A paper presented at the 14th annual colloquium of the Board Game Studies Association, Bruges, 2011
When and how did games begin? And how silly is this question? In 1866 the Société Linguistique de Paris reportedly banned all papers on the origins of language as being too speculative to be of any scientific value. They could have said the same of games.
Australopithecus afarensis, Homo habilis, H. erectus,
H. Neanderthalensis, H. sapiens
Nowadays any new game can be seen as the development of a previous game, or as a recombination and refinement of pre-existing game ideas. Keep tracing back this way and you reach a time of the earliest attested games. The oldest written references take us back to about 3000 BP, and the oldest archaeological remains to some 7000 BP. Further back it seems impossible to go. Games were played with ephemeral and ad-hoc materials that either no longer exist, or, if they do, are not identifiable as games. This is a sobering thought, when you consider that Homo sapiens has been occupying the planet for nigh on 200,000 years. What, if anything, was going on in the 193,000 years before the earliest attestations? When, during all that time, did Homo ludens first appear?
There is of course a sense in which Homo has always been ludens. All higher mammals play (birds too), and not only as juveniles. So, even before the appearance of H. sapiens, something playlike must have been practised by such primate predecessors as Australopithecus, Homo habilis, Homo erectus, and our old friends, neighbours and occasional sex partners the Neanderthals.
All intelligent animals play, but there are obvious differences from human play:
- The most significant is that only humans play formal games. Animals play instinctively and stop playing when they get tired, hurt or interrupted; humans play by design and agreement. They are conscious of the fact that they are playing and can verbalise their agreement to have come to an end or to recognise a win.
- Animal play is typically categorised as locomotor, object, or social, with little or no overlap between them, but even human children can mix and match behavioural elements derived from all three, making it difficult to classify human play in such terms.
- The function of play in animals is to practise life-sustaining skills, which is one reason why juveniles play more than adults. (Another is that juveniles are less busy earning a living.) But this can hardly be the role of the sophisticated play practised by human children. If not, then what is it?
Peter Carruthers points out that human children are unique in that they engage extensively in pretend-play in infancy and childhood. No other species does so. Surely the pretend play of childhood is intimately connected to adult creativity? They share essentially the same cognitive basis, in that both involve exercises of the imagination. The function of human pretend play must be to practise for the kind of imaginative thinking later manifested in the creative activities of adults. Both activities involve a capacity to generate, and to reason with, novel suppositions or imaginary scenarios. Carruthers's insight may explain why humans continue play into adulthood more than animals, and even extend and develop it to the point that games can be likened to other art forms and be taken as seriously. Human games are creative, imaginative scenarios in a way that is not possible for the most intelligent of animals.
The earliest human games we have any record of are formal games. But you obviously cannot jump overnight from instinctive, undefined animal play to the sophistication of formal human games. So formal human games must surely have been preceded by vast periods of unattestable informal games, and in accordance with the principles of evolution should be regarded as a development from them.
I'd better clarify what I mean by a formal game, and for this purpose I need to cover a broader perspective than that of our primary interest in board games.
- Play is any behaviour done for immediate enjoyment rather than for subsistence or reproduction
- A game in the broadest sense is any unstructured period of play
- A game in the narrowest sense is a recreational contest designed to end in a win
Notable properties of a formal game are that it is:
- definite - it comes to an agreed end
- competitive - it is designed to end in a defined win
- symbolic - a win is virtual, or conventional (not life or death)
- regulated - to ensure equality (may be enforced by arbiters)
- cooperative - players agree to abide by rules and arbitration
- interactive - players can hinder one another (legitimately!. )(A non-interactive game is a competition)
Undefined periods of play are common to human and non-human animals, so I believe the first stage of development is period of play that does come to an agreed end but has little or nothing in the way of governing rules of procedure. Such games are played by children, and are often hard to distinguish from play-acting. We might assume they were also played by primitive humans. To illustrate this, here's one that may be fictitious but could well be rooted in fact.
Quartermain (Stewart Granger) explains "Iyetsava"
In the 1950 film version of King Solomon's Mines some words are put into the mouth of its hero Allan Quartermain that do not appear in the novel on which it is based. Musing on the cycle of life and death in the African jungle he dismisses it as nothing but 'iyetsava'. (That's what it sounds like. I'm guessing at the spelling.) When asked what that is he replies 'Iyetsava? Oh, it's a game the natives play. It doesn't make sense. It's quite pointless. A chases B, B chases C, C chases A. Then they turn round and go the other way. They make a great fuss about trying to get things away from each other. Bits of nothing; twigs and leaves. Quite senseless. Except that the fellow who was here has now satisfied his desire to be over there; everybody's had the fun of running after something; everybody's grabbed the handful of stuff that they wanted because everybody else wanted it; and well, it's endless. Quite pointless. Silly game, iyetsava'.
