Relevance and Intertextuality

Extracted text from Pattemore_Intertext.pdf:

RELEVANCE THEORY, INTERTEXTUALITY AND THE BOOK OF REVELATION by Stephen Pattemore

d

r. Stephen Pattemore is a UBS Translation Consultant for the AsiaPacific Region and is based in New Zealand.

1. Relevance Theory and Literary Interpretation Although Relevance Theory (RT) was developed largely as a tool for understanding conversational exchanges, its implications for the interpretation of literary texts was immediately recognised and has subsequently been developed and explored further.1 But despite the fact that its potential as a tool of biblical interpretation was noted very early in its history by Cotterell and Turner, this has scarcely been developed.2 I do not intend to describe or defend RT here, but a brief statement of some of its strengths is appropriate as a background to my discussion of intertextuality.3 In relation to the interpretation of biblical text RT, with the insights it offers into the process of human cognition, is a valuable tool for use alongside others. It does not replace the disciplines of textual criticism, historical-critical analysis, discourse analysis, historical-sociological analysis and rhetorical criticism. Rather it provides a valid perspective from which to do these, a perspective with several important strengths.4 11 See, for example, D. Sperber and D. Wilson, Relevance. Communication and Cognition (Oxford. Blackwell, 2nd edn, 1995), pp. 61, 75, 158 and especially 217-243. The application of RT to literary texts has been helpfully discussed and debated in the journals Lingua 87 (1992), Language and Literature 5 (1996), Language and Literature 6 (1997). See also S. Uchida, Text and Relevance in R. Carston and S. Uchida (eds.), Relevance Theory. Applications and Implications (PBNS, 37; Amsterdam. John Benjamins, 1998), pp. 161-78; A. Pilkington, Poetic Thoughts and Poetic Effects. A Relevance Theory Account of the Literary Use of Rhetorical Tropes, (Unpublished PhD thesis, London. University College London, 1994); A. Furlong, Relevance Theory and Literary Interpretation, (Unpublished PhD thesis, London. University College London, 1996).

P. Cotterell and M. Turner, Linguistics and Biblical Interpretation (Downers Grove. IVP, 1989), pp. 270-71. A.J.P. Garrow, Revelation (New Testament Readings, London. Routledge, 1997) is one of the few commentaries to have been influenced by RT, though not to any significant extent. 3

2

For readers who are not familiar with RT, E.-A. Gutt, Relevance Theory. A Guide to Successful Communication in Translation (Dallas. SIL, 1992) provides a very readable introduction. I have discussed in more detailed the implications of RT for biblical interpretation in my Ph.D. thesis, S.W. Pattemore, The People of God in the Apocalypse. A Relevance-Theoretic Study. (Unpublished PhD thesis, Dunedin. University of Otago, 2000), ch. 2, pp. 16-88. This paper is a modified version of section 2.5, pp. 59-73.

4 Ernst Wendland, (On the Relevance of (Relevance Theory) for Bible Translation, BT 47 (1996), pp. 126-37; A Review of 'Relevance Theory' in Relation to Bible Translation in South-Central Africa. Part 1, JNSL 22 (1996), pp. 91-106) has criticized both RT and the attempt to use it as a single basis for a theory of translation proposed by E.-A. Gutt, Translation and Relevance. Cognition and Context (Oxford. Basil Blackwell, 1991). His criticisms of the latter have some validity. But in his own proposal for a multi-disciplinary approach to translation ( A Review of 'Relevance Theory' in Relation to Bible Translation in South-Central Africa. Part 2, JNSL 23 (1997), pp. 83-108, pp. 95-106) he comes very close to how I understand RT to work in relation to the other

43

First it involves treating the text as a real communication event and, as a consequence, emphasises the importance of context for understanding. This respect for the text and for the life-situation giving rise to it is a welcome change from some of the decontextualizing methodologies prevalent in post-modern criticism.5 At the same time what I have called a relevance perspective avoids the intentional fallacy and sidelines questions of literary dependence (or puts them in their proper place), by focussing attention primarily on how the text would be understood by the reader. Secondly RT gives a more precise definition to the notion of context, or to such a notion of context as is significant for understanding a text. Avoiding both the vagueness which usually accompanies the discussion of context, and the indefinability of the idea of mutual knowledge, the RT concept of mutual cognitive environments allows fairly explicit descriptions of such contexts to guide analysis, even if all the details of the wider socio-cultural environment are not known. Further, RTs assertion that contexts are not fixed in advance but progressively constructed as the understanding of a communication proceeds does justice not only to the processes of cognition in general but also to the more specific constraints of the linearity of text, and in particular of text designed to be heard. Thirdly RTs central thesis, that we understand a communication by trading off (often unconsciously) the benefits gained (contextual effects) against the amount of mental effort to be expended (processing effort), can be used as a criterion for analysing text (using any of the disciplines) from the readers/hearers perspective, and evaluating the likely priorities of competing cotexts and contexts. It also places limits on context and avoids the dilemma faced by discourse theorists, of the seemingly infinite regression of contexts.6 In this paper I will restrict my attention to two issues, and the second will only be touched on very briefly. First, I will show that relevance considerations clarify our understanding of how allusive references to earlier texts work. This will be addressed first against the wider background of intertextuality and biblical interpretation, and then more specifically from work in the book of Revelation. Secondly, I will mention briefly some implications of relevance relations to earlier texts for discourse structure, once again illustrated from Revelation.

2. Relevance and Intertextual Allusion Clearly the importance which RT places on the context (cognitive environment) of a communication relates it closely to the recent interest in intertextuality in both secular and biblical literary criticism.7 At the broadest level, according to Jonathan Culler, disciplines. Considerations of relevance in fact undergird many aspects of his model, such as discourse analysis, style and poetics. 5

For a radical deconstructive approach to Revelation, which results in it becoming unreclaimable, see A.M. Jack, Texts Reading Texts, Sacred and Secular (JSNTSup, 179; Sheffield. Sheffield Academic Press, 1999). See, for example, the problem of combinatorial explosion in R. de Beaugrande and W. Dressler, Introduction to Text-linguistics (London. Longman, 1981), pp. 32-33.

