Critical Summary of Ramo's The Seventh Sense

Grand summary: I wouldn't go as far as to say that Ramo's The Seventh Sense is terrible, but the book is definitely not worth the time (or money). Instead of reading 352 pages, read this ~1000 word critical summary (and see the end of the post for several superior alternatives).

McKinsey & Co. recently surveyed CEOs around the globe about what they were reading. Here is a quick recap of those who said they were reading Joshua Cooper Ramo's The Seventh Sense: Power, Fortune, and Survival in the Age of Networks:

  • Dominic Barton, McKinsey
  • Reid Hoffman, LinkedIn
  • Chuck Robbins, Cisco
  • Wendell P. Weeks, Corning

Not a bad list. It was also about networks, which I study. Then I saw that Malcolm Gladwell had positive things to say about the book, and I really liked all of Gladwell's books. I decided to purchase it, and boy was I disappointed. Indeed, I'm so disappointed I wrote this post. (Also,  these CEOs may have been reading the same book because "Maybe the book was in a hotel room at some conference they all attended.”)


First, Ramo begins by stating the obvious: networks matter. Second, he makes the entirely unoriginal argument that connection changes the nature of the objects connected. Third, Ramo makes the dubious claim that networks matter more now than ever -- we are on the cusp of a "new revolution," akin to the Enlightenment or the Industrial Revolution.

Fourth, and most important, he argues that if people are going to take advantage of this newfangled "networked" world, they have to develop a new "sensibility" that he calls the "seventh sense." This last point might be true, but Ramo fails to give us any useful insight into what that might mean in practice. In short, he diagnoses a problem but offers no remedy. To add insult to injury, Ramo's writing style is disorganized and chronically afraid of reaching a point. 


It is important to think about how social, economic, and political life is shaped by connections. Luckily, scholars have been doing just this for almost a century! Here are just a few concepts (some of which Ramo attempts to describe):

  • Nodes vs Relations: Although these concepts seem obvious (e.g. nodes are people or organizations, and relations are how those people or organizations are connected), the nature of nodes and relations can entirely change the definition and inner-working of the network under scrutiny. First, relations can be nearly everything, and mathematical speaking, they can even be the absence of connections. This is important because "seeing" the absence in networks is as important as "seeing" how nodes are connected (see, for example, research on structural holes and vacancy chains). Ramo focuses almost exclusively on computer mediated networks or information networks, and only occasionally discusses others. When he does, he elides that fact that depending on the definition of the node and relations, the network will have fundamentally different organizing principles.
  • Efficiency vs Redundancy: The hub-and-spoke network form is a highly efficient design, in it's extreme form, every node is separated by only two-degrees (i.e. a short path) with the fewest possible connections. Unfortunately, it is weak because one only has to remove one node -- the hub -- to disconnect the entire network. The alternative extreme is redundant network in which every node is connected to every other node (i.e. maximal degree). Unfortunately, this tends to be very resource intensive to maintain. This is an idea that Ramo struggles to convey at several points throughout the book, but only succeeds in flailing around.
  • Bonding vs Bridging: In social networks there is a tendency for "cliques" to form (i.e. bonding ties). This leads to information redundancy. Fortunately, it only takes a few ties to form between cliques (i.e. across structural holes) for new information to flow. This is referred to as a "small world network," and forms a happy middle ground between an efficient and a redundant network. As it relates to resources, it is generally argued that bonding ties help social cliques maintain what they already have, while bridging ties help cliques acquire what they do not yet possess.
  • Preferential Attachment vs Reciprocity: In some networks, when nodes enter a network they are more likely to try and connect to nodes that already have high degrees (i.e. preferential attachment). If this tendency proceeds unchecked, the result is referred to as a scale-free network (this naturally leads to a hub-and-spoke pattern). In networks -- especially social networks -- reciprocity acts as a check on preferential attachment. For instance, lets say a new kid tries to form a friendship with someone because they are popular. This popular kid has so many friends, however, that he or she can only spend so much time with all of them. There is a level of non-reciprocity in time at which the new kid will give up and find a different friend.
This Delta Airlines route map approximates a hub-and-spoke network. It is because of the tendency for preferential attachment in flight networks, that the distribution of degrees follows a power law (i.e. airports form a scale-free network).

This Delta Airlines route map approximates a hub-and-spoke network. It is because of the tendency for preferential attachment in flight networks, that the distribution of degrees follows a power law (i.e. airports form a scale-free network).

  • Selection vs Influence: These two concepts relate to nodes. On one hand, as a result of characteristics of the node, it may be more likely to join a particular network (i.e. it selects into a connection). For instance, two people who enjoy the same music are more likely to attend similar events, and have more to talk about, and therefore form a connection. On the other hand, once a connection is formed, a node may change its characteristics as a result (i.e. it is influenced by a connection). Continuing with the music example, one person may begin to listen to a band only after forming a friendship with someone who already listened to that band. These two mechanisms lead to the observation of "homophily" in many social networks, which means connected people tend to be much more similar (on an array of characteristics), than they are with people to which they are not connected. Ramo seems to have fallen into the trap of assuming that influence is the only mechanism at play, or that these cannot be empirically teased apart. Selection still matters a great deal. Anyone with the "seventh sense" would see this.

Now you know more about networks than Ramo! Go forth and lead your network to conquer other networks.

Read Instead

Overall, I agree that networks do matter (they *always* have) and that people should begin to think and imagine the world in network terms. Here are several popular and accessible books that do a much better job than Ramo's rambling tome.

Here are a few academic books on the matter, but are nevertheless accessible and useful for understanding network concepts (without much math).




Dustin Stoltz