Dogs’ play behavior has, over time, been attributed to a variety of factors. It provides puppies practice for adult behaviors. Play with humans is about “dominance.” Play with other dogs is about “dominance” or “status.” It’s a replacement behavior for hunting and killing prey. It’s a way to learn about your environment. Etc.
Turns out they are all wrong. Or partly right. Play can be any of these — depending on which dog is playing and when.
“Why Do Adult Dogs Play?” an article published by John Bradshaw in Behavioural Processes in 2015 (available online in 2014) presents a thorough overview of the research on dog play.
A sentence in the abstract grabbed my attention: “Solitary play with objects appears to be derived from predatory behavior: preferred toys are those that can be dismembered, and a complex habituation-like feedback system inhibits play with objects that are resistant to alteration.” This sentence is all a toy manufacturer needs to know. All those indestructible chew toys are going to (as I well know) collect dust at the bottom of the toy basket as Jana gnaws wood chips into small piles of splinters and works on the shrinking bones and antlers. That sentence also explains the proclivity of many dogs to shred soft toys (though it fails to address why Jana only ever shredded other dogs’ soft toys while lovingly carrying her own around and tending to them gently for years).
The section on solitary play with objects also mentions that while a dog may become bored with a toy, many dogs will perk up when presented with a new toy, even if it is identical to the old toy except for one key detail: “the new toy is not contaminated with its own saliva.” Would that it were true. Yet again, Cali confounds the research. The only tennis ball she wants to play with is the one coated with her saliva, then rolled in mud, then coated again, etc. When the layers get too thick for the Chuck-It to pick up the ball, the ball has a tendency to mysteriously disappear. Only then will Cali deign to play with a newer ball. And the cycle begins anew.
The most interesting section was on social play with objects, though. Through a survey of other research, Bradshaw concludes that there is a big difference in why dogs play with objects with other dogs and why they play with objects with humans. When playing with other dogs, possession of the object seems to be a goal, particularly in keep-away and chase games. But with humans, dogs often act in ways that extend the play, and the dogs show little or no interest in objects that humans are not holding. Those old myths about playing tug with dogs turning your dog aggressive or dominant are bunk. Dogs are likely to surrender an object to a human during play but not to another dog. Even with other dogs, the play tends to stay playful, with fights over objects relatively rare.
While the early attention of researchers to the evolutionary costs and benefits of play behavior is interesting, it is less relevant to dogs than to their wild cousins. Much of the recent research Bradshaw cites points to a motivation for play that has only recently become a subject of study: social connection. It’s possible that our breeding of dogs selected for playfulness, either intentionally or as a side effect of choosing other human-friendly traits. It’s also very likely that play behavior is motivated by social desire (on both the dog’s and the human’s part). It is enjoyable to both and enhances the relationship.
The appeal to humans is obvious; why have a companion animal if you aren’t going to enjoy spending time with him or her? That dogs also enjoy this type of interaction is clear from the appeal to many dogs of play as a training or work reward, but a short section of the article on the social learning aspects of play indicates that dogs might get benefits from playing that go beyond simple fun.