ALL ABOUT YOU

The combination of dogs' attention to us and their sensory prowess is explosive. We have seen their detection of our health, our truthfulness, even our relation to one another. And they know things about us at this very moment that we might not even be able to articulate.

The results of one study indicate that dogs pick up on our hormonal levels in interaction with them. Looking at owners and dogs participating in agility trials, the researchers found a correlation between two hormones: the men's testosterone levels, and the dogs' cortisol levels. Cortisol is a stress hormone—useful for mobilizing your response to, say, flee from that ravenous lion—but also produced in conditions that are more psychologically than mortally urgent. Increases in the level of the hormone testosterone accompany many potent elements of behavior: sex drive, aggression, dominance displays. The higher the men's pre-agility-competition hormonal levels, the higher the increase in the level of stress in the dogs (if the team lost). In a sense, the dogs somehow knew that their owner's hormonal level was high, by observing behavior or through scent or both—and they "caught" the emotion themselves. In another study, dogs' cortisol levels revealed that they were even sensitive to the style of play of human playmates. Those dogs playing with people who used commands during play—telling the dog to sit, lie down, or listen—wound up with higher postplay cortisol levels; those playing with people who played more freely and with enthusiasm had lower cortisol at play's end. Dogs know, and are infected by, our intent, even in play.

Being known and predicted by our dogs is no small part of our fondness for them. If you have experienced an infant's first smile at you as you approach, you know the thrill of being recognized. Dogs are anthropologists because they study and learn about us. They observe a meaningful part of our interaction with each other—our attention, our focus, our gaze; the result is not that they can read our minds but that they recognize us and anticipate us. It makes the infant human; it makes the dog vaguely human, too.

Noble Mind

It's dawn and I try to sneak out of the room without waking Pump. I can't see her eyes, so dark they're camouflaged against her black fur. Her head rests peacefully between her legs. At the door I think I've made it—tiptoed and breath-held to avoid her radar. But then I see it: the swell of her lifted eyebrows tracking my path. She's on to me.

The dog, as we've seen, is a master looker, a skilled user of attention. Is there a thinking, plotting, reflective mind behind that look? The development of the human infant's looking into using attention marks the blossoming of the mature human mind. What does the dog's looking tell us about the dog mind? Do they think about other dogs, about themselves, about you? And the timeworn but still unanswered question of dog minds: Are they smart?

DOG SMARTS

Dog owners, like new parents, always seem to have a handful of stories at the ready describing how smart their charges are. Dogs, it is claimed, know when their owners are going out, and when they are coming home; they know how to hoodwink us and they know how to beguile us. News reports buzz with the latest discovery of the intelligence of dogs: of their ability to use words, count, or call 911 in an emergency.

To verify this anecdotal impression, some have designed so-called intelligence tests for dogs. We're all familiar with intelligence tests for humans: pen-and-paper creations that require you to solve SAT-like problems of word choice, spatial relationships, and reasoning. There are questions that test your memory, your vocabulary, your declining math skills, and your simple pattern-finding ability and attention to detail. Even putting aside whether the result is a fair assessment of intelligence, the design does not translate obviously to testing dogs. So revisions are made. Instead of tests of advanced vocabulary, there are tests of simple command recognition. Instead of repeating a list of digits read aloud, a dog may be asked to remember where a treat was hidden. Willingness to learn a new trick may replace the ability to figure complex sums. Questions loosely mimic experimental psychology paradigms: of object permanence (if a cup is placed over a treat, is it still there?), learning (does your dog realize what foolish trick you desire him to do?), and problem solving (how can he get his mouth on that food you've got?).

Formal studies of groups of dogs on these kinds of abilities—mostly cognition about physical objects and the environment—yield what at first seem to be unsurprising results. By bringing dogs to a field baited with treats and timing dogs' speed in finding them, researchers have confirmed that dogs use landmarks to navigate and find shortcuts. This behavior is consistent with what their wolflike ancestors would probably have done in finding food and finding their way. Dogs are, of course, pretty good at all tasks that involve getting themselves to food. Given a choice of two piles of food, dogs have no trouble choosing the larger one—especially as the contrast between them grows. Turn a cup over a bit of food and dogs go right for it, knocking the cup and revealing the treat. Dog subjects have even learned how to use a simple tool—pulling a string—to get an attached biscuit that was otherwise out of reach.

But dogs don't pass all the tests. They typically make lots of mistakes when presented with piles of three versus four biscuits, or of five and seven: they choose the smaller amounts just as often as the larger. And they develop preferences for piles on the left or the right, which lead them to make even more blatant errors. Similarly, their skill at finding hidden food gets worse as the hiding gets more complicated. And their tool use also starts to look less impressive as the trials get trickier. When there are two strings, and only the more distant one is attached to an alluring biscuit, dogs nonetheless go for the nearer string, the one attached to nothing. They don't seem to understand the string as a tool: as a means to an end. Indeed, they may have succeeded in the original case simply by pawing and mouthing at the problem until accidentally solving it.

A dog owner tallying her dog's score in these dog intelligence tests might find that he's scoring closer to Dim but happy than Top of the obedience class. Is that it, then? Is he not smart after all?

A closer look at the intelligence tests and the psychological experiments reveals a flaw: they are unintentionally rigged against dogs. The flaw is in the experimental method, not in the experimented dog. It has to do with the very presence of people—experimenters or owners. Let's look more closely at a typical experimental setup. It might begin as follows: A dog is sitting at attention and being restrained by a leash. An experimenter comes before him and shows him a great new toy. This dog loves new toys.* The toy and a bucket are clearly shown to the dog, the toy is put into the bucket, and then the experimenter disappears with the booty behind one of two screens in the room. She returns with the bucket—emptied of its treat. This turns out not to be a cruel hoax, but a standard test of invisible displacement: wherein an object is displaced—moved to another location—invisibly—out of sight. This test has been regularly run with young children since Piaget proposed it as representing one of the conceptual leaps that infants make on their way to becoming incorrigible teenagers and then adults capable of having infants of their own. In this case, the conceptual understandings are of the continued existence of objects when they are out of sight—called object permanence—and some notion of that object's trajectory and continued existence in the world. If someone disappears behind a door, we realize not only that they still exist when we can't see them, but that we might find them by looking behind that door. Children master object permanence before their first birthday, invisible displacement by their second. Since Piaget reified this representational understanding as a stage in infant cognitive development, it is a standard test that is run with other animals, to see how they compare to little people. Hamsters, dolphins, cats, chimpanzees (who reliably pass), and chickens have all been tested. And dogs.

Dogs' performance is mixed. Oh, sure, if the test is run simply as described, then they have no trouble looking behind the screen for the toy. It looks as though they've passed the test. But complicate the scenario a little—carry the container behind two different screens, taking the toy out after the first screen and showing them that you have done so before going behind the second screen—and dogs fail: they race to the second screen first, where the toy clearly is not. Other test variations also result in dogs suddenly looking less smart in their searching. We could conclude that here too the dogs appear to be less than genius. Once the toy is out of sight, it may quickly fall out of mind.

But the very fact that dogs do succeed, sometimes, renders that conclusion suspect. Instead their behavior points to two explanations. First, it is likely that dogs remember the toy, but do not engage in detailed consideration of what its path might be when it vanishes. Though some dogs are indisputably keen to keep track of a toy, dogs nonetheless regard objects in their environment very differently than humans do. Significantly, what wolves and dogs do with objects is limited: some objects are eaten, and some are played with. Neither interaction requires complex rumination on the object. Dogs realize when a previously treasured object is missing, but needn't mull over possible stories for what happened to it. Instead they just start looking for it, or wait for it to show up.

The second explanation is more far-reaching. It appears that the very skill at social cognition that is their triumph as a companion to humans contributes to the dogs' failure at this and other physical-cognition tasks. Show your dog a ball, then conceal it from him while you place it under one of two overturned cups. Faced with the cups, and assuming he can't smell it out, a dog will look under either cup at random: a reasonable approach when he has nothing to go on. Lift one cup to reveal a peek of the ball underneath, and you won't be surprised that when allowed to search, your dog will have no trouble looking under that cup. But give a peek under the cup holding nothing, and researchers found that dogs suddenly lose their logic. They search first under the empty cup.

These dogs were stymied by their own skill. When presented with a problem of any kind, dogs cleverly look to us. Our activities are sources of information. Dogs come to believe that our actions are relevant—often leading, we might note, to some interesting reward or even food. So if an experimenter ducks behind a second screen, as she does in the more complicated invisible displacement tasks, why, there might be something of interest behind that screen. If she lifts up an empty cup, that cup becomes more interesting simply because of her attention to it.

If the social cues are diminished in the tests, dogs perform much better. When experimenters handle both cups even when showing the dog the empty one, dogs regain their heads. They see the empty cup, and by deduction search under the other cup, which holds the hidden ball. Similarly, dogs who are less well socialized—such as yard dogs kept outside for most of their hours—also set right to the problem, while dogs who live inside the house more often plead quietly with their owners to help.

If we revisit some of the problem-solving tests on which wolves performed so much better than dogs, we now see that the dogs' poor performance can there too be explained by their inclination to look to humans. Tested on their ability to, say, get a bit of food in a well-closed container, wolves keep trying and trying, and if the test is not rigged they eventually succeed through trial and error. Dogs, by contrast, tend to go at the container only until it appears that it won't easily be opened. Then they look at any person in the room and begin a variety of attention-getting and solicitation behaviors until the person relents and helps them get into the box.

By standard intelligence tests, the dogs have failed at the puzzle. I believe, by contrast, that they have succeeded magnificently. They have applied a novel tool to the task. We are that tool. Dogs have learned this—and they see us as fine general-purpose tools, too: useful for protection, acquiring food, providing companionship. We solve the puzzles of closed doors and empty water dishes. In the folk psychology of dogs, we humans are brilliant enough to extract hopelessly tangled leashes from around trees; we can magically transport them to higher or lower heights as needed; we can conjure up an endless bounty of foodstuffs and things to chew. How savvy we are in dogs' eyes! It's a clever strategy to turn to us after all. The question of the cognitive abilities of dogs is thereby transformed: dogs are terrific at using humans to solve problems, but not as good at solving problems when we're not around.

LEARNING FROM OTHERS

Yesterday Pump learned, courtesy of a pet supermarket's automatic doors, that when you walk toward walls, they open and let you pass through them. Today she unlearned it, in a spectacularly poignant display.

Once a problem is solved—a hidden treat is unearthed, an unjustly closed door is opened—with or without a person's help, the dog is quickly able to apply that same means to solve it again and again. He has identified a state of affairs, fashioned a response, and realized the connection between that problem and that solution. This is both his triumph and, at times, our misfortune. One success at jumping right onto the kitchen counter to get to the origin of that pleasing cheese odor will be followed by much jumping-on-counters. If you provide a sitting dog with a biscuit for sitting politely, expect to be inundated by polite sits. With this in mind it is easy to understand the admonishment that in training a dog you must reward only those behaviors you desire the dog to repeat endlessly.

Such is the dog's mastery of what in psychological circles is termed learning. There is no doubt that dogs can learn. It is the natural workings of any nervous system to adjust its actions over time in response to experience—and of every animal with a nervous system to thereby learn. Under the heading "learning" comes everything from the associative learning used in animal training, to memorization of a Shakespearean monologue, to finally understanding quantum mechanics.

Dogs' easy mastery of new procedures and concepts presumably stops prior to grasping what a quark is. What they learn is neither academic nor scholastic. Still, most of what we ask that dogs learn can only be described as capricious and arbitrary. Surely any animal recently wild will learn how to get its mouth on food. But typically the things we want dogs to learn—to obey—bear little connection to food. We ask dogs to change posture (to sit, jump up, stand up, lie down, roll over), to act in a very specific way on an object (get my shoes, get off the bed), to start or stop a current action (wait, no, okay), to change mood (cool it, go get him!), to move toward us or move away from us (come, go away, stay). This may not be quantum mechanics, but it is just as bizarre to these distant moose hunters. Nothing in a wild animal's life prepares him to be asked to maintain the state of holding his rump on the ground, unmoving, until released by your cheery okay! It is notable that dogs can learn these seemingly arbitrary things at all.

PUPPY SEE, PUPPY DO

One morning, on awakening lying on my belly, I pulled my arms over my head, stretched my legs into pointed toes, and pulled myself up onto my forearms. Aside me Pump stirred, and matched me move for move: she tensed her front legs, stretched them well out in front of her, then straightened her back legs, too, pulling herself forward into uprightness. Now we greet each other every morning with parallel wakening stretches. Only one of us swings her tail.

Even more interesting than learning commands would be the ability to learn by merely watching others—other dogs or even people. We know dogs can learn from our instruction, but can dogs learn from our example? It would seem to behoove a social animal like the dog to look to others for information about how best to negotiate the world. In many cases, though, the answer to this question is clearly no: dogs have plenty of opportunity to see us eating politely at the table—yet they never spontaneously pick up knife and fork and join us. Overhearing us talking is insufficient to get them talking; their only interest in clothes seems to be chewing them, not donning them. Amply exposed to our activities, dogs don't seem to know how to imitate us.

This is not a failing, though it would distinguish them from members of our own species, consummate imitators that we are. As children and into adulthood, we goggle at each other to see what to wear, what to do, how to act, and how to react. Our culture is built on our keenness in observing others act to learn how to behave ourselves. I need only see you opening a tin can with a can opener once before I can do it myself (one hopes). The stakes are higher than they might first seem, for success at imitation not only gets you the contents of the opened can, it is an indication of a complex cognitive ability. True imitation requires that you not merely can see what another is doing, not simply that you see how the means lead to an end, but also that you translate others' actions into your own actions.

In that case, dogs are not true imitators, for even after thousands of demonstrations with the can opener, no dog has shown an interest: the opener's functional tone is mute for them. But this is not a fair comparison, you might complain: dogs simply haven't the thumbs, nor the dexterity they allow, to operate can openers or cutlery. Similarly, they haven't the larynx for speech nor the need for clothing. And your complaint would be fair: the question is really if the dogs can be taught, by demonstration, how to do something new—not whether they are mini-humans.

Watch dogs interact for ten minutes and you will see what looks like imitation: one dog flaunts a gloriously large stick; the other finds a stick of his own and flaunts it back. If one dog finds a spot for digging, others will soon join him at the growing hole; one dog's discovery that he can swim leads another dog to self-baptize, suddenly finding himself swimming, too. By watching others, dogs learn the special pleasures of mud puddles and of bushwhacking through brush. Pump uttered nary a peep until one of her regular dog companions began barking at squirrels. All at once, Pump too was a squirrel-barker.

The question, then, is whether these are cases of true imitation, or of something else. The something else that it might be is opaquely called stimulus enhancement. A minor incident involving birds and home-delivered milk in mid-twentieth-century Britain demonstrates this phenomenon best. At the time, doorstep milk delivery was commonplace in Britain, and homogenization was not. Thus dawn found foil-capped bottles of separated milk idling unattended on front porches, the cream nearest to the top of the bottles. Up at dawn with the delivery men is much of Britain's bird population, for dawn is a propitious time to sing. One bird, the small blue tit, made a discovery: the foil on the bottles was susceptible to being pecked through, revealing a rich creamy drink just below. A few reports of vandalized milk bottles were lodged, soon a spate more, then a plague of them. Hundreds of birds had learned the milk-bottle trick. Cross with their skimmed milk, the Britons were not long in finding the culprits. For us, the question is not who but how: How did this discovery spread among the blue tits? Given the rapidity with which it spread, it seemed likely that some birds observed others getting the cream, and imitated them doing so. Clever, pudgy little birds.

By providing a captive population of chickadees with a similar setup, one group of experimenters observed the phenomenon recur step-by-step. Their studies suggest a more likely explanation than imitation. Instead of carefully observing and assimilating all that the first, cream-pilfering bird was doing, other birds simply saw that he was atop the bottle. This may have attracted them to the bottles. Once landed on the bottle tops, by doing a natural behavior—pecking—they discovered the foil's puncturability themselves. In other words, they were drawn to a stimulus, the bottle, by the first bird's presence. Its presence enhanced the likelihood that they too would become cream stealers, but it did not demonstrate how to do so.

This may seem nitpicky, but there is an important difference at work here. In a case of stimulus enhancement, I see that you are acting in some unspecified way on the door, after which it opens. If I amble over to the door and kick it, hit it, and otherwise maul it, I might get it to open, too. In a case of imitation, I watch exactly what you are doing with the door and reproduce just those actions—the seizing and turning of the knob, the application of pressure after turning, and so on—that lead to the desired outcome. I can do that because I can imagine that what you are doing is somehow related to your goal, your desideratum: to leave the room through the door. The blue tits, on the other hand, need not have been thinking about what the milk bottle tits wanted—and probably were not.

MORE HUMAN THAN BIRD

Dog researchers wanted to test whether the stick-flaunting dogs are acting more like a blue tit or more like a human being. The first experiment was designed to determine if dogs would imitate humans in a situation in which the people were acting to attain some desired object. The researchers were asking, in essence, whether dogs can understand that a person's actions function as a demonstration that can be followed if the dog is otherwise unsure how to get that desired object himself.

They set up a simple experiment in which a toy or a bit of food was placed in the crook of a V-shaped fence. The dog was seated on the outside of the point of the V, and was given a chance to try to retrieve the food. He couldn't go straight through or over the fence, but both routes around the fence—around the left stem or the right stem—were equally long, so equally good. When given no demonstration of how to get around the fence, the dogs chose randomly, preferring neither side, and eventually making their way to the inside of the V. But when given a chance to watch a person walking around the left side of the fence toward the reward—a person actively talking to the dog along the way—the observing dogs changed their behavior outright: they also chose the left side.

It looks as though these dogs were imitating. And what they learned by imitating stuck: when a shortcut through the fence was later introduced, they maintained the route they had learned by watching, ignoring the shortcut. The researchers ran a handful of other trials to be clear what exactly it was that the dogs were doing. They were not simply navigating by smell: laying down a scent trail on the left arm of the fence did not induce dogs to follow it.*

Instead it had something to do with understanding others' actions. Simply watching someone quietly walk around the fence was insufficient to get the dogs to follow the person's route: the person had to be calling the dog's name, grabbing attention, yammering away. Watching another dog who had been trained to retrieve the reward by the left-hand route also prompted observing dogs to go left.

