In my last post about route knowledge and survey knowledge, I wrote about my rideshire driver, disoriented in Vegas without her GPS, and about London taxi drivers whose famous navigation skills are enabled by extraordinary embodied memory and hierarchical route planning. But I suspect the tension between embodied, situated knowledge and more abstract high-level knowledge goes beyond navigation.
Narrative as Landscape
In the late 90's the multimedia designer Bob Hughes wrote about narrative as landscape; that stories should be understood not as linear directed paths, but as three-dimensional terrains through which we may travel in multiple ways. Given that so much of our communication involves sharing information with narratives (data storytelling is a familiar example to many readers) I am starting to think of our knowledge, what we believe we know, as a landscape that becomes familiar to us through experiences: sometimes by embodied and situated routes and sometimes by abstract, systemic surveys.
And when I think of knowledge in this way, I see many reasons why such different perspectives lead to significant misunderstandings between people on issues in science and society. I wonder if, by recognizing the value and limitations of these forms of knowledge, we may approach learning, expertise and collaboration very differently.
Rooted en route
It's common enough to talk of our journey through life. What we learn on that journey is, in effect, route knowledge of the world: a first-person understanding of how we arrived at our current position and what options appear to be available from where we stand. This form of understanding is often unspoken and intuitive, embedded in habits and emotional associations that may resist clear telling. We know because we know.
This route knowledge of the world carries the multisensory emotional memories and context that we don’t find in abstract models; it evolves with every decision, creating dynamic mental models that change with new experiences. The embodied nature of route knowledge makes it particularly compelling, creating a sense of certainty and immediacy that more theoretical understanding lacks.
A parent's intuition, a craftsperson's skill, or a community's shared understanding of its local geography and social dynamics are all forms of embodied knowledge.
Nevertheless, howsoever powerful and necessary this embodied route knowledge is, it remains inherently partial, constrained by the paths we have traveled and the vantage points we have occupied. Our mental maps are shaped by the unique terrain we traverse, including privilege, trauma and culture.
The view from above
Survey knowledge represents a more detached, systematic attempt to comprehend the world through models, theories, and frameworks. This form of understanding aims for comprehensiveness and generalizability; it may claim to transcend the limitations of any single perspective or experience.
Expert knowledge typically privileges the perspective of survey knowledge; the expert looks for patterns and principles that can be applied across many contexts and situations.
There's a power to this, too: the ability to reveal connections, structures, and possibilities that remain invisible from any single position we might reach by route knowledge. So, overarching frameworks, like scientific theories or economic models, offer the potential to see beyond the immediate horizon to connect seemingly distant experiences.
Nevertheless, despite its aspirations to objectivity, survey knowledge is also constructed: from specific intellectual and cultural positions that shape what is seen and valued. Even the most sophisticated models inevitably simplify and abstract away the messy complexity of our real experience, creating blind spots and limitations of their own.
Recursive knowledge
People frequently mistake their route knowledge for survey knowledge, assuming that their personal experience provides a comprehensive understanding of the full landscape. But to be fair, supposed experts also overreach when they treat their abstract models or surveys of a knowledge space as complete representations, failing to recognize that their frameworks also exclude or distort.
I myself (and many of my friends and colleagues) have a preference for survey knowledge, for systemic intelligence. I might dismiss others' experiential knowledge as uninformed or anecdotal in favor of scientific studies and expert analyses. But when I do so, I am ironically operating from a form of route knowledge that has shaped my own trust in these scientific and expert institutions. I have personal experiences and cultural narratives that have taught me to favor certain authorities over others. In other words, even the most abstract or supposedly objective forms of knowing remain grounded in situated, embodied histories of learning what to trust and why.
My friend Scott Davis (who writes the excellent Surf the Seesaw blog) said once …
People don’t trust data, they only trust other people.
Recognizing this is not to claim that all knowledge is merely subjective, but rather to recognize that all knowledge (even scientific knowledge) is always situated in particular contexts, histories, and communities of practice. The scientist who champions scientific, empirical method over intuition does so because of a personal history of experiences that taught them to value certain ways of knowing: in itself a form of route knowledge.
When maps diverge
It's a cliché that the map is not the territory. When expert knowledge and lived experience diverge this often stems from the disconnect between abstract and embodied knowledge. Clashes arise (often literally) when survey knowledge dismisses route knowledge as anecdotal or irrational. We see this right now, in the anti-vaccination movement, climate controversies, debates over immigration and crime and too many other examples.
