Nothing Happens for Just One Reason.

When something goes wrong — or right — the first question we ask is almost always some version of: what caused this?

It is a natural question. It feels like the responsible question. If we can identify the cause, we can address it, replicate it, or prevent it. Causation is the grammar of action: find the cause, control the cause, change the outcome.

The problem is not with the question. The problem is with the singular. In almost every situation of meaningful complexity, there is no single cause. There are causes — multiple, interacting, mutually reinforcing or counterbalancing — and the relationship between them is not additive but dynamic. The outcome is not the sum of the causes. It is the product of their interaction.

This distinction matters enormously, both for how we understand what has happened and for how we decide what to do next.

The Comfort of the Single Cause

Before examining why single-cause thinking fails, it is worth understanding why it persists — because it is not simply laziness or lack of rigour. It reflects something genuine about how human cognition works.

The mind craves narrative coherence. A story with a single cause has a shape that is easy to hold — a clear villain, a decisive moment, a turning point that explains everything that followed. Multi-causal explanations are harder to hold. They require simultaneously tracking several threads, understanding how they interact, and resisting the pull toward resolution that a single explanation provides.

Single-cause thinking is also deeply practical in its appeal. If there is one cause, there is one fix. The path from diagnosis to action is short and direct. Multi-causal understanding, by contrast, tends to produce a more complex and less satisfying prescription: several things need to change, their interactions need to be managed, and the effects will be partial and uneven.

Finally, single-cause explanations are socially useful in ways that multi-causal ones are not. They assign responsibility clearly. They support accountability narratives. They allow organisations and individuals to close the loop on what happened and move forward. Multi-causal explanations, which distribute responsibility across many factors and resist clean resolution, are harder to act on institutionally — even when they are more accurate.

All of this explains why single-cause thinking is so widespread and so persistent. But explaining why a cognitive habit is appealing is not the same as justifying it. And in complex situations, the habit of seeking a single cause leads, with considerable regularity, to misunderstanding what actually happened and therefore to decisions that address the wrong things.

What Interaction Actually Means

The claim that outcomes are produced by the interaction of multiple causes is easy to state and surprisingly difficult to absorb in practice. It helps to be precise about what interaction means.

Two causes interact when the effect of each depends on the presence or state of the other. This is different from two causes that both contribute independently to an outcome. When causes interact, the whole is not the sum of its parts — removing one cause does not simply reduce the effect proportionally. It can eliminate it entirely, reverse it, or produce an entirely different effect from what either cause would have produced alone.

Consider a straightforward example. A business initiative fails. The analysis identifies two contributing factors: an underfunded implementation and an unfavourable market environment. If these factors were simply additive, addressing either one partially would improve the outcome proportionally. But if they interact — if the market environment meant that even a well-funded implementation would have struggled, or if the underfunding was so severe that it would have failed even in a favourable market — then the analysis that treats them independently misses the actual dynamic.

In practice, the interactions between causal factors in significant situations are rarely this clean. They involve feedback loops, where an effect becomes a cause of further effects. They involve threshold effects, where a factor that was dormant becomes decisive once other conditions cross a certain point. They involve time lags, where causes and effects are separated enough in time that the connection is easy to miss.

These dynamics are what make complex situations complex — not the number of factors involved, but the nature of the relationships between them.

Why Root Cause Analysis Often Misleads

The concept of the "root cause" — the one underlying cause that, if addressed, will resolve everything downstream — is deeply embedded in how organisations approach problems. Root cause analysis is a standard tool in quality management, incident response, and strategic diagnosis. It has real value in bounded, well-understood systems where causal chains are relatively simple and linear.

In complex human systems, however, the root cause framework regularly misleads. Not because causation does not exist, but because the assumption of a single root — a primary cause from which all others derive — rarely matches the actual structure of how things happen.

When an organisation underperforms, is the root cause poor leadership? Misaligned incentives? Cultural dysfunction? Inadequate capability? Unfavourable competitive dynamics? Each of these factors is real. Each contributes. Each also affects and is affected by the others. Identifying any one of them as the root cause is not rigorous analysis — it is a choice about which thread to pull, often shaped by the analytical framework we brought in, the data that happened to be available, or the organisational dynamics that make certain explanations more acceptable than others.