6

7

For sample bibliographies see J. Clayton and E. Rothstein (eds.), Influence and Intertextuality in Literary History (Madison. University of Wisconsin Press,1991); H.F. Plett (ed.) Intertextuality (Berlin. Walter de Gruyter,1991); D.A. Carson and H.G.M. Williamson (eds.), It is Written. Scripture Citing Scripture. Essays in Honour of Barnabas Lindars, SSF (Cambridge. CUP,1988); T. Fornberg and D. Hellholm (eds.), Texts and

44

intertextuality is less a name for a works relation to prior texts than a designation of its participation in the discursive space of a culture.8 This is a perspective on communication that RT would affirm. Every act of communication is understood against a background of cognitive environments which the interlocutors share by their participation in a common culture, and which includes much more than the current discourse or any set of related discourses. But it is also necessary to say that not all parts of the discursive space of a culture are equally accessible to the participants in a particular act of communication. RT requires that that space be, if not partitioned, then carefully evaluated in terms of concentric and intersecting spheres of accessibility (the processing effort side of the equation) and productivity (the cognitive effects side). Such analysis reflects the pragmatics of real communication. Furthermore, to anticipate the discussion below, RT leaves ample room for the allusive/elusive/illusive power of weak implicature. Regardless of the breadth of scope of intertextuality, however, with respect to a literary text in a literary culture, some degree of special consideration must be given to the relationship of the text concerned to prior literary texts. C.H. Dodd provided a creative insight into the way in which New Testament writers used the Old Testament.9 He argued that the quotes and allusions on the surface of the NT documents were pointers to the broader context from which they came.10 Commenting on the methodology of early Christian exegesis he says. The method included, first, the selection of certain large sections of the Old Testament scriptures, especially from Isaiah, Jeremiah and certain minor prophets, and from the Psalms. These sections were understood as wholes, and particular verses or sentences were quoted from them, rather as pointers to the whole context than as constituting testimonies in and for themselves [I]n the fundamental passages it is the total context that is in view, and is the basis of the argument.11 More recently Richard Hays has developed an approach to intertextuality in the writings of St. Paul, based on similar principles, but much more nuanced and sensitive to narrative context and poetic effect.12 Drawing on the work of John Hollander, he describes the functioning of textual echoes in terms of a diachronic trope which Hollander calls transumption or metalepsis. When a literary echo links the text in which it occurs to an earlier text, the figurative effect of the echo can lie in the

Contexts. Biblical Texts in Their Textual and Situational Contexts, Essays in Honor of Lars Hartman (Oslo. Scandinavian University Press,1995); C.A. Evans and J.A. Sanders (eds.), Paul and the Scriptures of Israel (JSNTSup, 83; Sheffield. JSOT Press,1993); G. Aichele and G.A. Phillips (eds.), Intertextuality and the Bible (Semeia, 69/70; Atlanta. Scholar's Press,1995). Works specifically relating to the book of Revelation will be noted below. J. Culler, The Pursuit of Signs. Semiotics, Literature, Deconstruction (Ithaca. Cornell University Press, 1981), p. 103. 9 10 8

C.H. Dodd, According to the Scriptures. The Substructure of New Testament Theology (London. Collins, 1952).

See Dodd, According to the Scriptures, p. 61, where he holds there is reason to believe that the unit of reference was sometimes wider than the usually brief form of words actually quoted. 11 12

Dodd, According to the Scriptures, p. 126, (italics his).

R.B. Hays, Echoes of Scripture in the Letters of Paul (New Haven. Yale University Press, 1989). See his note on Dodds contribution, p. 222, n, 7.

45

unstated or suppressed (transumed) points of resonance between the two texts.13 This requires that the wider context of a quote or allusion is formative for its interpretation in its new setting, or in Hollanders words, the interpretation of a metalepsis entails the recovery of the transumed material.14 The explicit point of contact between two texts can thus be a nexus for an implicit flow of meaning between texts.15 Hays claims that not only faint echoes but also explicit quotes often function by means of metalepsis, and that sensitivity to the wider context of the original setting of a quote is thus a necessary hermeneutical strategy. Even Pauls overt allusions and quotations can be explored by using some of the same interpretive strategies ordinarily appropriate to subtler echoes, because Pauls citations of Scripture often function not as proofs but as tropes. they generate new meanings by linking the earlier text (Scripture) to the later (Pauls discourse) in such a way as to produce unexpected correspondences that suggest more than they assert.16 This approach is a refinement of Dodds work, in particular because of a more subtle treatment of the way in which the context of the quoted text is used by the new text. There is no claim that the context is treated in a historical-critical sense, but rather it is a field of suggestion and evocation of ideas.17 In this it is also strongly resonant with RTs emphasis on the contribution of implicatures established within mutual cognitive environments to the meaning of communication, and in particular the treatment of poetic effects as achieving their effect through a wide array of weak implicatures.18 Hays illustrates the sensitivity of his echo-sounding by exploring echoes of Job 13.16 in Philippians 1.19, where not only the explicit linguistic mimesis, but also the readily invoked parallels of personal situation between Paul and Job, subtly lead the reader to draw further conclusions about Paul. Yet the instability of such weak implicatures is also apparent. The volume of intertextual echo varies in accordance with the semantic distance between the source and the reflecting surface. Quotation, allusion, and echo may be seen as points along a spectrum of intertextual reference, moving from the explicit to the subliminal. As we move farther away from overt citation, the source recedes into the discursive distance, the intertextual relations become less determinate, and the demand placed on the readers listening powers grows greater. As we near the vanishing point of the echo, it inevitably becomes difficult to decide whether we are 13

Hays, Echoes, p. 20; see J. Hollander, The Figure of an Echo. A Mode of Allusion in Milton and After (Berkeley. University of California Press, 1981). Hollander, The Figure of an Echo, p. 115.

14 15

It is not claimed that every quote or allusion necessarily entails a large amount of transumed material. Metalepsis is, like other figures of speech, a deliberately and artistically used tool for the evocation of meaning. Again, as with other figures of speech, there will often be an asymmetry between its intended and received effects. See further below. Hays, Echoes, p. 24

16 17

On the relationship between literary and theological implications of such allusions see the debate in Evans and Sanders (eds.), Paul and the Scriptures of Israel. Sperber and Wilson, Relevance, p. 222. See their whole section on poetic effects, pp. 217-24.

