INTERACTION IN POETIC IMAGERY WITH SPECIAL REFERENCE TO EARLY GREEK POETRY INTERACTION IN POETIC IMAGERY with special reference to early Greek poetry M.S.SILK Lecturer in Classics, King's College London CAMBRIDGE UNIVERSITY PRESS CAMBRIDGE UNIVERSITY PRESS Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, Sao Paulo Cambridge University Press The Edinburgh Building, Cambridge CB2 2RU, UK Published in the United States of America by Cambridge University Press, New York www. Cambridge. org Information on this title: www.cambridge.org/9780521204170 © Cambridge University Press 1974 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 1974 This digitally printed first paperback version 2006 A catalogue record for this publication is available from the British Library Library of Congress Catalogue Card Number: 73—90813 ISBN-13 978-0-521-20417-0 hardback ISBN-10 0-521-20417-8 hardback ISBN-13 978-0-521-02460-0 paperback ISBN-10 0-521-02460-9 paperback CONTENTS Prolegomenon vii Acknowledgements xiii Abbreviations and Note xiv PART I INTRODUCTION i Interaction page 3 2 Dead metaphor and normal usage 27 3 Aesthetics 57 PART II THE CATEGORIES 4 Scope and procedure 79 5 Neutral-based interaction 85 Pivot and glide 87 Convergence 103 The explanatory function 114 The specifying function 127 Support 134 6 Intrusion 138 7 Interaction outside the grammar 150 Preparation 152 Anticipation of a theme 155 Link 157 Retrospective imagery 167 8 Aural interaction 173 Alliteration and assonance: preliminaries *73 Alliterative link 175 Alliteration as enforcement for interaction 184 Aural preparation and isolation 187 Aural suggestion 191 9 Combinations 194 Preparation 194 Pivot and convergence 200 Link 204 CONTENTS APPENDICES i 'Work on interaction to date' 209 11 Catachresis 210 in The characteristic quality of various authors' usage (' normal' or ' abnormal'): ancient testimony 211 iv Some notes on alliteration in Greek 224 v Confusion in ancient sty lis tics between metaphor and 'dead metaphor' 228 vi Apologetic cbcrrrep 230 VII Privative and limiting epithets as a criterion for the presence of live metaphor 231 VIII On the history of intentionalism 233 ix V- V relations 235 x Summary of poets' apparent characteristics 238 xi dcoTOS and flowers 239 XII TOTOS ydp in Arch.i 12 240 XIII Conditional metaphor 242 xiv 2N = 1V 243 BIBLIOGRAPHY Texts and Abbreviations 245 INDEXES 1 Greek words 251 11 Passages discussed 255 in General 259 PROLEGOMENON This book is not an attempt to apply to classical literature the habits of modern literary criticism, but as it might be supposed to be just that, I may as well forestall the supposition at the outset. Despite intermittent efforts in recent times, it is still comparatively rare for practising classicists to attempt such * applications' and, if anything, especially rare for Hellenists. But the present work represents something much less fashionable altogether: an attempt by a practising classicist to extend the * theory' of an aspect of litera- ture in general in the practical context of the literature of antiquity. But in case the claim should seem unduly immodest, it can be said at once that the 'aspect of literature' in question is, in itself, a small aspect, although not a trivial one. And by way of glossing the claim, let it be said also that the 'habits of modern literary criticism' and the theoretical apparatus (if any) that accompanies them are not simply separable from their 'traditional' counterparts. There is rather, as anyone familiar with the ancestry of modern criticism will know, a developing tradition, complex and many- sided, but continuous, which, in its development, sometimes modifies, sometimes innovates entirely and sometimes reconstitutes, in effect, earlier modes. The present study is not a matter of taking over (as from the outside) a tradition at a given point in its development, whether a more ' traditional' point or a more 'modern' one; but, as I hope, a matter of taking that de- velopment a little further, irrespective of which of the above-mentioned possibilities - modification, innovation or reconstitution - are incidentally involved. And this is to be attempted with regard to poetic imagery and, specifically, an aspect of imagery that I shall be calling 'interaction'. So much has been written about poetic imagery, in connection with particular literatures and in general, that some account of interaction was almost bound to have been taken. And in practice, some, at least, of the various forms and features of interaction have not been entirely unnoticed. Witness, for example, the commentators on Aeschylus, Agamemnon 966ft, an image in which the king, returning home and by his return bringing security to the royal household, is compared to a vine:* * Vine, as O^OCKOS (970) retrospectively suggests. The Greek text quoted follows Fraenkel , 969); the English paraphrase ('sun' for 'dogstar' etc.) is my own. vii PROLEGOMENON pigris y&p ouoris cpuAAccs TKET' IS 56pious OKIOCV UTrepTeivaaa creipiou KUVOS * Kai aou IJOAOVTOS 6co|iorriTiv sariocv OOCATTOS |i£V e*v \si\i6bvi or||jiaiveis IJIOAOV. If the root lives, the leaves reach home, stretching Their shade upwards against the sun. And you, Home with the family, betoken warmth Coming in winter. On IK£T' es SOJJIOUS ('reach home') Fraenkel says: 'Vahlen. . .puts the passage into its proper setting; he points out that. . .the image which serves as a comparison and the thing compared are not always. . .kept strictly apart, that, on the contrary, the image frequently assimilates elements of the thing compared.. .' — and he goes on to cite other eminent Hellenists including Verrall, Hermann Frankel and Wilamowitz. In English studies, Mrs Nowottny supplies an instance: *a further advant- age of metaphor to the poet (simile shares this) is its power to play in with the faded metaphors of ordinary language and to draw special effects from the interplay'.