Narrative Order, 1789–1819 Life and Story in an Age of Revolution
Gavin Edwards
Narrative Order, 1789–1819
Books by...
10 downloads
706 Views
620KB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Narrative Order, 1789–1819 Life and Story in an Age of Revolution
Gavin Edwards
Narrative Order, 1789–1819
Books by the same author GEORGE CRABBE’S POETRY ON BORDER LAND GEORGE CRABBE: Selected Poems (ed.) WATKIN TENCH: Letters from Revolutionary France (ed.)
Narrative Order, 1789–1819 Life and Story in an Age of Revolution Gavin Edwards
© Gavin Edwards 2006 All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No paragraph of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, 90 Tottenham Court Road, London W1T 4LP. Any person who does any unauthorised act in relation to this publication may be liable to criminal prosecution and civil claims for damages. The author has asserted his right to be identified as the author of this work in accordance with the Copyright, Designs and Patents Act 1988. First published 2006 by PALGRAVE MACMILLAN Houndmills, Basingstoke, Hampshire RG21 6XS and 175 Fifth Avenue, New York, N.Y. 10010 Companies and representatives throughout the world PALGRAVE MACMILLAN is the global academic imprint of the Palgrave Macmillan division of St. Martin’s Press, LLC and of Palgrave Macmillan Ltd. Macmillan® is a registered trademark in the United States, United Kingdom and other countries. Palgrave is a registered trademark in the European Union and other countries. ISBN-13: 978–1–4039–9211–6 hardback ISBN-10: 1–4039–9211–8 hardback This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources. A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data Edwards, Gavin. Narrative order, 1789–1819 : life and story in an age of revolution / Gavin Edwards. p. cm. Includes bibliographical references and index. ISBN 1–4039–9211–8 1. English literature—19th century—History and criticism. 2. Narration (Rhetoric)—History—19th century. 3. France—History—Revolution, 1789–1799—Literature and the revolution. 4. Literature and society— Great Britain—History—19th century. 5. Literature and society—Great Britain—History—18th century. 6. English literature—18th century— History and criticism. 7. Narration (Rhetoric)—History—18th century. 8. English literature—French influences. 9. Social change in literature. 10. Revolutions in literature. I. Title. PR468.N29E34 2005 820.9′358—dc22 2005049751 10 15
9 8 7 6 14 13 12 11
5 4 3 2 10 09 08 07
1 06
Printed and bound in Great Britain by Antony Rowe Ltd, Chippenham and Eastbourne
For Deborah Ferris
This page intentionally left blank
Contents Acknowledgements
viii
Part I 1 Narrative Order
3
2 Samuel Johnson and the Order of Time
21
Part II 3 Edmund Burke: Middles versus Beginnings and Ends
35
4 Watkin Tench and the Cold Track of Narrative
56
5 William Godwin: Stories and Families
81
6 Wordsworth’s Moving Accidents
100
7 Crabbe’s Parables
123
8 Relations: Mary Wollstonecraft and Mary Shelley
139
9 The Still Unravished Bride of Lammermoor
159
Notes
179
Bibliography
193
Index
201
vii
Acknowledgements I do not know when I started to think about narrative but I know that a conversation with Martin Golding in a Bristol café sometime in the late 1960s helped me to think about it in ways that eventually led to the writing of this book. In more recent years, conversations with Deborah Ferris have done more than anything else to help me understand the relationship between life and story. I am glad to have the opportunity to thank J. Hillis Miller for his encouragement; Sue Hirst and her family for providing me with a place to live while I was writing Chapter 3; and Diana Wallace for persuading me to write Chapter 8. An earlier version of Chapter 2 appeared in Sydney Studies in English 7 (1991-–2) and a version of Chapter 4 in Southerly, 60: 3 (2000). Permission to use that material is gratefully acknowledged. A version of Chapter 5 appeared in Studies in Romanticism 34: 1 (Winter 2000). I am grateful to the editors of SiR and to the Trustees of Boston University for permission to reproduce that material here.
viii
Part I
This page intentionally left blank
1 Narrative Order
Narrative The full title of Wordsworth’s poem ‘Simon Lee’, published in Lyrical Ballads in 1798, is ‘Simon Lee, the Old Huntsman, with an incident in which he was concerned’. That reference to an ‘incident’ may lead us to expect something which, after a lengthy description of the Old Huntsman’s previously active life and present decrepit condition, Wordsworth tells us he is not going to provide: My gentle reader, I perceive How patiently you’ve waited, And I’m afraid that you expect Some tale will be related. O reader! Had you in your mind Such stores as silent thought can bring, O gentle reader! You would find A tale in everything. What more I have to say is short, I hope you’ll kindly take it; It is no tale; but should you think, Perhaps a tale you’ll make it.1 (lines 69–80) The poet then tells us how on one occasion he came to the aid of Simon Lee as the old man was vainly attempting to cut through a rotten 3
4
Narrative Order, 1789–1819: Part I
tree-stump. This is indeed a kind of ‘incident’ but one that develops into a thought rather than a ‘tale’: – I’ve heard of hearts unkind, kind deeds With coldness still returning. Alas! The gratitude of men Has oftener left me mourning. (lines 101–4) This conclusion is not at all self-explanatory. What Wordsworth has described is, as so often in Lyrical Ballads, ‘an incident whose meaning can be supposed only after active deciphering in the mind of an alerted reader’.2 Whether Wordsworth thinks that this process of decipherment ought to lead us, as he believes it may well lead us, to our ‘making a tale’ out of the raw materials he has supplied is not clear. It seems more likely that the poet is nudging us in the direction of a different kind of mental activity altogether: the kind exemplified by Tom Paine’s proposal for old-age pensions in Part Two of Rights of Man (1792) or by William Godwin’s critique of gratitude in An Enquiry Concerning Political Justice (1793). Paine had proposed a tax to support the aged which would be ‘not of the nature of a charity, but a right’. Godwin had questioned the response which charity required. ‘Gratitude’, he wrote, if by gratitude we understand a sentiment of preference which we entertain towards another, upon the ground of my having been the subject of his benefits, is no part either of justice or virtue.3 We can in any case tell, from Wordsworth’s address to his ‘gentle reader’, that his doubts about storytelling are social and political as well as moral ones. In the context which Wordsworth has established, ‘gentle’ must be ironic, exposing a gap between two meanings of the word which its use would normally encourage people to confuse. The word refers to the likely social rank of the reader but also to the virtue of kindness ideally associated with that rank: hence the poem’s reference in its final stanza to ‘hearts unkind, kind deeds’. The point here is that while the poet’s contemporary readers very likely are gentlemen or gentlewomen, Simon Lee is not. The desire for a story, Wordsworth suggests, stands between the reading classes and the undramatic suffering of an old poor man like Simon Lee. ‘Simon Lee’ is only one, especially explicit, example of a widespread scepticism about narrative which appeared in British writing from 1789. Even those writers who were committed to the aesthetic and ethical
Narrative Order
5
value of narrative betray a sense that the narrative ordering of experience is either superficial, or dangerous, or impossible; while even those who were most hostile to narrative recognized, like the author of ‘Simon Lee’, its seductive power. ‘Simon Lee’ is also a good example of two things which were frequently linked to scepticism about narrative. The first is the important part which the semantic complexity of certain everyday individual words often played in the texts of narrative scepticism. Wordsworth makes his case, in ‘Simon Lee’, partly by the deft way in which he plays with the word ‘gentle’, separating the semantic elements it normally combined. Wordsworth here displays an awareness of the political significance of semantic complexity which was widespread in the poetry and polemic of the period, most programmatically in Charles Pigott’s satirical Political Dictionary of 1795.4 However, the writers were not always so alert to the complexity of the complex words they deployed as Wordsworth was with ‘gentle’ in ‘Simon Lee’. Variant meanings of such words could often take an argument in a direction that its author could not wholly control, or allow a writer to follow a line of argument with the left hand of which the right hand was not wholly aware. Godwin’s use of ‘character’ and ‘family’, Scott’s of ‘figure’ are slippery in this way, as is Wordsworth’s own use of ‘accident’. It is a slipperiness which arises in many cases from the fact that while the semantic complexity of some words (such as ‘gentle’) was of long standing, other words (such as ‘character’ and ‘family’) were undergoing semantic transformation in this period. There is always semantic change and instability; but in an age of revolution the meaning of many key words (including, it has been argued, ‘revolution’ itself) may have been particularly unstable.5 Wordsworth’s ‘Simon Lee’ also demonstrates the link, pervasive in this period, between scepticism about narrative and scepticism about the making of promises and contracts. If the poem’s subtitle, with its mention of an ‘incident’, seems to promise a story, the poem itself deliberately breaks the promise, or does not keep it in the way the reader has been led to expect. This procedure is turned into a principle of Lyrical Ballads as a whole by the 1800 Preface: It is supposed, that by the act of writing in verse an Author makes a formal engagement that he will gratify certain known habits of association, that he not only thus apprises the Reader that certain classes of ideas and expressions will be found in his book, but that others will be carefully excluded.
6
Narrative Order, 1789–1819: Part I
But Wordsworth then cautions his readers: I will not take upon me to determine the exact import of the promise which by the act of writing in verse an Author in the present day makes to his Reader; but I am certain it will appear to many persons that I have not fulfilled the terms of an engagement thus voluntarily contracted. (pp. 243–4) Wordsworth is of course talking here not specifically about the promise of a narrative but, more broadly, about the expectation that may be raised by ‘the act of writing in verse’. And indeed writers of all kinds (philosophers and pamphleteers as much as novelists, historians or ballad makers) can raise expectations which they then either live up to or disappoint. Nevertheless, it is more appropriate to talk about the ‘formal engagement’ to tell a story than, for instance, to expound a philosophical argument. A story is – for reasons I shall explain – particularly like a promise that is kept, and a story that turns out not to be one is particularly like a broken promise.
Order The word ‘order’ in my title requires to be understood in two senses: as regular and stable arrangement (the opposite of disorder), and as sequence. Sometimes these meanings come together, as in Lady Macbeth’s injunction to her noble guests to ‘stand not upon the order of your going, but go!’ Hierarchy is often expressed as sequence. Nevertheless, it is important – particularly in a discussion of narrative sequence in an age of Revolution – to be aware that this identification between orderliness and finite sequence is not a necessary one. Indeed, Narrative Order aims to draw attention to changes in people’s conception of the relationship between these two meanings. My argument is that British enlightenment writers, despite their very real differences, shared an assumption that there was indeed a very close relationship between the two meanings. These writers shared what Alasdair MacIntyre has called ‘a narrative idea of life’. On this view, we see our lives as they really are by standing back from them so as to get them into perspective. In particular, we see our own individual lives properly when we see them in essentially the same way as we more easily see the lives of others, as narratives in the third person. Since we can witness and record the birth and death of another person but not of ourselves, the
Narrative Order
7
life of another is more obviously like a narrative than our own life. Biography is not just a different form of narrative from autobiography, but a more narrative form. The narrative view of life assumed that, as MacIntyre puts it, ‘I am what I may justifiably be taken by others to be in the course of living out a story that runs from my birth to my death.’6 This was a view of the world – embodied in normal pre-nineteenthcentury usage of terms such as ‘character’, ‘story’ and ‘history’ – which strongly identified order in the sense of systematic arrangement, including social order, with what Samuel Johnson called ‘the order of Time’. In fact the phrase ‘the order of Time’ itself, as Johnson uses it, identifies systematic arrangement and discipline with the kind of finite sequence characteristic of biography. It is not surprising then that the narrative idea of life found its most explicit eighteenth-century expression, so far as the understanding of individual lives was concerned, in Johnson’s privileging of biography as a genre. For that reason, the present introductory chapter of Narrative Order is followed by a detailed reading of Johnson’s own most powerful biography, the Life of Richard Savage (1744). While no single text can, of course, suggest the range and the conflicting tendencies of a cultural period, the Life of Savage does effectively demonstrate certain important features of pre-1789 British enlightenment writing. It demonstrates, for instance, that what was specific to it was not a commitment to order as such but to a particular version of order: one that linked stable arrangement to finite sequence (a sequence with a beginning and end). The Life of Savage also demonstrates that the belief in narrative order should not be thought of as necessarily a confident, let alone a complacent commitment, immune from the sense of contrariety and internal division more obviously present in writing from the age of revolution. Johnson’s relentless commitment to narrative order – to the bond, that is to say, between finite sequence and systematic arrangement – is a response to his strong sense of what threatens that order. Furthermore, the natural facts and social practices upon which, I shall suggest, any narrative ordering of life depends – in particular, the natural facts of mortality and the social practice of making promises and contracts – are demonstrably as central to the Life of Savage as they are to the later texts. In other words, the Life of Savage provides an instructive contrast to the later texts because its difference from them stands out against a background of considerable similarity. This is particularly the case for its relationship to the first text from the age of revolution to be
8
Narrative Order, 1789–1819: Part I
discussed here in detail, Edmund Burke’s Reflections on the Revolution in France (1790). Indeed, the most specific reason for offering a detailed reading of the Life of Savage is that its preoccupations and values are particularly close to those of the Reflections, the text which first seriously challenged the narrative idea of life. Both texts are preoccupied with the problematic relationship between ends and beginnings, death and birth, because both are haunted by the possibility that the past can be relived and old battles refought. In the Life of Savage these are, on the face of it, purely personal and figurative battles; while the Reflections, as a political text, is more obviously concerned with literal ones. But the personal and political are in fact strongly linked in both texts. Johnson and Burke were both haunted by the possibility that the political past might be relived, and that old – seventeenth century – battles might be refought to a different conclusion, whether under the banner of jacobitism or of regicide ‘enthusiasm’. Johnson wrote the Life of Savage in 1744, when the temptation presented by the Stuart cause was still real for many British tories, a temptation effectively destroyed by the military defeat of the jacobite forces at Culloden in 1746 and the subsequent incorporation of the highland clans into a commercially and juridically unified British state.7 The social and religious conflicts of the seventeenth century remained closer to the surface in Ireland than in Britain, and the French revolution of 1789 was so dangerous for the Anglo-Irish Edmund Burke partly because it immediately threatened to revive (and to revive in himself) those seventeenth-century conflicts. It threatened to revive them in a quite new way however, a way which caused Burke to see finite narrative sequence as the enemy rather than – as it had been for Johnson – the ally of social order. I have selected the Life of Savage to represent the earlier period, then, not because it is wholly different from the later texts but because it is in important ways very similar to them. It is against a background of extensive similarity that difference most vividly reveals itself and allows us to claim that a seismic change really did take place between the two periods, a change of the sort required to constitute them as different periods. It will be evident from what I have said about ‘Simon Lee’ and about Burke’s Reflections that narrative scepticism could take various forms and have a very wide range of ethical and political implications, depending on the aspect of narrative which is at issue. It is important to stress this because, if Burke were left out of the picture, it would be possible (though it would still be misleading) to argue that just as
Narrative Order
9
support for the existing social order went with support for narrative order in Johnson, so criticism of the status quo was linked to scepticism about narrative following the revolution. A number of texts examined in Narrative Order could be cited in support of such an argument. Watkin Tench’s account of forced labour and near-starvation in the new convict colony of New South Wales would be the clearest case. In the middle of his narrative Account of the Settlement at Port Jackson (1793) Tench suddenly turns away from ‘the cold track of narrative’ in favour of the original journal notes – written in the heat of the moment and full of rebellious ‘ardour’ – on which his narrative has up to that point been based. It is an almost Blakian moment, in which Tench recoils simultaneously from narrative and from his own participation in the attempt to reconstitute British social order on the other side of the world. The fact is however that the narrative idea of life, with its assumption of a strong link between the different meanings of ‘order’, was first systematically and influentially challenged not by Blake, or by Wordsworth in his more radical years, but by Edmund Burke; in recoil from the possibility of radical change, not in support of it.8 Burke did retain a commitment, as we shall see, to certain aspects of narrative sequence, but he systematically detached social and moral order from a key feature of narrative, its commitment to beginnings and endings. Burke’s response to the revolution in France was to identify beginnings and endings with enlightenment. The model of enlightenment against which Burke reacted was not, however, the dominant British one, which has been described as ‘conservative’.9 This British model was based, it has been argued, on an acceptance of the self-consciously ‘moderate’ order in Church and State established by the 1688 Settlement. This was an order to which Burke too was committed but which he came to see as threatened by enlightenment because, following the revolution in France, he identified enlightenment with its more frequently oppositional and rationalist continental forms. The French revolutionaries, Burke argued, were committed to ‘beginning [the state’s] inauguration by its destruction’,10 destroying existing society so as to found a completely new one. Recoiling from this convergence of endings and beginnings – of death and birth – Burke asserted middles, mediations. These mediations were the temporal mechanisms of anticipation, retrospection and anticipated retrospection embodied most concretely, Burke believed, in a characteristically English contractual arrangement, the form of property inheritance known as ‘entail’.
10
Narrative Order, 1789–1819: Part I
It must, however, be stressed that the connection between Burke’s political thinking and the question of narrative is not a connection Burke himself makes or would have recognized. Unlike most of the other writers discussed in this book, who knew that they were interested in the ethics and politics of narrative, Burke was probably not aware that the Reflections was raising issues of this sort. Other writers – notably the Wordsworth of ‘Tintern Abbey’ and the Prelude, the Scott of the Waverley Novels – certainly drew practical lessons about storytelling as well as about politics from the Reflections. Burke helped them to develop less narrative forms of narrative, forms of lyricized and temporalized narrative less committed to beginnings and endings. But they did not think of their debt to Burke in this way. It is twentieth-century thinking about narrative which has allowed us to more clearly recognize and conceptualize Burke’s epoch-making challenge to narrative order. Relevant aspects of this modern thinking will be detailed below, but the essential point is this: the temporal mechanisms of ‘entailed inheritance’ as Burke conceived them are also the temporal mechanisms of narrative as modern narrative theorists have described them. Peter Brooks, for instance, has argued that ‘the essence of narrative plot is the anticipation of retrospection’.11 Recoiling from the revolutionary bringing together of beginnings and endings, Burke foregrounded these narrative mechanisms of anticipation and retrospection so as to obscure and postpone what they are normally designed to bring into being: narratives, with their beginnings and endings.
An age of revolution The reference to ‘1789’ and to an ‘age of revolution’ in the title of this book will have indicated from the start my allegiance to that tradition of analysis which credits the revolution in France with an epoch-making (or period-making) role in British literary culture. However, it is possible to overestimate or misconstrue this impact. The events which took place in France between 1789 and 1794 did not single-handedly transform British culture. For one thing, as Marilyn Butler has argued, Britain had already experienced ‘a period of rapid change or expectation of change’ since about 1760, ‘and its restiveness is conveyed in literature’.12 As for the period after 1789, there has been a tendency to overestimate the impact of the revolution itself and underestimate the importance of the counter-revolutionary and Napoleonic wars which the revolution
Narrative Order
11
precipitated.13 This tendency reflects a group of connected facts about British life. During the two centuries between the Battle of Culloden (1746) and the Battle of Britain (1940), none of Britain’s numerous military actions took place in Britain itself. What is more, in the period of the revolutionary and Napoleonic wars themselves, the writers now thought of as significant were not military people, and did not directly participate in the death and destruction in which their country was so heavily involved. As for the French revolution itself, ‘too much attention has been devoted to [its] spectacular impact . . . at the expense of the peculiarly English conditions which met that impact’.14 Indigenous features of the relationship between social classes and between men and women put both formal contractual arrangements and narrative ordering under intense strain in ways that were largely independent of events overseas but which help to explain the nature of the impact which those overseas events did have. Furthermore, it is not only peculiarly English conditions we need to consider, but British, Irish and colonial conditions. This is obviously true in the case of Walter Scott. It is equally true of Edmund Burke: the Reflections do not mention Ireland but Ireland is as real a presence in that book as Britain and France. It is also true of the important but – in Britain – little-known figure of Watkin Tench. Indeed, it is in the context of these provisos – this recognition of the various ways in which the French revolution did not single-handedly transform British culture – that the work of Watkin Tench acquires its particular importance here. Well known in Australia, Tench and his writing need an introduction to readers elsewhere. A Captain of Marines on the fleet of convict ships that reached Botany Bay in 1788, Tench was also the author of two books which together constitute the most astute and lively narrative account of the founding of New South Wales as a British colony, A Narrative of the Expedition to Botany Bay (1789) and An Account of the Settlement at Port Jackson (1793). Tench had previously fought in the American War of Independence and spent time in the West Indies. After his return from New South Wales in 1792 he saw service against the new French Republic in the Channel and Atlantic fleets. His third and final book, an account of six months as a prisoner of war in Brittany (following a sea battle off Brest), was published in 1796 as Letters Written in France to a Friend in London between the Month of November 1794 and the Month of May 1795. Tench is important here in two principal respects. First, he is probably the most-talented British writer to represent the military and political conflicts of the period as a participant in the fighting. He is indeed
12
Narrative Order, 1789–1819: Part I
self-consciously the soldier-storyteller, alluding – at the start of Letters Written in France – to Caesar’s Gallic Wars and to Othello’s speech to the Venetian senate, to support that self-image. Secondly, his military and literary careers not only spanned the globe but crossed the watershed of 1789. Indeed, while the ‘1789’ in my title does refer to the revolution in France, it also refers to A Narrative of the Expedition to Botany Bay, published in London in 1789 but wholly written in New South Wales in 1788. The Narrative allows us to assess the strain imposed on narrative order by an episode of social dislocation and innovation – the British colonization of New South Wales – which took place immediately prior to the French revolution. Offering a narrative not only of the ‘expedition’, but of the first months of colonization, the Narrative demonstrates the difficulty of writing a narrative of the immediate past hard on the heels of the journal on which it is based and in circumstances so volatile and unpredictable that Tench’s claim to be writing from a stable point of hindsight – the claim which any narrative makes – is palpably unconvincing. The immediately pre-revolutionary Narrative was succeeded by a sequel, An Account of the Settlement at Port Jackson. The Account, published in London in 1793, continues the story of the colony’s foundation up to June 1791, the month in which Tench himself left New South Wales for London, where he arrived in March 1792. This sequel reveals the same problems of establishing a stable point of hindsight that was evident in the Narrative, but now aggravated by the new perspectives on the convict colony suggested by events in France in 1789 and 1792–3 as news of them had reached the author, first in Sydney and then in London. It is not possible to establish exactly the compositional strata of either the Narrative or the Account but what is fascinatingly evident in both, and in the relationship between the two, is the process of rewriting (‘rewriting history’ as we would now say) produced by changing social circumstances first within the colony and then under the impact of successive French revolutionary events. Writing in 1796 about the previous six years, Mary Wollstonecraft noted ‘the immense multitude of important events which [France] has compressed within so narrow a circle’.15 She made these remarks in the course of reviewing Tench’s account of his experiences as a prisoner of the French, and it is not surprising that she should do so. As the full title of Tench’s book makes particularly clear, he had now exchanged continuous narrative for a form – the precisely dated letter – able to register the volatile and unpredictable character of revolutionary reality in which historical processes are speeded up, systems of government and value
Narrative Order
13
succeeding one another almost week by week. He is able, using the letterform, to present us with a future that cannot be anticipated and a changing present in which one point of hindsight is continually being pushed aside by another. In a letter from Quimper to his (real or imaginary) friend in London, dated ‘4th of April 1795’, Tench writes: In the centre of [the market-place] stands, on a square pedestal, a statue of Liberty, with inscriptions on each side, some parts of which have been recently white-washed, to obliterate them. Among these I could decypher the word ‘Montagne’ and a few others of analogous signification, which a change of opinion has suddenly expunged from the vocabulary of French patriotism.16 In the Reflections, Burke had asked his readers to imagine the revolutionary of 1789 bringing ‘himself to that pitch of presumption, to consider his country as nothing but carte blanche, upon which he may scribble whatever he pleases’ (p. 266). In the description of the whitewashed statue of Liberty of April 1795 Tench shows Burke’s metaphor brought to life as repeated attempts are made to wipe the slate clean and start again; with, as Deirdre Coleman points out, ‘Montagne . . . not entirely effaced because it is still too soon to know which party will supersede it.’17 These are circumstances in which it would be difficult to believe with any real conviction either in the present as a secure point of hindsight or in the idea of anticipating a secure point of hindsight in the future. And if circumstances of revolution made narrative problematic, they made promising problematic too. Tench spells this out in the very first of his ‘letters’, written on board the prison-ship Marat in Brest harbour: On board le Marat Brest, 9th Nov. 1794. My Dear Friend A performance of those flattering promises, which we exchanged at parting, to meet for a few days in London, about Christmas next, provided the exigencies of service would permit, must be suspended for the present – to be fulfilled when – is one of those secrets of futurity, which I dare not trust my imagination to anticipate. (p. 5)
14
Narrative Order, 1789–1819: Part I
It was not only in Tench’s Narrative of the Expedition to Botany Bay and Account of the Settlement at Port Jackson that the impact of radical, repeated and unpredictable change revealed itself so materially in the compositional stratification of a narrative text. Elsewhere too, in texts from the revolutionary 1790s, the point of view from which a narrative is written has clearly changed in the process of writing. Nowhere is this more vividly evident than in Things as They Are; or the Adventures of Caleb Williams (1794) and nowhere is this narrative instability more relevant to the substance of the novel in which it appears. Caleb Williams takes the form of Caleb’s ‘memoirs’ and Godwin had first ended his novel as Caleb’s mind and syntax disintegrate under the impact of his defeat by his master, Count Falkland. However, shortly before he delivered the manuscript to the printer, Godwin substituted a quite different ending in which Caleb triumphs over Falkland (though he experiences his triumph as a moral defeat). The interesting thing is that this change of mind on Godwin’s part echoes a change of mind which Godwin attributes to Caleb. The reader discovers, just two chapters from the end of the novel, that Caleb’s autobiographical narrative is not, after all, coextensive with the novel itself. Instead, Caleb starts a new narrative, with the words: ‘a revolution has taken place in my mind’. A revolution is something that changes your view of things so radically that you have to stop your story and start again. Focused so firmly on the present – on ‘things as they are’ – Caleb Williams brilliantly evokes one ‘present time’ changing into another one so completely different that the story you are writing or reading has to be rewritten or re-read. Caleb Williams had two endings. Wollstonecraft’s novel The Wrongs of Women; or Maria had none. Wollstonecraft’s death from puerperal fever, following the birth of her daughter Mary, meant that the novel was never finished. The manuscript was edited for publication by Godwin in 1789 and includes, as the penultimate chapter, Wollstonecraft’s notes for a number of possible ways in which the novel might have proceeded. And we are aware, even more than in Godwin’s novel, of the extent to which the choice of ending would have effected our reading of the whole novel. We are aware that events which happen early in the novel remain obscure – both as to fact and interpretation – without the solutions which an ending would have provided. This unfinished novel reveals in exemplary fashion the fact that a narrative without an ending can only have a provisional beginning. There is of course nothing specific to a period of revolution about the death of a woman in such circumstances. And in one sense the incomplete Maria represents the threat which mortality always poses for the
Narrative Order
15
narrative ordering of experience. And yet this particular death, because of the writer the mother was and the writer the daughter turned out to be, did come to have an impact, through Frankenstein, on the narrative ordering of experience in some ways comparable – as a violent bringing together of ending and beginning – to the revolution in France. In both cases, the two inseparable aspects of mortality are brought face to face, without mediation. Deprived of a relationship with her mother, what the daughter inherited was her mother’s life-story. Frankenstein explores the tragic situation in which people deprived of a relationship (in the case of Dr Frankenstein’s ‘Creature’, any relationship at all) attempt without success to bring one into being by means of a relation, by telling their own life-story to another person. It is no accident that the sense of compositional interruption – narratives palpably caught on the wrong foot by radical alterations of circumstance, shifts in the point of hindsight – should be so peculiarly a feature of the revolutionary 1790s. In the first two decades of the following century writers developed forms which could incorporate, and in a sense contain, such volatility: forms that on the one hand took for granted the absence of a strong, finite narrative order, and on the other hand frequently thematized such an order as an object of insatiable desire. The two most influential genres to emerge in the early nineteenth century – the short lyric and the historicist historical novel – were complementary genres in this respect, as the title of my final chapter aims to suggest. The short lyric that reflects upon its own passing present complements the historical narrative that offers itself as an episode in a story that is as fundamentally unfinished as the history in which it is being written. Scott’s Bride of Lammermoor (1819) is particularly appropriate as a text with which to conclude a period sequence that begins with Burke’s Reflections. Scott’s most Burkian novel, this study of marital and national contracts ends by directly facing, in a way Burke was afraid to do, the murderous acts which Burke feared were inseparable from such inaugurating ‘acts of union’.
Life and story The importance of what Frank Kermode called ‘the sense of an ending’ was most vividly articulated, as Kermode acknowledged, by Jean-Paul Sartre in his 1936 novel Nausea. Roquentin, the novel’s protagonist and narrator, is a biographer by profession, but a biographer who has become disillusioned with his work. He has come to distrust our desire
16
Narrative Order, 1789–1819: Part I
to see lives as stories. Living is one thing, he argues; telling stories about living is quite another: When you are living, nothing really happens. The settings change, people come in and go out, that’s all. There are never any beginnings. But when you tell about life everything changes; only it’s a change nobody notices. The proof of that is that people talk about true stories; whereas in fact events take place one way and we recount them the opposite way. You appear to begin at the beginning: ‘It was a fine autumn evening in 1922. I was a solicitor’s clerk at Marommes’. And in fact you have begun at the end. It is there, invisible and present, and it is the end which gives these few words the pomp and value of a beginning. The fellow is already the hero of the story. His morose mood, his money troubles are much more precious than ours, they are all gilded by the light of future passions. And the story goes on in reverse: the moments have stopped piling up on one another in a happy-go-lucky manner, they are caught by the end of the story which attracts them and each of them in turn attracts the preceding moments. We have the impression that the hero lived all the details of that night like annunciations, promises, or even that he lived only those that were promises. We forget that the future was not yet there.18 According to Sartre we are, in Jonathan Rée’s words, ‘constantly transmuting our lives into anticipatory biographies or obituaries, rather than living them first hand’.19 Sartre sees the storytelling impulse as, in Galen Strawson’s words, ‘our fundamental existential failing, the heart of our inauthenticity or mauvais foi’, a tendency rooted in the human mind rather than in the realities of material or social life.20 Sartre’s outright hostility to storytelling so conceived was echoed by Louis Mink, who argued that stories are not lived but told. Life has no beginnings, middle or ends. There are meetings, but the start of an affair belongs to the story we tell ourselves later, and there are partings, but final partings only in the story. There are hopes, plans, battles and ideas, but only in retrospective stories are hopes unfulfilled, plans miscarried, battles decisive, and ideas seminal. Only in the story is it America which Columbus discovered.21 That final sentence – ‘only in the story is it America which Columbus discovered’ – sums up an essential element in recent discussions of ‘the
Narrative Order
17
invention of tradition’ and of the link between ‘nation and narration’. Paul Carter’s remarkable book The Road to Botany Bay: A Study in Spatial History which offers readings of eighteenth-century British narratives of Pacific exploration (starting with James Cook’s Journals) is a particularly relevant contribution to these discussions in the present context. While drawing on other philosophical sources, Carter’s book helps us to link narrative scepticism of a Sartrian kind with the period subject matter of Narrative Order. Indeed, the first part of Carter’s title (The Road to Botany Bay) is taken from Tench’s Narrative of an Expedition to Botany Bay, a book which embodies, according to Carter, a ‘cause-and-effect empirical’ view of history in which the past is seen in the light ‘of events that have not yet happened’.22 Or, as Sartre’s Roquentin puts it, when we represent the past as a story, ‘we forget that the future was not yet there’. However, the sharp distinction which Mink, Sartre and Carter wish to draw between living and stories about living can be questioned. Alasdair MacIntyre, for instance, responded to Mink’s claim that ‘in life there are no endings’, by asking: ‘but have you never heard of death?’ (p. 197). Death, that is to say, is an ending provided by nature, not a mental, cultural or social construction. On the other hand – and as we shall repeatedly have cause to notice – while death is what most clearly guarantees the possibility of story, it is also what presents the most powerful challenge to it, as the interruption of Maria by the death of its author made brutally clear. While death provides a position of hindsight, it is a position which, in relation to our own individual lives, we may anticipate but can never reach. It is this aspect of our relationship to death which makes autobiography a less narrative form than biography. As Eric Hobsbawm puts it, towards the end of his own autobiography, ‘biographies end with the subject’s death. Autobiographies have no such natural termination.’23 The close relationship between death and narrative, strangely doubleedged as it is, does put a question mark against Sartre’s argument in one respect. For Sartre, it would seem, the narrativizing of our lives is a wholly mental process, ungrounded in material or social realities. Moreover, if nature provides, in the facts of birth and death, a kind of beginning and ending to individual lives, the social practice of making contracts and promises can be said to provide the specific sort of beginning that narrative requires: the sort that anticipates retrospection. Sartre unwittingly suggests as much: We have the impression that the hero lived all the details of that night like annunciations, promises, or even that he lived only those that were promises.
18
Narrative Order, 1789–1819: Part I
Sartre uses the word ‘promises’ in a figurative sense but it may also be helpful, as I have suggested, to think about narrative in connection with literal promises. If, as Kierkegaard argued, life is lived forwards but understood backwards, then promising is a way of trying to live a tightly integrated life, linking ends to beginnings in a way that gives your life the structure of a narrative.24 Louis Mink may be right to say that ‘the start of an affair belongs to the story we tell ourselves later’, but the same is not true of a marriage. What the texts of the Revolutionary period – so often preoccupied with promises and contracts – demonstrate is that it is through actions in which people formally commit themselves to a certain future line of conduct that they attempt to give their lives a narrative form. They attempt to do so, but they do not necessarily succeed. In relation to narrative, promises are as double-edged as death. The act of promising is certainly an inaugurating act, an act of establishing a beginning, with all the power which the concept of a ‘speech act’ attributes to utterances of this kind, whether spoken or written. And yet, to some extent, the proof of a promise is in the keeping: the speech-act depends for some of its power on the subsequent performance of other kinds of act.25 The beginning exists securely as such only in its anticipation of continuance through a designated period of time. It is only if they are kept that promises and contracts structure lives as narratives. The power to which promises and contracts aspire is indeed constantly associated, in the texts we shall be examining, with the power to which stories aspire. ‘Listening to stories’, says David Simpson, ‘holds us passive, even spellbound, awaiting the evolution of a narrative that we do not want to interrupt’ (p. 72). And this power is not necessarily confined to the moment of storytelling itself. The idea that a story can bind like a spell and can lead people, thus bound, to do things they would not otherwise have done, has one of its classic expressions in Shakespeare’s Othello. At the beginning of the play Othello stands accused by Desdemona’s father, Brabantio, of winning the hand of his daughter by spells and magic potions. The speech which Othello then makes in his own defence demonstrates to the Venetian Senate that the ‘story of my life’ which he had told to Desdemona would have been quite spellbinding enough. Godwin, Crabbe, Tench and Wordsworth all refer to this speech when they ponder the power of storytelling, Wordsworth alluding repeatedly, in poems and letters, to Othello’s tale of ‘moving accidents by flood and field’. More recently, Ross Chambers’ Story and Situation: Narrative Seduction and the Power of Fiction takes its cue from Othello, whose storytelling figuratively seduces the senators because they can believe it would
Narrative Order
19
have literally seduced a girl like Desdemona. As the Duke puts it, ‘I think this tale had won my daughter too.’ The point is, however, that Othello’s story not only seduces Desdemona, it persuaded her to marry him; while his subsequent storytelling before the court (the Duke’s court acting as a court of law) persuades a reluctant Brabantio to give the couple his retrospective paternal blessing. The real power of storytelling here, Shakespeare implies, lies in bringing about a contract and then in correcting the contract so as to reaffirm its proper sequence as a rite of passage. Story and Situation is a study of ‘how to do things with stories’, Chambers alluding to J.L. Austin’s exposition of ‘speech act’ theory, How to do things with words.26 And if Othello’s storytelling has illocutionary power (the power to make things happen) that is because it leads directly to the actual performatives of the father’s blessing, the marriage ceremony and the court judgement. Stories do not only resemble contractual relationships; they may also, it seems, derive their power from their ability to bring such relationships into being. For Wordsworth at least, Othello is a play about the morally dubious capacity of stories to persuade us to become characters in the social narrative constructed by contractual events such as marriages. George Crabbe, by contrast, hoped that stories could have that power, but wondered if in fact it was possible. His first embryonic verse tales appeared in ‘The Parish Register’ (1807), a poem based on the ‘fundamental social document’ in which the promises and performatives of baptisms, marriages and funerals are officially inscribed. Committed – both as a poet and as an Anglican clergyman – to maintaining narrative order, Crabbe nevertheless seriously doubted that such order could be maintained.27 He knew that some stories – romantic ones, stories with happy endings – were all too seductive, but suspected that true stories were less so. Indeed, his hostility to stories with happy endings is sometimes more than that, a suggestion that a truthful storyteller is a contradiction in terms: To me it seems their Females and their Men Are but the Creatures of the Author’s Pen; Nay, Creatures borrow’d and again convey’d From Book to Book – the Shadows of a Shade: Life, if they’d search, would show them many a change; The Ruin Sudden and the Misery strange! With more of grievous, base and dreadful things, Than Novelists relate or Poet sings:
20
Narrative Order, 1789–1819: Part I
But they, who ought to look the World around, Spy out a single Spot on Fairy-Ground; Where all, in turn, ideal Forms behold, And Plots are laid and Histories are told.28 In this passage from the 1810 poem ‘Ellen Orford’, endings – ‘Ruin sudden’ – are of the sort that undermine narrative, not of the sort that make it possible. Despite his religious convictions Crabbe’s stories do not anticipate a retrospective view from after death. It must be insisted of course that the ‘sudden’ and the ‘strange’ did not have to take the form – as Crabbe suggests they must – of ‘Misery’ and ‘Ruin’. Unpredictable ‘change’ could also be experienced as release, when the links of an imprisoning chain of events are broken. Nevertheless, Crabbe’s insistence that there is a force of circumstance which plays havoc with all ‘Plots’ and ‘Histories’ makes his a particularly radical statement of a widespread scepticism.
2 Samuel Johnson and the Order of Time
No narrative is perfectly chronological, but some are more chronological than others. Histories and biographies tend to be more chronological than novels and autobiographies. Chronology seems to claim an association with objective fact, significant deviation from chronology with artifice or subjectivity. Some twentieth-century narratives deliberately make it impossible for us to reconstruct the supposed actual order of events from the order in which the text presents them to us; and while we can reconstruct the actual order of events in The Prelude or Wuthering Heights we can do so only with great difficulty, and we cannot easily hold the actual order of events in our minds. By contrast Samuel Johnson’s Account of the Life of Mr Richard Savage, Son of the Earl Rivers (usually known as the Life of Savage) is methodically chronological, using analepsis (flashback) and prolepsis very sparingly. Apart from a brief introduction and a brief conclusion, the Life of Savage, like Savage’s life, begins at his birth and ends with his death. It is this effect of verisimilitude – this apparent similarity between the life and the Life – which commits biography to chronology. Introducing his own Life of Johnson (1791), Boswell wrote: I cannot conceive a more perfect mode of writing any man’s life, than not only relating the most important events of it in their order, but interweaving what he privately wrote, and said, and thought; by which mankind are enabled as it were to see him live, and to ‘live o’er each scene’ with him, as he actually advanced through the several stages of his life. Had his other friends been as diligent and as ardent as I was, he might have been almost entirely preserved. As it is, I will venture to say that he will be seen in this work more completely than any man who has ever yet lived.1 21
22
Narrative Order, 1789–1819: Part I
Johnson himself believed that ‘no species of writing is more worthy of cultivation than biography’ and that ‘there is no life of which a judicious and faithful narrative would not be useful’.2 It is of vital importance to Johnson that we should all learn to see that every life, including our own life, has the shape of a Life, a biography, a narrative in the third person. The importance Johnson attaches to biography is probably unique. It nevertheless represents an influential codification of what I have referred to, following Alasdair MacIntyre, as a ‘narrative idea of life’, an idea that was the common currency of British enlightenment writing; the narrative idea, for instance, implicit in the standard eighteenthcentury usage of a constellation of words for human existence and its narrative ordering: words such as ‘life’, ‘story’, ‘history’ and ‘character’. Nowadays we use the word ‘character’ to mean either a person’s moral identity – their personality – or a representation of a person in a fictional narrative. Normal eighteenth-century use of the word does not distinguish in that radical way between a person and their representation.3 Boswell offers us Johnson’s character rather than a description of his character. One’s character is one’s proper description. Similarly, in the passage I have just quoted, he talks about ‘writing a man’s life’ rather than writing the story of a man’s life. This is not because a human being and a text are thought to be the same (though his talk of ‘preserving’ Johnson comes oddly close to that) but because a life is believed to have the same shape as a Life and therefore to appear in its proper shape when it appears as a Life. Johnson is committed to this narrative idea of life with an urgency which suggests the strength of those self-deluding impulses – that ‘vanity’ – which it was required to correct. As Donald Davie said of the syntactic discipline of Johnson’s verse, ‘it is the mind which knows the power of its own potentially disruptive propensities which needs and demands to be disciplined’.4 With Johnson, narrative order is very much a matter of narrative discipline.5 The urgency of Johnson’s commitment to a narrative idea of life is suggested nicely by two particular features of his belief in biography. The first is that he lived a considerable part of his life very consciously as a biography-in-the-making: ‘I had the scheme of writing his life continually in view’, wrote Boswell, and ‘he was well apprised of this circumstance’ (p. 5). And the second is the human significance which attaches, in Johnson’s most powerful biography, the Life of Savage, to that chronological ordering of events characteristic of almost all biography.
Samuel Johnson and the Order of Time
23
Chronology and deviations from chronology have always been recognized as an important aspect of narrative. It is in part to help us think more precisely about this issue that Russian formalism distinguished between fabula and sjuzhet.6 The sjuzhet of Wuthering Heights begins on page one, with Lockwood’s first visit to Wuthering Heights; the fabula begins some 30 pages later and 30 years earlier with Mr Earnshaw’s departure for Liverpool. Readers who are trained literary critics can find plenty to say about narratives of this kind, where it is difficult, and perhaps impossible, to derive the fabula from the sjuzhet. Narratives where fabula and sjuzhet coincide, chronological narratives such as biographies and histories, have often proved less discussible: they seem to leave the reader with nothing to do. In this chapter I shall try to remedy this situation so far as the Life of Savage is concerned. I shall argue that the acknowledged power of this text does in fact have a lot to do with its chronological ordering. The text starts with three paragraphs of general philosophical reflection, in which Johnson asserts the vanity of human wishes and announces the biography that is to follow as an addition to those ‘mournful narratives’ which describe ‘the miseries of the learned’.7 The text ends with a description of Savage’s physical appearance and a summary of his positive and negative qualities as a writer and a human being. Between these prefatory and concluding paragraphs, Johnson gives us the events of Savage’s life more or less in the order in which he believes they occurred,8 from the unusual circumstances of his birth in ‘January 1697–8’ to his death in a Bristol prison in August 1743 and his burial ‘in the Church-yard of St Peter, at the Expence of the Keeper’ (pp. 5, 135). But no narrative is perfectly chronological, and the Life of Savage is no exception. The title itself tells us that we are about to read a biography and that the life of Savage has therefore come to an end: at the very beginning, ‘the end is already there’, as Sartre’s fictional biographer Roquentin put it. All narratives are written with the benefit of hindsight and Johnson’s is no exception: the end of the fabula (Savage’s death) puts in a kind of appearance at the beginning of the sjuzhet. However, although Roquentin was a biographer, it is notable that the example he gave was autobiographical: You appear to begin at the beginning: ‘It was a fine Autumn evening in 1922. I was a solicitor’s clerk at Marommes’. And in fact you have begun at the end. The essentially duplicitous character of narrative in Roquentin’s account surely depends, at least to some extent, on this choice of the
24
Narrative Order, 1789–1819: Part I
autobiographical form. Or at least, in biography, and notably in the Life of Savage, there is less attempt to conceal the fact that the end has already happened and that the sjuzhet could not begin unless the fabula was already over. The distance provided by hindsight is, for Johnson, the distance required for us to get our lives into perspective, to see them as they really are; to see, for instance, in the chronological order of events not just a random temporal succession but a real logic, an ‘order of Time’ as he calls it. There is a second, and equally inevitable, way in which Johnson’s narrative separates fabula and sjuzhet. The first sentence provides a number of examples: In the year 1697, Anne Countess of Macclesfield, having lived for some Time upon very uneasy Terms with her Husband, thought a public Confession of Adultery the most obvious and expeditious Method of obtaining her Liberty, and therefore declared, that the Child, with which she was then great, was begotten by the Earl Rivers. (p. 4) We are told about the Countess’s confession of adultery before we are told about her pregnancy and we are told about her pregnancy before we are told about the ‘begetting’. But no reader with an elementary knowledge of English grammar and human reproduction will have any difficulty converting this sjuzhet into a fabula in which the sequence of events is: conception, pregnancy, confession of adultery. However, readers of narratives are sometimes confused – at least for a moment – by such breaks with chronology; and some narratives trade on and deliberately prolong such confusion. Johnson, I believe, is very conscious of the possibility of confusion and wishes at all costs to minimize it. Evidence for this is provided by the fact that there is only one major departure from chronology in the Life of Savage, and by the fact that Johnson signals to the reader very deliberately that it is taking place. After describing Savage’s trial and conviction on a charge of murder (he had killed a man during a fight in a coffee-house), Johnson tells us that Mr Savage had now no Hopes of Life, but from the Mercy of the Crown, which was very earnestly solicited by Friends, and which, with whatever Difficulty the Story may obtain Belief, was obstructed only by his Mother. To prejudice the Queen against him, she made use of an Incident, which was omitted in the order of Time, that it might be mentioned together with the Purpose which it was made to serve. (p. 36)
Samuel Johnson and the Order of Time
25
Johnson then describes this ‘incident’ at some length. But he wishes to make it clear that the break with chronological narrative – with what Boswell called ‘events in their order’ and what Johnson here calls ‘the order of Time’ – is exceptional. And his mention of his own narrative procedure directs the reader to revise the story as quickly as possible: we must put the incident back in the place in the story from which it has been ‘omitted’. Johnson’s commitment to writing in ‘the order of Time’ is especially significant in the Life of Savage. One reason is that a methodically chronological narrative, which moves continually forwards in a progress from birth to death, throws into vivid relief the backwardlooking and regressive tendency of Savage’s life. From the moment that the 15-year-old apprentice shoemaker discovered that he was born the son of the Countess of Macclesfield, he devoted himself to an endless and obsessive attempt to undo what had subsequently been done to him. He wished to turn the clock back, return to zero, start again as the acknowledged son of his aristocratic mother. The relentless forward movement of the narrative helps to remind us that all such attempts to cancel out what separates us from the beginning of our lives are doomed to failure. The Life of Savage reveals the life of Savage as a series of attempts to abolish the ever-widening space that separates him from his origins. Freud, in Beyond the Pleasure Principle, speculated that the instincts may be inherently conservative and ‘tend towards the restoration of an earlier state of things’; a tendency which produces, according to Freud, the ‘impulse to repeat’.9 Whatever the general truth of Freud’s claim, the link between repetition and regression is demonstrably real in Savage’s case. This is so in a number of respects but especially in his ambivalent relationship to a long sequence of patrons and benefactors. Patronage is, quite openly, a quasi-parental relationship: but in Savage’s case the parental dimension becomes especially strong and peculiarly explicit. Sir Richard Steele asserted that ‘the inhumanity of his Mother had given him a Right to find every good Man his Father’ (p. 13). On the death of Mrs Oldfield, Savage is described as ‘wearing Mourning as for a Mother’ (p. 19). Savage himself at one point complained that a group of benefactors were keeping him on ‘leading strings’ (p. 116). Most of the relationships of this kind into which Savage enters are brief and discordant, and from the way Johnson describes them it seems clear that each involves an attempt – a necessarily futile attempt – to find his way back to his first relationship. So the forward-moving narrative, the narrative that follows the order of time, contrasts with and highlights Savage’s obsessive attempts to go
26
Narrative Order, 1789–1819: Part I
back to the beginning. But if this contrast separates Savage from Johnson and ourselves, we should not therefore assume that Johnson feels immune to the regressive tendencies to which Savage succumbs. On the contrary, I believe that Johnson regards these tendencies as highly contagious both for himself and for his readers. He is afraid that any narrative deviation from the order of time may lead him and his readers, like Savage, to believe that it is in fact possible to alter the order of time, and in particular to believe that the past can be altered. While analepsis certainly involves us in altering our understanding of the past, we may sometimes be confused because it seems to alter the past itself. In the case of the Life of Savage, is the ‘incident’ which was ‘omitted in the order of Time’ really compatible with the story-so-far into which it now has to be inserted? It probably is; but in many narratives it is possible to feel that the analeptic separation of fabula and sjuzhet allows a writer to covertly alter a character’s character (to use the word ‘character’ in its two modern senses). Analepsis can allow us to rewrite history without appearing to do so, to relive the past (live it over again, but differently). The new information about Henry Morton’s early life supplied in Chapter 13 of Scott’s Old Mortality is a notable case in point. Common sense and most philosophy tell us that the past, unlike the present and the future, is over and done with and is not open to human intervention.10 We constantly alter our understanding of our collective and individual pasts but we cannot, surely, alter the past itself. Analepsis, fantasy, Orwell’s Ministry of Truth and some postmodernist theory may tell us otherwise but common sense tells us that life follows the order of time. But in Savage’s case is this really so? If he seems set on undoing his own past and starting again, this may be because, contrary to common sense, he knows from bitter experience that it can be done. When Richard Savage was 15, the ‘Nurse, who had always treated him as her own son, died’ (p. 10), and he discovered from her papers that he was the son of the Countess of Macclesfield. In other words, Savage’s understanding of his own origins and history is abruptly and radically altered. He is now a different person from the one he thought he was, except of course that he is also the same person as he really was: he is a person who has, as we say, ‘lived a lie’ (his mother’s lie). For such a person, at such a moment – and well beyond that moment – the distinction between one’s understanding of the past and the past itself, while intellectually evident, may in emotional practice be impossible to make.
