6 March 2011

Norms as Mutual Intentionality

Filed under: Culture, Language, Psychology — ktismatics @ 12:37 pm

According to experimental psychologist Michael Tomasello, human language, learning, culture, and cooperation are all built on a common foundation of shared intentionality. Humans are like other animals in that they act with intent: they seek food, flee from predators, pursue mating opportunities, and so on. Humans differ from other animals in recognizing that other humans also act with intent. In Tomasello’s paradigmatic exemplar, a 12-month-old (pre-linguistic) human infant points to an object that someone else is seeking. By pointing, the infant simultaneously demonstrates that he understands the other’s intentions with respect to the pointed-at object, and he expects the other to understand the pointing gesture as an intentional act of information-sharing.

In Why We Cooperate (2009), Tomasello proposes that shared intentionality serves also as the basis for norms. Norms of cooperation make sense in this context: if each of us wants food, and each of us knows that the other wants food, then we are both functioning within a joint intentional frame. Establishing a norm for sharing food is useful for avoiding a fight, during which some third individual might sneak up and steal the food from both of us or a predator might eat us both for lunch.

But what about norms of conformity? If no one is actually harmed by violating a social convention, why do humans both follow and enforce the convention? The preschool teacher tells a kid to put the coat over there in the corner: why does the kid do it, rather than just tossing her coat wherever she feels like it? Piaget claimed that social rule-following is motivated initially by authority: if I toss my coat the teacher will punish me. Piaget asserted that only later, after kids have internalized authority-driven conformity, lose their thoroughgoing egocentricity, and start seeing one another as autonomous agents, do kids start enforcing the coats-go-over-there rule on one another.

So here’s a study that Tomasello conducted to evaluate the Piagetian theory of developmental acquisition of norm-based behavior:

“Three-year-old children were shown how to play a one-person game. When a puppet later entered and announced that it, too, would play the game, but then did so in a different way, most of the children objected, sometimes vociferously. The children’s language when they objected demonstrated clearly that they were not just expressing their personal displeasure at a deviation. They made generic, normative declarations like, ‘It doesn’t work like that,’ ‘One can’t do that,’ and so forth. They do not merely disapprove of the puppet playing the game differently; he is playing it improperly. This behavior is of critical importance, as it is one thing to follow a norm — perhaps to avoid the negative consequences of not following it — and it is quite another to legislate the norm when not involved oneself.”

In this study the rules don’t regulate the players’ cooperative social interaction within the game; rather, the rules are the game — they define the game — and playing the game is not a social activity but a solitary one. In learning the game the kids in this study were not subjected either to reward for following the rules or to punishment for breaking them. The puppet doesn’t play the game less effectively by not following the rules, since changing the rules changes the game. The puppet isn’t cheating an opponent or keeping a teammate from succeeding — isn’t violating mutual cooperative expectations — by not playing properly. In learning from the adult how the game is “supposed” to be played, the adult doesn’t have to describe the rules or make a mistake and then correct herself. Merely demonstrating how the game is played is enough to activate expectations in the observing kids about how the game should be played.

According to Tomasello, results of this study demonstrate that

“even young children already have some sense of shared intentionality, that is to say, that they are part of some larger ‘we’ intentionality. I contend that without this added dimension of some kind of ‘we’ identity and rationality, it is impossible to explain why children take it upon themselves to actively enforce social norms on others from a third-party stance, expecially those norms that are not based on cooperation but rather on constitutive rules that are, in an important sense, arbitrary.”

Tomasello contends that this sense of collective intentionality witnessed in three-year-olds is a generalized manifestation of the “he is me” identification with the other that is already manifest in the joint intentional behavior of those 12-month-old pointing infants he previously told us about. Conforming to the arbitrary rules of a game and enforcing them on others is not that different from participating in “language games,” in which specific meanings are arbitrarily assigned to specific sounds or physical markings — a game that can be played by oneself as well as with other people.