18

46

really hearing an echo at all, or whether we are only conjuring things out of the murmurings of our own imaginations.19 Here again we are very close to the RT explanation for the range of possible contextual implications that may be drawn from an utterance. Compare Sperber and Wilsons discussion. there may be no cut-off point between assumptions strongly backed by the speaker and assumptions derived from the utterance but on the hearers sole responsibility. The fiction that there is a clear-cut distinction between wholly determinate, specifically intended inferences and indeterminate, wholly unintended inferences cannot be maintained. Clearly the weaker the implicatures the less confidence the hearer can have that the particular premises or conclusions he supplies will reflect the speakers thoughts and this is where the indeterminacy lies.20 It would appear then, that the approach to the relationship between texts pioneered by Dodd and refined by Hays is strongly supported by RTs cognitive description of the communication process. Evans has argued that we must broaden the field in which the original voices of the echoes are to be sought, to include both interpreted scripture and non-scriptural texts.21 This too is affirmed by RT, with the additional criterion that the accessibility of the textual environments, the relative availability of the context to author and audience, will determine the strength of the echoes they return. It is significant then to note that the first five of Hays criteria for the validity of an echoavailability, volume, recurrence, thematic coherence and historical plausibilityall relate to the RT criterion of the accessibility of a context in terms of processing effort, while the remaining twohistory of interpretation and satisfactionrelate to the RT concept of contextual effects.22 But we are still talking here about what might be called textual spaces, or cognitive environments defined by other texts. RT asserts that anything, so long as it can be mentally represented, can contribute to the cognitive environment within which relevance is to be sought. As I observed at the beginning of this section, this is in line with the broadest definitions of intertextuality, where the potential field of reference is the whole discursive space of a culture. Two things need to be said in this regard. First, while affirming the dependence of all texts on a broad resonant space for their meaning, RT does not require the imposition of an ideologically driven destabilisation of meaning, prevalent in some discussions of intertextuality.23 It is inherently 19 Hays, Echoes, p. 23. Hays explicitly acknowledges that texts give rise to meanings other than those intended by the author, or even those allowed by the later exegete. texts can generate readings that transcend both the conscious intention of the author and all the hermeneutical strictures that we promulgate. (Hays, Echoes, p. 33). Here too he is close to Dodd. It would not be true of any literature which deserves to be called great, that its meaning is restricted to that which was explicitly in the mind of the author when he wrote it. On the contrary, it is a part of what constitutes the quality of greatness in literature that it perpetuates itself by unfolding ever new richness of unsuspected meaning as time goes on (Dodd, According to the Scriptures, p. 131). 20 21

Sperber and Wilson, Relevance, pp. 199-200.

C.A. Evans, Listening for Echoes of Interpreted Scripture in C.A. Evans and J.A. Sanders (eds.), Paul and the Scriptures of Israel (JSNTSup, 83; Sheffield. JSOT Press, 1993), pp. 47-51. 22 23

Hays, Echoes, 29-31.

Hays rejects the criticism that he works with a minimalist understanding of intertextuality, since he refuses to use it as an ideological tool. He has, as he points out, deliberately set aside the radical cynicism of Bloom,

47

constructive rather than deconstructive. Second, it may be that at precisely this point, RT can help to clarify the terminology. For if intertextuality is taken to mean the influence of everything on a text, it ceases to be a useful term, and its core morpheme, /text/, is drained of all distinctive meaning or significance. If everything is text then /text/ is no longer a useful discriminatory word. In RT terms, I suggest then that intertextuality be reserved for relevance found within textually defined cognitive environments, where a text is a (relatively stable) sequence of language. The remainder, the influence of lives and events, geography and culture, is relevance found within non-textually defined cognitive environments. The sum of the two is simply relevance.

3. Relevance and Revelations Intertextual Allusions 3.1 Three Recent Approaches Johns Apocalypse is unique among NT documents in that, while it contains no explicit quotes from the Old Testament, its whole fabric is a closely woven tapestry of verbal allusion and evocation of both biblical and extra-biblical text. This situation, which has fascinated and frustrated generations of commentators and scholars, provides a good opportunity to explore the usefulness of the Principle of Relevance in relation to intertextual relationships.24 G. K. Beale has provided a useful review of several recent authors who attempt to describe Revelations relationship to earlier scriptures.25 Three of these are evaluated here from a relevance perspective. Paulien begins with the assumption that Revelation was intended to be understood without difficulty by its original audience and that the best way to participate in the vision of the books author is to attempt to understand it in the way that the original readers would have, to tune our ears to the meanings they would have taken for granted.26 RT provides theoretical undergirding for this contextual perspective. However, problems arise when Paulien considers the ways in which a writer uses earlier texts, and in particular when he attempts to distinguish between a conscious allusion and what he calls an echo. Kristeva and Barthes, for the more positive appreciation of the functioning of textual echoes in Hollander. The more radical approach to intertextuality could be illustrated by the statement from Barthes essay The Death of the Author. We now know that the text is not a line of words releasing a single "theological" meaning (the "message" of an Author-God) but a multi-dimensional space in which a variety of writings, none of them original, blend and clash. The text is a tissue of quotations drawn from innumerable centres of culture. See R. Barthes, Image, Music, Text, Translated by Stephen Heath (Glasgow. Fontana/Collins, 1977), p. 146. Literature on the use of Scripture and other writings in Revelation has multiplied rapidly in the past 25 years. As a small sample see A. Yarbro Collins, The Combat Myth in the Book of Revelation (HDR, 9; Missoula. Scholars Press, 1976); G.K. Beale, John's Use of the Old Testament in Revelation (JSNTSup, 166; Sheffield. Sheffield Academic Press, 1998) (with references to his earlier work on the subject); D.E. Aune, Intertextuality and the Genre of the Apocalypse in E.H. Lovering Jr. (ed.), SBL Seminar Papers (Atlanta. Scholars Press, 1991), pp. 142-60; R. Bauckham, The Climax of Prophecy. Studies in the Book of Revelation (Edinburgh. T. & T. Clark, 1993); J. Fekkes, Isaiah and Prophetic Traditions in the Book of Revelation. Visionary Antecedents and their Development (JSNTSup, 93; Sheffield. JSOT Press, 1994); S. Moyise, The Old Testament in the Book of Revelation (JSNTSup, 115; Sheffield. Sheffield Academic Press, 1995); J. Paulien, Decoding Revelation's Trumpets. Literary Allusions and the Interpretation of Revelation 8.7-12 (AUSDDS, 11; Berrien Springs, Michigan. Andrews University Press, 1988); J. Paulien, The Book of Revelation and the Old Testament, BR 43 (1998), pp. 61-69 . 25 26 24

Beale, John's Use of the Old Testament, pp. 13-59. Paulien, Decoding, pp.5-6.