Simple as this sounds, it is still more sophisticated than animal play, for it combines all three of the categories associated only individually in animals. It is locomotor play, in that everyone is running around; object play, in that they are grabbing at 'bits of nothing: twigs and leaves'; and it is social play, because they recognise one another as individuals and have fun interacting with one another.
It is competitive, but not seriously so, since nobody can be said to have won. I see the next stage of evolution as being a competitive game designed to end in a win but not governed by articulated rules. In fact, I suggest that the concept of a win is more central to the definition of a formal game than its governance by rules. If I suddenly say 'Race you to that tree', and we both start racing to that tree, we won't have agreed on a set of rules, but the fact that we have started running automatically implies an agreement as to when the game ends and who will have won it.
The simplest of all rules need be no more than an implicit agreement that what we are doing is 'only a game': that we know what constitutes winning, and that winning is pleasurable but not a matter of life and death. Stephen Sniderman recounts the story of the group of scientists who tried teaching dolphins to play water polo. Although the dolphins learnt how to net the ball, and apparently enjoyed doing so, when the trainers tried to get them to stop the other team from "scoring," the dolphins launched an all-out war on the other team's players, using methods that no fair-minded human player would have dreamt of. In other words, they were unable to make the imaginative leap required to realise that what they were doing was 'only a game'.
A fully-fledged formal game is one played in accordance with previously agreed rules of procedure. This must be the latest stage in the evolution of games - partly because rules need to be verbally articulated, but also because the primary purpose of rules is to ensure that everyone is playing on terms of equality so that no one has an unfair advantage, and it is a well-known fact that hunter-gatherer societies are essentially egalitarian.
I want to survey some key stages in hominid evolution and ask what degree of game development could conceivably have been reached at each stage. Two critical factors to bear in mind are the development of language and the availability of leisure.
It seems obvious that language is needed for the expression of rules, but what degree of language development is needed to match what degree of formality? Even before the advent of writing (about 9000 BP) complicated rules would have required a degree of verbal formulation probably not achievable before the development of fully syntactic language. I see no reason why contests with formal rules should not have emerged from the elaboration of informal contests still classifiable as win-defined games, like the sort of ad-hoc foot racing I spoke of just now. This does not require highly developed language. Indeed, Merlin Donald quite explicitly advises us not to overestimate the significance of language in the development of games:
It is unreasonable to expect that the gulf between ape and human can be bridged by language alone. There are any number of non-verbal tasks that apes cannot master. They cannot acquire our athletic or play skills, for example. Human children play rule-governed games by imitation, often without any formalised instruction. They invent and learn new games, often without using language. No other animal can do this. Most informal games have little or no verbal component, and deaf-mute children have no difficulty mastering them. In human society games are identified with childhood and tend to come at the bottom of the cognitive pecking order, yet even children's games are beyond the reach of apes'.
Leisure and free time
Murray argues that formal games could not have appeared before the establishment of sedentism, which he associates with the beginnings of agriculture some 10,000 years ago. He sees this as primarily a question of leisure. Play time is essentially leisure time, as we are reminded by the dismissal of games as pastimes. 'All that primitive man needs for play [says Murray] is the opportunity. So long as he is fully occupied in the struggle to keep himself and his family alive, he has no time for it.'
Van Binsbergen disagrees. Referring to Sahlins's 1972 study of 'Stone Age economics' he counters that hunter-gatherers spend on average only about 20% of their time on productive activities, and are now thought to live in a world of relative plenty. A recent study of native Amazonian societies suggests that so long as production was adequate they valued free time more than excess production. If they did not have board games, says van Binsbergen, it is not because they had no time to spare, but because their lifestyle did not call for such inventions.
The truth surely lies somewhere between these two extremes. Leisure is indeed an obvious external necessity, and if it did not appear on a large scale before the emergence of sedentism this is not to say that it was never at all enjoyed by our hominid predecessors.
The difference between play and work, says Murray, is merely one of purpose: any human activity can be work in some circumstances and play in others. Susanna Millar similarly writes: "Perhaps play is best used as an adverb; not as a name of a class of activities but [as a description of] how and under what conditions an action is performed". This viewpoint suggests that games arise from a process of 'playful adaptation'.
Adaptation of what? In the simplest case, natural objects readily to hand. As Murray says:
Given leisure, people's innate urge to busyness impels them to action, if only to the manipulation of objects at hand, whether natural like pebbles, or man-made like beads - aimlessly at first, but, once their attention is seized, with the aim of exploring their capabilities for new uses.' Such objects serve first as playthings, then as materials for games. In the cave dwellings of meat-eaters, knucklebones must have been the earliest of playthings.
The adaptation of everyday activities must have led to the earliest physical games through the development of practice turned competition. For example, the primary function of bows and arrows in a hunting society is to secure food and to guard hunting grounds from rival tribes. A derivative activity would be to practise their use in order to improve one's skills. A subsequent development would be to establish, by competition, who the best marksmen are in order to allocate more efficiently the various co-operative labours of hunting. Some archers would gain more pleasure from the exercise of skill itself, or from the kudos attaching to the best of them, than from fulfilling their original purpose. What began as the purely practical thus gradually evolves into the purely recreational.
Then there is the playful adaptation of otherwise functional equipment. The most notable example of this process is that of divinatory equipment. In 1891 Stewart Culin argued for the origin of board games in divination in a passage too long to quote here but with which Murray disagrees in 1952. Murray himself, however, has been challenged by Wim van Binsbergen, who supports Culin's view at great length in his on-line paper listed below.
How can we envisage some of the stages of human play prior to the emergence of formal games?
Australopithecines were little more than walking apes, but they occupied open savannas, used simple tools, and developed greater cooperation among small bands with longer hunting times and distances. Their powers of communication were restricted to little more than gestures and vocalizations (grunts, shrieks, sighs, etc). With no evidence of symbolic thought, their play can hardly have differed from that of other adult apes.
Habilenes - Much the same applies to H. habilis. Habilis was a scavenger and lacked weapons, but did make simple stone tools and apparently was able to control fire. Its bigger brain supported larger bands, achieving occasional food surpluses. It is the first hominid exhibiting the bulge of Broca's area, a region of the brain essential to speech production. Some believe it could have had a form of communication intermediate between that of modern humans and non-human primates. It leaves no sign of manufactured artefacts, long-distance migrations, or base camps.
Homo erectus represents a major advance in hominid evolution, with a larger brain enabling it to invent in unprecedented ways. It was the first hominid to move out of Africa, leaving fossilized remains in Africa, Europe, Indonesia, Vietnam, and China. It was the first to hunt in coordinated groups, use complex tools (the Acheulean industry), and care for infirm or weak companions. It was the first to live in small, familiar band-societies with stable seasonal home bases similar to those of modern hunter-gatherers. Its manufactured implements demanded a considerable increase in cognitive sophistication, involving a solid mental template of the end product to be kept in mind over a protracted period of toolmaking. If erectus was capable of that, I submit, then it was capable of grasping the concept of a finite game as a complete structure, and if it could establish seasonal home bases it must have enjoyed at least a minimum degree of free time.
What about language? The Boxgrove site in England shows they hunted large dangerous animals, such as aurochs and horses, with wooden spears in elaborate coordination. Such planning, coordination and social organisation demand some form of language, though on physiological grounds it is unlikely to have been speech as we know it. Bickerton describes it as 'lexical protolanguage', capable of referring to phenomena other than those of the here and now. He sees it as a system of representation rather than communication, and hence as evidence of a rich perceptual/cognitive world predating modern humans.
To describe Erectus's form of communication and cultural transmission Merlin Donald employs the term 'mimetic'. The mimetic system, he writes, is a seminal hominid cognitive innovation, a mode of cognition that remains dissociable from language even in modern humans, and, is the logical basis of the first truly human culture. Its by-products include mime, play, games, skilled rehearsal, non-linguistic gesticulation, toolmaking, other creative instrumental skills, many non-symbolic expressive devices used in social control, and reproductive memory in general. We cannot guess what sort of play they found time to indulge in, but, following a flight of fancy, I would love to think of them as the inventors of Iyetsava.
In Donald's survey, mimesis remained the expressive mode of early modern humans and their close relations the Neanderthals.
Neanderthals were a predominantly European hominid first appearing between 300,000 and 230,000 years ago. Though anatomically distinct from H sapiens they are now thought to have sprung from the same ancestor and in their later history occasionally to have interbred with them, to the extent that up to 4 percent of the DNA of all present-day non-African people comes directly from Neanderthals. In severe glacial periods they migrated to the Middle East, but even in the interglacials they experienced a harsh and frigid climate. Sceptics may endorse the cartoon image of Neanderthals as primitive cavemen, pointing to their relatively static stone toolkit, apparent lack of art, and eventual displacement by Homo sapiens, but they were clever enough to prosper long in the most challenging of habitats. They gathered shellfish, plants and reptiles, and were extremely efficient hunters of large game, which they killed through strategy and cooperation rather than superior weapons. Fossils demonstrate that they held hides with their front teeth when fashioning warm clothing, like today's Inuit. Their tools were highly sophisticated; they delighted in personal ornamentation; they cared for their lame; and they buried their dead.
Such factors lead some to credit Neanderthals with a rudimentary language as dexterous as that of later Homo sapiens, indicating frequent and fluent speech with the basic elements of grammar and syntax. They could seemingly transcend the immediate necessities of everyday life (food, heat, sex) by mentally objectifying the achievements of the day, and analysing and qualifying them so as to be prepared to achieve better on the morrow. Their highly successful lifestyle must have afforded them some leisure time, and if we can't easily imagine them playing organised games in the great but freezing outdoors (except, perhaps, snowballs) we might at least envisage some form of Neanderthal night-life playing with discarded knucklebones by the communal fire.
Neanderthals are credited with complex thought processes possibly enabled by complex sentences, allowing the development of speech-based societies, albeit with a limited phonetic system. All physical features necessary for speech as we know it today are present by about 150,000 years ago in H. sapiens.
Homo sapiens. The fossil record shows anatomically modern humans appearing in East Africa some 200kya. This designation is carefully phrased. For at least 100,000 years they remained anatomically sapiens but cognitively and culturally at about the same level as their Neanderthal cousins. A glance at the timeline shows that they did not start expanding out of Africa until about 70kya, and that nearly all the major cultural achievements that we come to expect of them do not begin to kick in until somewhere between 50 and 40kya, undoubtedly not uniformly but in different bursts of development in different places. Colin Renfrew refers to the preceding long period of stasis as 'the sapient paradox', and Steven Mithen argues an extensive explanation for it by reference to the concept of "mental modularity".
Van Binsbergen agrees with Murray in situating the invention of board-games and in a Neolithic context of emergent agriculture but on grounds other than the mere expansion of leisure. In a rather high-flown passage he views both board games and material divination as arising from man's most drastic redefinition of space and time: space, in the demarcation of fields and the innovative straight lines of ploughed furrows, drainage trenches and irrigation ditches; and time, in the implementation of calculated anticipation through soil preparation, planting, weeding, and harvesting.
While not wishing to argue against that conclusion I still believe that formal games did not spring from nowhere but must have been preceded by and built upon the foundations of many millennia of informal games, that perhaps even Homo erectus might have had sufficient brain and found sufficient time in which to enjoy a round of Iyetsava. (And perhaps that Neanderthals played snowballs.) The origin and evolution of human play is a fascinating if tantalisingly elusive subject, and one that might be thought impossible to trace in the inevitable absence of fossil remains. There is, however, one field of study that has already proved capable of throwing suggestive light on the matter, and that is the relatively recent one of cognitive evolution, or the prehistory of the mind.
Bickerton, Derek (co-ed with Arbibd, M A), The Emergence of Protolanguage: Holophrasis Vs Compositionality (2010) (Return)
Binsbergen, Wim van, Board-games and divination in global cultural history (1999, on-line publication)(Return)
Bock, John, New Evolutionary Perspectives On Play (Guest editorial) (2004)
Carruthers, Peter, The Architecture of the Mind (2006) (Return).
Chick, Garry, Games in Culture Revisited: A Replication and Extension of Roberts, Arth, and Bush (1998)
Chick, Garry, The Adaptive Qualities of Leisure: a cross-cultural survey (1995)
Culin, Stewart, East Indian Fortune-telling with dice (1891) (Return)
Deacon, Terrence W, The Symbolic Species (1997)
Donald, Merlin, The Origins of the Modern Mind (1991). (Return)
Dunbar, Gamble, Gowlett (eds), Social Brain, Distributed Mind (2010)
Fagen, R. M., Animal Play Behaviour (1981)
Fischer, Steven Roger, A History of Language (2005)
Fitch, W Tecumseh, The Evolution of Language (2010)
Groos, Karl, tr. Baldwin, E, The Play of Animals (1898)
Groos, Karl, tr. Baldwin, E, The Play of Man (1901)
Iwaniuk, Andrew N et al, Do Big-Brained Animals Play More? Comparative Analyses of Play and Relative Brain Size in Mammals (2001)
Millar, Susanna, The Psychology of Play (1977) (Return)
Mithen, Steven, The Prehistory of the Mind (1996) (Return)
Mithen, Steven, The Singing Neanderthals (2005)
Murray, Harold J R, A History of Board Games Other than Chess (1952). (Return)
Renfrew, Colin, Prehistory: the Making of the Human Mind (2007) (Return)
Roberts, John M, et al, Games in Culture: a cross-cultural survey (1959)
Sahlins, M., Notes on the Original Affluent Society (1968) (Return)
Sahlins, M., Stone Age Economics (2000)
Smith, P. K., Does Play Matter? Functional and Evolutionary Aspects of Animal and Human Play (1982)
Tylor, Edward B, The History of Games (1879)