This result showed that dogs can see others' behavior as a demonstration of how to get to a goal. But we know from experience with our dogs that not every relevant behavior we do is seen as a "demonstration." Pump may watch me navigate around strewn chairs, books, and clothes piles as I head to the kitchen, but she will herself charge right through piles to take the quickest route. Other tests are necessary to determine if dogs are truly putting themselves in our shoes, and not just prone to follow that human, wherever we go.

Two experiments have tested just this imitative understanding. The first asked what exactly dogs see in others' behavior: the means or the end. A good imitator would see both, but would also see if the particular means isn't the most expedient way to the end. From a young age, human infants can do just that. They will religiously imitate—sometimes to a fault*

—but they can also be astute. For instance, in one classic experiment, after watching an adult turn on a light in an unusual way—with his head—the infant subjects could imitate this novel action, if asked to do so. But they did not spontaneously imitate if the adult was grasping something in his hands, making him unable to use them to turn on the light: the infants used their hands, reasonably enough. If the adult held nothing in his hands, infants were more likely to turn on the light with their heads, too—inferring, perhaps, that there must be good reason, besides one's hands being full, for this new maneuver. They seemed to realize that the adult's actions could be imitated, and they selectively imitated them only insofar as it seemed necessary to do so.

In the dog variation on this paradigm, a wooden rod taking the place of a light, one "demonstrator" dog was taught to press the rod with his paw to release a treat from a spring-loaded dispenser. The researchers then had the demonstrator dog perform his newfound trick in front of other dogs who were being restrained to watch. In one trial the demonstrator pressed the rod while holding a ball in his mouth; in the other, he had no ball. Finally, the observer dogs were let at the apparatus.

It should be noted that dogs are not naturally drawn to mechanical dispensers, even ones with wooden rods. And pressing is not the first approach of most dogs when facing a problem: dogs can use their paws handily, but they typically go at the world mouth first and paws second. Though they can be trained to push or press an object, dogs' first approach at an object such as this one is not one of intuitive understanding. They will bump it, mouth it, knock into it. If they can, they will push it over, dig at it, jump on it. But they do not consider the scene for a moment and then calmly press the rod. Thus the first approach of the observer dogs was particularly interesting: Would the demonstration change their behavior?

These dog subjects behaved just like the human infants with the light switches: The group that saw the demonstration with no ball imitated faithfully, pressing the rod to release the treat. The group that saw the demonstrator acting while holding a ball in his mouth also learned how to get the treat, but used their (ball-less) mouths instead of paws.

That the dogs so imitated is remarkable. This is no mere mimicry, copying for copying's sake. Nor is it just an attraction to the source of activity. It looks more like the behavior of an animal who is considering what another animal is doing: what his intention is, and how—or how much—to reproduce that behavior themselves, if they have the same intent.

If these experiments represent the performance of all dogs, it looks as though we could say that dogs are, at the very least, able to learn by watching others in particular social contexts—when food is at stake, for instance. One final experiment suggests something even more impressive: that dogs may actually understand the concept of imitation. The single subject, an assistant dog trained to work with the blind, had already learned by operant conditioning to do a number of non-obvious actions on command: to lie down, turn around in a circle, put a bottle in a box. What the experimenters wondered was whether he could do these actions not just to a command, but after seeing someone else do the action themselves. Sure enough, the dog ably learned to turn around in a circle not after the Turn around in a circle command, but on simply seeing a human do such a thing, followed by the imitation request Do it! They then examined what he would do when seeing a human do a new, completely odd action, such as running off to push a swing, tossing a bottle, or suddenly walking around someone else and returning to their starting spot.

He did it. It was as though this dog had learned the concept imitate, and, given that notion, could apply it more or less in any direction. To do this, he had to map his body onto a human's: where a person tossed a bottle by hand, the dog used his mouth; he used his nose to push the swing. This is not the final word about imitation (just ask your dog to copy your swing-pushing, and you can see how results do not always generalize), but these dogs' abilities are suggestive of something besides mindless mimicry. Dogs may be enabled to imitate by the same ability—almost compulsion—to look at us that allows them to use us to learn how to act. That is what I see in Pump's morning stretch alongside me.

THEORY OF MIND

I open the door stealthily and Pump's there, not two feet away, walking toward the rug with something in her mouth. She stops in her tracks and looks over her shoulder at me, her ears down, her eyes wide. In her mouth is an unidentifiable curved form. As I slowly approach, she wags low, ducks her head, and in the moment that she opens her mouth to get a better grip on her find I see it: the cheese left out on the counter to warm. The brie. The entire enormous round of brie. She gulps two gulps and it's gone, down the gullet.

Think of the dog caught in the act of stealing food from the table … or looking at you squarely in the eyes with a plea to go out, be fed, be tickled. When I see Pump, mouth full of brie, seeing me, I know she's going to make a move; when she sees me seeing her, does she know I'm going to try to thwart it? My strong impression is that she does: the moment I open the door and she looks at me, we both know what the other is going to do.

The study of animal cognition reaches its pinnacle in addressing just this kind of scene: one raising the question of whether an animal conceives of others as independent creatures with their own, separate minds. This ability seems more than any other skill, habit, or behavior to capture what it is like to be a human: we think about what others are thinking. This is called having a theory of mind.

Even if you've never heard of theory of mind, chances are you nonetheless have a very advanced one. It allows you to realize that others have perspectives different from your own, and therefore have their own beliefs; different things they know and don't know; a distinct understanding of the world. Without one, others' behavior, even the simplest acts, would be utterly mysterious, arising from unknown motivations and leading to unpredictable consequences. Trying to guess what a man approaching you, mouth agape, arm raised high, hand waving frantically, is going to do is greatly aided by having a theory of mind. It's called a theory because minds are not directly observable, so we extrapolate backward from actions or utterances to the mind that prompted that act or remark.

We aren't born thinking about others' minds, of course. It is quite likely that we aren't born thinking about much at all, even our own minds. But each normal child develops a theory of mind eventually, and it appears that it is developed through the very processes discussed so far: through attending to others, and then noticing their attention. Children with autism often don't develop some or any of these precursory skills: they may not make eye contact, point, or engage in joint attention—and many don't seem to have a theory of mind. For most people, it is but one large theoretical step from an awareness of the role of gaze and attention to realizing that there is a mind there.

The gold standard experiment for theory of mind is called the false belief test. In this design, the subject, typically a child, is presented with a minidrama played out by puppets. One puppet places a marble in a basket in front of her, in full view of the subject and a second puppet. Then the first puppet leaves the room. Promptly, the second puppet wickedly moves the marble over to her basket. As the first puppet returns, the subject is quizzed: Where will the first puppet go looking for her marble?

By age four, children answer correctly, realizing that they and the puppet know different things. Before that age, though, children surprisingly and unambiguously fail. They say the puppet will look for the marble where the marble actually is—in the second basket—showing that they aren't thinking about what the first puppet really knows.

To design a verbal false belief task for animals, who cannot be expected to communicate their answers (nor be engaged by a puppet marble-switching drama) is nigh impossible, so nonverbal tests have been developed. Many take their cue from anecdotal reports of compellingly mindful animal behavior seen in the wild: of deception or clever competitive strategies. Chimpanzees are the most frequent subjects since, as close relatives to humans, one might expect that they would have the most similar cognitive abilities.

While the results with chimpanzees have been equivocal, lending credibility to the notion that only humans have a fully developed theory of mind, a wrench has been thrown in the experimental works. That wrench is the dog: whose attention to attention, whose seeming mind reading, looks anecdotally just like what we call acting with a theory of mind. To go from my living-room theorizing about a dog's understanding of mind to solid scientific standing, researchers have begun to run dogs on the same tests used with chimps.

THEORY OF DOG MIND

Here's what one dog, an unsuspecting experimental subject, found awaiting him at home one day. Instead of the usual ready availability of his favorite tennis balls, every ball in the house had been collected, and an extra lot of people were standing about gazing at him. Fine so far: Philip, the three-year-old Belgian Tervuren in question, didn't freak out. But he might have been bemused when, one by one, the balls were shown to him and then placed in one of three boxes and locked up. This was new stuff. Whether game or threat, what was clear was that the balls were being methodically placed somewhere other than his favorite place: right in his mouth.

When released by his owner, Philip went, naturally, straight toward the box where he saw a ball hidden, and he nuzzled the box. This turned out to be the right thing to do, for it prompted the humans to exclaim merrily, open the box, and give him the ball. Despite just having his mouth on the ball, the dog found that the people around him kept taking it away and securing it in one or the other box—so he kept playing along. Then they started locking the boxes and putting the key elsewhere, so the whole thing took even longer after he selected the right box: someone must find the key, bring it to the box, and open it. The final twist involved one person who locked the box, hid the key, then left the room. Another person came in—surely one who, like all other people around, would be able to use these key-things to open these lock-things.

This was the moment the experimenters were waiting for: they wanted to know if the dog saw the new person as unknowledgeable about the location of the key. If so, then not only should Philip indicate which box has the beloved ball, he should also help the person find the key that would enable access to that ball.

On repeated trials, that's more or less what the dog did: ever patient, Philip looked toward the spot where the key had been hidden, or headed that way. Note that he didn't actually take it in his mouth and open the box: that'd be some trick, but even the most ardent dog enthusiast will admit it's unlikely. Instead, Philip used his eyes and his body as communications.

Philip's behavior could be interpreted in three ways: one functional, one intentional, one conservative. The functional interpretation is this: the dog's gaze served as information for the person, whether the dog meant it to or not. The intentional: the dog did in fact mean it to: he looked because he knew the person was ignorant of the key's location. The conservative: the dog looked reflexively, since someone was recently over there where the key was.

The data do the interpreting. They show that the functional is definitely true: gaze did serve as information to the person nearby. But the intentional take is also true: the dog looked at the location of the key more often when the person in the room with them was ignorant where it was—as if meaning to inform the person with his gaze. That nixes the conservative interpretation. Philip seemed to be thinking about these crazy experimenters' minds.

This is but one dog—maybe a particularly astute one. Remember the begging experiment run with chimps and dogs? Unlike chimps, all the tested dogs immediately followed the knower's (non-blindfolded or bucketed person's) advice as to which box was baited with food. Hoorah for these dogs, who thus all found food inside. This looks good for the theory of dog mind: they acted as though thinking about the knowledge states of the strange people pointing in front of them. But after this seeming cognitive accomplishment, a strange thing happened. When run again and again on the same test, these dogs changed their strategies. They began to pick the guesser just about as often as the knower. Does this mean they were prescient and then grew dimwitted? Although dogs will do impressive convolutions for food, this doesn't make sense as an explanation. Perhaps it indicates that the first round was a fluke.

The best interpretation is that the dogs' performance on the task makes a methodological point. There may be other cues the dogs are using to make their decisions that are, to them, just as strong as the presence or absence of the guesser is to us. Consider, for instance, that all humans are on the whole highly knowledgeable about the sources of food, from a dog's point of view. We are regularly around food, we smell like food, we open and close a cold box filled with food all day long, and sometimes we even have food dribbling out of our pockets. This is such a well-learned feature of us that it might be hard to overturn on the basis of a few trials one afternoon. This hypothesis is borne out by the fact that the dogs did use the people to make their decision: they never chose a third box, unselected by either the guesser or the knower.

However we interpret the results, though, the dogs are not going out of their way to prove to us that they have a theory of mind. Of course, one of the difficulties of designing experiments for any animal is that, as the procedure grows more complicated in order to test for a very specific skill, it risks becoming an exceedingly strange scenario for the animal. One might suggest that massive confusion on the part of the subjects is not unreasonable. They are often thrust into situations that are bizarre: that are, in fact, intentionally unlike anything they've seen before. People appear with buckets over their heads; trials go on endlessly; it is in every way not normal. Dogs nonetheless sometimes manage to perform well at the tasks in front of them.

Still, their natural behavior—in a natural setting—is a better indication. What do dogs do without the peculiarities of baited and locked boxes and uncooperative humans to puzzle over? Their most representative behavior will appear in dealing naturally with other dogs or with humans. If it is socially helpful for a dog to consider what other dogs are thinking, the ability to do so may have evolved—and may still be visible in social interactions. This is why I spent a year watching dogs play: playing in living rooms and veterinary offices, down hallways and pathways, on beaches and in parks.

PLAYING INTO MIND

Pump appears in the corners of all the videos: in one, she hops nimbly to avoid collision with a dog approaching too fast—then pursues him as he rushes out of the frame. In another, she lies prone with another dog, feigning bites with open mouths. In a third she tries and fails to join two dogs in play; as they run off she is left wagging alone in the camera's eye.

I should correct myself: I was lucky enough to spend a year watching dogs play. What is called, appropriately, "rough-and-tumble" play between two competent, athletic dogs is a gymnastic marvel to witness. The playing dogs seem to give a perfunctory greeting to each other before they suddenly mutually attack, teeth bared; tumbling together in precarious free fall; jumping on and over each other; bodies bent and tangled. When they stop, suddenly, at a noise nearby, they may be the pictures of quiet. It takes only a look or a paw raised in the air to engage in their shared havoc again.

Play might seem just like that thing dogs do, but it has a very particular scientific definition. Animal play, science intones, is a voluntary activity incorporating exaggerated, repeated behaviors, extended or truncated in duration, varied in fortitude, and atypically combined; and using action patterns that have identifiable, more functional, roles in other contexts. We don't just define play this way to take the pleasure out of it: we define it to reliably recognize it. Play also has all the attributes of a good social interaction: coordination, turn taking, and, if necessary, self-handicapping—playing at the level of one's play partner. Each partner takes the abilities and behavior of the other into account.

The function of animal play is a bit of a puzzle. Most animal behaviors are described by how they function to improve the survival of the individual or species. The search for a function of play is paradoxical, as it looks like behavior which is clearly function less: at the end of play, no food has been gained, no territory secured, no mate wooed. Instead two dogs pantingly collapse on the ground and wag their tongues at each other. One might thus suggest that the function is to have fun—but this is frowned upon as a true function, because the risks are too great. Play takes a lot of energy, can cause injuries, and, in the wild, increases an animal's danger of predation. Play-fighting can escalate into true fights, causing not just injury but social upheaval. Its riskiness makes the case for a real, undiscovered function of play even more compelling: it must be terribly useful to play, if this behavior survived the evolutionary process. It might serve as practice: a context in which to hone physical and social skills. Strangely, though, studies have shown that play is not essential to adult proficiency at the skills practiced in play. Maybe play serves as training for unexpected events. It does seem that volatile and unpredictable play is deliberately sought. In humans, play is part of normal development—socially, physically, and cognitively. In dogs, it may be the result of having spare energy and time—and owners who live vicariously through their dogs' tumblings.

Play among dogs is particularly interesting because they play more than other canids, including wolves. And they play into adulthood, which is rare for most playing animals, including humans. Although we ritualize play into team sports and solo video game marathons, as sober adults we rarely spontaneously blindside and tackle our friends, tag them and run, or make faces at each other. The hobbling, slow-moving fifteen-year-old dog on the block looks warily at the enthusiasm of young puppies approaching him, but even he occasionally play-slaps and bites at a younger dog's legs in play.

In my study of dog play I shadowed dogs around with a video camera rolling, and controlled my own delighted laughter at their fun long enough to record bouts of play, from a few seconds to many minutes long. After a few hours of this the fun stopped, the dogs would get packed into the backs of cars, and I would walk home, reflecting on the day. I'd sit down in front of my computer and play back the videos, at an extremely slow rate: slowed enough to see each frame—thirty of which fill a second—individually. Only at this speed could I really see what had happened in front of me. What I saw was not a repeat of the scene I'd witnessed at the park. At this speed I could see the mutual nods that preceded a chase. I saw the head-jockeying, open-mouth volleys that blurred into unrecognizability in real time. I could count how many bites it takes, over the course of two seconds, before a bitten dog responds; I could count how many seconds it takes for a paused bout to resume.

And, most important, I could look to see what behaviors dogs do, and when. Watching the play deconstructed into these subsecond moments enabled me to record a long catalog of the behaviors of each dog: a transcript of the play. I also noted their postures, their proximity to one another, and which way they were looking at every moment. Then, so deconstructed, the play could be reconstructed to see what behaviors match what postures.

In particular, I was interested in two kinds of behaviors: play signals and attention-getters. Attention-getters, as we've seen, are obvious things: they serve to get attention. Specifically, they are acts that alter the sensory experience of someone else—someone whose attention you're keen to have on you. They can be an interruption of the visual field, as when Pump suddenly puts her head between me and the book I'm holding. They can interrupt the auditory environment: a car's honk is so intended, and dogs' barks are so as well. If these methods fail, attention can be gotten by interacting physically: a hand on the shoulder; a paw on the lap; or, between dogs, a bump with the hip or a light bite on the rump. Clearly, many things we do are in some way attention-getting, but not every behavior is equally good at the task.

Calling your name out may be a way to catch your attention—but not if we are in Yankee Stadium in the bottom of the ninth. Then a more extreme method (and possibly an organist) would be necessary. Similarly, dogs' attention can be more or less easy to get. Between dogs, what I called an in-your-face—presenting oneself in front of, and very close to the face of, another dog—is effective at getting attention—but not if the dog is engaged in rollicking play with someone else. Then more forceful means are needed—thus explaining those dogs who circle a playing pair for minutes barking barking barking. (Better, perhaps, to interject some nice rump bites in there with the barking, if you are truly eager to break up the game.)

Play signals, the other behaviors, are requests for play or announcements of interest in playing: they could be translated as saying something like Let's play or I want to play or even Ready? because I'm about to play with you. What the specific words are is not as important as their functional effect: play signals are reliably used to begin and to continue play with others. They are a social requirement, not just a social nicety. Dogs typically play together rambunctiously and at a breakneck pace. Since they are doing all manner of actions that could easily be misinterpreted—biting each other on the face, mounting from behind or fore, tackling the legs out from under another dog—the playfulness of their actions has to be manifest.* If you fail to signal before biting, jumping on, hip-slamming, and standing over your playmate, you are not in fact playing; you are assaulting him. A bout wherein only one participant thinks it's play is no longer playful. All dog owners who walk their dogs among others know what then happens: a play bout becomes an attack. Without the play signal, a bite is a bite, worthy of rancor or retribution. With it, a bite is just part of the game.