The deeper problem is that root cause analysis, by design, directs intervention toward a single point. If the actual situation is produced by the interaction of several factors, an intervention at any single point will address part of the dynamic — but the remaining factors will continue to generate effects, often compensating for or undermining the intervention. Progress stalls. The problem reasserts itself. The analysis is declared incomplete, and the search for the real root cause begins again.

This cycle is familiar to anyone who has spent time in complex organisations. It is often less a failure of intelligence than a failure of the analytical framework being applied.

Confluence, Not Causation

A more useful way to think about complex outcomes is through the lens of confluence — the coming together of multiple streams.

Confluence does not require that any single factor be sufficient to produce the outcome. It requires only that a particular combination of factors, at a particular moment, creates the conditions in which the outcome becomes possible or inevitable. Change any element of that combination — remove a factor, alter its intensity, shift its timing — and the outcome may be entirely different.

This way of thinking has several practical implications.

It changes how you investigate what happened. Instead of asking "what caused this?" you ask "what combination of conditions made this possible?" The investigation becomes less about finding the one thing to blame and more about mapping the landscape of contributing factors and understanding how they related to each other.

It changes how you think about prevention. If an outcome requires a confluence of factors, preventing it does not necessarily require addressing all of them. It may require only disrupting one element of the combination — the most accessible one, the most changeable one, the one whose removal would most significantly alter the dynamic. This is often a more efficient intervention than attempting to address the "root cause," which may be deeply entrenched or only partially within your control.

And it changes how you think about replication. If a positive outcome was the product of a particular confluence of factors, reproducing it requires reproducing that confluence — not just the most visible element of it. The initiative that succeeded brilliantly in one context may fail in another not because it was poorly executed but because the surrounding conditions that made it effective are not present.

Reading the Interaction in Real Time

The most demanding application of multi-causal thinking is prospective rather than retrospective — not understanding why something happened, but reading how multiple forces are currently interacting and where their combined effect is likely to lead.

This is demanding because interactions are dynamic. The relationship between factors changes as circumstances change. A factor that was dormant becomes active. A combination that was stable becomes unstable. The trajectory of a situation is not determined by any single force but by the evolving relationship between all of them.

What this requires, in practice, is a habit of holding multiple threads simultaneously — tracking not just each factor but the space between them, attending to how they are affecting each other, noticing when a relationship that was stable is beginning to shift. This is genuinely difficult to sustain, partly because it resists the natural pull toward simplification and partly because it requires tolerating a degree of irreducible uncertainty that single-cause thinking, for all its limitations, does not.

But the reward for developing this capacity is significant. Those who can read the interaction of forces in a complex situation — who can hold the moving parts in view simultaneously and sense where their combined momentum is heading — are in a genuinely different position from those who are tracking only the most prominent single factor. They see more, earlier. They anticipate developments that others encounter as surprises. They make interventions that address the actual dynamic rather than its most visible symptom.

A Different Kind of Honesty

There is, finally, something worth saying about the relationship between multi-causal thinking and intellectual honesty.

Single-cause explanations feel more confident. They speak in declaratives. They assign responsibility clearly. Multi-causal explanations, by contrast, require a willingness to say: this is complicated, several things were operating simultaneously, the relationship between them is not fully clear, and any account of what happened is necessarily partial.

This kind of honesty is harder to produce and harder to receive. It does not satisfy the desire for resolution that a clear causal story provides. In many organisational and social contexts, it can even feel like evasion — as if the refusal to name a single cause is a refusal to take responsibility.

But the alternative — naming a single cause with false confidence — produces decisions that are built on a misunderstanding of what actually occurred. And decisions built on misunderstanding, however confidently made, tend to encounter reality on reality's terms.

The world is not organised around our preference for simple stories. Understanding it requires the willingness to hold complexity without prematurely resolving it — and to act well, nonetheless.