Samuel Johnson and the Order of Time
27
The discovery that what has established itself as your past is not your past is the stuff of which numerous plays, novels and autobiographies are made. Narratives of this kind – Oedipus Rex, Great Expectations – try to make us share something of the experience of their protagonists. They normally do this by separating the order of events in the fabula from the order of events in the sjuzhet, withholding from the audience or the reader some of the same information that has been withheld from the protagonist. It is as if Johnson discovered something close to a fictional stereotype – and indeed to the fantasy stereotype of superior birth which Freud called ‘the family romance’ – in the real life of his friend Savage and then wrote that life up in a way that is diametrically opposed to the standard fictional way. The Life of Savage does not participate, and does not allow us to participate, in the discovery and peripeteia experienced by Savage himself. Committed to the order of objective time, Johnson does describe Savage’s discovery, but only after he has described those facts of Savage’s birth and early life which are the substance of the discovery. More needs to be said at this point about just what it is that Savage discovered. What he discovered was that at a very early period of his life – within two months of his birth, according to Johnson – his past had in fact been changed. In trying to change the past Savage was only trying to do what had, he now discovered, been done to him. The following passage, in which Johnson describes the very complicated circumstances of Savage’s birth, explains this bizarre situation: In the year 1697, Anne Countess of Macclesfield, having lived for some Time upon very uneasy Terms with her Husband, thought a public Confession of Adultery the most obvious and expeditious Method of obtaining her Liberty, and therefore declared, that the Child, with whom she was then great, was begotten by the Earl Rivers. Her Husband, being as may be easily imagined, thus made no less desirous of a Separation than herself, prosecuted his Design in the most effectual Manner; for he applied not to the Ecclesiastical Courts for a Divorce, but to the Parliament for an Act, by which his Marriage might be dissolved, the nuptial Contract totally annulled, and the Child of his Wife illegitimated. This Act, after the usual Deliberation, he obtained, tho’ without the Approbation of some, who considered Marriage as an Affair only cognizable by Ecclesiastical Judges; and next year on March 3rd was separated from his Wife, whose Fortune, which was very great, was repaid her; and who
28
Narrative Order, 1789–1819: Part I
having, as well as her Husband, the Liberty of making another Choice, was in a short Time married to Colonel Bret. While the Earl of Macclesfield was prosecuting this Affair, his Wife was, on the tenth of January 1697–8 delivered of a Son, and the Earl Rivers, by appearing to consider him as his own, left none any Reason to Doubt of the Sincerity of her Declaration; for he was his Godfather, and gave him his own name, which was by his Direction inserted in the Register of St Andrew’s Parish in Holbourn, but unfortunately left him to the care of his Mother, whom, as she was now set free from her Husband, he probably imagined likely to treat with Great Tenderness the Child that had contributed to so pleasing an Event. It is not indeed easy to discover what Motives could be found to overbalance that natural Affection of a Parent, or what Interest could be promoted by Neglect or Cruelty . . . But whatever were her Motives, no sooner was her Son born, than she discovered a Resolution of disowning him; and in a very short Time removed him from her Sight, by committing him to the Care of a poor Woman, whom she directed to educate him as her own, and injoined never to inform him of his true Parents. (pp. 4–6) An annulment, as distinct from a divorce, alters the past. The couple who were married are retrospectively declared not to have been married; and the declaration is a performative utterance which actually brings about the (retroactive) change of status to which it refers. It does not simply change people’s understanding of the past; it changes the actual past. Infrequent and puzzling as the phenomenon of annulment may be, it is made possible by the nature of the marriage contract itself, on which it throws a sharp – if oblique – light. Like any contractual arrangement, any formal promising, the marriage contract – like the Act of Parliament which annuls it – depends upon performative language. And – again like any contract – it makes present and future strongly interdependent. In one sense, a promise made in good faith can subsequently be broken; but in another sense, as we have seen, the good faith of a promise, its force, can only be judged by the evidence of the future conduct to which the promise commits the promiser. Consequently, any breaking of the contractual promise can feel to the parties concerned like a weakening of the original contract, even though the original contract has not been altered. What annulment does is to give the force of law to this feeling that the original contract, which certainly took place, did not take place.
Samuel Johnson and the Order of Time
29
Confirmation of the challenge which annulment represents to narrative order comes from an unexpected source: Jorge Luis Borges’ short story ‘The Other Death’ (‘La Otra Muerte’), which tells the story of a man who died twice.11 Don Pedro Damián, who had survived the battle of Masoller in 1904, died in bed at his home in Entre Ríos in 1946; but he also died heroically at Masoller in 1904. The solution to this puzzle is that, as he lay on his death bed in Entre Ríos, in a delirium, he ‘lived his battle over again [revivió su batalla], conducted himself as a man, and in heading the last charge he was struck by a bullet in the middle of the chest’ (p. 216). Borges’s whimsical logic depends, of course, on his taking not only ‘relive’ literally; but also ‘annul’: In the Summa theologica, it is denied that God can unmake the past, but nothing is said of the complicated concatenation of causes and effects which is so vast and so intimate that perhaps it might prove impossible to annul [anular] a single remote fact . . . without invalidating the present. To modify the past is not to modify a single fact; it is to annul the consequences of that fact, which tend to be infinite. In other words, it involves the creation of two universal histories. In the first, let us say, Pedro Damián died in Entre Ríos in 1946; in the second, at Masoller in 1904. (p. 216) What the use of the dead metaphor of ‘annulment’ tells us, however, is that what is possible for marriage is not possible for death, since one is a cultural event and the other is a natural one. Weddings and marriage contracts bring about marriages; funerals do not usually bring about deaths. The case of annulment demonstrates the extent to which, contrary to common sense, the past can be altered. The fact of death demonstrates a common sense truth, that there are limits to the extent that the past can be altered. What is true of annulment is true, certainly in the Life of Savage, of ‘illegitimation’, except that the latter is an event that seems even stranger and more painful than annulment. This is because legitimacy is a cultural fact (like marriage) which can easily be mistaken for a natural one (like death). Johnson emphasizes the performative and retroactive force of the word in a subsequent paragraph: Such was the Beginning of the Life of Richard Savage: Born with a legal Claim to Honour and to Riches, he was in two Months illegitimated by the Parliament, and disowned by his Mother, doomed to Poverty
30
Narrative Order, 1789–1819: Part I
and Obscurity, and launched upon the Ocean of Life, only that he might be swallowed by its Quicksands, or dashed upon its Rocks. (p. 6) The extent of the paradox involved in the annulment and illegitimation can be grasped if we realize that there is no way in which the English tense system will allow us to describe such a situation without our description immediately requiring correction. If one had been speaking prior to the Act of Parliament, it would have been correct to say that the Earl and Countess of Macclesfield were a married couple and that the Countess’s son was legitimate. But if one is speaking after the Act has been passed, it becomes equally true to say that prior to the Act the couple were not married and the son was not legitimate. There are, in effect, two incompatible true versions of that earlier period, requiring what Borges calls ‘two universal histories’. Johnson perhaps gets round this problem by saying that Savage was ‘born with a legal Claim’, a statement that (since it may be taken to mean either that Savage had the right to inherit or only that he may have had the right to inherit) can be taken as more of less true whether it is said from a point of view prior to or subsequent to the Act of Parliament. Johnson registers the paradox most openly and vividly in the metaphor that follows, when he describes Savage as ‘launched upon the Ocean of Life, only that he might be swallowed by its Quicksands, or dashed upon its Rocks’. This is an image of being driven back (from ocean to sands and rocks); it even suggests that Savage never reached the ocean it says he reached. The ship was launched, and then unlaunched. Part of the resonance of the Life of Savage derives from the fact that Johnson has chosen to write about kinds of event – especially annulment and illegitimation – that are very specific, indeed bizarre, but which have a very wide and complex cultural resonance. The Borges short story demonstrates this, reminding us that annulment is now the everyday figurative name for an alteration of the past. But it is equally important to recognize that in Johnson’s time ‘illegitimation’ would have had a similarly powerful resonance. The meanings of ‘legitimate’ were much the same then as they are now: the word describes a child born out of wedlock; it is also the most general word in English for whatever is contrary to law or acceptable standards. However, although the meanings have not changed, the degree of perceived linkage between the two meanings has. In a society based – as eighteenth-century British society was based in so many
Samuel Johnson and the Order of Time
31
respects – on patrilineal inheritance (of names, property, titles, the crown), the link between the meanings of ‘legitimate’ was often strong and active. Two kinds of persons in particular drew the meanings close together: monarchs and sexually unreliable mothers. Thomas Hobbes had argued in Leviathan (1651) that ‘in the condition of mere nature, where there are no matrimonial laws, it cannot be known who is the father, unless it be declared by the mother: and therefore the right of dominion over the child dependeth on her will and is consequently hers’.12 Matrimonial laws could be thought of as transferring dominion to the father (or, in Locke’s view, to both parents) and as securing the safe passage of name, property, titles and crown down the male line. However, the apparatus of patrilineage remained vulnerable to the sexual and testamentary unreliability of women. At the heart of images of disorder in British enlightenment writing are the ‘ “savage” mothers’:13 the Goddess of Dullness in Pope’s Dunciad, Johnson’s Lady Macclesfield. The other context in which the connection between the two meanings of ‘legitimate’ was always close to the surface was in arguments about royal legitimacy and particularly the rival claims of the reigning Hanoverian and the exiled Stuart lines. The Life of Savage says nothing explicitly about such things; it says nothing, for instance, about jacobitism. But Johnson and Savage shared jacobite sympathies, and it is now argued that the extent of jacobitism among tories has been underestimated because of the understandable need to express those allegiances in coded forms.14 To a remarkable extent, the pattern of Savage’s life, as Johnson presents it, echoes the jacobite attempt to turn the clock back, to retie the broken Stuart line, to ‘restore an earlier state of things’ as had been done by the earlier Restoration in 1660. Furthermore, many tories, wrestling with the distinction between de jure and de facto legitimacy, must have felt that there were two legitimate royal lines in contention, producing (to use Borges’ phrase) two universal histories, two equally plausible but incompatible story lines. After the Battle of Culloden in 1746 most tories had to accept that the Hanoverian interlopers had now become the legitimate line; the lapse of time and the force of arms had now combined to push the Stuarts definitively into the past (the place where, since Scott, most people have seen them as having belonged from the start). The Life of Savage was published when that hope was still alive, immediately prior to the jacobite rising which culminated not only in the defeat of the Stuarts but in the final brutal incorporation of the highland clans into the commercially and juridically unified British State.
32
Narrative Order, 1789–1819: Part I
People who look for coded messages will usually find them, and I do not wish to argue that the Life of Savage is really a coded jacobite tract, or even that it is consciously intended as an allegory of the serious temptation which the jacobite cause could still represent for some tories in 1744. However, it is, among other things, precisely such an allegory. For Johnson, the ‘order of Time’ is in part the order of historical time. His commitment to narrative order is, among other things, a commitment to the status quo, the self-consciously progressive Hanoverian state, in tension with a subversive jacobite desire to return to the status quo ante. Johnson’s ordering of his materials in the Life of Savage reveals just how necessary such ordering is, just how strong and dangerous is the desire for a disorderly reordering of time.
Part II
This page intentionally left blank
3 Edmund Burke: Middles versus Beginnings and Ends
The French revolution of 1789 revived, for Edmund Burke, the fear that events in the recent British past might be ‘annulled’ (to use the term which he, like Samuel Johnson, used for the act of retroactive cancellation). In particular, the settlement in Church and State brought about by the ‘Glorious Revolution’ of 1688 might, he feared, be undone and the doctrinal, social and literal battles which had preceded it might be refought. This was something which Burke feared, but – and in this too he was like Johnson – his fear of it was in part a fear of his own desire for it. However, the way in which the Life of Savage and Reflections on the Revolution in France conceive of the interaction between past and present is very different, and the way in which they position themselves with respect to narrative is correspondingly different. For Johnson, as we have seen, the idea that it might be possible to restore an earlier state of things, the temptation to reject the status quo in the favour of the status quo ante, was what gave urgency and sinew to his commitment to ‘the order of Time’. For him, order in the sense of sequence – specifically, a sequence with a clear beginning and end – was an essential element of order in the sense of discipline and systematic arrangement. The French revolution of 1789 introduced an ingredient which, in the hands of Burke, transformed the Johnsonian recipe. This new ingredient was revolution conceived of as the creation of a radically new society, a society grounded on the destruction of the past. The French revolutionaries, Burke insisted, were set on ‘beginning [the state’s] reformation by its subversion’ (p. 194). This violent bringing together of ending and beginning converted a fundamental feature of narrative – its commitment to beginnings and endings – into a threat to social order. Recoiling from this bringing together of beginnings and endings, Burke emphasized the importance of middles, mediations. The Reflections 35
36
Narrative Order, 1789–1819: Part II
foregrounds those temporal processes of retrospection, anticipation and anticipated retrospection which are the mechanisms of narrative, but so as to postpone or put into the background what in other circumstances these mechanisms are designed to produce: narratives, with their beginnings and endings. Burke was indeed as keen on unbroken sequence as Johnson had been; indeed keener, since he feared those radical breaks which, in the form of beginnings and endings, are required to turn sequences into narratives. The temporal mechanisms of narrative – anticipation, retrospection and the anticipation of retrospection – are also the mechanisms of contract and promising, and these too are foregrounded by Burke. The Reflections is committed to a strange species of contractual promise that occludes the initial making and the anticipated termination of the contract, and occludes with them the material and bodily – and often violent – circumstances that are the necessary context for all contractual beginnings and endings. Paradoxically, however, the intensity with which Burke responded to the threat of what he took to be a new sort of revolution – a revolution unprecedented in its rejection of precedent – is partly explained by his half-acknowledged sense that it was not unprecedented at all. Events in France threatened to revive the dangerous passions of England’s own seventeenth-century Civil Wars, events which Burke was probably the first person to refer to as a ‘revolution’ (Reflections, p. 99). If the revolution in France was allowed to succeed, it might change the meaning of that part of the seventeenth-century British past, turning what had only been a dangerous interlude, an ‘interregnum’, into a beginning. Samuel Johnson had faced a choice between conservatism and reaction, the status quo and the status quo ante. The appearance of a third option – revolution as an absolute break with the past – created a relationship between past, present and future of a new kind, and one full of paradox. These complex temporal relationships invite the use of psychological concepts. Perry Anderson was right, if melodramatic, to argue that in the eighteenth century, by a classic process of psychological suppression, the Civil War was forgotten and its decorous epilogue, the Glorious Revolution of 1688, became the official, radiant myth of creation in the collective memory of the propertied class.1 For Burke, the French revolution (and in the Reflections the Revolution of 1688 always has a capital initial letter while the revolution of 1789
Edmund Burke: Middles versus Beginnings and Ends
37
never does) threatened to disturb that myth and to reorganize historical memory. Moreover, the phrase ‘psychological suppression’ encourages us to realize that the remembering of the Civil War, if it should happen, might be something like a re-membering of it. To recall it, in the sense of calling it back to memory, might be to recall it in the sense of bringing it back to life, in the first place by bringing back the murderous passions associated with it.2 If this is so, however, the task of warning the ‘propertied class’ of the danger facing them from the French revolution, which is the principal purpose of the Reflections, becomes particularly difficult, since it risks raising the very passions it seeks to keep suppressed. The peculiar form of address which Burke adopts in the Reflections is partly determined by the need to square this circle. He had to find a way of saying to his propertied readers: ‘don’t remember the Civil War!’ The Reflections is what we would now call an ‘open letter’. Nominally addressed to a young Frenchman – M. Dupont, a ‘gentleman in Paris’ – it is in fact intended for a readership of propertied ‘English’ (by which he probably means British) gentlemen. It is particularly intended for those of their number – probably a majority – who, at the time Burke was writing, were either welcoming events in France or suspending judgement in the expectation that the French were only trying to establish the kind of constitutional monarchy which the British had already achieved. Burke’s indirect mode of address helps him to tell English gentlemen what they should be thinking while appearing only to tell the young Frenchman what they (the English gentlemen) already are thinking; all of them, that is, except those dissenting intellectuals whose support for the French revolution was dangerously marked by ‘enthusiasm’ (p. 100). These latter, arguing with passion for the similarity between the Revolution of 1688 and the revolution of 1789, are presented as a wholly unrepresentative minority, scarcely a part of the English nation at all, an enemy within. By adopting an indirect way of addressing his intended English readers, Burke can disguise the extent to which he himself is in a minority and, indeed, the extent to which he himself is not wholly English. We should remember, after all, that if Burke had spoken of ‘we in England’ in the House of Commons – as he writes of ‘we in England’ in the Reflections – he would have spoken in an Irish voice.3 Indeed, it may be significant that Burke uses the expression ‘we in England’ but never ‘we English’. His use of the pronoun ‘we’ is itself slippery. Since he is addressing a Frenchman, Burke’s ‘we’ always means ‘I and they’; but since he is
38
Narrative Order, 1789–1819: Part II
really addressing English gentlemen, the other sense of ‘we’ (to mean ‘I and you’) is seldom wholly absent. Burke as well as his readers are subtly positioned by this shifty pronoun. It helps him to procrastinate his identity as an adopted Englishman: an Irishman with Catholic connections who became an Anglican member of the English landed class.4 Ostensibly written by an English gentleman to a young Frenchman, the Reflections is not quite written by the former or to the latter. The Reflections begins as an attack on A Discourse on the Love of Our Country, an address to the Revolution Society delivered by the Reverend Dr Richard Price on the one hundredth anniversary of the 1688 Revolution after which the society was named. Burke’s aim is to contest Dr Price’s interpretation of the 1688 Revolution, and particularly the suggestion that it resembles the contemporary revolution in France. However, although it is these two revolutions which Burke spends most of the Reflections talking about, he also refers, as we have seen, to a third revolution: ‘These gentlemen’ in the Revolution Society, Burke suggests, in all their reasonings on the Revolution of 1688, have a revolution which happened in England about forty years before, and the late French revolution, so much before their eyes, and in their hearts, that they are constantly confounding all the three together. It is necessary that we should separate what they confound. (pp. 99–100) But the process of ‘separating what they confound’ is less clear-cut than this suggests. Burke certainly does wish to ‘separate’ the ‘Revolution of 1688’ from the ‘revolution’ of 1789; but what he wants to say about the relationship between the ‘revolutions’ of 1648 and 1789 is far less clear. He does in fact, as I have suggested, wish both to separate them and confound them. Indeed, the relationship between all three ‘revolutions’ is in fact far less clear, in the Reflections, than Burke wishes us to believe. The Revolution of 1688, Burke insists, was far less radical, and the French revolution of 1789 is far more radical, than dissenters like Dr Price claim. The former was not an act of popular sovereignty in which the people established the right to choose their own leaders; it was rather ‘an act of necessity’ (p. 102) which made ‘a small and a necessary deviation from the strict order of a regular hereditary succession’ (p. 101), following the ‘virtual abdication’ (p. 112) of James II. The eagerness of Dr Price and his friends to identify 1789 with 1688 only demonstrates to Burke that they are in fact identifying 1789 with
Edmund Burke: Middles versus Beginnings and Ends
39
1648 (‘the revolution that happened in England about forty years before’). The true ancestry of the English dissenters gathered in the ‘Revolution Society’ should not be traced, as they themselves claimed, to the Revolution of 1688 but to the regicide ‘enthusiasts’ (p. 100) who instigated the ‘rapture of 1648’. The seventeenth-century regicide, the Reverend Hugh Peters is the ‘precedent’ (p. 158) for the ‘political Divine’ Dr Price (p. 99). The style of argument Burke is using here was not wholly new. Most eighteenth-century political argument had taken the form of disagreement over the true interpretation and ownership of the 1688 Revolution Settlement. It was indeed one of the strengths of the Settlement that it was possible for quite a wide range of interests among the propertied classes to lay plausible claim to it. Correspondingly, it had been common, certainly till mid-century, to indict one’s opponents by claiming that, far from being committed, as they claimed, to the status quo established by the Revolution Settlement – the moderate middle way embodied in constitutional monarchy and the Anglican Church – they were really extremists, aiming to restart the Civil War and, in doing so, to draw the common people into politics as the Civil War had done. Whigs would suggest that tories were secret jacobites, tories that whigs were really republicans. And there was often some truth in this rhetoric. Or at least there was frequently a tension, which we have already identified in Johnson’s Life of Savage, between a commitment to the status quo and a residual desire to return to some version of the status quo ante, whether Stuart or Commonwealth. Unlike Johnson or Swift, Burke was not a tory; but when he is attacking Dr Price and the dissenters, he is writing very much in the spirit and style of the so-called ‘tory satirists’ of the immediate post-1688 period. Indeed he recalls Swift in the immoderate intensity and venom with which he defends the moderate centreground. He thereby betrays the fragility of his own attachment to it, the insecurity of his allegiance, an insecurity associated, as it had been for Swift, with his mixed English–Irish identity. The new factor, which transforms this familiar type of eighteenth-century disputation, is the French revolution. Two years before the execution of the French king and queen, Burke was claiming in the Reflections that for the French revolutionaries regicide, and parricide, and sacrilege, are but fictions of superstition...the murder of a king, or a queen, or a bishop, or a father, are only common homicide. (p. 171)
40
Narrative Order, 1789–1819: Part II
It was passages like these which made most contemporary readers think the Reflections exaggerated when they read it in 1790, and which led many of the same readers, looking back from the events of 1792–3 (the September Massacres, the execution of the king and queen, the founding of the Republic), to think it had been remarkably prophetic. Iain McCalman has plausibly argued that ‘Burke’s early and extreme hostility to the French revolution owed much to his personal experiences of popular violence in the 1780s’, in particular the anti-Catholic Gordon Riots of June 1780. The Gordon Riots in their turn, McCalman suggests, had their effect partly because Burke – and others such as Gibbon and the Burneys – ‘saw in them an eruption of seventeenthcentury enthusiasm’.5 Burke therefore anticipated ‘regicide’ in France (saw violent death written into the Declaration of the Rights of Man and Citizen) because regicide had, for the English, already taken place. This is not exactly what Burke says of course; but then McCalman’s word ‘trauma’ – like Anderson’s phrase ‘psychological suppression’ – rightly suggests events or impulses whose strength is registered in the fact that they cannot be fully or directly acknowledged. The Reflections does speak about the English Civil War and it does suggest a link between the Civil War and the revolution in France; but it is also at crucial moments silent about the Civil War and explicitly denies that link between the seventeenthcentury English ‘revolution’ and the modern French one. Burke is as determined to ‘separate’ the French revolution from the ‘rapture of 1648’ as he is to ‘confound’ them. When Burke talks about the French revolution’s British supporters, the ‘rapture of 1648’ is certainly prominent. But when he goes on to talk about the French revolution itself, his argument has an exactly opposite tendency, since he is determined to demonstrate that the events in France have absolutely no precedent, either in France or anywhere else. He spends some time examining, one by one, acts of rebellion – often bloody and apparently thoroughgoing – which might be put forward as precedents for the revolution in France. But he brings these examples forward only in order to demonstrate that the French revolution is fundamentally different from all of them:
Other revolutions have been conducted by persons, who whilst they attempted or effected changes in the commonwealth, sanctified their ambition by advancing the dignity of the people whose peace they troubled . . . These disturbers were not so much like men usurping
Edmund Burke: Middles versus Beginnings and Ends
41
power, as asserting their natural place in society . . . Such was . . . our Cromwell. Such were your whole race of Guises, Candés, and Colignis. (pp. 136–7) No doubt Burke is attempting to separate Cromwell from his less gentlemanly associates with their more democratic intentions. Nevertheless, this is the same Cromwell whom Burke himself, in a letter to his son Richard Burke written in 1792, was to describe as a ‘regicide usurper . . . sent to subdue and confiscate . . . the Irish nation;’6 Cromwell was, that is to say, at the very centre of the ‘rapture of 1648’, a prime example of the sort of ‘theological politician’ Burke is so keen to twin with the ‘political theologians’ Hugh Peters and Richard Price. It has been argued, notably by Jon Mee, that Burke links the French revolutionaries to their English dissenting supporters and to seventeenthcentury English sectaries in terms of their common ‘enthusiasm’, a dangerously contagious zeal paradoxically linking cold rationalist to religious fanatic.7 It can, however, be argued that it is precisely the protean and chameleon character of the term ‘enthusiasm’ which is significant at this point: a term which, looked at from one angle, links the French revolution back to the English one, but which, looked at from another angle, makes no such precise claim. The term, as Burke uses it, can be seen as a hinge between opposed tendencies in Burke’s argument about the French revolution. He wishes both to ‘separate’ and to ‘confound’ the English and the French revolutions, to suggest that the former was a precedent for the latter and that the latter is absolutely without precedent. In arguing that the French revolutionaries are doing something unprecedented, Burke is taking his cue from the claims which, he believed, the revolutionaries were themselves making. The French National Assembly is determined, he insists, to abolish the past and to start human society again from scratch. It has ‘chosen to consider the French as a people of yesterday, as a nation of low-born servile wretches until the emancipating year of 1789 . . . as a gang of Maroon slaves, suddenly let loose from the house of bondage’ (p. 123). Royal pensioners do not get the pensions they were promised because ‘their services had not been rendered to the country that now exists’. The complete break with the past takes the form of a rejection of inherited promises and legislative enactments: ‘No acts of the old government of the kings of France are held valid in the National Assembly, except for its pecuniary engagements’ (p. 208). The Assembly is now considering ‘how far it is bound by the treaties made with other nations under the former government’ (p. 209). The revolution does not just break
42
Narrative Order, 1789–1819: Part II
inherited promises; it makes a principle of doing so. The revolutionary brings ‘himself to that pitch of presumption, to consider his country as nothing but carte blanche, upon which he may scribble whatever he pleases’ (p. 267). The French revolutionaries begin the new by destroying the old. They bring ends and beginnings, death and birth, together. We can describe Burke’s vision in this way, as associating the end of a regime with the death of a person, regicide with parricide, because Burke does so himself: We have consecrated the state, that no man should approach to look into defects or corruptions but with due caution; that he should never dream of beginning its reformation by its subversion: that he should approach to the faults of the state as to the wounds of a father, with pious awe and trembling solicitude. By this wise prejudice we are taught to look with horror on those children of their country who are prompt rashly to hack that aged parent in pieces, and put him into the kettle of magicians, in hopes that by their poisonous weeds, and wild incantations, they may regenerate the paternal constitution, and renovate their father’s life. (p. 194) In this passage he is adapting traditional analogies – between the polity and the body, between the polity and the family – so as to suggest a perverse entanglement of ending and beginning, the dead and the newborn, father and child.8 The idea of the body politic and the family politic are linked in the expression ‘the paternal constitution’, and in the two lurid images of unnatural family and bodily relationship. Both the latter are images of unnatural relations between fathers and children. The second – perhaps adapted from the story of Medea and her offspring – is an image of parricide. The first alludes to the biblical story of Noah and his sons in which Ham ‘saw the nakedness’ of his drunken father. Edmund Leach claimed that the biblical phrase ‘saw the nakedness’ indicates homosexual incest, and this is certainly an image of a son in too close a contact with the sexual being of a disabled father.9 The two images taken together – parricide and father–son ‘incest’ – can usefully be seen in the Levi-Straussian context suggested by Leach’s discussion of the biblical story, the two images forming a binary pair, one of too much distance between close kin, the other of too little distance. Against this fearsome marriage of ending and beginning – of ‘beginning [the state’s] reformation by its subversion’ – Burke asserts
Edmund Burke: Middles versus Beginnings and Ends
43
middles, those mediations he believes make change bearable and which he sees so benignly at work in English life: From Magna Charta to the Declaration of Right, it has been the uniform policy of our constitution to claim and assert our liberties, as an entailed inheritance derived to us from our forefathers, and to be transmitted to our posterity; as an estate specially belonging to the people of this kingdom without any reference whatever to any other more general or prior right. By this means our constitution preserves an unity in so great a diversity of its parts. We have an inheritable crown; an inheritable peerage; and an house of commons and a people inheriting privileges, franchises, and liberties, from a long line of ancestors . . . A spirit of innovation is generally the result of a selfish temper and confined views. People will not look forward to posterity who never look backward to their ancestors . . . Whatever advantages are obtained by a state proceeding on these maxims, are locked fast as in a sort of family settlement; . . . Our political system is placed in a just correspondence and symmetry with the order of the world, and with the mode of existence decreed to a permanent body composed of transitory parts; wherein, by the disposition of a stupendous wisdom, moulding together the great mysterious incorporation of the human race, the whole, at any one time, is never old, or middle-aged, or young, but in a condition of unchangeable constancy, moves on through the varied tenour of perpetual decay, fall, renovation, and progression . . . In this choice of inheritance we have given to our frame of polity the image of a relation in blood; binding up the constitution of our country with our dearest domestic ties; adopting out fundamental laws into the bosom of our family affections. (pp. 119–20) According to this passage, ‘we’ are always in process, never at the beginning or ending; always in transit, never setting out or arriving. And among the elements of English life which help to make this so, one is given particular prominence, being printed in italics. ‘It has been the uniform policy of our constitution to claim and assert our liberties as an entailed inheritance derived to us from our forefathers, and to be transmitted to our posterity.’ In the system of property inheritance known as ‘entail’ the owner at any one time is no more than life tenant, constrained by the terms of his inheritance to keep the property intact to be passed on to his eldest
44
Narrative Order, 1789–1819: Part II
son or, in the absence of sons, to the closest male relative (in the case of the Bennet property in Pride and Prejudice, to Mr Collins).10 Entail was for Burke both an important instance of and metaphor for the English way of life because – Mr Collins notwithstanding – it ensured that so many different things tended to get handed down from father to eldest son: estate, name, title, crown, and seat in parliament. Furthermore, it was a system that embodied and enforced that particularly strong interdependence of past, present and future which Burke wished to preserve, tying people to past generations by the same bond that tied them to future ones. There are potential difficulties for Burke in basing his social philosophy as much as he does on entail and the ‘family settlement’. For one thing, although it provides a powerful image of continuity, an over-emphasis on it – making it a front-line defence against the revolutionary enemy – may threaten the continuity it is supposed to embody. The reason for this is to be sought in the way Burke understands ‘family’. Where the traditional analogy between family and polity had normally envisaged the family as the household, Burke sees the family, whether in itself or as a model for the polity, principally as the lineage – the patrilineage – and the kin group (the ‘relation in blood’). The problem with this is that although patrilineage does privilege kin relations (especially the father–son relationship) it cannot be exclusively constituted by them. A lineage requires, if it is to survive (if it is to avoid the fate of Poe’s House of Usher), the continual introduction of women from outside who are not kin. By conflating family and kin, and family and nation, Burke leaves very little space for women in their sexual capacity. Other families (other countries) become enemies with whom no intermarriage is possible. Foregrounding patrilineage as the English nation’s front line of defence against the French revolution, associating the English nation with the family in this particular way, Burke shows that he has forgotten Levi-Strauss’s first rule: the fundamental relationship in society is not the family relationship but the relationship between families. A patrilineal family-nation turned defensively in on itself will aggravate the parricidal and incestuous tendencies it attributes to the external revolutionary enemy. It will also oppose itself to the foreign enemy in a way that has some of the characteristics of an internecine struggle or civil war, in which different families – which need each other in order to survive – turn upon one another. Burke is, in effect, preparing the English for two important developments: a sexual crisis (comparable to the crisis identified by Malthus in his 1798 Essay on the Principle of Population) and a ‘civil war’ against revolutionary France.
Edmund Burke: Middles versus Beginnings and Ends
45
Another problem which might follow for Burke from tying his vision of English life so closely to entail is that this system for the holding and transmission of landed property had only come to encompass a substantial proportion of such property quite recently, in the seventeenth century. Burke gets round this problem by means of a characteristic rhetorical manoeuvre. We assert our liberties, he claims, ‘as an entailed inheritance’: the phrase ‘as an’ can be read to mean either ‘like an’ (in which case the reference to entail is metaphoric) or ‘in the form of’ (in which case the reference is literal or, more strictly, metonymic). Indeed the italicization of ‘entailed inheritance’ can be read in either fashion as well: as emphasizing the (literal) importance of the institution (in which case the italics are like underlining), or as emphasizing its figurativeness (in which case the italics are like quotation marks). This oscillation between the literal and the metaphoric – sometimes slow and foregrounded, but sometimes (as in this case) so fast as to be subliminal – is how Burke’s arguments characteristically breath. It is, as we shall see, an oscillation which Burke characteristically uses to contain the existence, within his argument, of contrary tendencies. Burke’s reference to Magna Carta may seem, at first sight, to raise similar problems to those raised by his reference to entail. Is he tracing the English ‘idea of inheritance’ back to Magna Carta as the original contract, despite his apparent opposition to beginnings and despite the radical tendencies of contract theory? After all, as Marilyn Butler observes, the ‘notion that society originated in a combination for mutual protection’ was popular among radicals in this period because it ‘implied that government rested on the consent of the governed’.11 However, Burke’s preferred version of contract is designed to avoid both these problems. ‘Society is indeed a contract’, he writes, but the state ought not to be considered as nothing better than a partnership agreement in a trade of pepper or coffee, callico or tobacco, or some such other such low concern, to be taken up for a little temporary interest, and to be dissolved by the fancy of the parties . . . it is not a partnership in things subservient only to the gross animal existence of a temporary and perishable nature . . . It [is] a partnership not only between those who are living, but between those who are living, those who are dead, and those who are to be born. Each contract of each particular state is but a clause in the great primaeval contract of eternal society, . . . connecting the visible and invisible world, according to a fixed compact sanctioned by the
46
Narrative Order, 1789–1819: Part II
inviolable oath which holds all physical and all moral natures, each in their appointed place. (pp. 194–5) There never was an inaugurating contractual act, or not, in any case, one that can be known by human beings. Rather, any human being at any one time is always already implicated in an endless and beginningless process of keeping a promise. In fact when Burke says that ‘from Magna Charta to the Declaration of Right, it has been . . .’, he is not so much identifying Magna Carta as a beginning as saying ‘wherever in our past we look’. The crucial phrase in this respect is ‘it has been’, and the present perfect tense in which that phrase is written. Burke’s use of this tense plays a central part, throughout the Reflections, in the occluding of beginnings and endings. This tense, the present perfect, was to become, just a few years later, a staple of Wordsworth’s lyricized narratives, notably in such openings as ‘Five years have passed’ (‘Tintern Abbey’, 1798) and ‘Strange fits of passion I have known’ (1800). Its prominence in his verse may be one of the most important things Wordsworth took from Burke, though he characteristically changes Burke’s ‘we’ to ‘I’. In the context of Wordsworth’s use of it, this tense has been called ‘the perfect of experience’12 and it exhibits two features which, taken together, define its paradoxical relationship to narrative order. In the present perfect, unlike the past simple, the past opens out into the present of the utterance in which the past is represented, drawing the past into the orbit of that present. The same movement causes the beginning of the temporal process (and the end of whatever preceded that process) to disappear from view. And if, in the particular case we are discussing, Magna Carta is a beginning, it is a beginning only visible, as it were, from the inside of the temporal envelope linking it to our present. Its dependence on a prior ending (the acts of force which were a precondition for the signing of the Charter) can only be dimly seen. It is only when the whole process in which we are involved has itself come to an end – when we speak of ‘what we did’ rather than of ‘what we have done’ – that the beginning comes into proper focus. The present perfect tense, then – the effects of which will be examined in more detail when we discuss the journals and narratives of Watkin Tench – is a tense which draws our attention away from the inseparability of endings and beginnings, an inseparability which the French revolution had made so terrifyingly visible to Edmund Burke.
