Sketch maps have that same structure, pieces that connect places. We asked hungry students to sketch maps or write down directions from where they were to a nearby fast-food restaurant. Although sketch maps could represent distances and directions and turns and wiggles in paths in an analog fashion, they don’t. They are discrete in the same ways that directions in words are discrete. Sketch maps elongate distances where there are many turns, just as verbal directions are longer for many turns. Sketch maps straighten wiggles in roads and draw turns as 90 degrees regardless of their real angle, just as verbal directions say “go down” without specifying wiggles or say “turn right” or take a left” without specifying the angle of the turn. In short, the same schematic mental representation of a route seems to drive both the verbal route directions and the sketch maps.
Your perspective or mine? The you that I’ve dragged to find my bike or to explore midtown Manhattan is a hypothetical you, any you, even me. But what about face-to-face situations when I choose between your perspective or mine? Say there are two wine glasses on the table and you ask, “Which is mine?” I’m more likely to answer from your perspective than from mine, even if it means switching right and left, which is, as you know, a difficult switch. Taking your perspective rather than mine might be in part a matter of politeness, though Japanese people, members of the most stereotypically polite culture, take your perspective no more than Americans, about 70 percent of the time, depending on the circumstances. And that’s just it—whose perspective seems to depend on the circumstances, in particular, on the relative cognitive loads, yours and mine. In the case of the wine glass, I know whose is whose, but you don’t, so your cognitive load is greater than mine. You have to both understand the utterance and map it to the glasses; I know the mapping and only have to create the utterance. On the other hand, if my cognitive load is greater than yours, say, you know the answer but I don’t and I’m asking you, I’ll favor my perspective, “Is mine on my right?” If we are offered a neutral way to specify the location, neither your perspective nor mine, we’ll both prefer that one. If the designated glass is closer to the salt shaker, we’ll use the salt shaker as a landmark and say, “Your glass is the one closer to the salt shaker.”
Fans of route descriptions claim that they are common and compelling because they are natural: routes are the way we experience the world, by moving through it.
Then, in order to describe where things are in an environment, we reexperience it, we imagine ourselves moving through that environment and describe where everything is from our changing point of view. Persuasive as that claim might be, and it did persuade many, people have made maps for eons, and mapmaking requires an allocentric perspective, one from above. Sure enough, when people describe environments and even routes, they often adopt an allocentric perspective in part or in whole.
Allocentric perspective. Mapmaking has to be one of the truly remarkable achievements of the human mind. No one has found a chimpanzee who makes maps—yet. In contrast to exploration, maps don’t use an egocentric perspective, they use an allocentric perspective, an egoless “other-centric” perspective, one outside the body and typically one from above. The most general frame of reference that ordinary mortals use is the one of the world, north-south east-west. Geologists and meteorologists and some other -ologists add a third dimension, altitude. Sailors use the coordinates of the vessel, port-starboard (left-right) and bow-stern (front-back). Actors and directors use those of the stage from the point of view of an actor, facing an audience. Hence, upstage. Neuroscientists and physicians, those of the body, naturally obscure terms derived from Latin, dorsal/ventral, caudal/rostral, medial/lateral. In case you’re curious, dorsum means “back” and venter means “stomach.” No wonder they use Latin.
Before we can draw maps, we need to conceive of them. We need to imagine the world from a perspective above the world, a bird’s-eye or survey or overview perspective. We experience the world from within it in bits and pieces but can imagine the pieces patched together from outside, a feat accomplished by the hippocampus working with nearby entorhinal cortex. For the details, go back to Chapter Three. Don’t worry about the names, they are simply parts of the brain and the brain has far too many parts for most mortals to remember. Our brains can extract our egos from their place embedded in the world and arrange things relative to each other rather than relative to ourselves. That is, we, in common with rats, can enter a room or a neighborhood from different directions and know where things are. Unlike rats, we can shrink that world in our minds and then put it out in the world, onto a page.
Add language to a map-like mental representation and you can talk about the world from an allocentric perspective. Tourist guides also use allocentric perspectives to give readers an overview of the interesting landmarks in a city. Here’s a way to describe the heart of DC: “The National Mall is bounded by the Washington Monument on the west and the Capitol on the east. From west to east on the north side of the Mall are the National Museum of American Art, the National Museum of Natural History, and the National Gallery of Art. From west to east on the south side of the Mall are…” You get the picture. This perspective has been called a survey perspective or an overview perspective or a bird’s-eye perspective or even an absolute perspective. Instead of locating landmarks relative to “you” using body coordinates, the reference frame of a survey perspective locates landmarks relative to each other using an external perspective, typically the cardinal directions, north-south and east-west. I made things easier by using a simple environment, one that doesn’t have landmarks on many different streets that are parallel and perpendicular. By the way, did you pick up that the description started in the west, on the left, and proceeded eastward, rightward, in (Western) reading order?
