W. H. Walsh on colligatory concepts
The term 'colligation' derives from the first self-identified philosopher of science William Whewell. In his Philosophy of Inductive Sciences (1847), Whewell develops a more sophisticated account of induction than that of simple enumeration, prevalent in the years since Francis Bacon. In this process 'colligation' has a central role. 'Colligation' is an 'act of thought', which brings a number of empirical 'facts' together by 'superinducing' upon them a conception that integrates and makes them in this way capable of being expressed by a general law. Colligation provides the 'true bond of Unity by which the phenomena are held together', writes Whewell (Whewell 1847, 46-48; see also Snyder 2012). An example could be the known data points of the orbit of Mars and their colligation under the conception of an elliptical curve. We might also note that while Whewell saw that one may choose among several colligatory concepts, he insisted that their choice is not arbitrary. Colligations are chosen through a 'special process in the mind' that involves some kind of inference or inferences (Snyder 2012).
W. H. Walsh borrowed the term 'colligation' from Whewell and applied it in the philosophy of history and historiography. The earliest paper that Walsh wrote about colligation in historiography is his 'The Intelligibility of History' from 1942. The process of colligation is part of an attempt to discover intrinsically related coherent wholes from historical 'facts'. Similarly to Whewell's conception, the historian '"colligates" different events according to "appropriate conceptions"' (1942, 133), the functioning of which can be illustrated by historical phrases such as the 'Industrial Revolution' and the 'Enlightenment'. In his Introduction
to Philosophy of History, Walsh explicitly relates the concept of colligation to the issue of explanation in history. When the historian is asked to explain a specific phenomenon, such as the British general strike of 1926, the historian tries to trace connections between that event and others with which it stands in an 'inner relationship'. The underlying assumption is thus that different historical events can be seen as 'going together to constitute a single process, a whole of which they are all parts and in which they belong together in a specifically intimate way' (Walsh 1958, 23; similarly 59-62).
In his subsequent critical comments in 'Colligatory Concepts in History', Walsh remarks that his earlier use of 'colligation' was part of the attempt to find a plausible version of an idealist theory of history. The concept was used to integrate historical events that are not 'externally connected' (Walsh 1974, 134; but cf. Walsh 1942, 134) via an explanation that sees an agent or a group of agents pursuing a long-term policy over a long period of time, creating an 'internal link' between diverse phenomena. 'Colligation', says Walsh now, refers to an essential part of the historian's interpretative process in which the historian is confronted with a mass of material that seems unconnected at first sight - but the historian 'goes on to show that sense can be made of it by revealing certain pervasive themes or developments' (Walsh 1974, 136; my emphasis).
How can a historian reveal 'pervasive themes or developments'? According to Walsh, the historian specifies what is significant in events by identifying those aspects that point beyond themselves and connecting them with other events 'as phases in a continuous process' (Walsh 1974, 136). Colligating is to organize, and Walsh suggests that this is something that any type of historiography just has to do (Walsh 1974, 137).
The recognition of the subject-sidedness of colligation is important. Especially in his later writings, Walsh emphasizes the interpretative and subjective nature of colligation, as opposed to Whewell, who seemed to think that facts can be colligated correctly and naturally, that is, objectively, on the basis of induction. For Walsh colligation and the principles it implies mean that the historian adds something 'non-objective' to the historical reality.