Edmund Burke: Middles versus Beginnings and Ends
47
Burke is in any case very careful to stress that Magna Carta was not, and did not offer itself as, a beginning. It presented the liberties it asserted as inherited ones. In this respect, as in many others, the Reflections is written very much with William Blackstone’s 1767 Commentaries on the Laws of England in mind. Blackstone had asserted the importance in England of the common law, and it is one of the characteristic marks of English liberty, that our common law depends upon custom. And one thing required ‘to make a particular custom good’ is that it have been used so long, that the memory of man runneth not to the contrary. So that if any one can show the beginning of it, it is no good custom.13 None of this should be taken to mean, however, that Burke denies the existence of beginnings and endings, or believes they began in 1789. But while an estate or a crown cannot always have been inherited, we have always tended, he argues, to forget our origins; and it is as well that we should do so. This would particularly be the case in those parts of Ireland most directly effected by the colonial ‘plantation’ of the country under James I and Oliver Cromwell. As Burke had written to a fellow Anglo-Irishman, Captain Thomas Mercer of Newry, County Down earlier in 1790: It is possible that many estates about you were originally obtained by arms, that is, by violence . . . . but it is old violence; and that which might be wrong in the beginning, is consecrated by time, and becomes lawful.14 Burke prefers old money and old violence (old money is old violence). ‘It is possible’ that many estates in Captain Mercer’s neighbourhood were originally obtained by violence but, Burke implies, that is a good reason for not nosing about in the records to establish the truth of the matter. It is best not to try disperse the mists of time. The importance which Freud was to attach to our forgetfulness of our early individual lives is quite directly relevant to Burke’s position here, since what we find if we do investigate our origins is, for both writers, parricide and (though of course more explicitly in Freud) incest. Furthermore, while the beginning of the life of an individual is quite
48
Narrative Order, 1789–1819: Part II
different from the beginning of the life of a society or a state, the characteristics of the two are brought closer by Burke’s constant use of the pronoun ‘we’. ‘We’ is the controlling pronoun of the Reflections, as the present perfect is its controlling tense, the two combining to constitute the proper voice of the patrilineal family-nation. This ‘we’ both helps to give the individual access to the quasi-immortality of the nation and attributes to the national collective the sort of memory of its own past (and the sort of amnesia about its own origins) characteristic of an individual. What then exists at the beginning of this collective memory, as of an individual’s memory, is not in fact a beginning but – as Prospero put it, explaining to his daughter Miranda why she could not tell the story of her own early life – ‘the dark backward and abysm of time’. The technical name – in political theory – for the view of rights expressed in Burke’s letter to Thomas Mercer is prescriptionism. In a fine essay, Terry Eagleton has argued that rights, for a prescriptionist like Burke, are validated by ‘the recounting of a certain narrative’. It is clearly Nietzsche’s or Sartre’s understanding of narrative Eagleton has in mind: What bestows rights and titles for the prescriptionist is history itself . . . It is the temporal lapse between an origin and the present which retroactively justifies that origin, so that history moves forwards while its rationales travel backwards. In this sense, authority is always deferred, so that my right to the peerage I fraudulently acquired last Thursday will become luminously self-evident in two hundred years’ time.15 But if this is how prescriptionism works, suggests Eagleton, have not we found the logical flaw in Burke’s argument? After all, if a title will eventually be legitimated by the sheer passage of time, why not prospectively read this authority back into its origin, fraudulent or violent as that origin may have been? (p. 42) In other words, why not allow the French revolutionary regime time to become immemorial? However, while Eagleton’s argument is a good one, Burke anticipated it: If [the French] had set up this new experimental government as a necessary substitute for an expelled tyranny, mankind would anticipate
Edmund Burke: Middles versus Beginnings and Ends
49
the time of prescription, which, through long usage, mellows into legality governments that were violent in their commencement. (p. 276) The National Assembly has not, however, presented itself as doing any such English thing: ‘we have their own word for it that they have made a revolution’ (p. 276). One of the crucial differences between 1688 and 1789, and what prevents us from allowing the 1789 regime time to become immemorial seems, then, to be the difference between the way in which the protagonists in the two events represented what they were doing. And yet this is not quite the right way to state the matter. The French National Assembly’s ‘own word’ for what they are doing is ‘revolution’ (or rather, révolution). But Burke’s phrase ‘their own word’ has a double meaning: it also refers to the performative act of promising, ‘giving your word’. Burke is, in effect, referring to the French revolutionaries as ritually naming their undertaking as a ‘revolution’ and swearing allegiance to it. This focus upon what we would now call ‘speech acts’ points to the subtlety (indeed the slipperiness) of Burke’s understanding of the relationship between acts and representations of acts. On the one hand he insists that these crucial written and spoken acts – contracts, promises, Acts of Parliament – are indeed actions, things that people do. On the other hand, he also wishes to preserve the distinction between acting (in the sense of doing things) and the act of representing people doing things (both in the sense of describing their actions and in the sense of imitating their actions). When he says, of Dr Price and the dissenters, that ‘we must recall their erring fancies to the acts of the Revolution which we revere, for the discovery of its true principles’ (p. 100), we are likely to think that the contrast between acts and principles – with the terms italicized for emphasis – tells us that ‘acts’ very definitely means actions. By their acts, not their words, shall ye know them, is what he seems to be saying. However, it turns out that most of the acts he has in mind are actually upper-case ‘Acts’ of Parliament and that those actions which might not square with the Acts are the kind of act (in particular, acts of naked force) which we should be prepared, if at all possible, not to inspect too closely. The italicization of acts turns out to be as slippery in its meaning as the italicization of ‘entailed inheritance’. Read at first as an alternative to underlining, it can subsequently be understood as having been an alternative to quotation marks. Burke wishes to both make and blur the distinction between, on the one hand, those speech acts which initiate our performing (acting) lives – our lives as characters
50
Narrative Order, 1789–1819: Part II
in the Johnsonian sense – and, on the other hand, those acts which make Acts possible and necessary, acts which he strongly associates with nakedness (sexual acts, childbirth, death, acts of naked violence). He wishes to reveal these naked acts to us in something like a subliminal fashion, for just so long as is required to remind us how dangerous it is for us to dwell upon them. The acts performed in 1688 which Burke regards as most important were Acts of Parliament. He quotes at considerable length from parliamentary statute, demonstrating how packed with quotations from previous similar commitments those statutes themselves are. The 1688 Settlement, he writes, includes a clause, containing a most solemn pledge, taken from the preceding act of Queen Elizabeth, as solemn a pledge as ever can be given in favour of an hereditary succession, and as solemn a renunciation as could be made of the principles by [the Revolution Society] imputed to them. ‘The lords spiritual and temporal, and commons, do, in the name of the people aforesaid, most humbly and faithfully submit themselves, their heirs, and posterities for ever; and do faithfully promise, that they will stand to, maintain, and defend their said majesties, and also the limitations of the crown, herein specified and contained, to the utmost of their powers. (pp. 103–4) Burke is quoting the 1688 Act which is itself quoting an Elizabethan Act; and the Elizabethan Act is making a promise not only on behalf of its actual parliamentary authors but on behalf of future generations that include not only the drafters of the 1688 Act and Edmund Burke’s own generation but – so Burke would have us accept – subsequent generations of English people then not born, including our own. In the system which Burke’s text describes and aims to embody, it is as it is under entail: the individual – person, landowner, generation, legislative assembly, monarch – is only the life-tenant, the current keeper of a promise. If this is so, then the Sartrian logic which Eagleton describes, according to which ‘history moves forward but its rationales move backwards’, is not only the logic of narrative; it is also the logic of promising. And if it is ‘time’ we are dealing with here, it is very much socialized time, time as promises and contracts construct it. And what you are doing in anticipating the time of prescription if not just hanging around waiting for time to pass, but keeping a promise: something that
Edmund Burke: Middles versus Beginnings and Ends
51
might involve, if you were an eighteenth-century landed gentleman, raising rents and marrying profitably so that you could hand the entailed property on unburdened by debt. Moreover, it needs to be emphasized once again that what Burke foregrounds is the mechanism of narrative, that cat’s cradle of anticipation, retrospect and anticipated retrospect that Wordsworth enrols Dorothy to help him weave for himself in ‘Tintern Abbey’ and which governs so much of the poetry of Wordsworth, Coleridge, Keats and Shelley. And what is, by the same token, pushed into the background, always deferred, is what those mechanisms are, in other circumstances, there to bring into being: narratives, with their beginnings and endings. The mechanism of narrative and of promising, as I have described it, is one in which past, present and future are very tightly interrelated, with the lines of cause and effect travelling in both directions, from future to present to past as well as from past to present to future. If narrative beginnings depend upon endings and the anticipation of endings, and if promises are only properly made if they are kept, then what is done in the present retroactively determines the shape and meaning of the past, and the meaning of our own actions will be defined, in their turn, by future generations. This is an understanding of time that is palpable in the poetry of Keats and Shelley: it begins in Burke’s Reflections. And one feature of the interrelation needs particular attention if we are to fully understand the threat which the revolution in France represents for Burke. One of the things he most fears is the threat posed by the revolutionary present, and the likely revolutionary future, to the British past: the threat that the past may be remembered. However, Burke presents this threat in a displaced way. Nothing at all is said about it in his discussion of the French events themselves. He does not dwell, for instance, on the effect which the French revolutionaries’ cancelling of inherited treaties and engagements might have on their own past. He focuses instead on the dangers presented to the English past by the English dissenting supporters of the revolution in France. If, for the dissenters in the Revolution Society, 1688 marked the moment when the principle of popular sovereignty – the belief that we ‘may choose our own governors’ – was properly established, then, Burke asks do these theorists mean to imitate some of their predecessors, who dragged the body of our antient sovereigns out of the quiet of their tombs? Do they mean to attaint and disable backwards all the kings that have reigned before the Revolution [of 1688], and consequently to stain the throne of England with a continual usurpation? Do they
52
Narrative Order, 1789–1819: Part II
mean to invalidate, annul, or to call into question, together with the title of the whole line of our kings, that great body of our statute law which passed under those whom they treat as usurpers? To annul laws of inestimable value to our liberties . . . ? (p. 107) What is involved in this bizarre process of ‘disabling backwards’? Surely very much what was involved, for Samuel Johnson, in the annulment of the Countess of Macclesfield’s marriage and the illegitimation of Richard Savage. Indeed, the word ‘annul’ appears twice in the passage just quoted. Annulment, we recall, is a speech act, a juridical-legislative event, which corresponds in illocutionary force to the marriage vows which it retroactively cancels. The difference is that for Johnson the possibility of changing the past is, on the face of it, restricted to this one unusual phenomenon of marital annulment. The idea that past and present may always interact in this way and that other old battles (including literal ones) may be refought is no more than implicit in the Life of Savage. Johnson is committed to the irreversibility of events, the ‘order of Time’. Burke, by contrast, has foregrounded the narrative mechanisms of contract, the two-way interactions between past, present and future which they involve. He has done so in order to hold beginnings and endings at bay but he finds, as in this description of ‘disabling backwards’, that he has only displaced the problems he is trying to escape. History’s rationales are travelling backwards with a vengeance. The process by which the kings who ruled before 1688 might be ‘disabled backwards’ is like the process by which, in his theory of prescription, the passage of time can ‘mellow into legality governments that were violent in their commencement’. The mellowing and the disabling backwards are respectively the malign and the benign forms of the same mechanism, by means of which what happens later can alter what happens earlier. It is in this context that Eagleton rightly makes a link between the doctrine of prescription and Freud’s idea of ‘deferred action’ or nachträglichkeit. ‘Deferred action’ describes the way in which an event may acquire its traumatizing charge retroactively, in the light of subsequent events.16 This is clearly relevant to the way in which we may think of events such as the Gordon Riots or the French revolution as traumatic for the author of the Reflections. The riots may have been traumatic at the time, but they may also have become so retroactively in the light of the subsequent events in France. Burke’s stress on the two-way relationship between present and past
Edmund Burke: Middles versus Beginnings and Ends
53
necessarily brings this type of double-logic to the surface, making it possible, for instance, for the French revolution to convert the seventeenth-century Civil War into a ‘revolution’ (something it had not been called before), turning an isolated republican interlude – the ‘interregnum’ – into the ‘precedent’ of 1789. In discussing the Life of Savage I described the annulment of Lady Macclesfield’s marriage as an unmarriage, an act (and an Act) which really did rewrite the past, insofar as the past was constituted in the first place by written – or spoken – agreements. This was, I suggested, a process which made it possible to tell two incompatible true stories of the one marriage. And I suggested that, in the light of the close connection between the different meanings of ‘legitimate’ in this period, it might be appropriate to see a political story – a jacobite story – echoed in Johnson’s story of marriage and bastardization. The Reflections makes the same connection between the political and the familial, but more explicitly and from the opposite – the political – direction. Burke’s political sense of ‘annul’ has open familial connotations because the annulment of a royal marriage had been important in England’s political history, as all sexual matters must be in a hereditary system. But the question then arises of whether the alternative true story is in Burke’s case too a jacobite one. While it is not possible to know what all the impulses are – in himself and others – which Burke is afraid the French revolution will rejuvenate, one of them is surely a jacobite one, and very like the desire which led Don Pedro Damian in Borges story to refight the battle of Masoller. It is the desire to drag the body of Charles I from its grave and fight the civil war, and the Cromwellian subjugation of Ireland over again, to different conclusions. Dr Price, whose sermon ‘differs only in place and time, but agrees perfectly with the spirit and letter of the rapture of 1648’ (p. 158), is so dangerous partly because he may give Burke the opportunity to do that. Burke’s jacobitism appears subliminally, by courtesy of that curious relationship between the literal and the metaphoric which, I have suggested, allows his arguments to breathe: If they had set up this new experimental government as a necessary substitute for an expelled tyranny, mankind would anticipate the time of prescription, which, through long usage, mellows into legality governments that were violent in their commencement. All those who have affections which lead them to the conservation of civil order would recognize, even in its cradle, the child as legitimate, which has been produced from those principles of cogent
54
Narrative Order, 1789–1819: Part II
expediency to which all just governments owe their birth, and on which they justify their continuance. But they will be late and reluctant in giving any sort of countenance to the operations of a power, which has derived its birth from no law and no necessity; but which on the contrary has had its origin in those vices and sinister practices by which the social union is often disturbed and sometimes destroyed. This assembly has hardly a year’s prescription. We have their own word for it that they have made a revolution. (p. 176) It is difficult to decide, in this passage, whether it is the birth of the new government or of the new baby that seems the more literal. To speak about the beginnings of things as births is and was such a commonplace metaphor that it only regains its force when, as in this case, it is brought into close proximity with its literal source. And while we can say that Burke is really speaking about the infancy of governments and not of individuals, the fact that, in the hereditary system Burke is defending, the legitimacy of the one so often depends on the legitimacy of the other means that the real child and the figurative child are here on an almost equal footing. This is so much the case that – with a weirdness characteristic of Burke’s language – it is possible to believe for a moment that he is talking about a real baby ‘produced by principles of cogent expediency’ rather than by the usual sexual methods. In any case, given that he is talking about the intrigues of royal courts and high assemblies, the ‘vices and sinister practices’ in ‘the social union’ are likely to have as much to do with perverse sexual practices as perverse political ones. This strange conjunction of the literal and the metaphorical might be thought to give Burke’s argument special force. But if it does so it is only by allowing that argument to be subliminally questioned. The strangeness of this baby’s presence, at once so literal and so very much not literal, has a very specific political meaning which most of Burke’s original readers would have been in a position to recognize. A crucial political role had been played in the 1688 succession crisis by the arrival of just such a baby: the birth of a son to James II in June of that year, a birth represented, in anti-Catholic rumour, as being smuggled into the Queen’s bed. In the light of this historical episode, Burke’s metaphor could well seem, to contemporary readers, particularly pertinent and vivid. The problem is, however, that the message communicated by that particular literal baby is a jacobite one, wholly at odds with the official message – of support for the Revolution Settlement – which it is
Edmund Burke: Middles versus Beginnings and Ends
55
the task of the metaphorical baby to support. Once again, the oscillation between the literal and the metaphorical is what allows Burke’s argument to breathe. In this case, a strong jacobite sentiment is effectively ‘contained’ by the subliminal role it is allowed to play in celebrating the defeat of the Stuart cause. Burke has found, once again, a way of saying: ‘don’t remember the Civil War!’
4 Watkin Tench and the Cold Track of Narrative
H.M. Green, in his History of Australian Literature, claims that Watkin Tench ‘represents what was best in our Eighteenth Century cultural heritage’.1 Green’s ‘our’ is Australian, but since he is referring to Australia’s British, rather than its indigenous heritage, it is strange that histories of our (British) literature and culture make no mention of Tench at all. He needs to be recognized as, like Edmund Burke, a writer who travels with two passports. Burke’s mixed affiliation is now widely recognized on both sides of the Irish Sea: a similar recognition needs to be extended to other writers who moved between Britain and the various parts of its empire. Watkin Tench was a Marine officer as well as a writer. It was primarily as a marine that he sailed with the convict fleet to New Holland in 1787 and participated in the establishment of New South Wales. However, he also sailed as a writer, having made some sort of arrangement with a publisher before he left London. ‘Like many of his colleagues’, argues Tench’s modern editor, ‘he was well aware that he was making history’,2 and making history meant, among other things, writing it. His narrative accounts of the expedition and of the colony’s first years were partly written at the time, partly written up later from his journals. Indeed the question of the relationship between journals and narratives is crucial to our understanding of his work. A Narrative of the Expedition to Botany Bay was based on journals written in 1787 and 1788, during the voyage to New South Wales and the first months of colonization. The book itself was also written within that period: the finished manuscript left New South Wales on 14 July 1788 with the first boat to leave for London (where the book was published in 1789) while Tench himself stayed in the colony till December 1791. It was in June 1791 – after the departure of Tench’s manuscript but before the 56
Watkin Tench and the Cold Track of Narrative
57
departure of its author – that news of the French revolution reached the colonists in New South Wales. An Account of the Settlement at Port Jackson, Tench’s sequel to his Narrative, was written up from his New South Wales journals partly in the colony, partly when Tench was back in Britain, and partly, no doubt, on the voyage home. It was published in London in 1793. By then, following the outbreak of war with Republican France in June 1792, Tench had joined the Channel Fleet. Captured in a sea battle off the coast of Brittany on 6 November 1794, he was held for two months on prison-ships in Brest harbour, followed by four months on parole in the Breton town of Quimper. Following his return home on a prisonerexchange in May 1795 he wrote, or revised, his account of these experiences. Letters Written in France to a Friend in London was published by Joseph Johnson in 1796. Tench’s work is relevent to the question of narrative in a number of respects. In the first place, it provides a very distinctive instance of the commitment to narrative order under intense pressure. This is not the same kind of pressure which, in Johnson’s Life of Savage, gave shape and muscle to the commitment itself. Rather, it was a pressure which made the commitment to narrative evidently ragged and fragile; to such an extent that, at a moment of remarkable insight, Tench turns against the literary work in which he is engaged, leaving ‘the cold track of narrative’ for the more impassioned and rebellious journal entries from which he is attempting to construct his narrative. Secondly, the example of Tench, taken together with Burke’s Reflections, demonstrates that while attitudes towards narrative always had ethical and political implications, there was no simple polarization of attitudes to narrative order along ideological lines. Many permutations were possible. Burke identified certain crucial features of narrative order with a revolutionary continental enlightenment, leading him to reject the narrative idea of life characteristic of eighteenth-century British writing. Tench shared many of Burke’s basic commitments – notably to constitutional monarchy and the Anglican Church – but continued, even after 1789, to identify narrative order with a self-consciously moderate Enlightenment; and when, briefly, he rebelled against narrative order he did so as a social rebel, more as a Blakian than a Burkian. What Burke’s rejection of narrative order, and Tench’s momentary rejection of it have in common, is the sense of its ‘coldness’. Thirdly, Tench specifically links his military role with his role as a narrator. I have suggested that Tench stands out as the only significant writer in the age of revolution who writes as a participant in the
58
Narrative Order, 1789–1819: Part II
fighting. And this is particularly significant because of the way he himself foregrounds that relationship, alluding, at the start of his Letters written in France, to Othello and Caesar as soldier-storytellers. The contrast with Wordsworth is then particularly illuminating, since it is through allusion to the same speech from Othello – to Othello’s story of ‘moving accidents by flood and field’ – that Wordsworth attempted to distance himself from what he perceived to be the linked dangers of storytelling and military violence. Finally, the case of Tench is instructive because it allows us to assess from an unusual angle the impact of the revolution in France on British writing. And here I have in mind not the Letters written in France themselves, but the Narrative and the Account. These two books, written partly in and partly about the conditions of material scarcity and social insecurity of early New South Wales, allow us to see how circumstances that had nothing to do with the French revolution could produce a crisis of narrative order which the revolution itself, in its successive stages, then intensified. As the circumstances of these books’ composition and publication attest, Tench converted his journals into continuous narrative from various points in time and space between 1787 and 1793. Narrative takes shape from an actual or anticipated position of hindsight to an extent that a journal does not: Tench’s narrative was written up in circumstances that made the attainment of any stable position of hindsight difficult to achieve or anticipate. With the poets – with Wordsworth in particular – we know how important and difficult it is to establish the compositional strata of an author’s texts in this period. Poems were being continually revised under the pressure of changing views and changing circumstances. That the same is true of Tench is evident from, for instance, the description in the Account of the Settlement at Port Jackson (1793) of the arrival in New South Wales of the first ship bringing supplies and news (and more convicts) from Britain. Up to this point – June 1790 – ‘no communication whatever had passed with our native country since 13th May 1787, the day of our departure from Portsmouth’ (p. 162). Tench describes the ship’s arrival as follows: ‘Letters! Letters!’ was the cry. They were produced and torn open in trembling agitation. News burst upon us like meridian splendor on a blind man. We were overwhelmed with it; public, private, general and particular. Nor was it until some days had elapsed, that we were able to methodize it, or reduce it into form. We now heard for the first time of our sovereign’s illness, and his happy restoration to health.
Watkin Tench and the Cold Track of Narrative
59
The French revolution of 1789, with all the attendant circumstances of that wonderful and unexpected event, succeeded to amaze us. (p. 170) The passage raises interesting questions about the compositional history of the Account. The body of the book consists of a chronological narrative of events up to December 1791 when Tench left the colony. The voyage home is not described, though some incidents from it are referred to in footnotes written, we are to presume, after the main narrative text. This narrative text seems keen to avoid any reference to a retrospective British point of vantage on the events it describes. In this context the reference to ‘the French revolution of 1789’ is particularly intriguing. Why does Tench feel the need to mention the date? If we understand Tench’s narrative to be written in New South Wales – as news sent back, in effect, from New South Wales to London – he might mention the date so as to draw the attention of his British readers to how far in space and time news of the French revolution had to travel before it reached New South Wales. However, another reason may occur to us when we read the following passage from Letters Written in France: My political principles are . . . unchanged since we parted; and I still think a limited monarchy the best of governments. Had I been born a Frenchman, I should have struggled as hard for the revolution of 1789, as I should have resisted with all my might that of 1792. Much as I hate despotism, I am scarcely less a foe to democracy. (p. 100) But the republican revolution of 1792 – the overthrow of the monarchy, the September Massacres, the Terror – was a series of events which Tench only heard about, and some of which only took place, after his arrival back in England in June 1792. Consequently it seems likely that the phrase ‘the French revolution of 1789’ is a late revision, written in England in 1793. It is possible too that the placing of news of the King’s good health first is also a late revision, tailored to a situation in which there was increasing pressure to declare loyalist and royalist sentiment. Furthermore, this evidence of rewriting appears in a paragraph which is already about the process of composition, about the way in which the ‘news’ from England took time to ‘methodize’ and ‘reduce into form’. Tench is referring to the process of composing one’s mind: but it is
60
Narrative Order, 1789–1819: Part II
a process which goes on, under the pressure of changing political events, in the activities of writing and rewriting. And while the French revolution in its various stages precipitated some of this rewriting, the process was well under way for Tench before news of the revolution reached New South Wales. The fleet of convict transports and support ships in which Captain Tench sailed for Botany Bay in 1787 is known to historians and to most Australians as ‘the First Fleet’. However, as Alan Atkinson points out, in his recent history, The Europeans in Australia, to gentlemen in Downing Street it seemed likely that the First Fleet (the name we now give it) would also be the last to New South Wales. There was no certain vision of a Second Fleet, a Third Fleet and so on indefinitely. This point is hard to grasp from where we stand, in their distant future, especially when we think of the hundreds of convict ships which did in fact follow.3 Indeed, ‘the First Fleet’ is an essentially retrospective designation, especially with those initial capital letters which give it the status of a proper name. A thing can only ever be securely identified as the first – in the sense of the inaugurating – instance once others have followed. And yet nothing sounds more like a beginning than ‘the First Fleet’. As Sartre’s Roquentin says, ‘we think we begin at the beginning, but the end is already there’, or if not the end, at least a point of hindsight. No expression could more neatly encapsulate the link between ‘nation and narration’ than ‘the First Fleet’. And yet the beginning of New South Wales was just that, a self-conscious act of beginning. It was, like its near-contemporary, the French revolution, an ambitious and problematic attempt to found a new – and a new kind of – society from scratch, on the terrain of an ancient one. So we can imagine that, just as the colonists proclaimed the founding of the colony by – in Tench’s words – ‘the reading of the Commission, and taking Possession of the Settlement, in form’ on the beach at Port Jackson on 7 February1788, so the gentlemen in Downing Street, or the officers of the fleet might indeed have called their fleet ‘the First Fleet’. They could nevertheless have done so only in a spirit of confidant imperial anticipation: the name could only have been secured, possessed without risk of subsequent irony, with the arrival of a second fleet. Readers of Paul Carter’s fine book The Road to Botany Bay: A Study in Spatial History will recognize in these remarks the influence of his discussion of ‘the practice of naming’ in the journals and narratives of
Watkin Tench and the Cold Track of Narrative
61
European exploration and empire and in subsequent Australian historiography.4 As we have seen, Carter takes his title – The Road to Botany Bay – from Tench’s Narrative of the Expedition to Botany Bay, written in the first few months of settlement and published in London in 1789. The phrase comes from Chapter IX – ‘The Taking Possession of Port Jackson. With the Disembarkation of the Marines and Convicts. January, 1788’ – in which Tench reports that groups of convicts, attempting to escape from Port Jackson, ‘soon found the road to Botany Bay’ (p. 39). Carter’s point is a simple one: there was no road to Botany Bay. That is, at the time Tench is referring to – January 1788 – there was almost certainly no established two-way track, known to the Europeans, between Port Jackson and Botany Bay. The phrase embodies, in this respect, a central feature of what Carter calls variously ‘imperial narrative’ and ‘cause-and-effect empirical narrative’ – and which he identifies with the writing of Tench and his fellow officers and with prominent twentieth-century historians of their period. It is a kind of writing, embodied equally in the phrase ‘the road to Botany Bay’ and ‘the First Fleet’, which, in Carter’s words, ‘has as its focus facts which, in a sense, come after the event’ (p. 309). What is not clear from Carter’s account, however, is when he thinks Tench actually wrote the phrase ‘the road to Botany Bay’. Is it a phrase written hard on the heels of the event it offers to record, in effect part of a journal entry, but one that (illicitly, in Carter’s view) anticipates a future road, in that spirit of confidant imperial anticipation which Tench would also have shown if he had referred to the fleet he sailed with as ‘the First Fleet’? Or are we to understand Tench as writing from some time after the event (mid-February 1788 at the earliest), when a recognized route between Port Jackson and Botany Bay had been established: in which case we are to understand Tench as (again illicitly) reading the existence of that ‘road’ back into the convicts’ ‘finding’ of it? In which case, the Narrative of the Expedition to Botany Bay is indeed to be read as a narrative rather than a journal, and the date at the head of the chapter – January 1788 – is to be read as referring only to the time of the events narrated and not, as would be the case for a journal, also to the time of their narration. That remains unclear in Carter’s account because the question of the compositional status – the genre – of Tench’s books is not directly addressed, though it should be a crucial issue for a book which values the letters home, the explorers’ journals, the unfinished maps – written traces which, but for their spatial occasion, would not have come into being
62
Narrative Order, 1789–1819: Part II
over writing which ‘conforms to the rules of cause and effect empirical history’ (p. xxii). The headline ‘the Journals of Watkin Tench’ runs right through the modern edition of Tench’s New South Wales writing, Sydney’s First Four Years. No such title appears, however, in early editions of either the Narrative or the Account. And in fact these are, and are presented as, continuous narratives, although they are clearly written up on the basis of a regular journal. The journal no longer exists but the second book, the Account, includes verbatim extracts from it. What Tench seems to be aiming at is a form of continuous narrative which is as close as possible to the journal from which it has been written up. That is, it remains a ‘first person’ text in which Tench appears as a significant actor in the events described, and it is divided, like a journal, into chronological episodes; but it is nevertheless a continuous narrative which offers itself as a disinterested history of the events recorded. Johnson’s Tour of the Western Islands of Scotland (1775) is one of many possible models. Another, especially relevant for Tench the soldier-writer, could have been Caesar’s Gallic Wars: Caesar who, as Tench put it later, in the Letters Written in France, knew ‘not only how to execute, but to narrate deeds of glory’ (p. 2). However, where Tench calls himself ‘I’, Caesar famously called himself ‘Caesar’. And in this respect a particularly intriguing possibility is that Tench took as his model the official narrative of James Cook’s first Pacific voyage, the book which would have been the main source of information about Botany Bay for Tench and his fellow officers: Sir John Hawksworth’s Account of the Voyages Undertaken by the Order of His Present Majesty for Making Discoveries in the Southern Hemisphere (1773).5 This remarkable book consists of a sequence of first-person narratives ‘drawn up’ – in Hawksworth’s words – ‘from the journals that were kept by the Commanders of the several ships’. They were drawn up, however, not by the Commanders themselves, but by Hawksworth, writing as the ‘I’ of each Commander in turn. Where Caesar wrote about himself in the third person, Hawksworth wrote about Cook in the first person. We might now call this ghost writing, except that the ghost has his name on the title page and provides an account and justification of the authorial procedure in his Introduction. Tench, in the Narrative and the Account, has been Hawksworth to his own Cook. Of course Hawksworth was ‘drawing up’ the narrative after the event, while Tench, at least in the case of the Narrative, was not. Indeed, the confusion shared by Fitzhardinge and Carter about the genre in which Tench’s books are written is understandable. The books testify to the
Watkin Tench and the Cold Track of Narrative
63
difficulty, in certain respects the impossibility, given the circumstances of their writing, of doing what Tench set out to do: convert a journal into a narrative. The Narrative and the Account have certain distinct formal features which, as we shall see, follow from the rather different circumstances of their composition. Nevertheless, they also have important features in common. Each consists of a sequence of narrative chapters followed by a few chapters with titles such as (in the Narrative) ‘The Face of the Country; its productions, Climate &c’, or (in the Account) ‘Miscellaneous Remarks on the Country. – On its Vegetable Productions. On its Climate. – On its Animal productions. – On its Natives, &c.’ But this apparently neat organizational principle is belied by its management in detail. The difficulty of producing the narratives from the journals is frequently evident. I shall look at a number of places where it is interestingly evident, concluding with an analysis of a remarkable passage in the second book where Tench refers to ‘the cold track of narrative’, and to the conversion of his journal into such a narrative as a work of ‘servile adaptation’. The last event described in the narrative chapters of the first book is the execution of the convict Samuel Paton. ‘After this’, Tench writes, nothing occurred with which I think it necessary to trouble the reader. The contents of the following chapter could not, I conceive, be so properly interwoven in the body of the work; I have, therefore, assigned them to a place by themselves, with a view that the conclusions adopted in them may be more strongly enforced on the minds of those, to whom they are more particularly addressed. (p. 63) It is the first sentence I want to comment on: ‘after this nothing occurred with which I think it necessary to trouble the reader’. What that tells us is that the narrative is finished: not only because there are no more events to narrate, but because the sentence that tells us so is written in the past simple tense. This past tense tells us that Tench is aiming to write narrative in a rather strict sense, the sense we may associate with a history book.6 It is of course a history book written by a man who participated, and presents himself as participating, in the events narrated; and yet he wishes to present his own participation so far as possible in the way that a historian or biographer might do. He is writing, to be sure, a first-person narrative, but while ‘I’ the marine is in the frame, ‘I’ the writer is largely excluded from it, becoming, like the reader, an observer of the events narrated. Tench the writer is cut off from the events so that the latter may form
64
Narrative Order, 1789–1819: Part II
themselves into a homogeneous sequence with a beginning and an ending, removed to that distance at which they come into perspective both for us and for Tench himself. It is particularly true of a non-fictional narrative defined in this way – defined principally by tense – that it cannot be written until all the events to be narrated have finished. Consequently, an event must be identified as an ending, even if the ending is supplied only by the calendar (‘Britain in the Nineteenth Century’, for instance). And after that ending there must be a space, a space of time. This space is the epistemological space that sets the narrative of events at their proper distance, but we need to believe that this space of time is empirical, a matter of a certain number of weeks or months or years that would allow, among other things, for the narrative to get written. Since what Tench is telling us in that concluding narrative sentence is that, following the execution of Samuel Paton, nothing happened, the real ending becomes, by default, Tench’s own departure from New South Wales, the step from land to boat or, a few months later, from boat again to land, providing the time for writing the narrative, on board ship or back in England. The problem is that while the manuscript of A Narrative of the Expedition to Botany Bay returned to London on board ship, its author, Tench, stayed in New South Wales. Tench cannot write a complete narrative of the events in which he has been involved in New South Wales for the same reason that nobody can write a complete autobiography, the story of their own life from cradle to grave: there is no available vantage point.7 There was no ending of the sort that the past simple tense asserts and requires. Indeed, we may well be suspicious of Tench’s claim that ‘after [Samuel Paton’s execution] nothing occurred’, since in one of the subsequent overview chapters, Tench does in fact report ‘an event that happened a few evenings since’ (p. 72). Tench has, I believe, invented eventlessness as a firebreak, an equivalent of the water he would have crossed to get to a boat, or to get back to England. He has fabricated an ending which can separate the whole sequence of events recorded from the event of recording them which itself disappears from view. This is always what historical narrative does, but in this case, because we know that the author was separated from his manuscript, we can see how the disappearing trick works: the distance which allows events to form themselves into a compact narrative, putting events into perspective, has, in this case, lost its empirical alibi. The difficulty of converting journal into narrative is naturally most intense at the end of the first book, as the time of the events narrated
Watkin Tench and the Cold Track of Narrative
65
converges with the time of their narration. But it is there throughout both books as the circumstances of their composition should lead us to expect. Tench had signed on for a three-year tour of duty to New South Wales and had made some kind of arrangement with a London publisher before he left England. As it turned out he was able to leave after about four years and take manuscripts – some combination of journal and partly completed narrative Account – back with him to complete on the voyage and in Britain. Furthermore, by late 1791 a number of ships had already left the colony for England, a second fleet had arrived and the threat of starvation had for the moment receded. Tench may well have felt able to anticipate, in organizing the writing of his second book in New South Wales, that events would fall out in something like the way they did. Nonetheless, the circumstances of writing continued to be critically unpredictable. There was every possibility of not getting back to England, of dying from hunger, disease, violence or shipwreck, of his writing having to be sent back in advance of him or posthumously. The second book is certainly constructed in a more relaxed fashion: there is more explicit reference in it to a position of hindsight from which earlier judgements can be measured, of the whole thing being recast to conform to an overall design.8 Nevertheless, it is true of both books, if particularly of the first, that Tench must have written with a special sort of deadline hanging over him, knowing that the need to produce a completed narrative might come at any moment, as a thief in the night. Each chapter of the first book is headed with the date of the month and year – ‘January, 1789’ for instance – of the events it narrates. This makes the chapters look very like consolidated monthly journal entries, which is perhaps what Paul Carter takes them to be. But unlike a journal entry, events in a Tench chapter do not move closer to the present of the writing as the month progresses. It is all in the past simple tense – never the present or present perfect. ‘I’ the marine may be involved in the events narrated, but ‘I’ the writer is out of sight, joined with the reader in following all the events of the month from an equal distance. If we try to reconstruct Tench’s actual compositional practice, we may need to think of the chapters as being written up soon after the next month has started. After all, if the narrative as a whole requires an ending, so does each chapter of it, endings supplied by the calendrical break between one month and the next. Or rather, we can think of Tench partly doing that and partly trying to write his chapters up
66
Narrative Order, 1789–1819: Part II
within the month they are about but as if from the other side of the calendrical break, just as he wrote the end of the narrative as a whole as if from after a kind of ending – his departure from New South Wales – which did not in fact take place. It is as if each chapter is written for possible separate dispatch, or for adding to an accumulating pile of chapters for dispatch, episodes of a non-fictional serial novel with the rudimentary predictability of the calendar providing the only outline of the future plot. At the end of Chapter IX of the Narrative Tench refers to ‘what I was afterwards told by Monsieur De Perrouse’ (p. 37).9 This is one of very few places in the book at which a position of narrative hindsight is indicated. And yet there are no instances in either book of what we might expect to go with that: evidence of those alterations of perspective which are inevitable in a journal and which are the lifeblood of narrative fiction. Peripeteia, the kind of reversal which makes us realize that we have misunderstood the events that have already been narrated, is out of place in the kind of narrative to which Tench aspires, as it was out of place in Johnson’s Life of Savage. History books and biographies are not supposed to reveal, half way through, that we have been listening to an unreliable narrator. As for the Account, while there are still no significant changes of perspective (though, as we shall see, there is a very important reference to one), there are a lot more references to the view from the end of the story than there were in the Narrative. There is a lot more indication, that is to say, of the distance between the narrative we are reading and the journal on which it is based. Indeed, that distance allows Tench to quote, in this second book, from his daily journal and consequently allows us a clearer sense of what was involved in converting journal into narrative. A journal entry quoted in Chapter XVI reads: December 3rd, 1791. Began my survey of the cultivated land belonging to the public. The harvest has commenced. (p. 247) We may compare this with a narrative passage such as the following, taken from Chapter VI (‘Transactions of the Colony from the Beginning of the Year 1790, until the End of May following’): On the 5th of April news was brought, that the flag on the Southhead was hoisted. Less emotion was created by the news than might
Watkin Tench and the Cold Track of Narrative
67
be expected; everybody coldly said to his neighbour, ‘the Sirius and Supply are returned from Norfolk island.’ To satisfy myself that the flag was really flying, I went to the observatory, and looked for it through the large astronomical telescope, when I plainly saw it. (pp. 163–4) We can assume that the second passage – the narrative passage – is written on the basis of a journal entry because it begins, like the actual journal entry, with a date. Indeed, one of the things which makes the conversion from journal to historical narrative relatively easy is, as I have suggested, the strong commitment of both genres to chronology, to time ordered – as in this militarized European proto-society time was determinedly ordered in every possible respect – by clock, bell and calendar. Nevertheless, despite the continual presence of chronology as a hinge between journal and narrative, our examples of the two forms of writing show that they differ in at least two respects. First, the narrative is written in complete sentences whereas the first sentence of the journal entry misses out the subject of the verb (‘Began my survey’). Secondly, the narrative passage is written in the past simple tense (‘I went to the observatory’), whereas the journal entry is written in a mixture of the past simple (‘began my survey’) and the present perfect (‘the harvest has commenced’). The historical significance of the distinction between the past simple and the present perfect tenses has already been noted, in our discussion of Burke’s Reflections. Burke’s recoil from beginnings and endings leads him precisely to foreground the present perfect. Tench, by contrast, wishes to marginalize the present perfect, so that his present and his presence as narrator may disappear in deference to a free-standing and dispassionate narrative. The opening of the past into the present in which it is remembered and narrated, which Burke and then Wordsworth embrace, could be said, with hindsight, to be what Tench, in aspiring to convert journal into narrative, is attempting, with some difficulty, to avoid. Other grammatical choices were involved too however. When Tench refers, in a supplementary chapter of the Narrative, separated from the narrative itself, to ‘an event which happened a few evenings since’, he is excluding from narrative a deictic term (‘since’ in the sense of ‘ago’) which Walter Scott put at the centre of his new type of historicist historical novel with the publication of Waverley; or, ’tis sixty years since. It is a title which from the start links the past events to be narrated to the present moment in which they are narrated. It also suggests
68
Narrative Order, 1789–1819: Part II
a future that is not yet known, has not yet occurred at the time of narrating. The work of converting journal into narrative would of course involve other operations beside these grammatical and lexical ones: extensive selection, addition, condensation and expansion. But the grammatical and lexical changes would be fundamental, the necessary minimum required for the conversion of one genre into the other. They are also quite easy if you know the rules, as Tench had learnt the rules, probably at a grammar school. We can imagine him performing these operations quite quickly, more or less as he goes along, perhaps, a tedious but simple mechanical task. Indeed, in some cases the work of rewriting journal entries as passages of narrative may have involved literally rewriting – transcribing – and nothing more. For instance November, 1791. A very extraordinary instance of folly stimulated to desperation, occurred in the beginning of this month, among the convicts at Rose Hill. (p. 243) This is an episode in a narrative, but could well have originally been a journal entry. No words would have had to be changed, only the meaning of one of the words, the word ‘this’. As part of a narrative, ‘this’ means ‘that’ (‘this month I am talking about’). If however these lines were a journal entry, ‘this’ would mean both ‘this month I am talking about’ and ‘this month I am talking about it in’. The word would refer to the events and to the act and time of recording them. In becoming part of a narrative, it is not simply that the temporal perspective on the event changes, from present to past; it is rather that only the reference to the events survives. The reference to their recording simply and silently disappears. Even if mechanical changes of grammar, or a change of context, were often all that was required to convert journal into narrative, the difference between journal and narrative is profound. The journal entry for 3 December 1791 happens to record incomplete processes (the survey and the harvest have just begun). Narrative can record incomplete processes too, but not, as the journal entry does, in an incomplete way, in a shorthand prose and a verb form (‘has started’) which opens the past of the events recorded into the present of their recording. We may remember Paul Carter’s linking of ‘the explorers’ journals, the unfinished maps’.
Watkin Tench and the Cold Track of Narrative
69
The narrative passages by contrast make no reference to writing: either to the original writing of the journal entry – perhaps later on the same day as the events concerning the flag took place – or to the writing up of that journal entry into the narrative passage that was actually published. The journal entry links Tench the marine and Tench the writer. The conversion of journal to narrative makes Tench the writer disappear, so that the ‘I’ becomes as single as the misleading expression ‘first-person singular’ implies it always is. And since the narrative makes no mention of all the time other officers were also spending writing journals, the narrative can be said to conceal the work of writing in the same way that soap-operas conceal the time their characters would in fact spend watching and talking about soap-operas. The journal entry for 3 December 1791 – ‘Began my survey of the cultivated land belonging to the public. The harvest has commenced’ – encourages us to link a linguistic and epistemological understanding of narrative to a social one. It tells us that the writing of the journal is an activity among others, competing for time with other forms of labour (the survey, the harvest). That is one thing the shorthand form means: lack of time and energy for fully elaborated recording at the end of a day full of the events that are being recorded. Though the other thing shorthand means – and which marks one of the similarities between a journal entry and a visual artist’s sketch – is that this is writing that may well be designed, at the time, to be written up later – at leisure – into narrative. Tench’s second book was published in 1793 and Tench begins it by reminding his readers of how things had stood in the colony ‘on the Date of my former Narrative, in July 1788’: Previous to commencing any further account of the subject which I am about to treat, such a retrospection of the circumstances and situation of the settlement, at the conclusion of my former Narrative, as shall lay its state before the reader, seems necessary, in order to connect the present with the past. (p. 133) And at the end of these two chapters of ‘retrospection’ he writes: New matter now presents itself. A considerable part of the foregoing chapters had been related before, either by others or myself. I was however, unavoidably compelled to insert it, in order to preserve
70
Narrative Order, 1789–1819: Part II
unbroken that chain of detail, and perspicuity of arrangement, at which books professing to convey information should specially aim. (p. 138) Tench is spelling out a conventional conception of serious non-fictional narrative, as an unbroken chain of events, each event leading to – in the sense of causing – the event that followed it in time: post hoc ergo propter hoc. The narrative is therefore both factual and explanatory, exactly and explicitly what Paul Carter condemns as ‘cause and effect empirical narrative’. And it is disinterested and objective in the sense that it is essentially no different – or so Tench assumes – from those ‘relations’ of the same events that have been offered by others. Indeed, what we are dealing with here is not just certain views about narrative. It is a view of explanation – so far at least as human affairs are concerned – which is essentially a narrative one. One thing narrative so conceived must avoid is gaps, which would be both gaps of time and explanatory gaps. And the insertion of the retrospect, to ‘preserve unbroken that chain of detail’, suggests that he sees his two books – as I suggested he may have conceived their individual chapters – as instalments of a single extendable narrative. However, people who read the Account in 1793 would either never have read the first instalment (the Narrative) or, if they had, would very likely have read it three or four years previously, when it first appeared. Hence the need to supply readers with the-story-so-far. Modern readers, by contrast, will almost certainly read the two books together, one straight after the other, in the Fitzhardinge or Flannery editions. We are therefore far more likely than Tench’s original readers to register the quite startling change of perspective between the story offered in the first book and the-story-so-far offered by the first chapter of the second book. In one very important respect Tench acknowledges that, between writing the two books, his views have changed. Looking back to how things stood in 1788, he now writes: With the natives we were very little more acquainted than on our arrival in the country. Our intercourse with them was neither frequent or cordial. They seemed studiously to avoid us, either from fear, jealousy or hatred. When they met with unarmed stragglers, they sometimes killed, and sometimes wounded them. I confess that, in common with many others, I was inclined to attribute this conduct, to a spirit of malignant levity. But a farther acquaintance
Watkin Tench and the Cold Track of Narrative
71
with them, founded on several instances of their humanity and generosity, which shall be noticed in their proper places, has entirely reversed my opinion; and led me to conclude, that the unprovoked outrages committed upon them, by unprincipled individuals among us, caused the evils we had experienced. (p. 135) He has ‘reversed’ his opinion, and reversal is very much the word for what he describes. He has had to reverse his cause-and-effect narrative explanation of the antagonism between the colonists and the natives. If you read this account of a ‘reversal of opinion’ straight after reading the narrative it offers to correct, the effect of it, or so it seems to me, is of more than a specific correction, vitally important as that specific correction is. The experience is more like the kind of reversal that historical narrative is supposed to exclude, Aristotelian reversal, peripeteia, the realization that the story as you have read it so far is quite different from the story as you now have to rewrite it in your mind. Tench is saying, in fact, that in this respect he was an unreliable narrator. We may, if we wish, see this effect from a postmodern viewpoint, as a vindication of current scepticism about historical narrative as such. But if we do so, we should nevertheless realize that the effect is neither produced by nor produces scepticism of that type in Tench at this point. He has been led to his ‘reversal of opinion’ precisely by the rigour and integrity of his commitment to the discovery of true causes and real beginnings. I say that ‘at this point’ Tench does not write out of scepticism about his narrative enterprise as such, because in two passages from the Account a strongly sceptical impulse does appear. Tench acknowledges that the-story-so-far is not, with respect to relations between colonists and natives, the story we have in fact been reading. But in other respects too, ones that Tench does not acknowledge, these stories do not coincide. These ‘retrospective’ chapters offer a perspective on social relations among the Europeans themselves which has no real precedent in the account offered, nearer the time, by Tench’s first book. When ‘the first fleet of ships left for Europe, on the 14th of July 1788’, Tench now writes, a few of the convicts had got into huts, but almost all the officers, and the whole of the soldiery, were still in tents. In such a situation, where knowledge of the mechanic arts afforded the surest recommendation to notice, it may be easily conceived, that attention to the parade duty of the troops gradually
72
Narrative Order, 1789–1819: Part II
diminished. Now were to be seen officers and soldiers not ‘trailing the puissant pike’, but felling the ponderous gum-tree, or breaking the stubborn clod. And though ‘the broad falchion did not in a ploughshare end,’ the possession of a spade, a wheelbarrow, or a dunghill, was more coveted than the most refulgent arms in which heroism ever dazzled. Those hours, which in other countries are devoted to martial acquirements, were here consumed in the labours of the sawpit, the forge, and the quarry . . . On the convicts the burden fell yet heavier: necessity compelled us to allot to them the most slavish and laborious employments. Those operations, which in other countries are performed by the brute creation, were here effected by the creations of men. (pp. 133–4) There is a footnote to the second of these paragraphs, a quotation from Voltaire’s History of Charles XII (1732): The Swedish prisoners, taken at the battle of Pultowa, were transported by the Czar Peter to the most remote parts of Siberia, with a view to civilise the natives of the country, and teach them the arts the Swedes possessed. In this hopeless situation, all traces of discipline and subordination, between the different ranks, were quickly obliterated. The soldiers, who were husbandmen and artificers, found out their superiority, and assumed it: the officers became their servants.10 The paragraphs from the main body of the text describe the general undermining of an established European division of labour in circumstances where the familiar European infrastructure does not yet exist. Soldiers – including Tench himself – have become labourers, while convicts – who were mostly labourers in Britain and would be expected to labour in the colony – have become ‘beasts of burden’. Furthermore, political and cultural revolutions are latent in the social one, prestige following the possession of the spades and wheelbarrows of ‘mechanic arts’ rather than the tools and skills of the officer and gentleman, and – implicitly – following those who call a spade a spade rather than those who might wish to talk about ‘puissant pikes’ and ‘broad falchions’ and use words like ‘refulgent’. When were these passages written? The narrative chapters of the Account do not encourage us to envisage a distant retrospect from Britain in 1792–3 but may nevertheless involve such a retrospect, as we saw in the case of Tench’s reference to ‘the French revolution of 1789’. On the
Watkin Tench and the Cold Track of Narrative
73
other hand these early story-so-far chapters must have been written up towards the end of his sojourn in New South Wales and therefore after news of that revolution of 1789 had reached the colony. By contrast, the Voltaire footnote, because it is a footnote, is likely to be a late addition, written after the revolution Tench opposed, ‘the revolution of 1792’. The Voltaire footnote makes the political implications of the social and cultural developments described in the main text more explicit. It would be interesting to know how far – either at the time referred to ( July 1788) or at the time of writing (1791 or 1792) – the fact that ‘a few of the convicts had got into huts, but almost all the officers, and the whole of the soldiery, were still in tents’ carried the sense of threat – of the social and political world about to turn upside down – that the passage from Voltaire retrospectively gives it. In any case, we are dealing here with an original situation twice reframed in the light of Tench’s changing present. The footnote reframes the main text, which is itself part of a retrospective retelling of events in 1788 originally described (and described differently) in Tench’s first book. That first book, for its part, was itself written up from journals. The narrative is supposedly constituted as a narrative from a consistent position of hindsight; but the point of hindsight in fact keeps changing, changing the narrative as it does so. As for the social content of the passage, the terms in which Tench thinks about ‘subordination’ and the division of labour here hinge on a fundamental distinction between ‘mechanic arts’ and ‘liberal’ ones. I shall need to describe that distinction in some detail so that Tench’s application of it, here and in a later passage from the same book, may be properly understood. In so doing, I shall be describing the particular British enlightenment framework within which Tench views events in New South Wales, but also noting the effects which events in New South Wales have on this framework. The distinction could be illustrated from numerous eighteenthcentury sources, but the most likely source for Tench’s understanding of it is Adam Ferguson’s Essay on the History of Civil Society (1767), the book from which the Account of the Settlement at Port Jackson takes – in the form of a quotation – its own final paragraph. In a chapter entitled ‘Of the Subordination Consequent to the Separation of Arts and Professions’, Ferguson writes: There is one ground of subordination in the difference of natural talents and dispositions; a second in the unequal division of property; and a third, no less sensible, in the habits which are acquired by the practice of different arts.11
74
Narrative Order, 1789–1819: Part II
It is this last – the ground of subordination in the practice of different arts – which concerns Ferguson at this point. By ‘arts’, he means employments and skills, as Tench does when he refers to ‘mechanic arts’; nevertheless the arts in a narrower sense – a sense that would include the writing of a philosophic history or a narrative of colonial settlement – are among the arts to which Ferguson refers and are among the employments he has in mind as his argument develops: Some employments are liberal, others mechanic. They require different talents, and inspire different sentiments; and whether or not this be the cause of the preference we actually give, it is certainly reasonable to form our opinion of the rank that is due to men of certain professions and stations, from the influence of their manner of life in cultivating the powers of the mind, or in preserving the sentiments of the heart. There is an elevation natural to man, by which he would be thought, in his rudest state, however urged by necessity, to rise above the consideration of mere subsistence, and the regards of interest. He would appear to act only from the heart, in its engagements of friendship, or opposition; he would show himself only upon occasions of danger or difficulty, and leave ordinary cares to the weak or the servile. (p. 184) One feature of mechanical employments is that, as the word ‘mechanic’ implies, they fit a person for the repetition of a specific and limited task and, by the same token, preclude them from an overview of human life. Such employments are therefore ‘servile’ in a double sense: they involve subjection to rulers, but also to the rules and regulations of a particular skill or specialist calling. Correspondingly, an affinity is presumed to exist between those not bound to a particular task and those able to take an overall and disinterested view of the kind that is required in those who rule. It is an overall view also required in those who may aspire to participate in public life by influencing those who rule, as Tench himself does in wishing that his ‘conclusions’ should be ‘strongly enforced on the minds of those, to whom they are more particularly addressed’, which is to say, those in a position to act on the basis of those conclusions, Alan Atkinson’s ‘gentlemen in Downing Street’. A crucial issue for Ferguson is how, in a commercial society, that link can be embodied in men of the liberal professions such as himself and Tench: men who are masters of particular skills and earn their livelihood by means of those
Watkin Tench and the Cold Track of Narrative
75
skills, but who aim to align themselves with gentlemen of leisure in the service of the public. With respect to Tench we might put the issue as follows. Has a man who is obliged to spend the day ‘surveying the cultivated land belonging to the public’ been able to cultivate his mind in a way that would allow him to make a survey of the attempt to found a colony, a survey that can aspire to influence public policy. As for Ferguson, he deals with the question in the following way. ‘In polished society’, he writes, man’s desire to avoid the character of sordid, makes him conceal his regard for what relates merely to his preservation or his livelihood. In his estimation, the labourer, who toils that he may eat; the mechanic, whose art requires no exertion of genius, are degraded by the object they pursue, and by the means they employ to obtain it. Professions requiring more knowledge and study; proceeding on the exercise of fancy, and the love of perfection; leading to applause as well as to profit, place the artist in a superior class, and bring him nearer to that station in which men are supposed to be highest; because in it they are bound to no task; because they are left to follow the disposition of the mind, and to take that part in society, to which they are led by the sentiments of the heart, or by the calls of the public. (pp. 184–5) It is through a process of concealment – concealment of the mechanical element of their own employment – that the liberal professions distance themselves from those in purely mechanical employments and bring themselves into close proximity with ‘that station in which men are supposed to be highest; because in it they are bound to no task’. It is in Joshua Reynolds’s Discourses on Art that this issue is clarified in terms of the actual practice of a liberal art, and therefore in such a way as to bring it close to the questions raised by Tench’s own practice as a writer of narrative. ‘The first endeavour of a young painter’, says Reynolds in his Third Discourse, ‘must be employed in the attainment of mechanical dexterity’; but the students who, having passed through the initiatory exercises . . . and who, sure of their hand, have leisure to exert their understanding, must now be told, that a mere copier of nature can never produce anything great.12
76
Narrative Order, 1789–1819: Part II
And in the Second Discourse, he wrote that the artist who has his mind thus filled with ideas, and his hand made expert by practice, works with ease and readiness . . . The well-grounded painter . . . has only maturely to consider his subject, and all the mechanical parts of his art follow without his exertion. (p. 37) ‘With ease’, ‘without exertion’: these are the hallmarks of a gentleman. To become a fine artist is to become like a fine gentleman, through the expert practice of mechanical arts whose function is precisely to conceal themselves by becoming second nature. You will then be able to call on those mechanical skills and they will follow you, as a landed gentleman might call on those in mechanical employment and they will follow him. And this is more than an analogy because to become an artist of this successful sort is to become a gentleman; and to fail to do so is to become a tradesman: The value and rank of every art is in proportion to the mental labour employed in it, or the mental pleasure produced by it. As this principle is observed or neglected, our profession becomes either a liberal art or a mechanical trade. In the hands of one man it makes the highest pretensions, as it is addressed to the noblest faculties: in those of another it is reduced to a mere matter of ornament; and the painter has but the humble province of furnishing our apartments with ornament. (p. 57) What happens then, it may be asked, when a practitioner of narrative art like Tench finds himself in a situation where he has been forced to become what Ferguson refers to as a ‘labourer, who toils that he may eat’? This is how Tench responds in a passage describing – at first from a position of narrative hindsight – the colony’s second period of drought and reduced rations, in April 1791: The hardships which we . . . suffered, were great; but not comparable to what had been formerly experienced. Besides, now we made sure of ships arriving soon to dispel our distress: whereas, heretofore, from having never heard from England, the hearts of men sunk; and many had begun to doubt, whether it had not been resolved to try how long misery might be endured with resignation.