Back to the bike. Using an allocentric perspective, I could’ve said, “The bike’s against the south side of the house” (after giving you an allocentric description of the house’s location). So many complexities in what seems like a simple task: Where’s my bike? And I’ve only told you the easy parts. I’ve skipped many other fascinating and revealing opportunities for ambiguity, error, and confusion. Just in descriptions of space. This is one of many subtleties that keeps linguists in business. An aside: this is why we use gesture to locate things when face-to-face. Gesture has dual benefits: it’s more direct and it’s less ambiguous.
You might be discouraged by the complexity of crafting descriptions for what should be straightforward, describing locations in space. You might think, Just give me a map. And you’d be right: how simple and straightforward it is to sketch the relative locations of the landmarks. The simplicity of the depiction in contrast to the clumsiness of the description. Both to create and to grasp. For one thing, maps use space to represent space, a direct mapping. They can show many spatial relations all at once, not just a single route. They enable taking many different perspectives and exploring many different routes. Indeed, maps typically give us overhead or survey perspectives, but we often use them to create routes, routes that can start and end in many different places. We can also use maps, in the mind or on the page, to estimate distances and directions, so maps are not only far more compact and direct than descriptions but also far more informative.
Real-life in-the-world on-the-ground spatial descriptions. Back to talk. How do people actually talk on the ground in real life? The descriptions I provided are clearly concocted. Real-life spontaneous descriptions and directions, like any real-life talk, are messier. Even written ones: here’s how a package sent from Sweden to Kabul, the capital of Afghanistan and a city of millions, in 2017 was addressed: “To Atta Mohammed, next to Sajadee Mosque.” “Next to”—no clear perspective, vague frame of reference. And where is Sajadee Mosque? To make things worse, Atta Mohammed is a common name. It took hours of detective work and many false turns by a postal worker on a bike, but, amazingly, the package was delivered. It turns out that uncertain addresses are not at all unusual. On a recent Scandinavian Airlines
flight, I was informed by the September 2018 Scandinavian Traveler that “half the people on our planet, or four billion individuals, lack a precise address.” SAS is funding a group to rectify that (http://mapproject.se/).
When people actually talk, they casually mix frames of reference and switch perspectives, and they do that without signaling. Even more surprising, the rest of us understand the mixed perspectives and mixed reference frames and make sense out of them. This flies in the face of claims and presuppositions not only in psychology but also in linguistics: that we need a uniform, coherent perspective not only to talk and understand but also to think. Apparently not so. Actually, people quite frequently—though not always—make sense even out of nonsense. It’s the sense we’re after, the meaning.
A simple laboratory task, describing an assortment of environments learned from maps, showed decisively that people don’t consistently adopt a unified perspective. We gave people many maps to study, a convention center, a small town, a museum, an amusement park, and more, and asked them to study each map and then describe each environment in turn from memory. The spontaneous descriptions were quite good, complete and accurate enough for other people to sketch more or less the same map. Most of the descriptions mixed perspective and frame of reference, that is, they used both egocentric and allocentric perspectives, without signaling when they switched. Skip the examples below if you trust us—they’re fragments, the environments aren’t real, you don’t have the maps, and spatial descriptions, even when they’re simple, are hard.
Convention center fragment:
After you pass the bulletin board (on your left) and Camera Stores (on your right) there is an office in the back of the building in the left-hand corner. Right next to it on its right side are restrooms and a Cafeteria, respectively.
The Cafeteria is in the far right corner of the building and the restrooms are between the office and cafeteria. Next to the Cafeteria along the North side of the building is a CD store and a Stereo components store. The Stereo component store is in the NE corner of the building. There are TV and VCR stores across from the CD Store and Stereo component store.
Small town fragment:
The River Hwy. runs E-W from the river to the Mountain Rd. and further; it is to the S of the mountains. Going E on the River Hwy., at the intersection with the Mtn. Rd., there is a gas station and a restaurant on the left. On the right, before the intersection are the stables. Going N on Mtn. Rd. from the intersection towards the mountains, to the right is the town hall and to the left is the park with a gazebo.
Mixing perspectives seems to be the rule rather than the exception. It can take us a bit more time to understand descriptions that use mixed perspectives rather than consistent ones, but once we have, our own perspectives are more flexible. We get good at switching perspective. Our mental representations become perspective-free. Then we are equally fast and accurate answering questions from each perspective, irrespective of the perspective we studied. While we’re listening, it might take us longer to understand mixed perspectives, but if they are reasonably consistent and unambiguous, we can get them. Making sense of mixed perspectives might be difficult but ultimately has benefits: it makes our own thinking more flexible. Reverse the First Law of Cognition: Every cost has a benefit.
