Variants of incongruity resolution

Graeme D Ritchie

Research output: Contribution to journalArticlepeer-review


For some time now, there has been widespread support for the idea that incongruity resolution (IR) is at the heart of much, perhaps all, humour. However, the two central terms, »incongruity« and »resolution«, are as yet unclear. This means that there could be a number of different IR theories, depending on the exact definitions adopted for these terms. The dimensions of variation between the possible theories can be organised by analysing the terms »incongruity« and »resolution«, and the ways in which they are put together, in greater depth.

The first of these aspects is the scope: how much of humour the theory claims to cover. Some authors offer IR as an explanation of all humour, others regard IR as a subclass of humour. Which of these positions is adopted has consequences for how the theory can be tested, as it radically affects the notion of a counter-example.

A fundamental attribute of most versions of IR is that the model is sequential: that is, it characterises humour in terms of a process (in which resolution follows incongruity) rather than as a static configuration of elements. Although this is generally regarded as an essential aspect of IR, some recent suggestions raise the possibility of a non-sequential variant of IR.

Perhaps the most central area of variation between IR theories concerns the nature of the incongruity. Within a sequential IR mechanism, there are number of stages where incongruity (in the sense of a clash between two meanings) can occur. It can be embedded within the situation which the text (or image) conveys, or it can consist of a clash between two alternative interpretations of the material, or it can be defined in terms of a relation between information presented initially (in the set-up) and information presented later (the punch-line). These are not all equivalent definitions. The point here is not that incongruity can arise in a number of different ways (which may well be the case); the issue is that different theorists define incongruity in logically different ways.

It is also useful to distinguish between the way in which the incongruous situation differs from normality (what we call the deviation) and the cognitive route by which the perceiver of the humour arrives at the incongruity (the discovery). Some writings imply that one of these can be the »resolution« but neither is involved with lessening or removing the incongruity, which further blurs the notion of resolution.

There is also a lack of clarity within the literature about what exact facet of the processing of the humour constitutes the »resolution«: is it the information (in the joke or other material) on which the reasoning depends, or the general rule which the reasoning invokes, or the actual process of applying some inference? Or is it some combination of these?

Perhaps the most obviously varying feature of IR theories is the extent of the resolution: the incongruity can be eliminated (full resolution), merely mitigated (partial resolution) or not affected (null resolution). Some IR frameworks allow only full resolution, others allow partial and/or null resolution.

The IR model was primarily developed in the 1970s, with articles by Suls and by Shultz being highly influential. This work can be seen as setting out the traditional IR model, which is highly sequential, and typically involves full resolution. More recent proposals (including the »general theory of verbal humour«) are non-sequential, with only partial resolution.

Despite this lack of agreement amongst theoreticians, experimental studies of psychological and neurological aspects of humour sometimes assume that there is an unproblematic notion of IR theory, usually treating IR as a subclass of humour involving sequential processing. Even in these cases, it is apparent (from the way that the investigators analyse the texts which they use in the tests) that different authors can have quite different views on what defines the IR model.

A further complication is that the 3WD test – a prominent psychological classification of people's reactions to humour – uses two categories which are sometimes glossed as corresponding to IR and non-IR (nonsense) humour. However, the design of the 3WD test says only that there seem to be two clusters of humorous items which people react to differently; it does not provide a formal analysis of how these classes of humorous items differ, and the two classes can, strictly speaking, overlap.

There appears to be a widely accepted informal notion of IR, but in view of the variation in versions of this framework, it is not clear that all who espouse this terminology are actually in agreement on the substantive details. If we try to extract a minimal essence which is present in all IR proposals, then we arrive at something like the following: All humour involves some degree of incongruity, but this incongruity is not random or arbitrary – it is systematically related to other aspects of the setting.

Although few writers on IR would disagree with this, it has the methodological disadvantage of being a very weak statement, not claiming a great deal about the nature of humour.
Original languageEnglish
Pages (from-to)313-332
Number of pages20
JournalJournal of Literary Theory
Issue number2
Publication statusPublished - 2009


  • humour theory


Dive into the research topics of 'Variants of incongruity resolution'. Together they form a unique fingerprint.

Cite this