Watkin Tench and the Cold Track of Narrative
77
Notwithstanding the incompetence of so diminished a pittance, the daily task of the soldier and convict continued unaltered. I never contemplated the labours of these men, without finding abundant cause of reflection on the miseries which our nature can overcome. – Let me for a moment quit the cold track of narrative: – let me not fritter away by servile adaptation, those reflections, and the feelings they gave birth to: – let me transcribe them fresh as they arose, ardent and generous, though hopeless and romantic. – I every day see wretches pale with disease and wasted with famine, struggle against the horrors of their situation. How striking is the effect of subordination; how dreadful is the fear of punishment! – The allotted task is still performed, even on the present reduced subsistence: – the blacksmith sweats at the sultry forge; the sawyer labours pent-up in his pit; and the husbandman turns up the sterile glebe. – Shall I again hear arguments multiplied to violate truth, and insult humanity! – Shall I again be told that the sufferings of the wretched Africans are indispensable for the culture of our sugar colonies: that white men are incapable of sustaining the heat of the climate! – I have been in the West Indies: – I have lived there. – I know that it is a rare instance for the mercury in the thermometer to mount there above 90°; and here I scarcely pass a week in summer without seeing it rise to 100°; sometimes to 105°; nay, beyond even that burning altitude. But toil cannot be long supported without adequate refreshment. The first step in every community, which wishes to preserve honesty, should be to set the people above want. The throes of hunger will ever prove too powerful for integrity to withstand. – Hence arose a repetition of petty delinquencies, which no vigilance could detect, and no justice reach. (pp. 220–1) The very layout of this passage is unique in Tench’s two books. Other excerpts from the original journals are separated on the page and marked by dates. Here it is hard to know when the excerpt from the journal begins and ends. The approximate boundaries are indeed marked by the changes of tense: from the past of narrative to the present and present perfect of the journal entry (‘I have lived in the West Indies’) and back to the past simple again. But the present tense and the hectic journal-like dashes between sentences begin before the journal entry, as Tench starts questioning his whole narrativizing enterprise: ‘Let me quit the cold track of narrative.’ And it is not clear
78
Narrative Order, 1789–1819: Part II
whether we return to the narrative – retrospective and framing – with the whole final paragraph or only with the past tense of the final sentence. The journal has flooded its narrative banks. ‘The throes of hunger will ever prove too powerful for integrity to withstand’: was that written at the time in his journal (April 1791), from a position of narrative hindsight later in 1791, or from the very different position of narrative hindsight of a Europe in the turmoil of revolution and war in 1792 or 1793? We can also see, in this passage, ways in which the language of Ferguson and Reynolds is altered by the circumstances it is here being required to articulate. When Ferguson talked about ‘subordination’ he was talking about an inherited and variously inculcated hierarchy of ranks and skills. In the Tench passage it refers to something more specifically military and carceral, thinner, more brittle; more to do, we might now say, with power than with hegemony. What we saw in the previous excerpt, from Chapter One of the Account, was a threat to subordination arising from the failure of what Ferguson called ‘the separation of arts and professions’. What we see in this later passage is the forcible reassertion of that separation by military means. Indeed, what Tench is presenting here is precisely the insecurity, in these extreme colonial circumstances, of that ‘separation of arts and professions’, where those accustomed to practice liberal arts may find themselves having to perform mechanical ones and can only maintain subordination through the giving of orders backed up by the open threat of military violence. This is what leads Tench to foreground – in sadness or disgust – the mechanical and ‘servile’ character of his own work in producing a text in which all traces of that work will disappear. Working in the intense heat, whether as a labourer himself, an officer ordered to order others to labour, or a writer converting a journal into a narrative to fulfil a contract with a London publisher, the latter task – the production of the narrative – is revealed as a mechanical one, a work of ‘servile’ adaptation: that is to say, the repetitive application of a specific set of rules (grammatical rules) and a species of labour which links him, he believes, with the labouring convicts and with the slaves in the British sugar-colonies to whom his mind quickly turns. The disinterested distance which narrative offers to supply, the perspective it offers on events (a distance and perspective which the description of the journal extract as ‘hopeless and romantic’ attempts to salvage), is now identified as a human distance created between himself and practitioners of more openly mechanical tasks.
Watkin Tench and the Cold Track of Narrative
79
But what exactly does Tench mean by calling narrative a ‘cold track’ and contrasting it with his journal reflections, ‘ardent and generous, but hopeless and romantic’? Reynolds’s Seventh Discourse provides an illuminating context: What distinguishes oratory from a cold narration, is a more liberal, though chaste, use of those ornaments which go under the name of figurative and metaphorical expressions; and poetry distinguishes itself from oratory by words and expressions still more ardent and glowing. This is what you might call a thermometric theory of genre, from cold (narrative) to hot (poetry). The context of genre theory would suggest that when Tench calls his original journal reflections ‘hopeless and romantic’ there is a specific link to ‘the romance’, by contrast with historical narrative of the sort he is attempting. ‘Romantic’ here also means unrealistic – romance opposed to reality – and is close to ‘hopeless’ in that sense. On the other hand, while ‘hopeless’ has the sense of unrealistic – unrealistic in its passionate empathy across the distances of class and the distances of empire – it also refers to the despair and ironic bitterness of the journal entry, the sense of living in a world that does not make sense. Just as ‘subordination’ takes on a new meaning in the context of forced labour, so does the theory of artistic genres as the metaphors of heat and cold are moved into the context of actual body heat and actual thermometers. For Reynolds, the warmer genres, poetry and oratory, are characterized by a greater use of ‘figurative and metaphorical expressions’ than is ‘cold narrative’. And indeed, to move from Tench’s narrative to the journal reflections is to move from a text in which things are linked by chronology and cause and effect to a text where they are linked by elements of similarity and difference. In the heat of the moment, different things are welded together: his own work of ‘servile adaptation’ is associated with the servile work of convicts and labourers and with slavery, leading his thoughts from New South Wales, to that very English sounding ‘glebe’, to the West Indies. Tench’s thoughts are moving from one form of oppressive labour to another and one continent to another, in long interrogative sentences populated by genre figures of labour. We may be reminded that, sometime between Tench’s journal entry and his narrative, William Blake was producing his Visions of the Daughters of Albion. It is as if Oothoon’s cries echoed back to her across the Atlantic by Albion’s daughters are being echoed back to her from across the Pacific too.
80
Narrative Order, 1789–1819: Part II
Finally, why does Reynolds’s ‘cold narrative’ become Tench’s ‘cold track of narrative’? Narrative is a track because it is a place where one thing leads to another; or rather, he here suggests, it is a track which (like so many Australian tracks for Europeans) does not seem to lead anywhere. It is also the hunter’s track – hunters of animals, escaped convicts and aborigines – which is cold because the scent has gone cold and because the hunting is a cold-hearted thing. In any case, that metaphorical cold track leads off into all corners of Tench’s work. It even leads us back to ‘the road to Botany Bay’ which is exactly, for Paul Carter, a section of ‘the cold track of narrative’.
5 William Godwin: Stories and Families
It is now widely accepted that both Godwin’s treatise, An Enquiry Concerning Political Justice (1793), and his major work of narrative fiction, Things as They Are; or, Caleb Williams (1794), are ‘designed to achieve change and also designed to refute the case for the status quo familiarized, above all, by Burke’.1 Nevertheless, the two books must be designed to fulfil this objective in different ways, if only because the design – the form – of a novel and a treatise are different. That the political argument had something to do with narrative would not, however, have been equally recognized by all parties. Burke identified politics closely with aesthetics, but my attempt to place his work in the context of thinking about narrative would probably have puzzled him. William Godwin, by contrast, would not have been surprised to find his work discussed in these terms. As Gary Kelly, Pamela Clemit, Jon Klancher and others have demonstrated, Godwin was interestingly alert to the relationship between politics and narrative form.2 My aim in this chapter is to take their argument further. The relationship between politics and narrative is, I shall suggest, a principle preoccupation of both Caleb Williams and Political Justice. I shall also argue, however, that Godwin’s treatment of this relationship is as interesting for its uncertainties as for its intelligence, uncertainties which derive in part from the semantic instability of the vocabulary available for its discussion. To focus a discussion of narrative and politics, as the title of this chapter suggests we should, on ‘stories and families’ is immediately to beg one of the questions it is my purpose to answer. That is, is a family a political institution? The answer to that question will clearly depend on what is meant by the word ‘family’; and it will also depend, I shall suggest, on what is meant by the word ‘story’. In exploring the relationship 81
82
Narrative Order, 1789–1819: Part II
between stories and families in Caleb Williams and Political Justice we need to pay serious attention to Godwin’s often puzzling use of two groups of words: on the one hand, words used to describe narrative or features of narrative, including ‘story’, ‘history’, ‘character’ and ‘narrative’ itself; on the other hand, words for significant social relationships, including ‘family’, ‘domestic’, ‘servant’ and ‘master’. The importance of both the groups has already been noted, for instance in the use of the word ‘character’ by Johnson and Boswell, and in Burke’s use of ‘family’ to mean lineage. In Godwin it is significantly harder for us to be sure what these important words mean. These words – the narrative words and the social words – have always been complex. They were especially complex in the period of Godwin’s writing life because they were all undergoing semantic transformation. They are of course distinct groups of words and a change of meaning within one group does not necessarily or immediately entail a shift of meaning in the other. The two lexical groups are nevertheless connected, if only because all of these words helped to alter the way in which the distinction between public life and private life was conceived. Private life came to be associated with, on the one hand, the family as James Mill defined it in 1829, ‘the group which consists of a Father, Mother, and Children’,3 and, on the other hand, with a conception of personal identity that was inward in the specific sense of being detached – in a sense I shall explain – from any open connection to narrative representation. It is interesting to watch some of these words at work in specific passages of Godwin’s writing, first of all by putting a passage from the Preface to Caleb Williams beside the opening sentences of the 1796 edition of Political Justice. ‘The following narrative’, Godwin writes in the Preface, is intended to answer a purpose more general and important than immediately appears upon the face of it. The question now afloat in the world respecting Things As They Are is the most interesting that can be presented to the human mind. While one party pleads for reformation and change, the other extols in the warmest terms the existing constitution of society. It seemed as if something would be gained for the decision of this question, if that constitution were faithfully developed in its practical effects. What is now presented to the public is no refined and abstract speculation; it is a study and delineation of things passing in the moral world. It is but of late that the inestimable importance of political principles has been
William Godwin: Stories and Families
83
adequately apprehended. It is now known to philosophers, that the spirit and character of the government intrudes itself into every rank of society. But this is a truth highly worthy to be communicated, to persons, whom books of philosophy and science are never likely to reach. Accordingly it was proposed, in the invention of the following work, to comprehend, as far as the progressive nature of a single story would allow, a general view of the modes of domestic and unrecorded despotism, by which man becomes the destroyer of man. If the author shall have taught a valuable lesson, without subtracting from the interest and passion by which a performance of this sort ought to be characterized, he will have reason to congratulate himself upon the vehicle he has chosen.4 When Godwin refers to philosophers and says that it is but of late that the importance of political principles has been ‘adequately apprehended’, he is referring mainly to himself and to Political Justice. He is suggesting that the two books are to communicate the same general truth – that the spirit and character of government intrudes itself into every rank of society – but that they will communicate different aspects of this truth, by different means, to different audiences. When Godwin came to revise Political Justice for its second – 1796 – edition, he rewrote the first sentence of it with a view, it would seem, to emphasizing this complementarity between the treatise and the novel: The object proposed in the following work is, an investigation concerning the form of public or political society, that system of intercourse or reciprocal action, extending beyond the bounds of a single family, which shall be found most to conduce to the general benefit. (4. 11) Putting these two passages side by side – one from the Preface to Caleb Williams, the other introducing the 1796 edition of Political Justice – we may notice how the word ‘story’ and the word ‘family’ seem to attract each other, through their use in the similar phrases ‘a single story’ and ‘a single family’. More broadly, we may be tempted to conclude that for Godwin distinct forms of discourse (abstract philosophical speculation on the one hand, narrative fiction on the other) correspond to distinct areas of subject matter (the public world of government on the one hand, the private world of family and domestic life on the other) and to the talents and preoccupations of different kinds of reader (the learned on the one hand, the unlearned but literate on the other).
84
Narrative Order, 1789–1819: Part II
Furthermore, all these distinctions can seem, to modern readers, strongly gendered: women playing a central role in family life, as characters in novels, and as writers and readers of novels, but a relatively minor part in public life or as the writers, readers or subject matter of philosophical treatises. There are however two difficulties involved in understanding Godwin to mean that his treatise and his novel are complementary in these particular ways. One difficulty has to do with narrative, the other with family. The truth which both books aim to communicate is to be communicated, in Caleb Williams, by a narrative fiction. But Godwin’s Preface betrays an uncertainty about whether he has succeeded in doing this and even whether it can be done. The novel has ‘a purpose more general . . . than immediately appears upon the face of it’. Godwin is conscious that there may be some gap between the ‘general view of the modes of domestic and unrecorded despotism’ and the ‘progressive [i.e. sequential] nature of a single story’; between the general and the single; between the aim of ‘teaching a valuable lesson’ and the means – the generating of ‘interest and passion’ – by which this is to be done. Godwin is concerned that readers may find it hard to move from descriptions of what a particular set of people did (or, since he is writing fiction, what they may be imagined to have done) to generalized statements about what people, or particular kinds of people, do (what worried Godwin may have reassured his publisher, Benjamin Crosby, who was prepared to publish the novel but not the Preface in 1794, presumably in the belief that the novel’s politics were less dangerously clear than those of the Preface). The similarity between the phrases ‘a single story’ and ‘a single family’ does indeed suggest that there is an affinity between stories and families by virtue of their particularity. The public world of government and the abstract world of philosophy are, it could be said, the realms of universally applicable rules; while the domestic worlds of the family and of the novel are the realm of variety and particularity. And it can seem to a modern reader of these passages that Godwin is addressing just this assumption, in a way that partly accepts it and partly challenges it. His novel will, he may appear to be saying, deal with relations between individuals in a domestic setting; but it will show that, as George Eliot was later to put it in Felix Holt, the Radical (1866), ‘there is no private life that has not been determined by a wider public life’. Or, closer to home, we could take Godwin to be writing a story of ‘domestic tyranny’ as Mary Wollstonecraft did in her unfinished novel
William Godwin: Stories and Families
85
The Wrongs of Women; Or Maria (written in 1796 and 1797; edited by Godwin for posthumous publication in 1798). But if these are some of the expectations raised by Godwin’s references to family life and domestic despotism, the novel itself should come as something of a surprise. Sexual relationships, relationships between husbands and wives, relationships between parents and children or other close kin: these things are confined to important but brief episodes. The novel’s central relationships are non-sexual ones between men who are not kin, notably between Count Falkland and Squire Tyrrel and between Falkland and his secretary, Caleb Williams. This secretary is sometimes called a servant but he is not what we would call a domestic servant, and it is not what we would call a domestic or a family relationship. If this is so, perhaps we should reconsider that first sentence of the 1796 Political Justice. Strictly speaking, we should note, Godwin is not distinguishing in that sentence between the public realm on the one hand and the private realm of the family on the other; rather, he describes the public realm as ‘extending beyond the bounds of a single family’. And this could mean that the public realm consists not of the world outside families but of the collectivity of families, while the private consists of any particular family (one’s own, for instance). And if that is so it could well be that Godwin is not talking about ‘the group which consists of a Father, Mother, and Children’ but – to use one of the two definitions of ‘family’ in Johnson’s Dictionary (1755) – ‘all who live in the one house; household’. ‘All who live in the one house; household’ is the sense of ‘family’ which Naomi Tadmor has tracked through mid-eighteenth-century journals, conduct books and popular novels. Used in this sense, ‘family’ could certainly be used to refer to a family consisting of parents and children but it did not mean that, would not group the people in that capacity. The word could be used, as Tadmor demonstrates, when what was being referred to was a household which did not include anyone related by either blood or marriage, such as a household consisting only of a master and two apprentices. As she explains, ‘the boundaries of these household-families are not those of blood and marriage; they are the boundaries of authority and of household management’.5 That is, this is an explicitly political and economic unit, whatever else it may also be. In the same vein, Anna Kussmaul has even argued that ‘early modern English has no word whose meaning was “all in the household except the servants” ’.6 ‘Family’, used unequivocally as a word for this grouping, seems to emerge somewhere during the second half of the
86
Narrative Order, 1789–1819: Part II
eighteenth century. Wollstonecraft’s Maria uses it, I believe, in this sense, as does Malthus, responding to Political Justice in his first Essay on Population in 1798. The new usage appears in the work of these two very different writers because both are distancing themselves in a determined fashion from some of the core assumptions on which the concept of the household-family is based: Wollstonecraft because she sees men rather than simply fathers (heads of households) as oppressors; Malthus because he is focusing on the (proletarian) family specifically as a unit of procreation existing outside the proper remit of government. In Caleb Williams, Godwin is probably using the word ‘family’ to mean household. Tadmor, arguing that ‘the household-family framework . . . can expand and contract and include many individuals’, notes that ‘one phrase that recurs in the context of such changes is “to be taken into the family”’ (p. 119). ‘Family’ means household when Falkland invites Caleb ‘into his family’ (p. 6) as his secretary, or when Tyrrel invites the younger Hawkins ‘into his family’ as his ‘whipper-in-of-hounds’ (p. 37). Neither Caleb Williams nor Political Justice seem to use the word in its principal modern sense. Perhaps more surprisingly, only rarely does either book use the word to mean lineage or kin group (which Johnson’s Dictionary, in the first of its two definitions, calls ‘those that descend from one common progenitor; a race, a tribe, a generation’).7 Since these are the meanings which are so important to Burke in the Reflections we might have expected that Godwin would have attributed to Falkland a preoccupation with family in this sense. Interestingly, however, while Falkland is indeed concerned with the preservation of his ‘name’, he will preserve it by controlling his ‘story’, not by producing male heirs. For Falkland, the story line takes the place of the male line. As for the word ‘domestic’, that appears in Caleb Williams mainly as a noun, the name for an indoor servant. In this capacity it is the focus of an interesting uncertainty on Caleb’s part, and possibly on Godwin’s part as well: With respect to the domestics in general, they saw but little of their master. None of them, except myself from the nature of my functions, and Mr Collins from the antiquity of his service and the respectableness of his character, approached Mr Falkland, but at stated seasons and for a very short interval . . . One day when I had been about three months in the service of my patron I went into a closet or small apartment which was separated from the library by a narrow gallery. (pp. 7–8)
William Godwin: Stories and Families
87
While it is grammatically possible, it is nevertheless rather awkward to include the first person ‘myself’ among the third person ‘them’. Caleb is in any case talking about a difference between himself and the other domestics; but it is as if his sense of that difference is greater than the one which he openly acknowledges. It is as if he is not sure whether he and Mr Collins are really domestics. A consciousness of this uncertainty is suggested by Caleb’s reference to Mr Falkland, a few sentences later, not as his ‘master’ but as his ‘patron’: a word whose probable complement here would be secretary rather than servant. There is an uncertainty about status which is in part an uncertainty about nomenclature, and which appears more openly in the following exchange between Caleb and a magistrate: ‘And pray, sir, who is this Mr Falkland? And what may have been the nature of your connection with him?’ ‘Mr Falkland is a gentleman of six thousand per annum. I lived with him as his secretary’. ‘In other words you were his servant?’ ‘As you please’. (pp. 242–3) What we are dealing with in these two passages is the interaction between the language of family and the language of class. Godwin does not use the phrase ‘domestic servant’, which seems to be a nineteenthcentury locution that belongs with Mill’s sense of ‘family’. But Caleb’s discomfort in thinking of himself as either a domestic or a servant suggests that the two words do belong together. In wanting to differentiate himself by one degree from the domestics and the servants, Caleb is also suggesting a set of relationships within the patrician household more polarized between the family (in Mill’s sense) on the one hand and the living-in employees on the other, with himself poised rather uncertainly between the two. Godwin, then, is probably using ‘family’ to mean household; but he is using it, and the other social words associated with it, to define social relationships which put that meaning under considerable strain. It may of course be argued that since the novel takes the form of Caleb’s memoir, we are dealing, strictly speaking, with Caleb’s usages rather than with Godwin’s. Nevertheless, the history of authorial alterations to the novel shows Godwin wrestling with some of the same uncertainties of nomenclature as his protagonist. The two passages I have just quoted were not themselves altered in later editions – Falkland
88
Narrative Order, 1789–1819: Part II
is still referred to there as ‘master’, ‘patron’ and ‘Mr Falkland’, while Caleb is still Falkland’s ‘domestic’, ‘servant’ and ‘secretary’ – but in the novel as a whole, Godwin alters the balance, changing ‘my master’, in a majority of cases, to ‘my patron’ or ‘Mr Falkland’. If Godwin’s use of words like ‘family’ and ‘domestic’ should give us pause, so should his use of words such as ‘character’ and ‘story’. In the course of a bitter argument between Falkland and his neighbour Tyrrel, the latter exclaims: It has been my character when I had once conceived a scheme of vengeance never to forego it; and I will not change that character. (p. 44) What we have here is a use of the word ‘character’ which we have encountered in Johnson and Boswell but which is as foreign to us as the use of ‘family’ to mean household. It may be worth spelling out here in rather more detail the difference between this use of ‘character’ and our own. The word’s principal modern meanings are ‘personality’ and ‘person in a novel or play’. The relationship between these two modern meanings is by no means a simple one. A character in a novel is more or less the fictional representation of a personality; but only more or less, because a fictional character has a personality, as well as other features (such as clothes and a face). What nevertheless makes these terms almost complementary is their relationship to representation. ‘Character’ in the sense of personality contains no open reference to representation at all, while a character in a novel is precisely a particular kind of representation (a fictional one). The two meanings therefore more or less hinge along the line between things and the representation of things. Consequently, when the word is used to mean personality, that meaning is indeed linked to the other meaning, but the link is not visible. On the face of it, ‘character’ in the sense of personality has no relation to representation at all. As Tyrrel uses the word however, and as the word is normally used in eighteenth-century writing, the link between personality and representation (though not specifically fictional representation) is never broken. ‘Character’ might be translated, in Tyrrel’s declaration about himself, as something like ‘acknowledged identity’. It embodies precisely that ‘narrative idea of life’ which Alasdair MacIntyre identified in prenineteenth-century moral thinking and to which reference has already been made. According to this narrative idea, MacIntyre argued, ‘I am what I may justifiably be taken by others to be in the course of living
William Godwin: Stories and Families
89
out a story that runs from my birth to my death.’ In this idiom, to see yourself as you really are is to see yourself as a figure in a third-person narrative; so that, as we have seen, Boswell may give us Johnson’s character, rather than a description of his character, or his Life rather than a narrative of his life. A person’s life – including one’s own life – has the structure of a narrative in the third person. The place of this idiom in Caleb Williams is not easy to define. Falkland and Tyrrel’s commitment to it does not prove Godwin’s commitment to it. On the contrary, a novel that so consistently exposes the gap between what people are really like and the stories that are told and believed about them – the gap between ‘truth’ and ‘reputation’ – may well put a strain on this idiom. Indeed, it is tempting to read Caleb Williams precisely as an attack on MacIntyre’s ‘narrative idea of life’, seen as the animating spirit of the aristocratic society epitomized by such figures as Tyrrel and Falkland. Falkland is after all overwhelmingly concerned with his ‘name’: that is what his belief in honour turns out to amount to. He conceives of his life as a biography in the making and himself as its hero. Indeed, he models himself, as he explains to Caleb, on heroic fictional and historical figures. What he says about one of these figures – Alexander the Great – also applies to himself: ‘he formed to himself a sublime image of excellence, and his only ambition was to realise it in his own story’ (p. 98). Falkland’s position is written into his vocabulary here. His use of the word ‘story’, like Tyrrel’s of the word ‘character’, refuses to distinguish between an existence and the narrative of an existence. Falkland and his kind retain their social power partly by their control over stories, in a number of respects. Falkland in effect combines the skills of a patrician Othello and a patrician Iago. He alludes to Othello’s rhetoric of the ‘plain unvarnished tale’ (p. 94) and he has Iago’s ability to cut a plausible figure. As an actual master – man of property and magistrate – he already has the power which the self-made Iago has worked to acquire, the power to arrange people and objects so that they constitute effective evidence, tell-tale signs, signs that appear to tell their own story. Bearing in mind MacIntyre’s definition of ‘character’ as ‘what I may justifiably be taken by others to be in the course of living out a story that runs from my birth to my death’, we could even say that Caleb Williams, even though he is not a thief, could ‘justifiably’ be taken by others to be a thief. Within a world dominated by a narrative idea of life it is unlikely, Godwin seems to suggest, that Caleb Williams can ever be found innocent. Godwin’s critique of the justice offered by the Law can
90
Narrative Order, 1789–1819: Part II
be read as a critique of what is ‘justifiable’ in MacIntyre’s sense. The Law’s criterion of truth is plausibility. And it is not simply that lies can sometimes be plausible, but that they usually will be more plausible than truth. Justice, like art, deals with the probable rather than the possible. Justice is always poetic. A good story is always, to some crucial degree, an untrue story. Or so Caleb Williams often leads us to suspect. However, one reason why the novel is so claustrophobic is that Godwin does not offer any concepts to oppose to the narrative idea. He is, it would seem, committed to this idiom himself. Consequently, while the novel can be read as identifying this idiom with aristocratic despotism it can also be read – and with at least equal plausibility – as a defence of that idiom against its aristocratic corruption. It is, after all, in a court of law that Caleb, in the novel’s published ending, finally gets his story believed. Nor, as I have suggested, is Caleb’s vocabulary ever clearly distinct from the one employed by Tyrrel and Falkland. The aim of his continual storytelling, and of the memoir which describes it, is to ‘vindicate [his] character’ (p. 277). And while it is wrong to treat Godwin and Caleb as if they are the same person, it is impossible to define a clear and constant distance between them of the sort that might allow us to distinguish confidently between Godwin’s and Caleb’s understanding of ‘character’ and ‘story’. The difficulty of interpreting the novel in this respect becomes particularly acute, and particularly interesting, on those occasions when storytelling itself is a subject of discussion. For instance, about a third of the novel consists of the story which Falkland’s bailiff Mr Collins tells Caleb about his master’s earlier life. After communicating this story to us, Caleb reflects on the impact which hearing it had on his own life: It will soon be perceived how essential [Collins’s] narrative is to the elucidation of my own history. . . . Hitherto I had had no intercourse with the world and its passions; and, though I was not totally unacquainted with them as they appear in books, this proved to be of little service to me when I came to witness them myself. The case seemed entirely altered, when the subject of those passions was continually before my eyes, and the events had happened but the other day as it were, in the very neighbourhood where I lived. There was a connection and progress in the narrative, which made it altogether unlike the little village incidents I had hitherto known. My feelings were successively interested for the different persons that were brought upon the scene. My veneration was excited for Mr Clare, and my applause for the intrepidity of Mrs Hammond. I was
William Godwin: Stories and Families
91
astonished that any human creature should be so shockingly perverted as Mr Tyrrel. I paid the tribute of my tears to the memory of the artless miss Melvile. I found a thousand fresh reasons to admire and love my master. At present I was satisfied with thus considering every incident in its obvious sense. But the story I had heard was for ever in my thoughts, and I was peculiarly interested to comprehend its full import. . . . In the original communication it appeared sufficiently distinct and satisfactory; but, as I brooded over it, it gradually became mysterious. There was something strange in the character of Hawkins. So firm, so sturdily honest and just, as he appeared at first; all at once to become a murderer! (pp. 95–6) This intriguing passage may remind us of Wordsworth’s critique, a few years later, of the power of malicious gossip in his play The Borderers (1796) or in poems such as ‘Simon Lee’ and ‘The Thorn’ from Lyrical Ballads (1798). Such poems question the desire, of readers or of villagers, to turn what Caleb Williams calls ‘little village incidents’ into stories. The relationship between Godwin and Wordsworth in this respect – their shared interest in the social relations of storytelling – is as important as the more obviously political connection, and just as hard to define. Wordsworth’s critique of the storytelling impulse is sometimes of a very radical kind. Caleb Williams’s description of the impact on him of Mr Collins’s story suggests that Godwin’s critique of the impulse may be less thoroughgoing. Caleb tells us that he has lived in the real world and that he has also read about human passions in books, but that only now, when he can see the real lives of the people amongst whom he lives as persons in a narrative – a non-fictional adventure story – can he develop a passionate interest in them. This kind of everyday storytelling is what we call gossip if we wish to be pejorative about it. Caleb does not wish to be pejorative about it, but does Godwin? We may usefully link this question to the importance which Falkland attaches to the stories of famous fictional and historical figures such as Alexander the Great. For both Falkland and Caleb, relationships to other people become of passionate interest only when they are mediated through narrative representations: only when they can see themselves as participants in a story can they see themselves as truly participating in society. Does Godwin himself, however, wish to distinguish,
92
Narrative Order, 1789–1819: Part II
ethically or epistemologically, between different kinds of storytelling (Collins’s on the one hand, romances and romantic histories on the other) or does he wish to associate the two and condemn both equally? Clearly, there is some criticism of Collins’s storytelling: it wrongly identifies Hawkins as the murderer of Tyrrel. And the novel certainly seems to be offering some more general criticism of the obsessive curiosity about Falkland which Collins’s storytelling engenders in Caleb’s mind. On the other hand, there does not seem to be any suggestion that Collins’s story is not, in the main, a true one, let alone that a true story is a contradiction in terms. On the contrary, truth, it is suggested, has a narrative structure, so that when Caleb tells us that Collins’s ‘narrative is essential to the elucidation of my history’, the phrase ‘my history’ refers both to the historical narrative he is writing and to the fact that this narrative is about himself. ‘My history’ is a phrase that attributes a narrative structure to reality in the same way that Tyrrel’s reference to his own ‘character’ does. In learning to see the lives of others and of himself as histories, Caleb is learning to see them as they are. The distinction between ‘the narrative’ and ‘the little village incidents’ is therefore probably not between reality and stories about it but between reality as Caleb had previously experienced it and reality as, by courtesy of the narrative, he now finds it to be. Many stories or parts of stories may be false but life is nevertheless shaped like a story. It is when you see it as a story that life makes sense: the otherwise meaningless fragments fall into place, the confused scene comes into focus, you get things into perspective. The ‘narrative’ therefore does not refer only to a particular organization of discourse but also to the organization of events themselves which a discourse organized like this allows one to see. If this is so, the criticism of Falkland’s interest in narrative models – and, in a slightly different way, of Caleb’s own previous interest in novels – is not that they are stories but that they are untrue stories. They encourage people to adopt what Caleb subsequently calls a ‘fictitious’ or ‘counterfeit’ ‘character’. The contrast the book makes is not between roles and selves, but between true roles and false ones. One should present oneself to people in one’s proper character rather than a counterfeit character. What is wrong with Collins’s story is not that it naively – or malignly – represents life as an adventure story, but that it has identified the wrong person as the villain. Nevertheless, although Godwin provides no language to challenge Caleb’s, the language they share is put under great strain, even rendered incoherent by the human material with which it is required to deal. This is most evidently the case, in the passage under discussion, when
William Godwin: Stories and Families
93
Caleb tells us, with respect to Collins’s ‘story’, that ‘there was something strange in the character of Hawkins’. We may want to ask whether the phrase means that Hawkins’s personality (as the story represents it) is an odd one, characterized by extremes (the initial appearance of virtue, the subsequent murderous conduct); or whether it means that as a literary character Hawkins is not very well drawn, and the story not a very good story in this respect. These questions are however difficult to answer because they are probably couched in a language which is not Godwin’s. ‘There was something strange in the character of Hawkins’ is a phrase which probably assumes that, at least in this instance, there is an unbroken continuum between the following things: a badly constructed story, a story that is not a good story, an implausible story and an untrue story. And one element of that assumption, or a corollary of it, is that while a personality and the narrative representation of a person are of course different things, they nevertheless have an identical structure and coherence. Caleb bears witness to this assumption, stressing the importance of coherence, when he tells us that ‘my story will appear to have that consistency which is seldom attendant but upon truth’ (p. 5). ‘There was something strange in the character of Hawkins’ makes the same assumption, but also makes that assumption vulnerable. The sentence itself is indeed a very strange one which threatens to make a breach between semantic elements of the word ‘character’ as Caleb is trying to use it. The breach threatens to occur because, as the novel shows, people’s lives and identities often do not have the unity required of a narrative representation. It seems that Caleb is right to be led by a feeling of inconsistency to doubt Collins’s identification of Hawkins as the murderer of Tyrrel. But why should the criterion of consistency be relevant when precisely what marks the true culprit (and perhaps any murderer) is inconsistency? There is, after all, something just as strange in the character of Falkland, murderer and man of honour. Godwin talks implicitly about narrative in the chapter of Political Justice entitled ‘On Free Will and Necessity’. Explaining the ‘doctrine of necessity’, he writes: Trace back the chain [of acts] as far as you please, every act at which you arrive is necessary. That act, which gives the character of freedom to the whole, can never be discovered; and, if it could, in its own nature includes a contradiction. (3. 165)
94
Narrative Order, 1789–1819: Part II
One thing this passage makes very clear is that the doctrine of necessity involves privileging a retrospective view of life: it embodies the wisdom of hindsight. Furthermore we may well be reminded, reading a passage like that, of Godwin’s description of composing Caleb Williams, starting with the final volume and working backwards, ‘so that an entire unity of plot would be the infallible result’; strong compositional evidence for Peter Brooks’s claim that ‘the essence of narrative plot is the anticipation of retrospection’. And it is in this context, of being able to anticipate the future, that Political Justice talks about ‘character’: The idea correspondent to the term character, inevitably includes in it the assumption of necessary connexion. The character of any man is the result of a long series of impressions communicated to his mind, and modifying it in a certain manner, so as to enable us, from a number of these modifications and impressions being given, to predict his conduct. Hence arise his temper and habits, respecting which we reasonably conclude, that they will not be abruptly superseded and reversed; and that, if they ever be reversed, it will not be accidentally, but in consequence of some strong reason persuading, or some extraordinary event modifying his mind. If there were not this original and essential connection between motives and actions, and, which forms one particular branch of this principle, between men’s past and future actions, there could be no such thing as character, or as a ground of inference enabling us to predict what men would be from what they have been. (3. 161) The doctrine of necessity embodies a narrative view of life, a conjunction which is the basis of Gary Kelly’s influential conception of ‘the Jacobin novel’. However, Kelly’s theory does not take account of the extent to which Godwin is speaking a foreign language. According to the doctrine of necessity we are all ‘characters’, but in an eighteenthcentury sense of the word that has already been outlined and not in any of its modern senses, although, in giving this everyday eighteenthcentury sense a precise philosophical articulation, Godwin also makes it vulnerable – as we shall see – to a modern distinction between personality and representation. If one important fact about the composition of Caleb Williams is that it was written backwards, starting at the end, the other is that, at the last moment between manuscript and first edition, the ending was completely changed. In the manuscript ending of the novel Caleb fails
William Godwin: Stories and Families
95
to ‘vindicate’ his ‘character’, while in the published ending he succeeds in doing so. This is, paradoxically, an intensely and patently teleological novel in which the telos has been changed at the last moment. These two contrary tendencies correspond to the backward and forward movements involved in any process of composition: but in Caleb Williams these contrary tendencies are continuously visible in a way we are most familiar with from reading modern experimental novels or very badly written novels. We should remember of course that a person’s ‘temper and habits’, the continuity and predictability of which are, for Godwin, the basis of ‘character’, can nevertheless, in his view, be superseded and reversed ‘in consequence of some strong reason persuading, or some extraordinary event modifying his mind’. This proviso is however only that: it is an exceptional circumstance, and needs to be so if it is to be compatible with the concept of character as Godwin defines it in the context of the doctrine of necessity. Nevertheless, the presence of this proviso points to the possibility of the very different view of the relationship between past, present and future which we find in Godwin’s discussion of contracts and promises. In this section of Political Justice, particularly as it was rewritten for the 1796 edition, we have an emphasis on the discontinuity of a human life – a rejection of any entanglement of past, present and future – as insistent as the emphasis on its continuity in the doctrine of necessity.8 Godwin opposes social contracts as manacles which bind people together into a society by binding the present to the past and the future. This binding is effected by the act of promising: ‘the whole principle of an original contract rests upon the obligation under which we are conceived to be placed to observe our promises’ (3.87). But Godwin believes that we are morally obliged to do what is just in the particular circumstances in which we find ourselves, whether we have promised to do it, promised not to do it or not promised at all. Promising, argues the 1796 edition of Political Justice, involves a denial of the reality of time, since we abridge, and that in the most essential point, the time of gaining information, if we bind ourselves today, to the conduct we will observe two months hence. He who thus anticipates upon the stores of knowledge, is certainly not less improvident, than he who lives by anticipating the stores of fortune. (4. 93) Godwin here offers a view of human life in time that is – and is surely intended to be – the exact reverse of Burke’s description, in the Reflections,
96
Narrative Order, 1789–1819: Part II
of human life as a ‘contract . . . between those who are dead, those who are living, and those who are yet to be born’. For both Burke and Godwin, the binding together of the synchronous elements of society on the one hand and of past, present and future generations on the other are complementary aspects of a single ‘contract’. But Godwin wishes to unbind in both directions, even unbinding the links of the chain joining a person’s individual past to their own present and future (so that, for instance, people should not be held responsible for crimes they may have committed at an earlier stage of their lives). However, while Godwin may believe he is rebutting Burke, the actual relationship between them is – as their different uses of the term ‘family’ may already have led us to suspect – as much a matter of crosspurposes as of simple opposition. Godwin’s hostility to contractual promising is certainly at odds with his own commitment to narrative order; but the unusual and attenuated form of contract which Burke supports – a contract which has no knowable beginning or ending but has always already been made – itself represents, as we have seen, a challenge to narrative order. Burke does indeed foreground the temporal mechanisms which are at the heart of narrative, the mechanisms of anticipation, retrospection and anticipated retrospection. Social order, for Burke, is certainly linked to unbroken sequence, to that element of narrative. But he foregrounds the mechanisms of narrative so as to put into the background what they are normally designed to produce: narratives, with their beginnings and endings. And these temporal mechanisms are most firmly embedded, for Burke, in the ‘entailed estate’ or ‘family settlement’, a version of the ancient link between family and state, but a temporalized version in which ‘family’ means lineage rather than household. Godwin is especially sceptical about those contracts through which families – in the sense of households – are formed. The marriage contract is of course one of these and Godwin’s scepticism about marriage was in part an instance of his wider hostility to contractual promising. However, the principle example of the evils of contractual promising in the 1796 version of Political Justice is of another sort of contract that establishes a family relationship. Furthermore, it is in this passage on contractual promising that the two groups of words that I have been discussing – the narrative words and the social words – are most clearly interdependent: Let the case be of an indentured servant. Why should I, unless there be something in the circumstances obliging me to submit to this
William Godwin: Stories and Families
97
disadvantage, engage to allow him to reside for a term of years under my roof, and to employ towards him a uniform mode of treatment, whatever his character may prove in the sequel? Why should he engage to live with and serve me, however tyrannical, cruel, or absurd may be my carriage towards him? We shall both of us hereafter know more of each other, and of the benefits or inconveniences attendant on our connection. Why preclude ourselves from the use of this knowledge? Such a situation will eventually generate a perpetual struggle between the independent dictates of reason, and the conduct which the particular compact into which we have entered, may be supposed to prescribe. (4. 93) If anything in Political Justice should remind us of Caleb Williams it is this passage, where ‘character’ is discussed in the context of the relationship between a servant and master bound to one another through time. What is particularly noticeable is the effect which Godwin’s opposition to contractual promising has on the word ‘character’. If the master cannot predict the future conduct and temper of his servant, nor the servant the future conduct and temper of the master, how can either be said to have a character at all, in the light of the definition of ‘the term character’ offered in the context of the doctrine of necessity? Why should there be ‘something strange in the character of Hawkins’, if peoples’ personalities can ordinarily and as a matter of course – and not just as a result of ‘some strong reason or some extraordinary event’ – be expected to change radically over time or to turn out to be radically different from what we took them to be? Or perhaps ‘strong reasons and extraordinary events’ are, for Godwin writing in 1793, the rule rather than the exception. Godwin wishes to use ‘character’ in a way that embodies a narrative idea of life, but he also, here, wants to insist that unpredictable inconsistency, a lack of predictable ‘connection’ between present and future, is the rule rather than the exception. The crucial phrase here is ‘whatever his character may prove in the sequel’. What does this mean? Does ‘in the sequel’ mean ‘in what follows’ or ‘in a subsequent instalment of the narrative’, both meanings available at the time? Probably Godwin has the former in mind, but it is not wholly anachronistic to think of the latter – specifically literary – meaning, or therefore of the modern literary meaning of ‘character’. It is as if, asserting the view that human beings can normally be expected to change frequently and radically over time, Godwin finds himself attributing to human personality only that degree of consistency visible
98
Narrative Order, 1789–1819: Part II
in those literary characters who appear in a sequence of linked narratives, such as Falstaff or Faust. Caleb Williams enters Falkland’s family as the result of a verbal agreement rather than strict written indentures. Nevertheless, the bond between the two men, the feeling each has that he cannot escape from the other and that the other cannot escape from him, is exactly as it would be if they were bound by indentures. Furthermore, when Falkland discovers that Caleb ‘has formed an intention of quitting his service’, the bond is ruthlessly enforced by means of a kind of Faustian D-notice. That is, Falkland exchanges his ‘story’ – the corrected version of the story communicated to Caleb by Collins – for Caleb’s silence. This possibly unbreakable contract intensifies the master–servant dialectic, the ambivalence, the exchange of qualities and powers between the two men. Initially it puts Falkland on top, but in a precarious way because he is now, more than ever, dependent on his dependent. And when, finally, Caleb decides to break the contract and (in the published ending) reverses the positions of power, the see-saw starts to move fast: Falkland throws himself at Caleb’s feet in humility and admiration, an action that only serves to elevate him once more in Caleb’s eyes. The hierarchical society which Godwin presents in Caleb Williams is one in which the master–servant relationship is central. Caleb vows at one point ‘never to fill the part of the oppressor or the sufferer’, but this turns out to be impossible. Attempts to find a degree of freedom or equality outside the world defined by this relationship – in vagrancy, in wage-labour, in leasehold tenancy – do not enable people to escape from what Caleb calls ‘the general policy of landowners’. Nor is it really possible to escape inwards into one’s mind, satisfied with one’s innocence in one’s own eyes. One’s character as an innocent person needs to be confirmed by another; but there is no god and the confederacy of masters controls all the mechanisms of social recognition. In A Vindication of the Rights of Woman Mary Wollstonecraft had written: ‘I wish to show that the first object of laudable ambition is to obtain a character as a human being, regardless of the distinction of sex’, a use of ‘character’ identical to Caleb’s and meaning, approximately, an acknowledged identity or justified reputation. In the world Godwin describes, it seems unlikely that your identity can ever be acknowledged, but without such acknowledgement there is no identity. It may not therefore be surprising that Caleb (in the 1796 revision of the novel) thinks about committing suicide in a way which suggests that the only truly private life is private death: ‘No human being must see me’, said I, ‘at the moment that I vanish forever’ (p. 322).