Nearly every play bout begins with one of these signals. The quintessential signal is the play bow, in which the dog's body genuflects in front of a desired play partner. A dog bent on his forelegs, mouth open and relaxed, with his rump in the air and tail high and wagging is pulling out all the stops to induce someone to play. Even tailless, you can mimic this pose yourself; expect a response in kind, a friendly nip, or at least a second look. Two dogs who are regular playmates may use a bow shorthand: familiarity allows abbreviations in formality, just as between human acquaintances. Just as How do you do? became Howdy, the play bow can be shortened into the aforementioned play slap, the front legs clapping the ground at the beginning of the bow; the open mouth display, the mouth opened but without the teeth bared; or the head bow, a bobbing of the head with opened mouth. Even panting in quick bursts can be a signal to play.

It is how dogs might use these play-signal and attention-getting behaviors together that could reveal or refute that dogs have a theory of mind. In just the way the false belief task shows that some children are thinking about what other people know, and some are not, one's use of attention in communicating is meaningful. The key question I asked of my data of playing dogs was this: Did they communicate, using play signals, intentionally—with attention to the attention of their audience? And did they use attention-getters when they didn't have their play partner's attention? Just how were those bumps, barks, and bows of play used?

It's hard to give a good account of what's happened in a bout of play you have just watched. Sure, I could create a very simplistic story line between two dog protagonists—Bailey and Darcy ran around together … Darcy chased Bailey and barked … they both bit at each other's faces … then they split—but it glosses over the details, such as how often Darcy and Bailey self-handicapped, intentionally throwing themselves on the ground on their backs to be bitten, or using less force in a bite than they could. Whether they took turns in biting and being-bitten; chasing and being-chased. And, most critically, whether they signaled to each other when the signal could be seen and responded to—with play or by hightailing it out of there. For this, you need to look at the moments between the seconds.

What I found there was remarkable. These dogs play-signaled only at very particular times. They signaled reliably at the beginning of play—and always to a dog who was looking at them. Attention might be lost a dozen times in a typical play session. One dog gets distracted by a ripe smell underfoot; a third dog approaches the playing pair; an owner wanders away. What you might notice is simply a pause followed by a resumption of play. In fact, in these cases, a quick series of steps needs to be followed. For the play not to be permanently severed, the interested dog must regain his partner's attention and then ask him to play again. The dogs I observed also play-signaled when the play had paused and they wanted to resume the game—again, almost exclusively to dogs able to see the signal. In other words, they communicated intentionally, to an audience able to see them.

Even better, in many cases the record of where the dogs were looking revealed that a dog who had paused play was distracted—looking elsewhere, playing with someone else. One option for his erstwhile partner would be to play-bow madly, hoping to lure someone over to play. But more mindful would be just what they did: used an attention-getter before doing a bow. Importantly, they used attention-getters that matched the level of inattention of their playmates, showing they understood something about "attention." Even in the middle of play, they used mild attention-getters—such as an in-your-face or an exaggerated retreat, leaping backward while looking at the other dog—when their partner's attention was only mildly diverted. If a dog's desired playmate was just standing there staring at him, these attention-getters might indeed be enough to rouse him, as a wave hello? in front of a daydreaming friend. But when the other dog was very distracted, looking away or even playing with another dog, they used assertive attention-getters—bites, bumps, and barks. In these cases, that mild hello? would not do. Instead of using a brute-force method of trying to get attention by any means necessary, they chose types of attention-getters that were just sufficient, but not superfluous, to get the desired attention. This was truly sensitive behavior on the part of the players.

Only after these attention-getters were successful did the dogs signal their interest in playing. In other words, they were using an order of operations: get attention first, then send an invitation to rumble.

This is just what good theorists-of-mind do: think about their audience's state of attention and only talk to those who can hear and understand them. The dogs' behavior looks tantalizingly close to a display of theory of mind. But there's reason to believe that their ability is different than ours. For one thing, in both the experiments and my play study, not all dogs acted equally mindfully. Some dogs are oblivious in their attention-getting. They bark, get no response—and then bark and bark and bark and bark. Others use attention-getters when attention has already been gotten, or play signals when play has already been signaled. The statistics show that most dogs act mindfully, but there are plenty of exceptions. We can't tell yet whether they are just the underperformers or whether they indicate that the species has an incomplete understanding.

It may be a little of both. Rather than contemplating the mind behind the dog, most dogs are likely to simply interact. Their skill at using attention and play signals hints that they may have a rudimentary theory of mind: knowing that there is some mediating element between other dogs and their actions. A rudimentary theory of mind is like having passable social skills. It helps you play better with others to think about their perspective. And however simple this skill may be, it may be part of an inchoate system of fairness among dogs. Perspective-taking underlies our agreement to a code of conduct between humans that is jointly beneficial. Watching play, I noticed that dogs who violated the implicit rules for attention-getting and play-signaling—simply barging in on others' play without following the proper, mindful procedures, say—were shunned as playmates.*

Does this mean that your dog is aware of and interested in what's on your mind right now? No. Does it mean that he might realize that your behavior reflects what's on your mind? Yes. Used to communicate with us, this is a large part of dogs' seeming humanity. Sometimes it is even used in nefarious ways only too human.

WHAT HAPPENED TO THE CHIHUAHUA

We can now revisit the wolfhound and Chihuahua we met at the start of this book. Their hillside encounter is no less remarkable now, but it does perfectly encapsulate the flexibility and variety of behaviors of the species. The explanation for that play begins in the history of their social ancestors, the wolves; it is apparent in the hours of socializing between humans and dogs; in the years of domestication; in the dialogues of speech and behavior between us. It is explicable in the sensorium of the dog: the information he gets from his nose, what his eyes take in. It is in the capacity of dogs to reflect on themselves; it is explained in their different, parallel universe.

And it is in the particular signals they use with each other. The wolfhound's high-rumped approach: the play bow, an invitation to a game—making perfectly clear his ardent intent to play with, not eat, the little dog. In return, the Chihuahua bowed: accepting the offer. In the language of dogs that is enough to see each other as equals in play. Their disparate sizes aren't irrelevant—and this explains the hound's drop to the ground: he handicaps himself. By putting himself at the little dog's height—taking the Chihuahua's point of view—and exposing himself to her attacks, he levels the playing field.

They endure jostling body on body. Bodies in full contact is a reasonable social distance for dogs. They bite with impunity: every bite is matched, or explained with a play signal—and every bite is restrained. When the hound hits the little dog too hard, sending her scurrying backward, she could for a moment be seen as small, fleeing prey. But the difference between dogs and wolves is that dogs can put aside their predatory instincts. Instead the hound takes back that swipe with an apologetic play slap, a milder version of the bow. It works: she rushes right back into his face.

Finally, when the hound is pulled up and away by his owner, the Chihuahua tosses a bark to her departing playmate. Had we kept watching them, had he turned around, we might have seen her open her mouth or leap a tiny leap—calling out in the hopes of continuing the game with her giant friend.

NON-HUMAN

The study of dogs' cognitive abilities emerged from a context of comparative psychology, which by definition aims to compare animals' abilities with those of humans. The exercise often winds up splitting hairs: they communicate—but not with all the elements of human language; they learn, imitate, and deceive—but not in the way that we do. The more we learn of animals' abilities, the finer we have to split the hair to maintain a dividing line between humans and animals. Still, it is interesting to note that we seem to be the only species spending any time studying other species—or, at least, reading or writing books about them. It is not necessarily to the dogs' discredit that they do not.

What is revealing is how dogs perform on tasks that measure social abilities we thought only human beings had. The results, whether serving to show how alike or unalike dogs are to or from us, have relevance in our relationships with our dogs. When considering what we ask of them and what we should expect from them, understanding their differences from us will serve us well. Science's effort to find distinctions illustrates more than anything else the one true distinction: our drive to affirm our superiority—to make comparisons and judge differences. Dogs, noble minds, do not do this. Thank goodness.

Inside of the Dog

Her personality is unmistakable and omnipresent: in her reluctance to climb the steep steps out of the park—but then forging ahead of me strongly and gamely; in her great spasms of running and scent rolling of younger days; in her delight at my return from a long trip—but not dwelling on it; in her checking back for me on our walks but also always keeping a few paces apart. For a dog who is in fact wholly dependent on me, she is incredibly independent: her personality is forged not just in interaction with me, but in the times wandering outside without me, in exploring her space alone. She has her own pace of life.

Despite the wealth of scientific information about the dog—about how they see, smell, hear, look, learn—there are places science doesn't travel. It perplexes me that some of the questions I have most often been asked about dogs, and that I have about my own dog, are not addressed by research. On matters of personality, personal experience, emotions, and simply what they think about, science is quiet. Still, the accumulation of data about dogs provides a good foothold from which to extrapolate and reach toward answers to those questions.

The questions are typically of two kinds: What does the dog know? and What is it like to be a dog? So first we will ask what dogs know about things of human concern. Then we can further imagine the experiences—the umwelten—of the creatures who have this knowledge.

I

WHAT A DOG KNOWS

Claims about what dogs know are made constantly. Oddly, they tend to cluster around the academic and the ridiculous. The former prompts researchers to ask if a dog knows how, for instance, to count sums. In one experiment the dogs looked longer—evincing surprise—when there were either more or fewer biscuits revealed behind a screen than they had been shown being hidden there one by one—indicating that they were keeping track of number and noticing when there was a discrepancy. Ta-da: counting dogs.

The other kind of claims are the far-flung: that dogs have ethics, rationality, a metaphysics. I admit to entertaining the notion more than once that my own dog seems to act ironically (whether or not she intends to).* One ancient philosopher maintained that dogs understand disjunctive syllogisms. As evidence, he gave the observation that in tracking an animal to a branching path, dogs can deduce that if the animal is neither down the first nor second of three trails, they realize, even without scent, that it must be down the third.*

Starting with an interest in math or metaphysics and working downward does not get us very far in understanding dogs. But start with their snuffling approach of the world, their striking attention to humans, and knowledge of the various means by which dogs learn about the world—and we might be able to learn what they know. In particular, we might approach an answer to whether they experience life as we do: whether they think about the world as we do. We mind our own autobiographical journeys through life, managing daily affairs, plotting future revolutions, fearing death, and trying to do good. What do dogs know about time, about themselves, about right and wrong, about emergencies, emotions, and death? By defining and deconstructing these notions—making them scientifically examinable—we can begin to answer.

Dog days (About time)

Back home, Pump gives me a perfunctory greeting, executes an unlikely pirouette, and then races off. Over the course of the day she has located all the biscuits I left around the house for her, and has waited until now to consume them, gobbling from the one balanced on the chair's edge to the one on the doorknob to the tricky one on a towering pile of books, which she delicately plucks off and spirits away.

Animals exist in time, they use time; but do they experience time? Surely they do. At some level there is no difference between existing in time and experiencing time: time must be perceived to be used. What many people mean, I suspect, in asking whether animals experience time is, Do animals have the same feelings about time that we do? Can a dog sense the passage of a day? And, critically, are dogs bored all day, at home alone?

Dogs have plenty of experience of the Day, if no word day to call it. We are the first source of their knowledge of days: we organize the dog's day in parallel with ours, providing landmarks and surrounding them with ritual. For instance, we provide all sorts of cues about when the dog's mealtime is. We head for the kitchen or pantry. It may be our mealtime, too, so we begin to unload the refrigerator, wafting food smells about, and making a racket with pots and plates. If we glance at the dog and coo a little, any remaining ambiguity is erased. And dogs are naturally habitual, sensitive to activities that recur. They form preferences—places to eat, to sleep, to safely pee—and notice preferences of yours.

But in addition to all those visible and olfactory cues, does the dog naturally know that it is dinnertime? I know owners who insist they can set the clock by their dog. When he moves to the door, it's precisely the time to go out; when he moves to the kitchen, sure enough, it's time to be fed. Imagine removing all the cues the dog has about the time of day: all of your movements, any environmental sounds, even light and dark. The dog still knows when it's time to eat.

The first explanation is that dogs wear an actual clock—though internally. It is in the so-called pacemaker of their brain, which regulates the activities of other cells of the body through the day. For a few decades neuroscientists have known that circadian rhythms, the sleep and alertness cycles that we experience every day, are controlled by a part of the brain in the hypothalamus called the SCN (suprachiasmatic nucleus). Not only humans have an SCN: so do rats, pigeons, dogs—every animal, including insects, with a complex nervous system. These neurons and others in the hypothalamus work together to coordinate daily wakefulness, hunger, and sleep.* Deprived entirely of cycles of light and dark, we would all still go through circadian cycles; without the sun it takes just over twenty-four hours to complete a biological day.

This morning I heard her barking in her sleep—the muffled, jowl-puffing bark of dreaming. Oh, does she dream. I love her dream-barks, falsely severe, often accompanied by twitching feet or lips curled into a teeth-baring growl. Watch long enough and I'll see her eyes dancing, the periodic clenches of her jaw, hear her tiny whimpers. The best dreams inspire tail-wags—huge thumps of delight that wake herself and me.

We humans experience the day according to our ideas about what typically or ideally will happen throughout it—what meals, work, play, conversation, sex, commuting, naps—and also according to the cycle of our circadian rhythms. Given our attention to the former, though, we sometimes hardly notice that our bodies are charting a regular course through the day. That midafternoon sleepiness, the difficulty in rising at five in the morning—both are due to our activities clashing with our circadian rhythms. Take away some of those human expectations and you've got the dog's experience: the bodily feelings of the passage of the day. In fact, without the societal expectations to distract them, they may be more attuned to the rhythms of their body telling them when to rise and when to eat. As per their pacemaker, they are most active as dark gives way to dawn, and markedly reduce their activity in the afternoon, with a burst of energy in the evening. With nothing else to do—no papers to shuffle, no meetings to attend—dogs nap straight through that afternoon slowdown.

Even without regular mealtimes the body goes through feeding-related cycles. Right before it is time to eat, animals tend to be more active—running about, licking, salivating—in anticipation of food. We see this food-sense when a dog pursues us relentlessly with panting mouth and appealing eyes. Eventually we figure out it is time to feed the dog.

So in fact one can set the clock by the dog's belly. And, even more impressive, dogs maintain a clock operated by other mechanisms not yet fully understood, which seem to read the day's air. Our local environment—the air in the room we are in—indicates (if we have the right indicator) where we are in the day. Although we do not typically sense it, it is just the sort of thing a dog might notice. If we attend carefully, we might notice the gross changes of the day: the cool at the moment the sun sets, or the time of day registered in the amount of light streaming in the window—but the day's changes are infinitely more subtle than this. With sensitive machinery, researchers can detect the gentle air currents that form as a summer's day ends: warmed air pulled up along the inner walls creeps across the ceiling, spilling into the center of the room and falling along the outer walls. This is no breeze, nor even a noticeable puff or waft. Yet the sensitive machinery that is the dog evidently detects this slow, inevitable flow of air, perhaps with the help of their whiskers, well positioned to register the direction of any scent on the air. We know they can detect it because they can also be fooled: brought into a room that was warmed, a dog trained to follow a scent trail may search first by the windows when the track is really closer to the room's interior.

She is patient. How she waits for me. She waits as I duck into the local grocery store: looking plaintively, then settling down. She waits at home, warming the bed, the chair, the spot by the door, for me to return. She waits for me to finish up what I'm doing before we go outside; for me to finish talking with someone during our walk; for me to figure out when she is hungry. She waited for me to finally realize where she liked to be rubbed. And for me to finally begin to figure her out. Thanks for waiting, kiddo.

Dogs have not been tested on their ability to detect a specific length of time; but bumblebees have. In one study, bees were trained to wait for a fixed time interval before sticking a proboscis through a tiny hole for a bit of sugar. Whatever the interval, they learned to restrain themselves for just that long … and then no longer. When you're a bee waiting for sugar water, a half minute is a long time to wait. But they patiently tapped their many feet and did so. Other well-experimented-on animals—rats and pigeons—do the same: measuring time.

It is probable that your dog knows just how long a day is. But if so, a horrible thought occurs: Mustn't dogs be terribly bored enduring that day all alone at home? How can we tell if a dog is bored? Like other concepts whose applicability to dogs we are curious about, we first need to get a handle on what boredom looks like. Any child will tell you when he is bored, but dogs don't—at least, not verbally.

Boredom is rarely discussed in the non-human scientific literature, because it is one of the classes of words whose application to animals is thought suspect. "Man is the only animal who can be bored," the social psychologist Erich Fromm declared; dogs should be so lucky. Human boredom is rarely the subject of scientific scrutiny, either, perhaps because it is seen simply as a part of the experience of life, not as a pathology to scrutinize. Its very familiarity gives us a way to define it: we experience it as a profound ennui, as an utter lack of interest. And we can recognize it in others: in their flagging energy, in an uptick in repetitive movements and a decline in all other activities, and in rapidly waning attention.

With this definition, the subjective becomes objectively identifiable, in dogs as well as humans. Flagging energy and reduced activity are simple to recognize: less moving and more lying and sitting. Attention may wane straight into protracted bouts of sleep. Repetitive movements include stereotyped (aimlessly and endlessly repeated) or self-directed behaviors. We twiddle our thumbs when bored; we pace. Animals kept in barren zoo enclosures often pace madly—and, thumbless, have twiddle-equivalents: licking or chewing skin or fur obsessively and constantly, pulling out their own feathers, rubbing their ears or face, rocking back and forth.

So is your dog bored? If you return home to find apparently restless socks, shoes, or underwear that have magically migrated some small distance from where you left them, or straggled bite-sized reminders of what you threw in the garbage yesterday—the answer is both Yes, your dog was bored, and No, at least not during one manic hour of chewing. Imagine a child complaining, There's nothing to do: that is just the case for most dogs left alone. Left without anything to do, they will find something. Your solution, for the sake of your dog's mental health, and for the sake of your socks, is as simple as leaving something for them to do.

Even if you return to find the house a bit unkempt, a warm depression on the forbidden couch cushion, what is also reliable is that the dog is still alive and usually looks well. We get away with leaving them, with boring them, because they generally adapt to their situations without much complaint. In fact, dogs take comfort in habit, in reliable occurrences, just as we might. If so, then their boredom may be tempered by resignation to the familiar. And they may even know how long they typically need to stay in the suspended animation of waiting at home for you. It is one reason why your dog may be waggily waiting at the door even when you try to quietly sneak in at the workday's end. And it is why I leave more treats hidden around the apartment the longer I will be gone. I'm telling Pump I'll be away—and leaving something to mind the time.

