Implicature and literary texts (research digest)


Reading time: 9 minutes

Poetic effects

This post digests Adrian Pilkington’s influential paper on ‘poetic effects’, which shows how ideas from pragmatics can be used in exploring literary texts and in accounting for literary interpretations. The paper refers to Seamus Heaney’s poem ‘Digging’.

Pilkington focuses on the notion of ‘implicature’, originating in the work of H. Paul Grice. Grice coined the term to refer to what we indirectly communicate. We can illustrate this using an exchange discussed in the paper by Pilkington:

Alan:                    Drink?

Beth:                    I’m a Mormon.

Implicature:     Beth does not want an alcoholic drink.

Here Beth directly communicates that she is a Mormon and indirectly communicates (‘implicates’) that she does not want an alcoholic drink. This follows as long as Alan access the following contextual assumption:

Mormons do not drink alcoholic drinks.

Grice developed an approach which aimed to explain how implicatures are communicated. Pilkington’s paper focuses on more recent ideas, developed by Dan Sperber and Deirdre Wilson in their work on relevance theory. They suggest that utterances typically communicate a range of implicatures with varying amounts of evidence for each one. While Grice’s approach would have focused on the implicature that Beth does not want an alcoholic drink, Pilkington points out that Beth’s utterance also makes it possible to infer further conclusions, including:


Beth does not want to drink other alcoholic drinks.

Beth does not drink coffee.

Beth does not smoke.

Beth might practise polygamy.

These depend on Alan accessing and using the assumptions that Mormons don’t drink coffee, don’t smoke, and that they can be more likely to practise polygamy. There are other possible conclusions which Alan might arrive at (depending on whether appropriate contextual assumptions occur to him) and which he is likely to be less sure of, such as:

Possible implicatures:

I will not have much in common with Beth.

Beth might be offended by my way of living.

These are weaker implicatures than those about coffee, smoking and marriage, which are in turn less strong than the implicature that Beth would not like an alcoholic drink. In fact, it might be hard for Alan to decide whether Beth intends to convey an assumption about her attitude to Alan’s way of life at all. Two key ideas developed within relevance theory are: first, that implicatures can be stronger or weaker; second, that there is not always a clear distinction between intentionally communicated implicatures and conclusions for which the addressee is wholly responsible.

Pilkington goes on to consider how weak implicatures can contribute to what have been termed ‘poetic effects’, which are often thought of as vague, impressionistic and hard to capture. Relevance theorists have suggested that poetic effects involve the communication of a relatively wide range of relatively weak implicatures. Pilkington discusses the effects of repetition in the Biblical utterance:

Oh Absalom, my son my son!

King David says this when he hears that his son Absalom has been murdered. The repetition encourages thoughts about David’s relationship with his son and his emotions on hearing of his son’s death. Pilkington suggests that this repetition gives rise to poetic effects because we can access a wide range of assumptions about their relationship, how it might feel for David to have lost his son, and so on. By contrast, repetitions such as the following will seem pointless or ridiculous in most contexts because we cannot access such a rich set of possible implicatures:

Oh Anna, my colleague, my colleague!

The pubs have closed, closed.

The relevance-theoretic account suggests that poetic effects arise when an utterance communicates a rich set of weak implicatures.

Pilkington’s discussion of ‘Digging’ focuses in particular on the ending of the poem. He suggests that this gives rise to a fairly wide range of implicatures because the rest of the poem has made this possible. The poem is ‘spoken’ by a poet (who clearly has a lot in common with Heaney himself) sitting at a table near a window and writing. The poem begins:

Between my finger and my thumb

The squat pen rests; snug as a gun.

Outside, his father is digging in the garden. The poem leaps back in time to when his father was younger and digging in potato fields. His father ‘could handle a spade’ and comes from a long line of men who dug the land. The poem contrasts the father with the poet who has not joined in this tradition. The poem ends back at the window:

Between my finger and my thumb

The squat pen rests.

I’ll dig with it.

What inferences do readers make at the end of the poem? Pilkington focuses on the key metaphor evoked by the use of dig as something the poet will do with a pen. He suggests that the earlier mention of the father and his digging in potato fields, along with other contextual assumptions, make it possible for us to access a wide range of relatively weak implicatures. Pilkington cannot come close to listing all of the possibilities but he lists a small set as a starting point, including these:

Digging is how the poet’s forefathers earned their living

Digging is an activity with a long tradition in the community.

Digging involves hard work.

Digging involves intense concentration.

Digging is a manly occupation.

Digging is a worthy occupation.

Digging involves taking things from below the surface.

For each one, he suggests, we can go on to think about ways in which writing poetry is similar or different to the kind of digging his father was engaged in. This poem is successful because the effort involved in thinking about the broad range of implicatures it suggests is rewarded by the relatively rich interpretations this leads to. He also considers how deriving this set of weak implicatures might lead to emotional responses and to what extent this approach might account for more broadly aesthetic experiences.