48

To summarize, allusive references to previous literature can enter a work in two ways. The author may use a source directly and consciously with its original context in mind. Such an allusion is "willed into being." The author is fully conscious of the source as well as of its relevance to his composition. He/she is assuming the reader's knowledge of the source and of his/her intention to refer to that source. On the other hand, an author may "echo" ideas, the origin of which he/she is unaware. In an echo, the author does not point the reader to a particular background source, but merely utilizes a "live symbol" that would be generally understood in his original audience."27 This distinction between echo and allusion is valid in itself, but the criterion Paulien uses for making the distinction in any particular case remains problematical. He appears to assume that we have access to the author's mind to know when a phrase is consciously employed and when it is just a live symbol. Form is no reliable guide because what appears a fleeting echo may well be deliberate, in the same way that a more definite resemblance may almost be accidental or in the air.28 The consciousness of the author cannot be a valid criterion for our determination of what is allusion and what is echo. Furthermore, the question must still be asked as to what difference it would make if we could determine whether or not the author was conscious of the origin of his language. Does the absence of authorial consciousness really make the original context irrelevant, as Paulien appears to imply? Fekkes appeals even more directly and explicitly to authorial intent. In attempting to distinguish between an informal quotation and an allusion he says. In the end, the distinction between an informal quotation and an allusion must be based on perceived authorial motive. This will include an estimate of the function of an OT text in a given passage and particularly its interrelationship with the surrounding context. To be more precise, the boundaries between an informal quotation and an allusion depend on the level of consciousness or visibility attached to an OT text by an author, and consequently, on the degree of recognition that that author expects of the reader.29 He goes on to suggest that this may be discerned in the degree to which a text is broken up, by study of the authors own conventions, by the prominence and accessibility of OT texts in the social and liturgical environment (e.g. testimonia). The fact that the authors intentions are a part of the communication event and that the assumption of intentionality influences the audiences understanding is not in question.30 But the attempt to make the distinction from our perspective by appeal to the authors motivation begs the question of whether this is directly accessible to us at 27 28

Paulien, Decoding, pp. 172-73. Like Hays, he bases his ideas on Hollanders Figure of an Echo.

Beale, Johns Use of the Old Testament, p. 20, criticizes the lack of objectivity in Pauliens criteria for recognizing echoes. Fekkes, Isaiah, pp. 64-65. For Fekkes, a formal quotation is one which is introduced as such in the text by at least a minimal formula. There do not appear to be any formal quotations in Revelation, hence the importance of informal quotations. His allusion would appear to be close to Pauliens echo. See R.W. Gibbs, Intentions in the Experience of Meaning (Cambridge. CUP, 1999) , especially pp. 5, 15, for a defence of this assertion. 30 29

49

all. On the other hand, Fekkes criteria for determining this motivation are largely centred on the text and the cognitive environment in which it communicates, and as such are generally compatible with relevance considerations. What must be questioned, however, is whether the distinction between an informal quotation and an allusion (certainly a deliberate allusion) is a valid or a necessary one. There is the further complication that the NTs use of an OT text also involves a language change. So even if we could establish criteria for an informal quotation in terms of contiguous text (say two or three words) it would still be difficult to establish whether the author might be (a) quoting (perhaps from memory) a translation known to us, (b) quoting an existing translation known to his audience but not to us, or (c) making an extempore translation from the MT, which might show semantic but not lexical similarity to a known translation. When we assume an allusion, how do we know it is not (b) or (c)? And does it really advance our understanding of a text if we distinguish between a quote and an allusion? If an allusion is deliberate and recognizable, then surely it is equivalent to a quote because it conveys (communicative) propositions like I have taken this idea from the former text and the former text has something to contribute to your understanding of my meaning as cognitive effects. Even a formal quote may convey no more than this (in terms of communicative intents).31 Steve Moyise was the first to explicitly apply Hays intertextual approach to Revelation in a stimulating article.32 He develops this further in The Old Testament in the Book of Revelation, where his particular interest is not so much which texts are echoed by which verses, but how John appropriates the older material, to what extent he imitates, controls or reinterprets them, and whether or not he provides a definitive interpretation of them for the reader.33 He examines Rev 1.5a in the light of Ps. 89, using Hays figure of the dissimile, in which connotations bleed over from the denied images to the entity with which they are discompared.34 Moyise concludes that John does reinterpret his sources by bringing them into use in a new context, forcing an interaction between the old and the new. [B]y incorporating the words and images of the Old Testament into his own composition, the reader is unsure whether he or she is reading Johns words or words from another context. This can be resolved by allowing one context to swallow up the other without remainder, but this misses the true dynamics of the work. Meaning in Revelation is in the tension between its previous contextual definition and its present context.the genius of Revelation is not that it borrows Old Testament language or that it has created something completely newbut in the dynamic intersection of shared language, imagery and style.35 31 32 33 34 35

See the passage from Hays, Echoes, p.24 quoted above. S. Moyise, Intertextuality and the Book of Revelation, ExpTim 104 (1993), pp. 295-98,. Moyise, The Old Testament, especially pp. 20-23 and chap. 6, Revelation and Intertextuality, pp. 108-138. Hays, Echoes, pp. 142; Moyise, The Old Testament, pp. 115-118.

Moyise, The Old Testament, pp. 137-38. The imbedded quote is from H. Davidson, T.S. Eliot and Hermeneutics. Absence and Interpretation in the Waste Land, (Baton Rouge. Louisiana State University Press, 1985), p. 117. Beale, Johns Use of the Old Testament, pp. 42-45, criticizes Moyise for being unwilling to resolve the tension between the meaning of a text in its original context and in its new context. His own view is that John views the OT with presuppositions which have grown organicallyout of the OT itself (p. 45) and that the OT context is thus determinative. We shall see later how RT can help to resolve this dilemma.