Personally, I'm strongly in favour of vaccination. But it's difficult to say if that is rooted more in my scientific understanding (which is limited: I'm no virologist) or in my family's experience. Three cousins of my mother died as children, in one night, in the same room, from measles: an awful family memory has been passed on. Others have different stories that no doubt move them too.
Effective persuasion and communication require recognizing that we must meet others where their knowledge is situated, not where we wish it would be.
What can we do about this divergence?
Non-linear landscapes
Bob Hughes observed that if narrative really is landscape, then there is no radical difference between the kinds of narrative you can deal with in linear and non-linear media: any well-enough realized landscape should be traversable by a variety of routes.
In a simple book or movie (linear media), you experience the world in just one way: following the main character on a specific journey. But in non-linear media, like video games, you can choose different paths.
Hughes is saying that if a story world is detailed and rich enough (if, for example, the author has really developed how everything works and connects), then it doesn't matter whether you experience it linearly like a book or non-linearly like a game. The world itself exists either way, and you should be able to take different journeys through it (even if only in your imagination) that all make sense.
A richly-textured novel, such as The Master and Margarita or Weir of Hermiston, will contain many possible narratives within its scope.
I think there's a hint here about how we might reconcile different forms of knowledge. The richness of a knowledge domain is measured not by its adherence to a single path or perspective but by the multiplicity of meaningful journeys it can support. A truly rich subject matter isn't just about memorizing one correct sequence of facts or one right way to think about it. Instead, what makes a subject interesting and valuable is that different people can approach it from different angles and still learn something worthwhile.
When a landscape of understanding is sufficiently detailed and coherent, it becomes navigable from multiple entry points and along many paths. So the real test of whether someone deeply understands something isn't if they can recite one explanation (as I can about vaccines) but whether they can approach it from multiple angles, explain it different ways and still capture what's important about it (which I can't do with the vaccine story.)
Complementary cartographies
Hughes has yet another way of thinking about this problem, that is also very useful. He illustrates the versatility of narrative landscapes by distinguishing between story maps and grid maps which are very much related to route and survey knowledge.
Story maps are experiential, relational, and emotionally resonant. They are built from route knowledge and deeply embedded in cultural contexts and personal significance. They excel at conveying how it feels to traverse a landscape and what meanings emerge along the journey. Indigenous knowledge systems typically privilege story maps, encoding ecological wisdom, cultural values, and practical know-how. They are representations of route knowledge.
Grid maps, by contrast, organize territory according to abstract, standardized coordinates. They prioritize consistency, universality and measurement. Grid maps allow for precise location and navigation regardless of one's subjective experience of the terrain. They facilitate communication across cultural contexts precisely because they attempt to strip away the subjective and the particular. They record and layout survey knowledge for our future learning.
The distinction is not one of superiority but of different strengths and limitations. A story map might better convey the emotional significance of a sacred site, while a grid map might better enable someone to locate it from a distance. Both are valid ways of knowing the same territory.
If we accept that knowledge exists as a landscape rather than a fixed path, then communicating across different perspectives requires a particular kind of cognitive flexibility: the ability to traverse the same territory via multiple routes and to translate between different mapping systems.
Gadamer's fusion of horizons
This process of translating across perspectives reminds me of Hans-Georg Gadamer's concept of the fusion of horizons: Horizontverschmelzung. For Gadamer, understanding is never a matter of one perspective simply replacing another. Rather, genuine understanding occurs when different horizons of interpretation (each shaped by history, culture, and experience) engage with each other and create a new, expanded horizon.
In our landscape metaphor, the fusion of horizons might be visualized as two travelers who have explored the same territory via different routes coming together to create a more comprehensive map. Each brings unique vantage points and observations; together, they construct an understanding richer than either could achieve alone. Neither abandons their journey or perspective, but each recognizes it as one path through a more expansive landscape.
This shift from conflict to co-creation requires particular practices:
Humility: Recognizing that one's own map, no matter how detailed, captures only certain aspects of the territory.
Reciprocity: Developing the capacity to retell one's understanding in terms that honor another's mapping conventions and starting points.
