Updated: Sep 15, 2020

Causality describes a relationship that exists between two or more things where a change in one thing – or set of things – causes a change in another.[1] The essence of causality is a phenomenon being dependent on some other effect. As such causality is a connection or linkage between states or events through which one thing – the cause – under certain conditions gives rise to or causes something else – the effect.[2]

Causality forms a fundamental and pervasive part of our perception and interpretation of the world around us. Equally, our ability to act in the world depends on our grasp of causal relationships among things; the ways they act and interact. In everyday life humans – and other animals – rely on the assumptions of causality literally every waking second; in academia, much of science is the study of systematic cause and effect relations.


The philosopher David Hume argued there are no necessary connections between events in this world, in his words “all events seem entirely loose and separate. One event follows another, but we can never observe any tie between them. They seem conjoined but never connected.” In his book An Enquiry Concerning Human Understanding he expands upon this writing “That after a repetition of similar instances, the mind is carried by habit, upon the appearance of one event, to expect its usual attendant, and to believe that it will exist. This connexion, therefore, which we feel in the mind, this customary transition of the imagination from one object to its usual attendant, is the sentiment or impression from which we form the idea of power or necessary connexion. Nothing farther is in the case. Contemplate the subject on all sides; you will never find any other origin of that idea.”[3]


Perceived causality involves the process of inductive inference. The basic process of induction is one of inference from a set of things that have something in common to generalizing what we observe about them as being true to all instances of that kind. We generalize from a sample of previous experiences to a whole population. We project past uniformities in our experience onto future events through expectation. We have seen a ball move off in the opposite direction every time it has been hit with a bat in the past, so we assume it will be the same in the future and this assumption through inductive inference creates the perception of there being a causal relationship between them. We perceive cause and effect and then extrapolate from that to infer cause and effect relations about things that we can not see such as black holes, quarks or events in the future. Of course, when we generalize we are going beyond the evidence; by the definition of what we are doing, generalizing. Our conclusion covers things that we have not or can not know. The rising and setting of the Sun in the past can not guarantee it will rise and set again tomorrow.

The philosopher Immanuel Kant reconciled this insight by positing that our mind contains certain primordial categories, such as causality, which contextualize and condition events to create the impression of the causal relationship between things being real. As such causality is a subjective layer that we place on events in the world. We do not perceive the world in frames of isolated events but in fact, perceive it instead as one continuous event through processes of cause and effect.

Axiom of Causality

The Axiom of Causality is the proposition that everything in the universe has a cause and is thus an effect of that cause. This means that if a given event occurs, then it is the result of a previous, related event. If an object is in a particular state, then it is in that state as a consequence of another object interacting with it previously.[4] Plato states this in Timaeus when he writes “In addition, everything that becomes or changes must do so owing to some cause; for nothing can come to be without a cause.”

Establishing Cause and Effect

Determining cause and effect is one of the primary processes for generating new “knowledge.” Thus the central goal of most scientific research is the identification of causal relationships or demonstrating that a particular independent variable – the cause – has an effect on the dependent variable of interest – the effect. The three criteria for establishing linear cause and effect are correspondence, time precedence, and non-spuriousness.[5]


The first step in establishing causality is demonstrating correspondence or association. This means asking the question is there a relationship between the independent variable and the dependent variable? Correspondence or correlation means that the cause and effect occur within some unit of analysis. For example, if being exposed to cold is more likely to make you sick then the people who are exposed to cold should be more often ill.


Time precedence means that the cause must occur before the effect. If one wants to say that being well educated causes one to earn a high salary, then the cause of being educated must precede the effect of earning a high salary.

Elimination of Alternatives

A spurious or false relationship exists when what appears to be an association between the two variables is actually caused by a third extraneous variable. This is captured in the saying “Correlation does not imply causation” a phrase used to emphasize that a correlation between two variables does not mean that one causes the other.[6] A classic example of a spurious causal relationship might be the relationship between children’s shoe sizes and their academic knowledge: as shoe size increases so does knowledge, but of course, both are also strongly dependent upon the child’s age.[7] Likewise, a false correlation might be drawn between the amount of ice cream sold and the sale of sunglasses. Again there is a hidden variable of temperature that is causing both to change together without there being a direct cause and effect relation between them.