A key feature of the paper is that it shows how ideas from pragmatics are useful in accounting for literary interpretations and in exploring literary texts. Exploring the implicatures of texts is a fairly accessible way of beginning to explore texts and interpretations of texts more fully.

This is a digest of the following article:

Pilkington, Adrian. 1992. Poetic Effects. Lingua 87.1: 29-51.

You can read ‘Digging’ and hear Seamus Heaney reading it at the Poetry Foundation website here.

Using this in teaching and learning

You can use the ideas in this paper in the following ways to develop understanding of pragmatics and literary texts:

  1. The notion that implicatures can be stronger or weaker is a key notion in relevance theory. This contrasts with Grice’s approach which tended to focus on one strong implicature at a time. Students can explore this contrast by looking at short texts and analysing them in three stages:

a. identify one strong implicature of the text

b. make a list of other implicatures which might follow from the same text

c. for each implicature, identify contextual assumptions which give rise to the implicature

This activity could lead to broader discussion of what kinds of texts are more likely to be read for a wider range of weaker implicatures and which are more likely to be read for a narrower range of stronger ones (see activity 4 below).

  1. Students can develop their own spoken and written communication by rewriting utterances to adjust the likelihood of particular inferences being made by hearers or readers of their utterances. This might also focus on the extent to which different versions give rise to stronger or weaker implicatures. They could explicitly explore the effect of this in their recreative writing (Writing about society, A level; People and places, AS level)’
  1. Students can explore the novel and poems they are studying (Imagined worlds and Poetic voices A level; Views and voices, AS level) by making a list of possible implicatures which they think the text suggests and organising them into stronger or weaker implicatures. Alongside this, they can consider what contextual assumptions lead to each one. Which conclusions depend only on assumptions derived from the text? Which derive from more general contextual assumptions? Do some of them depend on knowledge about the author? Or the context in which the text was produced?
  1. Students can aim to develop richer or shallower interpretations of particular texts across the specification and on the NEA (Making Connections, A-level) by focusing to a greater or lesser degree on stronger or weaker implicatures. A fairly shallow interpretation of a text might focus mainly on ‘what happens’ in the text. A richer one would focus on broader themes, the psychology of characters, the emotional impact on audiences, and so on. Students can reflect on this explicitly by comparing different kinds of interpretations of texts and different kinds of reading practices. What would a very shallow interpretation of a particular text look like? What would a much richer one look like? Are there texts which are more likely to give rise to richer interpretations than others?
  1. Students can also compare different parts of particular texts and alternative possible versions of the same text. This might inform the study of any text on the AS and A level specifications. Pilkington suggests that the connection between the pen and a gun at the start of ‘Digging’ (where the pen rests ‘snug as a gun’) does not lead very far and so has a much less rich interpretation than the notion of the pen as something to dig with at the end. Students could explore this in greater detail, listing possible implicatures of both parts. Alternatively, they might consider other utterances which lead to weaker interpretations. Pilkington suggests, for example, that the simple sentence ‘The poet digs with his pen’ would be far less evocative or interesting than Heaney’s poem. Students could discuss why this is so.

Reading suggestions

There has been a huge amount of work on pragmatics since Grice developed his ideas over 50 years ago (his key texts began to be shared in the 1960s and published in the 1970s but some of the work predates this). Here are some sources on pragmatics in general, on relevance theory in particular, and on pragmatics and literary interpretation.

Chapman, Siobhan. 2011. Pragmatics. Palgrave Macmillan, Basingstoke.

An accessible introduction to pragmatics, covering Grice’s approach and other approaches developed from that, including discussion of relevance theory.

Blakemore, Diane. 1992. Understanding Utterances. Wiley-Blackwell, Oxford.

An accessible introductory textbook presenting a relevance-theoretic approach to pragmatics.

Clark, Billy. 2013. Relevance Theory. Cambridge University Press, Cambridge.

A more comprehensive accessible account of relevance theory, including discussion of how it developed from the work of Grice.

Dan Sperber’s website, here.

This site provides access to downloadable versions of a wide range of work by Sperber, including work with Deirdre Wilson.

Relevance Theory Online Bibliographic Service, here.

This site gathers all academic sources which discuss or apply ideas from relevance theory. This is probably not the most helpful place to find sources when beginning to study pragmatics but it is interesting to see just how much work has been published on this approach to pragmatics.

Pilkington, Adrian. 2000. Poetic Effects. John Benjamins, Amsterdam.

A book-length study of poetic effects from the perspective of relevance theory. Develops further the ideas in the paper discussed in this digest, including the limits of how far this approach can go in accounting for emotional and affective responses to literary and artistic works.

Chapman, Siobhan and Billy Clark (eds.) 2014. Pragmatic Literary Stylistics. Palgrave Macmillan, Basingstoke.

A collection of chapters applying ideas from pragmatics in accounting for interpretations of literary texts. It includes an introduction with general comments on different approaches to pragmatics and their relevance in approaches to literary texts.