William Godwin: Stories and Families
99
What Caleb Williams lacks is words with which to name the sort of private life associated with ‘family’ in its modern sense. That this lack is palpable is suggested by the widespread feeling that parental and conjugal relationships are present in this novel, but hidden: hidden within the master–servant relationship, trying to get out. Readers often feel they can make more recognizable sense of the novel’s central relationship if they think of it as a father–son relationship – Caleb and Falkland bound by strong psychic indentures – or as a partially sexual one (a recent British television version of the novel saw it in this way). These are in one sense anachronistic readings which fail to come to grips with the foreign world of masters and servants, and with the dead language of ‘family’ and ‘story’. They are not wholly anachronistic however. Godwin himself, after all, later noted ‘a certain similarity between the story of Caleb Williams and the tale of Bluebeard. . . . Falkland was my Bluebeard. . . . Caleb Williams was the wife’ (v.ii).9 Godwin’s modern editors agree with him, suggesting that the novel ‘rings a bell as the characteristic woman’s fantasy of domestic incarceration’, anticipating a ‘classic nineteenth century genre . . .: the fictional autobiography of the intensely aware housebound woman’ (v.45). That may have been what Mary Wollstonecraft saw in it, liberating some of these features of Things as They Are; or the Adventures of Caleb Williams in The Wrongs of Women; or, Maria. And what happens in the process is that the meaning of certain key words also changes. ‘Domestic incarceration’ is what both novels are about, but in distinct senses of domestic. We move from the domestic tyranny of masters to the differently domestic tyranny of fathers, husbands and brothers. Writing about indentures, Godwin described ‘the perpetual struggle between the independent dictates of reason, and the conduct which the particular compact into which we have entered, may be supposed to prescribe’. If we add ‘feeling’ to ‘reason’, this is the conflict which Maria and Caleb each describe in their memoirs and which each finds the courage to resolve by repudiating the contract in their passionate speeches in court. But in Caleb Williams both would be family contracts; while in Maria the only family contract – the only contract through which you join or constitute a family – is the marriage contract.10 And while telling your life-story to other people is an equally fundamental activity in both novels, its purpose is quite different. In Caleb Williams its purpose is to establish your name and character; in Maria, as we shall see, it is to establish personal solidarity, a politically aware personal relationship.
6 Wordsworth’s Moving Accidents
‘Othello’s Relation of his Courtship to the Senate’ Most potent grave and reverend signors, My very noble and approved good masters, That I have ta’en away this old man’s daughter, It is most true; true I have married her; The very head and front of my offending Hath this extent, no more. Rude am I in my speech And little blessed with the soft phrase of peace; For since these arms of mine had seven years’ pith Till now some nine moons wasted, they have used Their dearest action in the tented field; And little of this great world can I speak More than pertains to feats of broil and battle; And therefore little shall I grace my cause In speaking for myself. Yet, by your gracious patience, I will a round unvarnished tale deliver Of my whole course of love: what drugs, what charms, What conjuration and what mighty magic – For such proceedings am I charged withal – I won his daughter . . . Her father loved me, oft invited me, Still questioned me the story of my life From year to year – the battles, sieges, fortunes That I have passed. I ran it through, even from my boyish days To th’ very moment that he bade me tell it: Wherein I spake of most disastrous chances, 100
Wordsworth’s Moving Accidents
101
Of moving accidents by flood and field, Of hair-breadth scapes i’th’imminent deadly breach, Of being taken by the insolent foe, And sold to slavery; of my redemption thence, And portance in my travels’ history . . . . . . This to hear Would Desdemona seriously incline: But still the house affairs would draw her thence, Which ever as she could with haste dispatch She’d come again, and with a greedy ear Devour up my discourse; which I observing Took once a pliant hour, and found good means To draw from her a prayer of earnest heart That I would all my pilgrimage dilate Whereof by parcels she had something heard, But not intentively. I did consent, And often did beguile her of her tears When I did speak of some distressful stroke That my youth suffered. My story being done, She gave me for my pains a world of sighs: She swore, in faith ’twas strange, Twas pitiful, ’twas wondrous pitiful; She wished she had not heard it, yet she wished That heaven had made her such a man. She thanked me, And bade me, if I had a friend that loved her, I should but teach him how to tell my story, And that would woo her. Upon this hint I spake: She loved me for the dangers I had passed, And I loved her, that she did pity them. (Othello, Act 1, scene 3, lines 76–94, 127–38, 144–67, quoted from Elegant Extracts in Prose and Verse, London, 1791, Book III, p. 89) Caleb Williams claimed that he, like Othello in the extract quoted above, could persuade his listeners by telling a ‘plain unvarnished tale’, a tale he hoped would discredit the varnished one told by his Master, Count Falkland. George Crabbe, at the start of his most systematic exploration of the relationship between life and story, ‘The Parting Hour’ (1812), quotes the lines in which Othello tells the senators how Desdemona’s father first ‘questioned me the story of my life’ and how he, Othello, then ‘ran it through, even from my boyish days / To the
102
Narrative Order, 1789–1819: Part II
very moment that he bade me tell it’.1 Watkin Tench, in his Letters Written in France, identifies the French as the contemporary version of Othello’s ‘insolent foe’, implicitly attaching himself to a line of soldierstorytellers that also includes Julius Caesar who ‘knew not only how to execute, but to narrate deeds of glory’ (p. 5). However, it was Wordsworth who most frequently cited Othello’s speech to the senators, usually by quoting from Othello’s reference to ‘moving accidents by flood and field’. In our own time, the time of what has been called the ‘narrative turn’ in the human sciences, Othello’s speech has once again become a touchstone in discussions of narrative and storytelling. Stephen Greenblatt uses Othello to develop his concept of ‘narrative self-fashioning’, linking Othello’s talent for autobiographical storytelling in Act I, scene 3 to his subsequent vulnerability to Iago’s plotting. ‘Othello characteristically responds to his experience by shaping it as a story’, and this is why, Greenblatt argues, he is so susceptible to Iago’s manipulation of tell-tale signs, the tales they tell ‘based on good Aristotelian principles of probability’. ‘It is’, for instance, ‘eminently probably that a young, beautiful Venetian gentlewoman would tire of her old, outlandish husband and turn instead to the handsome, young lieutenant: it is, after all, one of the master plots of comedy.’2 Othello’s speech is also, as we have seen, behind the subtitle of Ross Chambers’s study, Story and Situation: Narrative Seduction and the Power of Fiction. Othello’s storytelling figuratively ‘seduces’ the senators because they can well believe it would have literally seduced a young girl like Desdemona: ‘I think this tale had won my daughter too’ is the Duke’s response to Othello’s speech. The implication, once again, is that courts of law, like Aristotelian dramas, deal in probabilities. Courts judge whether somebody did something on the basis of an assumption about whether that kind of person would do that kind of thing. Hence the power of the storyteller, who deals in such probabilities too. Ross Chambers is interested, as the other half of his subtitle suggests, in narrative ‘power’, in ‘how to do things with stories’. However, and as we have also noted, the link between ‘doing things with words’ and ‘doing things with stories’ is closer and more interesting, in Othello, than Chambers suggests. The editors of Elegant Extracts were right to refer to Othello’s ‘courtship’ of Desdemona; Brabantio’s complaint was not just that his daughter had been seduced by Othello but that he had ‘ta’en [her] away . . . married her’. Othello’s seductive storytelling led directly to the actual performatives of the marriage contract. And then Othello’s second act of storytelling – his
Wordsworth’s Moving Accidents
103
‘relation of his courtship to the senate’ – persuades a reluctant Brabantio to give that marriage contract the authority of his paternal consent. Wordsworth is reported to have described Othello as one of the three ‘most pathetic of all human compositions’.3 Its influence on his work in the second half of the 1790s, after he left revolutionary France and Annette Vallon, is very striking, notably where storytelling is presented as playing a significant part in the exercise of political, military and sexual power. In his own tragic drama, the 1797 Borderers, with its Jacobinical Iago-figure Rivers, stories – mainly untrue stories, especially the untrue story told by Rivers – provide ‘the ground and motivation for every significant thought and action’.4 The 1800 Lyrical Ballads poem ‘Ruth’ tells the story of a young English girl attracted by the glamour of ‘a Youth from Georgia’s shore’ (line 13), an exotic freedom fighter, of English descent though ‘from Indian blood you deem him sprung’ (line 29): Among the Indians he had fought And with him many tales he brought Of pleasure and of fear, Such tales as told to any Maid By such a Youth in the green shade Were perilous to hear. (lines 37–42) Of march and ambush, siege and fight, Then did he tell; and with delight The heart of Ruth would ache; Wild histories they were, and dear; But ’twas a thing of heaven to hear When of himself he spake!5 The soldier-storyteller’s tale leads, tragically for Ruth, to their marriage: “We in the Church our faith will plight, A Husband and a Wife.” Even so they did; and I may say That to sweet Ruth that happy day Was more than human life. (lines 98–102)
104
Narrative Order, 1789–1819: Part II
The thematic allusions to Othello in the poem are pinned down by the verbal echo. Desdemona’s ‘I wish that heaven had made me such a man’ is echoed by ‘such tales as told to any maid / By such a youth . . . were perilous to hear’, underlining Wordsworth’s response to that element in Othello’s speech and in the senators’ response (‘this tale had won my daughter too’) which suggests that when people participate in acts of seductive storytelling, as tellers or listeners, they are being led to behave (to take up Greenblatt’s point) in an Aristotelian way. They are being led to do what is ‘probable’, to behave not as an individual but as a kind of person, a copy (of a model, or of another person). Aristotle had explained the existence and power of the mimetic arts by the centrality and power of imitation in everyday life, especially in growing up: ‘for it is an instinct of human beings, from childhood, to engage in mimesis . . . it is through mimesis that [man] develops his earliest understanding’.6 Wordsworth, like Aristotle, sees that imitation is deeply rooted in human action; but where Aristotle accepts and celebrates such imitativeness, Wordsworth would resist it, while acknowledging its power. As he puts it in the ‘Ode’ (‘Intimations of Immortality’, completed 1804): Behold the Child among his new-born blisses, A four year’s Darling of a pigmy size! See where mid work of his own hand he lies, . . . See, at his feet, some little plan or chart, Some fragment from his dream of human life, Shap’d by himself with newly-learned art; A wedding or a festival, A mourning or a funeral; And this hath now his heart, And unto this he frames his song: Then will he fit his tongue To dialogues of business, love, or strife; But it will not be long Ere this be thrown aside, And with new joy and pride The little Actor cons another part, Filling from time to time his ‘humorous stage’ With all the Persons, down to palsied Age, That Life brings with her in her Equipage; As if his whole vocation Were endless imitation.7
Wordsworth’s Moving Accidents
105
In opposition to Aristotle and Othello, Wordsworth seems to be expressing the desire for an uncopied, an unscripted self. This would be a kind of individuality established in opposition to the parts – or, as we would now say, in another version of the theatrical metaphor, the ‘roles’ – which people may be called upon to play, for instance in the ritualized acts through which people become ‘a Husband and a Wife’ and organize their lives into a temporal sequence of linked stages or ‘ages’. Wordsworth recognizes, like Aristotle, that acting (people doing things) cannot be clearly separated from acting (people imitating people doing things), least of all in the prescribed performances of rituals and contracts; but his response to that recognition is an attempt, through his poetry, to encourage a scepticism about acting. This is a view of life and art which attempts to establish a radical alternative to the narrative idea of life as we have seen that at work in Johnson and, more precariously, in Tench and Godwin. What Wordsworth is attacking, as a species of self-alienation, is the way in which, through mimesis, humans become ‘characters’ in the eighteenthcentury sense. He is making this critique, in this passage from the ‘Ode’, in terms of the theatre rather than in terms of narrative as such (and through allusions to Hamlet, As You Like It and the history plays rather than Othello). But of course the narrative idea of life – which urges us to be see ourselves as figures in a third-person narrative – can also be thought of in terms of the theatre, with each of us as spectators of a drama in which we are also actors. It was nevertheless principally in terms of narrative, and through allusions to Othello’s storytelling that Wordsworth attempted to work out his attitude towards action and its representation. The fact that these allusions to Othello are often brief does not make them any less pivotal. Jonathan Bate has described a ‘process in many of [Wordsworth’s] poems whereby a phrase is quoted from Milton or Shakespeare, and the reader is asked to bring the context of the original to bear on the passage into which it is appropriated’.8 The fact that Wordsworth repeats the phrase ‘moving accidents’ on a number of occasions suggests that, despite its brevity, it is an important allusion whose meaning and context he may expect most of his readers to recognize and ponder. Many of his readers would have been in a position to live up to Wordsworth’s expectations because the long extract from Othello’s speech reproduced at the start of the present chapter was included in the much reprinted anthology Elegant Extracts in Verse and Prose, which – Wordsworth wrote in his third Essay Upon Epitaphs (1810) – ‘must be known to most of my Readers, as it is circulated every
106
Narrative Order, 1789–1819: Part II
where and in fact constitutes at this day the poetical library of our Schools’.9 The narrator of ‘Hart-Leap Well’ (1800) asserts that ‘the moving accident is not my trade’. The old peddler, Armitage in The Ruined Cottage (1798), urged by the narrator to continue his story of Susan and her family, warns that his story will not be full of ‘moving accidents’. In both cases the Shakespearean phrase appears as something like a manifesto statement on Wordsworth’s part, and has often been read as such. Jonathan Bate argues that when Wordsworth uses the phrase he is distancing ‘his kind of tragedy from Shakespeare’s’ as well as ‘his lyrical ballads from the ballad-mongers stock in trade of supernatural and extraordinary incidents’ (p. 99). James Averill, defining these ‘moving accidents’ as ‘the exciting and emotionally stimulating adventures of the heroic soldier’, argues that throughout Lyrical Ballads Wordsworth ‘pursues a narrative strategy of avoiding what he calls the moving accident’.10 It has of course long been held that Wordsworthian romanticism entailed, and can perhaps be defined by, a movement away from action and event – from narrative as conventionally understood – towards feeling and lyric meditation. Wordsworth’s statement in the Preface to Lyrical Ballads that ‘the feeling therein developed gives importance to the action and situation and not the action and situation to the feeling’ is often adduced in support of this view. Jonathan Wordsworth puts it like this: ‘Action’, Wordsworth writes in his play, The Borderers (1796–7), ‘is transitory – a step, a blow, / The motion of a muscle this way or that . . .’. ‘Suffering’, by contrast, ‘is permanent, obscure and dark, / And has the nature of infinity’. They are strange words to choose, but they tell us a lot about the poet and his beliefs. Actions, incidents, exciting events, have no place in his work – they themselves are too soon over, and so is the pleasure they stir in the reader. Suffering is different. It does not just happen, it goes on.11 However, if the whole passage from The Borderers had been quoted it could not so readily have been used to support Jonathan Wordsworth’s argument: Action is transitory, a step, a blow – The movement of a muscle – this way or that – ’Tis done – and in the after vacancy We wonder at ourselves like men betray’d.
Wordsworth’s Moving Accidents
107
Suffering is permanent, obscure and dark, And has the nature of infinity.12 The action and suffering are tied together here: the suffering is surely the result of the action. This is indeed a passage which turns away from action as such, as intrinsically perilous, intrinsically tragic, to the frequently silent and uneventful world of suffering which is its consequence. A turning away from the representation of action – that is, from storytelling or the drama – might follow from this. However, it is more likely that the author of these lines will turn to an exploration of the turning away itself, so as to understand the real meaning of the action – and the storytelling – turned away from, understanding them in other than their own action-packed terms, refusing to take them at their own valuation. This view is perhaps supported by those critics who have seen the title, Lyrical Ballads, as appropriate to the poems it names because it draws attention to the issue of genre, suggesting that the poems are written about – as well as in – the combination of lyric and narrative.13 Indeed, Averill’s argument is that Wordsworth’s strategy of ‘avoiding the moving accident’ entails in practice an appeal to the reader’s desire for a gripping story and, only then, a deliberate frustrating of that desire. ‘The evasion of the climax is the primary device shaping the plots of the 1800 Lyrical Ballads’. These are often poems designed to make us ponder our desire – and that may include Wordsworth’s own desire – for narrative. Wordsworth’s attitude towards these ‘moving accidents’ is then not straightforward, not a matter of simple rejection. He recognizes, and wishes in some respects to tap, their power. But just what are these dangerously seductive ‘accidents’? Averill calls them ‘exciting and emotionally stimulating adventures’, but ‘adventures’ seems, on the face of it, an odd translation of ‘accidents’. In fact I believe that we cannot properly grasp Wordsworth’s attempts to think through his attitude to narrative and action unless we recognize the extraordinary semantic complexity of the word ‘accident’ as he used it, particularly in the context of his allusions to Othello.14 First of all, we have to recognize that the word had a different range of meanings in 1800 from the range of meanings it has for us today. And secondly, we have to recognize that it had a meaning for Shakespeare which, while it seems to have been known had certainly become archaic by the time Wordsworth alluded to ‘moving accidents’. Writing to William Mathews from Racedown in 1795, Wordsworth says that he and Dorothy ‘do not see a soul. Now and then we meet a
108
Narrative Order, 1789–1819: Part II
miserable peasant in the road or an accidental traveler.’15 Wordsworth’s usage here is not difficult to translate: the accidental traveller is the chance traveller, the traveller who happens to pass that way even though it is not a regular thoroughfare. Nevertheless, this example should alert us to the fact that, when he uses the word ‘accident’, Wordsworth may not always be speaking in exactly our language any more than Johnson or Godwin are when they talk about ‘character’. Wordsworth does, it is true, frequently use the word ‘accident’ in our own most common sense, to mean a physical mishap (such as a boating accident). One of Samuel Johnson’s three definitions of the term in his Dictionary (1755) was ‘that which happens unforeseen; casualty, chance’, a definition which could include both physical mishap and Wordsworth’s ‘accidental traveler’. However, Wordsworth’s use of the word is often complicated by the possible presence of two other meanings with which we are less familiar. One of these meanings is philosophical and derives ultimately from Aristotle. This is the meaning Johnson gives as ‘the property or quality of any being, which may be separated from it, at least in thought’. Johnson’s example of this sense is taken from the 1745 edition of Isaac Watts’s Logick (1725): ‘An accidental mode, or an accident, is such a mode as is not necessary to the being of a thing; for the subject may be without it, and yet remain of the same nature that it was before; or it is that mode which may be separated or abolished from its subject.’ This philosophical usage, in which ‘accidental’ and ‘necessary’ frequently constitute a binary pair, is not used today outside of academic philosophy but was quite widely used in polite discussion of intellectual topics in the eighteenth century. Writing to John Wilson in June 1802 – he is criticizing Wilson for calling gorse ‘unsightly and unsmooth’ – Wordsworth commented that there are many aversions of this kind, which, though they have some foundation in nature, have yet so slight a one, that though they may have prevailed hundreds of years, a philosopher will look upon them as accidents. So with respect to many moral feelings, either of [lo]ve or dislike what excessive admiration was payed in former times to personal prowess and military success it is so with [the] latter even at the present day but surely not nearly so much as hereto[fore] . . . The loathing and disgust which many people have at the sight of an Idiot, is a feeling which, though having som[e] foundation in human nature is not necessarily attached to it. (p. 356)
Wordsworth’s Moving Accidents
109
It is notable here that ‘military prowess’ is mentioned as an example of an ‘accidental’ moral feeling. But it is notable too that it is an example which puts some pressure on the distinction it is designed to illustrate (the distinction between the ‘accidental’ and the ‘natural’). Wordsworth, writing in wartime, is clearly not quite sure how far ‘excessive admiration’ for military prowess is an historically specific rather than a universal moral feeling: it survives ‘even at the present day but surely not so much as heretofore’. The other unfamiliar meaning of ‘accident’ comes into play in Wordsworth’s writing only in the context of his allusions to Shakespeare because it is a meaning that seems to have largely gone out of use by the time Wordsworth wrote. Certainly, modern editors of Shakespeare need to footnote the word ‘accident’, in Othello’s speech and elsewhere, to tell readers that the word means ‘incident’ or ‘event’, rather than chance event or mishap. This older sense does not appear in Johnson’s Dictionary; OED’s most recent example is from 1750. The only later example I have come across (from Southey, writing in 1815) appears, significantly, in the discussion of battle narratives: ‘a battle can only be made tolerable in narration when it has something picturesque in its accidents, scene etc, etc.’.16 It is presumably this sense which Averill has in mind when he defines ‘moving accidents’ as ‘the exciting and emotionally stimulating adventures of the heroic soldier’. Averill assumes too readily that Shakespeare’s phrase must keep its Shakespearean meaning intact when it appears two hundred years later in Wordsworth’s poems. Nevertheless, we shall see that this probably was one of the meanings Wordsworth sometimes had in mind when he used the expression ‘moving accidents’. Between the various meanings of ‘accident’ there is of course considerable semantic overlap. In particular, Johnson’s definitions (the philosophical sense and the more familiar senses of ‘chance’ and physical mishap) can be very close: they both have to do with contingency and it is often hard to tell, in texts from this period, whether one rather than another of these meanings is being offered. Between these meanings on the one hand and the Shakespearean meaning – accident as incident – on the other, the semantic distance is much greater. But even in this case there can be some overlap: contingent incidents are, after all, a species of incident. It is only in certain contexts that the difference between these meanings of the word becomes significant. It is most importantly in the context of discussions of narrative that a semantic chasm does open between the Shakespearean meaning on the one hand and the philosophical and
110
Narrative Order, 1789–1819: Part II
modern meanings on the other. This is because incidents or events are normally thought to be the very essence of narrative while contingent or chance happenings are precisely what narrative (on most accounts) should avoid or minimize. Certainly Wordsworth seems to have associated the word ‘incident’ very specifically with narrative. We have seen how the subtitle of ‘Simon Lee, the Old Huntsman, with an incident in which he was concerned’ is designed to lead readers to expect that ‘a tale will be related’. In the context of the discussion of narrative, therefore, the word ‘accident’, unless it is very carefully defined, is in danger of contradicting itself, as if we were to use the same word for good and bad. When Wordsworth attempts to establish a position on narrative through an allusion to Shakespeare’s ‘moving accidents’, he brings into play – whether deliberately or not – two meanings that stand to one another as binary opposites. We have already examined, in our discussion of Godwin, the link between narrative and the term ‘necessity’, to which ‘accident’, in its philosophical sense, is often opposed. We have seen that Godwin’s doctrine of necessity, which impressed Wordsworth so much when he read Political Justice in 1794, embodied a narrative idea of life: In the life of every human being there is a chain of events . . . in consequence of which it was impossible for him to act in any instance other than he has acted. However: The contrary of this having been the conception of the mass of mankind in all ages, and the ideas of contingency and accident having perpetually obtruded themselves, the established language of morality has been universally tinctured with this error. (3: 168) This opposition between a ‘chain of events’ on the one hand and ‘contingency and accident’ on the other were derived ultimately from Aristotle and it is in the Poetics that one kind of happening (the sort that link together into a chain) are defined as what tragedy and narrative are made of, while another sort (accidents) are what tragedy and narrative try to exclude. For human beings, argues Aristotle, ‘happiness and unhappiness consist in action’ (p. 51) and tragic drama is the representation of a complete action. The most important element of
Wordsworth’s Moving Accidents
111
a tragedy – as of ‘narrative mimesis in verse’ (p. 115) – is ‘the structure of events’ (p. 51), their linking together by probability or necessity so as to constitute the whole action – with a beginning, middle and end – which the plot imitates. In life and its artistic imitation, accidental (contingent) events are non-essential events, while important events are those which are linked by probability or necessity, those which constitute a chain. In Shakespeare too the word ‘accident’ has a crucial role to play in the definition of narrative: but to name what narrative should consist of rather than what it should not consist of. If this is the case in Othello, it is even more explicitly the case at the end of The Tempest when Prospero promises the Neapolitans that he will spend part of the coming night With such discourse as I not doubt shall make it Go quick away: the story of my life, And the particular accidents gone by Since I came to this isle; . . . (5.i. 303–6) Accidents, for Prospero as for Othello, are what the story of one’s life is properly made up of. They are the kinds of event which naturally link together to form a chain of events, a story. For the academic tradition derived from Aristotle, as for Othello and Prospero, a story is made up of incidents that lead to one another in a chain of cause and effect, not of incidents that just happen to happen. But while ‘accidents’ is the name given to the former by Othello and Prospero it is the name given to the latter in the Aristotelian tradition. Whether this means that Shakespeare and the Aristotelian tradition differ in their terminology but agree in substance is a moot point. Samuel Johnson certainly wished to believe that Shakespearean drama was in line with Aristotelian principles, arguing, in his Preface to Shakespeare (1765), that Shakespeare ‘always makes nature predominate over accident’ and that his ‘plan has commonly what Aristotle requires, a beginning, a middle and an end; one event is concatenated with another, and the conclusion follows by easy consequence’.17 Wordsworth, by contrast, may have wished to see a distinction between Othello and Othello in this respect, with Shakespeare offering a critique, rather than an embodiment, of the human and aesthetic values embodied in the tradition. In any case, even if the difference between Shakespeare and the Aristotelian tradition is mainly terminological, that makes Wordsworth’s
112
Narrative Order, 1789–1819: Part II
meaning, quoting late seventeenth-century texts in his own late eighteenth-century ones, all the more puzzling and all the more interesting. How much of its original meaning and context does the phrase ‘moving accidents’ bring with it on its journey from Shakespeare’s text to Wordsworth’s? This is a familiar type of intertextual question, raised by all literary allusion; but it is raised especially sharply in this case because of the semantic transformation the word seems to have undergone between 1600 and 1800. To answer the question we need to look more closely at the particular contexts – the poems and the letters – in which Wordsworth alludes to Othello’s speech. As we have seen, Wordsworth first used the phrase ‘moving accidents’ in The Ruined Cottage (1798). If ‘the evasion of the climax is the primary shaping device of the 1800 Lyrical Ballads’, the same strategy had already been used in the earlier poem. The phrase ‘moving accidents’ occurs in the middle of the poem, when Armitage’s story about Margaret and her family has come to a halt and the person to whom he is telling the story, the poem’s narrator, begs for more: In my own despite I thought of that poor woman as of one Whom I had known and loved. He had rehearsed Her homely tale with such familiar power, With such a countenance of love, an eye So busy, that the things of which he spake Seemed present, and, attention now relaxed, There was a heartfelt chillness in my veins . . . . . . and impelled By a mild force of curious pensiveness, I begg’d of the old man that for my sake He would resume his story. He replied, ‘It were a wantonness, and would demand Severe reproof, if we were men whose hearts Could hold vain dalliance with the misery Even of the dead, contented thence to draw A momentary pleasure never marked By reason, barren of all future good. But we have known that there is often found In mournful thought, and always might be found, A power to virtue friendly; were’t not so, I am a dreamer among men – indeed
Wordsworth’s Moving Accidents
113
An idle dreamer. ’Tis a common tale, By moving accidents uncharactered, A tale of silent suffering, hardly clothed In bodily form, and to the grosser sense But ill adapted, scarcely palpable To him who does not think. But at your bidding I will proceed.18 In this context what face does ‘accident’ show to us, its Shakespearean face or its philosophical face, accident as incident or accident as chance? Or are both faces visible? Is Wordsworth simply confused, unaware that he has foregrounded the word in such a way as to highlight antagonistic semantic elements within it? Or should we see the signifier ‘accident’ as a means for papering over the crack between the two signifieds, containing a semantic conflict which Wordsworth cannot at the moment see a way of resolving? Or – a third possibility – can we read him as being fully in control of what is happening, and offering, through a quite deliberate foregrounding of the antithetical meanings, a provocative argument which might word as follows: the world of action and exciting events in which Othello and Aristotle (and perhaps Shakespeare) trade are in reality a non-essential, contingent aspect of human life compared with human feelings; furthermore, what are normally taken as chains of event – events linked together by probability or necessity – are really always, when you look at them closely, contingent events; all events are in fact chance events, all actions merely ‘the movement of a muscle, this way or that’. So far as the use of ‘accident’ in The Ruined Cottage is concerned it is hard to choose between these readings, although the third is probably least plausible. The terms in which Armitage agrees to continue his story in The Ruined Cottage were echoed, though in a very different tone, seven years later in a letter to Coleridge from Nordhaven dated 1805. Wordsworth describes his renewed attempts to complete the ‘Salisbury Plain’ poem which he had started in 1793 soon after his return from France: I also took courage to devote two days (O Wonder) to the Salisbury Plain. I am resolved to discard Robert Walford and invent a new story for the woman. The poem is finished all but her tale. Now by
114
Narrative Order, 1789–1819: Part II
way of a pretty moving accident and to build together in palpable knots the story of the piece I have resolved to make her the widow or sister or daughter of the man whom the poor Tar murdered. So much for the vulgar. Further the poets invention goeth not. This is my way of giving a physical totality to the piece, which I regard as finished minus twenty-four stanzas the utmost tether allowed to the poor lady. (pp. 256–7) Wordsworth is here admitting to doing exactly what Armitage in The Ruined Cottage refused to do. The peddler’s ‘scarcely palpable’ tale becomes the ‘palpable knots’ that will bind the Salisbury Plain poem’s story. The peddler’s reference to his story’s lack of ‘bodily form’ is echoed by the ‘physical totality’ which Wordsworth will give to the Salisbury Plain poem. The peddler’s dislike of appeals to ‘the grosser sense’ is echoed by Wordsworth’s embarrassment at appealing to ‘the vulgar’ (and itself echoes Brabantio’s ‘gross in sense’ from Othello). In other words, the ‘pretty moving accidents’ will give the poem a clearly articulated story line, make it a ‘totality’, give his youthful composition – the unfinished Salisbury Plain poem – what ‘the vulgar’ require (which is also, ironically, what Aristotle would require). The letter to Coleridge is referring, even more definitely than The Ruined Cottage did, both to the Shakespearean and to the philosophical senses of ‘accident’. The letter is jocular; its foregrounding of Shakespearian and Aristotelian formulations has the character of a shared joke, or a rather shamefaced attempt at one.19 Wordsworth writes, in a very Aristotelian way, that he has introduced Shakespearian ‘moving accidents’ into his poem so as to ‘build the story of the piece together into palpable knots’. And the content of these moving accidents echoes Aristotle too: Aristotle’s principal model in the Poetics is Sophocles’s Oedipus Rex and Wordsworth has ‘resolved to make [the woman] the widow or sister or daughter of the man whom the poor Tar murdered’. The paradox is of course that it is precisely the closeness to The Poetics (with their privileging of action and event) which confirms that when Wordsworth uses the phrase ‘moving accidents’ he is employing not Aristotle’s but Shakespeare’s meaning of ‘accidents’. But since such ‘accidents’ are precisely the opposite of what were called accidents in the Aristotelian tradition, Wordsworth’s usage can only be utterly – and it must be to some extent deliberately – paradoxical. It is far more likely here than in The Ruined Cottage that Wordsworth is saying (in a shared private joke with Coleridge): ‘all accidents (incidents) are really accidents (contingent incidents)’.
Wordsworth’s Moving Accidents
115
The situation is different again in ‘Hart-Leap Well’. This was the first poem in the new second volume of the 1800 Lyrical Ballads and the expression ‘moving accidents’ occurs at the beginning of the second half of the poem, as it did in The Ruined Cottage. The first part of this ballad-poem tells a romantic tale of a hunt, ending with the death of the hunted hart and the subsequent building of a Pleasure House at the behest of the huntsman, to commemorate the hart’s death-leap, to which the huntsman later takes his lover. The second part of the poem describes the narrator’s chance meeting, at the site of the now decayed memorial, with the shepherd from whom he hears the story that has already been told in the first part. While the first part appears to be written in sympathy with the heroic attitudes of the huntsman, the second shares the shepherd’s feeling for the hart in its ‘cruel leap’ and urges us Never to blend our pleasure or our pride With sorrow of the meanest thing that feels. (lines 179–80) Averill says that this refers to the literary pleasure the ballad encourages us to take in representations of cruelty, not to the hunting itself. But this is surely wrong. While the adventure and the adventure story are not the same thing, Wordsworth is nevertheless linking them together and questioning both. In this case, the hunting and the memorials in stone or verse share the values – imitate the anticipatory excitement – of heroic adventure. The word ‘moving’ suggests the same link: action is both mobile and inspiring; it raises the blood of the person acting and of the storyteller and of the listener. Wordsworth implies the existence of a dangerous collusion between heroic action and heroic storytelling, a perilous circuit which he sets himself to break by means of Averill’s ‘narrative strategy of avoiding the moving accident’. The second part of the poem begins: The moving accident is not my trade, To freeze the blood I have no ready arts; ’Tis my delight, alone in summer shade, To pipe a simple song to thinking hearts. (lines 97–100) Once again, moving accidents are associated with, rather than distinguished from, black arts (the description of them taken, in this instance,
116
Narrative Order, 1789–1819: Part II
from Hamlet rather than Othello). Jonathan Bate argues, as we have seen, that when Wordsworth uses this phrase he is distinguishing his kind of tragedy from Shakespeare’s and ‘his lyrical ballads from the balladmongers stock in trade of supernatural and extraordinary incidents’. On the other hand, a number of critics have argued that Wordsworth’s opposition to ‘moving accidents’ in ‘Hart-Leap Well’ is less clear-cut than Bate suggests. After all, Wordsworth is trading in the moving accident, at least in the sense that he has relied on rousing the anticipatory emotions of hunting and ballad-reading in his readers: emotions which the reader is only then, in retrospect, required to question. Indeed, it is clear that the narrator’s character and sympathies are as evident in the first part of the poem as in the second so that the allusion helps to express what Don Bialostosky has aptly called the narrator’s wish for a ‘simpler tale and purer pleasure than he in fact enjoys’.20 The phrase ‘moving accidents’ then, whatever ‘accidents’ may be taken to mean, refers to something which Wordsworth feels the power of even as he distrusts it in himself and in his readers. They have a power, a glamour, that exerts a perilous attraction, a power which he wishes in various ways to resist, but also to tap, or modify so that they may become ‘a power to virtue friendly’. In the case of ‘Hart-Leap Well’, there are clear reasons for reading ‘accident’ to mean incident rather than mishap; but there are also clear reasons for reading it to mean mishap rather than incident. In favour of the first reading, the old Shakespearean phrase can be thought of as matching as well as describing the archaism of the hunting ballad, and so keeping its old Shakespearean meaning. Moreover, one of the compositional contexts of ‘Hart-Leap Well’ seems to point us in the same direction. The first part of the poem is modelled quite closely on the German poet Bürger’s ballad ‘Der Wilde Jager’, the Wild Huntsman. In late 1798, soon after reading Bürger, Wordsworth wrote to Coleridge from Goslar as follows: In poems descriptive of human nature . . . character is absolutely necessary, &c: incidents are among the lowest allurements of poetry. Take from Burger’s poems the incidents, which are seldom or ever of his own invention, and still much will remain; there will remain a manner of relating which is almost always spirited and lively, and stamped and peculiarized with genius. Still I do not find those higher beauties which can entitle him to the name of a great poet. (p. 234)
Wordsworth’s Moving Accidents
117
Clearly, if Wordsworth is writing the first half of ‘Hart-Leap Well’ with Bürger in mind, then when he writes about moving accidents in the second part he means moving incidents. However, the hart in the poem does die what we would call an accidental death. Furthermore, while one of the compositional contexts of the poem points towards incidents, another points towards physical mishaps. The huntsman is called Sir Walter, which doubly alludes to Walter Scott because Scott had published a translation of Bürger’s poem and because in that translation the huntsman was called Earl Walter. Scott was at this point Britain’s best-selling poet and could well stand for that popular taste for heroic adventure that Wordsworth finds so perilous. Wordsworth seems to have admired Scott’s poetry much to the degree and in the terms he admired Bürger’s. Replying to a letter from Scott in November 1805 after the Scotts had visited the Wordsworths at Grasmere, Wordsworth wrote: Your Letter was very welcome; I am not apt to haunt myself with fears of accident from flood and field, &c; it was nevertheless pleasant to hear that you had got home well. (p. 640) This light-hearted allusion relies for its effect on our sense of the semantic deflation suffered by all three of Othello’s words – ‘accident’, ‘flood’ and ‘field’ – when they are moved out of their original heroic and military context into a discussion of the problems of getting about in modern Britain. Flood and field, which in Othello are a binary pair dividing the world into land and water, the scenes of sea-battles and land-battles respectively, are here metonymically reduced to inundation and agricultural field. Accident is likewise reduced, from incident to mishap. The metonymic reduction of Shakespeare’s meanings – a mishap is a particular kind of incident, a flood a particular kind of watery expanse, a field a particular kind of land – is a move from the language of heroic adventure to a more heterogeneous and less simply plotted world, a world that is not only more accident-prone but more accidental in the philosophical sense. However, while the letter to Scott may encourage us to read the ‘moving accidents’ of ‘Hart-Leap Well’ as mishaps, it is also true to say that in the letter itself Wordsworth is using ‘accidents’, more consciously and purposefully than in any of our previous examples, to mean both incidents and mishaps. That is, he is playing openly on the double meaning, and expecting Scott to recognize that fact.
118
Narrative Order, 1789–1819: Part II
We need to ask why it should be in this context, in a letter to Scott, that Wordsworth achieves a semantic control that has elsewhere eluded him. There are two possible answers to this. One is that while the allusion in the letter is light-hearted, it is not wholly light-hearted. Wordsworth is writing to Scott just two weeks after the defeat of the French at Trafalgar on 21 October 1805. Scott, an enthusiastic Volunteer, travelling back to Scotland some weeks earlier, when fears of invasion were widespread and intense, may well have expected British fields to soon become battlefields, and Wordsworth would have been aware of that expectation. The other possible reason for Wordsworth’s semantic self-consciousness in the letter points in a rather different direction, and can be summed up in the word ‘history’. The allusion to Othello’s ‘moving accidents by flood and field’ is especially apt to Scott as a writer of historical verse narratives of heroic deeds on land and sea who was actually living in a world of accidentally flooded fields. Indeed, it is prophetically apt to him, anticipating the more radically historicist solution to the relationship between these two worlds – the heroic and the mundane – which Scott began to elaborate in Waverley (1814). In the context of Scott’s historicism, Wordsworth can distinguish clearly between incident on the one hand and mishap and contingency on the other because he can accept the reality of both kinds of ‘accident’. He can accept both, because each is presented as belonging to a different period, one to the past, the other to the present. Heroic narrative was valid and is invalid. We can simultaneously entertain the divergent meanings of Shakespeare’s words because the move from one set of meanings to the other is presented as a historical move. The Shakespearean meanings are appropriate, we are to believe, to Shakespeare’s world, but not to our own. Of course this act of historicization is such a familiar move to us that we may not notice its significance. But it was a momentous move. And while Wordsworth was not a historicist of Scott’s stamp, it is a move that can help us to understand the most interesting and elaborate of all Wordsworth’s Othello allusions, in the two-book Prelude of 1799: Ere I had seen Eight summers (and ’twas in the very week When I was first transplanted to thy vale, Beloved Hawkshead! when thy paths, thy shores And brooks were like a dream of novelty To my half-infant mind) I chanced to cross
Wordsworth’s Moving Accidents
119
One of those open fields which, shaped like ears, Make green peninsulas on Esthwaite’s lake. Twilight was coming on, yet through the gloom I saw distinctly on the opposite shore Beneath a tree and close by the lake side A heap of garments, as if left by one Who there was bathing: half an hour I watched And no one owned them: meanwhile the calm lake Grew dark with all the shadows on its breast, And now and then a leaping fish disturbed The breathless stillness. The succeeding day There came a company, and in their boat Sounded with iron hooks and with long poles. At length the dead man ‘mid that beauteous scene Of trees and hills, and water, bolt upright Rose with his ghastly face. I might advert To numerous accidents in flood or field, Quarry or moor, or mid the winter snows, Distresses and disasters, tragic facts Of rural history that impressed my mind With images, to which in following years Far other feelings were attached, with forms That yet exist with independent life And, like their archetypes, know no decay. There are in our existence spots of time . . .21 The sequence of events that Wordsworth records here as following one another in time do not, he insists, succeed one another by any probability or necessity but, like so many important events in the Prelude, by chance and mischance. The Wordsworth who ‘chanced to cross one of those open fields’ is a reincarnation of the ‘accidental traveler’ of the 1795 letter to William Mathews; and happening to cross that field he comes across a physical mishap – an accidental drowning – which brings ‘numerous accidents’ to his mind. When we read the word ‘accidents’ here the broader Shakespearian meaning (accident as incident) may not enter our minds immediately. The substitution of ‘numerous accidents’ for ‘moving accidents’, in the context of an accidental death by drowning, for the moment keeps Shakespeare and his meanings completely out of sight. But then, with the phrase ‘flood and field’, the Shakespearian context surfaces: by the end of the line we have therefore registered both meanings of ‘accident’, but
120
Narrative Order, 1789–1819: Part II
separately, one after the other. Consequently, as in the letter to Walter Scott, we are made sharply aware of the semantic transformation which all the words have undergone in being transferred from Shakespeare’s world into Wordsworth’s world. ‘Flood’ has become inundation, ‘field’ has become meadow (the kind Wordsworth ‘chanced to cross’ earlier in the passage). As we read further, we come across many more altered fragments of Othello: Othello’s ‘tented field’ has become Wordsworth’s ‘open field’, his ‘rough quarries’ (not included in the Elegant Extracts selection) have become Wordsworth’s ‘quarry’, his ‘disastrous’ Wordsworth’s ‘disasters’. Even the words ‘moor’ and ‘tragic’ come from The Tragedy of Othello, The Moor of Venice, although it would be more appropriate to use the word ‘signifiers’ than ‘words’ here, since in the case of ‘moor’ the word has only survived severely disfigured. Only Shakespeare’s signifier has floated to the surface, attached to a quite unrelated signified. The passage has a number of implications for storytelling. One thing that it does is make a democratic challenge to what is still a powerful academic tradition of thinking about tragedy. This tradition tells us for instance, that when newspapers refer to ‘tragic accidents’, they are simply being ignorant, or using the word ‘tragic’ in a way that has nothing in common with the way we need to use it when we talk about literary tragedy. The Tragedy of Othello, The Moor of Venice has nothing to do, this tradition asserts, with the everyday mishaps of ordinary country people getting accidentally drowned in floods or stranded on moors. That is the sad material of local storytelling and gossip, not of tragic drama. It is outside the proper remit of the tragic poet on both formal and social grounds.22 From this perspective, Wordsworth’s reference to ‘tragic facts of rural history’ is inadmissible since, in an Aristotelian perspective, history deals with what particular people happened to do while tragedy deals with the kind of thing particular kinds of people are likely to do. Wordsworth does here seem to be questioning the high claims of literary tragedy and ‘poetry that imitates in narrative verse’, suggesting that there is no such thing, in reality, as incidents that are linked by probability or necessity together to form a whole action with a beginning, middle and end. If you look at any action closely it may reveal itself as no more than ‘the motion of a muscle, this way or that’. The events of Othello’s story, even of Othello itself, may separate, under scrutiny, into unconnected, and therefore chance incidents. The true significance of the play may lie in the insight it gives us into its characters’ – and our – all too human tendency to find a story where there is none.23
Wordsworth’s Moving Accidents
121
However, action and narrative are not rejected out of hand. Instead, as in the letter to Scott, they are assigned to the past. They were real and appropriate but they no longer are. Wordsworth’s lines suggest that the world of Shakespeare and Aristotle – the world of narrative events – is dead, but that the signifiers which used to bear the meanings of that world have survived, floating on the surface of the verse, attached now to other meanings but still recognizable, linking us to that disappeared past. And by talking in the figurative way I have – of meanings hidden beneath the surface, of floating signifiers and words disfigured – I am echoing the analogy Wordsworth himself seems to be making between this story about a dead man brought to the surface of Esthwaite Lake and those Shakespearean words surviving only as signifiers: an analogy between human decay and what, by an apparent coincidence, we have subsequently learned to call semantic decay. In the 1805 Prelude Wordsworth described himself ‘incumbent o’er the surface of past time’: As one who hangs down-bending from the side Of a slow-moving boat, upon the breast Of a still water, solacing himself With such discoveries as his eye can make . . .24 If we put this passage together with the drowned man passage we get an association between the surface of the lake and the present, the depth of the lake and the past, and also a conception of semantic transformation as words are transferred – by metaphor, by allusion – from the depths of the past to the superficial present. The difficult linking lines between the drowned man episode and the first reference to ‘spots of time’, tell us that the tragic facts of rural history impressed my mind With images to which in following years Far other feelings were attached – with forms That yet exist with independent life, And, like their archetypes, know no decay. (First part, lines 283–7) There is a similarity, these lines seem to suggest, between the way images continue unchanged over time in the individual mind (but get
122
Narrative Order, 1789–1819: Part II
attached to different feelings at different times) and the collective historical process of semantic development in which signifiers – sound images – continue unchanged (but attached to altered meanings). In a footnote to his discussion of the Othello allusions in the drowned man episode, Jonathan Bate asks: ‘does the recollection of the Moor of Venice lead by sub-conscious association to Wordsworth’s “Quarry or moor” (i.e. moorland)?’ (p. 258) Whether or not the association is ‘sub-conscious’ (and I tend to think it is conscious), Bate’s use of the psychoanalytic term is too easily appropriate to what Wordsworth is doing here. It is appropriate because, like Wordsworth’s lines, it associates what is deep (‘sub’) with what is old (Shakespeare). It is too easily appropriate because it is exactly this idea of a stratified self (of which psychoanalysis is a later formulation) which Wordsworth himself, in lines such as these, was helping to bring into being. And one function of this new stratified self was precisely to allow an event to be both accidental (in relation to the speaker’s conscious intention) and purposeful (in relation to a sub- or un-concious mind). Jonathan Arac has written of Wordsworth and Coleridge ‘laying the grounds for a new specification of self in the nineteenth century’, an ‘inward self’ which could be a defence against the ‘threat of traumatic change’ associated with the French revolution and which in fact ‘offers an individual equivalent to the Burkian state, which also requires both the veiling of origins and the possibility of always going further back’.25 I would only add that this inward, temporalized self was created through a process of stratification which links time to place in the fashion Wordsworth refers to as ‘spots of time’. This stratification was produced when, in lines such as these from the 1799 Prelude, Wordsworth brought together two ways of thinking that had probably never been brought together before. One was the association between what is beneath the surface and what is past (an association embodied most obviously in buried or drowned bodies); the other was the association between what is beneath the surface and what is richly significant (embodied in the metaphor of ‘deep’ thoughts, by contrast with superficial ones). Neither of these associations was new; Wordsworth’s innovation was to merge them, and in so doing find a way of resolving the set of problems – to do with action and narrative – focused, for him, in the complex word ‘accident’: we live in a plotless and accident-prone world, governed by contingency; but we can sometimes have access, in some places, to traces of our individual or collective pasts that were otherwise.
