that shape how communities understand and move through space. Consider the contrast between Mediterranean plaza culture and Japanese concepts of ma (間) - the meaningful negative space. In Mediterranean cities, the plaza pulses with social energy, creating strong fields of public life that draw people into shared experience. The Japanese ma generates fields of contemplative emptiness, spaces charged with potential meaning.

A traditional Japanese tea house generates fields of meaning that encourage careful, deliberate movement and mindful presence. A Spanish plaza generates fields that encourage spontaneous social interaction and lingering conversation. These aren't just different uses of space - they're different ways of understanding what space means and how we participate in its significance.

From spatial autobiography to place identity

While our movements through space tell our story, this relationship between self and space runs even deeper. The fields of meaning we navigate don't just record our paths - they become interwoven with our very sense of self. As environmental psychologist Harold Proshansky recognized, our identities extend beyond our bodies to encompass the meaningful spaces we inhabit. (See here.)

The integration of place identity with fields of meaning suggests something profound about human consciousness and its relationship to space. When Proshansky first developed his theory of place identity, he recognized that our sense of self extends beyond our physical bodies to include the environments we inhabit. But viewed through the lens of spatial fields, we can now understand this relationship with unprecedented clarity: we don't just identify with places; we become entangled with their fields of meaning in ways that transform both the fields and ourselves.

Consider how a graduate student's identity becomes intertwined with their university library. Traditional AI models might see this as a simple pattern of repeated visits to a location. Even advanced language models, for all their sophistication in processing text, would struggle to capture the deep resonance between person and place that develops over time. The library's field of meaning - its atmosphere of scholarly focus, its quiet corners conducive to deep thought, its subtle daily rhythms - gradually becomes woven into the student's way of being. They don't just study at the library; they begin to think differently because of their entanglement with its field of meaning.

This phenomenon exposes both the limitations of current AI architectures and the exciting possibilities ahead. Language models work with discrete tokens and can masterfully process text about places and experiences. But they cannot yet grasp how meaning exists as continuous fields that shape human identity and behavior. A spatial transformer architecture, by contrast, could model how fields of meaning and human identity become entangled over time. It could understand not just where people go, but how their movements through space both express and shape who they are.

This isn't just a theoretical distinction - it's crucial for the future of AI systems that aim to understand and support human behavior. When critics point out that language models can't truly understand human experience, they're right but missing the larger point. The goal isn't merely to defend current architectures but to evolve them. By combining the linguistic sophistication of transformer models with new architectures that can process continuous fields of meaning, we move closer to AI systems that can comprehend the full richness of human spatial experience.

The convergence of place identity theory with spatial transformers suggests new ways of thinking about urban design and personal development. If our identities are shaped by the fields of meaning we inhabit, then designing cities isn't just about arranging physical structures - it's about crafting fields of meaning that support human flourishing. Similarly, personal growth might be understood not just as internal psychological development but as a process of intentionally engaging with and being transformed by meaningful spatial fields.

This perspective reveals why changes to familiar places can feel so personally disrupting. When a beloved café closes or a neighborhood transforms, we're not just losing a location - we're experiencing a disruption in the fields of meaning that have become part of our extended self. Our spatial transformer architecture would need to account for this deep entanglement between personal identity and spatial fields, understanding how changes in one ripple through and affect the other.

The Mesh of Meaning

To understand how fields of meaning work in space, think about how the influence of a café spreads through its neighborhood. At its strongest point - right at the café itself - the field of meaning is most intense. As you move away, this influence gradually diminishes, like ripples spreading out from a stone dropped in water. But unlike simple ripples, these fields of meaning have complex patterns of interaction.

When we say a field of meaning spreads continuously, we're describing something quite precise: the way the café's influence smoothly transitions from strong to weak as you move through space, never jumping suddenly from one level to another. This continuous nature is what sets it apart from traditional ways of thinking about locations as distinct points.

But these fields aren't static - they pulse with time and context. A café's field of meaning might be strongest during morning hours, weaken during the afternoon lull, and transform entirely in the evening. It might strengthen when friends are nearby, weaken during bad weather, or change character depending on who else is present.

Most importantly, these fields interact with each other. When a café is near a bookstore, their fields of meaning don't just overlap - they interact in ways that create new meanings. This is why certain combinations of places create particularly rich environments for specific activities or experiences. A café near a university generates a different field of meaning than the same café near a shopping mall.

