There are few activities that are more joyful than watching dogs at play. Dogs of all sizes, shapes, and ages voluntarily seek it out relentlessly, run around wildly clearly loving what they're doing, wrestle with one another to their hearts content, engage in what seem to be interminable tug-of-wars, and rest a few minutes and go at it again and again. Dog play, consisting of frenetic and kaleidoscopic bouts of behavior consisting of actions that are borrowed from other contexts such as aggression and dominance (biting accompanied by head-shaking, pinning to the ground), predation (stalking and biting accompanied by head-shaking), and reproduction (mounting and humping), is a contagious activity that makes human observers smile. People really want to know more about dog play, and I'm pleased to alert you to a new and beautifully illustrated and well-referenced book by German dog trainer Mechtild Käufer called Canine Play Behavior: The Science of Dogs at Play (for which I wrote one of the two forewords) that provides an excellent up-to-date summary of research on what dogs do, know, think, and feel when they play and why they do it. Ms. Käufer also offers conjectures about what might be happening during play that surely should serve to stimulate more rigorous studies.
Canine Play Behavior consists of seven parts and eleven chapters and covers such topics as what is play, what does play feel like, solitary and social play, play soliciting signals, the role of stress, age, gender, and breed in play, the benefits of play (why it has evolved), and how humans can play with their dogs. Although I've been studying play behavior for many decades this book made me think more about how far we've come in learning about the details of dog play - what we know and don't know - and I rethought some ideas I've been pondering for decades and learned of studies of which I was unaware.
While we know quite a lot about play in dogs and these areas are nicely covered, there are areas that surprisingly still lack rigorous study. This book offers detailed descriptions and explanations of different aspects of dog play that are goldmines for further and more quantitative studies. For example, we don't really know much about how dogs who know one another ask one another to play and what they do when they play and if and how this differs compared to dogs who don't know one another. The author writes about how familiarity might influence play and offers that there are differences such that dogs who know one another engage in self-handicapping (e.g. they don't bite or slam into one another as hard as they could) and role-reversing (e.g. more dominant dogs don't assert dominance and put themselves in compromising positions by rolling over on their back or running away to initiate chase) more than dogs who don't know one another. Future research will tell whether she is right or wrong and inform us of possible subtleties that could generate even additional studies. There really is so much to learn.
We also don't have a good handle on the factors that account for the large amount of variability that has been observed in dog play. For example, how do locale, gender, age, size, familiarity, early experience, individual differences in personality and temperament, and breed influence play patterns? I was recently told that in a study of play in adult dogs that biting accompanied by head-shaking was never observed, however, my students and I have seen it many times among adult dogs. When I asked others about this they, too, were surprised that it was never observed. Why might this be so? Are we talking about the same behavior or are there really these sorts of differences? Just last week I was watching three dogs playing on campus and they were jumping on one another's back and biting and head-shaking rather vigorously. The guy who was with them told me that they play like this all of the time and never once has it escalated into an assertion of dominance. To the untrained eye it looked as if they really were beating the hell out of one another. I've seen interactions like this among young and old animals in dog parks, on campus, and among wild coyotes, wolves, and red foxes.
Ms. Käufer's book also contains very valuable information about, for example, questions that continually arise among people at dog parks and researchers including, "How often does social play escalate into serious aggressive encounters?" Although my students and I haven't kept detailed records on this aspect of play for dogs, we all agree that play didn't turn into serious fighting in more than around two percent of the thousands of play bouts we've observed. Current observations at dog parks around Boulder, CO support our conclusion. And, for the approximately 1,000 play bouts that my students and I observed in wild coyotes, mainly youngsters, on only about five occasions did we see play fighting escalate into serious fighting. Along these lines, Ms. Käufer reports on a study of which I was unaware (Shyan, Fortune, and King 2003) that discovered that fewer than 0.5 percent of play fights in dogs developed into conflict, and only half of these were clearly aggressive encounters. In this case our intuitions were right on the mark.
The reason I mention this is because I often hear people say something like, "When dogs play roughly with one another it always/usually results in a real fight." This is not so. Of course, there may be dogs who simply bite too hard or slam too hard into their play partners when they get highly aroused and lost in play, and this results in an aggressive encounter of varying intensity. But that is the exception rather than the rule, for play fighting only very rarely escalates into real fighting.
Play is a foundation of fairness (see also this) and there is a good deal of cooperation among the players as they negotiate the ongoing interaction so that it remains playful. Perhaps dogs even know what their playmates are thinking and feeling? Do they have a theory of mind? While I think so we still need more data on this aspect of play as well.
Dogma about dog play doesn't work
People are also keenly interested in why dogs mount and hump one another when they play and there is no single answer to this question. A few weeks ago at a dog park in Boulder I was told that mounting and humping always is related to sex, and when I tried to explain that this isn't so the person wouldn't listen to me. Simply put, dogma about dog play doesn't work and this book shows this to be so.
I highly recommend Canine Play Behavior and hope it enjoys a wide readership. Ms. Käufer's book is rich in ideas and hypotheses about how and why dogs play. It also will be very useful to those studying play in other nonhuman animals and perhaps even those studying play in humans. I always feel that when we study animal play we learn a lot about ourselves. While some people might quibble about the author's views about what we know and don't know, it truly is a book that will get readers to think more about the general topic of play behavior, including those who have a lot of experience studying play in dogs and other animals. Not only is it a good and comprehensive review and a fun and easy read, it also covers many areas in which we clearly need more detailed studies and exposes readers to studies about which they were unaware or published in languages they cannot read. I'm sure I will return to it time and time again as I continue my own studies of play.