A few years ago, several of my children asked me to explain a joke from The Reader's Digest. This is always a delicate parental moment. Even when the source is as safe as The Reader's Digest, it is still possible that one will be confronted with humour that would be difficult, if not illegal, to explain to children. Fortunately, this joke was clean. It went like this:

What were my children confused about? After all, they understood each of the words in the sentence and understood the sentences as sequences of words. Had I asked them to draw me a picture, they could have done so. Some of them caught the frame-breaking humour of a bartender self-aware enough to know he is in a joke. But they did not understand why he would draw that conclusion in the first place. As I reflected on the reasons for this, it became obvious that this joke depended for its humour on a convergence of two joke traditions: Jokes about diverse religious figures or professionals on the one hand, and jokes about bar room conversations on the other. The bartender was clearly well-informed about these joke traditions, so he suspected what was happening.Footnote ² My children lacked that information, and missed the point completely.

This insight into the workings of humour was reinforced and generalised when I watched the animated movie, Shrek, a movie that I now tell my students is a gold mine of hermeneutical insight. All the funny parts of that film assume that the viewer has information that the movie does not provide, information from three main sources: nursery rhymes, fairy tales and popular culture, especially movies. During one scene, for instance, the evil Lord Farquaad is torturing the gingerbread man in his dungeon – dunking him in milk, breaking off his legs, threatening to remove his gumdrop buttons. Finally, he gets the gingerbread man to talk, and this exchange ensues:

In another part of the film, Shrek fights his way into a castle tower to rescue the beautiful Princess Fiona, who is being held prisoner by a dragon. Shrek's commotion wakes the Princess, who speaks in stilted fairy-tale English, wonders why she has not been kissed and is horrified to learn that the dragon is still alive: ‘You haven't killed the dragon?’ she asks. ‘It's on my to-do list’ is Shrek's response. Finally, when Shrek and Fiona are waylaid by a group of mincing merry men in the forest, we suddenly learn that Fiona has had some martial arts training. During the fight she leaps into the air and the action stops as the ‘camera’ rotates slowly around the scene, an allusion to the film techniques of The Matrix. At the end of the film the wedding scene echoes the wedding scene of the Disney film, Beauty and the Beast. Fiona is under a curse that turns her into an ogress every night, but she hopes that when she receives true love's kiss she will be released from the curse and remain a beautiful princess forever. When Shrek kisses her, she begins spinning and flies up into the air raining sparks, just as the Beast does in the Disney film. When the Beast comes back down to earth he has been transformed into a handsome prince. In one of the most delicious subversions in the film, when Fiona descends, she is still an ogress. This is only a small sampling, but the point is clear: Shrek is impenetrable unless the viewer comes armed with a cache of nursery rhymes, fairy tales, and recollections from pop culture.Footnote ³ A viewer ignorant of these resources does not miss some marginal features of the film. He misses the entire meaning. He doesn't get it.

Like the jokes in Shrek, I submit, every text depends for its meaning on information lying outside the text. This is true in the trivial sense that no text defines all its terms for the reader. At least, the author intends for the reader to bring a basic understanding of the language to his reading. Beyond that, and even if meaning is defined rigidly in terms of authorial intent, meaning is never confined within a single text, but always arises, in small or large degree, from the intertextual relations between the focus text and other texts. Every text is a joke, and all interpretation is a matter of ‘getting it’. All texts mean the way jokes mean. Humour thus provides a vantage from which to address various issues in contemporary hermeneutics.