50

The question of whether the sources of Old Testament allusions or the present life situation of the readers forms the primary context within which to understand the text itself is, for Moyise, focussed by the letters to the seven churches. He asks, Which has priority, the Old Testament context or the contemporary situation?36 While respecting the work of Ramsey and Hemer on the importance of local situations, he rejects the idea that the local context determined Johns choice of texts. He concludes. Far more probable is that John composed the vision narrative using a mosaic of Scripture texts and then applied what was relevant to the various churcheswith respect to local allusions in the seven messages, it is more natural to see John applying already chosen Scripture to the local situations than the local situations suggesting the Scripture texts.37 Two comments need to be made on this from the perspective of RT. First, what is true of the letters to the seven churches is equally true of the whole book, namely that cognitive environments defined by prior texts must always be evaluated alongside those defined by local situations to determine which yields the most positive cognitive effects for the least processing effort.38 The priority claimed here of earlier texts over situational context cannot be taken for granted and must be evaluated in each instance.39 Second, while Moyise has here developed a view of the construction of the text from the authors perspective, the same considerations apply from the readers/hearers point of view.40 Furthermore, the evaluation of the accessibility of contexts from the audiences point of view is a separate exercise which may yield different results to that from the authors perspective. 3.2 Relevance criteria for intertextual relationship How does an author like John use prior texts? Does the original context determine use in the new context, or vice versa? Is there a useful distinction between a conscious allusion and an accidental echo? Relevance considerations would appear to offer a way out of these somewhat confusing dichotomies.41 RT would not attempt to prejudge the authors intentionality by appeal to other criteria, but would seek to access it through an evaluation of how an audience might 36 37 38

Moyise, The Old Testament, p. 21. Moyise, The Old Testament, p. 36.

As Moyise, The Old Testament, p. 125, notes in another context, As well as playing on the evocative power of texts, John may well be playing of the evocative power of shared experiences. Bauckham, Climax of Prophecy, pp. x-xiii, emphasizes the fact that both prior text and situational context are important for understanding Revelation. 40 Moyise deals with this to a limited extent in his chapter on intertextuality and his epilogue. The reader brings a range of experiences and expectations to the text and these also interact with the voices of the text and the texts within the text. The Old Testament, p. 126. See also pp. 142-43. Here he appears to agree with the hermeneutics of reader-response theory. Yet in the earlier parts of the book the focus is on how the author uses the OT, rather than how the audience perceives this use (which was Pauliens focus). 39

Beale, Johns Use of the Old Testament, p. 28, tends to cloud the issue emotively when he speaks of the vexed question of whether or not John uses the Old Testament contextually in line with the original meaning of Old Testament passages or whether he merely uses Old Testament wording, twisting and disregarding the original meaning to suit the new situation (italics mine). This does not allow for the possibility that Johns use may be both responsible to the OT context and yet creative.

41

51

optimise the relevance of the text itself. First we concentrate on the audience's perception. We assume that the OT and other prior texts are part of a mutual cognitive environment, a context which the author can presume he shares with his audience.42 Then we assume that the author has communicated with a view to optimising relevance. Thus whether the relationship of a statement to a previous text is an informal quotation, an allusion or an echo is scarcely important. What is important is whether the author has made manifest only the assumptions expressed by the statement (its explicatures) and those entailed by the immediate context, or in addition, the assumption that he used a particular source for his language. This distinction is to be made on the criterion of optimising relevance. In the former case, even if the audience perceives a source, any further implicatures taken from the source are entirely the audience's responsibility. In the latter case the audience may reasonably assume that the author wishes them to explore the source context for further implicatures. In the former case the source context could be said to be weakly open, in the latter case it is strongly open.43 It would seem, then, that a more valid distinction between what we shall call contextual evocation and conventional usage might be established using RT: An OT (or other source) text is contextually evoked if the audience needs to access the original context of the text, and add it to their cognitive environment, in order to optimise the relevance of the authors statement in their contemporary context. An OT (or other source) text is used conventionally if the audience can optimise relevance within the mutual cognitive environment of author and audience, without access to the original context of the imbedded text.44 But even this cannot be a bi-polar distinction. As we have already seen, RT requires that the two conditions described be seen as points on a continuum.45 Every use of an earlier text brings with it some context. The audience assumes optimal relevance and searches available contexts for ones that will provide this. If adequate contextual effects are obtained in a near context (e.g. 1st century Christian practice and belief) then there is no need to open less accessible contexts (e.g. the book of a minor prophet, or an obscure text in 2 Chronicles). On the other hand, there may be clues in the text that further processing of the source context will be rewarded by additional Bauckham, Climax of Prophecy, p. xi, goes much further than this when he says that the OT forms a body of literature which John expects his readers to know and explicitly recall in detail while reading his own work. This appears to read his own presuppositions back into the mind of John. His view that canonical scripture will be more important than non-canonical texts may well be correct (their status may mean they are more readily evoked), but it is not a necessary starting assumption. 43 Thus even in the first case, further contextual implicatures drawn from the source context may contribute good contextual effects and increase relevance. But these are largely the audiences responsibility and illustrate the asymmetric nature of the communication event. 44 42

This is very similar to Pauliens echo. Note what he says (Paulien, Decoding, pp. 175-76). In order to understand an "echo" as John did, we must go back to the origin of the idea(s), but without the assumption that John was consciously pointing to a particular background passage when he included the idea(s) in Revelation. Such a "live symbol" has become divorced from its original context... What matters is the basic meaning of the concept which had attained a fairly fixed content by New Testament times. See the quotes above, from Sperber and Wilson, Relevance, p. 23, and Hays, Echoes, pp. 199-200.