This dynamic interaction of fields helps explain why certain locations emerge as natural gathering points or why some neighborhoods develop distinct characters over time. It's not just about individual places, but about how their fields of meaning combine and influence each other.

Learning to Read Spatial Stories

How might artificial intelligence learn to read these spatial stories? Current AI systems might recognize that someone visits a café daily, but they miss the rich layers of meaning embedded in this pattern. A spatial transformer architecture would need to understand both universal patterns and cultural variations in how meaning flows through space.

The system would learn by observing how people move through and linger in spaces, how patterns change with time and social context, and how different types of spaces influence each other. It would discover that Spanish plazas show strong activity fields late into the night, while Japanese shopping districts pulse with different energies during seasonal festivals.

Most importantly, the system would learn that these patterns aren't rigid - they flex and adapt, blend and evolve. New cultural forms create new spatial languages, and individual variations create personal dialects within cultural patterns.

Beyond Current AI Architectures

To understand why we need a new approach to artificial intelligence for spatial understanding, let's examine how current AI systems process information. The transformer architecture that powers today's most advanced AI systems works much like a careful reader parsing a sentence. When you read 'The café is busy,' you understand each word in relation to the others. Similarly, transformer models break down information into discrete pieces (called tokens) and figure out how these pieces relate to each other.

Think of it like this: each word in a sentence gets converted into a kind of coordinate in a vast space of meaning. In this space, similar words cluster together - 'café' might be near 'restaurant' and 'bistro.' When processing information, the transformer determines how much attention each word should pay to every other word, much like how we might emphasize certain words in a sentence to understand its meaning.

But here's where we encounter a limitation: this approach works wonderfully for language, where meaning is naturally broken into discrete units like words and sentences. But human spatial experience isn't discrete - it flows continuously. When you approach a café, its influence on your behavior doesn't suddenly switch on like a light. Instead, it gradually intensifies as you get closer, interacting with other nearby influences in subtle ways.

This is why we need a new kind of transformer architecture - one that can handle continuous fields of meaning rather than discrete tokens. Instead of jumping from point to point like stepping stones across a river, this architecture would flow like the river itself, understanding how meaning spreads and blends across space.

Consider how this might work in practice. Rather than labeling a location simply as 'café' and calculating its relationships with other fixed points, a spatial transformer would understand how the café generates a field of influence that varies smoothly with distance, time, and context. It would recognize how this field strengthens during morning hours, weakens in the afternoon, and transforms in the evening. It would understand how the field interacts with nearby influences - how a café near a university creates a different kind of space than the same café near a shopping mall.

This shift from discrete to continuous understanding mirrors a similar revolution in physics, when scientists moved from thinking about distinct particles to understanding fields and waves. Just as quantum mechanics needed new mathematical tools to describe how particles exist as probability waves, our spatial transformer needs new tools to capture how meaning exists as continuous fields in space.

Implications for Urban Life

This framework has profound implications for how we design and understand cities. If we can model how fields of meaning flow and interact, we can create spaces that better support human flourishing. We can understand how changes to urban space will affect not just movement patterns but the very way people experience and create meaning in their environments.

Moreover, this understanding might help us create cities that support multiple cultural spatial languages while allowing new forms to emerge. Just as multilingual societies enrich human experience, cities that support multiple spatial dialects might create richer, more vibrant urban life.

Conclusion: Participating in Fields of Meaning

In the end, this framework suggests a profound truth about human spatial behavior: we are not just observers or creators of spatial meaning - we are participants in fields of meaning that have their own reality and patterns. Our movements through space are not just choices but participations in these fields, creating a dance between human intention and spatial significance.

As we move forward in developing artificial intelligence and designing urban spaces, this understanding of spatial meaning could help create systems and environments that better support human spatial intelligence. Just as quantum mechanics revolutionized our understanding of the physical world, a quantum theory of spatial meaning might revolutionize our understanding of human behavior and urban life.

The city becomes not just a collection of locations but a living text, written in the language of movement and meaning, waiting to be read and understood in all its complexity and beauty.​​​​​​​​​​​​​​​​

1. ‘Fields of Meaning’ is a phrase that’s too good not to use in this context. Markus Gabriel, a German philosopher, has written extensively on ‘fields of sense’ ontology. I find much to applaud in what he’s developed. Necessary here to highlight the similarity in name even if the use isn’t equivalent.