For readers with a philosophical orientation to normativity as it relates to human cognition and society, you might want to check in with bloggers Duncan Law, Deontologistics, Planomenology, and Minds and Brains.


  1. Hi John,

    Good post, very informative. I’ve been thinking through this in relation to our former conversations on Paul’s law versus spirit. It’s also interesting to extrapolate these thoughts a bit and ponder their political ramifications. I resonate with anarchist and libertarian perspectives, but if one is not an anarchist or libertarian, then presumably one supports some sort of societal regulations and an army of force to enforce it. This is our thinking in the States.

    In the States we incarcerate more of our population (as a percentage) than anyone in the world. Our thinking is to force conformity with the law. This “philosophy” (if it can be called that) is faulty in many ways and ill-conceived, but in light of your post, we might ask whether or not our national approach to social relations has produced a genuine and sincere concern for the collective good. In other words, has the authoritarian enforcement of rules produced a spirit of care for one another or a greater community consciousness?

    Is the answer more rules? Greater enforcement?

    Or do we need another approach? One that affects change at a deeper level.

    Thanks for passing along the info.


    Comment by Erdman — 7 March 2011 @ 6:33 pm

    • Hey Erdman. Tomasello cites quite a few research studies, but the one I describe in the post involves a set of arbitrary rules. The findings suggest that even very young kids are apt to adopt such rules for themselves and to enforce them in others as if these rules were the “right way” to do things. The implication is that humans carry an innate tendency to engage in rule-governed behavior. This tendency is helpful within a group or community, making sure that everyone is on the same page with respect to mutual expectations and coordination of complex coordinated activities. But since the rules can be arbitrary, it’s distinctly possible that two groups will arrive at very different sets of rules. In fact, a group might even opt for an idiosyncratic rule set that distinguishes its members from other groups; i.e., who’s in versus who’s out. Says Tomasello:

      “Imitation and conformity can create high degrees of intra-group homogeneity and inter-group heterogeneity, and on a faster time scale than that of biological evolution. Because of this peculiar fact — presumably characteristic of no other species — a new process of cultural group selection became possible. Human social groups became maximally distinctive from one another in language, dress, and customs, and they competed with one another… Indeed, recent evolutionary models have demonstrated what politicians have long known: the best way to motivate people to collaborate and to think like a group is to identify an enemy and to charge that ‘they’ threaten ‘us.’ The remarkable human capacity for cooperation therefore seems to have evolved mainly for interactions within the local group. Such group-mindedness in cooperation is, perhaps ironically, a major source of strife and suffering in the world today. The solution — more easily described than attained — is to find new ways to define the group.”

      It’s pretty evident that, while some of the Old Testament laws have some rational logic behind them, many other seem quite arbitrary. Further, Yahweh repeatedly tells the Jews that they are not to be like the other nations, are not to follow their ways, and so on. So it’s evident that the in-group versus out-group distinctions are being laid down. Paul says he’s dead to the law, neither Jew nor Greek, and so on, presumably as a means of overcoming the in-group versus out-group dialectic inherent in Biblical Judaism. He does, however, replace it with a believer versus unbeliever distinction that’s got just as much bite as the Law. Anyhow, building community based on mutual enforcement of a set of distinctive laws and customs and social norms might build on innate human tendencies, but the fascistic us-versus-them groupthink is a serious risk.

      Here’s Ronald Reagan’s solution to this dilemma:


      Comment by ktismatics — 7 March 2011 @ 8:39 pm

      • Ha! UFO’s as the answer to world peace….why didn’t Reagan win a Nobel Peace Prize? Damn liberals.


        Comment by Erdman — 8 March 2011 @ 6:21 pm

  2. “I contend that without this added dimension of some kind of ‘we’ identity and rationality, it is impossible to explain why children take it upon themselves to actively enforce social norms on others from a third-party stance, expecially those norms that are not based on cooperation but rather on constitutive rules that are, in an important sense, arbitrary.”

