You Are Here (6 page)

Human beings, on the other hand, continue to search in the center of the square as defined by all four landmarks, even though the distances between them have changed. It is as if the “code” for the location of the target is made up of the relationship between the target and
all
of the landmarks, rather than any one of them.
⁷

It isn’t completely clear what these differences in behavior might mean, especially since other animals, such as certain birds, also treat landmark arrays in the human fashion. But one possibility is somewhat related to the Inuit tendency to navigate according to named landmarks connected by stories. Suppose that the four orange pylons in our hypothetical experiment were mentally encoded not as a set of four discrete objects but instead as a single shape—a square. We could even imagine that this shape is encoded in our memory not as a very specific collection of corners of a particular size and in a particular location but as something much simpler—like the mental equivalent of the word
square
.

This kind of encoding would have a couple of advantages. For one thing, it would ease our memory load considerably. But in addition, it would allow us to make predictions about how the shape of the collection of landmarks might look from different points of view without our necessarily
being
at those viewpoints. Like an Inuit navigator describing the route through a storm-ridden Arctic fjord from the comfort of the fireside, we could describe the location of the target as being at the center of the square. The whole geometric problem is boiled down to, in this case, two simple ideas—center and square. Encoding locations in this highly schematized form, though it has some advantages for memory load and mental portability, also has liabilities. Because such forms jettison geometry, we
are prone to make errors when landmarks move around. What if it
is
the case that one or two of the landmarks we’ve been using have moved? How do we then find our target? In such cases, a rat might have the advantage over us because one of the landmarks could still be used to define the position of the target. For us, on the other hand, if the four landmarks no longer define a nice square, we are left to guess where the center of the square used to be, and any guess would lack precision.

Recently, I visited a colleague in Santa Barbara at one of the campuses of the University of California. I had never been to this campus before, and so my friend sent me directions to his office. The first part of his instructions consisted of a short list of commands to be followed while driving. Provided that I paid attention to the road signs listing highway interchanges and followed the right sequence of left and right turns, I was assured of landing in the correct parking lot at the university. The crucial part of his directions dealt with what happened to me after I left the car. He instructed me to leave the parking garage and then walk so that the mountains were behind me and the ocean was in front of me. In a way, this seems to be entirely in accord with everything we discussed in the last chapter—I was following a simple taxic strategy by using the ocean as an attractive target and the mountains as a repulsive one. In spite of appearances, though, there is a crucial difference. The real target of my movements was not the ocean but my friend’s office. In other words, I was using the mountains and the ocean in an entirely different way—these large, obvious, and clearly visible objects were serving as landmarks for a target that, from my starting position, was entirely invisible. Fortunately for me, the man I was meeting was an expert in spatial navigation, so his directions, efficient and
precise, guided me to his office with laser accuracy, almost as if I were performing like a small insect on a homing mission.

When I used landmarks to find my colleague’s office in Santa Barbara, I didn’t need to think about the shapes of collections of landmarks at all. I got out of the car, set the mountains behind me and the ocean before me, and I walked. Why are situations such as these so different from those such as looking for lost keys in the grass? In part, because the landmarks involved are very large and very distant. These types of landmarks, sometimes called distal or global landmarks by scientists interested in navigation, behave differently from the clumps of grass or orange pylons that we have been discussing.

Distal landmarks would not work at all to help locate a lost wedding ring on a forest floor, but they
can
help a monarch butterfly find its way from a summer home in Canada or the United States to overwintering sites in Mexico, or a savvy hiker find her way out of a dense forest and to the nearest highway. Landmarks like the sun, moon, distant mountain ranges, or even skyscrapers or cityscapes enjoy special status among the tools used by navigators both human and otherwise because they have an ideal collection of properties that can assist wayfinders. If they are both very distant and visible, then they are also probably very large and therefore immobile. Large, immobile objects can be relied upon to stay where they are, and so define locations and directions. (The sun, moon, and stars are exceptions. Such objects change position as the earth rotates, but because such movements are predictable, we can learn their patterns and use these objects as landmarks anyway.) The other kind of landmark—pinecones near a wasp nest, clumps of grass, or bits of rock in a meadow—are not nearly as reliable. They might remain in place, or they might be kicked away, eaten, or even disturbed by curious scientists.

Many animals have such an implicit understanding of the differences between local and global landmarks that these can be shown to play different roles in their lives. Small, local landmarks are quickly disregarded when they show signs of being unreliable. Cues derived from large, distant landmarks are clung to with much more tenacity, and some studies show that animals are more reluctant to abandon representations of space based on the orientation of such masses. Determination may move mountains, but it doesn’t happen very often.

