Play Will Persist

We live in the third human era, industry, which followed the farming and foraging eras. Each era introduced innovations that we expect will persist into future eras. Yet some are skeptical. They foresee “post-apocalyptic” scenarios wherein civilization collapses, industrial machines are lost, and we revert to using animals like mules and horses for motive power. Where we lose cities and instead spread across the land. We might even lose organized law, and revert to each small band enforcing its own local law.

On the surface, the future scenario I describe in my book The Age of Em looks nothing like a civilization collapse. It has more better bigger tech, machines, cities, and organizations. Yet many worry that in it we would lose an even more ancient innovation: play. As in laughter, music, teasing, banter, stories, sports, hobbies, etc. Because the em era is a more competitive world where wages return to near subsistence levels, many fear the loss of play and related activities. All of life becomes nose-to-the-grindstone work, where souls grind into dust.

Yet the farming and foraging eras were full of play, even though they were also competitive eras with subsistence wages. Moreover, play is quite common among animals, pretty much all of whom have lived in competitive worlds near subsistence levels:

Play is .. found in a wide range of animals, including marsupials, birds, turtles, lizards, fish, and invertebrates. .. [It] is a diverse phenomenon that evolved independently and was even secondarily reduced or lost in many groups of animals. (more)

Here is where we’ve found play in the evolutionary tree:


We know roughly what kind of animals play:

Animals that play often share common traits, including active life styles, moderate to high metabolic rates, generalist ecological needs requiring behavioral flexibility or plasticity, and adequate to abundant food resources. Object play is most often found in species with carnivorous, omnivorous, or scavenging foraging modes. Locomotor play is prominent in species that navigate in three-dimensional (e.g., trees, water) or complex environments and rely on escape to avoid predation. Social play is not easily summarized, but play fighting, chasing, and wrestling are the major types recorded and occur in almost every major group of animals in which play is found. (more)

Not only are humans generalists with an active lifestyle, we have neoteny, which extends youthful features and behaviors, including play, throughout our lives. So humans have always played, a lot. Given this long robust history of play in humans and animals, why would anyone expect play to suddenly disappear with ems?

Part of the problem is that from the inside play feels like an activity without a “useful” purpose:

Playful activities can be characterized as being (1) incompletely functional in the context expressed; (2) voluntary, pleasurable, or self rewarding; (3) different structurally or temporally from related serious behavior systems; (4) expressed repeatedly during at least some part of an animal’s life span; and (5) initiated in relatively benign situations. (more)

While during serious behavior we are usually aware of some important functions our behaviors serve, in play we enter a “magic circle” wherein we feel safe, focus on pleasure, and act out a wider variety of apparently-safe behaviors. We stop play temporarily when something serious needs doing, and also for longer periods when we are very stressed, such as when depressed or starving. These help give us the impression that play is “extra”, serving no other purpose than “fun.”

But of course such a robust animal behavior must serve important functions. Many specific adaptive functions have been proposed, and while there isn’t strong agreement on their relative importance, we are pretty confident that since play has big costs, it must also give big gains:

Juveniles spend an estimated 2 to 15 percent of their daily calorie budget on play, using up calories the young animal could more profitably use for growing. Frisky playing can also be dangerous, making animals conspicuous and inattentive, more vulnerable to predators and more likely to hurt themselves as they romp and cavort. .. Harcourt witnessed 102 seal pups attacked by southern sea lions; 26 of them were killed. ‘‘Of these observed kills,’’ Harcourt reported in the British journal Animal Behaviour, ‘‘22 of the pups were playing in the shallow tidal pools immediately before the attack and appeared to be oblivious to the other animals fleeing nearby.’’ In other words, nearly 85 percent of the pups that were killed had been playing. (more)

Play can help to explore possibilities, both to learn and practice the usual ways of doing things, and also to discover new ways. In addition, play can be used to signal loyalty, develop trust and coordination, and establish relative status. And via play one can indirectly say things one doesn’t like to say directly. All of these functions should continue to be relevant for ems.

Given all this, I can’t see much doubt that ems would play, at least during the early em era, and play nearly as typical humans in history. Sure it is hard to offer much assurance that play will continue into the indefinite future. But this is mainly because it is hard to offer much assurance of anything in the indefinite future, not because we have good specific reasons to expect play to go away.

GD Star Rating
a WordPress rating system
Tagged as: , ,
Trackback URL:
  • J Storrs Hall

    You need to read chapter 13 of Beyond AI.

    • Don Geddis

      Your chapter 13 seems to be speculation about the future design of AIs. What is the connection between that subject, and Robin’s Em scenario, which involves minds that are merely copied from humans, not designed?

      • J Storrs Hall

        Read it. You will find that play is a necessary part of the learning architecture, which I believe to be closer to the evolved human one than most synthetic AI approaches.

      • RobinHanson

        But I already think that, for similar reasons.