45

52

contextual effects. These could include elements in the text that remain unexplained by the near context, or that require considerable effort to process without returning significant contextual effects. In any case, how much of the source context is explored is a continuously variable quantity, and is itself determined by the principle of relevance: just sufficient to optimise the relevance of the authors usage.46 This way of looking at intertextual relationships does not eliminate all subjectivity, but it does go some way to reducing it because we are no longer trying to discriminate the same categories as before. Relevance Theory suggests a way of understanding intertextual allusion that respects the original context without making prior assumptions about the way it is being used. It gives access to the authors probably intentions, not as though they can be directly read off the text, but by assuming that the author has communicated ostensively and that human communication is guided by the quest for optimal relevance.47 3.3 Relevance criteria applied A paper by Daniel Olson provides a practical example of sensitivity to intertextualechoes in keeping with an RT approach.48 Olson focuses on the problems associated with the phrase from Rev 14.4, these who have not defiled themselves with women, for they are virgins. Though the exegetical issues are of considerable interest, what concerns me here is not the content but the methodology. Olson rejects or finds inadequate the usual hermeneutical strategies employed with this verseexcision, literal interpretation, interpretation in terms of spiritual, military or priestly metaphors. He finds a clue in Yarbro Collins suggestion of a parallel with the Book of Watchers (BW) in 1 Enoch 1-36, where the fallen angels are said to be those who have defiled themselves with women.49 Whereas Yarbro Collins did not exploit the implications of this link, Olsen finds it to be central to the interpretation of Rev 14.4, depicting the 144,000 followers of the Lamb as taking the place of the fallen angels in BW. In doing this, Olsons strategy is clearly consonant with that of Hays and with RT. If John intends to allude to the fallen angels of the BW here in Rev 14.4a, why this particular phrase? The answer to this lies in Johns understanding of the way good literature should work. The single line those who have not defiled themselves with women, for they are virgins serves effectively as a trigger to call up an entire narrative from outside See D. Blakemore, Semantic Constraints on Relevance (Oxford. Basil Blackwell, 1987) , p. 112, the Principle of Relevance entitles the hearer to interpret every utterance in the smallest and most immediately accessible context to yield an adequate range of contextual effects. This does not deny the possibility of extensive contextual reference built up by a network of allusions to the same passage, as suggested by Bauckham, Climax of Prophecy, p. xi. A second allusion is more readily perceived because the context has already been accessed. Contextual effects are able to multiply as more of the original context is evoked. 47 Beale, Johns Use of the Old Testament, pp. 50-59 argues for the importance of the authors intentions by appeal to Hirschs concept of the willed type. His discussion here is helpful and in many ways similar to the approach I have taken through RT. See especially his handling of the implications of a statement, and the varying responsibility of author and audience in determining these, pp. 54-56. 46

D.C. Olson, Those Who Have Not Defiled Themselves With Women". Revelation 14.4 and the Book of Enoch, CBQ 59 (1997), pp. 492-510 . 49 A. Yarbro Collins, Women's History and the Book of Revelation in K.H. Richards (ed.), SBL Seminar Papers (Atlanta, Georgia. Scholar's Press, 1987), pp. 80-91 . The references quoted by Olson are 1 Enoch 7.1; 9.8; 10.11; 12.4; 15.2-7.

48

53

the book. It has the marks of a literary allusion. the words are vivid and they grab the readers attention, but nothing can be found elsewhere in Revelation to which they pointIn short, Rev 14.4 sends one off looking for a storyFor anyone deeply familiar with the BW, those who have defiled themselves with women would be a reminder of the fallen watchers. A few words would call to mind the whole lurid episode.50 Of course RT does more than open a door to a wider context suggested by an allusion. It also limits the information to be retrieved from the text or its context to that which provides good contextual effects for minimum processing effort. But Olson is already there when, in countering the views of those who take offence at the male-specific and supposedly misogynist language, he says. The male language of Rev 14.4 may be no more than an accurately worded reference to a specific story in which angels lost their priestly status (were defiled) by taking wives.We may wish to pursue the implications of sexual defilement and the male priesthood for women and their bodies, but that would entail further study, not of the thought of Rev 14.4 but of the BW to which it points. The deeper we go into such an analysis, however, the less certain it will become that we are bringing back anything that will help elucidate Revelation 14, since a single brief allusion does not tell us how much or how little of the BW can be presumed in Johns thought.51 This is a very significant study from a RT perspective. Olson has employed hermeneutic strategies consistent with the search for optimum relevance, both positively, by establishing a cognitive environment within which to understand the phrase from Rev 14.4, and negatively by refusing to push the allusion beyond what it is there to achieve. Explicitly using RT categories would seem to add further confidence to his conclusions in several ways, while yet raising a question mark about the completeness of the solution offered. First, the literal meaning yields inadequate contextual effects, as it is totally unrelated to any existing hypotheses about the followers of the Lamb. It also involves excessive processing effort to make sense. RT suggests that literal interpretations involve a correspondence between the text and its referent in all features, and should thus only be resorted to if all non-literal interpretations do not yield adequate cognitive effects. Secondly, the spiritual interpretation, where virginity is taken as spiritual faithfulness (an idea present within the context of canonical Scriptures), leaves a significant amount of (male) specific information unexegeted in the text. Why are women in general considered to be defiling? Why is the ideal presented as male celibacy? The same result (i.e. indicating the spiritual faithfulness of the 144,000) could have been achieved with far less processing effort required. This suggests that optimum 50 51

Olson, Those who have not defiled themselves, p. 500.

Olson, Those who have not defiled themselves, pp. 509-510. The fact the BW is itself an intertextual echo of earlier Scripture is highlighted by K.E. Pomykala, A Scripture Profile of the Book of Watchers in C.A. Evans and S. Talmon (eds.), The Quest for Context and Meaning. Studies in Biblical Intertextuality in Honor of James A. Sanders (Leiden. Brill, 1997), pp. 263-84 ; and J.C. VanderKam, Enoch and the Growth of an Apocalyptic Tradition (CBQMS, 16; Washington. Catholic Biblical Association, 1984) , pp. 110-40, who also explores the influence of Greek literature and history on the BW, and the intertextual relations between different subsections of it.

54

relevance has not yet been achieved.52 Thirdly, while the passage clearly evokes elements of the holy war metaphor, there is no semantic or contextual link between temporary abstinence (of soldiers or priests) and virginity. To make the link requires a surplus of processing effort. Fourth, at this point in the audiences experience of the Apocalypse, the Book of Watchers may well be a context which is already open, an easily accessible mutual cognitive environment.53 No other justification for finding the echo is required, beyond that it is available. Finally, once optimal relevance is achieved by linking the 144,000 as taking the place of the fallen angels in BW, no further processing is needed. There are many other implications and derivations of the propositional form which are not to be derived once this optimal relevance is obtained. The language of male celibacy has found its target. However, it must be noted that most of what has been said applies specifically to the phrase who have not defiled themselves with women. The addition by John of for they are virgins is still left largely unexplained. If it was merely a repetition or confirmation of the intertextual reference, then it involves more processing effort than can reasonably be expected. Olson attributes the strength of the echo to the fact that four passages in the BW use language similar to Rev. 14.4a, but in only one of these is there anything approaching the explicit reference to virgins.54 Thus there remains a query as to whether Olsons discovery of a metalepsis has exhausted the intended meaning of the text. This study also illustrates the limitation of RT used in this way. However likely and however satisfying we may find this echo, its discovery is contingent on the coincidental fact that we also have access to the source of the echo. How many more examples of metalepsis are we missing simply because we do not know the texts they link to? The (diachronic) relevance we find when we read an ancient text for ourselves will be determined by our own cognitive environment, including our knowledge of other texts. But our investigation of the (synchronic) relevance found by the first readers will necessarily be limited by our ignorance of their cognitive environments.55

52 Olson, Those who have not defiled themselves, p. 494, is again close to this when he observes The apparent confusion could have been so easily avoided with different phraseology that we must suspect that something is being overlooked.

Olson, Those who have not defiled themselves, pp. 498-99, points out that Rev. 22.1-2,14 makes use of Ezekiel 47.12 in much the same way as does 1 Enoch 25.4-6, and with many similar modifications. Although the first-time hearer has not yet heard this fuller description when he/she hears 14.4, already Rev. 2.7 has spoken of the fruit of the tree of life as the reward for those who conquer, an idea which requires not only Genesis 2-3 but also 1 Enoch 25 to form its sounding board. In 1 Enoch 15.2-7 God says to the angels I did not provide wives for you, because for celestial spirits heaven is their dwelling place. Olson, Those who have not defiled themselves, p. 497. 55 cf. Hollanders comment in The Figure of an Echo, p. 65. The readers of texts, in order to overhear echoes, must have some kind of access to an earlier voice, and to its cave of resonant signification, analogous to that of the author of the later text. 54

53

55

4. Structural Effects of Intertextual Relationships Regina Blass has discussed some of the implications of RT for discourse analysis.56 She argues that discourse meaning and discourse structure are not inherent in the text itself, but in the interaction between text and context.57 Approaches to discourse analysis based on cohesion, coherence and topic often pay too little attention to the role of context. 58 But the definition and limitation of context itself is also problematic for many discourse analysts. Blass therefore proposes an approach to discourse analysis based on an RT understanding of context (as a mutual cognitive environment), and in which the optimisation of relevance is the criterion for determining text units. A unit of text is a unit of relevance if relevance is optimised over it.59 This does not replace or eliminate considerations of cohesion, coherence, topic, staging, participant structure and the like, but does provide a more reliable criterion for determining how they influence the overall structure of the text. Within the focus of this paper, on intertextual relations, I wish simply to illustrate, from the book of Revelation, a few of the implications of prior texts, as part of the context accessed in order to optimise relevance, for discourse structure.60 (1) The inclusio formed by the words in Rev. 1.1; 22.6 is reinforced by allusion to a similar inclusio in Daniel 2.28, 45 (Theodotion) where it both opens and closes Daniels speech which contains both the description of the kings dream and its interpretation. This functions to underline the integrity of the Apocalypse as a single discourse unit. (2) The reference to the four winds which are prevented from acting on land or sea (Rev. 7.1) creates possible evocations of two prior texts or traditions, themselves interrelated, namely Daniel 7 and the Synoptic Apocalypse. The effect of the latter is to bind ch.7 closely to the opening of the seals in ch.6 while the former creates a tension which is unresolved until the appearance of the beast from the sea in Rev 13.1. Thus prior texts may serve to resolve ambiguities in structure, to confirm overt surface-level structuring, and to create another layer of structure which is not apparent on the surface of the text. (3) Whether Rev. 11.19 closes the first half of chs. 4-22 or opens the second half is much controverted. Relevance relations with earlier texts argue in favour of the former both positively and negatively. Positively, the unprecedented mention of the ark may create weak evocations of 2 Macc. 2.4-8 which is consistent with the text sequences preceding 11.19 but not those that immediately follow. Another possible context of interpretation is the establishment of the ark in a tent outside Jerusalem with instructions to blow trumpets regularly before it (1 Chr. 16.6), again linking the appearance of the ark to the trumpet sequence. Negatively, Aune argues on grounds that appear to have some support from relevance, that 11.19 is linked to 12.6 by allusion to Isa. 66.6-7. But there is nothing in 11.19 in itself to justify the opening 56

R. Blass, Relevance Relations in Discourse. A Study with Special Reference to Sissala (Cambridge. CUP, 1990). Blass, Relevance Relations, pp. 9-10. See, for example, Blass, Relevance Relations, pp. 19-20, 24, 27-28. Blass, Relevance Relations, p. 80. For a fuller treatment of relevance-guided discourse analysis, see my People of God, pp. 46-58, 90-171.

57 58 59 60

56

of the context of Isa. 66 and thus the link can only possibly work in retrospect, and then only tenuously. (4) The cry of the martyrs under the altar (Rev. 6.9) requires several contexts to be opened in order to optimise its relevance. Prominent are Ps.78 (79) and Zech.1, both of which place it in the context of the appeal of the exiles for vindication, for the punishment of Israels enemies and the restoration of Jerusalem. Such expectations, aroused by the martyrs cry, find their fulfilment in the climax of the book of Revelation, where the destruction of Babylon is followed by the appearance of the New Jerusalem. Again a relevance-created structure is stretched over the skeleton of the more obvious numerical framework.

5. Conclusions and Implications for Translation Relevance Theory, by its handling of context, provides an insightful way to understand the dynamics of intertextual allusion which allows for the influence of the total situation of a communication and yet can handle the special case of the relationship of a text with prior texts. This applies to structure, to meaning and to the interactive relationship between them. It remains to point the way, however briefly, to the implications for the translation of the scriptures. The requirement of faithfulness to the original meaning of scripture demands that we not only recognise the existence of quotations and allusions, but that we attempt to understand how they are being used by a biblical author. While the apparatus and appendices of the UBS Greek New Testament provide essential checklists for the former task, RT offers very significant help in the latter. Further, contemporary audiences are obviously at a huge disadvantage in terms of sharing a cognitive environment with the authors of scripture. Part of the task of translation must therefore be to pass on not merely lexical and semantic information, but also such information as will best help the audience to establish a mutual cognitive environment with the author. I would suggest the following. (1) For a new audience to come anywhere near the understanding of the original audience, they must be able to optimise relevance by accessing the same prior scriptures. This has two consequences. First the prior texts must be available to them. This is clearly a requirement for complete canonical translation, rather than piecemeal ad hoc collections of texts perceived to be important. But second, new audiences will probably not have the same cognitive geography as the original audience and will need to be given maps and sign-posts to direct them to the prior texts which are required to optimise relevance. Footnotes are probably the way to do this, but something more nuanced than bare cross-references might be needed. (2) As well as the cognitive environment provided by earlier scriptures, there is that created by other non-scriptural texts, and indeed the whole discursive space of a culture. Once again the contemporary audience needs to have something of this replicated if it is to establish the correct relevance relations. These two points together are a strong argument for the Study Bible format as a vital, perhaps indispensable, mode of translation.