I have used the term ‘intertextual’, and before moving on I need to define more precisely what I mean. Because using the term ‘intertextuality’ might register my agreement with Derrida's claim that ‘there is nothing outside the text’, let me make it clear that, in this paper, I neither defend nor critique that position. Except when summarising other writers’ views, I will avoid the term in order to side-step the complicated questions that Derrida poses. Further, the relations that constitute the meaning of a text are not only relations to other texts but to extra-textual realities. It is confusing to describe a poet's evocation of, say, the Second Battle of the Marne as a piece of ‘intertextuality’ if the reference is to extra-textual experience (i.e. the poet fought in the battle). In place of the term ‘intertextuality’, I will talk about ‘porous texts’ or ‘permeable texts’. Individual texts have their own integrity and uniqueness, but a good reader will find his attention being directed outward to other texts and will find other texts penetrating and leaving traces on the text he reads. ‘Porous text’ is more felicitous than the popular ‘open text’ because it emphasises both that texts have boundaries and that these boundaries are permeable.

In this paper, I examine one large question from the perspective that texts mean in the same way that jokes mean; that issue is the connection between authorial intention and the permeability of texts. Are the two compatible? Does an announcement that textual boundaries are porous amount to an obituary for the author? The argument develops in several stages. First, I examine Kevin Vanhoozer's treatment of authorial intention and his critique of intertextuality. Though I have reservations about various aspects of Vanhoozer's theory, here I largely accept it for the sake of argument. Then, drawing on the work of Jonathan Culler, I argue that even Vanhoozer's strong view of authorial intention is compatible not only with the recognition that texts are permeable, but with the stronger claim that authors are able to use texts to communicate only because texts are porous. I close by noting several other issues that might be illumined by reflections on the hermeneutics of humour.

The death of the porous text

In recent years Kevin Vanhoozer has offered one of the most sophisticated, thorough, and vigorous responses to the postmodern ‘undoing’ of the author, the text and the reader.Footnote ⁴ At the heart of Vanhoozer's ‘realist hermeneutic’ is a proposal for restoring authorial intention to a central place in the theory of meaning and the practice of interpretation. Postmodernism announces the death of the author, and once that anchor is loosed, Vanhoozer argues, the death of the text and the death of the reader quickly follow. In his effort to ‘resurrect’ the author after his postmodern postmortems, Vanhoozer rejects a Humpty-Dumpty relativism, where authorial intention trumps public linguistic conventions, and also rejects psychological accounts of intention, which assume that the interpreter is capable of a kind of mind-meld with the author. At the same time, he combats the depersonalisation of authorship and interpretation so pervasive in structuralist and poststructuralist theory.

Vanhoozer steers clear of these errors through an appeal to speech-act theory.Footnote ⁵ Drawing on the work of J. L. Austin, John Searle and, more recently, William Alston, he argues that the text is an act and that its meaning is what the agent who produced the text intended by the action. Meaning is thus not precisely ‘in’ a text, but is something done by an agent. Taking note of intention is the only way to determine what an action is to ‘count as’. A twitch of the eye during a game of bridge might be taken as a wink, and only an investigation of the intention of the winker can determine whether it counts as an illicit signal to a partner or an innocent physical foible. In a text, the author's intention determines what the text is to ‘count as’: ‘An author intends his or her text as a particular kind of communicative act, with a particular object and a particular force’.Footnote ⁶

Vanhoozer (characteristically) develops this point through a series of puns on ‘intend’. An author means by ‘tending’ his words in a particular fashion. By describing the author's work as a form of ‘tending’, Vanhoozer hopes to ‘catch both the sense of “inclining” or “straining towards” on the one hand, and “taking care of” on the other. What does an author do in tending to words? Just this: the author intends both an object (e.g., the propositional content to which the author's activity is directed) and a disposition (e.g., the stance an author takes with regard to the object)’.Footnote ⁷ Interpreting authorial intention demands ‘attending’ to how the author ‘tended’ his words, and discovering authorial intention is the point of interpretation. Thus, ‘the author, lost as Cartesian thinking subject, thus returns as communicative agent – one who means, one who puts a language system and a literary form to work in a particular way for a particular purpose’.Footnote ⁸