* She exemplifies this power with When to the sessions of sweet silent thought I summon up remembrance of things past, ' where the metaphor of the sessions of. . . thought plays in with the faded metaphor in summon up9 and 'the familiarity of the phrase summon up acts as a guarantee of the propriety of the metaphor of the sessions of thought; the old metaphor ratifies the new metaphor, supports it... The reanimation of dead metaphor may become a means of contriving effects of great subtlety and power.. .' And on a Shakespearian passage again, Wolfgang Clemen, apropos the * associative rise of the image': You are too shallow, Hastings, much too shallow, To sound the bottom of the after-times. 'Shallow, meaning here superficial in character, assumes a concrete signifi- cance in Shakespeare's mind and thus leads to the (unuttered) image of the ocean.. ,'f These three discussions are representative of ordinary adequacy: they haven't been chosen as containing any gross errors. But for my purposes they all evidence a simple but basic inadequacy. Compare a fourth discussion, this * Nowottny 70. f Clemen 74f. viii PROLEGOMENON time from an ancient theorist. For one can in fact point to examples of Greco-Roman 'criticism' (so-called) which show the critic as having occasion to discuss an interaction, but-and here the same inadequacy presents itself conveniently conspicuously - lacking the conceptual apparatus needed to make his discussion articulate. The critic may have a perception about it, may have an inkling of something which prompted him to discuss the interaction in the first place, but can't interpret his perception with any particularity. Quintilian 5.11.24: admonendum est rarius esse in oratione illud genus, quod EIKOVOC Graeci vocant. . .quam id, quo probabilius fit quod intendimus: ut si animum dicas excolendum, similitudine utaris terrae, quae neglecta spinas ac dumos, culta fructus creat. . . it is to be noted that the kind of comparison that the Greeks call EIKCOV is less common in oratory than the kind that helps us enforce a point. If, for instance, your point is that the mind should be cultivated, you might use a comparison drawn from the soil: in a state of neglect, the soil produces thorns and brambles, whereas under cultivation it yields fruit... It isn't entirely clear that this is an interaction. That would depend on the precise form that the recommended comparison took: with excolendum animum ('the mind should be cultivated') as an actual part of the image, we would appear to have another example of Mrs Nowottny's broadly charac- terized 'playing in with the faded metaphors of ordinary language'. But in any case, 'broadly characterized' brings us back to the main point. Quin- tilian, like Mrs Nowottny, deals in broad terms, too broad. The most he can give us to indicate the nature of his prescription is the vague phrase ' the kind that helps us enforce a point', a phrase that doesn't, in itself, imply interaction at all. His problem is that he doesn't know what interaction is - and the same goes for our modern critics. None of them, by way of en- lightening themselves or their readers, can appeal to 'interaction', for the simple reason that the name is lacking and the concept lying behind the name. None of them can do what I did, blandly, just now - relate their varied instances as instances of one thing. The boundaries aren't drawn. Fraenkel's slightly opaque expression, 'the image assimilates elements of the thing compared', is in fact a reference to one particular category of inter- action; Mrs Nowottny's 'plays in with the faded metaphors of ordinary language' is a phrase, a rather unfortunate phrase, that would cover many instances of several different categories, but not all the instances of those categories and probably very few of the instances of any other category; and ix PROLEGOMENON Clemen's 'associative rise of the image' is a formula both too narrow and too wide for interaction, a formula that would exclude many of its categories and include many instances that are not to be thought of as instances of inter- action at all. This brings us to a related consequence of conceptual inadequacy. The critic isn't able to differentiate, when occasion arises, between instances of inter- action and instances of some other thing, some different aspect of imagery. Or worse, he can't see when the occasion might arise, can only press on regardless. So Clemen, thinking in terms of his 'associative rise of the image', is unable to allow for, let alone do anything with, the difference between the interactive imagery of' You are too shallow, Hastings...' and the imagery of the non-interactive passage that he cites immediately before it: And shall our quick blood, spirited with wine, Seem frosty? O, for honour of our land, Let us not hang like roping icicles Upon our houses' thatch... Clemen's gloss is, 'frosty was here the occasion for the associative image of the icicles', and he then passes on to 'You are too shallow. . .' without further comment. His reflections, though not devoid of interest, could hardly be said to evince true discrimination or to encourage it in his readers. My conclusion is that discussions of this kind - and that means any discus- sion of imagery — will stand to profit from a comprehension of interaction and the conceptual framework that the word will imply. A little while ago, I made mention - blandly, once again - of categories of interaction, and before that I noted that Quintilian, representatively, isn't able to define the nature or effect of the interaction he cites. My suggestion now is that the comprehension of imagery could profit from not only the concept of interaction, but also - for the purpose of defining natures and effects — a framework of categories within it. Look at our instances. ' En- forcing a point' is apparently a function of Quintilian's interactive (?) soil- image, but it isn't clear whether it's a function of its interactive aspect or of some other aspect; nor is it clear whether the interactive aspect is to be credited with any other function. Clemen's 'shallow.. .leads to the... image of the ocean' might be essentially a statement about the poet's mind, rather than (or as well as) about the effect of his words, but taking it as a statement about the effect, there's certainly room for doubt as to whether 'leads to' covers the whole effect. If nothing else, it might be said that 'shallow' serves to add a concrete detail to the image of the ocean represented by 'sound the bottom' and probably to make it more decisively 'of the