BOULDER, Colo. -- We tend to think of the mysteries of science as remote from daily life, residing in the movement of distant stars or hidden in the minuscule architecture of our cells. Yet they can take a far more familiar form.
They can gaze at you from the affectionate eyes of your own dog.
One of the questions that most puzzles scientists is whether animals can think -- and if they can, what form their thinking takes.
Marc Bekoff, a biology professor at the University of Colorado in Boulder, has for years been working to discover how dogs, wolves and coyotes think by watching them play.
He is watching his dog, Jethro, playing with a canine friend on the sandy pathway outside his mountain home.
For the average observer, there are simply two dogs on the path, leaping, nipping and wrestling.
But Mr. Bekoff sees something much more complex: an intricate dance of approach and withdrawal, dominance and submission, friendliness and aggression.
The steps by which dogs initiate play are familiar to most dog owners: They bow their front legs, raise their butts, wag their tails, bark excitedly or paw at each other. Sometimes their front paws drum the floor in an odd little dance.
During play, "dogs use behavior patterns from different categories: predatory behaviors, aggressive behaviors, reproductive behaviors," Mr. Bekoff says.
But "somehow they have to tell each other: I don't want to prey on you; I don't want to dominate you; and I don't want to mate with you."
If one dog bites the other too hard, he must find a way of communicating that his intentions remain friendly for play to continue, Mr. Bekoff says.
Speculation on the inner lives of animals is relatively new. The influential 17th-century philosopher Descartes believed that thinking was the sole province of human beings, and that animals were mere automatons.
His view about animals prevailed well into this century, but few reputable scientists would endorse it now.
Rather, debate in the field centers on exactly how animals feel and reason.
William A. Mason of the University of California at Davis, who has experimented with primates, says: "Claims about whether animals feel things are very well founded. Obviously, they do."
But it does not mean they feel the same things humans do, or in the same way -- or that in their mental lives animals are merely humans in different bodies.
Animal thought processes are difficult to explore, since animals communicate only through movement, body language, sounds and odors.
But experiments can be constructed.
"You might have an animal vocalize when they make a discrimination between colors," says Mr. Bekoff. "Or have them do a particular behavior."
He has spent years filming dogs playing and has analyzed the images frame by frame. He sees play shade into aggression when animals feel unsure of their status, not unlike the aggressions a person might show when encountering a stranger.
"It's easy to get a very high-ranking and a low-ranking animal playing. If you're higher ranking than I am and I bite you too hard, you know I'm not trying to usurp you," he says.
But when status is unclear, animals are far more cautious.
"Those data show that the animals are doing some thinking," Mr. Bekoff says. "Play really requires the animals to exchange detailed information about their beliefs and desires and intentions."
He sometimes gets down on all fours to play with Jethro: "I nip him. I bite him on the muzzle. That's how dogs interact."
He has also played with wild wolves and coyotes, suffering no worse consequence over the years than a single bite on the rear when he bent to look into a coyote den. "It was my fault," he says. "I shouldn't have been that close when there were pups."
The fact that there's no general consensus on what thought actually is makes his field more complex, says Mr. Bekoff, whose work has been supported by the National Science Foundation and the National Geographic Society.
"Some people say we think in sentences. If you don't have sentence language, you can't think," he says. "Other people look at thinking as purely information processing and computation. Some people say it has to be visual because we're visual."
He suggests another model -- one based on the nose. "There's no doubt that my dog thinks olfactorily," he says. "I believe my dog has olfactory and auditory representations in his brain.
"So do we."
The idea that animals can feel and think troubles some people because it blurs the boundary between them and human beings.
Over the centuries, there have been innumerable attempts to define that boundary.
"First it was Man the Toolmaker," Mr. Bekoff says, "until they found out that chimpanzees make tools. Then it was language, until they discovered that animals have very sophisticated ways of communicating.
"It's just a debate whether you call the way they communicate language."
A lot of human behavior is reflexive: "We see a red light and we hit the brake, but we don't really think about it" -- or, we are not aware of that thinking.