57

(3) Given the importance of structure for meaning and for the creation of relevance links both within the text itself and with earlier texts, the structure which is discovered by exploration of relevance units of text should be reflected in formatting the final text for printing.

Bibliography Aichele, G. and G.A. Phillips eds. 1995. Intertextuality and the Bible in Semeia, 69/70. Atlanta: Scholar's Press. Aune, D.E. 1991. Intertextuality and the Genre of the Apocalypse in E.H. Lovering Jr. ed. SBL Seminar Papers, Atlanta: Scholars Press. Pp. 142-60. Barthes, R. 1977. Image, Music, Text, Translated by Stephen Heath Glasgow: Fontana/Collins. Bauckham, R. 1993. The Climax of Prophecy. Studies in the Book of Revelation, Edinburgh: T. & T. Clark. Beale, G.K. 1998. John's Use of the Old Testament in Revelation, JSNTSup166, Sheffield. Sheffield Academic Press. Blakemore, D. 1987. Semantic Constraints on Relevance, Oxford: Basil Blackwell. Blass, R. 1990. Relevance Relations in Discourse. A Study with Special Reference to Sissala Cambridge: CUP. Carson, D.A. and H.G.M. Williamson, eds. 1988. It is Written. Scripture Citing Scripture. Essays in Honour of Barnabas Lindars. SSF. Cambridge: CUP. Clayton, J. and E. Rothstein, eds. 1991. Influence and Intertextuality in Literary History, Madison: University of Wisconsin Press. Cotterell, P. and M. Turner, 1989. Linguistics and Biblical Interpretation. Downers Grove IL: IVP. Culler, J. 1981. The Pursuit of Signs. Semiotics, Literature, Deconstruction, Ithaca: Cornell University Press. de Beaugrande, R. and W. Dressler. 1981. Introduction to Text-linguistics. London: Longman. Dodd, C.H. 1952. According to the Scriptures. The Substructure of New Testament Theology. London: Collins. Evans, C.A. 1993. Listening for Echoes of Interpreted Scripture in C.A. Evans and J.A. Sanders eds., Paul and the Scriptures of Israel JSNTSup, 83, Sheffield. JSOT Press, pp. 47-51. Evans, C.A. and J.A. Sanders eds. 1993. Paul and the Scriptures of Israel. JSNTSup, 83, Sheffield: JSOT Press.

58

Fekkes, J. 1994. Isaiah and Prophetic Traditions in the Book of Revelation. Visionary Antecedents and their Development. JSNTSup, 93. Sheffield: JSOT Press. Fornberg, T. and D. Hellholm 1995, eds. Texts and Contexts. Biblical Texts in Their Textual and Situational Contexts, Essays in Honor of Lars Hartman. Oslo: Scandinavian University Press. Furlong, A. 1996. Relevance Theory and Literary Interpretation, Unpublished PhD thesis, London. University College London. Garrow, A.J.P. 1997. Revelation, New Testament Readings. London: Routledge. Gibbs, R.W. 1999. Intentions in the Experience of Meaning, Cambridge: CUP. Gutt, E.-A. 1992. Relevance Theory: A Guide to Successful Communication in Translation. Dallas: SIL. . 1991. Translation and Relevance: Cognition and Context, Oxford: Basil Blackwell. Hays, R.B. 1989. Echoes of Scripture in the Letters of Paul, New Haven: Yale University Press. Hollander, J. 1981. The Figure of an Echo. A Mode of Allusion in Milton and After, Berkeley: University of California Press. Jack, A.M. 1999. Texts Reading Texts, Sacred and Secular, JSNTSup, 179. Sheffield: Sheffield Academic Press. Moyise, S. 1993. Intertextuality and the Book of Revelation in ExpTim 104 pp. 295-98. . 1995. The Old Testament in the Book of Revelation Sheffield: Sheffield Academic Press. JSNTSup, 115,

Olson, D.C. 1997. " `Those Who Have Not Defiled Themselves With Women' ". Revelation 14.4 and the Book of Enoch, CBQ 59, pp. 492-510. Pattemore, S.W. 2000, The People of God in the Apocalypse. A RelevanceTheoretic Study, Unpublished PhD thesis, Dunedin. University of Otago. Paulien, J. 1998. The Book of Revelation and the Old Testament in BR 43, pp. 61-69. . 1988. Decoding Revelation's Trumpets. Literary Allusions and the Interpretation of Revelation 8.7-12, AUSDDS. 11. Berrien Springs, Michigan: Andrews University Press. Pilkington, A. 1994. Poetic Thoughts and Poetic Effects. A Relevance Theory Account of the Literary Use of Rhetorical Tropes, Unpublished PhD thesis, London. University College London. Plett, H.F. ed. 1991. Intertextuality: Berlin: Walter de Gruyter. Pomykala, K.E. 1997. A Scripture Profile of the Book of Watchers in C.A. Evans and S. Talmon, eds. The Quest for Context and Meaning. Studies in Biblical Intertextuality in Honor of James A. Sanders, Leiden. Brill. Pp. 26384.

59

Sperber, D. and D. Wilson, 1995. Relevance. Communication and Cognition, 2nd edn. Oxford: Blackwell, Uchida, S. 1998. Text and Relevance in R. Carston and S. Uchida, eds. Relevance Theory: Applications and Implications. PBNS, 37, Amsterdam: John Benjamins, pp. 161-78. VanderKam, J.C. 1984. Enoch and the Growth of an Apocalyptic Tradition, CBQMS, 16. Washington: Catholic Biblical Association. Wendland, E. 1996. On the Relevance of "Relevance Theory" for Bible Translation in BT 47 pp. 126-37. . 1996, 1997. A Review of 'Relevance Theory' in Relation to Bible Translation in South-Central Africa. Part 1 in JNSL 22, pp. 91-106;"Part 2 in JNSL 23, pp. 83-108. Yarbro Collins, A. 1976. The Combat Myth in the Book of Revelation (HDR, 9; Missoula. Scholars Press,. . 1987. Women's History and the Book of Revelation in K.H. Richards ed.. SBL Seminar Papers, Atlanta, Georgia: Scholar's Press. Pp. 80-91.

60