Diversity: Valuing different knowledge (mapping) traditions not merely as cultural curiosities but as legitimate ways of knowing that reveal different aspects of reality.
Boundary objects: Creating shared reference points that can be recognized and valued across different knowledge systems, serving as translation points between story maps and grid maps.
When these practices are cultivated, seemingly incompatible knowledge systems do not need to abandon their distinctions but can recognize themselves as different journeys through a shared reality. In fact, their differences can become resources to help co-create a new understanding rather than obstacles.
How can this help us with the debate about vaccination?
A co-creation approach wouldn't start by trying to “correct" vaccine-hesitant people or dismiss their concerns. Instead, it would recognize that both groups are travelling through the same landscape of health and safety, but using very different maps.
Scientists typically navigate via grid maps: systematic studies, statistical analyses and clinical trials that plot knowledge according to standardized coordinates. They see vaccines through data points showing efficacy rates and population-level risk reduction.
Vaccine-hesitant individuals often navigate via story maps: personal or community experiences, narratives about bodily autonomy, and accounts of medical interactions that have shaped their trust or distrust in institutions. Their concerns are often rooted in real experiences of feeling unheard, disrespected, or harmed by medical systems.
The majority of people making decisions about vaccination are women. Their experience of the medical profession is often a story of arrogance, intrusion, neglect and incompetence: countless women have these experiences. Books like The Lady's Handbook for Her Mysterious Illness by Sarah Ramey or Elinor Cleghorn's Unwell Women: Misdiagnosis and Myth in a Man-Made World are shocking both in their account of suffering and the familiarity of their story.
For a woman who has been through this and has painfully-earned route knowledge of the failures of medicine, why would she trust the abstract voice of medicine coming from on high? All her real-world experience tells her to be doubtful of their claims.
An approach that values the co-creation of knowledge might try harder to identify common ground; both groups care deeply about health, safety, and protecting vulnerable people. Scientists and health officials would acknowledge that personal experiences are real and matter rather than being merely anecdotal. If someone had an unsatisfactory encounter with healthcare, that is real and has shaped their route knowledge about all healthcare.
In other words, we might accept that vaccine hesitancy is a legitimate response to experiences of modern healthcare even if not scientifically grounded or objectively best for children.
That feels like a shocking thing to write; I have edited it many times now. But what a different conversation we could have from that perspective. The dialogue wouldn’t be structured as a one-way “education” campaign where experts correct the "misinformed." Instead, it becomes a process of mutual learning where both forms of knowledge are valued.
Health systems might evolve practices that better address individual care. Information resources might use both statistical evidence and first-person narratives of the many complex experiences we have with vaccination.
This approach doesn't guarantee agreement, but it creates a space where different perspectives can engage productively rather than talk past each other. The goal isn't to replace one mapping system with another but to create a new, shared landscape of understanding where both have value and can inform each other. There's a good example of this in Minnesota, where anti-vaccination claims led to a dramatic fall in childhood protection, which is now slowly being reversed with a very similar approach of shared horizons.
Donald, as always your thoughts raise memories and questions.
First, a memory—when working on the Extreme Ultraviolet Explorer, there were two engineers who were deep wells of experience and practical knowledge, and there was one chief mechanical engineer, whose route knowledge, as you say, was from ground based medical equipment, not space flight hardware. All three of them, and the many students, astronomers and engineers, shared survey knowledge, the same or similar coursework, bodies of knowledge, compendium and frameworks. The discussions around material selection and tolerances were… Interesting. These memories from almost 40 years ago are helping me to, or perhaps filtering, my understanding of your thoughts.
For me, what you are writing is not so much about knowledge, but about questioning and learning, about individuals forming communities to make or block decisions. About how we come to have our inherent recognition of cause and effect, regardless of correlation and the arrow of time, and how we form conceptualizations of our place and what surrounds us. It also tells me why I continue to feel that machines don’t learn and our current approaches, even using causal inference, and even if useful, will never lead to artificial intelligence at any level.
Your words are well written; thank you for sharing them.
Ok, this is not a blog post. It is a masterclass worthy of a series of posts or its own book. You've connected so many ideas into a web of useful revelation, I will have to spend a bit of time evey day for a month to contemplate small pieces of this stunningly expansive essay. The only appropriate thing I can add is ... thank you for the gift of pointing out these things and makiing them easier for us to see.