The inner dog (About themselves)

The best scientific tool proposed to determine if dogs think about themselves—if they have a sense of self—is a simple one: the mirror. One day the primatologist Gordon Gallup pondered his reflection while shaving and wondered if the chimpanzees he studied would ponder their reflections in mirrors, too. Certainly using a mirror for self-examination—smoothing a shirt over belly, patting down a wayward hair, testing a coy smile—is a display of our own self-awareness. And before we are self-aware, as young children, we do not use mirrors as adults do. A short time before children pass theory-of-mind tests, they begin to consider their mirror images.

Gallup promptly placed a full-length mirror outside his chimpanzees' cages and watched what they did. They all did the same thing first: they threatened and tried to attack the mirror. Suddenly, it seemed, there was another chimp right outside their cage; this must be addressed at once. Despite the no doubt confusing result—the mirror image seemed to attack back, only for the affair to resolve without ado—their first days with the mirrors were full with social displays toward this new, glaring chimpanzee. After a few days, though, the chimpanzees seemed to come to a realization. Gallup watched as his chimps approached the mirrors and began to use them to examine their own visages and bodies: picking at their teeth, blowing bubbles, making faces toward their mirror image. They were especially interested in parts of their bodies that are ordinarily visually inaccessible: the mouth, the rump, up the nostrils. To be sure that they were thinking about the mirror images as themselves, Gallup devised a "mark" test: he inconspicuously applied a prominent dab of red ink to the head of the chimps. These first subjects in this test needed to be anesthetized to apply the mark; later researchers would affix the mark while doing ordinary grooming or medical care of their animals. When the marked chimps again stood in front of the mirrors, they saw a red-tagged chimp—and they touched the spot on their own heads, bringing their hands down to examine the ink with their mouths. They passed the test.

There is considerable debate about whether this indicates that chimpanzees are thinking about themselves, have a concept of self, recognize themselves, are self-aware, or none of the above*

—especially since it would be disruptive of our ideas about animals to suddenly grant them self-awareness. But the mirror tests have continued alongside the debate, and to this date dolphins (by moving their bodies to explore the mark) and at least one elephant (using her trunk) have passed the test; monkeys have not. And dogs? Dogs have not been shown to pass the test. They never examine themselves in the mirror. Instead they behave more like monkeys do: they sometimes look at the mirrors as though it were another animal, and sometimes look at it idly. In some cases, dogs will use mirrors to get information about the world: to see you tiptoeing up behind them, for instance. But they don't seem to see the mirror as an image of themselves.

There are a few explanations why dogs might behave this way. The dogs may indeed not have any sense of self—thus no sense of who that handsome dog in the mirror might be. But as the debate over this test indicates, it is not universally accepted as a conclusive test of self-awareness; thus neither can it be a conclusive determination of lack of self-awareness. Another possible explanation for the dogs' behavior is that the lack of other cues—specifically olfactory cues—coming from the mirror image leads dogs to lose interest in investigating it. Some fantastical odor-mirror that wafts the dog's own scent while reflecting the dog's own image would be a better medium for this test. Another issue is that the test is predicated on a specific kind of curiosity about oneself: one that leads humans to examine what is new on our own bodies. Dogs may be less interested in what is visually new than what is tactually new: they notice strange sensations and pursue them with nibbling mouth or scratching paw. A dog is not curious why the tip of his black tail is white, or what the color of his new leash is. The mark needs to be noticeable, and also worth noting.

Even so, there are other dog behaviors suggestive of their self-knowledge. In most actions, dogs do not grossly misestimate their abilities. They surprise themselves by jumping into water after ducks—only to find that they are natural swimmers. They surprise us by leaping to scale a fence—which they may in fact be able to clear. On the other hand, one regularly hears that dogs don't know a very basic fact about themselves: how big they are. Small dogs strut up to enormous dogs: their owners proclaim that their dogs "think they're big." Some big-dog owners who endure lap sitting likewise assert that their dogs "think they're small." In both cases, the dogs' accompanying behaviors lend more credibility to the notion that they do know their sizes: the small dog is compensating for his small size by trumpeting his other qualities extra loudly; the large dog raised with a lap to sit on continues with this close contact just as long as he is tolerated, and then finds a large-dog-sized pillow to sit on elsewhere.

Both small and large dog are tacitly acknowledging an understanding of their own size. It might seem unlikely that this means they are thinking about the categories big or small. But look at how they act on objects in the world. Some dogs will attempt to pick up a felled tree, but most dogs with stick-carrying habits will choose similarly sized sticks at every opportunity, as though they have gauged what can be picked up and held in their mouths. From then on, all sticks in the path of a searching dog are quickly assessed: too big? too thick? not thick enough?

Further suggestive evidence that dogs know their size comes from their rough-and-tumble play. One of the most characteristic features of dog play is that socialized dogs can, by and large, play with almost any other socialized dog. This includes the pug who leaps onto the hocks of the mastiff, reaching his knee. As we've seen, big dogs know how to, and often do, moderate the force of their play to smaller playmates. They can withhold their fiercest bites, jump halfheartedly, bump into their more fragile playmates more gently. They might willingly expose themselves to attack. Some of the largest dogs regularly flop themselves on the ground, revealing their bellies for their smaller playmates to maul for a while—what I called a self-takedown. Older, learned dogs adjust their play styles to puppies, who don't yet know the rules of play.

Play between dogs of mismatched statures often does not last long, but it is usually an owner, not a dog, who moves to stop it. Most socialized dogs are considerably better at reading each other's intent and abilities than we are. They settle most misunderstandings before owners even see them. It's not the size or the breed that matters; it's the way they talk to each other.

Working dogs provide another glimpse into what dogs know about themselves. Sheepdogs, raised from their first weeks of life with sheep, do not grow up to act like sheep. They do not bleat or scream, chew their cud, aggressively head-butt, nor suckle from the ewe, as sheep do. Their cohabitation leads dogs to interact socially with sheep—using social behaviors characteristic of dogs. Those who study sheepdogs observe, for instance, that dogs will growl at sheep. Growling is a dog communication: the dog is treating the sheep more like a dog than like a possible meal. These dogs' only fault is to overgeneralize: not only are they clear on their own identity, in some sense—they also think that everyone else is a dog, too. One could call this foible very human: they talk to sheep as though they were dogs, just as we talk to dogs as though they were humans.

Between play bouts, stick-retrieving, and sheepherding, do dogs sit around thinking, My, but I'm a fine medium-sized dog, aren't I? Certainly not: such continued reflection on size or status or appearance is peculiarly human beings' lot. But dogs do act with knowledge of themselves, in contexts where such knowledge is useful. They respect (for the most part) the limits of their physical abilities, and will look pleadingly at you when you ask them to leap a too-high fence. A dog will hop discreetly around a pile of his own defecation encountered on the ground: he recognizes the smell as his. If the dog is reflecting on himself, one might wonder if he thinks about himself in the past—or in the future: if he is quietly writing his autobiography in his head.

Dog years (About their past and future)

As we round the corner Pump stops in her tracks. She moves as if to sniff something a half-step back; I slow to indulge her; and she darts back around the corner. There are still twelve blocks, a brief park, a water fountain, and a right turn until we get there, but she knows this walk. She'd been glancing up at me for blocks, and with that final turn, it's confirmed. We're going to the vet.

Psychologists report that those people with the most prodigious memories—able to flawlessly recite a string of hundreds of random numbers read to them once, as well as identifying every moment the reader blinked, swallowed, or scratched his head—are sometimes the most tortured by what they recall. The complement of remembering so thoroughly can be the strange inability to forget anything at all. Every event, every detail, piles on the garbage heaps that are their memories.

The overflowing garbage, collector of the day's past, is more than a little evocative when considering the memory of a dog. For if anything is on the dog's mind, it is that wonderful, odoriferous pile that we teasingly preserve in our kitchens, off-limits to the dog as a special form of torture. In that pile go the leavings of so many dinners, the extra-rank cheese that was discovered in the back of the fridge, clothes that have smelled too much for too long to be worn. Everything goes there but nothing is organized.

Is the dog's memory like this? At some level, it just might be. There is clear evidence that dogs remember. Your dog plainly recognizes you on your return home. Every owner knows that their dog won't forget where that favored toy was left, or what time dinner is supposed to be delivered. He can forge a shortcut en route to the park; remember the good peeing posts and quiet squatting sites; identify dog friends and foes at a glance and a sniff.

However, the reason we even pose the question "Do dogs remember?" is that there is more to our memory than keeping track of valued items, familiar faces, and places we've been. There is a personal thread running through our memories: the felt experience of one's own past, tinged with the anticipation of one's own future. So the question becomes whether the dog has a subjective experience of his own memories in the way that we do—whether he thinks about the events of his life reflexively, as his events in his life.

Though usually skeptical and reserved in their pronouncements, scientists often implicitly act as though dogs have memories just like ours. Dogs have long been used as models for the study of the human brain. Some of what we know about the diminishing of memory with age comes from tests on the diminishment of the beagle's memory with age. Dogs have a short-term, "working" memory that is assumed to function just as the psychology primers teach that human memory works. Which is to say: At any moment, we are more likely to remember just those things that we bring a "spotlight" of attention to. Not everything that is happening will be remembered. Only those things that we repeat and rehearse for later recollection will get stored as longer-term memories. And if a lot is happening at once, we're bound to remember only some of it—the first and last things sticking best. The dog's memory works the same way.

There is a limitation to the sameness. Language marks the difference. One reason why as adults we don't have many—arguably any—true memories of life before our third birthdays is that we were not skilled language users at the time, able to frame, ponder, and store away our experiences. It might be the case that while we can have physical, bodily memories of events, people, even thoughts and moods, what we mean by "memories" is something facilitated only by the advent of linguistic competence. If that's the case, then dogs, like infants, don't have that kind of memory.

But dogs certainly remember a large amount: they remember their owners, their homes, the place they walk. They remember innumerable other dogs, they know about rain and snow after experiencing them once; they remember where to find a good smell and where to find a good stick. They know when we can't see what they are doing; they remember what made us mad last time they chewed it up; they know when they are allowed on the bed and when they are forbidden from it. They only know these things because they have learned them—and learning is just memory of associations or events over time.

Back, then, to the matter of the autobiographical memory. In many ways, dogs act as if they think about their memories as the personal story of their life. They sometimes act as though they are thinking about their own future. Unless sick or asleep, there was usually nothing that could stop Pump from eating dog biscuits—and yet she often refrained when home alone, opting to wait for my return. Even when accompanied, dogs regularly hide bones and squirrel away other favored treats; a toy may be abandoned outside with seeming insouciance only to be beelined-for the next week. Their actions can often be traced to events of their own past. They remember and avoid ground that was rough underfoot, dogs who turned suddenly gruff, people who acted erratically or cruelly. And they evince familiarity with creatures and objects they encounter repeatedly. Besides their quick recognition of their new owners, young dogs come to know their owners' visitors over time. They play best, and with the least ceremony, with those dogs they have known the longest—as though they are stamped together. These longtime playmates need not use elaborate play signals with each other: they use their own shorthand, signals abbreviated into mere flashes, before fully engaging.

It is somewhat dispiriting to find that our knowledge about a dog's autobiographical sense has not advanced beyond Snoopy's affirmation half a century ago, "Yesterday I was a dog. Today I'm a dog. Tomorrow I'll probably still be a dog." No experimental study has specifically tested the dog's considerations of his own past or future. But a few studies with other animals examine part of what might be considered their autobiographical consciousness. For instance, a test run on the Western scrub-jay, a bird that naturally caches food for later consumption, has shown what in humans would be called willpower. If I'm hankering for chocolate-chip cookies, and someone gives me a bag of chocolate-chip cookies, it is extremely unlikely that I would put them away until the next day. The jays were taught that when given a preferred food—their chocolate-chip cookie equivalent—they would not be given food on the subsequent morning. Despite what we can presume is a strong interest in eating the food straightaway, they saved some and consumed it the next day. And me, without my cookies.

We might ask whether dogs act similarly. If prevented from eating in the mornings, does your dog begin to stash food the night before? If so, that would be suggestive evidence that they can plan for the future. As we know from finding uneaten unidentifiables in refrigerated takeout containers, not all saved food is equally good over time. If your dog buries a bone in the dirt or in the corner of the couch each month for three months, does he remember which is the oldest, the foulest, and which is the freshest? Putting aside any overpowering odors emanating from your couch, it is not likely. If we consider the dog's environment, it is apparent that they simply do not need to use time in this way, as they, unlike scrub-jays, are provided with a regular supply of food. In addition, discriminating food by its expiration date, or saving food for later when you're hungry now, may be a difficult task for an animal descended from opportunistic feeders, who eat as much as they can when food is available, then endure long stretches of fasting when food is not. Some suggest, reasonably, that dogs' bone-burying behavior is tied to an ancestral urge to stash some food aside for the lean times.* Evidence that a dog can distinguish the freshest bone from the one that has rotted—or leaves some aside just to enjoy it later—would bear this out. It is more likely that most of the time dogs are not thinking about time when they are thinking about food. A bone is a bone is a bone, buried or in the mouth.

On the other hand, a dearth of evidence verifying dogs' time-telling with bones does not mean that dogs do not distinguish past from present from future. When encountering a dog who had once—but only once—been aggressive, a dog will first be wary and gradually, with time, grow more emboldened. And dogs certainly anticipate what is in their near future: with growing excitement on beginning the walk that leads to the dog food store; or anxiety at the car ride that suggests a visit to the veterinarian.

Some thinkers treat the dog as having no past: as enviably ahistorical, happy because they cannot remember. But it is clear that they are happy even despite remembering. We don't yet know if there is an "I" there behind the dog's eyes—a sense of self, of being a dog. Perhaps there need only be a continuous teller for the autobiography to be written. In that case, they are writing it right now in front of you.

Good dog (About right and wrong)

When Pump was a young dog, a common scene in our household went like this: I turn my back or go into another room. Milliseconds later, Pumpernickel has her nose at the kitchen trash can, peering in for good bits. If I return and catch her in this vulnerable spot, she immediately pulls her nose out of the can, her ears and tail drop, and she wags excitedly, slinking away. Caught.

When researchers asked a sample of dog owners what kinds of things dogs know or understand about our world, the owners most frequently claimed that dogs know when they have done something wrong: that dogs have knowledge of a kind of category of things one must never, ever do. These days that category includes things like tearing into the garbage, devouring footwear, and snatching just-cooked food off the kitchen counter. The punishment in our enlightened age is, one hopes, not terribly severe: a stern word; a frown and a stamped foot. It was not always so: in the Middle Ages and earlier, dogs and other animals were brutally punished for misdeeds, from the "progressive mutilation" of the ears, feet, and on to the tail of a dog in correspondence with the number of people he had bitten, to the capital punishment, after legal trial and conviction, of a dog for homicide;* to earlier, in Rome, the ritual crucifixion of a dog on every anniversary of the evening the Gauls attacked the capital and a dog failed to warn of their approach.

The guilty look of a dog responsible for lesser trespasses is well-known to anyone who has caught a dog in Pump's pose, with her snout deeply plunged in the trash can, or discovered with bits of stuffing in his mouth and surrounded by tufts of what had until recently been the innards of the couch. Ears pulled back and pressed down against the head, tail wagging in quick time and tucked between the legs, and trying to sneak out of the room, the dog gives every appearance of realizing that he's been caught red-pawed.

The empirical question this raises is not whether this guilty look reliably occurs in such settings: it does. Instead, the question is what it is, exactly, about those settings that prompts the look. It may in fact be guilt—or it may be something else: the excitement of sniffing the trash, a reaction to being discovered, or anticipation of the unhappy, loud noises her owner tends to make when encountering trash out of its can.

Can dogs know right from wrong? Do they know that this particular action is clearly, maddeningly, wrong? A few years back, a Doberman employed to guard an expensive teddy bear collection (including Elvis Presley's favorite bear) was discovered in the morning with the devastation of hundreds of maimed, mauled, and beheaded teddies around him. His look, captured in news photos, was not of a dog who thought he had done wrong.

It would seem to defy reason if the mechanism behind the guilty or the defiant look were the same as ours. After all, right and wrong are concepts that we humans have by virtue of being raised in a culture that has defined such things. Excepting young children and psychotics, every person winds up knowing right from wrong. We grow up in a world of oughts and oughtn'ts, learning some rules for conduct explicitly and others by a kind of observational osmosis.

But consider how we know that other people know right from wrong when they cannot tell us so. A two-year-old sidles up to a table, gropes toward an expensive vase, and knocks it over, shattering it. Does the child know that it is wrong to break things that belong to other people? This might be an occasion on which, given the probable explosive reaction from any adults in the vicinity, she begins to learn. But at age two, she does not yet understand the concepts: she did not maliciously destroy the vase. Instead, she is an ordinary two-year-old who is clumsily trying to master moving her own body. We get an indication of her intent by watching what she did before and after the vase fell. Did she head directly for the vase and act to push it over? Or was she reaching for the vase and was uncoordinated in doing so? After it fell, did she evince surprise? Or did she look, well, satisfied?

Essentially the same method can be applied to dogs by allowing them to break expensive vases and watching how they react. I designed an experiment to determine if those guilty looks come from being guilty or from one of the something-elses. Though my method is experimental, the setting is ordinary, so as to best capture the animals' natural behavior: in the "wild" of their own homes. To qualify for subjecthood, dogs had to have been exposed to an owner's disallowance—for instance, by the owner pointing at an object to be left alone and loudly stating No!—and must know to therefore leave it be.

In the place of expensive vases, I use highly desirable treats—a bit of a biscuit, a cube of cheese—that will not be shattered, but will be expressly forbidden. Given that the claim being tested is that a dog knows that engaging in a behavior that has been disallowed by the owner is wrong, I designed this experiment to provide an opportunity to do that very behavior. In this case, the owner is asked to bring the dog's attention to the treat and then clearly tell the dog not to eat it. The treat is placed in an enticingly available spot. Then the owner leaves the room.

Remaining in the room are the dog, the treat, and a quietly observing video camera. Here's the dog's chance to do the wrong thing. What the dogs do is only the beginning of the data for our experiment. In most cases we assume that if given the opportunity, the dog's first move is to get the treat. We wait until he does. Then the owner returns. Here is the crucial data: How does the dog behave?

