Modern Sociometry:
how the deep science of teams returns.

You have more data about your employees than any organization in history. You also have less structural insight into your teams than the United States Navy had in 1942. That is not a metaphor. During the Second World War the Navy used a quantitative structural science of human groups, and its measures of group cohesion tracked accidents, sick days, morale collapse and disciplinary failure from the relational pattern of a unit alone. The Navy was not alone in this. The US Army used it too, and the British Army drew on it in officer selection, through the War Office Selection Boards. Industrial psychologists used it. Sociometry had its own journal and a serious academic following across American universities.

Then the science was hollowed out.

Its name and its surface survived: in classroom seating charts, in the badge-and-sensor analytics that measure who talks to whom. What was abandoned was its depth, the original ambition to read the structural composition of a team rather than merely its traffic.

This publication restores that depth.

I. The 1934 Invention

In 1934, a Romanian-American psychiatrist named Jacob Levy Moreno published Who Shall Survive? A New Approach to the Problem of Human Interrelations. The book was the founding document of sociometry, Moreno’s term for the quantitative science of human relational structure.

Moreno’s claim was both simple and radical. He wrote that human groups “have an actual, dynamic, central structure underlying and determining all its peripheral and formal groupings.” That structure was not the organisational chart. It was not the survey result. It was not what people said about each other when asked. It was the invisible architecture of who carried whom, who relied on whom, which combinations produced spontaneity and which produced friction, the load-bearing pattern beneath every visible team.

Moreno invented a method for drawing this structure. He called the diagram a sociogram. He defined the field that produced such diagrams as “the inquiry into the evolution and organization of groups and the position of individuals within them.” He was pointing, in the careful language of his era, at the inner structure rather than the outer.

He did not work alone. His collaborator Helen Hall Jennings co-authored the founding work. In 1938, more than two decades before Erdős and Rényi formalised the random graph in mathematics, Moreno and Jennings had already published the first stochastic network model in social science, which they called a chance sociogram. They were, by any honest historiography, the first network scientists.

II. What Replaced It

In the decades after Moreno, sociometry was displaced, for a plain reason: other methods were easier to run at scale. The engagement survey could be dashboarded. The personality test could be branded and shared. The people-analytics platform could be procured and defended in a budget meeting. Each is a real instrument, and each measures something real: what people say they feel, what a person looks like on their own, what can be counted from the outside. Each answers a useful question.

But each measures a proxy. None of them reads what Moreno built sociometry to read: the structural composition of a team, how its members actually fit together, where the load sits, where it is heading. The easier methods became standard. The deeper one fell out of practice.

III. The Return — Through Network Science

While organisational practice forgot sociometry, theoretical sociology did not. In the 1970s at Harvard, sociologist Harrison White built a generation of students around what was, in effect, a sociometric revival. Two of them changed the world. Mark Granovetter published The Strength of Weak Ties in 1973, demonstrating that the structural position of a relationship (not its emotional intensity) was the load-bearing variable in information flow. Ronald Burt formalised the theory of structural holes, the gaps in the social fabric where strategic value accumulates. Both men, by their own description, were building on Moreno.

This same lineage went on to construct the foundations of social network analysis (SNA), the field that underlies Facebook’s friend graph, LinkedIn’s recommendation engine, Google’s PageRank, and the contact-tracing models built during COVID-19.

Sociometry didn’t die in the 1960s. It went underground. It became the foundational science of how connection works at scale, applied to everything except its original purpose: reading the structural health of a human team in an organisation.

IV. What The Sociographer Does

The Sociographer is a practice that publishes modern sociometric readings of human teams, rendered in the editorial register of a serious publication, on demand rather than on a schedule.

Every reading reads three structural layers, and renders what they compose into.

A note on the word sociogram. Moreno’s original technical sociogram was a graph of dyadic choices within a group. In our use, a sociogram is the composed structural reading itself: the species composition, the soil composition, the flow, and the health state that emerges from them. It is the deliverable of the method Moreno opened, applied in the way the present-day instrument allows: read from each person, composed into the team. The historical sociogram and the published Sociographer reading sit in the same intellectual tradition, ninety years apart.

The deliverable is the kind of considered reading a serious 19th-century scientific publication once delivered: the lectern, the masthead, the structural rigor, now augmented with the computational descendants of Moreno’s own method. Each report is a sociogram in this sense, and a work of editorial structural intelligence in our own.

V. A Different Question

Sociometry is easily mistaken for two instruments it sits beside but does not replace.

A personality test answers who is this person on their own, and answers it usefully. Sociometry asks a different question: how does this person compose with a team? The individual is fully present in the reading; they are simply never read in isolation, because the unit sociometry measures is the combination, not the person alone.

This holds for one person as much as for a team. A single person is already a structure: a species pair, a soil they are planted in, energy moving through them. The Sociographer reads one person as readily as it reads twelve. A single reading is a real mirror of your species, your fit, and whether your energy is rising or draining, long before any team is in the room.

An engagement survey answers how does the team feel right now, a real signal with its place. The Sociographer reads the layer beneath it: the structural conditions that produce the feeling, often months before sentiment registers the change. We do ask people about their own experience, but as one structured input into a reading of composition, soil, and energy, rather than as a sentiment score averaged over time.

Both are worth asking. The Sociographer asks the structural one.

VI. The Founding Claim

He was saying, in the precise language of his era, that the science of teams had to be rebuilt from scratch, because the methods imported from psychology, statistics and operational management were not equipped to read what was structurally happening in human groups. He was right. He was forgotten. The field he founded was absorbed into the network science that now underlies the largest information systems on Earth, while its original organisational purpose was abandoned.

This publication restores it.