7 Crabbe’s Parables
25. And, behold, a certain lawyer stood up, and tempted him, saying, Master, what shall I do to inherit eternal life? 26. He said unto him, What is written in the law? How readest thou? 27. And he answering said, Thou shalt love the Lord thy God with all thy heart, and with all thy soul, and with all thy strength, and with all thy mind; and thy neighbour as thyself. 28. And he said unto him, Thou hast answered right: this do, and thou shalt live. 29. But he, willing to justify himself, said unto Jesus, and who is my neighbour? 30. And Jesus answering said, A certain man went down from Jerusalem to Jericho, and fell among thieves, which stripped him of his raiment, and wounded him, and departed, leaving him half dead. 31. And by chance there came down a certain priest that way: and when he saw him, he passed by on the other side. 32. And likewise a Levite, when he was at the place, came and looked on him, and passed by on the other side. 33. But a certain Samaritan, as he journeyed, came where he was: and when he saw him, he had compassion on him. 34. And went to him, and bound up his wounds, pouring oil and wine, and set him on his own beast, and brought him to an inn, and took care of him. 35. And on the morrow when he departed, he took out two pence, and gave them to the host, and said unto him, Take care of him; and whatsoever thou spendest more, when I come again, I will repay thee. 36. Which now of these three, thinkest thou, was neighbour unto him that fell among the thieves? 37. And he said, He that shewed mercy on him. Then said Jesus unto him, Go, and do thou likewise. (Luke 10:25–37. King James Authorized Version) 123
124
Narrative Order, 1789–1819: Part II
One of the things which Christianity offered to the Reverend George Crabbe was a tradition of storytelling, the biblical parable. He transmitted Gospel parables to his parishioners in sermons and in readings from the Prayer Book;1 he also wrote moral tales in verse which may be compared to biblical parables. Referring, in a sermon dated 29 April 1821, to the parable-like story of Isaac and Rebecca in Genesis 24, Crabbe wrote: We may dress our Stories in the Habits of Pomp and Circumstance. We may add Adventures and descriptions, but these simple Relations of natural Manners and common Events are not only more true but they are more affecting; they do not bring to our View extraordinary Situations and Trials that no one undergoes, but they teach us by Example what is our Duty.2 It is tempting to read this as a manifesto for Crabbe’s own verse tales. On the other hand, L.J. Swingle has argued that for Crabbe each moral situation is unique. As we read through the 1812 Tales, Swingle argues, we are invited to realize that the general truth to which a single tale may seem to lead is always pointedly undermined by the counterexample of another tale.3 If this is so, we can only treat a Crabbe tale as a moral example by ignoring its complicating context. My own view is that Crabbe did wish his stories to have the force of moral examples, but that he was also sceptical about their chances of success in this respect. Evidence both for Crabbe’s interest in this issue and for his divided mind about it is provided, I want to suggest, by those of his verse tales which themselves explore the question of whether a person can tell a story that has real moral power and whether that power can be retained when the story is retold (by word of mouth or in print) to other people in other circumstances. In the 1812 Tales, two poems in particular explore this issue, though in different ways. ‘The Widow’s Tale’ describes a widow who successfully uses her own life-story as an example, a parable, to persuade a younger woman to change her mind about marriage. ‘Procrastination’ brings parable and allusion together in a brilliantly targeted allusion to the parable of the Good Samaritan. The significance for Crabbe of a parable like this one is that, while it does offer to ‘teach us by example what is our duty’, it is also very carefully designed to take into account the difference between the storytelling circumstances it describes and the circumstances in which subsequent listeners and readers are likely to find themselves.
Crabbe’s Parables 125
Neither the Gospel writers nor Jesus as they represent him expected the world to last long enough for their stories to appear, many centuries later, translated into English, under the authority of a state church and its royal head. Nevertheless, the parable reproduced at the head of this chapter is clearly designed to survive transplantation into circumstances other than those it describes or those in which it was first told. The situation it describes is designed to strike readers or listeners in a very wide range of personal and cultural circumstances as significantly comparable to situations in which they may find themselves. Even without the authority of the Churches, this particular story might well have survived, if only because there have always been ‘roads’, in the sense of recognized two-way tracks, with people languishing beside them in need of help, and other people tempted to ‘pass by on the other side’. Circumstances have continued to remind people of the original story’s literal features. Some of its words find their way into people’s minds, the end of a thread which can bring the whole story line with it. Even though they do not live in first-century Palestine or seventeenth-century England, people still sometimes believe they know what it means to ‘go and do likewise’. The story can still prompt people to action, or at least stir their consciences. It can still have illocutionary force. When a poet alludes to a biblical parable, as Crabbe does in ‘Procrastination’, we may be drawn to think about the similarity between the way parables work and the way literary allusion works. The force of an allusion depends, like the continuing force of a parable, on the perception of similarity in difference. But literary allusion is less constrained in this respect than parable. A literary allusion may draw attention to elements of similarity between the old story and the new one so as to intensify our sense of the essential difference between the two. Wordsworth’s allusion to Othello’s speech in the 1798 Prelude certainly does that, as we have seen, and does it in a way that helps to produce a partially historical understanding of the human condition. Of course, parables too depend upon the perception of difference to some degree. When Jesus said ‘Go, and do thou likewise’ he did not mean that the young lawyer to whom he was telling the story should patrol the road from Jerusalem to Jericho in search of people beaten up by thieves. Nevertheless, if a story is to work as a parable, as distinct from working as literature, the perception of difference has to serve the perception of similarity. A parable like the Good Samaritan, the sort of parable that is a moral tale, only works if, perceiving comparable circumstances in their own lives, people are prompted to acts of mercy comparable to the Samaritan’s own; or when, recognizing that it is one who is not of
126
Narrative Order, 1789–1819: Part II
‘our kind’ who has behaved in a neighbourly fashion, we act differently towards ‘Samaritans’. Can Crabbe’s tales work as parables, or, alluding to parables, can they only work as literature? Many of Crabbe’s stories contain characters who tell stories that have a powerful effect on their listeners. Such poems lead us to compare, in this respect, the spoken stories they describe and the printed stories they are. They encourage us to ask what happens to stories when they move from one teller to another, one medium to another, one society to another. ‘The Widow’s Tale’ tells the story of a farmer’s daughter, Nancy Moss, sent at the insistence of her socially ambitious mother to a genteel boarding school. The mother has now died, and Nancy returns home to a father who expects his daughter to make herself useful and accept that her destiny is to be a farmer’s wife, not a lady of leisure. Seeking relief from what she sees as the crude ways of her family, Nancy sets out to befriend Lucinda, a widow living nearby, whose genteel appearance suggests she will be a kindred spirit. The widow accepts the offer of friendship; but, instead of echoing Nancy’s romantic views, she successfully counters them by telling the story of her own youthful romantic aspirations and their disappointment. Re-educated to accept her station in life, Nancy busies herself about the farmhouse, and before long is contentedly married to a young farmer, Harry Carr. Her father, pleased but puzzled, asks his son what caused this sudden change in Nancy. ‘ “It is the Widow’s deed” ’, replies the son. The father philosophizes: ‘Some are made for action, some to speak, And, while she looks so pitiful and meek, Her words are weighty, though her nerves are weak.’ (lines 407, 421–3) This is a poem about words as deeds, or – to borrow, once again, Ross Chambers’ adaptation of J.L. Austin’s title – ‘how to do things with stories’. Austin took most of his examples of performative utterances from religious, juridical and legislative acts. I have argued elsewhere that Crabbe’s supposedly documentary realism in ‘The Parish Register’ (1807) looks different when we realize that writing in a parish register is an integral part of the ritual event which the register itself documents.4 Parish registers help to bring baptisms and marriages about: they are a part of the reality they document, as is, though in a different way, ‘The
Crabbe’s Parables 127
Parish Register’. Crabbe’s later tales do not base themselves explicitly on social documents in the manner of ‘The Parish Register’. Nevertheless, the widow’s tale in ‘The Widow’s Tale’ does bring about a marriage service: in Austin’s terms, it has ‘illocutionary’ force, bringing into being – by removing the principle obstacle to – the more strictly performative or ‘perlocutionary’ speech of the marriage service itself. In this respect its power is closely comparable to the power to move of Othello’s ‘moving accidents’. Told to Desdemona, Othello’s story brings a marriage about, but improperly (because without her father’s permission); told to the Senate it retrospectively removes that obstacle, securing the father’s blessing. The widow’s tale seems to have this illocutionary power, but is this power shared by ‘The Widow’s Tale’, the poem in which the widow’s tale appears? There is some reason to believe that Crabbe thought his poem could share the power of the widow’s story. The poem is certainly one of a number of tales in which Crabbe and his storytelling character are closely aligned by their intention to reveal truths obscured by untruthful storytelling. The first tale of that kind, ‘Ellen Orford’, has already been mentioned. In it, the unfortunate Ellen Orford’s story of her own life – a veritable chapter of accidents – is contrasted with the neatly plotted fantasies and happy endings of romantic fiction. And the moral which Ellen drew from her own life-story was quite explicitly also the moral which Crabbe wishes us readers to draw from it and from his poem: Ellen Orford knows, That we should humbly take what Heav’n bestows. (lines 124–5) As for Nancy in ‘The Widow’s Tale’, her early life demonstrates the dangers of reading romantic fiction, and the tale the widow tells teaches her to distinguish between reality and these romantic fictions. Crabbe’s tale does seem to align itself with the widow’s tale, just as ‘Ellen Orford’ seems to align itself with Ellen Orford. However, if we put the opening lines from ‘Ellen Orford’, with their radical questioning of ‘plots and histories’ together with Crabbe’s comments on the biblical story of Isaac and Rebecca, which ‘teach us by Example what is our Duty’, there does seem to be a different emphasis, a difference which required that we think carefully about what it means for a life or a life-story to be an ‘example’.
128
Narrative Order, 1789–1819: Part II
To ‘teach us by Example’ a story must be an ‘example’ in two senses: an instance and an object lesson, an object lesson by virtue of being an instance. That is to say, the people or situations it describes are not unique and it shows us that (contrary perhaps to what we might have thought) we – or the situations we are in – are not unique. An exemplary tale – a tale, in effect, as a parable – is based on the notion of likeness, comparability, of individual lives as in fundamental respects the same as one another. It embodies, in other words, a narrative idea of the moral life in the sense that has been described, an idea which encourages us to see our own life as a narrative in the third person. Another way of putting this is to say that to be an ‘example’ is to be classifiable. And one thing this means for Crabbe is that we can each be identified as members of a social class. It is the ambiguous class position occupied by Nancy Moss in ‘The Widow’s Tale’ that made her reading of romantic fiction so powerfully pernicious. Romantic fiction helped to give her ideas above her station, even encouraged her to pay no heed to station. And yet, as we shall see, the grounds for thinking that the widow’s life can act as an ‘example’ to Nancy are themselves shaky, and precisely because of the obscurity surrounding the widow’s own class position. Crabbe’s commitment to ‘exemplary’ storytelling is matched by a radical scepticism about it, not just about romantic storytelling but about the ‘plots and histories’ integral to storytelling as such and the assumptions of comparability and likeness on which they rest. As with Godwin, we can never be sure whether he wishes to contrast life and story or true stories and false ones. In the detailed readings which follow we shall see Crabbe trying to negotiate this problem, and the problems of social classification with which it is associated. In ‘Procrastination’ he does it by alluding to a biblical parable, the parable of the Good Samaritan, which seems designed, with great precision and subtlety, to cope with just such problems of comparability, likeness and classification. Many parables are indirect answers to direct moral questions, such as (in the case of the Good Samaritan) ‘Who is my neighbour?’ It is in a similar spirit that the widow, in ‘The Widow’s Tale’, tells her own lifestory. ‘Quitting precept’, she says, ‘let example show / What joys from Love unchecked by Prudence flow’ (lines 210–1). The widow’s tale is explicitly designed, that is to say, to ‘teach [Nancy Moss] by Example what is [her] duty’. But what is Crabbe’s design on his readers? Of course he would not expect his readers, having read ‘The Widow’s Tale’, immediately to marry young farmers. On the other hand, he may hope
Crabbe’s Parables 129
to persuade them to act in comparably prudent ways when faced with comparable choices between romantic desire and the constraints of circumstance. If Nancy can recognize something of her own situation in the widow’s story, perhaps Crabbe expects readers to be able to recognize something of their own situation in his story. However, if Crabbe believes his tale can be not only as truthful but as directly effective as the widow’s, he may be disappointed. The similarities between ‘The Widow’s Tale’ and the widow’s tale may simply highlight the far more significant differences between them. The persuasive power which the widow’s tale has over Nancy may remind us how different our own ethics, or our own social problems, are from those which Crabbe and the widow seem to share. The poem may interest us precisely because we do not share its values, or because it illuminates a society different from our own. For instance, few modern readers will accept that disaster necessarily follows from marrying someone of a different social class. Consequently, we may not be persuaded that we should go and do likewise, or be sure what kind of behaviour would, in our circumstances, be ‘like’ the behaviour presented for our approbation in the poem. This is not, after all, a story told to us by a friend, but a printed poem, written by someone we have never met, which has (only just) survived radical cultural transformations. Nothing could more clearly distinguish the poem from the story it contains than the fact that Nancy can interrupt the widow who can then, in response, change her story’s emphasis somewhat. The interaction between poet and reader cannot be a dialogue in that literal sense. Furthermore, the distinction which the widow makes between ‘precept’ and ‘example’ means one thing in the context of her own relationship to Nancy, another in the context of Crabbe’s relationship to his readers. It can be a distinction between moral utterances and moral actions; but it can also refer, more narrowly, to different kinds of utterance: those that directly propound moral truths and those that communicate truths indirectly through narrative instances. The widow offers Nancy an ‘example’ in two senses: she offers the evidence both of herself and of her life-story. Crabbe’s poem can only be an example in the narrower sense. Finally, ‘The Widow’s Tale’ cannot move us in the sense that the widow’s tale moves Nancy: out of the cottage, back home and into church to marry Harry Carr. The poem will influence our feelings and understanding and therefore, presumably, in some way, our conduct; but it will do so indirectly and unpredictably, working variously on different readers. The only thing it may move all of us to is acts of interpretation.
130
Narrative Order, 1789–1819: Part II
At this point, however, pondering the influence which the poem may have on ourselves, we may be drawn to look more closely than we have done so far at the influence the widow’s tale has on Nancy. That the tale did influence Nancy is impossible to deny. After all, the widow told her story to persuade Nancy to do what, having heard the story, Nancy does; and Nancy’s brother asserts that there is a direct causal link between the widow’s words and the ensuing marriage. Nevertheless, the way in which the storytelling caused this effect is not at all clear. ‘Quitting precept .. . let example show, / What joys from Love unchecked by Prudence flow’, and in one sense the example does support the precept. It was doubtless imprudent of Lucinda and the apprentice to get romantically entangled. But when the father objects to the liaison, we probably share Nancy’s expectation that more imprudence will follow: ‘But you were true’, exclaim’d the Lass, ‘and fled The tyrant’s power who fill’d your soul with dread?’ (lines 248–9) In fact, however, the couple were endlessly obedient and prudent. We never see the ‘unchecked’ love we were promised. We see the results of being half-heartedly imprudent. Lucinda and the apprentice come to exemplify one of Lucinda’s other precepts – ‘Passion to Reason will submit’ – rather than the dangers that follow if Passion does not submit. Even if the couple were imprudent to fall in love, this seems to be something that they could only have fully known afterwards, when the father – fallen on hard times – ‘revealed his former fears’. This phrase seems to mean that the father had opposed the relationship for reasons other than those indicated at the time. The real reason would not have been the apprentice’s actual poverty but the combination of that and her father’s own anticipated poverty. The father’s social insecurity would have led him to see that, paradoxically, his daughter could only maintain her present class position by improving it, by marrying above herself. But nobody – including Crabbe – spells this out. The official moral – what Reason officially entails – is that in order to maintain your class position you should marry within your class: ‘Passion to Reason will submit – or why Should wealthy maids the poorest swains deny? Or how could classes and degrees create
Crabbe’s Parables 131
The slightest bar to such resistless fate? Yet high and low, you see, forbear to mix . . .’ (lines 198–202) Perhaps what Nancy really learns from the widow is named by the title of my other Crabbe tale, ‘Procrastination’. As the widow puts it: ‘ “Let not romantic views your bosom sway, . . . Observe his merits and his passion hear!” ’ (lines 351, 354). Notice, however, that ‘passion’ – at least the suitor’s passion – has now moved to the other side of the equation. At the start of the story, Passion was linked with Romance and opposed to Reason; it is now on Reason’s side, and opposed to Romance. The complex social particulars of the example do not fit the precept it is supposed to exemplify, so the precept is made to shift surreptitiously in the course of the story. Something about the widow and her story persuades Nancy, but what it is we cannot be sure. It could simply be that the story has shown the widow’s life to have been quite unlike what it was in Nancy’s romantic anticipation. Nancy’s confidence in her own judgement and in romantic plots destroyed, she may now be more willing to accept her father’s judgement. Or perhaps she now realizes that Passion and Romance can point in opposite directions and decides that she prefers the young farmer’s passion to her own romance. Furthermore, she may now realize that it is easier for her to ‘know her place’ than it was for the widow. The widow had to choose between Passion and Reason, but perhaps Nancy is offered both in the person of Harry Carr. Perhaps it is the difference between her own situation and the widow’s which impresses Nancy rather than the similarity. The widow’s story is inconsistent. It succeeds, but we do not know why. Nancy would have had to interpret it, understand it selectively, for it to persuade her to follow the widow’s advice. In which case Nancy is – so we once again discover – rather like us readers, and the widow’s tale rather like ‘The Widow’s Tale’. What the poem now seems to suggest is that no story – whether spoken or written – can be counted on to bring about, of its own accord, a particular end. This does not simply mean that no parable carries a guarantee of success. It is rather that, after analysing Crabbe’s poem, we may feel that it is hard to be sure how far Nancy’s situation and the situation of Lucinda as a young woman are similar and how far different. The widow’s ethics depend on the language of social description she employs, a language in which individuals can be allocated to ‘classes
132
Narrative Order, 1789–1819: Part II
and degrees’. This is not to say that people in different classes have to face the same choices or obligations – far from it – but that they face comparable ones. However, in the social circumstances to which Crabbe often seems to point, this language of ‘classes and degree’ and the ethical language that depends upon it obscure the situations they are intended to clarify. They aggravate, by hiding, a situation of generalized social insecurity and unpredictable social mobility. Nancy’s parents had enough money to educate her into genteel expectations but not enough to marry her into a genteel life. Lucinda’s problem also involved social ambiguity, but in a murkier form, one less easy for her – or Crabbe – to articulate. The combination of Lucinda and the story she tells certainly persuades Nancy to change the course of her own life; but by what mental process, what juggling of similarities and differences, we are not told. And if Crabbe’s story speaks strongly to us, one thing it tells us about is the distance across which this early nineteenth-century voice has carried. The parables of Jesus were certainly designed to be effective beyond the circumstances of their original use. Nevertheless, the Gospel writers seldom present the parables without also presenting the contexts in which they were first told and their effect on the people listening. Parables are embedded within the Gospel stories as the widow’s tale is embedded in ‘The Widow’s Tale’. Furthermore, while the parables are supposed to be persuasive for readers in other times and places, they are not always presented as having instant persuasive force even as they were first told. The way in which Jesus’s listeners learn to interpret his parables – or fail to do so – is often as important as the parables themselves. If Crabbe was looking to biblical parables as models for how to write stories that ‘teach us by Example’, they are complex models, which foreground the problems of interpretation, the problem of working out how the story applies to oneself. Jesus’s parables are of two kinds.5 Most are allegories, often deliberately cryptic ones. Their listeners sometimes respond – as Jesus seems sometimes to wish them to respond – by failing to understand the encoded messages. Like the Samaritan parable, these allegorical parables – which often begin ‘the Kingdom of Heaven is like . . .’ – tell stories set in a domestic and social world familiar to their listeners. But, unlike the Samaritan story, they do so as a way of talking about an utterly different world which can only be described figuratively (for instance, the Kingdom of Heaven is not literally a Kingdom). Crabbe the clergyman sometimes used this sort of allegorical parable, but Crabbe the poet seldom did. Most of his stories-within-stories are
Crabbe’s Parables 133
certainly mirrors held up to the lives of their tellers and listeners: but if Nancy Moss is to see the widow’s life as analogous to her own the comparison is between two lives lived in the same world. If she – or we – are to ‘go and do likewise’ this ‘likeness’ (unlike the likeness of ‘the Kingdom of Heaven is like’) will be an action performed under different circumstances in the same world as the actions in the story. And that is also true for the lawyer to whom Jesus tells the Samaritan story. Another way of formulating this distinction between types of parable would be to say that some of the parables stand in a metaphoric relation, some in a metonymic relation to the world they are really about. The parable of the Good Samaritan certainly works metonymically, in a sequence of sideways movements: the dialogue between Jesus and the young lawyer is a continuous passing of the buck, as each attempts to get the other to have the last word. Each asks general questions which elicit specific answers, or specific questions which elicit general answers. Asked ‘which of these three . . . was neighbour unto him who fell among thieves?’ the lawyer does not say ‘the Samaritan’ but ‘he that shewed mercy on him’, drawing from the story the general truth which Jesus, when originally asked ‘who is my neighbour?’, refused to provide. In refusing to provide the direct answer and veering off into a story instead, Jesus was making the same sort of evasive – parabolic – movement. The lawyer’s motives for conversing in this way are political, Jesus’s political and pedagogical. The lawyer learns the precept through the example. Then – continuing the sequence of creative evasions – Jesus does not say ‘you have answered rightly’, but ‘Go, and do thou likewise’, leaving the final moral of the story for the lawyer to establish, by discovering, in practice, what ‘doing likewise’ may involve. The parable is therefore only really understood when a person discovers, in different particular circumstances, a comparable need to show mercy. At those moments the circumstances of a person’s own life and the circumstances of the parable illuminate each other. That a reader stands in a different relationship to the story from the young lawyer is therefore written into the kind of story this is and the kind of interpretation it requires. It assumes that circumstances change but asks us to see likenesses between these different circumstances; but to see them not so much when we hear or read the parable as when, later, we find ourselves in particular circumstances which call the biblical circumstances and words to mind. The appearance of words derived from this parable (‘passed him on the other side’ from the parable’s ‘passed by on the other side’) at the end of Crabbe’s poem ‘Procrastination’ is wonderfully illuminating
134
Narrative Order, 1789–1819: Part II
precisely because the manner of their appearance is appropriate both to the poem as a whole and to the way in which the original parable seems to have been designed to work. On the other hand, it remains the case, I shall suggest, that these biblical words illuminate the difference between their old context and their new one in a way that literary allusion can happily do but which must stretch the tolerance of parable. ‘Procrastination’ tells the story of a young couple – Dinah and Rupert – and of Dinah’s wealthy aunt with whom she lives. This triangular relationship recalls the one between the apprentice, Lucinda and Lucinda’s father in ‘The Widow’s Tale’. In both poems, lovers procrastinate owing to an obscure combination of financial and sexual anxiety. In ‘Procrastination’ Rupert eventually goes overseas in the hope of earning enough money to marry Dinah. A succession of disasters keeps him away for forty years, while the increasingly prim and pious Dinah, who expects one day to inherit her aunt’s money, gradually transfers her affections from the end (marriage to Rupert) to the means (her aunt’s wealth). When Rupert, now a poor and weather-beaten sailor, bursts into Dinah’s drawing room, expecting to take up where they left off years before, he is made coldly unwelcome. Rupert does stay on in his native place, but supported by the parish, not by Dinah. The poem concludes: Thence as she walks the street with stately air, As chance directs, oft meet the parted pair: When he, with thick-set coat of Badge-man’s blue, Moves near her shaded silk of changeful hue; When his thin locks of grey approach her braid, When his frank air, and his unstudied-pace, Are seen with her soft manner, air, and grace, And his plain artless look with her sharp meaning face; It might some wonder in a stranger move, How these together could have talked of love. Behold them now! See there a tradesman stands, And humbly harkens to some fresh commands; He moves to speak – she interrupts him – ‘Stay!’ Her air expresses, ‘Hark! to what I say’: – Ten paces off, poor Rupert on a seat Has taken refuge from the noon-day heat, His eyes on her intent, as if to find What were the movements of that subtle mind: How still! – how earnest is he! – it appears His thoughts are wandering through his earlier years;
Crabbe’s Parables 135
Through years of fruitless labour, to the day When all his early prospects died away; ‘Had I,’ he thinks, ‘been wealthier of the two, Would she have found me so unkind, untrue? Or knows not man when poor, what man when rich will do? Yes, yes! I feel that I had faithful proved, And should have soothed and raised her, blessed and loved.’ But Dinah moves – she had observed before, The pensive Rupert at an humble door: Some thoughts of pity raised by his distress, Some feeling touch of ancient tenderness; Religion, duty urged the Maid to speak In terms of kindness to a man so weak: But pride forbad, and to return would prove She felt the shame of his neglected love; Nor wrapped in silence could she pass, afraid Each eye should see her, and each heart upbraid; One way remained – the way the Levite took, Who without mercy could on misery look; (A way perceived by Craft, approved by Pride,) She crossed and passed him on the other side. (lines 304–49) These pairs of lines present a human ‘pair’ for whom familiarity and strangeness, similarity and difference intensify each other. These opposed qualities co-exist because Rupert and Dinah’s past and present exist together in the present. Although the couple ‘meet’ in the present and ‘parted’ in the past, these opposed movements seem painfully simultaneous.6 Such effects bear out Christopher Ricks’s description of Crabbe as a ‘moral geologist’.7 ‘Procrastination’ is about the gradually – but increasingly firmly – buried past forcing its way to the surface once again. Even before Rupert himself bursts into Dinah’s drawing room, unwelcome thoughts of him sometimes intrude into her mind. For instance, she sits at home one day with friends who are discussing How tender damsels sailed in tilted boats And laughed with wicked men in scarlet coats; And how we live in such degenerate times, That men conceal their wants, and show their crimes;
136
Narrative Order, 1789–1819: Part II
While vicious deeds are screened by fashion’s name, And what was once our pride is now our shame. Dinah was musing, as her friends discoursed, When these last words a sudden entrance forced Upon her mind, and what was once her pride And now her shame, some painful views supplied; (lines 184–93) The words uttered in the present draw to the surface of Dinah’s mind the memory of an old relationship to which the words also apply. It is by a similar process of association that the modern street draws to the surface of Crabbe’s own mind the old story about an incident on the road from Jerusalem to Jericho. A mutual attraction takes place between parable and poem. They illuminate one another by finding for each other a comparable element in a different situation. The parable seems to come into Crabbe’s mind via the literal meaning of the words he takes from it. It is a satirically appropriate allusion: the Reverend Crabbe has special authority to transmit the parable and give it contemporary relevance, while Dinah is a follower of the same religion who is caught following in the Levite’s footsteps. The allusion sharpens the parable’s cutting edge for use in circumstances not originally envisaged for it, its life in an institutionally Christian society. However, if Crabbe nicely updates the parable he also demonstrates effects of social transformation which open a chasm between the two contexts in which the phrase ‘to pass by on the other side’ appears. He does this by changing the context of the biblical phrase from ethnic to class antagonism. Of course, numerous Gospel parables describe the relationship between classes, normally for allegorical purposes. The proper relationship between kings and masters on the one hand and stewards, servants or labourers on the other is used as a way of talking about the relationship between God and man. These social relationships, however, are properly called ones of rank or degree, rather than of class, since there is no sense of them as relationships which are made rather than inherited. In the Gospels, at least in their translation into seventeenth-century English terms, a master is a master and a servant is a servant. These roles are the taken-for-granted background and necessary basis for any discussion of proper reciprocal duties between people and, therefore, of the relationship between God and man which these earthly relationships allegorize.
Crabbe’s Parables 137
To a remarkable extent Crabbe’s poems are populated by people with the same occupational titles as people in the King James version of the Gospel parables: master, steward, servant and so on. Furthermore, as Raymond Williams pointed out, although Crabbe is one of the first writers to use ‘class’ as ‘a singular social term’, he uses it in a way that is ‘virtually equivalent’ to rank or degree. However, while ‘individual mobility could be seen as movement from one . . . degree . . . or rank to another’,8 Crabbe is preoccupied with social insecurity in a way and to an extent that puts the older language under intense pressure, as it did in ‘The Widow’s Tale’. Crabbe wrote in his ‘commonplace book’ that he took his characters from those ‘placed between the humble and the great’.9 While this may refer to ‘the middling sort’ as a distinct stratum, most of his protagonists are placed very precariously between the humble and the great. When Crabbe focuses closely on a social stratum, on the individuals occupying it, it usually turns into a stratification line. His protagonists are seldom quite sure which side of the master–servant divide they are on (in this respect too, Crabbe’s tales are very like Caleb Williams). The context of known ranks in terms of which individuals are defined as deviant is implicitly questioned. In ‘The Widow’s Tale’, Lucinda’s language of ‘classes and degrees’ – and the ideas about marriage which depend upon it – aggravates ambiguities in her own life that such language is intended to clarify. When Rupert wonders whether ‘man when poor’ knows ‘what man when rich will do’, he combines a traditional sense of hierarchy with a sense that individual social mobility is normal. His self-questionings suggest that the unity of a person’s own life through time may consequently be impossible to grasp. The poem presents the social division between Rupert and Dinah as something obscurely latent in the original relationship between the couple and Dinah’s aunt, and which developed as an imperceptibly slow process of adjustment in Dinah that eventually hardened into a new identity, a second identity. To recall Godwin’s discussion of Master and Servant in Political Justice, Rupert is suggesting a world in which we cannot securely know what a person’s ‘character’ will be ‘in the sequel’. The gap between rich and poor is, for Dinah and Rupert, a gap between past and present, and between an individual and himself or herself. The social stratification line runs between Rupert and Dinah, but it also runs through each of them. In circumstances of this kind we may well wonder what acting mercifully – being ‘a Good Samaritan’ – could possibly involve. Since
138
Narrative Order, 1789–1819: Part II
Dinah passes by on the other side both literally and figuratively, it is appropriate to ask what would be actually involved in ‘binding up the wounds’? Would it simply mean speaking ‘in terms of kindness to a man so weak’, or should Dinah have ‘soothed and raised [him], blessed and loved’; should she have married him? And if she did marry him, it is a moot point whether she would be ignoring the constraints of rank, since the two were more or less engaged when they were equals but are now at least a class apart. Where likeness and difference are so obscurely entangled, it is hard to ‘go and do likewise’. These are poems which may set out to be parables but perhaps end up as what we now call literature.
8 Relations: Mary Wollstonecraft and Mary Shelley
The pun on ‘relations’ (relations as family, relations as stories) is ours rather than theirs, but it does illuminate important features of the narratives written by Mary Wollstonecraft, William Godwin and their daughter Mary.1 We have already seen that this is true of Godwin’s Things as They Are; or, Caleb Williams (1794). It is equally though differently true of Wollstonecraft’s The Wrongs of Woman; or, Maria (1797) and Mary Shelley’s Frankenstein, or the Modern Prometheus (1818). Maria and Frankenstein, like Caleb Williams, foreground storytelling – mainly autobiographical storytelling – as a central feature of human relationships and one which has, or aspires to have, illocutionary force. These are novels about ‘how to do things with stories’. They are also novels which, like Caleb Williams, explore family relationships, though their understanding of ‘family’ is closer to our own than to Godwin’s, referring to those who are linked, like these three writers themselves, by marriage or by close ties of kinship. That most of the stories in Caleb Williams and Maria are autobiographical (Caleb’s, Maria’s, Darnford’s, Jemima’s) seems to lend support to Gary Kelly’s contention that ‘autobiography is the central form of the Godwin–Wollstonecraft fictions’ and that this is so because in oneself are written the evils of ‘things as they are’, and the institutionalised ‘wrongs of woman’. The autobiographical text then becomes enlightening and thus reformative for others, as in seventeenthcentury spiritual autobiography, which is one of the sources for this kind of writing. Self-knowledge is textualized, and results in social change, if only on the level of other individual lives; but since 139
140
Narrative Order, 1789–1819: Part II
society is an agglomeration of these individual lives, that is how society will change, through the ‘spread of truth from mind to mind, self to self’.2 However, while Kelly’s argument is broadly true, it minimizes the difference between Godwin (and, for that matter, seventeenth-century spiritual autobiography) and Wollstonecraft. That is, the sharp distinction between biography and autobiography implied by Kelly would have been more acceptable to Wollstonecraft than to Godwin. Although Caleb Williams is autobiographical in form – a first-person narrative – both Caleb and Godwin are committed to that ‘narrative idea of life’ according to which – in Alasdair MacIntyre’s words – ‘I am what I may be taken by others to be in the course of living out my life from birth to death.’ They are committed, that is to say, to an essentially third-person idea of the self, the idea which was the basis of Samuel Johnson’s belief in the importance of biography but which was also, as we have seen in the case of Watkin Tench, perfectly compatible with certain forms of first-person narrative. Caleb’s autobiography attempts, though with difficulty and doubtful success, to establish his ‘character’ in a sense consistent with the narrative idea of life. Moreover, the illocutionary purposes which Kelly sees as specific to autobiography match rather closely the aims of Godwin’s 1798 Memoirs of Wollstonecraft, another text which attempts to resist a sharp distinction between biography and autobiography: ‘It has always appeared to me’, wrote Godwin in the introduction to his Memoirs, that to give the public some account of the life of a person of eminent merit deceased, is a duty incumbent on survivors . . . The justice that is thus done to the illustrious dead, converts into the fairest source of animation and encouragement to those who would follow them in the same career.3 Godwin clearly believes that, writing as an intimate survivor, he can convert the life of an ‘I’ who is not himself into that of a ‘she’ without radically transforming the kind of power the story can have. By contrast, the autobiographical storytelling in Wollstonecraft’s Maria is quite radically distinct from biography; and in this respect Maria belongs with Frankenstein which, however, takes the distinction one step further – perhaps as far as it has ever been taken – in the autobiographical story of Dr Frankenstein’s Creature.
Relations: Mary Wollstonecraft and Mary Shelley 141
Narrative is associated with power in all three novels, but in different ways, with Wollstonecraft and Shelley having more in common with each other than either does with Godwin. Godwin was fascinated by the connection between political power and the ability to persuade people that your story (the story you told about yourself) did indeed accurately represent your story (your life). This kind of interest does, it is true, put in an appearance in both Maria and in Frankenstein. In Maria, it appears in Mr Venables’ ability to tell a ‘plausible tale’ in a court of law.4 In Frankenstein, it appears in Jessica’s inability to persuade the court that she is innocent of little William’s murder, in the face of overwhelming circumstantial evidence to the contrary. These two episodes however – both very direct echoes of similar court-room scenes in Caleb Williams – are untypical of the novels in which they appear. The power to which storytelling characteristically aspires in Maria and Frankenstein is of a more specifically personal sort. It is the power to bring into being a bond of affection, solidarity or obligation between individuals who feel themselves to be otherwise excluded from such relationships. While Maria is not itself cast in the form of autobiography, it is dominated by the autobiographical stories embedded in it and by their aspiration to illocutionary force. The extent to which those aspirations are achieved is debatable. Gary Kelly argues that Maria’s own memoir ‘moves [Darnford] to love Maria and moves Jemima to help her’. It ‘converts’ Jemima, he suggests, frees her from her misanthropy, ‘revives her female feelings, and makes of her an ally and assistant’. The ‘autobiographical story of oppression’ does indeed, on this reading, ‘become a source of power, rhetorical power, releasing love’ (p. 41). Claudia Johnson is slightly more tentative: ‘To the extent that freedom is achieved in this fractured and unfinished work, it is in the cooperative and mutually respecting partnership Jemima and Maria seem on the verge of achieving.’5 Barbara Taylor describes their relationship as ‘hardly a utopian vision . . . but a prescient one: a century further on, and it was alliances like these – fragile, bias-ridden, courageous – that were to become the driving force of a mass feminist politics’.6 In any case, as both Johnson’s and Taylor’s formulations suggest, this relationship is one of the important respects in which the novel’s unfinishedness, its lack of a conclusion, seems profoundly appropriate to its substance. Frankenstein takes up and takes further Maria’s interest in the aspiration to illocutionary power of autobiographical storytelling. The form of the two novels reflects this relationship. Maria begins in medias res, so that the novel can be a mixture of events directly narrated and earlier events
142
Narrative Order, 1789–1819: Part II
communicated through the stories which the characters themselves tell. Frankenstein goes further, beginning not in the middle but near the end, at a point in the action where almost nothing except the principal characters’ own storytelling is left to happen. Both novels, then, are frame narratives, but Shelley’s far more self-consciously and systematically so. Its structure – a tale within a tale within a tale – has been elaborated by Beth Newman,7 and summarized by Peter Brooks as follows: In the outer frame, explorer Robert Walton writes to his sister Mrs Saville, and tells of meeting Frankenstein in the Arctic; in the next frame, Frankenstein recounts his life story to Walton; in the innermost tale, the monster at a crucial moment tells his tale to Frankenstein. When the monster has finished, Frankenstein resumes speaking in his own right; when he has done, Walton resumes. The nested narrative structure calls attention to the presence of a listener for each speaker . . . and to the interlocutionary relations thus established. Each act of narration in the novel implies a certain bond or contract: listen to me because . . . The structure calls attention to the motives of telling; it makes each listener – and the reader – ask: Why are you telling me this? What am I supposed to do with it? (p. 199) All the acts of storytelling in Frankenstein are designed to maintain or bring into being intimate relationships (of kin, friendship or sexual union) of a kind from which the storyteller is otherwise excluded. The Creature’s storytelling is specifically illocutionary in its intent, designed to elicit from Dr Frankenstein a promise to produce a female equivalent of himself who can be his mate: the purpose of the Creature’s relation is to acquire a relation. At first is seems it may be successful, eliciting a promise which, if kept, will lead to a kind of marriage; but the promise is broken. The link between storytelling and promising in Caleb Williams has already been discussed; Beth Newman has explored the same link in Frankenstein. It is also an important theme, arguably the central theme, in Maria, although ‘theme’ may be the wrong word since the importance of the link between storytelling and the marriage contract comes from the fact that the novel itself remained unfinished at Wollstonecraft’s death: a cruel ending that left Maria’s story, Wolsltonecraft’s story and the attitude of both women towards the bonds of marriage unresolved. In all three novels the issue of promises and contracts is part of an exploration of situations in which one person is linked to another by
Relations: Mary Wollstonecraft and Mary Shelley 143
bonds (usually, but not always, formalized by a promise) that are experienced as chains. In Caleb Williams the bond is between master and servant and combines two features which Hegel too later saw in the master–servant relationship. On the one hand it is a relationship – a ‘family’ relationship in Godwin’s terms – based on inequality. On the other hand it is a relationship of interdependence which is always potentially reversible since the master is dependent on his dependent. In effect, Maria and Frankenstein separate out these two elements – inequality and reversibility – which in Caleb Williams are combined. Maria is a story of inequality, not reversibility. The marriage between Maria and Mr Venables imprisons Maria, not Mr Venables. There is no danger in Maria of confusing the gaolers with the gaoled. By contrast, Frankenstein returns to Godwin’s preoccupation with reversibility. The bond between Creator and Creature is deeply ambivalent: it becomes difficult to say who is pursuer and who pursued, just as it does of Caleb and Falkland. Correspondingly, Shelley downplays issues of inequality. The difference between the elements of Caleb Williams which Wollstonecraft and Shelley develop – one exploring inequality, the other ambivalence or reversibility – is particularly evident in the way they address the topic which Godwin, in Caleb Williams, restricts to the margins: the relationship between men and women. While Frankenstein is less obviously or explicitly a feminist novel than Maria, it can certainly be read as one. The picture of relationships between men and women which Shelley presents is, in its way, quite as grim as the picture presented by Wollstonecraft and it is the actions and attitudes of the men which are presented as largely responsible for this. However, it is the separation rather than the inequality of men and women on which Shelley focuses, a radical polarization of roles from which the men in the end suffer as much as the women. Caleb Williams focused on the ‘family’ in the now unfamiliar sense of the household. Wollstonecraft’s novel focuses on the family in a more modern sense of the word: the unit composed of husband and wife, parents and children. She reveals marriage, supposedly a sustaining bond, as a contract which effectively makes women the property of their husbands till death do them apart. The novel does not focus exclusively on the family however. Rather, it places the ‘domestic tyranny’ of men at the centre of a wider society in which the same rule applies as applies within families: some people are in effect the property of others. Mary Shelley shares her parents’ interest in looking at the links between intimate personal relationships and the wider society in which they are set. Indeed, Frankenstein seems to be assembling materials for
144
Narrative Order, 1789–1819: Part II
an evaluative taxonomy of personal relationships in the context of different kinds of polity (aristocratic, republican). These consist of family relationships (those between spouses, siblings, parents and children), sexual relationships and those relationships of ‘friendship’ with other men, outside the familial or sexual bond, which the principal male characters in the novel are seeking. However, the limits of every kind of personal relationship and political constitution are exposed by their inability to incorporate the unclassifiable being created by Frankenstein. This is a novel in which the possibility of having a relationship with someone is presented as being fundamentally dependent on knowing how to classify them and your relationship to them (as ‘brother’, ‘sister’, ‘friend’, ‘husband’, ‘wife’ and so on). The Creature challenges all personal and social relationships by challenging available classifications. The Creature is sufficiently human to require social classification, but seems too different to be classifiable. We have seen from our reading of Wordsworth and Crabbe that the power of storytelling depends, for them, on the perception of likeness between different people and on the possibility of co-ordinating the similarity and the difference. Wordsworth attempted to distance himself from likeness and from storytelling – from ‘imitation’ in life and art – while acknowledging their seductive power. Crabbe, by contrast, like Samuel Johnson, attempted to write the kind of exemplary story that depends upon an acceptance that we all are more like other people than we tend to think; but his stories nevertheless reveal to us a kind of class society in which it is actually very difficult to identify oneself securely as a member of a class. In Frankenstein one living being is felt to be radically different from all others, defeating efforts at classification. It is a problem, at least in the first place, of naming, and it applies to proper names (such as ‘Frankenstein’) as well as class names (such as ‘sister’, ‘friend’, ‘monster’). It applies most obviously to the living being created by Frankenstein, but also to Frankenstein himself and to the relationship between the two of them. As Chris Baldick says, ‘there is some uncertainty whether best to define the being created by Victor Frankenstein as “monster”, “wretch”, “daemon”, “creature” or “fiend” ’,8 all terms which are applied to him in the novel. Peter Brooks, in the summary of the novel which I have quoted, calls him ‘the Monster’ but, as Marilyn Butler argues, there are ‘good reasons for the critic to avoid Frankenstein’s harshest, most unpaternal acts of naming’.9 The problem remains, however, that there is no term within the novel which catches the Creature’s essentially hybrid identity so well as the prejudicial word ‘monster’ does (no word as ethically neutral as ‘hybrid’, for instance).