Every psychological and biological experiment is designed to control one or more variables, while leaving the rest of the world unchanged. A variable can be anything: ingestion of a drug, exposure to a sound, presentation with a set of words. The idea is simply that if this variable is important, the subject's behavior will be changed when exposed to it. In my experiment, there are two variables: whether the dog eats the treat (the one owners are most interested in) and whether the owner knows whether the dog has eaten it (the one I guessed the dogs are most interested in). Over a handful of trials, I alternate these variables one at a time. First the opportunity to eat the treat is varied: either removing the treat after the owner leaves, providing the dog the treat, or letting the dog stew over it (and eventually disobey). What we tell the owner of the dog's behavior is also varied: in one trial the dog eats the treat, and the owner is informed on return to the room; in another, the dog is surreptitiously given the treat by the videographer, and the owner is misled into thinking that the dog has obeyed the command not to eat.

All the dogs survive the experiment looking well fed and a little bewildered. In many of the trials, the dogs could be models for the guilty look: they lower their gaze, press their ears back, slump their body, and shyly avert their head. Numerous tails beat a rapid rhythm low between their legs. Some raise a paw in appeasement or flick their tongue out nervously. But these guilt-related behaviors did not occur more often in the trials when the dogs had disobeyed than in those when they had obeyed. Instead there were more guilty looks in the trials when the owner scolded the dog, whether the dog had disobeyed or not. Being scolded despite resisting the disallowed treat led to an extra-guilty look.

This indicates that the dog has associated the owner, not the act, with an imminent reprimand. What's happening here? The dog is anticipating punishment around certain objects or when seeing the subtle cues from the owner that indicate he may be angry. As we know, dogs readily learn to notice associations between events. If the appearance of food follows the opening of the large cold box in the kitchen, why, the dog will be alert to the opening of that box. These associations can be forged with events of their making as well as those they observe. Much of what is learned is based, deep down, on making associations: whining is followed by attention, so the dog learns to whine for attention; scratching at the trash can causes it to tip and spill its contents, so the dog learns to scratch to get what's inside. And making certain kinds of messes is sometimes followed much later by the presence of the owner, which is itself quickly followed by the reddening of the owner's face, loud verbiage coming out of the owner, and punishment by that reddened loud owner. The key here is that the mere appearance of the owner around what looks like evidence of destruction can be enough to convince the dog that punishment is imminent. The owner's arrival is much more closely linked to punishment than the garbage emptying the dog engaged in hours earlier. And if that's the case, most dogs will assume a submissive posture on seeing their owners—the classic guilty look.

In this case, a claim about the dog's knowledge of his misdeed is importantly off the mark. The dog may not think of the behavior as bad. The guilty look is very similar to the look of fear and to submissive behaviors. It is no surprise, then, to find so many dog owners who are frustrated with attempts to punish a dog for bad behavior. What the dog clearly knows is to anticipate punishment when the owner appears wearing a look of displeasure. What the dog does not know is that he is guilty. He just knows to look out for you.

A lack of guilt does not mean dogs do nothing wrong. They not only do plenty of human-defined wrong things, they sometimes seem to flaunt these things: a half-chewed shoe is paraded in front of a busy owner; you are greeted by a dog merrily exhausted from rolling in defecation. The teddy-bear guard dog looked nothing if not proud when photographed surrounded by the teddy-bear remains. Dogs do seem to play with the fact of our knowing and not knowing something—to get attention (which it generally does) and perhaps just for the sake of playing with knowledge. This is not unlike a child testing the limits of his understanding of the physical world by sitting on his high chair, dropping a cup to the floor … and again … and again: he is seeing what happens. Dogs do this with different states of attention, knowledge, or alertness of their owners. In this way they come to learn more about what we know, which they can then use to their advantage.

In particular, dogs are quite capable of concealing behavior, acting to deflect attention from their true motives. Given what we know about their understanding of mind, it is entirely within their reach to deceive. And given that it is a rudimentary understanding, their deception is not always very good. This too is childlike, as in the two-year-old child who puts his hands over his eyes to "hide" from a parent: partway to hiding, but not quite getting the essence of "hidden." Dogs show both imaginative insights and inadequacies. They do not work to hide the spoils of an overturned trash can or a messy roll in the grass. But they do act in ways to conceal their true intent. To stretch forward idly next to a dog playing with a treasured toy—only to get close enough to snatch it. To shriek overly dramatically when bitten in play, thereby ending a momentary disadvantage as the playmate stops in shock. These behaviors may begin fortuitously, with accidental actions that turn out to yield happy consequences. Once noticed, they will be produced again and again. It only remains now for an experimenter to provide an opportunity for dogs to intentionally deceive one another—unless they are too clever to let their scheming be revealed.

A dog's age (About emergencies and death)

With age she uses her eyes less; she looks at me less.

With age she would rather stand than walk, lie than stand—and so she lies next to me outside with her head between her legs, nose still alert to the smells on the breeze.

With age she has become more stubborn, insisting on hoisting herself up stairs without help.

With age the difference is amplified between her day mood—reluctant to walk, extra-sniffy—and her evening mood—pulling me out the door, a spring in her step, willing to forsake smells for a jaunty tour around the block.

With age I have been given a gift: the details of Pump's existence have become even more alive. I started seeing the geography of smells she checks up on in the neighborhood; I feel how long are the periods she waits for me; I hear the way she speaks volumes by simply standing; I see her efforts to cooperate when I goad her to trot across the street.

Every dog that you name and bring home will also die. This inescapable, dreadful fact is part of our lot for introducing dogs into our lives. What is less certain is whether our dogs themselves have any inkling of their own mortality. I inspect Pump for any sign that she notices the age of her sniffmates on the sidewalks; notes the disappearance of the old droopy-eared fella with the cloudy eyes from down the block; observes her own slowed and stiff gait, graying fur, and lethargic mood.

It is our grasp of the fragility of our own existence that makes us wary of risky undertakings, cautious for ourselves and those we love. Our mortal knowledge may not be visible in all of our moves, but it shines through in some: we shrink back from the balcony's edge, from the animal with unknown intent; we buckle up for safety; we look both ways before crossing; we don't jump in the tiger cage; we refrain from the third serving of fried ice cream; we even entertain not swimming after eating. If dogs know about death, it might show in how they act.

I would prefer that dogs not know. On the one hand, when I have been confronted with a dying dog, I wanted to be able to explain to her her situation—as though an explanation would be a comfort. On the other, despite many owners' habit of giving explanations to their dogs for every command or event (come ON, I overhear regularly in the park, we've got to go home so Mommy can get to work …), dogs do not seem comforted by explanations. A life untrammeled by knowledge of its end is an enviable life.

There are a few indications that we should not envy them much. One comes from their own balcony aversion: for the most part dogs reflexively withdraw from true danger, be it a high ledge, a rushing river, or an animal with a predatory gleam in its eye. They act to avoid death.

But so does the lowly paramecium, beating a hasty retreat from predators and toxic substances. Avoidance behavior is instinctual, seen in some form in nearly all organisms. Instincts, from the knee jerk to an eye blink, do not require that the animal understands what it is doing. And we are not ready to grant the paramecium an understanding of death. But that reflex is not trivial: a more sophisticated understanding could be bootstrapped onto it.

And here are two ways dogs differ from the paramecium: First, they are not only avoidant of injury, they act differently once injured. They are aware of when they are damaged. Hurt or dying, dogs often make great efforts to move away from their families, canine or human, to settle down and perhaps die someplace safe.

Second, they are attentive to the dangers that others put themselves in. One need not wait long for a story of a heroic dog to pop up in the local news. A child lost in the mountains is kept alive by the warmth of dogs who stayed with him; a man who falls through the ice of a frozen lake is saved by the dog who came to him at the ice's edge; a dog's barking attracts a boy's parents before he can reach into the hole of a poisonous snake. Heroic dogs tales abound. My friend and colleague Marc Bekoff, a biologist who has studied animals for forty years, writes of a blind Labrador retriever named Norman who was roused to action by the screams of the family's children, caught in the current of a raging river: "Joey had managed to reach the shore, but his sister was struggling, making no headway, and in great distress. Norman jumped straight in and swam after Lisa. When he reached her, she grabbed his tail, and together they headed for safety."

The end result of all the dogs' actions is clear: someone was able to avert death for another day. Given that the dogs needed to overcome their own instinct of self-preservation to preserve another self, the usual interpretation is that the dogs are heroic, not inadvertent, actors. An understanding of the dire straits faced by the various humans might seem the only explanation.

But the trouble with anecdotes is that one does not have the full story of what happened, since the teller, with his own umwelt and particular perception, is necessarily restricted in what he sees. One could reasonably ask whether Norman did not as much intend to save Lisa as, say, follow her brother's instruction to swim out to her; or maybe Lisa herself was able to swim to shore on seeing her faithful companion near; or maybe the current shifted and carried her to shore. There is no videotape to rewind and examine to carefully consider what happened here—or in any of the rescues described. Nor do we know the long-term behavior of the dogs. It is one thing if a dog suddenly barks in order to alert others that a boy is imperiled; it is another if that dog is barking all the time, day and night. An understanding of the dogs' life histories is also important to correctly interpret what happened.

Finally, what of all the cases when a dog didn't save the drowning child or the lost hiker. The newspaper headlines never crow, LOST WOMAN DIES AFTER DOG FAILS TO FIND AND DRAG HER TO SAFETY! If the heroic dogs are taken to represent the species, so should the non-heroes be given consideration. There are certainly more unreported non-heroic acts than there is reported heroism.

Both the skeptical and the heroic talk can be displaced by a more powerful explanation, wrought by looking more closely at the dogs' behavior. Scrutiny of these dog stories reveals a recurring element: the dog came toward his owner, or stayed close to the person in distress. The warmth of a dog saves a lost, cold child; a man in a frozen lake can grab on to his dog waiting on the ice. In some cases the dog also created a ruckus: barking, running around, calling attention to himself—and to, say, the venomous snake.

These elements—proximity to the owner, and attention-getting behavior—are by now familiar to us as characteristic of dogs, and go into their being such fine companions for humans.

And in these cases, they were also essential for the survival of the person whose life was at risk. So are the dogs truly heroes? They are. But did they know what they were doing? There is no evidence that they did. And they don't know they're acting heroically. Dogs certainly have the potential, with training, to be rescuers. Even the untrained dog may come to your aid—but without knowing exactly what to do. Their success is due instead to what they do know: that something has happened to you, which makes them anxious. If they express that anxiety in a way that attracts other people—people with an understanding of emergencies—to the scene, or allows you leverage out of a hole in the ice, great.

This conclusion is affirmed by one clever experiment performed by psychologists interested in whether dogs show appropriate behavior when there is an emergency. In this test, owners conspired with the researchers to feign emergencies in the presence of their dogs, in order to see how their dogs responded. In one scenario, owners were trained to fake a heart attack, complete with gasping, a clutch of the chest, and a dramatic collapse. In the second scenario, owners yelped as a bookcase (made of particleboard) descended on them and seemed to pin them to the ground. In both cases, the owners' dogs were present, and the dogs had been introduced to a bystander nearby—perhaps a good person to inform if there has been an emergency.

In these contrived setups, the dogs acted with interest and devotion, but not as though there were an emergency. Dogs frequently approached their owners, and sometimes pawed or nuzzled these seeming victims, now silent and unresponsive (in the heart attack case) or crying out for help (in the bookcase scenario). Other dogs, though, took the opportunity to roam around in the vicinity, wandering and sniffing the grass or the floor of the room. In only a very few cases did a dog vocalize—which might serve to get someone's attention—or approach the bystander who might be able to help. The only dog who touched the bystander was a toy poodle. The poodle leaped into the bystander's lap and settled down for a nap.

In other words, not a single dog did anything that remotely helped their owners out of their predicaments. The conclusion one has to take from this is that dogs simply do not naturally recognize or react to an emergency situation—one that could lead to danger or death.

A killjoy conclusion? Hardly. If dogs lack the concepts emergency and death this is not to their discredit. One might as well ask a dog if he understands bicycles and mousetraps and then censure him for responding with a puzzled tilt of the head. A human child is also naïve to these concepts: an infant has to be screamed at as he zeros in on an open electrical outlet; a two-year-old who saw someone hurt would likely do little but cry. They will be taught to understand emergency situations—and then the concept of death. So too are some dogs trained, for instance, to alert a deaf companion to the sound of an emergency device, such as a smoke alarm. The teaching of children is explicit, with some procedural elements—If you hear this alarm, get Mommy; the dogs' training is entirely reinforced procedure.

What the dogs seem to know is when an unusual situation occurs. They are masters of identifying the usual in the world you share with them. You often act in reliable ways: in your own home, you move from room to room, spending long pauses in armchairs and in front of refrigerators; you talk to them; you talk to other people; you eat, sleep, disappear for long stretches into the bathroom; and so on. The environment is fairly reliable, too: it is neither too hot nor too cold; there is no person in the house apart from the ones who have come in the front door; water is not pooling in the living room; smoke is not drifting in the hallway. From that knowledge of the usual world comes some acknowledgment of the unusual fact of someone's odd behavior when injured, or of the dogs' own inability to act as they customarily can.

More than once Pumpernickel got herself in dire straits (once, trapped on a catwalk heading off a building edge; another time, her leash stuck in the elevator doors as the car began to move). I was amazed at how unfazed she appeared—especially as contrasted with my own alarm. It was never she who got herself out of the fix. I believe that I was more worried about her well-being than she was about mine. Still, much of my well-being hinged on her—not on her knowing how to fix dilemmas, great or small, in my life, but rather on her unremitting cheer and constant companionship.

II

WHAT IT IS LIKE

In our attempt to get inside of a dog, we gather small facts about their sensory capacities and build large inferences upon them. One inference is to the experience of the dog: what it actually feels like to be a dog; what his experience of the world is. This assumes, of course, that the world is like anything to a dog. Perhaps surprisingly, in philosophical and scientific circles there is a bit of debate about this.

Thirty-five years ago, the philosopher Thomas Nagel began a long-running conversation in science and philosophy about the subjective experience of animals when he asked, "What is it like to be a bat?" He chose for his thought experiment an animal whose almost unimaginable way of seeing had only recently been discovered: echolocation, the process of emitting high-frequency shouts and then listening for the sound being reflected back. How long the sound takes to bounce back, and how it is changed, gives the bat a map of where all the objects are in the local environment. To get a rough sense of what this might be like, imagine lying in a dark room at night and wondering if someone is standing at your doorway. Sure, you could resolve the question by turning on the light. Or, bat-like, you could hurdle a tennis ball at the doorway and see if (a) the ball comes back toward you or flies out of the room, and (b) if a grunt is heard at about the time the ball arrives at the threshold. If you're very good, you might also use (c) how far the ball bounces back, to determine if the person is very tubby (in which case the ball loses most of its speed in his belly) or has washboard abs (which will reflect the ball nicely). Bats use (a) and (c), and in lieu of tennis balls they use sound. And they do it constantly and rapidly, as quickly as we open our eyes and take in the visual scene in front of us.

This, appropriately, boggled Nagel's mind. He thought that the bat's vision, and thus the bat's life, are so wildly odd, so imponderable, that it is impossible to know what it is like to be that bat. He assumed that the bat experiences the world, but he believed that that experience is fundamentally subjective: whatever "it is like," it is that way only to that bat.

The trouble with his conclusion has to do with the imaginative leap that we do make every day. Nagel treated an interspecies difference as something wholly unlike an intra species difference. But we are perfectly happy to talk about "what it is like" to be another human being. I do not know the particulars of another person's experience, but I know enough about the feeling of being human myself that I can draw an analogy from my own experience to someone else's. I can imagine what the world is like to him by extrapolating from my own perception and transplanting it with him at its center. The more information I have about that person—physically, his life history, his behavior—the better my drawn analogy will be.

So can we do this with dogs. The more information we have, the better the drawing will be. To this point, we have physical information (about their nervous systems, their sensory systems), historical knowledge (their evolutionary heritage, their developmental path from birth to adults), and a growing corpus of work about their behavior. In sum, we have a sketch of the dog umwelt. The parcel of scientific facts we have collected allows us to take an informed imaginative leap inside of a dog—to see what it is like to be a dog; what the world is like from a dog's point of view.

We have already seen that it is smelly; that it is well peopled with people. On further consideration, we can add: it is close to the ground; it is lickable. It either fits in the mouth or it doesn't. It is in the moment. It is full of details, fleeting, and fast. It is written all over their faces. It is probably nothing like what it is like to be us.

It is close to the ground …

One of the most conspicuous features of the dog is one of the most conspicuously overlooked when contemplating their view of the world: their height. If you think that there is little difference between the world at the height of an average upright human and that at the height of an average upright dog—one to two feet—you are in for a surprise. Even putting aside for a moment the difference in sound and smell close to the ground, being at a different height has profound consequences.

Few dogs are human-height. They are human-knee height. One might even say they are often underfoot. We are magnificently obtuse when it comes to imagining even the simple fact of their being less than half our height. Intellectually we know that dogs are not at our height, yet we set up interactions such that the height difference is a constant problem. We put things "out of reach" of dogs, only to be frustrated by their attempts to get them. Even knowing that dogs like greeting us at eye level, we typically do not bend down. Or, bending down just far enough to allow them to reach our faces with a leap, we may get annoyed when they then leap. Jumping up is the direct result of desiring to get to something one needs to jump up to reach.

Scolded enough for jumping up, dogs happily find there is plenty of interest underfoot. There are, for instance, lots of feet. Smelly feet: the foot is a good source of our signature odors. We tend to sweat pedally when we are mentally taxed: stressed, or concentrating hard. Clumsy feet: sitting, we dangle them, but not with dexterity. They act as single units, with toes only existing as places between which extra odors may be discovered by a roving tongue.

If the foot smells so interesting, of course, then the way we treat them must be awfully frustrating: damned shoes. We cloister our odors. On the other hand, shoes left behind smell just like the person who had been in them, and they have the additional interest of carrying on their soles whatever you squishily stepped in outside. Socks are equally good carriers of our odor, hence the gaping holes that regularly appear in socks left bedside. On examination, each hole has been lovingly poked by the incisors of a dog with a sock in her mouth.

Besides feet, at dog height the world is full of long skirts and trouser legs dancing with every footfall of their wearer. The tight whirling motions the warp of a pant leg presents to a dog's eye must be tantalizing. Between their sensitivity to motion and their investigatory mouths, it is no wonder one can find one's pants being nipped by the dog at the end of your leash.

The world closer to the ground is a more odoriferous one, for smells loiter and fester in the ground, while they distribute and disperse on the air. Sound travels differently along the ground, too: hence birds sing at tree height, while ground dwellers tend to use the earth to communicate mechanically. The vibration of a fan on the floor might perturb a dog nearby; likewise, loud sounds bounce more loudly off the floor into resting dog ears.