Linearity and Nonlinearity

Conceptions of causality can be roughly divided into linear and nonlinear. Linear causality is the idea that cause and effect follow a single direction between events, from A the cause, to B the effect. Nonlinear causality is the idea that causality may follow a bidirectional path from A to B or from B to A, or even both at the same time.

Linear Causality

With linear causality, cause precedes effect in a sequential pattern. There is seen to be a direct link between cause and effect. Linear causality has a clear beginning and a clear end. There is one or a limited number of causes for any given effect; additional linkages of causes or effects create a line or pattern of domino causality.[8]

Linear causality is a keystone of the analytical, reductionist approach, whereby a closed system is defined, and linear cause and effect interactions are searched for as an explanation for how elements in the system behave. Within this paradigm cause and effect relations are seen to move in one direction from the bottom-up, and not the reverse direction; lower-level phenomena are seen to cause higher-level events. For example, in asking why the body functions as it does we would refer to the internal constituent parts of the organs and tissues to derive an upward causal relation, instead of looking for a cause within the system’s environment; which would be a form of downward causation. Linear causality leads to the conception of determinism, in that it defines a closed system and reduces the number of causes to a limited set acting in a single direction. Reductionism and linear causality try to reduce the cause of an effect to a single determinant, and the fewer the component determinants, the greater the determinism.

Nonlinear Causality

Nonlinear causality sees causation flowing in a bidirectional or multidirectional pattern. Nonlinear causality involves cyclical processes where one thing impacts another which in turn impacts the first; although this chain of events leading to feedback may be mediated through several events or over a prolonged period.[9] Nonlinear causality is part of the holistic, synthetic paradigm that looks at systems within their context or environment. As such it is much more focused on downward causation, looking at how the environment causes the behavior in the system. The holistic paradigm posits that effects can be the product of many causes. To gain a full understanding of the effect we need not drill down to find a single cause but instead look at multiple different factors and how they interact to give rise to the outcome as an emergent phenomenon.

Whereas in the linear model, the relationship between cause and effect is seen to derive from one of the components affecting another. Within the synthetic approach, the relationship between the parts is seen to create the effect. For example, from this perspective, it is not that one chemical substance causes another to react in a particular way it is, in fact, the type of relationship between them that generates the emergence of a particular type of outcome.

With nonlinear causality, cause and effect can flow in both directions through time. However, this requires some kind of control system being involved.[10] Purely physical processes result in a unidirectional flow to causality; from the past to the future. But once there is a control system involved this can define some future desired state – the goal – and then affect events in the present based upon the future. For example, whether I spend lots of money now may be contingent on whether or not I think I will get paid at the end of the week. Thus with nonlinear causality causes for events may be derived – at least partially – from the future. But this would appear to be only possible under the condition of goal-orientated behavior, where current events are controlled by projections surrounding some goal in the future.

1. (2020). Definition of CAUSALITY. [online] Available at: [Accessed 15 Sep. 2020].

2. (2020). The Principle of Causality. [online] Available at: [Accessed 15 Sep. 2020].

3. Hume, D. (n.d.). An Enquiry Concerning Human Understanding. [online] Available at:

4. Wikiwand. (2020). Axiom of Causality | Wikiwand. [online] Available at: [Accessed 15 Sep. 2020].

5. Statistics Solutions. (2020). Establishing Cause and Effect - Statistics Solutions. [online] Available at: [Accessed 15 Sep. 2020].

6. Wikiwand. (2020). Correlation does not imply causation | Wikiwand. [online] Available at: [Accessed 15 Sep. 2020].

7. Statistics Solutions. (2020). Establishing Cause and Effect - Statistics Solutions. [online] Available at: [Accessed 15 Sep. 2020].

8. Six Causal Patterns LINEAR CAUSALITY. (n.d.). [online] Available at:

9. Six Causal Patterns LINEAR CAUSALITY. (n.d.). [online] Available at:

10. Copernicus Center for Interdisciplinary Studies (2012). On the Nature of Causality in Complex Systems, George F.R. Ellis. YouTube. Available at: [Accessed 15 Sep. 2020].

Systems Innovation

  • LinkedIn
  • YouTube
  • Twitter
  • Facebook