Relations: Mary Wollstonecraft and Mary Shelley 145
‘Creature’, the term which many of us now use, has the advantage and limitation of an essentially provisional name. Having no secure class name, not being unambiguously human and having no parents, the Creature has no species name, family name or given name. Within the novel there is no word which captures that condition as well as ‘Monster’, which however also damns him (damns him partly for being unclassifiable, unnamable). The uncertainty about the name ‘Frankenstein’ is of a rather different kind. Nevertheless, the common mistake of calling the Creature by the name of his Creator is a significant mistake, produced by a combination of factors, starting with the book’s title, Frankenstein; or, the Modern Prometheus. As we start to read through the book we are necessarily waiting for the appearance of the person or (following the signal provided by ‘Prometheus’) the godlike figure to whom we can attach the name offered by the book’s title. The first – and very likely – candidate is the ‘being which had the shape of a man, but apparently of gigantic stature’ (p. 12) seen crossing the polar icepack by Captain Walton and described in the latter’s fourth letter home to his sister, Mrs Saville. While we probably change our minds about who the title-name refers to soon after the real Dr Frankenstein starts to tell his story, some uncertainty remains since Frankenstein does not actually name himself till well on in his story. Furthermore, if the relationship between the Creator and his Creature is indeed (as most readers agree) a quasi-paternal one, then if the Creature did have a family name it would indeed be Frankenstein. The uncertain relationship between Frankenstein, his Creature and their names is very carefully planned. It is partly because we cannot name the Creature, say who or what he is, that we find ourselves continually reaching for things he is ‘like’. This process goes on in a number of different, though overlapping, areas: in literary analysis of the novel itself, in attempts to identify its real-life and literary sources and, most notably, in the continual activity, in British society, of finding new ‘Frankenstein’s monsters’ (‘Frankenstein foods’ and so on). We cannot pin down the Creature or the nature of his relationship with his maker; and our inability to say what they are is matched, in everyday life and in literary criticism, by an apparently inexhaustible ability to say what they are like. They are like parent and child, God and man, colonizer and colonized; like the bourgeoisie society which produces, in the proletariat, its own gravedigger. However, our inability to name the Creature or his relationship to Dr Frankenstein is only part of the explanation for these proliferating
146
Narrative Order, 1789–1819: Part II
attributions of likeness. What is going on here is not only a series of attempts to make sense of Frankenstein and his Creature but a series of attempts to make sense of the things to which Frankenstein and his Creature are being likened. What all such things have in common is that they are constituted by processes we would now call alienation. That is, they are constituted by the giving of life to a being separate from its Creator, a being which then – in what can be seen as a distinct stage of the process – turns against the source of its life. This type of situation is a challenge to naming precisely because we cannot be sure whether what we are naming is one entity, two entities or the changing relationship between them. We are dealing here, after all, with something that can become in part other than itself and even be opposed to itself. When we proclaim that Dr Frankenstein’s Creature is like something else there is, consequently, a process of mutual attraction, mutual need, taking place. One thing that is difficult to name becomes the surprisingly apt name for other instances of things which, for similar reasons, challenge naming. What is happening in such cases is something like what we see happening when we find ourselves recalling the biblical parable of the Good Samaritan: illumination takes place simultaneously in both directions, our own puzzling situation is illuminated by the parable to precisely the extent that our own situation makes sense of the puzzling parable. However, this process by which each discovers a name in the other provides an illumination which is always precarious, since we are always liable to remember that likeness is not sameness. Frankenstein’s Creature is not a child, not the proletariat, not Caliban, not a genetically modified crop. The analogies always end up highlighting the differences, the irreducible unnamability of this entity who can only ever be securely named in inverted commas, as ‘creature’, ‘monster’, ‘fiend’. The search for and discovery of likeness and the continual discovery of its limits begins in the novel itself. It is Frankenstein himself who sees his relationship to his Creature as paternal and as godlike but not the relationship of a father or a god: ‘a new species would bless me as its creator and source . . . No father could claim the gratitude of his child so completely as I should deserve theirs’ (p. 36). As for the Creature, he attempts to make himself at home in the world by finding a being like himself. He searches both amongst the people he encounters and among the people he reads about in the four texts that constitute his formal self-education: Milton’s Paradise Lost, Goethe’s Werther, ‘a volume of Plutarch’s Lives’ and Volney’s Ruins of Empire. Paradise Lost is his best source in this respect, but even though he decides that he is
Relations: Mary Wollstonecraft and Mary Shelley 147
‘like Adam’ in his isolation, ‘[Adam’s] state was far different from mine in every other respect’. In some respects Satan was ‘the fitter emblem of my condition; for often, like him, when I viewed the bliss of my protectors, the bitter gall of envy rose within me’ (p. 105). The Creature summarizes his predicament as follows: As I read...I applied much personally to my own feelings and condition. I found myself similar, yet at the same time strangely unlike the beings concerning whom I read, and to whose conversation I was a listener. (p. 103) His predicament is like our predicament as we ponder him (and yet how different!). Any attempt to offer analogies for Frankenstein – including the ones I am going to offer here – must always be aware that whenever we find an illuminating similarity it will always draw our attention to ‘something strangely unlike’. Mary Shelley brings uncompromisingly out into the open the kind of challenge to classification with which Crabbe wrestled in ‘The Widow’s Tale’. In Frankenstein, the other characters allow the Creature’s difference from themselves completely to overshadow the similarities so that the similarities become nothing more than an additional source of fear and loathing. As Chris Baldick and others have pointed out, some of the most compelling analogies, which are also likely sources, come from Mary Shelley’s parents or their generation, from the revolutionary 1790s. For instance, Baldick quotes the conservative Abbé Barruel speaking of ‘that disastrous monster Jacobin . . . engendered’ by a conspiracy of enlightenment intellectuals originating in Ingolstadt (p. 19). On the enlightenment intellectuals’ side of the argument, the same Jacobin monster was seen as the unwanted progeny, not of people like themselves but of that ancien regime represented by such people as the Abbé Barruel. The Frankenstein-logic could also be seen at work in English society, if not in the formation of a proletariat, then certainly in the individual lifehistory of a working-class woman, Jemima in Wollstonecraft’s Maria. ‘I cannot help attributing the greater part of my misery’, says Jemima to the misfortune of having been thrown into the world without the grand support of life – a mother’s affections. I had no one to love me; or to make me respected, to enable me to acquire respect. I was an egg dropped on the sand; a pauper by nature, hunted from family to family, who belonged to nobody – and nobody cared for me. (p. 106)
148
Narrative Order, 1789–1819: Part II
Jemima confesses that, ‘shut out from society’ (p. 114), she persuaded a man to throw his current mistress out into the street so that she could take her place (the woman subsequently died from exposure). She explains that ‘I was famished: wonder not that I became a wolf . . . a monster’ (p. 116). ‘Who ever’, she asks, ‘risked anything for me? – Who ever acknowledged me to be a fellow-creature?’ (p. 119). Key elements of the Creature’s story are here: the lack of a human parent, particularly a mother, the absence of a childhood ( Jemima was a ‘pauper by nature’), the experience of being hunted from family to family. Born in isolation, she was treated as a monster and therefore became one. The principle difference between Jemima’s situation and the Creature’s is in this case simple but radical: what is metaphorically true of her becomes something like literally true of him. You can turn Jemima’s story into science fiction by having her actually born out of ‘an egg dropped on the sand’, and become literally a monster, part human part monotreme. It is by thinking about the relationship between the literal and metaphorical that Edmund Burke’s contribution to Frankenstein can be seen in its proper light. Baldick cites a passage from the Reflections as a close source or analogy: [We] should approach to the faults of the state as to the wounds of a father, with pious awe and trembling solicitude. By this wise prejudice we are taught to look with horror on those children of their country who are prompt rashly to hack that aged parent in pieces, and put him into the kettle of magicians, in hopes that by their poisonous weeds, and wild incantations, they may regenerate the paternal constitution, and renovate their father’s life. (p. 194) I have myself already cited this passage, which compares the French revolution to a perverse relationship between parents and children, in support of my claim that Burke’s challenge to a narrative idea of life originated in his recoil from the bringing together of death and birth, ends and beginnings. The passage is, as we shall see, equally but differently significant for the attitude to narrative displayed in the novels of Mary Wollstonecraft and Mary Shelley. But what we need to notice in the first instance is that what is metaphor in the source – ‘a monstrous dismemberment, a castratory hacking which precedes the reassembly of limbs into an abortive body politic’ as Baldick puts it (p. 17) – becomes literal in Frankenstein. However, the other main
Relations: Mary Wollstonecraft and Mary Shelley 149
feature of the passage – the perverse relationship between parents and children – and the way in which that too moves from the metaphoric to the literal was of greater importance, not only for Frankenstein but for Maria too, and then for the relationship between the two novels.10 The revolutionary logic according to which the old must die in order for the new to be born is imaged by Burke as a perverse relationship between generations within the family. What we saw in Jemima’s story was something similar. Her sense of having been from the very start ‘shut out from society’ (not of being thrown out of society but of being ‘thrown into’ society from the outside) was rooted in her radical separation from her mother. As for Maria, she writes her ‘memoir’ for the infant daughter she believes she will never see again. And then – in a repetition that surely is uncanny – that separation within the novel was repeated in the relationship between the novel’s author and her own daughter, Mary Wollstonecraft Godwin, the future Mary Shelley. It is hard to believe that Mary Shelley, reading her mother’s novel, was not struck by the way in which its content and its unfinished form mirrored and helped to constitute her own relationship to its author. The bringing together of ending and beginning within the family, so that there is no overlap between one generation and the next, no shared life, was a fundamental metaphor in the Reflections, one that helps to explain the emotional power behind Burke’s recoil from a narrative idea of life. Become literal, this relationship – or lack of relationship – between generations was equally fundamental to the way in which both Wollstonecraft and Mary Shelley conceived of the part played by storytelling in the attempt to establish relationships. The more or less completed portion of Wollstonecraft’s novel ends where Maria’s memoir, addressed to the daughter she does not expect to see again, itself ends: ‘Such, my child, are the events of thy mother’s life to this dreadful moment – Should she ever escape from the fangs of her enemies, she will add the secrets of her prison-house – and – ’ Some lines were here crossed out, and the memoirs broke off abruptly with the names of Jemima and Darnford. (p. 185) It is hard to believe that Mary Shelley, reading this, did not see Maria as her own mother and the child as herself. Gary Kelly has compared
150
Narrative Order, 1789–1819: Part II
Maria to the Girondist, Madame Roland, as a feminist autobiographer who addressed the text of her life to her absent daughter, as a self-vindication, a legacy, and a warning. Sons may inherit estates, daughters, unjustly deprived of estates by customs of male primogeniture, inherit texts designed to raise a feminist self- and social consciousness. (p. 39) It is not only that what is true of Madame Roland and Maria is true of Wollstonecraft but that the life-story is not only a girl’s alternative to an estate, it is also this girl’s alternative to the mother’s actual life. In other words: if parent and child cannot co-exist, neither can life and lifestory. Child and life-story can only exist when the mother’s life ends. A life-story is the alternative to a life. This situation of compound mutual exclusion is only manifested completely in the tragic relationship between Mary Wollstonecraft, her narrative and her daughter. It is nevertheless the limit case which brings into focus important features of all the principle relationships and acts of storytelling in both Maria and Frankenstein. The autobiographical life-stories in these novels are spoken or written by individuals who are emotionally and often physically isolated and who are now attempting to relate to others by telling their stories. The characters start out – at the start of the narrative or the start of their lives or both – in states of more or less crippling isolation, an isolation which may be imposed ( Jemima, Maria, the Creature) or chosen (Walton, Frankenstein). Telling their life-stories is either an alternative to sharing their lives or an attempt to sustain or bring into being a shared life. They attempt to relate to others by relating their life-stories. It is not surprising that the absence of a relationship between mother and offspring should play a central part in novels about people who start out excluded from society, since the interdependence of individuals and of generations is particularly visible and incontestable in maternity. The interdependence of generations – which Burke was so keen to assert and Paine so keen to reject – may be contrasted with the idea of human origins represented either by Adam in Genesis and Paradise Lost or by seventeenth- and eighteenth-century theories of social contract (theories which themselves took the form of stories). The original individual male envisaged by the Bible and by contract stories (he is present, for instance, in Volney’s Ruins of Empire) must then enter into relationships in order to gain certain benefits. And indeed, the two most remarkable
Relations: Mary Wollstonecraft and Mary Shelley 151
autobiographical life stories in Maria and Frankenstein – Maria’s and the Creature’s – bring to the fore in remarkably explicit, though radically different, ways the connection between narrative and contract for which I have been arguing in this book. Marilyn Butler, in her discussion of Frankenstein, has drawn attention to a passage in A Treatise on Morals (?1815) in which Percy Shelley suggested that if it were possible that a person should give a faithful history of his being from the earliest epochs of his recollection, a picture would be presented such as the world has never contemplated before . . . If it were possible to be where we have been, vitally and indeed . . . if the passage from sensation to reflection – from a state of passive perception to voluntary contemplation were not so dizzying and tumultuous, this attempt would be less difficult.11 Mary Shelley has created, in Frankenstein’s Creature, a being who starts life as an adult, with no childhood and no parents and whose ability to tell his own life-story from scratch seems to reflect those circumstances. And while these facts do distinguish the Creature from ordinary human beings, they link him both to influential images of the human in Genesis, Paradise Lost and contract theory, and to the lived experience of someone like Jemima in Maria. It is as if Frankenstein were written in answer to a question: ‘What would a being have to be like that matched both the isolated individual imagined by contract theorists and the consciousness of an actual individual such as Jemima?’. If the Creature is unlike us in having no childhood, he is nevertheless like us in not being born with language. He needs, like a human child, to acquire language. It is of course frequently the case in reading autobiographies that we are aware of a tension between the consciousness of the person telling the story and of the younger person whose story is being told (the ‘I’ of an autobiography is less than a first person plural but more than a first person singular). This tension is the greater the further back in time the autobiographer attempts to go, the further back we attempt (in Percy Shelley’s formulation) ‘to be where we have been’. It is especially intense if we attempt to speak about the feelings we had and observations we made before we could speak. The special character of the Creature – his lack of a childhood or of childhood amnesia together with his initial lack, like a child, of human language, and his need to learn it – make this normal autobiographical tension especially intense. We are very conscious, that is, that a speaking and reading
152
Narrative Order, 1789–1819: Part II
Creature is telling the story of himself as a Creature who for a long time could not use or understand language. Indeed, the disconcerting relationship between the Creature as a being who is spoken about and as a being who speaks must strike all readers sharply the moment he does start to speak. Mary Shelley in fact leads us carefully through the frames of her novel towards this startling moment. Within the first frame – Walton’s letters to his sister – Dr Frankenstein appears first of all in the third person, as a ‘he’. When he starts to tell his story to Walton, he becomes an ‘I’, and starts to construct the novel’s second frame. Within that frame we are first of all disconcerted by the moment at which the assembly of dead limbs – an ‘it’ – turns into a living male being, a ‘he’. But we are almost equally disconcerted when this ‘he’ – male but not definitively human – appears as an ‘I’, a Creature with that kind of subjectivity and that sense of participation in a shared world specifically associated with the use of a human language, especially its deictic elements. To say that the Creature is both a third person (‘he’) and an ‘I’, that he exists both inside and outside language, is of course only to say that he is like the rest of us. But Mary Shelley has created a situation in which this normal situation is experienced as unnervingly strange. She has created a kind of autobiography which heightens normal features of the genre in a particular way. The first person and third person pronouns do here come to stand for radically incompatible kinds of experience and knowledge. The kind of radical opposition between biography and autobiography which Gary Kelly appeared to take for granted in his association of the Godwinian novel with the latter really does exist in Frankenstein. Our sense of the Creature as an ‘I’ and our sense of him as a ‘he’ can scarcely be reconciled and we are pulled towards both. Dr Frankenstein too is pulled towards both. The Creature tells his story in order to persuade Frankenstein to create another being like himself, a female Creature to be his mate. The storytelling wins Frankenstein over (‘for the first time I compassionated him’), eliciting from him a promise to do what the Creature has asked. But the empathy induced by the autobiographical storytelling is not strong enough and the Creature becomes a ‘he’ again. The link between storytelling and promising in the texts of the age of Revolution tend, as we have seen, to take one of two forms, though both are frequently present to some degree. Stories may aspire to an illocutionary force that brings about a contract, as Othello’s story leads directly to his marriage to Desdemona and Crabbe’s widow’s tale moves Nancy Moss to marry Harry Carr. The Creature’s story is an example of this: a story that aspires to bring about a promise which will itself lead
Relations: Mary Wollstonecraft and Mary Shelley 153
to a kind of marriage. On the other hand, promises and contracts (the kind that actually inaugurate those marriages) attempt to structure lives as narratives. Wollstonecraft’s Maria links narrative and contract in both ways. Indeed, the two ways in which narrative and contract may be linked are hardly distinguishable in this case. The exchange of confidences, in the form of confessional life-stories, between Maria and Darnford may create a real and positive bond between them; but whether it does so is a question which perpetuates the main problem of Maria’s whole life: the problem of whether any kind of ‘marriage’ – a relationship with a guarantee of continuance – with a man can be entered into. This is an unfinished novel, and the question is not resolved, either for Wollstonecraft or for Maria. Consequently the novel foregrounds the structural similarities between narrative and contract. Maria tells a story of marriage as an unequal contract, a bond which is for women always a form of bondage: When I recollected that I was bound to live with such a being for ever – my heart died within me – . . . Marriage had bastilled me for life . . . This was the man I was bound to respect and esteem: as if respect and esteem depended on an arbitrary will of our own. (pp. 155, 158) One meaning of ‘bound to’ is ‘necessarily will’; but Wollstonecraft brings this meaning into close contact with ‘bound to’ in the senses of contracted to and forced to. She brings together the terminal points of a semantic chain, one end of which is necessity and the other end of which is a promise. Not only ‘bound’, but the associated words ‘bond’ and ‘bondage’ are constantly repeated in the novel. ‘Bond’ can of course be a very positive as well as a very negative term and that doubleness is important here too. Without bonds between people, the novel suggests, life is unbearable and yet such bonds – particularly those between men and women – repeatedly turn out to be bonds in the negative sense, bondage. It is that bind – that lived contradiction – which faces Maria with the possibility of suicide as the only possible course of action (as it had faced Wollstonecraft). What Maria and Wollstonecraft are doing in the novel is exploring the possible ways in which Maria might establish bonds which are not forms of bondage, partly by rejecting a contract which officially she cannot reject. Since this is a novel set in an asylum where Maria’s husband, Mr Venables, has incarcerated her, the material sense of these words – ‘bonds’, ‘bound’ and ‘bondage’, but also ‘ties’ and the more exclusively
154
Narrative Order, 1789–1819: Part II
negative ‘fetters’ – is never far from our minds. The connection is made explicit in that description of herself as ‘bastilled’ by marriage, an old metaphor given new life by the events of 1789. Paine had famously discussed the power of aristocratic titles in these terms: believing that people really are ‘Dukes’ and ‘Kings’, we are, he argued, ‘imprison’d within the bastille of a word’. In Wollstonecraft’s case the imprisoning words are those which are uttered and written when the marriage-bond is signed and we accede to ‘laws . . . which forge adamantine fetters to bind minds together’ (p. 165); and which eventually, in her case, lead to literal incarceration. We are made pointedly aware of the interaction between mind-forg’d manacles and the sort forged by blacksmiths. Indeed, we are very close to Blake’s world here. It is the world of ‘London’ (1794) with its ‘marriage hearse’ and its evocation of the performative utterances and documents – charters, ‘banns’ of marriage, curses – through which people imprison and are imprisoned.12 It is also close to Visions of the Daughters of Albion, where ‘she who burns with youth and knows no fixed lot, is bound / In spells of law to one she loathes’.13 In attempting to free herself from her marriage to Mr Venables and at the same time to forge a different kind of bond with Darnford, Maria adopts various strategies, all of which are problematic. In the first place she attempts to effect her own divorce in an improvised ritual act in front of Mr Venables’ attorney, ‘Mr S’: Then, turning to Mr S———, I added, ‘I call on you, Sir, to witness,’ and I lifted my hands and eyes to heaven, ‘that, as solemnly as I took his name, I now abjure it,’ pulled off my ring, and put it on the table; ‘and that I mean immediately to quit his house, never to enter it more. I will provide for myself and child. I leave him as free as I am determined to be myself – he shall be answerable for no debts of mine.’ (p. 162) From then on Maria does not consider herself to be really Mr Venables’ wife, though of course the law does consider her to be so. It is a moot point whether she has really freed herself from the bastille of the word ‘husband’. An attorney ‘advised me to make over to my husband – for husband he would term him – my property’ (p. 182). While she asserts the possibility of undoing a legal tie by means of an improvised performative ritual of her own she nevertheless also knows she must persuade a court of the justice of her case too. And she asserts the possibility of
Relations: Mary Wollstonecraft and Mary Shelley 155
doing that, even though the divorce laws are against her, by her appeal to the jury: I claim then a divorce . . . I appeal to the justice and humanity of the jury – a body of men, whose private judgement must be allowed to modify laws, that must be unjust, because definite rules can never apply to indefinite circumstances. (p. 198) She wishes to believe that within the law, which relies on the likeness between different people and different circumstances, there is an element which works on an opposite principle, the principle of the incomparability, uniqueness, of individual circumstances. Her position is antinomian, Blakian: ‘One Law for The Lion & Ox is Oppression’. However, the force of that position remains unclear. Chapter 17 concludes with the judge’s hostile summing up of Maria’s case. Then the ‘Conclusion, by the editor’ (that is to say, by Godwin) gives us, in note form, alternative continuations which appear to suggest indecision, on Wollstonecraft’s part, about the jury’s verdict. In effect, this is an indecision about the extent to which an individual, through force of personal conviction and rhetorical power can match and overcome the powerful performatives of Church and State. Can an antinomian position be accommodated, via the jury system, into the Law itself? By this point in the novel, the informal divorce from Mr Venables effected by Maria has been followed by an informal marriage to Darnford. In a pattern we are now familiar with, the crucial event which causes Darnford to fall for Maria is the reading, in his cell, of Maria’s memoir. They then meet in her cell and proceed to a kind of marriage: The lovers were, at first embarrassed . . . Darnford represented ‘that they might soon be parted’, and wished her ‘to put it out of the power of fate to separate them’. As her husband she now received him, and she solemnly pledged himself as her protector – and eternal friend. (p. 188) What is described is at once an act of sexual bonding and a species of unofficial marriage: ‘she called him by the sacred name of husband, and bade him hasten to her’ (p. 190). However, it is difficult for Maria and Wollstonecraft to be sure what the ‘marriage’ to Darnford – like the unofficial divorce from
156
Narrative Order, 1789–1819: Part II
Venables – really means. Since ‘the odium of society impedes usefulness’, Maria wishes to avow her affection to Darnford by becoming his wife according to the established rules . . . though her conduct would be just the same without the ceremony as with it, and her expectations from him not the less firm. (p. 194) Yet there are problems with this formulation by Maria of her own situation. For one thing, the unofficial act of bonding does have an element of ceremony about it, in a formal exchange of vows. Nor is it at all clear what vows are really exchanged, or how far Maria sees this event as a pledge of future loyalty (an informal contract) or, by contrast, as an act of love that is self-sufficient. Wollstonecraft’s description of the episode is radically inconclusive in all these respects, an inconclusiveness underlined by the dashes which end the passage and which end, with it, this unfinished draft chapter. There was one peculiarity in Maria’s mind: she was more anxious not to deceive, than to guard against deception; and had rather trust without sufficient reason, than be for ever the prey of doubt. Besides, what are we, when the mind has, from reflection, a certain kind of elevation, which exalts the contemplation above the little concerns of prudence! We see what we wish, and make a world of our own – and though reality may sometimes open a door to misery, yet the moments of happiness procured by the imagination, may, without a paradox, be reckoned among the solid comforts of life. Maria now, imagining that she had found a being of celestial mould – was happy, – nor was she deceived. He was then plastic in her impassioned hand – and reflected all the sentiments which animated and warmed her. (pp. 188–9) Whatever we think of the claim made here about the solidity of ‘the moments of happiness procured by the imagination’ it surely cannot be made ‘without a paradox’? Maria has in the past suffered from projecting her ideals onto men – notably Mr Venables – who turned out to be unworthy of them. What is claimed here is both that she is prepared to risk this happening again, and that it was not really a risk because at this moment, in this act of bonding, she did in fact turn him into what she imagined him to be. Nor is it possible to know whether, when she
Relations: Mary Wollstonecraft and Mary Shelley 157
says ‘nor was [Maria] deceived’, Wollstonecraft means that Darnford was in fact to prove trustworthy, or that, in the heat of this passionate moment there was an intimacy between them which was real while the moment lasted. If the latter is the case – if she is arguing for a kind of free love in the manner of Blake’s Oothoon – then we do have in this episode an exchange of vows which are indeed fully performative (in a way that a promise is not) in the sense that they do not depend on future conduct to prove themselves genuine. The main reason why we cannot clarify the meaning of this passage, or decide what weight should be given to Maria’s belief that her words can persuade a jury to grant a divorce, is that Wollstonecraft died before she was able to decide what would happen next. The novel has no conclusion: neither an ending, nor a solution in view to the important questions it raises. Indeed, uncertainties about how the novel might have proceeded and concluded are echoed, within the apparently completed chapters we do have, by surviving evidence of different compositional strata. These strata involve different decisions about, for instance, the character and role of Darnford. When he first appears in Maria’s cell, he exclaims: ‘This is extraordinary! – again to meet you, and in such circumstances!’ Still, impressive as was the coincidence of events which brought them once more together, their full hearts did not overflow. (p. 93) This passage presents us with a puzzle which is, in one sense, no more than the planned puzzle of a mystery story: we will read on to discover what actually happened before the moment in the action at which the novel began. However, this planned uncertainty is complicated and intensified by a real one caused by the fact – which the unfinished state of the novel would alert us to even if Godwin’s note did not – that Wollstonecraft herself may not yet have decided how Maria and Darnford will eventually turn out to have previously met. If a narrative is unfinished it is not completely begun. Godwin’s footnote to this passage tells us that ‘the copy which had received the author’s last corrections, breaks off in this place, and the pages which follow, to the end of Chap. 4, are printed from a copy in a less finished state’ (p. 93). These less finished pages consist of Darnford’s life-story as he tells it to Maria, a story which makes no reference at all to any previous meeting. That possible earlier meeting may be described much later in the novel, in the course of Maria’s own memoir where, in a passage twice broken up by asterisks, an unnamed
158
Narrative Order, 1789–1819: Part II
‘gentleman’ appears and protects Maria from her husband. At this point Godwin comments, in a footnote: The introduction of Darnford as the deliverer of Maria, in an early stage of the history, is already stated (Chap 3) to have been an afterthought of the author. This has probably caused the imperfectness of the manuscript in the above passage; though, at the same time, it must be acknowledged to be somewhat uncertain, whether Darnford is the stranger intended in this place. It appears from chapter 17, that an interference of a more decisive nature was designed to be attributed to him. (p. 175) We are here in a world where the unintentionally unfinished narrative meets the deliberately fractured narrative of A Tale of Tub or Tristram Shandy. The final references to Darnford appear in the section called ‘Conclusion, by the Editor’, to which reference has already been made, and which offers Wollstonecraft’s brief, alternative, ways in which the story might have been continued. Some of these suggest that Darnford has been or later becomes unfaithful, others do not. The surviving final section, entitled ‘The End’, describes the pregnant Maria attempting suicide to escape from ‘this hell of disappointment’ (p. 202) and her rescue through the solidarity of Jemima. What emerges from these summaries – summaries of Maria’s uncertainties about the bonds that can be made with men, summaries of Wollstonecraft’s uncertainties about the direction in which her novel should go – is how closely connected the two forms of uncertainty are. Stories and promises both structure lives in a way that makes the beginning as dependent on the end as the end obviously is on the beginning. The fact that the link between the uncertain outcome of a promise and of a story is so palpable in this novel could be explained by its strongly autobiographical content, the resemblance between Maria’s life and Mary Wollstonecraft’s life. However, we need to remember that the resemblance is not only between Mary as a woman who struggled in her life with the problems Maria struggles with, but also between Mary as a novelist and Maria as author of her unfinished memoir. If one of the ways in which people attempt or are obliged to give their lives narrative form is through making commitments of a formalized kind, then it is especially appropriate that a radical uncertainty about the desirability of entering into such engagements should take the form of an unfinished story.
9 The Still Unravished Bride of Lammermoor
The actual murder of the king and queen, and their child, was wanting to the other auspicious circumstances of this “beautiful day”. The actual murder of the bishops, though called for by so many holy ejaculations, was also wanting. A group of regicide and sacrilegious slaughter was indeed sketched, but it was only sketched. It unhappily was left unfinished, in this great history-piece of the massacre of the innocents. What hardy pencil of a great master, from the school of the rights of men, will finish it, is to be seen hereafter. (Edmund Burke, Reflections on the Revolution in France, p. 166) The historical novel as it was developed by Walter Scott – the historicist historical novel in which the past is seen as ‘the prehistory of the present’1 – was, together with the short lyric, the most influential literary form to emerge from the dissolution of the narrative idea of life. Narratives which feel like instalments, which are merely parts of the continuing history in which they are written, are one side of a coin whose other side is the short lyric focused on the passing present. Two texts from post-Waterloo Britain, both written in 1819, make us especially aware of the complementarity of the two genres and demonstrate once again how closely the ordering of experience as a narrative is linked to its ordering by contracts and promises. The fact that these two texts – Scott’s Bride of Lammermoor and Keats’s ‘Ode on a Grecian urn’ – were probably published too close together in time (the novel in 1819, the poem in 1820) for either to have influenced the other is an additional indication that we are dealing here with widely shared preoccupations. The Bride of Lammermoor, like many of Scott’s novels, is narrated by Peter Pattiesson, village schoolmaster and antiquarian, collector and 159
160
Narrative Order, 1789–1819: Part II
reteller of stories from the seventeenth- and eighteenth-century Scottish past. The novels are, that is to say, frame narratives which draw our attention to the memories and documents from which these stories of the past are derived. The Bride of Lammermoor is ‘perhaps the most enigmatic of [Pattiesson’s] many frame narratives’.2 Stories which contain stories are often stories about storytelling, and even more than Scott’s other novels, The Bride of Lammermoor is ‘a reflection on the problems of telling stories about the past’.3 However, this is no self-contained metafictional exercise; the novel also encourages us to think about the connection between storytelling and two other practices that attempt to order life in a narrative way. One of these is the practice of making formal commitments or contracts. The other is the practice of narrative painting, especially in what was still its most prestigious form, and the form most directly relevant to Scott’s own project: history painting. While The Bride of Lammermoor’s focus on a disputed marriage contract scarcely distinguishes this novel from numerous others, the character of its interest does. For one thing, the contested marriage contract of the eponymous bride, Lucy Ashton, is linked, both through the plot and through the resonant vocabulary of marital ‘union’, with the Act of Union which, in 1707, had abolished the Scottish parliament and linked Scotland to England and Wales as the United Kingdom of Great Britain.4 Secondly, the novel is, as Fiona Robertson has noted, ‘based on an interlinked series of betrothals and burials’ (viii), pointing to the fact that Scott is interested not only in the making and breaking of contracts and promises but in the passage rites in which the signing of contracts and making of promises play a central part. The title of this novel announces the crucial importance of these rites: a ‘bride’ has no existence except as a protagonist in the marriage rite. Rites of passage are acts – in part, speech-acts – which can be understood as attempts to give life a narrative shape both because they frequently mark beginnings, middles and endings (such as baptisms, weddings and funerals) and because each rite is identified as having a similar tripartite structure, of separation, transition and incorporation.5 The connection between these rites and narrative is foregrounded by The Bride of Lammermoor, but so, remarkably, is the failure of the rites and their consequent failure to give lives a narrative shape. T.M. Kelley, writing about Keats’s ‘Ode on a Grecian Urn’, noted that ‘as a “still, unravished bride”, the urn is threatened by the sexual violence this figure keeps just at bay: it too can be violated, broken’.6 In Scott’s novel, the drive to get to the end of the story, where we will finally discover – or so we
The Still Unravished Bride of Lammermoor
161
anticipate – what has really happened, ends with the sexual violence which Keats’s poem holds at bay. In Scott’s novel, the urn is broken: the start of the bride’s new life turns out to be its end as well, a convergence of beginning and ending which prevents her story from being told. Scott’s most Burkian novel, The Bride of Lammermoor, shows us more of the violence associated with beginning and ending than Burke was prepared to look at, though it does so in a less collective context. However, the question of whether narrative is possible is broached first – and foremost – through the discussion of narrative painting. In the novel’s first chapter, the narrator, Peter Pattiesson, is arguing with his painter friend, Dick Tinto, about Tinto’s sketch for a projected history painting. Tinto suggests that Pattiesson use the events represented in the sketch as the basis for his next story. Pattieson responds (in terms we shall examine in detail) that no story can be derived from the sketch. Tinto is an interesting figure. Son of a tailor (though claiming gentle ancestry), he is what might now be called an ‘artisan painter’.7 That is to say, he is a painter without academic training who lives by painting such things as inn-signs and stage-scenery as well as portraits of local people. However, Tinto has aspirations to practise ‘fine art’ and to become (like his supposed ancestors) a gentleman. The plan to produce a ‘history piece’ is a part of this artistic-social aspiration. We may recall, reading the argument between Pattiesson and Tinto about the narrative legibility of Tinto’s sketch, our analysis of the relationship between Watkin Tench’s journals and his completed narrative. We may also recall that it was a theory of visual art – the theory propounded by Reynolds in his Discourses – which helped us to make sense of Tench’s position. It is Reynolds – ‘Sir Joshua’ (p. 7), as Tinto calls him with false familiarity – whom Tinto claims as authority for his ideas about art, especially Reynolds’s argument for the pre-eminence of history painting. History painting was a species of narrative painting, and the argument between Tinto and Pattieson draws on the eighteenth-century debate on visual narrative that is normally traced back to Lessing’s Laokoon (1766).8 The principal question at issue was whether a painting could tell a story, given that paintings are silent and only able to represent, it might be thought, arrested moments of time. The standard answer to the question was that they could do so by representing ‘pregnant’ moments, moments in which there is visible evidence of what has led up to them and what will follow from them. While Scott does not invoke the metaphor of pregnancy, Tinto’s sketch does include the representation of a young woman at a similarly public moment of
162
Narrative Order, 1789–1819: Part II
transition: the bride. It is the bride, as an actor in a rite of passage and as an ideal subject for narrative painting, with which this discussion of Scott’s novel will be principally concerned. In exploring the issue of visual narrative and its relationship to verbal narrative, Scott probably did have Reynolds’s Discourses in mind. While Reynolds did not address himself directly to the standard question about narrative painting – ‘can a painting tell a story?’ – his argument for the pre-eminence of history painting did involve an argument about narrative. One reason why history paintings had universal significance was, according to Reynolds, that they derived their subjects from existing classical and scriptural narratives. These pictures can tell stories, he argues, because the instants they represent are taken from sequences of events already known to us as stories. Another contemporary influence on Scott’s thinking may have been Austen’s Emma which he had reviewed in the Quarterly for March 1816, since the issues raised by Emma’s portrait of Harriet Martin are echoed in Scott’s description of the picture of Lucy Ashton – ‘the bride of Lammermoor’ – in the first chapter of his novel. Emma and The Bride of Lammermoor are, or include, significant instances of ekphrasis, the literary representation of visual art. Ekphrasis is a practice which – as James Heffernan’s more abstract definition of it as ‘the verbal representation of graphic representation’ suggests – tends to involve effects of mirroring, encouraging us to ponder the processes of representation in which we and the author are ourselves engaged.9 The connection between The Bride of Lammermoor and the best known example of ekphrasis in English – the ‘Ode on a Grecian urn’ – is necessarily of a different kind. Keats appears to have begun the ‘Ode’ in May 1819 and it was first published in Annals of the Fine Arts in January, 1820. The Bride was published, in Edinburgh and London, in June 1819, so that if there was any actual influence, which is unlikely, it must have been of Scott on Keats not the other way round. If no influence was involved, the simultaneous appearance of the two texts makes a strong statement about the experience of narrative in Britain in 1819. The ‘Ode’, like Scott’s novel, explores the impulse to derive stories from works of visual art, and specifically to derive from them stories about the past, histories. Scott, as we shall see, identifies the picture that is asked to divulge its story with the young woman – the bride – portrayed in the picture, an identification which appears in Keats’s poem in the metaphoric apostrophe to the urn: ‘thou still, unravished bride of quietness’. The urn in Keats’s poem is addressed as an historian (a ‘sylvan historian’) because the speaker of the poem is himself a kind of
The Still Unravished Bride of Lammermoor
163
historian, hoping that the ‘storied’ artefact from the past will give up its narrative secrets and itself thereby become an historian. This ‘attempt to double the historians’, as Susan Wolfson puts it,10 places Keats’s narrator in a tradition of historiographical metaphor recently explored by Bonnie Smith, Anne Curthois and John Docker, in which ‘the male historian despairs that he cannot finally capture and penetrate the mysteries of the female past’.11 Of course a short lyric poem like the ‘Ode’ is a long way from being an historical novel. But it is, in a sense, precisely this difference of genre between lyric and novel which the poem is about. It is about why neither the story of the urn nor the stories on the urn can be told. It is a short lyric about why it cannot be a long narrative. Scott’s novel both echoes and contrasts with Keats’s lyric in this respect. Unlike the ‘Ode’, it involves us in a serious attempt to penetrate the mysteries of the female past but, like the ‘Ode’, it makes no attempt to conceal the failure of this attempt. The distance between the short lyric and the long novel is measured rather exactly by the fact that Keats writes about a metaphorical bride and foster child rather than about the literal bride and foster child who are staple characters of narrative fiction. A bride who is a foster child is separated from her sexual origins and her sexual destiny; and it is our curiosity about that original parentage to which a novel will always respond, revealing the specific difficulties – of class, proximity of blood, national or ideological affiliation – which the bride’s unknown or double parentage may put in the way of her marriage. Dick Tinto plans to use his sketch as the basis for a history painting, and he offers it to Peter Pattiesson as the basis for a written history. Tinto had been complaining that Pattieson’s previous narratives (that is, Scott’s previous novels) had had too much dialogue in them, not enough description: ‘Your characters,’ said he, ‘my dear Pattieson, make too much use of the gob-box; they patter too much – (an elegant phraseology, which Dick had learned while painting the scenes of an itinerant company of players) – there is nothing in whole pages but chat and dialogue.’12 Tinto believes that description can communicate character and action better than dialogue because it can bring scenes before the mind’s eye as he, the painter, can bring them before the body’s eye. Pattieson disagrees: he says Tinto’s argument ‘ “confounded the operations of the pencil [brush] and the pen” ’ (p. 11). He nevertheless offers to ‘make one trial of a more straight forward style of composition, in which my
164
Narrative Order, 1789–1819: Part II
actors should do more, and say less, than in my former attempts of this kind’ (p. 12). Tinto then offers Pattieson a subject for his experiment: The story, he said, was, by tradition, affirmed to be truth, although, as upwards of a hundred years had passed away since the events took place, some doubt upon all the accuracy of the particulars might be reasonably entertained. When Dick Tinto had thus spoken, he rummaged his portfolio for the sketch from which he proposed one day to execute a picture on a canvas of fourteen feet by eight. The sketch, which was cleverly executed, to use the appropriated phrase, presented an ancient hall, fitted up and furnished in what we now call the taste of Queen Elizabeth’s age. The light, admitted from the upper part of a high casement, fell upon a female figure of exquisite beauty, who, in an attitude of speechless terror, appeared to watch the issue of an animated debate betwixt two other persons. The one was a young man, in the Vandyke dress common to the time of Charles I, who, with an air of indignant pride, testified by the manner in which he raised his head and extended his arm, seemed to be urging a claim of right, rather than of favour, to a lady, whose age, and some resemblance in their features, pointed her out as the mother of the younger female, and who appeared to listen with a mixture of displeasure and impatience. (p. 12) This is as far as Pattieson can get in making sense of the sketch, much to the sketcher’s annoyance: ‘Yet, on my honour’, said Dick, ‘I would swear you had been born blind, since you have failed at the first glance to discover the subject and meaning of that sketch. I do not mean to praise my own performance – I leave these arts to others – I am sensible of my own deficiencies, conscious that my drawing and colouring may be improved by the time I intend to dedicate to the art. But the conception – the expression – the positions – these tell the story to everyone who looks at the sketch; and if I can finish the picture without diminution of the original conception, the name of Tinto shall no more be smothered by the mists of envy and intrigue.’ I replied that I admired the sketch exceedingly; but that to understand its full merit, I felt it absolutely necessary to be informed of the subject. ‘That is the very thing I complain of,’ answered Tinto; ‘you have accustomed yourself to so much of these creeping twilight details of
The Still Unravished Bride of Lammermoor
165
yours, that you are become incapable of receiving that instant and vivid flash of conviction, which darts on the mind from seeing the happy and expressive combinations of a single scene, and which gathers from the position, attitude, and countenance of the moment, not only the history of the past lives of the personages represented, and the nature of the business on which they are immediately engaged, but lifts even the veil of futurity, and affords a shrewd guess at their future fortunes.’ (p. 13) Pattieson actually shares Tinto’s assumption that to understand the sketch means to deduce from it the sequence of events preceding and succeeding the instant represented. The picture’s meaning is its narrative. The difference between the two men is simply that Pattieson does not believe that this narrative meaning is recoverable: ‘I protest to you, Dick, that were I permitted to peep into that Elizabeth-chamber, and see the persons you have sketched conversing in flesh and blood, I should not be a jot nearer guessing the nature of their business, than I am at this moment while looking at your sketch.’ (p. 13) In any case, the upshot is that it is not so much Tinto’s sketch but his notes and ‘written memoranda’ which, Pattieson tells us, he proceeded to ‘weave’ into his ‘tale’ (p. 14). At one point Tinto swears ‘by the soul of Sir Joshua’ and, as I have suggested, Reynold’s views are very much in evidence throughout the argument between narrator and painter. However, Scott’s purpose in having Tinto allude to Reynolds’s views is complex and evidently ironic, or ironic at least in part. Scott seems to show Tinto comprehensively misunderstanding Reynolds. For one thing, Tinto is reversing the relationship of derivation between verbal and visual narrative as that relationship had been envisaged in the Discourses. Reynolds had argued that in historical painting ‘the subject . . . is commonly supplied by the Poet or Historian’, in particular by ‘the great events of Greek and Roman fable and history’ and of ‘scripture history’.13 An example of this would be a painting of ‘the massacre of the innocents’, the example chosen by Edmund Burke in the passage from the Reflections quoted at the head of the present chapter. Dick Tinto, by contrast, wants Pattieson to derive a written narrative from his own visual one.