The artist Jana Sterbak tried to capture a dog's-eye view by rigging a video camera to a girdle worn by Stanley, her Jack Russell terrier, and recording his perambulations along a frozen river and through Venice, the "city of doges" (pun probably intended). The result is a manic, jumbled rush of sights, the world akilter and the image never calm. At fourteen inches above the ground, Stanley's visual world is a glimpse of his olfactory world: what catches his olfactory interest he pursues in body and sight.

But by suiting up animals with critter-cams we are mostly getting an idea of their vantage on the world, not their entire umwelt. With most if not all wild animals, only by taking such a vantage may we have any information about their world, their day: we can't keep up with a diving penguin as a camera strapped to its back can; only an inconspicuous camera could capture the tunnel building of a naked mole rat underground. To watch Stanley from the vantage of his back is to be surprised at the view. There is the temptation, though, to think that by capturing a picture of Stanley's day we have completed the imaginative exercise. It is but the beginning.

… It is lickable …

She is lying on the ground, head between paws, and notices something potentially interesting or edible a short stretch away on the floor. She pulls her head forward to it, her nose—that beautiful, robust, moist nose—nearly but not quite in the particle. I can see her nostrils working to identify it. She gives a wet snort and brings her mouth to aid in the investigation: by turning her head ever-so-slightly on an angle her tongue reaches the floor. She test-licks it with quick swipes, then straightens up and sets to a more serious posture from which to lick, lick, lick the floor—long strokes with the fullness of her tongue.

Nearly everything is lickable. A spot on the floor, a spot on herself; the hand of a person, the knee of a person, the toes of a person, the face, ears, and eyes of a person; a tree trunk, a bookshelf; the car seat, the sheets; the floor, the walls, the all. Unidentifiables on the ground are especially ripe for tonguing. This is revealing, for licking—bringing molecules into oneself, not merely taking a distant safe stance toward them—is an extremely intimate gesture. Not that dogs mean to be intimate. But to be so directly in contact with the world, intentionally or not, is to define oneself differently with respect to one's environment than humans do: it is to find less of a barrier at the edge of one's own skin or fur from that which surrounds it. No wonder it is not unusual to see a dog duck his head fully into a mud puddle or twist his supine body in exaltation of spirit and the rank earth.

The dog's sense of personal space reflects this intimacy with the environment. All animals have a sense of comfortable social distance, the breaching of which causes clashes and the stretching of which they try to contain. While Americans balk at strangers standing closer than eighteen inches, American dogs' personal space is approximately zero to one inches. Repeating itself on sidewalks across the country this very second is a scene that demonstrates the clash of our senses of personal space: the sight of two dog owners as they stand six feet apart, straining to keep their leashed dogs from touching, while the dogs strain mightily to touch each other. Let them touch! They greet strangers by getting into each other's space, not staying out of it. Let them get into each other's fur, sniff deeply, and mouth each other in greeting. It is not for dogs the safe distance of a handshake.

As we have a limit to the proximity of others we'll endure, we also have a limit to the distance we prefer: a kind of social space. Sitting over five or six feet apart makes for an uncomfortable conversation. Walking on opposite sides of the street, we do not feel we are walking together. Dogs' social space is more elastic. Some dogs happily walk in parallel but at great, owner-distressing distances from their owners; others like to trot at your heels. This extends to their sense of fit with us, resting at home. Dogs have their own version of enjoying the pleasantness of a book that fits closely but not too tightly into a box. Pump wanted to sit so that her body was cupped by the embrace of a small upholstered chair. She would fill the space created by my bent legs when lying on my side in bed. Other dogs position themselves with the length of their backs against the length of a sleeping body. The pleasure of this alone is enough for me to invite a dog onto the bed.

… It either fits in the mouth or it's too big for the mouth …

Of the innumerable objects we see around us, only a very few are salient to the dog. The array of furniture, books, tchotchkes, and miscellany in your home is reduced to a more simple classificatory scheme. The dog defines the world by the ways that he can act on the world. In this scheme, things are grouped by how they are manipulated (chewed, eaten, moved, sat upon, rolled in). A ball, a pen, a teddy bear, and a shoe are equivalent: all are objects that one can get one's mouth around. Likewise, some things—brushes, towels, other dogs—act on them.

The affordances—the typical use, the functional tone—that we see in objects are superseded by dog affordances. A dog is less threatened by a gun than interested in seeing if it fits in his mouth. The range of gestures you make toward your dog is reduced to those that are fearsome, playful, instructive—and those that are meaningless. To a dog, a man raising his hand to hail a cab says the same thing as a man reaching to high-five or one waving goodbye. Rooms have a parallel life in the dog's world, with areas that quietly collect smells (invisible detritus in the crook of the wall and floor), fertile areas from which objects and odors come (closets, windows), and sitting areas where you or your identifying perfume might be found. Outside, they do not so much notice buildings: too big; not able to be acted on; not meaningful. But the building's corner, as well as lampposts and fireplugs, wears a new identity each encounter, with news of other dog passersby.

For humans it is the form or shape of an item that is usually its most salient feature, leading to our recognition of it. Dogs, by contrast, are generally ambivalent about the shape in which, say, their dog biscuits come (it is we who think they should be bone-shaped). Instead, motion, so readily detected by the retinae of dogs, is an intrinsic part of the identity of objects. A running squirrel and an idle squirrel may as well be different squirrels; a skateboarding child and a child holding a skateboard are different children. Moving things are more interesting than still ones—as befits an animal at one time designed to chase moving prey. (Dogs will stalk motionless squirrels and birds, of course, once they have learned that they often spontaneously become squirrels running and birds on the wing.) Rolling quickly on a skateboard, a child is exciting, worth barking at; stop the skateboard, and the motion, and the dog calms.

Given their definition of objects by motion, smell, and mouthability, the most straightforward items—your own hand—may not be straightforward to your dog. A hand patting his head is experienced differently than one pressing continuously on it. Similarly, a glance, even many stolen glances, is different than a stare. A single stimulus—a hand, an eye—can become two things when experienced at different speeds or intensities. Even for humans, a series of still images shuffled fast enough becomes a continuous image: as though changing identity. To the common snail, wary of the world, a slowly tapping stick is risky to walk over; but if the stick is oscillated four times a second, the snail will move into it. Some dogs will endure a pat on the head but not a hand resting there; for others, the reverse is true.

These ways of defining the world can all be seen by watching a dog interact with the world. Dogs entranced by a blank spot on the sidewalk, those whose ears perk at "nothing," those transfixed by an invisibility in the bushes—you are watching them experience their sensory parallel universe. With age the dog will "see" more objects familiar to us, will realize that more things can be mouthed, licked, rubbed against, or rolled in. They also grow to understand that different-seeming objects—the man at the deli, and the deli man on the street—are one and the same. But whatever we think we see, whatever we think just happened in a moment, we are pretty much assured that dogs see and think something different.

… It is full of details …

Part of normal human development is the refinement of sensory sensitivity: specifically, learning to notice less than we are able to. The world is awash in details of color, form, space, sound, texture, smell, but we can't function if we perceive everything at once. So our sensory systems, concerned for our survival, organize to heighten attention to those things that are essential to our existence. The rest of the details are trifles to us, smoothed over, or missed altogether.

But the world still holds those details. The dog senses the world at a different granularity. The dog's sensory ability is sufficiently different to allow him to attend to the parts of the visual world we gloss over; to the elements of a scent we cannot detect; to sounds we have dismissed as irrelevant. Neither does he see or hear everything, but what he notices includes what we do not. With less ability to see a wide range of colors, for instance, dogs have a much greater sensitivity to contrasts in brightness. We might observe this in their reluctance to step into a reflective pool of water, in a fear of entering a dark room.* Their sensitivity to motion alerts them to the deflating balloon wafting gently curbside. Without speech, they are more attuned to the prosody in our sentences, to tension in our voice, to the exuberance of an exclamation point and the vehemence of capital letters. They are alert to sudden contrasts in speaking: a yell, a single word, even a protracted silence.

As with us, the dog's sensory system is attuned to novelty. Our attention focuses on a new odor, a novel sound; dogs, with a wider range of things they smell and hear, can seem to be constantly at attention. The wide-eyed look of a dog trotting down the street is that of someone being bombarded with the new. And, unlike most of us, they are not immediately habituated to the sounds of human culture. As a result, a city can be a explosion of small details writ large in the dog's mind: a cacophony of the everyday that we have learned to ignore. We know what a car door slamming sounds like, and unless listening for just that sound, city dwellers tend to not even hear the symphony of slams playing on the street. For a dog, though, it may be a new sound each time it happens—and one that sometimes, even more interestingly, is followed by a person arriving on the scene.

They pay attention to the slivers of time between our blinks, the complement of what we see. Sometimes these are not invisible things but simply those we would prefer they not pay attention to, like our groins, or the favored squeaking toy we stuff in a pocket, or the forlorn, limping man on the street. We could see those things, too, but we look away. Human habits that we ignore—tapping our fingers, cracking our ankles, coughing politely, shifting our weight—dogs notice. A shuffle in a seat—it may foretell rising! A scootch forward in the chair—surely something is happening! Scratching an itch, shaking your head: the mundane is electric—an unknown signal and a whiff of shampoo. These gestures are not part of a cultural world for dogs as they are for us. Details become more meaningful when they are not swallowed up in the concerns of the everyday.

That very attention that dogs bring to us may cause them to acclimate to these sounds over time, to be inculcated in the human culture. Watch a bookstore dog, who lives out the hours of his day surrounded by people: he has become inured to strangers coming by, standing close while they riffle the pages of a book; to being scratched on the head, to passing smells and ever-present footsteps. Crack your knuckles a dozen times a day and a nearby dog will learn to ignore this habit. By contrast, a dog unaccustomed to human habits is alarmed at every one: the most exciting and frightening thing that could happen to a dog left chained to guard a house is that it actually requires his guarding. Guard dogs may only occasionally see an unknown person walking by, a new smell on the air or new sound, let alone any rampant knuckle-crackers.

We can begin to make up for our human disadvantage in understanding the dog's sensory umwelt by trying to startle our sensory systems. For instance, to escape our bad habits of seeing things roughly in the same colors every day, expose yourself to a room lit by only one color—say a narrow bandwidth of yellow. The colors of objects under such light are washed out: your own hands are drained of their blood-filled vitality; pink dresses turn dully white; face stubble stands out like pepper in a bowl of milk. The familiar is made foreign. But for the yellow glow from above, this is much closer to what it might be like to have a dog's color perception.

… It is in the moment …

Ironically, attention to details may preclude an ability to generalize from the details. Sniffing the trees, the dog does not see the forest. Specificity of place and object is useful when you want to calm your dog on a road trip: you can bring his favored pillow to help calm him. A feared object or person put in a new context can sometimes be reborn as unscary.

That same specificity might indicate that dogs do not think abstractly—about that which is not directly in front of them. The influential analytic philosopher Ludwig Wittgenstein suggested that though a dog can believe that you are on the other side of the door, we cannot sensibly talk of his ruminating on it: believing that you will be there in two days' time. Well, let's eavesdrop on that dog. He has slowly zigzagged through the house since you left. He has run through all the interesting unchewed surfaces in the room. He has visited the armchair, where food was once left unattended long ago, and the couch, where food was spilled last night. He has napped six times, had three visits to the water bowl, lifted his head twice at faraway barks. Now he hears your shuffling approach of the door, quickly confirms by nose that it is you, and remembers that each time he hears and smells you, you appear visually next.

In sum, he believes that you are there. It is nonsense to suggest otherwise. Wittgenstein's doubt is not that dogs have beliefs. They have preferences, make judgments, distinguish, decide, refrain: they think. Wittgenstein's doubt is that before you arrive, your dog is anticipating your arrival: pondering it. It is doubt that dogs have beliefs about things not happening right now.

To live without the abstract is to be consumed by the local: facing each event and object as singular. It is roughly what it means to live in the moment—to live life unburdened by reflection. If it is so, then it would be fair to say that dogs are not reflective. Though they experience the world, they are not also considering their own experiences. While thinking, they are not consulting their own thoughts: thinking about thinking.

Dogs come to learn the cadence of a day. But the nature of a moment—the experience of moments—is different when olfaction is your primary sense. What feels like a moment to us may be a series of moments to an animal with a different sensory world. Even our "moments" are briefer than seconds; they are the duration of a noticeable instant, perhaps the smallest distinguishable time unit, as we normally experience the world. Some suggest that this is measurable: it is an eighteenth of a second, the length of time a visual stimulus has to be presented to us before we consciously acknowledge it. Thus we barely notice a blink of an eye, at a tenth of a second long. By this logic, with a higher flicker-fusion rate, a visual moment is briefer and quicker for dogs. In dog time each moment lasts less long, or, to put it another way, the next moment happens sooner. For dogs, "right now" happens before we know it.

… It is fleeting and fast …

For dogs, perspective, scale, and distance are, after a fashion, in olfaction—but olfaction is fleeting: it exists in a different time scale. Scents don't arrive with the same even regularity as (under normal conditions) light does to our eyes. This means that in their scent-vision they are seeing things at a different rate than we.

Smell tells time. The past is represented by smells that have weakened, or deteriorated, or been covered. Odors are less strong over time, so strength indicates newness; weakness, age. The future is smelled on the breeze that brings air from the place you're headed. By contrast, we visual creatures seem to look mostly in the present. The dogs' olfactory window of what is "present" is larger than our visual one, including not just the scene currently happening, but also a snatch of the just-happened and the up-ahead. The present has a shadow of the past and a ring of the future in it.

In this way, olfaction is also a manipulator of time, for time is changed when represented by a succession of odors. Smells have a lifetime: they move and they expire. For a dog, the world is in flux: it waves and shimmers in front of his nose. And he must keep sniffing—as if we had to repeatedly look at and attend to the world for a constant image to remain on our retinae and in our minds—for the world to be continually apparent to him. This explains so much familiar behavior: your dog's constant sniffing, for one,* and also, perhaps, his seemingly divided attention, which races from sniff to sniff: objects only continue to exist as long as an odor is emitted and he inhales. While we can stand in one place and take in a view of the world, dogs must do much more moving themselves in order to absorb it all. No wonder they seem distracted: their present is constantly moving.

The odor of objects thus holds the data of passing minutes and hours. As they note the hours and days, dogs can note the seasons through smell. We on occasion notice the passing of a season as marked by the smell of blooming flowers, decaying leaves, air about to burst into rain. Mostly, though, we feel or see the seasons: we feel the welcome sun on our winter-paled skin; we glance out the window on a bright spring day and never remark, What a beautiful new smell! Dogs' noses stand in for our sight and skin sense. The air of spring brings odors in every sniff-ful remarkably different from the air of winter: in its moisture or heat; the amount of rotting death or blooming life; in air traveling on breezes or emanating from the earth.

Navigating the world of human time with their expanded window of the present, dogs function a little ahead of us; they are preternaturally sensitive, a shade faster. This accounts for their skill at catching the tossed ball midair and also for some of the ways they seem out of sync with us, some of the ways we can't get them to do what we want. When dogs don't "obey," or have difficulty learning something we want them to, it is often that we are not reading them well: we don't see when their behavior has begun.* They are lunging toward the future a step before us.

… It is written all over their faces …

She has a smile. It's one of the panting faces she puts on. Not every panting face is a smile, but every smile is a panting face. A slight fold in her lip—it would be a dimple on a human face—adds to the smile. Her eyes can be saucers (engaged) or half-open slits (contented). And her eyebrows and eyelashes exclaim.

Dogs are ingenuous. Their bodies do not deceive, even if they sometimes cajole or trick us. Instead the dog's body seems to map straight to his internal state. Their joy when you return home or when you approach them is translated directly through their tails. Their concern is plotted by the lift of an eyebrow. Pump's smile is not an actual grin, but that deep lip retraction that gives a glimpse of teeth is used in a ritualized way, part of a communication with us.

You can tell a lot about a dog by observing how he carries his head. Mood, interest, and attention are writ in capital letters from the altitude of the head, the lay of the ears, and the radiance of the eyes. Think of a dog prancing around in front of other dogs, tail and head high, with a cherished or stolen toy: given dogs' usual way of negotiating around each other, this is a clear, intentional gesture—of something like pride. Young wolves too may cheekily flaunt food in front of older animals. The leader in interaction with the world, the head is usually aimed in the direction the dog is going. If a dog turns his head to the side, it is just momentary—to determine if there is something worth pursuing yonder. This is unlike us, who might turn our heads in contemplation, to strike a pose, or for effect. The dog is refreshingly free of pretense.

What the head doesn't tell of the dog's intent, the tail does. The head and tail are mirrors, conveying the same information in parallel media, the classic antithesis. But they can also be true pushmi-pullyus, differently sensitive at either end. A dog who balks at being sniffed in the face may be fine being examined at the rump, or vice versa. Either the tail or head is telling you what is inside.

I would be more surprised if I were entirely correct about "what it is like" inside of a dog than if I am entirely wrong. To address this question is to begin an exercise in empathy, informed imagination, and perspective-taking more than it is to discover the conclusive account. Nagel suggested that no objective account can ever be made of other species' experiences. The privacy of the dog's personal thoughts is intact. But it is crucial that we try to imagine how he sees the world—that we replace anthropomorphisms with umwelt. And if we look carefully enough, imagine skillfully enough, we may surprise our dogs with how much we get right.

You Had Me at Hello

I walk in the door and waken Pump with my arrival. First, I hear her: the thump-thumping of her tail against the floor; her toenails scratching on the ground as she rises, heavily; the jingle of collar tags as she wriggles a shake down the length of her body and out her tail. Then I see her: her ears press back, her eyes soften; she smiles without smiling. She trots to me, her head slightly down, ears perked and tail swinging. As I reach forward she snuffles a greeting; I snuffle back. Her moist nose just touches me, her whiskers sweep my face. I'm home.

Here's a possible explanation for why dogs were not the subjects of serious scientific inquiry until recently: you don't ask questions when you already know the answers viscerally. The delight of my twice-or thrice-daily reunions with Pumpernickel is matched by their ordinariness. Nothing could seem more natural than these simple interactions: they are wonderful, but it is not a wonder that at once demands scientific scrutiny. I may as well dwell on the nature of my right elbow: it is simply a part of me, all the time, and I don't puzzle over its helpful placement there precisely between my upper arm and my forearm, or ponder what it might be like in the future.