    Good stuff JD. As to Tomasello I got suspicious at the ‘I contend’, one of those stock-phrases most often seen dressing up a tired point in rhetorical finery, and even moreso at the ‘without [my hypothesis] it is impossible to explain’, another stock-phrase favored by those with poor explanatory imaginations.

    Some time ago G.H. Mead called the shared-intentionality effect in question ‘taking the perspective of the other’, and thought it could be understood essentially in terms of tracking objects in ballistic space. Explanatory parsimony doesn’t need there to be any ‘shared’ intentionality at all, just a learned understanding of the situated behaviors of the whatsits in the environment. Introduce me to whist or to carburetors or to waterfalls, it’s the same basic thing – I’m looking for their characteristic patterns in relation to my conduct. Mead uses baseball as an example. I don’t need to have any sense of the intentionality of the shortstop-object to play first base, I just need to know what shortstop-objects (and batter-objects, and ball-objects, and grass-objects, and wind-objects, etc.) characteristically do in given situations and act accordingly. In this sense I internalize a variety of ‘significant others’ in relation to which the sense of ‘me’ develops. In short, it’s a bunch of stimulus-response loops that get really complexly interactive and emergent very quickly. But there’s no warrant anywhere in here for leaping to shared intentionality, although that’s not automatically a bad way to describe what’s emergent.


    Comment by Carl — 7 March 2011 @ 6:36 pm

  3. Hey Carl. Even if Mead is right about perspective-taking, it still seems to be the case that only humans are any good at doing it. Chimps, for example, could watch a shortstop play for months, but if you gave him a glove and put him out on the field I’m pretty sure he wouldn’t try to “be the shortstop.” As a matter of fact, Tomasello’s group tested this out by having chimps play a game with humans in which there was no instrumental goal such as acquiring food. Says Tomasello:

    “Results were clear and consistent. The chimpanzees showed no interest in the social games, basically declining to participate. In the problem-solving tasks, on the other hand, they synchronized their behavior relatively skillfully with the human, as shown by the fact that they were often successful in bringing about the desired result. However, when the human partner stopped participating, no chimpanzee ever made a communicative attempt to reengage her — even in cases where they were seemingly highly motivated to obtain the goal — suggesting that they had not formed with her a joint goal. In contrast, the human children collaborated in the social games as well as the instrumental tasks. Indeed, they sometimes turned the instrumental tasks into social games by placing the reward back into the apparatus to start the activity again; the collaborative activity itself was more rewarding than the instrumental goal. Most importantly, when the adult stopped participating in the activity, the children actively encouraged him to reengage by communicating with him in some way, suggesting that they had formed with him a shared goal to which they now wanted him to recommit.”

    I.e., there has to be some motivation for the creature exposed to the S-R loops both learn the patterns but also to behave in accord with those patterns. Why would anyone want to learn how to play shortstop? Maybe because they get esteem points from performing well, which might lead to mating opportunities and valuable endorsement contracts later in life. However, Tomasello has studied little kids’ helping behaviors and found that kids who have been rewarded for helping are less likely to continue helping if the rewards are no longer given, versus kids who were never rewarded, who just seem to find helping in itself a rewarding activity.

    Does a motivated and cognitively skilled creature learns complex behaviors like shortstopping via exposure to repeated S-R loops? There have been studies in which an adult demonstrates to a kid how to perform some sort of complex intentional behavior sequence that includes some pointless behavioral epicycle. E.g., someone demonstrating how to play shortstop might drop the ground ball, then pick it up again and throw to first base. If kids were imitating according to S-R behaviorism they’d imitate the useless bits of behavior (e.g., dropping the ball) as well as the useful sequence (fielding the grounder and throwing to first). Kids generally don’t do that: they eliminate the extraneous actions even on the first trial. I.e., they seem to infer the other’s intent that motivates his/her behavior, adopting that intent in learning rather than blindly imitating.