Vanhoozer believes his account of authorial intention is able to stop up the holes in the porous text of postmodern theory and to establish a text's ‘single, determinate meaning’.Footnote ⁹ He assumes that establishing authorial intention as the foundation for stable textual meaning is incompatible with any strong notion of intertextuality.Footnote ¹⁰ Since the term was coined by Julia Kristeva, ‘intertextuality’ has been used in a wide variety of senses.Footnote ¹¹ On one side of the spectrum, intertextuality is little more than a rhetorical device of whispered allusion or metalepsis, and an interpreter's commitment to intertextuality might manifest itself as an effort to trace influences on, and sources of, a particular text.Footnote ¹² At the other end, the term implies a radical critique of supposedly traditional understandings of textuality and interpretation, decentring the text, challenging its autonomy, self-consistency and self-presence and deconstructing the ‘illusion’ of fixity. In this sense, ‘intertextuality’ is of a piece with the postmodern decentring of the Cartesian and Kantian self. By the latter definition, intertextuality, like many other trends in poststructuralist hermeneutics, wrests the text from the hands of authors to place it in the hands of readers. A theory of intertextuality might be used to justify something like the whimsical hermeneutics of Borges, who proposed a ‘new technique of the deliberate anachronism and the erroneous attribution’ and argued that reading the Odyssey ‘as if it were posterior to the Aeneid’, or attributing the Imitatio Christi to Celine or James Joyce, or interpreting Quixote as the work of (the fictional) Pierre Menard would fill ‘even the most placid works with adventure’.Footnote ¹³

Vanhoozer deals with the more radical account of intertextuality. He begins his attack with a summary of Ricoeur, about whom he has written a book-length study. For Ricoeur, intertextuality is a particular species in the genus metaphor: ‘Two texts that have no literal or logical connection are nonetheless seen together, and as a result of this dynamic interaction new meaning is produced – ‘resemblance’ is created’. Metaphor brings two semantic fields into connection, while intertextuality brings together two ‘textual fields’ in a way that creates ‘a new context in which to read a text’. On Vanhoozer's account of Ricoeur, intertextuality implies that ‘texts are open – open to the effects of past texts and to the context of present readers’. For Vanhoozer, Ricoeur's intertextuality is basically equivalent to Derrida's more rhetorically provocative claim that ‘there is nothing outside the text’.

Vanhoozer is particularly severe with the use of intertextuality in biblical studies, which, he claims, ultimately ‘explodes the idea of canon’. He acknowledges that in one sense ‘canon confirms intertextuality’: ‘New Testament texts refer directly and indirectly to certain Old Testament texts; the meaning of the Synoptic Gospels is in part a function of their differences from one another; later texts are shot through and through with the vocabulary and themes of earlier texts’. Yet, though the ‘canon encourages a play of meaning’ this is only ‘as it were’ and ‘only within carefully prescribed boundaries’.Footnote ¹⁴ When intertextuality has the ‘last word’, it challenges and then undoes canonicity. This happens in two ways. First, intertextuality implies that a text cannot have a ‘self-same meaning’. If texts are intertexts, then ‘the very same sequence of words can mean different things in different contexts’. If intertextuality is true then, for instance, ‘Psalm 22 means one thing by itself, another taken with the Psalms as a whole, and still another when cited by Jesus’ and presumably also another when written in the text of the gospels.Footnote ¹⁵ If texts are intertexts, meaning is never fixed, but always forming or reforming and God's word cannot speak clearly. Second, Vanhoozer says that intertextuality undermines the Protestant insistence that scripture interprets itself because it treats the canon as ‘an illegitimate fence around the Scriptures’ that the postmodern intends to dismantle so that he can ‘open the text to associations with other texts and contexts’. In short, ‘Intertextuality is the free association of diverse voices, the centrifugal force that explodes the centripetal constraint of canon. Meaning is not something located in texts so much as something that happens between them. It is precisely because this ‘between’ cannot be stabilized that intertextuality undermines determinacy of meaning’.Footnote ¹⁶

Vanhoozer summarises and critiques intertextuality in this fashion:

This is fun stuff, but ultimately it is a revealing and unfortunate paragraph. In place of substantive argument, Vanhoozer relies on several rhetorical ploys. He states Derrida's position. Then he mentions a set of anonymous thinkers – ‘some critics’ – who use the principle of intertextuality as an excuse for having their way with the text. Vanhoozer does not indicate whether there are ‘other critics’ who might have a more restrained view of intertextuality. Having opened the possibility that intertextuality might be used to dismantle the text, he closes the paragraph with a flourish, satirising postmodern jouissance.Footnote ¹⁸ He gives no attention to the possibility that new theory can shed light on an old text, and thus might not only serve a heuristic purpose but might even, by drawing attention to unexplored corners of a text, provide an enriching angle of interpretation. Nor is this a merely hypothetical possibility. Freud's interpretation of Hamlet highlighted Oedipal elements that had been embedded, unrecognised, in Shakespeare's play all along, and even the most traditional of Shakespearean scholars will find it difficult to expunge thoughts of Oedipus and Jacosta as he studies Hamlet and Gertrude. A Girardian perspective on Dostoevsky likewise throws into relief an obsession with twinning, rivalry and scapegoating, and new historicist readings of Shakespeare have enabled twenty-first-century readers better to understand what a sixteenth-century audience might have heard in a play.Footnote ¹⁹ There is no inherent reason why a Marxist or feminist reading might not illumine certain aspects of the Gospel of Mark as well, though I hasten to add that a consistently Marxist or feminist reading would also distort the gospel in many ways.

I do not wish to exaggerate my differences with Vanhoozer. Elsewhere, Vanhoozer endorses a plurality of interpretative strategies (while criticising pluralism),Footnote ²⁰ and his targets are partly chosen because they offer ‘ideologically-driven interpretation that ultimately reduces textual meaning to what a specific theory lets in’.Footnote ²¹ Further, he recognises that authors intend discourses within a particular cultural as well as linguistic context, and he admits that the scope of the relevant background varies from text to text. Given his emphasis on the agency of meaning, Vanhoozer is particularly concerned with the depersonalisation of texts and meaning often assumed in accounts of intertextuality.Footnote ²² In fact, he agrees with Searle that:

In practice, furthermore, what Vanhoozer calls the ‘natural sense’ is often rich in allusive qualities. In his interpretation of John 19:34 (where blood and water flow from Jesus), he draws attention to the links between this text and John 7:38–9, where Jesus speaks of the water of the Spirit that will flow from him, and John 2, where Jesus identifies himself as the temple. Vanhoozer also rightly hears the echoes of Numbers 20 and Ezekiel 47 in John's account of the crucifixion.Footnote ²⁴ Tying interpretation to authorial intention does not, for Vanhoozer, prevent interpreters from pursuing ‘thick interpretation’.

Yet, his (partly rhetorical) polarisation of authorial intention and intertextuality remains problematic. To those problems I now turn.

Resurrecting the porous text

Vanhoozer is a clever and witty writer (though some of his puns are, perhaps intentionally, groaners). The very title of his book, Is there a Meaning in This Text? jokingly alludes to Stanley Fish's 1980 book, Is there a Text in This Class?,Footnote ²⁵ in which Fish develops his theory of interpretative communities. In First Theology, Vanhoozer writes that ‘Deconstruction castrates the text; bereft of its true voice, must not the text necessarily speak falsetto?’ and in criticising the notion that reading is specific to interpretative communities, he suggests that ‘the notion that one should look “not to the meaning but to the use” is easily traced back to Ludwig von Wittgenstein, who is neither Fish nor Fowl but the philosophical presence behind both’.Footnote ²⁶ Examples could easily be multiplied.

My question is, can his theory of textual meaning account for his own use of humour? Does Vanhoozer's theory of meaning adequately account for what is going on in his own jokey prose? Or does his wit undercut his theory? Vanhoozer would surely admit that the humour is inherent in the meaning of these statements. Meaning is, after all, what authors do when they put the linguistic system into action in particular ways, and Vanhoozer certainly intended his sentences to ‘count as’ jokes.Footnote ²⁷ But what makes these jokes jokes? I suspect that Vanhoozer would explain the humour by reference to ‘intentional context’ or ‘background’. Vanhoozer's ideal reader is one sufficiently aware of contemporary hermeneutics to know (at least) the name of Fish's book and to remember that both Fish and Fowl have written on interpretative communities. Vanhoozer expects the reader of his sentence about deconstruction to have some basic background knowledge about deconstruction, to know that castration changes the voice, and to catch the pun on false/falsetto. These are all part of the ‘intentional context’.

To refer all this to ‘background’, however, is unsatisfactory. Locating the line between foreground and background is notoriously difficult, and the reasons for the difficulty get us to the heart of my argument. Vanhoozer would certainly include authorial intention in the ‘foreground’ of the text, but authors ‘tend’ not only linguistic conventions but also cultural and literary ones. In fact, the two are inseparable. The speaker of the sentence ‘I am running for the President of the United States’ is not only assuming a common set of ‘background’ facts but deploying those facts. What Vanhoozer calls ‘general background knowledge’ is precisely what makes the sound ‘President’ mean President, and thus is part of the foreground, contained (yet absent), as it were, in the very word on the page. The situation is more complex with Vanhoozer's crack about ‘Fish and Fowl’, since the words themselves do not always carry the cultural baggage that the word ‘President’ does. Still, the point remains that Vanhoozer is tending to bits of academic culture at the same time that he is tending to adages about piscatorial and avian creatures. A reader can attend properly to his communicative act only if he recognises that Vanhoozer is intending other texts as he tends to his own.Footnote ²⁸

Indeed, the fact that the sentence is intersected by and intersects other sentences and facts is precisely what enables the author to intend to mean in the first place. A text without pores, a text with impermeable boundaries, would be incapable of communicating anything. To develop this point, Jonathan Culler's treatment of intertextuality is most helpful. Culler notes that his own discussion of literary theory ‘is intelligible only in terms of a prior body of discourse’, and in this sense he asserts ‘the intertextual nature of any verbal construct’. Culler is not merely claiming that his book occurs against a backdrop of prior discourses, but that his book means what it does only because it has its place within that network of prior discourses or, in my terminology, because the boundaries of the book are permeable to prior texts and facts. When intertextuality is offered as part of an account of meaning, it ‘leads us to consider prior texts as contributions to a code which makes possible the various effects of signification’ and ‘becomes less a name for a work's relation to particular prior texts than a designation of its participation in the discursive space of a culture’. Intertextuality thus names ‘the relationship between a text and the various languages or signifying practices of a culture and its relation to those texts which articulate for it the possibilities of that culture’. Given this understanding, intertextuality does not merely encourage source criticism, but ‘casts its net wider to include anonymous discursive practices, codes whose origins are lost, that make possible the signifying practices of later texts’.Footnote ²⁹ Vanhoozer can mean ‘Fish nor Fowl’ in the way he does only through the code provided by contemporary texts in hermeneutics.

Culler notes that intertextuality is a difficult concept to work with,Footnote ³⁰ but the problems with the concept do not argue for abandonment but rather ‘suggest the need for multiple strategies, for different focuses and restrictions’. A fruitful way forward is to consider intertextuality by reference to the role of presuppositions in linguistics. The presupposition of a text ‘introduces a modest intertextuality in relating one sentence of a text to another set of sentences which they presuppose’. To illustrate, he refers to a poem of Baudelaire, which begins with the line, ‘When the low and heavy sky weighs like a lid on the spirit groaning from the long anxieties that prey on it’. By opening with the subordinate clause, Baudelaire treats the claims of the clause as already given. He did not write, ‘The low and heavy sky weighs like a lid’, but ‘when the low and heavy sky weighs like a lid’, thus presupposing that in fact the sky does or could weigh like a lid. When Blake asks of his Tyger ‘What immortal hand or eye/could frame thy fearful symmetry?’ he presupposes that the Tyger possesses that symmetry, that it was in fact framed by an immortal hand, that the fire of the eyes burned, and thus Blake treats these implicit assertions as ‘part of a discourse or mode of discourse already in place’. Culler notes that, however limited, these examples support Roland Barthes's claims that ‘a text refers to or cites bits of discourse which are “anonymous, undiscoverable, and nevertheless already read”, and a reader will treat these presupposed (but never written) sentences as “prior discourse”’.Footnote ³¹ In addition to logical presuppositions, texts have literary presuppositions. Opening a story with the sentence, ‘Once upon a time there lived a king who had a daughter’ implies few logical presuppositions. Though ‘poor in logical presuppositions’, the sentence is ‘extremely rich in literary and practical presuppositions’. The sentence ‘relates the story to a series of other stories, identifies it with the conventions of a genre, asks us to take certain attitudes toward it’.Footnote ³²

Culler's version of intertextuality does not undermine authorial intention in the least, and is better able to account for the how the textual permeability enables an author to mean with the communicative act of a text. Culler's treatment of the porous text, in short, is capable of accounting for jokes. The writers and director of Shrek want their film, and the particular jokes in the film, to ‘count as’ comedy, and to achieve that intention they put to work conventions from nursery rhymes, fairy tales and popular culture. Culler does not confine these sources to the background but instead treats them as materials that the author ‘tends’ right along with the words and syntactical rules. The author's pallet contains not only words but, inseparably, prior texts and facts; he writes as much with those prior texts as he does with individual words organised into sentences. Textual permeability, in short, is not an obstacle or solvent of meaning, but a condition of its possibility. A hermeneutic that minimises or ignores the ways an author pushes us toward outside texts, and sucks other texts into his own, violates the ethics of reading. Such reading is not attentive to the author.

Concluding observations

In conclusion, let me suggest several other avenues where it appears that exploring the hermeneutics of jokes and humour would assist us.

First, humour is often unintended, and an analysis of unintentional humour might provide a perspective for examining how textual meaning can exceed authorial intention. For instance, the statement, made at the end of the nineteenth century, ‘Everything that can be invented has been invented’ was certainly not intended humorously by the author. But if a reader of the early twenty-first century does not find that sentence at least mildly amusing, it seems correct to say that he does not know what the sentence means. Such an analysis might also open up ways to account for new meaning in scripture. Prior to the coming of Jesus, Matthew probably did not think that Hosea 11:1 was meant to describe Jesus, and neither did Hosea. Yet, in the light of Christ's coming, a new meaning is disclosed. Now, any reader who does not read Hosea 11:1 with reference to Christ doesn't get it. Like a man speaking with unintentional humour, Hosea was talking about Jesus without realising what he was saying.

Second, if texts function like jokes, then interpretation depends on the interpreter having something analogous to a good sense of humour. This model of hermeneutics thus challenges any strictly ‘procedural’ hermeneutics, and emphasises instead the character of the interpreter. What can you do with someone who has no sense of humour? Analysis and teaching might improve things marginally, but his main problem is not a technical but a spiritual one: A man without a sense of humour suffers from a contracted soul, and the only real solution is conversion. Similarly, interpretative skills can be taught and improved, but only the glad of heart make good interpreters.

Third, analysing the experience of humour is a helpful way to explore the experience of interpretation. For both the interpreter and the reader of interpretations, the commentator and the reader of commentaries, the experience of a good interpretation can be very like the experience of a good joke. An exegete pours over a text and finally, and often suddenly, a dozen pieces fall into place; the experience is one of sudden release (sex would also be a good analogy).

The exegete might actually laugh: ‘Now, I get it’, he says.Footnote ³³