Well, I should reconsider that elbow. For the nature of what in certain circles is termed the "dog-human bond" is exceptional. It is not just any animal awaiting my arrival, and it is not just any dog. It is a very particular kind of animal—a domesticated one—and a particular kind of dog—one with whom I have created a symbiotic relationship. Our interactions enact a dance to which only we know the particular steps. Two things—domestication and development—made the dance possible at all. Domestication sets the stage; the rituals are created together. We are bound together before we know it: it is before reflection or analysis.

The human bond with dogs is animal at its core: animal life has succeeded by individual animals associating with, and eventually bonding with others. Originally animals' connection with each other may only have lasted for one sex-filled instant. But the meeting of anatomy at some point evolved in myriad directions: into long-term pairings centered around raising young; groups of related individuals living together; unions of same-sex, non-mating animals for protection or companionship or both; even alliances between cooperative neighbors. The classic "pair bond" is a description of the association that forms between two mated animals. Bonded animals might be recognized by even a naïve observer: most pair bonds hang out together. They mind and care for each other, and they excitedly greet each other on reuniting.

This kind of behavior may seem unsurprising. After all, we humans spend much time trying to pair-bond, maintaining or discussing our current pair bonds, or trying to extricate ourselves from ill-advised pair bonds gone sour. But from an evolutionary point of view, bonding with others is non-obvious. The goal of our genes is to reproduce themselves: an inherently selfish aim, as sociobiologists observe. Why bother with others at all? The explanation of a selfish gene bothering to mind and greet other gene forms turns out to also be selfish: sexual reproduction increases the chance of helpful mutations. It also behooves the selfish gene to ensure that one's sexual mate is healthy enough to bear and raise the new, infant genes.

Sound far-fetched? A biological mechanism has been discovered that supports pair-bonding. Two hormones, oxytocin and vasopressin (known for their roles in, respectively, reproduction and body-water regulation), are released when interacting with one's partner. These hormones make changes at the neuronal level, in areas of the brain involved in pleasure and reward. The neural change results in a behavioral change: encouraging association with one's mate, because it simply feels good. In the small, mouselike prairie voles that the researchers studied, the vasopressin seems to work on dopamine systems, which results in the male vole being very solicitous of his mate. As a result, prairie voles are monogamous, forming long-lasting pair bonds, in which both parents are involved in raising the wee voles.

But these are intraspecific pair bonds: between members of the same species. What started cross-species bonding, which now results in our living with, sleeping with, and dressing up in sweaters our dogs? Konrad Lorenz was the first to describe it. He gave a description of what he called simply "the bond" in the 1960s, well before the current age of neural science, and before human-pet relationship seminars. In scientific language, he defined the bond as revealed in "behaviour patterns of an objectively demonstrable mutual attachment." In other words, he redefined the bond between animals not by its goal—such as mating—but by the process—such as cohabitating and greeting. The goal could be to mate, but it could also be survival, work, empathy, or pleasure.

This refocus opens the door to considering lots of other, non-mating kinds of pairings as true bonds—between members of the same species, or between two species. Among dogs, working dogs are a classic case. For instance, sheepdogs bond early in life with the intended subject of their work: sheep. In fact, to be effective herders, sheepdogs must bond with sheep in their first few months. They live among the sheep, eat when the sheep eat, and sleep where the sheep sleep. Their brains are in the throes of rapid development at an early age; if they don't meet sheep then, they don't become good shepherds. All wolves and dogs, working or not, have sensitive periods of social development. Early in puppyhood they show a preference for the caregiver, seeking her out and responding to her differently than to others, with a special greeting.* For young animals, it is adaptive to do so.

There's still a big leap, though, between a bond wrought of developmental advantage and one based in companionship. Given that humans neither mate with dogs nor need them to survive, why might we bond?

BONDABLES

The feeling of mutual responsiveness: that each time one of us approached or looked at the other, it changed us—it effected some response. I smiled to see her look or wander over; her tail would thump and I could see the slight muscle movements of the ears and eyes that indicated attention and pleasure.

We don't need to be herded; neither are we born to herd. Nor, as we saw earlier, are we a natural pack. What, then, accounts for our bond with dogs? There are a number of characteristics of dogs that make them good candidates for us to choose to bond with. Dogs are diurnal, ready to be awake when we can take them out and asleep when we can't. Notably, the nocturnal aardvark and badger are rare as pets. Dogs are a good size, with enough variation between breeds to suit different specs: small enough to pick up, big enough to take seriously as an individual. Their body is familiar, with parts that match ours—eyes, belly, legs—and an easy mapping on most of those that don't—their forelimbs to our arms; their mouth or nose to our hands.* (The tail is a disparity, but it is pleasing in its own right.) They move more or less the way we do (if more swiftly): they go forward better than backward; they have a relaxation to their stride and a grace to their run. They are manageable: we can leave them by themselves for long stretches of time; their feeding is not complicated; they are trainable. They try to read us, and they are readable (even if we often misread). They are resilient and they are reliable. And their lifetime is in scale with ours: they will oversee a long arc of our lives, perhaps from childhood to young adulthood. A pet rat might live a year—too brief; the gray parrot sixty—too long; dogs hit a middle ground.

Finally, they are compellingly cute. And by compelling, I mean a literal compulsion: it is part of our constitution that we coo over puppies, that we soften at the sight of a big-headed, small-limbed mutt, that we go ga-ga for a pug nose and a furry tail. It has been suggested that humans are adapted to be attracted to creatures with exaggerated features—the prime examples of which are human infants. Infants come with comically distorted versions of adult parts: enormous heads; pudgy, foreshortened limbs; teeny fingers and toes. We presumably evolved to feel an instinctual interest in, and drive to help, infants: without an older human's assistance, no infant would survive on its own. They are adorably helpless. Thus those non-human animals with neotenized (infantlike) features may prompt our attention and care because these are features of human juveniles. Dogs accidentally fit the bill. Their cuteness is half fur and half neoteny, which they have in spades: heads overly large for their bodies; ears all out of proportion with the size of the heads they are attached to; full, saucer eyes; noses undersized or oversized, never nose-sized.

All these features are relevant in attracting us to dogs, but they don't fully explain why we bond. The bond is formed over time—not just on looks, but on how we interact together. At its most general, the explanation may simply be that, as one of Woody Allen's characters says, we need the eggs. He describes his own crazy pair-bonding attempts with a joke about his brother, a fellow so off that he thinks he's a chicken. Sure, the family could send him to be fixed of this delusion, but they're too happy with the protein-rich spoils of his mental disease. In other words, the answer is a non-answer: it's simply in our nature to bond.* Dogs, who evolved among us, are the same way.

At a more scientific level, the question of how bonding came to be in the nature of dogs and humans is answerable in two ways: with explanations that in ethology are called "proximate" and "ultimate." An ultimate explanation is an evolutionary one: why a behavior like bonding to others evolved to begin with. The best answer here is that both we and dogs (and dogs' forebears) are social animals, and we are social because it turned out to confer an advantage. For instance, one popular theory is that human sociality allowed for the distribution of roles that enabled them to hunt more effectively. Thus our ancestors' success at hunting made it possible for them to survive and thrive, while those poor Neanderthals who stuck it out on their own did not. For wolves, too, staying in social family groups allows for cooperative hunting of large game, for the convenience of a mating partner, and for assistance in rearing the pups.

We might be social with any other social animal; but we do not, notably, bond with meerkats, ants, or beavers. To explain our particular choice of dogs, we must look one step more immediate. A proximate explanation is a local one: what immediate effect the behavior has that reinforces it, or rewards the "behaver." For an animal, reinforcement could be the meal that follows a hunt or the copulation that follows an ardorous, energetic pursuit.

It is here that dogs distinguish themselves from the other social animals. There are three essential behavioral means by which we maintain, and feel rewarded by, bonding with dogs. The first is contact: the touch of an animal goes far beyond the mere stimulation of nerves in the skin. The second is a greeting ritual: this celebration of encountering one another serves as recognition and acknowledgment. The third is timing: the pace of our interactions with each other is part of what can make them succeed or fail. Together, they combine to bond us irrevocably.

TOUCHING ANIMALS

Neither of us is truly comfortable but neither of us moves. He is on my lap, sprawled across my thighs, his legs already a little long and dangling down the side of the chair. He's settled his chin on my right arm, right in the crook of my elbow, his head tilted sharply upward just to keep in contact with me. To type, I must strain to pull my trapped arm up and just over the desktop onto the keyboard, with only my fingers able to move freely, and my body leaning precariously. We're both working to hold on to each other, to keep that gossamer of contact that says we are going to intertwine our fates—or they are already intertwined.

We named him Finnegan. We found him at a local shelter, in a cage among dozens of cages, in a room among a dozen rooms, all filled with dogs who we could just as easily have taken home. I remember the moment I knew it would be Finnegan. He leaned. Outside of his cage, on the tabletop where germ-carrying humans were allowed to interact with the sick dogs, he wagged, his ears flopped around his tiny face, he coughed long bursts of coughs, and he leaned against my chest, at table height, his face tucked into my armpit. Well, that was that.

Often it is contact that draws us to animals. Our sense of touch is mechanical, matter on matter: different than our other sensory abilities, and arguably more subjectively determined. The stimulation of a free nerve ending in the skin could be, depending on the context and the force of stimulation, a tickle, a caress, unen-durable, painful, or unnoticed. If we are distracted, what would otherwise feel like a painful burn might be a niggling irritation. A caress might be a grope if it comes from an unwanted hand.

In our current context, though, "touch" or "contact" is simply the erasing of a gap separating bodies. Petting zoos have arisen to satisfy the urge to engage that animal on the other side of the fence not only by looking at it, but by touching it. Better still if the animal is touching back—with, say, a warm tongue or worn teeth grabbing at the food in your outstretched hands. Children and even adults who approach me on the street as I walk with my dog want not to look at the dog, to watch her wag, to meditate on the dog—no, they want to pet the dog: to touch her. In fact, after a cursory rub, many people appear satisfied with that interaction. Even a brief touch is sufficient to bolster the feeling that a connection has been made.

Occasionally one might find one's toes, hanging off the end of the bed bare, being licked.

Dogs and humans share this innate drive for contact. The contact between mother and child is natural: by dint of the requirement for food, the infant is drawn to the mother's breast. Thenceforth, being held by the mother may be naturally comforting. A child who has no caregiver, male or female, will develop abnormally, in ways that it would be inhuman to experimentally test. Inhumane or not, in the 1950s a psychologist named Harry Harlow enacted a series of now notorious experiments designed to test the importance of maternal contact. He took infant rhesus monkeys away from their mothers and raised them in isolation. Some had the choice of two surrogate "mothers" in their enclosures: a wire-framed, monkey-sized doll covered in cloth, plumped with filling, and warmed with a lightbulb; or a bare wire monkey with a bottle full of milk. Harlow's first discovery was that the infant monkeys spent nearly all their time huddled against the cloth mother, dashing over to the wire mother periodically to feed. When exposed to fearful objects (demonic noisemaking robotic contraptions Harlow put in their cages), the monkeys tore for the cloth mothers. They were desperate for contact with a warm body—with just that warm body from which they had been removed.*

The long-term discovery from Harlow's work was that these isolated monkeys developed relatively normally physically, but abnormally socially. They did not interact with other monkeys well: terrified, they huddled in the corner when another young monkey was put into their cage. Social interaction and personal contact is more than desirable: it is necessary for normal development. Months later, Harlow tried to rehabilitate those monkeys whose early isolation so malformed them.

He found that the best remedy was regular contact with young normal monkeys, whom he came to call "therapy monkeys," in play. This restored some of the isolates to more normal social actors.

Watch an infant child, with limited vision and even more limited mobility, try to snuggle into his mother, his head rooting around for contact, and one is seeing just what newborn puppies look like. Blind and deaf at birth, they are born with the instinct to huddle with siblings and their mother, or even with any solid object nearby. The ethologist Michael Fox describes the head of a puppy as a "thermotactile sensory probe," moving in a semicircle until it touches something. This begins a life of social behavior reinforced by and embracing contact. Wolves are estimated to make a move to touch one another at least six times an hour. They lick—each other's fur, genitals, mouths, and wounds. Snouts touch snouts or body or tail; they nuzzle muzzles or fur. They are oriented to touch even in agonistic activity, which, unlike many other species, usually involves contact: pushing, pinning with a bite, biting the body or leg, seizing another's muzzle or head with one's mouth.

Directed toward us, the dog's youthful instinct becomes a drive to burrow a head under our sleeping bodies or to rest a head upon us; to push and bump us as we walk; to gently nibble or lick us dry. It seems no accident that dogs playing at full steam regularly run into any observing owners nearby, using them as living bumpers defining their playing field. In turn, dogs suffer being touched by us. This is to their infinite credit. We find them touchable: furry and soft, right under dangling fingertips and often wearing their neoteny to greatly cute result. The dog's experience of that touch, though, is likely not what we think. A child may rub the belly of a dog fiercely; we reach to pat a dog's head—unknowing whether they want to be either fiercely rubbed or head-patted. In point of fact, their tactile umwelt is almost certainly different than ours.

First, sensation is not uniform across one's body. Our tactile resolution is different at different points on our skin. We can detect two fingers one centimeter apart at the nape of our necks, but if the fingers are moved down the back we feel that they are touching the same spot. The resolution of touch to animals is likely different still: what we think is a gentle pat may be barely detectable or may be painful.

Second, the somatic—body—map of the dog is not the same as our somatic map: the most sensitive or meaningful parts of the body are different on dogs. As seen in many of the aforementioned agonistic contact actions, grabbing a dog's head or muzzle—the first part a guileless dog-petter reaches for—may be viewed as aggressive. It is similar to what a mother will do to an unruly pup, or an older dominant wolf will do to a member of his pack. Here too are the whiskers (vibrissae), which like all hairs have pressure-sensitive receptors at their ends. The whiskered receptors are specially important to detecting motion around the face or nearby air currents. If you are close enough to see the dog's muzzle whiskers, you might notice them flare when the dog feels aggressive (it might be inadvisable to be so close in that case). Pulling a tail is a provocation, but usually one for play, not aggression—unless you don't let go. Touching the underbelly might prompt a dog to feel sexually frisky, as genital licking often precedes an attempt to mount. A dog rolling over on his back is doing much more than simply revealing his belly: this is the same posture dogs use to allow their mothers to clean their genitals. The forceful belly-rubber may find himself urinated upon.

Finally, just as we have highly sensitive areas—the tip of the tongue, our fingers—so too does the dog. There is a species level to this—no person likes being poked in the eye—and an individual level—I might be ticklish on the bottoms of my feet, while you aren't at all. You can easily do a tactile survey and map your own dog's body. Not only are the favored and prohibited places to touch different, but the very form of contact is crucial. In a dog's world repeated touching is different than constant pressure. Since touch is used to communicate a message, holding a hand in one place on a dog's body conveys that same message writ large. At the same time, full-body contact is preferred by some dogs, especially young dogs, and especially when they are the initiators of the contact. Dogs often find places to lie down that maximize contiguity of body with body. This might be a safe posture for dogs, especially as puppies, when they are entirely reliant on others for their care. To feel light pressure along the whole body is to have assurance of your well-being.

It is hard to imagine knowing a dog but not touching him—or being touched by him. To be nudged by a dog's nose is a pleasure unmatched.

AT HELLO

Early in my life with Pumpernickel, I got a full-time job and she got a classic case of separation anxiety. Mornings as I prepared to leave the house after our walk, she began to whimper, shadow me from room to room, and, finally, vomit. I consulted with trainers who gave me very reasonable guidelines to reduce her stress at separation. I followed all known commonsensical procedures, and before too long Pump returned to a healthy mental and physical state. But there was one dictum I didn't follow. Don't ritualize your departure and later return, they advised; don't celebrate your reunion. I refused. Her snuffly, nosed greeting, our heaping together on the floor in a joyous commemoration of togetherness, was too good to let go.

Lorenz called the greeting between animals after being apart a "redirected appeasement ceremony." That nervous excitement one might feel on suddenly seeing someone else in one's den or territory could lead to two different results: an attack of the potential stranger, or a redirection of the excitement into a greeting. His idea was that there is very little difference between the attack and the greeting, besides a few subtle alterations or additions. Between mallards, one of the birds he studied extensively, two individuals meeting each other engage in a rhythmical "ceremonial to-and-fro movement" that could become aggressive, but for the male mallard, the drake, lifting his head and turning it away. This leads to a mutual ceremony of pretending to preen each other, and the greeting is complete: another fight inhibited.

The greeting among humans is similarly ritualized. We look each other in the eyes, wave hands at each other, hug or kiss once or twice or thrice depending on one's native country. These all may be redirections of an uncertain feeling upon seeing someone else. What is more, we may smile or chuckle. Nothing is more reassuring of the good intent of another person than laughter, Lorenz proposed. This paroxysm of noise is surely most often the expression of joy, but it might also be an eruption typical of alarm reformulated as delight or surprise (not unlike the rough play context in which dog laughter appears).

Having channeled one's excitement into a greeting in this Lorenzian way, one might add other components to the hello. Wolves and dogs do. Their greetings, and the greetings of all social canids, are similar. In the wild, when parents return to the den, the pups mob them, madly lunging at their mouths in the hope of getting them to regurgitate a bit of the kill they have consumed. They lick at their lips, muzzle, and mouth, take a submissive posture, and wag furiously.

As we have seen, what many owners cheerfully describe as "kisses" is face licking, your dog's attempt to prompt you to regurgitate. Your dog will never be unhappy if his kisses in fact prompt you to spit up your lunch. This greeting isn't complete without an excited approach and constant, energetic contact. Ears that were pricked to hear your arrival fall flat against the dog's head, which dips slightly in a submissive gesture. The dog pulls his lips back and drops his eyelids: in humans, markers of a true smile. He wags madly or beats a frantic rhythm with the tip of his tail against the ground. Both wags contain all the excited running-around energy that the dog suppresses in order to stay close to you. He may whine or yelp with pleasure. Adult wolves howl daily: among packs, a chorus of howling may help coordinate their travels and strengthens their attachment. Similarly, if you greet the dog with cries and vocal hellos, your dog may cry back at you. In every move he is breathing and exuding his recognition of you.

