Social Gravity

Definition

The shape of space makes people move, and the gravity of social attention shapes space. Individuals in a network have social gravity, some greater than others. When they move through a social network, more people are drawn in by this social gravitational field. With enough attention, areas of great social mass become approximate black holes, drawing thousands of views to a single web page or object.

How the metaphor works

Gravity is the right analogy for three specific reasons.

Mass attracts mass. Attention accrues to things that already have attention, and the rate of accrual rises with the amount already held. Robert Merton named this the Matthew effect in 1968, and network scientists later described the same behavior as preferential attachment: a new node joining a network is more likely to connect to a node that already has many connections. Neither mechanism requires the popular thing to be better, only for it to be visible first.

Gravity falls off with distance. Social gravity does too, though the distance is measured in hops through a network rather than in metres. A person two connections away from a high-mass node feels its pull; a person twenty connections away does not.

Mass curves the space around it. This is the part the definition above gets at with "the gravity of social attention shapes space." A sufficiently large node changes the paths available to everyone near it. Conversations route through it, other people position themselves relative to it, and the shape of the surrounding network is determined by where it sits.

Where the metaphor breaks

Physical gravity is symmetrical. Two masses attract each other in proportion to both. Social gravity is not. A node with a million followers pulls on each of them and is pulled on by none of them individually, which is the asymmetry described in Local vs. Global Connectivity.

Physical mass is also conserved and social mass is not. Attention can be manufactured, purchased, and destroyed, and a node can lose all of it in a day. The black hole comparison is useful for the accumulation and misleading about the permanence.

Interface as terrain

The first sentence of the definition is the more useful half, and it is the one most often skipped: the shape of space makes people move.

An interface is terrain. What is placed above the fold, what is default-on, what requires one click versus four, and what is ranked first all determine where people go, in the same way a valley determines where water goes. Designers set the topography and then read the resulting movement as preference.

This is where social gravity stops being a metaphor and becomes a design decision. A ranking algorithm that sorts by engagement is a machine for concentrating mass, because it routes new attention toward whatever already holds attention. The concentration is not an emergent property of human nature. It is the predictable output of a sorting rule, and a different rule produces a different distribution.

Related Reading

• Local vs. Global Connectivity
• Designing Social Interfaces
• The Presentation of Self in Everyday Life
• Interaction Ritual : Essays on Face-to-Face Behavior
• Behavior in Public Places: Notes on the Social Organization of Gatherings
• Collective Intelligence: Mankind's Emerging World in Cyberspace
• The Wealth of Networks: How Social Production Transforms Markets and Freedom
• Reassembling the Social: An Introduction to Actor-Network-Theory
• The Tipping Point: How Little Things Can Make a Big Difference
• Outliers: The Story of Success
• Society of the Spectacle
• Class: A Guide Through the American Status System
• Conspicuous Consumption
• Ambient Findability
• Equipotential Space: Freedom in Architecture
• Space is the Machine
• The Social Life of Small Urban Spaces

Key texts

• Merton, Robert K. "The Matthew Effect in Science." Science 159, no. 3810 (1968): 56-63.
• Barabási, Albert-László, and Réka Albert. "Emergence of Scaling in Random Networks." Science 286, no. 5439 (1999): 509-512.
• Salganik, Matthew J., Peter Sheridan Dodds, and Duncan J. Watts. "Experimental Study of Inequality and Unpredictability in an Artificial Cultural Market." Science 311, no. 5762 (2006): 854-856.