Different languages, different perspectives. Now an important qualification on frames of reference. Many of you will be wondering about other cultures, other languages. The answer is fascinating. We take the primacy of an egocentric perspective for granted. After all, we begin with our own bodies, we experience the world from our own viewpoint. There are so many voices in the research literature—I won’t name names, they’re friends!—claiming that it takes extra effort to get out of our own egocentric perspective. Yet, there are languages scattered all over the world that don’t use an egocentric perspective at all. If there are two glasses of wine on the table and the person across from you asks you whose is whose, you’re likely to say something like, Yours is on your right. If you happen to be from the outback of Australia and speak Guugu Yimithirr or if you happen to be from the highlands of Mexico and speak Tzeltal or if you happen to be from rural India and speak Tamil, you would be more likely to say, Yours is the north glass. These languages don’t use an egocentric reference frame. Instead, they use an allocentric reference frame based in the cardinal directions and sometimes called an absolute reference frame because it is a fixed reference frame, one that doesn’t depend on any individual’s perspective.
What’s especially notable about people who speak languages that rely solely on an absolute reference frame is that they know where they are with respect to the rest of the world. We’re so often clueless. They seem to keep track of their orientations in space much better than those of us who frequently rely on an egocentric frame of reference. To say or understand where one thing is relative to another, they need to know where things are relative to north-south east-west. If you blindfold them, lead them hither and thither, and then ask them to point home, they can! Do the same to residents of Amsterdam, and they’re hopeless, they point almost at random.
This fascinating discovery is the best evidence so far for the Whorf-Sapir hypothesis, the idea that the language you speak affects the way you think. There is more, but that’s for a different book. And all of it is controversial. Naturally.
Addresses and devices. Many of you are rolling your eyes again, all of this talk about talk about navigating in space is archaic. It’s like doing arithmetic; who does arithmetic anymore? Just give me an address. But an address can’t be used without an enormous amount of shared information, map-like information, maybe even a map. An address locates a building in a network of streets, it doesn’t have a start point or a way of navigating. Okay, then pull out your smartphone: that has both. But smartphones and other navigation systems depend on that same map-like information. It happens to be in the brains of the smartphone instead of in yours. Just as the operations of multiplication and division are in the brains of a calculator rather than burdening yours. You’ve outsourced parts of your brain. Soon even smartphones will be archaic; all the knowledge we will need will be embedded and updated in our minds or our bodies. We will never lose our glasses or smartphones because we won’t need them; we’ll never lose our kids in a mall or our friends in a museum because the knowledge embedded in us of where they are will be updated continuously with AI programs that understand us and our lives. Maybe we won’t even have to clarify whose wine glass is whose. We’ll see.
Places and paths. Talk is linear, one word after another. Thinking seems that way, too, one thought after another, however chaotic the order. We talk about and think about routes as sequences of nodes and links, where the nodes are places and the links are the paths that connect them. Years ago, some anthropologists asked river traders in New Guinea to sketch their routes. The traders had never been to school and had never seen a map. They drew a line for the river (straightening it!) with small circles, dots, along it for the settlements where they stopped to trade. Beads on a string.
Survey descriptions have that same character, nodes and links, places and paths. Routes arrange points, landmarks, along a single line, often turning in space to take you from A to B. Survey representations are like maps: they have many lines and many points, they arrange landmarks along multiple lines, parallel or intersecting or not connecting at all. A network.
There isn’t a single path; there’s the possibility of many. But as we said, talk is one word after another, so describing forces us to linearize a space, irrespective of perspective. Otherwise, route-like thought and survey-like thought differ deeply: a single path for a specific set of actions versus a web of paths for many possible sets of actions. These different views, insider or outsider, a single path or the possibility of many, have enormous consequences.
We began with talk, talk about the endlessly rich and complex and variable spaces in the world, the spaces that we experience constantly, that we cannot escape. Despite thei
r variability and complexity and richness (or maybe because of it), the mind abstracts them to a simple form, a network composed of places linked by paths. For networks of places, as well as for networks of events or networks of people or networks of just about any concept, we can assume a perspective from within or from without, from inside or from outside, egocentric or allocentric. Perspective, real or imagined, crucially determines what we can see. Things near our perspective loom larger: we see them more clearly and see the differences among them more clearly. Things in the distance get jammed together. Seeing things from within is vital for orchestrating our lives. Where we’ve been, where we’re going, what we just did, what we are going to do. Seeing things from above is fundamental to planning our lives and to maintaining orientation with things out of view. Switching perspective from inside to outside isn’t straightforward; we can do it if only we have the larger picture, but even then, switching perspective from inside to outside, from the here and now to the general, or from outside to inside, from general possibilities to specific plans, takes effort.
That simple structure, nodes and links, scaffolds thinking but in and of itself is devoid of meaning; the meanings of the nodes and links need to be filled in. A multitude of concepts beyond strictly spatial ones, paths, and places, can be mapped to that simple structure, and we turn to some of them in Chapter Seven, beginning with time. But even before that, a broader view, a detour to ways we organize the stuff in the world and the stuff in the mind. The same ways! Networks are only part of the story.
Mind in Motion Page 16