Green wrote a powerful article for the American Law Review (edited by Holmes), the first published in his collection Essays and Notes on the Law of Tort and Crime, entitled "Proximate and Remote Cause", the former said to determine its effect, the latter merely being necessary but not sufficient for the effect to occur.

Green debunked the popular idea of a single "chain of causation" and the notion of a single proximate cause as just a rationalization. He argued that there are always multiple causes for events. People simply choose a proximate cause from the multiplicity of possible causes to serve their purposes and interests in assigning responsibility.

Causation is the law of cause in relation to effect. Nothing more imperils the correctness of a train of reasoning than the use of metaphor.

There is no single "chain of causation"

By its over free use the subject of causation has been much obscured. The phrase "chain of causation," which is a phrase in frequent use when this maxim is under discussion, embodies a dangerous metaphor. It raises in the mind an idea of one determinate cause, followed by another determinate cause, created by the first, and that followed by a third, created by the second, and so on, one succeeding another till the effect is reached. The causes are pictured as following one upon the other in time, as the links of a chain follow one upon the other in space. There is nothing in nature which corresponds to this. Such an idea is a pure fabrication of the mind.

There is but one view of causation which can be of practical service. To every event there are certain antecedents, never a single antecedent, but always a set of antecedents,..

There are always multiple causes
for an event

From every point of view from which we look at the facts, a new cause appears. In as many different ways as we view an effect, so many different causes, as the word is generally used, can we find for it. The true, the entire, cause is none of these separate causes taken singly, but all of them taken together. These separate causes are not causes which stand to each other in the relation of proximate and remote, in any intelligible sense in which those words can be used. There is no chain of causation consisting of determinate links ranged in order of proximity to the effect. They are rather mutually interwoven with themselves and the effect, as the meshes of a net are interwoven. As the existence of each adjoining mesh of the net is necessary for the existence of any particular mesh, so the presence of each and every surrounding circumstance, which, taken by itself we may call a cause, is necessary for the production of the effect.

In this view of causation there is nothing mysterious. Common people conduct their affairs by it, and die without having found it beyond their comprehension. When the law has to do with abstract theological belief, it will be time to speculate as to what abstract mystery there may be in causation ; but as long as its concern is confined to practical matters it is useless to inquire for mysteries which exist in no other sense than the sense in which every thing is a mystery.
(Proximate and Remote Cause, pp.11-13

Green left Harvard Law School in 1873, objecting to its formalism, to become a dean at the new Boston University Law School.