    Comment by ktismatics — 7 March 2011 @ 9:15 pm

    • Fair enough. In the sense I’m getting at chimps do ‘it’ just as well as humans, but then so does river water. That is, we all navigate fields of objects that become relevant to us in situated ways, acting ‘as if’ we understood their intentions: chimps for example by avoiding where the tigers are, water by charting the topography and splitting around the rocks. In this sense we are all players in the games that suit our particular objectivity.

      And I’m saying that baseball is not qualitatively different than this from an input/output perspective for humans. That’s just a relevant game for us. I don’t see the need for the shared-we hypothesis, or rather if it applies to us it applies to rocks and water also.

      Btw, in Mead and my attempted translation of him here, imitation is a red herring. We don’t learn primarily by imitation any more than chimps learn about tigers by acting like tigers or water makes a river by acting like rocks. I think intention is (at least at this point of the discussion) also a red herring in the same way. We need not intend to play first base any more than water intends to make a river.


      Comment by Carl — 8 March 2011 @ 12:03 pm

  4. I’m still not sure I understand what you’re trying to say, Carl. (That was a mutual intentionality joke btw.)

    Chimps do chimp stuff as well as humans do human stuff, and a lot of chimp stuff is similar to human stuff. Tomasello doesn’t propose mutual intentionality as a magical miraculous capability of humans, without precedent or cause evolutionarily. Still, chimps fail at doing a lot of kinds of human stuff: language, imitative learning, tool use, social organization, cumulative culture. Even in understanding cause-effect relationships of inanimate objects appearing in their natural environment, chimps aren’t very good compared to humans. Tomasello contends that it’s not just the ability to read the other’s intention as a predictor of behavior that is the key distinctive between chimps and humans. It’s the ability to coordinate one’s own intentions with the other in joint activity that’s crucial in humans, and crucially lacking in chimps. It’s not that chimps wouldn’t benefit from joint intentionality in the wild. They could plan monkey hunts, figure out how to divide the carcasses without fighting, set tiger traps, plant banana plantations, etc. Rocks and streams could coordinate their activities to build hydroelectric dams, allowing them to heat the water so the rocks could enjoy a nice cup of tea or to keep the stream from freezing in the wintertime…

    You say that we need not intend to play first base — is this like the apocryphal roomful of chimps who need not intend to type Hamlet on a keyboard but who succeed eventually nonetheless? If you discount both imitation and intentionality, why and how would someone play first base in a way that other baseball players on the field would recognize as appropriate to playing a game together? Put it this way: I wouldn’t want that first baseman on my team.


    Comment by ktismatics — 8 March 2011 @ 2:04 pm

  5. Here Tomasello discusses the research I cited in this post, where children enforce arbitrary rules to a game that has no apparent instrumental purpose and that is played by individuals, in the context of Mead:

    “These studies demonstrate that even outside of such cooperative activities, children also value conformity to the group — both their own and that of others. Initially children base such ‘we-ness’ on identification with significant-other individuals such as parents and family and schoolmates (G.H. Mead’s significant other), and only later generalize them into truly impersonal norms based on identification with some type of cultural group (Mead’s generalized other). Young children thus become more adult-like in their understanding of social norms by coming to comprehend ever more about: (a) the ‘arbitrary’ nature of norms, based on consensus; and (b) the independence of norms from any specific individuals (their ‘agent-neutral’ status).”

    Even if the norms of baseball are arbitrary, there is a mutual agreement among players to abide by those norms. The norms aren’t just rules about what’s allowed and disallowed; they also imply standards of excellence in performance. The first baseman who repeatedly drops thrown balls is playing according to the rules, but he’s performing poorly according to the standards. It seems to me that the standards of performance have to do primarily with intentionality. I.e., the defensive team’s intent is to prevent the offensive team’s players from reaching base safely; if the first baseman drops the throw he thwarts the defense’s intent.

