Dunbar's Number

I came across an interesting article in "The New Yorker" the other day ("The Limits of Friendship" by Maria Konnikova) on something called "Dunbar's Number (see the TED talk "Can the internet buy you more friends?" by Robin Dunbar).  The concept makes a lot of sense to me - basically, there is a limit to the number of friends that one human being can possibly have at any given point in time.  Dunbar is an anthropologist and psychologist from the University of Oxford who came across the concept more or less by accident.  He was trying to determine why primates spend so much time grooming one another (I know, it's an interesting question!).  He found part of the answer in a 1980's hypothesis which is now called the "Social Brain Hypothesis" (formerly known as the Machiavellian Intelligence Hypothesis, apparently). 

According to the "Social Brain Hypothesis,"primates (apes, gorillas, chimpanzees, orangutans, humans) have larger brains because they live in socially complex societies - in this case, the larger the group, the larger the brain.  Dunbar extrapolated from regression modeling of data obtained comparing the size of different species of primates' social groups with the size of their frontal lobes.  Based on the average size of a human brain, the number of people that any one individual could have in his or her social group was approximately 150 - any number greater than that would be too difficult to handle from a brain capacity standpoint.

After constructing a rationale for the so-called "Dunbar's Number," Dunbar conducted a number of laboratory and field experiments to prove his hypothesis.  For example, he surveyed 43 different households and asked questions about (here we go again) how many Christmas cards they sent out and under what circumstances.  As it turns out, the average network size (i.e., how many different family members, friends, and acquaintances each household considered part of their social group) was around 153, supporting the concept of Dunbar's Number.  He reviewed anthropological records and found that the average group-size of so-called hunter-gatherer societies was 148.4 individuals.  Again and again, regardless of what data set he reviewed, Dunbar found evidence to support the concept of Dunbar's Number.

As it turns out, Dunbar surmised (and again found supportive evidence) that there are a series of relevant group network sizes, depending upon the degree of "closeness" between individuals.  For example, out of the 150 members of a person's social group, there are about 50 individuals who would be considered close friends (close enough to invite to a dinner party).  Of those 50 individuals, there are 15 or so individuals who are close confidants (people that you would tell most things to).  Finally, of that last group, there are about 5 individuals that are considered "best friends" and family members.  Conversely, group network size can increase to 500 individuals if you consider them to be merely acquaintances and 1,500 for people you would be able to "put a name to a face."  Note that all of these groups are multiples of three - hence, Dunbar's "rule of three."  Going back to his original Christmas card study, Dunbar was again able to provide supportive evidence of his "rule of three."

I know what you are thinking right now - what about social media (Facebook friends, LinkedIn contacts, Twitter followers, etc)?  For example, the average Facebook user has 338 "friends," while the average LinkedIn user has over 500 contacts.  Social media apparently does not conform to Dunbar's Number.  However, Dunbar has a lot to say about the impact of social media on group social behavior and whether one of your Facebook "friends" is truly someone that you would consider "close enough to invite to a dinner party."  That is not for me to answer here, but what I will say is that the new generation who grew up with social media may answer this question very differently from someone like me, who grew up in an age without social media.

So what does all of this have to do with leadership?  Well, as it turns out, leaders likely need a certain brain capacity to be able to process all of the things that they need to do to effectively lead and manage their group networks, and the number 150 seems to be important again.  Since the days of the Roman Empire, the size of military units have been based on a similar concept to match unit leaders with an appropriate number of soldiers to effectively lead in battle.  For example, the basic military unit is called a company and consists of approximately 150 soldiers.  Companies are comprised of platoons (around 30-50 soldiers), which are comprised of squads (around 8-12 soldiers).  Multiple companies form battalions (around 500 soldiers), and multiple battalions form regiments or brigades (1,000-1,500 soldiers), which eventually form divisions (10,000 soldiers).  In other words, the military, since antiquity, has been effectively divided into social groups that roughly correspond to Dunbar's "rule of three"!

Most organizations today are designed with "Dunbar's Number" (or some version of it) in mind, even if not fully cognizant of doing so.  There have been a number of articles that have asked the question, "How many direct reports should a manager have?" (there is a good Harvard Business Review article here).  The answer likely depends on a number of factors (some of these factors may include the leader's locus of control, level of responsibilities both within and outside the organization, and the degree of collaboration with other areas of the organization that is required by the job).  Most experts would recommend somewhere between 7-9 direct reports.  Here again, we see evidence of Dunbar's "rule of three"!  The direct reports would correspond (roughly) to the "best friends" and family members described above for individuals outside of the work setting.  I am not suggesting that you should be best buddies with your direct reports or have your brother or sister report to you.  Rather, what I am suggesting is that the level of interaction and number of daily or weekly contacts that you have with your direct reports should equate roughly to how many contacts you would have with your "best friend" or close relative.  In order to effectively lead and manage, the number of direct reports needs to be workable based on the leader's brain capacity!  As it turns out, it all comes back to primate behavior and the "Social Brain Hypothesis" - perhaps we should all be anthropologists!