Some of the most exalted of human navigators have learned to conquer vast tracts of what, to the untrained eye, can appear as the archetypal example of flat, featureless territory—the open sea. Long before there were global positioning satellites, maps, or even compasses, human beings found ways to complete long voyages over open water employing the most subtle of cues, including landmarks.

Puluwatese marine sailors of the South Pacific routinely embarked on voyages of up to 650 kilometers in large, wind-driven outrigger canoes, navigating completely without instruments. In many cases, these expeditions were conducted for the simplest of reasons— to obtain a new stash of tobacco, or even for the men to escape the pressures of village life for a few days by fishing on a remote island.

Among the Puluwatese, the navigators were a revered class. They underwent rigorous secret training and a long apprenticeship under a master navigator before being allowed to guide a crew on the high seas. Much of this training involved painstaking memorization of the sequence of appearances and disappearances of stars along the horizon as an evening’s sail progressed. Although the path that stars take across the heavens changes from season to season, where they appear and disappear is more stable.

Most voyages consisted of a series of segments connecting islands that were close enough to be visible from one another, but, given the distances involved, target and landmark islands could be tiny brown dots in a vast blue ocean. Puluwatese navigators learned clever strategies that increased the visibility of these distant objects. Seabirds often aggregate near particular islands, and so their appearance in the sky foretells the sighting of a landmass. By attuning themselves to the identity and habits of such birds, Puluwatese navigators learned to see over the horizon, into the future. Landmasses, even small ones, affect patterns of cloud formation. As well, land and water have different effects on the appearance of the sky that overarches them. Skilled navigators learned to tune keen vision to these subtle signs so that they not only could detect an island before its silhouette crossed the threshold of visibility but could often identify it based on the reflections cast by its trees, bushes, and lagoons on the sky above.

Puluwatese navigators also devised some subtle conceptual tools that could help fill in the blanks left by the organization of the human mind. They tracked their progress at sea using a method called
etak
, which involves lining up a succession of stars near the horizon with the positions of visible landmarks. The navigators memorized these star positions during their training, and their later observations allowed them to estimate their position in the sea. Interestingly, though Puluwatese navigators learned to master complicated lists of star-landmark combinations that could help them to gauge their progress, they apparently had no real understanding of the geometry of space that made these feats possible. Try as he might, Thomas Gladwin, an anthropologist who studied the Puluwatese people, could not make them see the geometric relationship between vessel, island, and star as it might appear to an overhead observer.
⁸
Gladwin reports, somewhat mysteriously, that
the Puluwatese were able to line up star positions with the locations of landmarks even when the landmarks were not visible. Although how is not clear to me from his account, my surmise is that they were somehow able to bootstrap the estimated location of an
etak
landmark from the sum of all the information and intuition that they were able to bring to bear on their current location, perhaps also using a little intuition and blind luck.

The navigational feats of the Puluwatese are almost beyond the imagination of any but the most experienced Western sailors. It is hard to comprehend the skill of a navigator who can be blown off a capsized canoe in an ocean gale in the middle of the night and not only regain his purchase on board the vessel but also manage to compute his location and bring his crew safely home. Such sensational feats of navigation, like those of the Inuit, were mostly based on a keen observational eye and arduous practice in the patterns of star movement and the locations of visual landmarks. Such knowledge was probably so deeply ingrained through training as to have become completely automatic, probably seeming almost mystical to the Western anthropologists who studied them. There is at least a hint of mystic reverie in the description given by David Lewis of the navigational prowess of one of his main informants, the indomitable Tevake:

Reading this passage and wondering about the abilities of human beings so acutely tuned to their environment as to appear to use methods beyond the grasp of mere mortals reminds me of a conversation I once had, somewhere near the end of a long bottle of rum on an idle summer day, with a retired Nova Scotia fisherman. He told me that one of the most important skills for a fisherman was to be able to return to the same spot in the ocean from one day to the next. One didn’t necessarily want to employ marker buoys, as these would also be visible to competing fisherman, so it was better if one was able to use fixes on distant shoreline landmarks to estimate position. I asked him whether he could also return to a good fishing position when visibility was poor and shoreline landmarks were invisible, and he assured me that he could do so. When I asked him what the secret was, he fixed me with a surprised stare as if he had suddenly discovered that he was dealing with a simpleton.

“You just know when you’ve gone too far.” All too often, I wish this were true.