    Animals and foragers have play because a) play is indeed useful and b) spending more time foraging/hunting wouldn’t gain them anything they could use, it would just deplete their energy and the land. They don’t really live at subsistence level when you consider leisure time a form of wealth (which any serious economist should): there are just enough resources to sustain them but they can acquire all these resources with a fraction of their time and effort. This is comparable to someone who chooses to work part time, to just be able to get by, and spends the rest of their time on fun things. Farmers could spend their entire waking day on work, but they were protected against merciless competition by religious institutions (the holy Sunday, for example) and frankly it would’ve been hard for a lord to find an army that would enforce a no-leisure society. In the industrial era there were still some religious protections and there were central governments that relied on the people’s cooperation and who could be petitioned by labor unions, even so there were many workers (including children) who did not have any leisure time outside of Sundays.

    Fast forward to an em society with no real government, almost unlimited potential economic growth, high population growth, no religious authorities, no taboo against killing “lazy” individuals and where new individuals already come with skills (so no playful childhood required). I think leisure time would rapidly disappear, save for forced, work related “play”, and as a luxury for the richest individuals.

    • RobinHanson

      Seems you don’t think play is functional – its just an accident that has lasted for 100 million years. “More time foraging/hunting wouldn’t gain them anything they could use” that’s crazy as a description of most animals. And for farmers how is it that all religions everywhere just happened to accidentally create leisure?

      • IMASBA

        “that’s crazy as a description of most animals”

        No it’s not, they stretch the carrying capacity of the land and it takes them less than their whole day to do so. If there’s only going to be one prey animal in your territory per day and that’s enough to keep your belly filled you would waste energy by searching for more prey, even if you could find something small it might not be worth your energy. And that’s why tigers spend most of their time being just as lazy as a domestic cat. It was same with human foragers: they had no refrigerators and they could only trade excess stuff for the same stuff because every tribe was more or less the same.

        Farmers could increase the land’s carrying capacity at the expense of more work hours, foragers and most animals cannot do this, Maybe some species like ants are exceptions, and ants indeed do not seem to have leisure time.

      • Ronfar

        Play is functional in juvenile animals but adult animals play much less. The function of play seems to be education – if you don’t need to learn new things, you don’t need play.

      • RobinHanson

        But humans have neoteny because things that are functional for other animals when young are also functional for us as adults.

    • Ronfar

      Farming work is very seasonal in many areas. You have periods of time that require intense, constant labor and others in which the only thing you can do is wait. Ancient Egypt, farmer society par excellence, had so much surplus labor that they were able to build and maintain giant pyramids. It took industry to make year-round work productive for the masses.

      • RobinHanson

        And yet farmers played for most of the year.

      • IMASBA

        They may have played for some most days of the year but in total they played significantly less than foragers. At least I don’t consider building monuments, repairing homes, weaving clothes and being called on for military duty during the off-season as “play”.

      • IMASBA

        *for some time on most days of the year*

  • Pingback: Overcoming Bias : Play Blindness

  • Ari

    Curious, how much do you play? I know quite many professors and they already seem like robots in terms of work. I have previously worked as much as them for shorter periods of time, and it felt horrible. There’s an implicit status-game here. If you prefer work, you will come up with reasons to prefer a work-future as it will raise your status. Are you sure if you were born in another time with other genes, you would still prefer the same future? This kind of philosophical question of ultimate value feels like we ought to put our best effort to get right answers.

    Obviously I think what matter quite much is how much effort is required. Someone with higher IQ will go quickly through work that takes someone with less IQ to go pass. I do enjoy work but I also enjoy play.

    By the way, people feel motivated to work a lot more when the work feels meaningful. Think war-time efforts. I wonder if such motivation could be created in EM scenario?

    Have you actually calculated how much play time would be competitive in EM scenario? And how much work would be within their skillset (eg. satisfying) ?

    • Andy McKenzie

      Although he doesn’t note it here, Robin before has said that he likes playing a lot, in ways such as board games. I think he gets in his fair share of good play =)

    • Don Geddis

      “Are you sure … you would still prefer the same future?”

      Robin isn’t writing about a future that he “prefers”. He’s writing about a future he views as somewhat likely, given our current state of knowledge. He’s not advocating that people attempt to achieve this scenario, especially at the exclusion of other possible scenarios.

      • Joe

        Eh. I don’t buy that. Much of Robin’s argument is that a world of free competition just isn’t nearly as bad as many people seem to think: the parts of our minds that we like, the parts of our lives we think make them worth living, aren’t just some weird aberration that have appeared in spite of competition, but are actual functional useful features, and so can mostly be expected to continue into the future, at least in some form or other, even if changed in many ways.

        This might not mean that doing nothing to shift the future from its natural course is the best possible choice to make. But it does certainly make that choice vastly better than it looks under the alternative worldview, where unless we slay the demon of free competition, sentient conscious beings will be outcompeted and consumed by homogeneous mind-slush, the natural conclusion of efficiency left unchecked.

        Robin has said a number of times that he does consider the em world pretty good – better than our world, even. And actually I think that, assuming his positive claims are true, that’s a very reasonable stance, probably the same one that most people would come to, if they similarly believed his positive claims.