If greeting and contact were all, we might expect a rash of monkeys bonded with wolves, of rabbits cohabitating with prairie dogs. They all require contact in infancy. And even ants greet homecomers to the nest. I suppose that, predatory issues aside (a big aside), the potential is there. A gorilla named Koko, taught to use sign language to communicate and raised in a human home, had his own pet kitten.

We are relieved of acting instinctually in the way few animals are. But there is one other aspect that makes human-dog bonding unique: timing. We act well together.

THE DANCE

On a long walk Pump stays near me, but not too. If I call her to me, she comes charging forth full-steam and stops just past me. She likes to be one step off. And yet when we walk together on a lean path and she is ahead of me, she checks—regularly looking back to see where I am. She only needs to turn her head partway round to see me, lifting it from its regular downward cast, surveying the ground. If I ever lag, she turns all the way round, ears up and attentive: waiting for me. Oh, I love to come to this beckoning stance of hers: I might gallop a bit as I near her, and this cues her to play-bow, or to pivot on her rear legs and assume her trot leading us on our walk.

He has begun, on this second day, to come to a snap: just picked it up right away. We snap him back and forth between us.

Dogs, though they do not hunt cooperatively, are cooperative. Watch the parade of leashed dog-person twosomes along a city street. Despite small diversions, they are dancing in masterful synchrony, traveling together. Working dogs are trained to heighten their sensitivity to the dance. Blind people and their guide dogs take turns initiating movement, completing each other.

It helps that dogs live at our speed. A house mouse, its heart beating four hundred times a minute at rest, is always in a hurry; a tick can wait for a month, a year, or eighteen years in suspended animation for that odor of butyric acid to come along; dogs function much more at our pace. Though we outlive them, their lives stretch across a generation. And they act at a pace sufficiently close to ours—if slightly quicker—to enable us to discern their movements, imagine their intent. They act in response to our actions, with alacrity. They dance with us.

A puppy initially balks at a leash, pulls at it unyieldingly, or simply fails to grasp that he is tethered to it—and thus to you—as he pulls toward that very interesting newspaper wafting down the sidewalk. In very little time, though, puppies learn to be highly cooperative walking partners, walking at roughly the same rate and often in step with their owners. They match their owners, almost mimicking us. In turn, we unconsciously mimic our mimickers. In ethology, this is called "allelomimetic behavior" and is implicated in the development and maintenance of good social relationships among animals. More than that, though, the puppy has learned about the sequence of behaviors that you repeat, that make up a walk—and anticipates them. Before long, he knows the series of steps to get the walk started, the corners you turn on your route to the park, the place where the leash is snapped off or the ball is brought out. He anticipates the long-walk turnaround point; the short-walk turnaround point; and knows how to evade the latter. Some dogs even seem to know exactly how far the parameters of a leash extend from our hands, and they dart about within those parameters, grabbing a stick or sniffing a passing dog without our breaking stride.

Once we take them off their leashes, the dance continues. My conception of the perfect walk, occasionally achieved, has my dog off-leash running not alongside me but in great circles around me, with our average forward progress over the miles more or less the same. Ideally, we encounter a dozen other dogs. There is little as therapeutic as watching two dogs at play together in a boisterous full-bodied brawl: it extends our pleasure at turn-taking games to high-speed, exuberant result. The rules of play—signaling, timing—are similar to our conversational rules. And so we can enter into a dialogue of play with our dogs.

I start it. I inch to where she's lying and I put my hand on her paw. She pulls it away—and puts her paw on my hand. I place my hand over her paw again; more quickly now, she mimics me. We trade slaps like this until it is too much: I laugh, breaking the spell, and she stretches forward over her paws, mouth open nearly in smile, to lick my face. There's a special intimacy of having her put her hand—its weight, the scratchiness of her pads, the feeling of each claw—on mine. Mostly it's the simple fact of the use of this appendage to communicate with me—it is not seen as a hand independent of its arm until she treats it as one, parallel to mine.

The elements that make play enjoyable are hard to pinpoint, just as a great joke always seems to be funnier than its deconstruction. Try getting a robot to play with you: they always seem to lack a certain … playfulness. A few years ago Sony developed a mechanical pet, "Aibo," designed to look like a dog—it is four-legged, has a tail, characteristic head form, et cetera—and to act something like a dog—it wags, barks, and performs simple trained-dog routines. What the Aibo does not do is play like a dog, and the designers wanted it to be more playfully interactive with people. With this in mind I studied dogs and humans playing together: wrestling, chasing, tossing and retrieving balls and sticks and ropes. I watched, videotaped, and then transcribed all the behaviors that each of the participants did. Then I looked for the elements that were consistent across the successful bouts of this interspecies play.

What I hoped to find were clear routines and games that could be modeled in a doggish toy such as Aibo. What I found was both simpler and more powerful. In every bout, the player's actions were importantly contingent on—based on and related to—the other's actions. This established a rhythm to the play.

Such contingency is easily seen in even very early human social interaction. At two months, infants coordinate simple movements with their mothers, such as mirroring facial expressions. In play, coordinated responses to actions, such as a ball leaving a thrower's hand, happened in as little as five frames of the videotape (approximately one-sixth of a second). Mirrored responses—lunging after being lunged at, for example—are rife during play. The timing is crucial: dogs respond to our movements in the time frame another human might.

A simple game of fetch, for instance, is a dance of call and response. We enjoy the game because of the dog's reactive readiness to respond to our actions. Cats, by contrast, are simply not enjoyable fetch playmates: they may in fact fetch you an object, but in their own time. Dogs participate in a kind of communion with their owners around the ball, with each responding at a conversational pace: in seconds, not hours. The dogs are acting like very cooperative humans. Another game is simply doing an activity in parallel: running together. In play between dogs parallelism is common. Two dogs may mimic each other's gaping mouths yawing back and forth. Often one dog will observe and then match the other's preoccupation: hole digging, stick chewing, ball trumpeting. As wolves hunt together collaboratively, this ability to act with others, matching their behavior, might come from their ancestry. To have your play-slap matched by a dog's is to feel suddenly in communication with another species.

We experience the dog's responsiveness as expressive of a mutual understanding: we're on this walk together; we're playing together. Researchers who have looked at the temporal pattern of interactions with our dogs find that it is similar to the timing patterns among mixed-sex strangers flirting, and to the timing among soccer players as they move down the field that feels like great teamwork. There are hidden sequences of paired behaviors that repeat in interaction: a dog looking at the owner's face before picking up a stick, a person pointing and a dog following the point to what it's directed. The sequences are repeated, and they are reliable, so we begin to get the feeling, over time, that there is a shared covenant of interaction between us. None of the sequences is itself profound, but none is random, and together they have a cumulative result.

Walk down Fifth Avenue in midtown Manhattan around lunchtime on a weekday and you experience the frustration and pleasure of being a member of the human species. The sidewalks are mobbed, jammed with tourists wandering and gawking; office workers rushing to grab lunch or dallying before returning; enterprising street vendors rushing from enforcement officers. It is a formidable sight, one you may not relish joining. On most days, though, you can take any pace you'd like, and just as easily wend your way through the crowds. It has been speculated that people walking en masse don't crash into each other because we are instantly and easily predictable. It only takes a glance to calculate when the oncoming person will reach you. You unconsciously veer subtly right to avoid him; he has done the same with you. It is not unlike (but not quite as completely successful as) the school of fish that abruptly, with one mind, turns tail and goes back from where it came. We are social, and social animals coordinate their actions. What dogs do is cross the species line and coordinate with us. Pick up the leash of any dog in your neighborhood and suddenly you are walking together, like old friends.

The significance of these three elements is corroborated by the kinds of feelings generated when they disappear: of mild betrayal, of momentary severance of the bond. There's a feeling of disconnect when a dog one reaches for ducks her head away, preventing contact. The frustration is immediate when a dog stops cooperating in taking turns in a game: refusing to bring the ball back, not seeing the toss or pursuing a seen toss. A betrayal is felt when the simple communication come! isn't followed by a dog coming. And it would be heartbreaking to approach your dog and to fail to prompt a tail to wag, ears to flatten to the head, or a stomach to be bared for scratching. Dogs whom we perceive as stubborn or disobedient are those dogs who flout these elements. But these elements are natural for both them and for us; a disobedient dog more likely simply does not realize what rules he is being asked to obey.

THE BOND EFFECT

Our bond with dogs is strengthened by contact, by synchrony, and by marking reunions with a greeting ceremony. So too are we strengthened by the bond. Simply petting a dog can reduce an overactive sympathetic nervous system within minutes: a racing heart, high blood pressure, the sweats. Levels of endorphins (hormones that make us feel good) and oxytocin and prolactin (those hormones involved in social attachment) go up when we're with dogs. Cortisol (stress hormone) levels go down. There is good reason to believe that living with a dog provides the social support which correlates with reduced risk for various diseases, from cardiovascular disease to diabetes to pneumonia, and better rates of recovery from those diseases we do get. In many cases, the dog receives nearly the same effect. Human company can lower a dog's cortisol level; petting can calm a racing heart. For both of us, this is a kind of placebo, which is not to say that it isn't real, but that a change is induced in us without a known agent of the change. Bonding with a pet can do the work that long-term use of prescribed drugs or cognitive behavioral therapy do. Of course, it can go wrong, too: separation anxiety is the consequence of a dog feeling so very attached that he cannot stand a moment of detachment.

What are the other results of the bond? We've seen how much they know about us—our smell, our health, our emotions—due not just to their sensory acuity but also to their simple familiarity with us. They come to know how we normally act, smell, and look over the course of our days, and then they are able to notice, many times in ways we cannot, when there is a deviation. The bond effect works because dogs are, at their best, acting as extremely good social interactants. They are responsive, and, crucially, they pay attention to us.

And this connection to us runs deep. A simple experiment consisting of dogs and yawning humans indicates that our link is instinctual—on the level of reflex. Dogs catch our yawns. Just as happens between humans, dog subjects who saw someone yawning themselves began uncontrollably yawning in the next few minutes. Chimpanzees are the only other species we know of for whom yawning is contagious. Spend a few minutes yawning at your own dog (trying not to glare, giggle, or give in to his inevitable complaints) and you can see for yourself this deep-seated connection between human and dog.

Yawning dogs aside, there is a limit to the science here. Science is quite intentionally not looking at the very feature that is most important to dog owners: the feel of the relationship between person and dog. That feel is made up of daily affirmations and gestures, coordinated activities, shared silence. It can be deconstructed somewhat with the dull butter knife of science, but it cannot be reproduced in an experimental setting: it is importantly non-experimental. Experimenters often use what is called a double-blind procedure to assure the validity of their data. The subject is always blind to the point of the experiment, and in a double-blind the experimenter is also blind to which subject's data—one from the experimental group or the control group—he is analyzing. In that way, one avoids inadvertently seeing a subject's behavior as fitting in just a little more tightly with the tested hypothesis.

Dog-human interactions, by contrast, are happily double-seeing. We have the feeling of knowing exactly what the dog is doing; the dog may, too. What we think we see is not the stuff of good science, but it is the stuff of a rewarding interaction.

The bond changes us. Most fundamentally, it nearly instantly makes us someone who can commune with animals—with this animal, this dog. A large component of our attachment to dogs is our enjoyment of being seen by them. They have impressions of us; they see us in their eyes, they smell us. They know about us, and are poignantly and indelibly attached to us. The philosopher Jacques Derrida ruminated on his cat seeing him nude: he was startled and embarrassed. To Derrida, what was startling was that the animal reflected his image back to him. When Derrida saw his cat, what he saw was his cat seeing him, in nakedness.

He was right to implicate our self-regard in our regard of our pets. (As far as I know, though, Derrida never had a dog: his discomfiture might have been greater at the dog's superior gaze.) Of course we revel in the animals themselves. Still, part of what we see when we look at a dog is: the dog looking at us. This is a component of our bond, too. I still imagine my own dog, Pumpernickel, looking at me, seeing herself in my eyes. And I look at her, seeing myself in hers.

The Importance of Mornings

Pump changed my own umwelt. Walking through the world with her, watching her reactions, I began to imagine her experience. My enjoyment of a narrow winding path in a shady forest, lined with low bushes and grasses, comes in part from seeing how Pump enjoyed it: the cool of the shade, of course, but also the pathiness, allowing her to zoom along unchecked, stopping only for rousing scents along the sides.

I now see city blocks, and their sidewalks and buildings, with their investigatory sniffing possibilities in mind: a sidewalk along an uninterrupted wall without fences, trees, or variation, is a block I'd never want to walk down. Where I'll choose to sit in the park—which bench, what rock—is based on where a dog at my side would have the best panoramic olfactory view. Pump loved large open lawns—to plop down in, to roll repeatedly in, to sniff endlessly—and high grass or brush—to lope regally through. I came to love large open lawns and high grass and brush in anticipation of her enjoyment. (The interest in rolling in unseen smells remains elusive …)

I smell the world more. I love to sit outside on a breezy day.

My day is tilted toward morning. The importance of mornings has always been that if I awoke early enough, we could have a long, off-leash walk together in a relatively unpeopled park or beach. I still have trouble sleeping in.

It is a very small bit comforting to realize how deeply she is in me, even over a year from the day when she was also aside me, willing to submit to a tickle of the dense curls under her chin as she rested it on the ground for the last time.

Sitting with a dog on my lap, considering what we know about dogs' abilities, experiences, and perception, I feel partway to full dogness myself. Also, right now, I am covered with dog hair.

Even without getting coated with fur, the knowledge of dog science brings us closer to an understanding of, and appreciation for, dog behavior: how it arises from the ancestral canid, from domestication, from their sensory acuteness, and from their sensitivity to us. With any luck it will get under your skin and you will see the dog from the dog's point of view. Along the way, here is a smattering of ideas of umwelt-ful ways of relating to your dog, of interpreting their behavior, and of considering them in our lives.

GO FOR A "SMELL WALK"

Most of us would agree that we go for walks with dogs for the dog's sake. It is for Pump's sake that I woke early every morning, to catch a permitted off-leash walk in the park; for her sake that I came home during the day to circle the block with her; for her sake that I shod myself before bed and sleepwalked a walk. Yet dog-walks are often not done with the dog's sake in mind, but strangely playing out a very human definition of a walk. We want to make good time; to keep a brisk pace; to get to the post office and back. People yank their dogs along, tugging at leashes to get noses out of smells, pulling past tempting dogs, to get on with the walk.

The dog doesn't care about making good time. Instead, consider the walk your dog wants. Pump and I had a good variety. There were the smell walks, where we made zero progress but she inhaled untold purple, mesmerizing molecules. There were Pump's-choice walks, where I let her choose which way we went at every intersection. There were serpentine walks, where I restrained myself instead of her as she weaved on leash from my left to my right and back again. As a younger dog, she tacitly agreed to go on runs with me when I agreed to occasionally stop and circle around her as she circled around an interesting dog. As she got older, there were even non-walking walks, where she lay down, and just stayed put until she was ready to move on.

TRAIN THOUGHTFULLY

Teach your dog the things you want in a way he can understand: be clear (about what you want him to do), consistent (in what you ask and how you ask it), and tell him when he has got it right (reward him straightaway and often). Good training comes from understanding the mind of a dog—what he perceives and what motivates him.

Avoid the missteps common to those who have the classic idea of what a dog should do: sit, stay, obey. Your dog is not born knowing what you mean by come here. You must teach it explicitly, in small steps, and reward him when he actually comes. Dogs are attuned to tiny cues from you, cues that may be the same when you call come as when you say go away!: a tone of voice, a body posture. It is up to you to make your request specific and distinctive.

Training can take a long time; be patient. When even a "trained" dog does not come to calling, too often people chase him down and then punish him—forgetting that from the dog's point of view, the punishment is linked with your arrival, not his earlier disobedience. This is a quick, effective way to get him to never come when you call him.

When come here has been learned, a good argument can be made that there is little else by way of commands that an ordinary dog needs to know. Teach them more if you both enjoy it. What a dog most needs to learn is the importance of you—and that is something he is born to see. A dog who cannot "shake hands" on command is just a little more doggy. Make clear what behaviors you dislike and be consistent in not reinforcing them. Few celebrate a dog who jumps at people as they

approach—but start with the premise that it is we who keep ourselves (and our faces) unbearably far away, and we can come to a mutual understanding.

ALLOW FOR HIS DOGNESS

Let him roll in whatever-that-thing-is once in a while. Endure some traipsing through mud puddles. Walk off-leash when you can. When you cannot walk off-leash, do not yank him along by his neck, ever. Learn to distinguish a nip from a bite. Let approaching dogs smell each other's rumps.

CONSIDER THE SOURCE

Why does he do that? I am asked almost daily. Many times my only answer can be that not every behavior a dog does has an explanation. Sometimes when a dog suddenly flops on the ground and looks at you, he is just lying down and looking—and nothing more. Not every behavior signifies something. Those that do mean something should be explained by taking into consideration the natural history of your dog—as an animal, as a canid, and as a particular breed.

Breed matters: A dog that stares down invisible prey or slowly stalks other dogs may be presenting very good "eye" behavior for a herder. So too with the dog who is aggrieved when one person leaves the room or who nips at everyone's heels as they wander down the hallway. Freezing at movement in the bushes slows down your walk, but it is very good pointing behavior. A bred dog with no task may be agitated, restive, keyed up: a drifter, not clearly driven to any activity. Give him some. This is the great science behind "tossing a ball": a retriever is made happy just to do it, over and over. He is fulfilling his capability. On the other hand, if your dog has a short nose and has trouble breathing, don't assume he can run with you. That same dog, with his near, central vision, may not care for the game fetch, while a retriever with a wide visual streak may care only for it. Give your dog a context to play out his innate tendencies—and indulge him a little staring at the bushes now and then.

Animalness matters: adapt to your dog's capacities rather than simply expecting him to adapt to our strange notions of how to be a dog. We want our dogs to heel—I have seen people turn furious when their dog does not—but dogs may be more or less prone to walk close to, and in step with, their social companions. Retrievers do, but sporting breeds might not (both will keep an eye on you). Also, most dogs exhibit handedness—pawedness—so while we shunt them to our left, as every training class has us do, we might be disadvantaging some dogs more than others (and leading to inevitable frustration if the good smells are all on the right side of the path). It would be a shame to punish a dog needlessly because we simply do not know his nature. Not every dog needs to heel in the same way: the essence is simply being safe and manageable.

Canidness matters: Your dog is a social creature. Do not leave him alone for most of his life.