    This distinction between rules and standards came up in a prior discussion about scientific practice. For now let’s table the possibility that the rules of science are arbitrary. It is possible to be a rule-following scientist who performs poorly; e.g., bad study design, sloppy data collection, math mistakes in doing statistical analysis, etc. It’s important for an individual scientist to play the game with both integrity and skill not only for his/her own success, but because science is a team sport. The other people in your lab count on you: other researchers in the same discipline rely on your findings.


    Comment by ktismatics — 8 March 2011 @ 5:04 pm

  6. “Introduce me to whist or to carburetors or to waterfalls, it’s the same basic thing – I’m looking for their characteristic patterns in relation to my conduct.”

    “we all navigate fields of objects that become relevant to us in situated ways, acting ‘as if’ we understood their intentions: chimps for example by avoiding where the tigers are, water by charting the topography and splitting around the rocks.”

    I think it’s reasonable to assert that intentionality is like a waterfall in the sense that cause-effect relationships are involved in both. Tomasello argues that children infer intentionality by observing that others are unhappy when their intentions are unsatisfied. Do little children infer that a stream is frustrated when its flow is impeded by rocks? They might, although in Descartes’ Baby Bloom cites empirical evidence that children typically do not ascribe intentional states to inanimate objects. Still, there’s not a lot to lose for a kid to generalize intent broadly if it helps the kid make accurate predictions about the world. So: the stream’s intent is impeded by the rocks, so what will the stream do in order to overcome this impediment? This is Dennett’s “intentional stance” — inferring intent has pragmatic value, even if there is no real intentionality involved. Even though Dennett is skeptical about intentionality even in humans, he agrees that it’s pragmatically useful to treat other humans as if they act with intent, whereas doing inferring intent with inanimate objects just isn’t necessary to account for what they do.

    It’s certainly possible to interpret Tomasello in terms that don’t conflict with Dennett’s. I personally am prepared to accept that humans really do act with intent — that the shortstop intends to catch the ground ball, that the batter intends to reach base safely, and that the shortstop knows the batter’s intent and vice versa. Whether shortstop’s and batter’s intentions are freely chosen by them or determined by interactions of neurons and muscles and eyes and environment isn’t really relevant at the level of analysis required for actually playing a game of baseball. Similarly, it’s not important to say whether the floor is really a solid mass when my intent is to walk across it from point A to point B.


    Comment by ktismatics — 9 March 2011 @ 6:01 pm

    • Thanks John, your replies here helpfully engage and trouble my responses. I especially like the point about the pragmatic value of inferring intention differently. I tend to be a deflationist where intention is concerned, but I don’t really believe it’s always self-flattery. Just more often than we care to think. And I therefore think that there’s sometimes a pragmatic value in not inferring intention, because it’s one of those default explanations that may conceal more than it reveals.


      Comment by Carl — 14 March 2011 @ 9:51 am

  7. Duncan Law and I have been discussing this post and Tomasello’s book here.


    Comment by ktismatics — 10 March 2011 @ 6:54 am

  8. […] most favored source On this subject see: https://ktismatics.wordpress.com/2011/03/06/norms-as-mutual-intentionality/ Additionally on this topic you can read: […]


    Pingback by Is 2012 Real ? Find out the truth! — 10 March 2011 @ 9:25 am

  9. Thanks for the pingback, but I’m curious about the context. From the link:

    “While we contemplate the planets and stars, in astrology this 2012 subject matter has strong weight. When looking at the constellations we realize that this particular date has even greater value on our lives as it is the year that the Age of Aquarius really shifts into maturity. This is the electrifying beginning of our age. The most favored source…”

    Maybe you perceive an intention in my post that I hadn’t consciously recognized. Or maybe the Age of Aquarius is pervaded by such harmony and understanding that we all share the same joint intentional frame — or at least the same joint.


    Comment by ktismatics — 10 March 2011 @ 9:48 am

RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Create a free website or blog at WordPress.com.

%d bloggers like this: