Just for fun?

  • An Erratum to this article was published on 30 March 2005

Working out why animals play is no easy task.

The Genesis of Animal Play: Testing the Limits

MIT Press: 2005. 518 pp. $50, £32.95 0262025434 | ISBN: 0-262-02543-4
Dolphins are thought to engage in play-fighting — but is it just for fun or does practice make perfect? Credit: D. PERRINE/NATUREPL.COM

A kitten batting a ball of yarn, kids on a swing, or an adult wielding a fishing-rod — few would disagree that these behaviours can be described as play. Yet in the study of animal behaviour, the phenomenon of play is an anomaly. It is said to be adaptive and yet it involves the expenditure of much energy, often with no apparent pay-off. When a certain behaviour is found to have obvious pay-offs or functions it is, almost by definition, no longer ‘play’ but is defined by its function, such as foraging, predator avoidance or mating.

According to the simplest, most short-term definition, play is ‘just for fun’. But in the long term it may also be practice for a future role, although the ultimate pay-off may only be determined over a lifetime. Consider the activity of batting a ball around. That's play, isn't it? But if someone got paid for it, would it still be play? Even without pay, it may be for practice. And what about others who expend time and resources to watch this senseless activity? It would be difficult to assign adaptive value to these behaviours, or to measure them objectively. Should play then be defined by internal motivation — pleasure, fun? But the motivation of internal pleasure can apply to many complex behaviours, such as hunting, birdsong, sex or chasing a frisbee, although only the latter is likely to be called ‘play’. Is play then only ‘senseless’ behaviour, or is it simply behaviour for which an ultimate function has yet to be discovered?

Play may be plagued by paradoxes and enigmas, but it is a genuine behavioural phenomenon. It is an appropriate subject for enquiry, if only because we know so little about it, despite the interest of scholars who for centuries have tried to define it, fix its boundaries and fathom its functions. For the most part there has been little progress — instead, the subject has become entangled in a web of definitions and controversies. It is clearly time to re-examine play.

The Genesis of Animal Play does not really explore the limits that I allude to above, but to date it is the most comprehensive and illuminating effort to come to terms with this enigmatic topic. Even though Gordon Burkhardt claims his book “is not meant to be a thorough review of play research in animals or people on either a narrow or broad scale”, I believe nevertheless that it does come close. Burkhardt reviews the literature (about 1,300 references are cited) and refers to most of the serious attempts to study play. In his attempt to understand its origins and nature, he incorporates both comparative and multidisciplinary approaches.

In the first part of the book, Burkhardt explores the diversity of play behaviour and the history of various theories, definitions and controversies surrounding it. He also proposes criteria for a modern definition of play. He champions the view of ethology laid down by Niko Tinbergen, which says that any definition of a behaviour must encompass four entirely different types of problem: its causation or mechanisms; its adaptation or function; its development or ontogeny; and its evolution and phylogeny. Burkhardt adds a fifth category: the world of private experience. Researchers using this last criterion would ask whether all play is accompanied by one or a few specific emotions (presumably after play has already been identified, not to identify it).

In the second half of the book, Burkhardt examines the phylogeny of play, reviewing examples from studies of mammals, birds, reptiles, fish and invertebrates. Most of the examples of play in the literature come from a small number of placental mammals, in which play is clear-cut. However, because the focus of the book is the origin and function of play, the most relevant examples are those at the boundaries where play is not readily distinguished from non-play. Even among the mammals and the few birds studied, there is a great variety of play or potential play behaviour, so ecological, social and other potentially relevant factors may shed light even within this narrow taxonomic grouping. At the borders, in fish and invertebrates, descriptions of putative play behaviour remain anecdotal. I suspect that few of these will sustain the five criteria for play that the author sets up, but he remains open-minded to the possibility.

Burkhardt concludes by strongly backing the ‘surplus resources’ theory as a way to predict where and when we might expect to find playful behaviour. This idea is an elaboration of one proposed in 1795 by the German playwright Friedrich Schiller, who wrote: “An animal may be said to be at work when the stimulus to activity is some lack, and it may be said to be at play when the stimulus is sheer plenitude of vitality.” Schiller's idea was further elaborated by Herbert Spencer in 1872. Burkhardt brings it into the social context and adds the adaptive significance — that play not only originates from, but also creates, surplus resources that may be useful on subsequent occasions. I am not sure how or at what point in the life of an animal such surplus resources would manifest themselves or how an ethologist could demonstrate their existence. However, the idea surely applies to the activity of scientific research, and perhaps even to the writing of a book.

Author information



Rights and permissions

Reprints and Permissions

About this article

Cite this article

Heinrich, B. Just for fun?. Nature 434, 273–274 (2005). https://doi.org/10.1038/434273a

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.