166
Narrative Order, 1789–1819: Part II
Furthermore, the classics and scripture are the ideal narrative sources for history painting because, Reynolds had argued, they were generally known and of general interest. That is to say, they are stories which early education, and the usual course of reading, have made familiar and interesting to all Europe, without being degraded by the vulgarism of ordinary life in any country. (p. 58) By contrast, Tinto takes his subject from a version of some events that supposedly took place over one hundred years before, told to him by the ‘aged goodwife’ of a farm in ‘the mountains of Lammermoor’. From this dubious piece of local oral history he hopes that he himself can produce a ‘history piece’ (p. 14), and that Pattieson will write an historical narrative. Finally, Tinto expects this story to be instantly legible in a sketch. For Reynolds, by contrast, sketches are distinguished from finished paintings precisely by being relatively ‘undetermined’, so that ‘the imagination supplies more than the painter himself, probably, could produce; and we accordingly often find that the finished work disappoints the expectation that was raised by the sketch’. The finished painting is nevertheless the higher form and ‘every thing shall be very carefully and distinctly expressed, as if the painter knew, with correctness and precision, the exact form and character of whatever is introduced into the picture’ (p. 164). On every point then, Tinto opts for what, in the context of the ideas of ‘Sir Joshua’, would be the procedure least compatible with narrative legibility: the source of the subject in the variable and particularistic popular storytelling of a single nation rather than universal and stable written narratives; the status of the work as a sketch rather than a finished painting. However, if we are supposed to recognize that Dick Tinto takes Reynolds’s name comprehensively in vain – and makes himself ridiculous in doing so – we cannot help being aware, as we quote Reynolds against Tinto, that Scott himself must be, to some extent, on Tinto’s side. Scott himself is after all engaged in deriving an historical narrative from a combination of the written and oral sources – most of them of course not already known to most of his readers – of ‘ordinary life’ in a particular country. Finally, the kind of authority to which Scott’s historical writing aspires is really quite different from the kind of authority to which
The Still Unravished Bride of Lammermoor
167
history painting as espoused by Reynolds aspired. Reynolds’s version of history painting aspired to the authority of history in the sense that Watkin Tench was trying to write it: a kind of narrative (Gerard Genette’s histoire) that gains authority to the extent that it breaks all links with the writing present. History is then pushed to that distance at which it can be seen for what it truly is, seen in perspective, as having the structure of a narrative in the third person and the past tense. Historical authority in Scott comes from the opposite direction, from drawing our attention to the relationships between then and now, an authenticity derived from foregrounding the written and oral sources which mediate present and past. In foregrounding these sources Scott does of course also often suggest their partiality and contradictoriness, nowhere more so than in The Bride of Lammermoor. The idea of a past seen in perspective, where the past and its representation will converge, does indeed survive in this novel. It survives, however, not as the takenfor-granted framework of the action but only as a powerful idea, an ever-receding object of desire, a desire focused on the figure of the bride and (as we shall see) on the language used to describe her. The relationship between Scott and Reynolds is a complex and awkward one, comparable to the relationship between Wordsworth and the Aristotelian tradition (to which indeed Reynolds belongs). In each case, the modern writer is trying to find support or sanction in the classical tradition for doing things which are in fact at odds with the tradition. If Scott’s idea of historical representation is really quite different from Reynolds’s, and in some ways similar to Tinto’s, Reynolds would nevertheless be important to Scott because of the aesthetic and social status which Reynolds, writing of a ‘sister art’, had given to historical painting. Wishing to agree with Reynolds, Scott is seeking Reynolds’s permission for doing in verbal art something very different from what Reynolds thought should be done in visual art. And it is the contrast which Reynolds makes between the sketch in verbal and in visual art which Scott probably has in mind in this context. In a letter dated 15 April 1819 (just two months before the publication of the novel), Scott wrote to his patron, the Duke of Buccleuch, about William Allen’s sketch of the murder of Archbishop Sharp on Magus Moor in 1679, an event which is the subject of partisan debate in this novel and which had precipitated the action of Old Mortality (1816): The savage ferocity of the assassins crowding one on another to strike at the old prelate on his knees contrasted with the old man’s figure and that of his daughter endeavouring to interpose for his
168
Narrative Order, 1789–1819: Part II
protection & withheld by a ruffian of milder mood than his fellows – the dogged fanatical severity of Rathillet’s countenance who remained on horseback witnessing with stern fanaticism the murder he did not choose to be active in lest it should be said that he struck out of private revenge are all amazingly well combined in the sketch. I question whether the artist can bring them out with equal spirit in the painting which he meditates – sketches give a sort of fire to the imagination of the spectator who is apt to fancy a great deal more for himself than the pencil in the finished picture can possibly present to his eye afterwards.14 Scott’s description of the sketch – a sequence of phrases linked by dashes that do not add up to a well-made sentence – echoes the urgency and incompleteness attributed to the sketch itself. Letter and sketch, it is implied, are linked, by their closeness to the event itself in its own urgency and incompleteness. In this they contrast with the finished painting or narrative which have the clarity and composure of a finished action. It is close to the contrast – also made by means of dashes and loosened grammar – between the journal entry in Tench’s Account of the Settlement at Port Jackson and the ‘cold track of narrative’. Scott’s preference for the sketch over the ‘finished picture’ here echoes a widespread view that has been taken as characteristic of romanticism.15 However, his final words – ‘sketches give a sort of fire to the imagination of the spectator, who is apt to fancy a great deal more for himself than the pencil, in the finished picture, can possibly present to his eye afterwards’ – also very closely echo Reynolds’s acknowledgement of the superior pleasures offered by the sketch over the finished painting in this particular respect, an acknowledgement that does not effect Reynolds’s view of the sketch as the lower form. Furthermore, Reynolds had made a distinction between verbal and visual art so far as the acceptability of sketchiness was concerned: Those general ideas, which are expressed in sketches, correspond very well to the art often used in Poetry. A great part of the beauty of the celebrated description of Eve in Milton’s Paradise Lost consists in using only general indistinct expressions, every reader making out the detail according to his own particular imagination – and his own idea of beauty, grace, expression, dignity or loveliness: but a painter, when he represents Eve on a canvas, is obliged to give a determined form, and his own idea of beauty distinctly expressed. (p. 164)
The Still Unravished Bride of Lammermoor
169
If we put these statements on sketches from the Discourses together, we can see that Scott could, if he felt the need, regard Reynolds as giving permission for a significant degree of indeterminacy, a lack of ‘finish’ in his novel, and perhaps particularly for the representation of its beautiful young heroine. The argument between Pattieson and Tinto, the writer and the painter, is in any case relevant to the remainder of The Bride of Lammermoor in two principal respects. First, it predisposes us to read through the novel in anticipation of eventually coming upon a clear, written representation of the scene which Tinto has sketched. Secondly, when the first chapter is seen in relation to the rest of the novel, the argument between Tinto and Pattieson encourages us to believe that there is a particular kind of human activity – rites of passage as formal acts of closure and inauguration – upon which narrative order of any kind, pictorial or otherwise, depends. What comes first in a text always has a special importance, predisposing us to read what follows in a particular way. This is especially true in the present case. What Tinto has shown Pattieson is a sketch, not the fully fledged ‘history piece’, the ‘picture of fourteen feet by eight’, that he hoped one day to produce from the sketch. As Burke’s use of the metaphor of ‘sketch’ and ‘history-piece’ in the Reflections makes clear, a sketch, particularly a sketch of a real or a supposedly real event, places itself firmly within time and history, pointing insistently towards the past and the future. It points back to the event it represents and forward to the planned painting. A sketch is always a sketch of something and often a sketch for something. By contrast, a finished painting corresponds to the finished character of the past as histoire: fixed, unalterable, separated from us. By starting the novel with the description of a sketch, Scott exaggerates and draws attention to two features of any novel and its reading. On the one hand, novel readers are always engaged in inventing and revising possible denouements. On the other hand, we know – or believe we know – that there will turn out to have been only one possible denouement, which will not be exactly like any of the ones we have anticipated. Starting the novel with the sketch, Scott both gives unusually explicit licence to the reader’s imagination, and seems to tell us – equally firmly – that the actual ending is already fixed and that the rest of the novel will only fill in what is already there. Special licence is given to the imagination on condition that imagination is recognized as fantasy. And of course this is relevant in a quite special sense to the historical novel, the course of whose action is determined from the start
170
Narrative Order, 1789–1819: Part II
in a particularly explicit way. The various possible futures we may imagine as we read are already very firmly adjudicated upon because the future is already the past.16 As we read through the novel we wait for the blurred still – the sketch – to appear as a sharply focused scene. This scene will, we anticipate, retrospectively make full sense of the initial sketch, and simultaneously of the supposedly real event of which it was a sketch. The finished novel itself will borrow some of the quality of a finished painting. However, when the scene does come into view late on in the novel, it does not completely succeed in doing these things. The gap between historical events and their representation is not completely closed. The Bride of Lammermoor makes us especially conscious of those processes of anticipation, retrospection and anticipated retrospection which are involved in establishing any kind of narrative; but although we are drawn to read on towards the moment of truth, what appears is a moment of violence in which the only person who could complete the story (the bride herself) loses her mind and dies. The scene in the ‘Elizabeth-chamber’ itself does, it is true, eventually come into a kind of focus. In Chapter 33, we find that the ‘female figure of exquisite beauty’ is Lucy Ashton. Just a few moments prior to the scene which is the subject of Tinto’s sketch, Lucy has finished signing ‘deeds of espousal’ between herself and Hurleston of Bucklaw. She has signed under pressure from her family led by her cruel and manipulative mother, Lady Ashton, who is the second woman in the sketch. But just as Lucy has finished signing, the man whom she loves and to whom she has already promised herself bursts into the room. This man, Edgar, Master of Ravenswood, whose family estates have been taken over by the Ashtons, is the third figure in Tinto’s sketch. In describing the sketch, Pattieson said that the female figure of exquisite beauty ‘appeared to watch the issue of a debate betwixt two other persons’. That is indeed what happens, and it is Lady Ashton who wins the debate or power struggle. The wedding of Lucy Ashton and Bucklaw takes place as planned, four days later. However, soon after the couple have left the celebrations, screams are heard coming from the bridal chamber. When the room is entered, Bucklaw is found stabbed and bleeding (but not dead) on the threshold and Lucy ‘gibbered’ and ‘made mouths’ in the corner of the old fireplace (p. 260). She soon dies and is buried ‘in a coffin bearing neither name nor date’ (p. 262). Lucy Ashton’s burial is the last of that ‘interlinked series of betrothals and burials’ on which the novel is based. However, their significance
The Still Unravished Bride of Lammermoor
171
here, and their relationship to the narrative ordering of experience which they are supposed to effect, arises from the uncomfortable way in which they are ‘interlinked’. A bride is indeed, as I have suggested, tailor-made for narrative painting because of her position in a rite of passage. ‘A woman at her marriage; a woman just about to be or very recently married’ (OED), ‘bride’ is a clear, a named, identity, but an explicitly transitional one, between the longer-term conditions of maid or daughter beforehand and wife afterwards. A bride, and the actions in which a bride is involved, are a natural subject for narrative painting in the context of the kind of views presented by Tinto, since they provide ready-made turning points, instantly legible moments of transition between the stages of a prescribed sequence. However, most of the rites in this novel are ‘maimed’: I take the word from Hamlet, to which the novel constantly alludes. Ophelia’s burial rites were described as maimed because she had literally maimed herself; Lucy Ashton’s funeral rites are maimed because she has literally maimed her bridegroom. Anthropologists refer to the central segment of a passage rite as its liminal or threshold stage and in The Bride of Lammermoor ‘the body of the bridegroom was found lying on the threshold of the bridal-chamber, and all around was flooded with blood’ (p. 260). The blood is presumably Bucklaw’s, but it is possible to think of it also as hymenial, with Scott making a silent pun on Hymen and hymen, linking the ravishing of the bride with the death of the groom – Hamlet’s two ‘consumations’ – just as Lucy’s funeral follows hard on the heels of her marriage. There is an overclose ‘interlinking’ of passage rites, so that they confuse the narrative order which they are designed to articulate. As Hamlet put it, ‘the funeral baked meats do furnish forth the wedding breakfast’. The maiming of the bride’s marriage rites perpetuates a bridal condition which has been stymied from the start. In fact both Lucy and Ravenswood are stymied in the same fashion. The impossibility of Ravenswood’s situation is that if he is to keep the vow he has made to Lucy – and thereby save his ancestral estates by accommodating himself to the Ashtons and the new British regime – he must break the vow of revenge made at his father’s grave. Lucy Ashton’s situation is that, used first by her father as sexual bait to catch Ravenswood and the Ravenswood estates, and then by the more powerful mother on behalf of Bucklaw, she finds herself claimed as ‘affianced bride’ by two men at once, bound by incompatible promises, utterly immobilized at a moment of supposed transition, stuck on the threshold.
172
Narrative Order, 1789–1819: Part II
The event presented by Tinto’s sketch is a scene in which the very possibility of moving through the rite, from one stage (the signing of the deeds of espousal) to the next (the wedding) is threatened. Not only is it difficult for Pattieson to generate a narrative meaning from the sketch; the sketch is, it turns out, of an event which helps to undermine the very possibility of narrative order of any kind. There are a number of other occasions in the novel where we are invited to think of events as painted scenes. The principal example is of another threatened rite of passage, but in this case the rite – the burial of the old Master of Ravenswood, Edgar’s father – is shown actually taking place, even though under threat. The old Master was tory, jacobite, feudal, episcopalian. He lost his estates to Lucy’s father, Lord Ashton, who is whig and presbyterian. Armed men, sent by Lord Ashton at the request of the presbyterian church authorities, attempt to intervene to stop a funeral being performed by an episcopalian priest, a ceremonial which ‘was readily complied with by the tory gentlemen, or cavaliers, as they affected to style themselves, in which faction most of his kinsmen were enrolled’ (p. 19). Tory episcopalian swords keep whiggish presbyterian swords at bay for the duration of the ceremony, which Pattieson describes as follows: The scene was worthy of an artist’s pencil. In the very arch of the house of death, the clergyman, affrighted at the scene, and trembling for his own safety, hastily and unwillingly rehearsed the solemn service of the church, and spoke dust to dust, and ashes to ashes, over ruined pride and decayed prosperity. Around stood the relations of the deceased, their countenances more in anger than in sorrow, and the drawn swords which they brandished forming a violent contrast with their deep mourning habits. In the countenance of the young man alone, resentment seemed for the moment overpowered by the deep agony with which he beheld his nearest, and almost his only friend, consigned to the tomb of his ancestry. A relative observed him turn deadly pale, when, all rites being now duly observed, it became the duty of the chief mourner to lower down into the charnel vault, where mouldering coffins shewed their tattered velvet and decayed plating, the head of the corpse which was to be their partner in corruption. He stept to the youth and offered his assistance, which, by a mute motion, Edgar Ravenswood rejected. Firmly, and without a tear, he performed that last duty. The stone was laid on the sepulchre, the door of the aisle was locked, and the youth took possession of its massive key. (p. 20)
The Still Unravished Bride of Lammermoor
173
Edgar then addresses his fellow mourners, concluding as follows: ‘Others bury their dead in sorrow and tears, in silence and in reverence; our funeral rites are marred by the intrusion of bailiffs and ruffians, and our grief – the grief due to our departed friend – is chased from our cheeks by the glow of just indignation. But it is well that I know from what quiver this arrow has come forth. It was only he that dug the grave who could have the mean cruelty to disturb the obsequies; and heaven do as much to me and more, if I requite not this man and his house the ruin and disgrace he has brought on me and mine!’ (pp. 20–1) Clearly, Scott sees this particular rite as having a dual function. It marks not only the end of an individual life and the passing of name and authority (and ‘massive key’) to the son, but also the passage from one stage in a wider history – feudal, episcopalian, Scottish – to another – capitalist, contractual, presbyterian, British. In both respects, however, the rite is maimed; or, as Edgar Ravenswood puts it, ‘marred’. The ghost of Edgar’s father does not actually appear to his son on the battlements of the family seat at Wolf’s Crag, but the obligation to avenge his father certainly weighs on his brain like a nightmare. And much of the novel is taken up with Caleb Balderstone’s farcical attempts to refabricate the family’s feudal past. The past is dead but not properly buried. However, if the transition between stages of the individual and national life are not clearly made, at least the lack of a clear transition is, in that particular episode, clearly articulated. Elsewhere in the novel that is not so, a failure which, however, intensifies the novel’s power as an exploration of narrative. In particular, Scott’s own uncertainty as to whether he wished to set the action before or after the Act of Union confuses the analogies between political Union and marital union in a way peculiarly appropriate to the character of this marital union itself as a marred rite.17 One feature of paintings is their silence: ‘that serene and silent art’, Tinto calls it. Do events themselves have to be silent in order to be ‘worthy of an artist’s pencil’? And if they are silent, how can they communicate narrative meaning? Of course the burial of Edgar’s father is not silent, even prior to Edgar’s fighting speech. The clergyman speaks the funeral service, and we are given a few summary words of it (he ‘spoke ashes to ashes and dust to dust’). But the implication is that neither the clergyman’s congregation nor the novel’s readers need to hear (or read) the words spoken. Indeed, to the extent that rites of
174
Narrative Order, 1789–1819: Part II
passage perform scripts derived from scripture they do indeed almost conform to at least one of Reynold’s criteria for history painting. The scene is ‘worthy of an artist’s pencil’ partly because in it the clergyman ‘rehearsed’ a script we all supposedly know already. Later on, when Edgar’s relative ‘stept to the youth and offered his assistance’, we can interpret this offering as a silent one or not, as we wish; Edgar’s response is a ‘mute [but intelligible] motion’. The implication is that these events lend themselves to pictorial representation, and to ‘description’ rather than ‘chat and dialogue’. Although things are said in the course of these events, hearing the words is not a prerequisite of understanding either them or the events of which they are a part. But this relation between the characteristics of the representation and of the events represented is one that Pattieson and Scott are more puzzled by than they will admit. This is evident in Pattieson’s promise to try ‘a more straight-forward style of composition, in which my actors should do more, and say less, than in my former attempts of this kind’. Does Pattieson mean that he will write about people who do not talk so much, or that he will represent people (talkative or otherwise) by describing their actions rather than by quoting their words? It is a double meaning that can be present of course – as Hamlet notably demonstrates – in the word ‘actor’, which means both ‘somebody doing something’ and ‘somebody representing somebody doing something’. It is present, above all, as we shall see, in Scott’s description of the bride in Tinto’s sketch. The funeral of Edgar’s father is ‘worthy of an artist’s pencil’ partly because it is a rite of passage. It is an event of conventional and heightened gesture and visible religious and political insignia, performed on the basis of a known script, whose narrative meaning is evident even if the words spoken in it cannot be heard. The relation between narrative, painting and silence in the scene sketched by Tinto is interestingly different from this, though compatible with it. In his argument with Tinto, Pattieson had seemed to suggest that it was the discrepancy between the silence of Tinto’s sketch and the talkativeness of two of the people sketched which undermined the sketch’s attempt to establish a narrative meaning: ‘I protest to you, Dick, that were I permitted to peep into that ElizabethChamber, and see the persons you have sketched conversing in flesh and blood, I should not be a jot nearer guessing the nature of their business, than I am at this moment while looking at your sketch.’ (p. 13)
The Still Unravished Bride of Lammermoor
175
Pattieson is envisaging a situation in which he could see the events but not hear them: almost as if the two people he was watching were miming an argument. The immobility of pictures and their silence are often cited together in discussions of visual narrative, and both are certainly at issue in the argument between Pattieson and Tinto; but at this stage of the argument it is the sketch’s silence which is said to inhibit narrative meaning. The third person in the sketch however, Lucy Ashton, is represented as being both silent and immobile. This makes her, in one sense, specially appropriate for pictorial representation: The light, admitted from the upper part of a high casement, fell upon a female figure of exquisite beauty, who, in an attitude of speechless terror, appeared to watch the issue of an animated debate betwixt two other persons. (p. 12) These lines point to a specially close relationship between the sketch and the young woman in the sketch. For one thing, there is a remarkable similarity between the triangular relationship within the sketch and the triangular relationship of which the sketch is itself a part (and then between each of those triangles and the triangle constituted by Scott, the reader and the novel). Pattieson and Tinto are arguing over the silent sketch; within the sketch the young man and the older woman are arguing over the younger silent woman. This young woman is a ‘female figure of exquisite beauty . . . in an attitude of speechless terror’ whose only action is to ‘watch’. Silence, stillness and beauty: qualities proper, it is implied, to pictures and to young women. The strongly gendered character of the language of aesthetic appreciation is of course evident in Reynolds’s discussion of Milton’s Eve, where the ‘beauty of the celebrated description’ is not clearly distinguished from the ‘beauty’ of the woman described. Equally to the point is Austen’s Emma, where it is partly this convergence of vocabularies which is shown to have dangerous human consequences. Mr Elton’s complimentary references to the picture of Harriet Martin can be taken by Emma as compliments to Harriet where they are intended as compliments to her own skill as an artist. Similarly, here, it is possible to understand the phrase ‘a female figure of exquisite beauty’ either as a compliment to Lucy Ashton or to Dick Tinto, since the word ‘figure’ can mean either a human form or the representation of a human form. The difference is that Austen wishes to reveal and question the
176
Narrative Order, 1789–1819: Part II
customary identification between the language of gender and the language of pictorial representation. Scott’s position is far less clear-cut: he is unwilling directly to question the bride as the object of a desire for the convergence of life and its narrative representation. It is worth making clear what is perhaps obvious, that we are dealing in these cases not just with an aesthetic language which is strongly gendered, but with an everyday language of gender which is strongly aestheticized. That is why it is possible for Reynolds and Scott, and for Emma and Mr Elton not to be wholly aware of what is happening as they move between talking about young women and talking about visual representations of young women. The association of real young women with ‘beauty’ and with designed form (‘figure’) seems to have made them the point in the culture which most clearly promised a convergence between reality and its formal representation or, as Keats put it, Truth and Beauty. But this promise is only that: an incitement to pursue what is never attained. Only so long as we have not yet seen things as they are can we believe it is possible to see things as they are. Only so long as truth and beauty have not yet converged can we believe that they do. Only so long as the contract has not yet reached its term can we believe that the contractual promise will turn out to have been truly made. Lucy Ashton’s silence, immobility, and beauty clearly make her, in one sense, more worthy of an artist’s pencil than the other two people in Tinto’s sketch. And as a bride, between two prescribed stages of the marriage rite, there should not be a person whose narrative meaning is more instantly legible. But this silence, which should be as eloquently liminal as the white of a bride’s dress, is – as silence can of course always be – the very sign of inscrutability. Lucy Ashton awaits the issue of others’ debate. Signed up to one bridegroom but already spoken for by another, she seems no longer to have a self to speak for. She is, despite her literal presence, incommunicado. The double-binding Lady Ashton even dictates the letters Lucy writes to Edgar, Scott punning powerfully on the word ‘dictate’. The novel’s penultimate chapter is preoccupied with the difficulty of establishing narrative meaning when the protagonists are speechless and can communicate only through incoherent sounds and physical violence. At the death of Lucy, ‘convulsion followed convulsion, till they closed in death, without her being able to utter a word explanatory of the fatal scene’ (p. 261). Nevertheless, there are strong pointers to possible meanings in these events, or at least to what it is about these events which makes them such a threat to a specifically narrative form of meaning.
The Still Unravished Bride of Lammermoor
177
There is a repeated reference, for instance, to thresholds. The bridegroom’s body is found bleeding ‘on the threshold of the bridal chamber’ (p. 260) and it is assumed that ‘the bride, in a sudden fit of insanity, had stabbed the bridegroom at the threshold of the apartment’ (p. 261). Lucy is herself discovered in the corner of the great old-fashioned chimney of the apartment...her eyes glazed, and her features convulsed into a wild paroxism of insanity. When she saw herself discovered, she gibbered, made mouths, and pointed at them with her bloody fingers, with the frantic gestures of an exulting demoniac. Female assistance was now hastily summoned; the unhappy bride was overpowered, not without the use of some force. As they carried her over the threshold, she looked down, and uttered the only articulate words that she had yet spoken, saying, with a sort of grinning exultation, – ‘So, you have ta’en up your bonny bridegroom?’ (p. 260) But who is it who speaks these last words, since the Lucy Ashton we have known never uses dialect words? Her speaking in dialect in fact brings together different aspects of identity – gender, nationality and class – more neatly than they are brought together elsewhere in the novel. The man she attacks is named Bucklaw, the law of the buck (the male). This law is associated at this point with British law and, linguistically, with a more English, more polite, and more upper-class-specific, form of English. Her speaking these words in dialect can therefore be read as a refusal to be a British lady, a pure object of property-exchange in a marital union linked to and standing for the Act of Union. Lucy’s words are in any case part of an episode which undermines, in a very precise way, the kind of narrative meaning generated by the marriage rite. Earlier, Lady Ashton had persuaded her daughter to ‘annul the engagement’ (p. 252) to Ravenswood; now, vividly putting into reverse the ritual of the bridegroom carrying the bride over the domestic threshold after the wedding, Scott describes Lucy being carried back across the threshold. And at that moment she utters those dialect words which in effect say ‘no’ to the question – ‘Do you take this man . . .?’ – which she would have already been asked, and have answered differently, in the marriage service. These are of course only speculations about what is really happening in these episodes, speculations about the meaning of Scott’s novel. But this is appropriate if one of the things that is happening is the
178
Narrative Order, 1789–1819: Part II
undermining of narrative meaning. Although no longer either as silent or as pretty as a picture, Lucy Ashton still does not get to tell her story: A tide of recollections seemed to rush upon her, which her mind and body were alike incapable of bearing. Convulsion followed convulsion, till they closed in death, without her being able to utter a word explanatory of the fatal scene. (p. 261) As for Bucklaw, echoing Coleridge’s ‘Ancient Mariner’, he ‘had arisen from the bed of sickness a sadder and wiser man’ (p. 339). Both Coleridge’s wedding guest and Bucklaw go through experiences which put them off weddings. But although we are never clearly told, in either case, what the men’s newly acquired wisdom consists in, at least the wedding guest got told a story. If Lucy Ashton told Bucklaw a story before she stabbed him the rest of us do not get to hear it. We are left, in the end, with a literary equivalent of a sketch. We are left with a vivid but only partly legible scene of violent death, not a ‘massacre of the innocents’ but something perhaps close to the scenes of seventeenthcentury violence which, in the Reflections on the Revolution in France, are screened by Burke’s allusion to the biblical event.
Notes
1
Narrative order
1. Wordsworth and Coleridge: Lyrical Ballads, the Text of the 1798 Edition with the Additional 1800 Poems and the Prefaces, ed. R.L. Brett and A.R. Jones, London: Methuen, 1963. All references to the poems and Prefaces will be to this edition. 2. David Simpson, The Academic Postmodern and the Rule of Literature: A Report on Half-Knowledge, Chicago and London: University of Chicago Press, 1995, pp. 55–6. Simpson’s whole discussion of storytelling in The Academic Postmodern has helped me to develop my own argument. 3. Thomas Paine, Rights of Man, ed. Henry Collins, Harmondsworth: Penguin, 1969, p. 265. William Godwin, ‘Enquiry into Political Justice’, in Political and Philosophical Writings of William Godwin, ed. Mark Philp, 7 vols, London: William Pickering, 1993, vol. IV, p. 81. All references to Political Justice will be to this edition (and to the 1793 edition of Political Justice unless otherwise stated). 4. Charles Pigott, A Political Dictionary: Explaining the True Meaning of Words, London: D.I. Eaton, 1795. See Corinna Wagner, ‘Symbolic Performances, Public Contests: Negotiating Politics in the 1790s’, British Association of Romantic Studies Bulletin and Review, 23, March 2003, pp. 11–14. 5. For excellent analyses of the part played by complex words in the political conflicts of the period, see John Barrell, Imagining the King’s Death: Figurative Treason, Fantasies of Regicide, 1793–6, Oxford: Oxford University Press, 2000; Jon Mee, Romanticism, Enthusiasm and Regulation: Poetics and the Policing of Culture in the Romantic Period, Oxford: Oxford University Press, 2003. For opposed views on the word ‘revolution’, see Mark Philp, ‘Revolution’, in An Oxford Companion to the Romantic Age: British Culture, 1776–1832, Oxford: Oxford University Press, 1999, pp. 17–25; Marilyn Butler, ‘Revolving in Deep Time: The French Revolution as Narrative’, in The French Revolution and English Romanticism: Politics and Rhetoric, eds Keith Hanley and Raman Selden, London: Harvester, 1990, pp. 1–22. 6. Alasdair MacIntyre, After Virtue: A Study in Moral Theory, Notre Dame: University of Notre Dame Press, 1981, p. 202. In an unpublished essay on ‘The idea of Narrative in Eighteenth Century Moral Writing’, Martin Golding had identified a ‘narrative idea of the moral life’. 7. Three decades after the publication of the Life of Savage, Johnson and Boswell made their tour of the Scottish highlands, and it is clear from their accounts of the journey that Johnson’s continuing sentimental attachment to the lost jacobite cause is inseparable from its association with a clan system which he believed could not now be revived. Johnson’s Journey to the Western Islands of Scotland (1775) is a brilliant analysis of the dissolution of this particularistic and ‘patriarchal’ order into the juridically and commercially unified British 179
180
8.
9.
10. 11. 12. 13.
14. 15.
16.
Notes state following ‘the late conquest’. His own analysis excludes, through its rigorous abstraction, his attachment to the clan system and to the surviving sites and personnel of the jacobite insurgency; elements which figure powerfully, as Johnson was aware they would do, in Boswell’s account, The Journal of a Tour to the Hebrides (1785). See Samuel Johnson, ‘A Journey to the Western Islands of Scotland’, and James Boswell, ‘The Journal of a Tour to the Hebrides’, ed. R.W. Chapman, Oxford: Oxford University Press, 1970. Donald Ault has discussed Blake’s poetry in a way that would provide a basis for comparing what Blake and Burke do to narrative order. Ault argues that the reader of Blake’s The Four Zoas (1797) is engaged in a ‘constant retroactive reconstitution’ of prior events within the narrative. See Donald Ault, Narrative Unbound: Re-visioning William Blake’s ‘The Four Zoas’, New York: Station Hill Press, 1987, p. xi. See J.G.A. Pocock, ‘Conservative Enlightenment and Democratic Revolutions: The American and French Cases in British Perspective’, Government and Opposition, xxiv, 1989, 81–105; ‘Post-Puritan England and the problem of Enlightenment’, in Culture and Politics from Puritanism to the Enlightenment, ed. P. Zagorin, Berkeley: University of California Press, 1980, 91–111. It will be evident that my own view differs from Pocock’s to the extent that I believe much British enlightenment writing (Johnson’s, for instance) is characterized by a tension between conservatism (support for the status quo) and reaction (support for some version of the status quo ante, whether Stuart or Commonwealth). The problem with Pocock’s view is exemplified by his presentation of the allegory of the three brothers in A Tale of a Tub (1704) as evidence of Swift’s commitment to a typically British middle way in Church and State (‘Conservative Enlightenment’, 85–6). The allegory does indeed represent this view, a view which Swift did hold, but which he held – or so other parts of the same text suggest – at a distance and even, sometimes, in contempt. That is to say, Pocock ignores the question of Swift’s irony. Edmund Burke, Reflections on the Revolution in France, ed. C.C. O’Brien, Harmondsworth: Penguin, 1969, p. 56. All references will be to this edition. Peter Brooks, Reading for the Plot: Design and Intention in Narrative, Oxford: Oxford University Press, 1984, p. 93. Marilyn Butler, Romantics, Rebels and Reactionaries, Oxford: Oxford University Press, 1981, p. 9. See Simon Bainbridge, Napoleon and English Romanticism, Cambridge: Cambridge University Press, 1995; Philip Shaw, Waterloo and the Romantic Imagination, London: Palgrave, 2002. Kelvin Everest, Coleridge’s Secret Ministry: The Context of the Conversation Poems, Brighton: Harvester, 1979, p. 5. Mary Wollstonecraft, Analytical Review, xxiv, September 1796, in The Works of Mary Wollstonecraft, eds M. Butler and J. Todd, London: Pickering, 1991, vol. 1, p. 467. Watkin Tench, Letters from Revolutionary France, ed. Gavin Edwards, Cardiff: University of Wales Press, 2001, p. 81. All quotations from the Letters Written in France are taken from this edition. For a fuller analysis of the Letters, see Gavin Edwards, ‘From Chester to Quimper via Sydney: Watkin Tench in Revolutionary France’, Literature and History, 11, 2, 2002, pp. 1–18.
Notes 181 17. Deirdre Coleman, ‘Chronicler of Two New Societies’, Southerly, 63, 1, 2003, pp. 206–10 (208). 18. Jean-Paul Sartre, Nausea, trans. Robert Baldrick, Harmondsworth: Penguin, 1965, p. 62. Frank Kermode, The Sense of an Ending, Oxford: Oxford University Press, 1967. 19. Jonathan Rée, ‘Bound to be in the wrong’, London Review of Books, 20 January 2005, pp. 20–2. 20. Galen Stawson, ‘Tales of the Unexpected’, Guardian Review, 10 January 2004, p. 15. 21. Louis Mink, ‘History and Fiction as Modes of Comprehension’, New Literary History, vol. 1, 1970, pp. 541–58 (557). 22. Paul Carter, The Road to Botany Bay: A Study in Spatial History, London: Faber, 1986, p. 309. See also, Homi K. Bhabha, Nation and Narration, London and New York: Routledge, 1990; The Invention of Tradition, ed. Eric Hobsbawm and Terence Ranger, Cambridge: Cambridge University Press, 1983. 23. ‘Biographies end with the subject’s death. Autobiographies have no such natural termination. However, this one has the advantage of ending at the moment of an undeniable and dramatic caesura in world history, in consequence of the attack of September 11, 2001 on the World Trade Centre and the Pentagon.’ Eric Hobsbawm, Interesting Times: A Twentieth-Century Life, Harmondsworth: Allan Lane, 2002, p. 411. 24. Hayden White links a range of relevant speculations as follows: ‘In the acquisition of language, Lacan had suggested, the child also acquires the very paradigm of orderly, rule-governed behaviour. Barthes adds that in the development of the capacity to assimilate “stories” and to tell them . . . the child also learns what it is to be that creature that, in Nietzsche’s phrase, is capable of making promises, of “remembering forward” as well as backwards, and of linking his end to his beginning in such a way as to attest to an “integrity” which every individual must be supposed to possess if he is to become a “subject” of (any) system of law, morality or propriety.’ Hayden White, The Content of the Form: Narrative Discourse and Historical Representation, Baltimore: Johns Hopkins University Press, 1987, p. 36. Nietzsche discussed remembering and promising in The Genealogy of Morals (1887); see Friedrich Nietzsche, The Birth of Tragedy and The Genealogy of Morals, trans. F. Golffing, N.Y.: Doubleday, 1956, pp. 189–230. Lacan described the sentence, out of which narratives are made, in a way that could be applied to narrative itself: ‘the sentence completes its signification only with its last term, each term being anticipated in the constitution of the others and, inversely, sealing their meaning by its retroactive effect’: Jacques Lacan, Ecrits: A Selection, trans. Alan Sheridan, London: Tavistock, 1977, p. 303. 25. See Carole Pateman, The Sexual Contract, Cambridge: Polity Press, 1988, pp. 58–9. 26. Ross Chambers, Story and Situation: Narrative Seduction and the Power of Fiction, Minneapolis: Minnesota University Press, 1984, p. 3. J.L. Austin, How to Do Things with Words, Oxford: Clarendon Press, 1962. The principal study of performative language in Romanticism is Angela Esterhammer, The Romantic Performative: Language and Action in British and German Romanticism, Stanford: Stanford University Press, 2000.
182
Notes
27. For an analysis of ‘The Parish Register’ in these terms, see Gavin Edwards, George Crabbe’s Poetry on Border Land, Lampeter and Lewiston: Edwin Mellen Press, 1990, pp. 77–144. Jerome McGann argues that ‘in “The Parish Register” Crabbe chooses the arbitrary order of a fundamental social document to organize his materials, and the consequence is the invention of a new kind of major form in poetry’: ‘George Crabbe: Poetry and Truth’, London Review of Books, 16 March, 1989, pp. 16–17. 28. From The Borough, 1810, ‘Letter 20: The Poor of the Borough: Ellen Orford’, lines 17–28. Text taken from George Crabbe: The Complete Poetical Works, ed. Norma Dalrymple-Champneys and Arthur Pollard, 3 vols, Oxford: Clarendon Press, 1988. All quotations from Crabbe are from this edition.
2
Samuel Johnson and the order of time
1. James Boswell, The Life of Samuel Johnson LLD, ed. R.W. Chapman, London: Oxford University Press, 1961, p. 22. 2. The Works of Samuel Johnson, III, ed. W.J. Bate and Albrecht B. Strauss, New Haven and London: Yale University Press, 1969, pp. 319–20. 3. See Deirdre Lynch, The Economy of Character: Novels, Market Culture and the Business of Inner Meaning, Chicago: University of Chicago Press, 1998, p. 35. 4. The Late Augustans: Longer Poems of the Later Eighteenth Century, ed. Donald Davie, London, 1958, Introduction, p. xxiii. 5. See Gavin Edwards, ‘Why Are Human Wishes Vain? On reading Samuel Johnson’s Vanity of Human Wishes’, Proceedings of the English Association North, vol. 2, Liverpool: Liverpool University Press, 1986, pp. 52–62. 6. See Jonathan Culler, ‘Story and Discourse in the Analysis of Narrative: Some American Discussions’, in his The Pursuit of Signs: Semiotics, Literature and Deconstruction, London: Routledge and Kegan Paul, 1981, pp. 169–87. 7. Samuel Johnson, An Account of the Life of Mr. Richard Savage, ed. Clarence Tracy, Oxford: Oxford University Press, 1971, p. 4. 8. On the factual accuracy or otherwise of Johnson’s account – an important issue which I do not address here – see Clarence Tracy, The Artificial Bastard: A Biography of Richard Savage, Toronto: University of Toronto Press, 1955; Richard Holmes, Dr Johnson and Mr Savage, London: Hodder and Stoughton, 1993. 9. Sigmund Freud, Collected Papers, trans. James Strachey, vol. 28, London: Hogarth Press, 1937, p. 19. 10. Michael Dummett, ‘Bringing about the Past’, in Truth and Other Enigmas, London: Duckworth, 1978, pp. 333–50. 11. ‘The Other Death’, in Borges: A Reader, ed. E.R. Monegal and A. Reid, New York: Dutton, 1981, pp. 213–17. I am grateful to Janet Turner Hospital for drawing my attention to the Borges story. 12. Thomas Hobbes, ‘Of Dominion Paternal and Despotical’, in Leviathan; or, The Matter, Forme, and Power of a Commonwealth Ecclesiastical and Civil (1651), ed. Michael Oakshott, New York: Collier, 1966, p. 152. 13. See Felicity A. Nussbaum, ‘ “Savage” Mothers: Johnson’s Life of Savage’, in Torrid Zones: Maternity, Sexuality, and Empire in Eighteenth Century English Narratives, Baltimore and London: Johns Hopkins University Press, 1995, pp. 47–72.
Notes 183 14. Howard Erskine-Hill, ‘Literature and the Jacobite Cause: Was There a Rhetoric of Jacobitism?’, in Literature and Jacobitism, ed. Evelyn Cruickshanks, Edinburgh: Edinburgh University Press, 1982; JCD Clark, Samuel Johnson: Literature, Religion and English Cultural Politics from the Restoration to Romanticism, Cambridge: Cambridge University Press, 1994. Clark argues for Jacobite implications in the Life of Savage. See also Ian Higgins, Swift’s Politics: A Study in Disaffection, Oxford: Oxford University Press, 1994.
3
Edmund Burke: Middles versus beginnings and ends
1. Perry Anderson, ‘Origins of the Present Crisis’, New Left Review, 23, Jan–Feb. 1964, pp. 26–53 (31). 2. My play on the word ‘recall’ derives from Shelley’s in Prometheus Unbound where Prometheus tells Jupiter that ‘The curse / Once breathed on thee I would recall . . .’. (lines 58–9), P.B. Shelley, The Poems of Shelley, vol. 2, eds Kelvin Everest and Geoffrey Matthews, London: Longman, 2000. ‘Recall’ here means simultaneously ‘remember’ and ‘retract’, as if his discourse (the curse), bearing within itself the life of its own subsequent material effects, were a root whose pulling up would kill the plant it has engendered, a physical weight which can and must be carried back to him. To remember the curse is to retract it, but it can only be remembered if it is called back (re-called) at him. The repressive power which Shelley attributes to the curse, and the liberating power he attributes to its retraction is relevant to his idea that poetic speech itself has ‘legislative’ force. See Gavin Edwards, ‘Mind-Forg’d Manacles: A Contribution to the Reading of Blake’s “London” ’, Literature and History, 5, i, Spring 1979, pp. 87–105 (103). 3. I am grateful to Iain McCalman for drawing this point to my attention. 4. See Conor Cruise O’Brien, ‘Introduction’ to Edmund Burke, Reflections on the Revolution in France, Harmondsworth: Penguin, 1969. For other Irish perspectives on the Reflections, see Seamus Deane, Strange Country: Modernity and Nationhood in Irish Writing since 1790, Oxford: Clarendon Press, 1997; Claire Connolly, ‘Reflections on the Act of Union’, in Burke’s ‘Reflections on the Revolution in France’: New Interdisciplinary Essays, ed. John Whale, Manchester: Manchester University Press, 2000, pp. 168–92; Terry Eagleton, Heathcliffe and the Great Hunger: Studies in Irish Culture, London: Verso, 1995, pp. 27–53. 5. Iain McCalman (ed.), An Oxford Companion to the Romantic Age: British Culture 1776–1832, Oxford: Oxford University Press, 1999, p. 526. 6. Quoted in O’Brien, Reflections, p. 36. O’Brien argues that Burke’s attempt to show that ‘the English Revolution, unlike the French one, had not been really revolutionary at all’ is ‘the most forced part of his argument’, an attempt ‘which we can judge successful only by choosing to forget about the contributions of the contemporaries of Henry VIII and Oliver Cromwell’ (p. 37). Indeed, we need also to forget those other parts of the Reflections where Burke says something like the opposite. O’Brien’s references to the ‘English Revolution’ (or ‘the English revolution’) are themselves significantly confusing: sometimes the phrase seems to refer specifically to the events of 1688 and sometimes much more broadly to the long national-protestant
184
7. 8.
9. 10.
11. 12. 13. 14. 15.
16.
Notes sequence beginning with the Henrican Reformation and culminating in 1688. Burke does indeed ‘choose to forget’, and wants his readers to ‘choose to forget’, the contribution of Oliver Cromwell – ‘regicide usurper’ and subduer of the Irish – to the Glorious Revolution of 1688: that is, to remember it, but subliminally. Jon Mee, Romanticism, Enthusiasm, and Regulation: Poetics and the Policing of Culture in the Romantic Period, Oxford: Oxford University Press, 2003, pp. 82–93. These images may strike us as so lurid that they must tell us more about Burke than they can tell us about the revolution. But there is plenty of evidence in the debates, legislative decisions, everyday events and public images of the revolution to suggest that Burke’s gothic extravaganza did anticipate some of the revolution’s effects on French people’s own sense of social – including familial – identity. Lynn Hunt, whose Family Romance of the French Revolution ably explores this aspect of the revolution, provides a nicely appropriate example: ‘As I was completing this book I came across a most unusual piece of information in the holiday gifts section of The New Yorker of 10 December 1990. Under the rubric “On and Off the Avenue” appeared a description of a mechanical toy dating from the French Revolution. The toy is shown on a videocasette titled The Marvellous Toys of Dr Athelstan Spilhaus; it consists of two blacksmiths hammering at the severed head of an aristocrat while another figure tends the forge fire. The toy is dated 1791 and its motto reads: “Ici on reforge les tetes de familles”.’ Lynn Hunt, The Family Romance of the French Revolution, Berkeley: University of California Press, 1989, p. 193. The idea of the body politic and the family politic are indeed joined here, at the neck, where they are most vulnerable to the guillotine. Edmund Leach, ‘The Legitimacy of Soloman’, in Genesis as Myth and Other Essays, London: Cape, 1969, p. 19. For a lucid account of entail, see Barbara English and John Saville, Strict Settlement: A Guide for Historians, Hull: University of Hull Press, 1983. For the literary significance of entail and family settlement in the poetry of Dryden and Pope – for the ways in which the poets linked literary to other forms of inheritance – see Christopher Ricks, ‘Allusion: The Poet as Heir’, in Studies in the Eighteenth Century, III, eds R.F. Brissenden and J.C. Eade, Canberra: Australian National University Press, 1979, pp. 209–40. Wonderfully illuminating as Ricks’s essay is, it does fall in with the poets it describes in ignoring the dependence of patrilineage on wives and mothers. Burke, Paine, Godwin, and the Revolution Controversy, ed. Marilyn Butler, Cambridge: Cambridge University Press, 1984, p. 237. Julian Boyd and Zelda Boyd, ‘The Perfect of Experience’, Studies in Romanticism, 16, 1, Winter 1977, pp. 3–13. William Blackstone, Commentaries on the Laws of England, 4 vols, Oxford: Clarendon Press, 1765–9, vol. 1, 1765, pp. 74, 76. Eagleton, Heathcliff, p. 42. The Correspondence of Edmund Burke, 10 vols, ed. T.W. Copeland, Chicago: University of Chicago Press, 1958–78; vol. 6, ed. A. Cobban and R.A. Smith, 1967, p. 95. See Steven Blakemore, Burke and the Fall of Language: The French Revolution as Linguistic Event, Hanover and London: University Press of New England, 1988. ‘What is at work in this prescriptivist ideology is the political version of Nachträglichkeit, the process by which past experiences may be revised so as
Notes 185 to align them with one’s current psychic state.’ Eagleton, Heathcliffe, p. 43. See also Brooks, Reading for the Plot, pp. 90–112.
4
Watkin Tench and the cold track of narrative
1. H.M. Green, A History of Australian Literature, Pure and Applied, vol. 1, 1789–1923, 2nd edn, revised by Dorothy Green, Sydney: Angus and Robertson, 1984, p. 19. 2. Sydney’s First Four Years, Being a Reprint of ‘A Narrative of the Expedition to Botany Bay’ and ‘An Account of the Settlement at Port Jackson’, by Captain Watkin Tench, ed. L.F. Fitzhardinge, Sydney: Library of Australian History, 1979, p. xxiv. All subsequent references are to this edition. The New South Wales texts are available in paperback as 1788, ed. Tim Flannery, Melbourne: Text Publishing Company, 1996. 3. Alan Atkinson, The Europeans in Australia: A History, Oxford: Oxford University Press, vol. 1, 1997, p. 59. 4. Paul Carter, The Road to Botany Bay: A Study in Spatial History, London: Faber, 1986. 5. John Hawksworth, An Account of the Voyages undertaken by order of the present Monarch for making Discoveries in the Southern Hemisphere, successively performed by Commodore Byron, Captain Wallis, Captain Cartaret, and Captain Cook in the Dolphin, the Swallow, and the Endeavour, drawn up from the Journals which were kept by the several Commanders, and from the papers of Joseph Banks Esq by John Hawksworth, LLD., in 3 vols. Illustrated by cuts, and a great variety of Charts and Maps relating to Countries now first discovered, or hitherto but imperfectly known. London: Strahan and Cadell, 1773. 6. Narrative of the kind I attempt to describe here corresponds to what the French linguist Emile Benveniste influentially described as histoire. Sometimes translated as ‘history’, sometimes as ‘narrative’, histoire is defined by contrast with what Benveniste calls ‘discours’. This distinction, subsequently extended by Gerard Genette, cannot however be applied directly to English texts because French tenses – on the characteristics of which the distinction principally relies – do not correspond to English ones in all respects. I have tried to adapt Benveniste’s important distinction to English grammatical circumstances, stressing the central importance of the exclusion of the present perfect in the construction of narrative. See Emile Benveniste, Problems in General Linguistics (1966), translated by Mary Meek, Coral Gables, Florida: University of Miami Press, 1971, pp. 205–16; Gerard Genette, Figures of Literary Discourse, translated by Alan Sheridan, New York: Columbia University Press, 1982, pp. 127–44. 7. Reference has already been made to Eric Hobsbaum’s formulation of this problem as he approaches the end of his autobiography, Interesting Times. It is appropriate that a historian, used to working in the past tense, should be especially conscious of the problems raised by narrating his own recent past. 8. For an interesting discussion of literary contrivance in the Account, see Adrian Mitchell, ‘Watkin Tench’s Sentimental Enclosures’, Australia and New Zealand Studies in Canada, 3, 1998, pp. 20–35.
186
Notes
9. Le Comte de La Pérouse commanded the French voyage of exploration which put in to Botany Bay a few days after the British in 1788 and subsequently disappeared in the Pacific. 10. Les Oeuvres completes de Voltaire/The Complete Works of Voltaire, vol. 4, Histoire de Charles XII, ed. Gunnar von Proschwitz, Oxford: Voltaire Foundation, 1996, p. 362. Tench’s English version may be his own. 11. Adam Ferguson, An Essay on the History of Civil Society, with an Introduction by Louis Schneider, New Brunswick and London: Transaction Publishers, 1980, p. 184. The discussion that follows owes a great deal to John Barrell’s exploration of these issues. See especially, English Literature in History, 1730–80: An Equal, Wide Survey, London: Hutchinson, 1983, and The Political Theory of Painting from Reynolds to Hazltt: ‘The Body of the Public’, New Haven, Conn.: Yale University Press, 1986. Barrell’s work is particularly relevant to Tench because of the importance which both writers attach to Ferguson’s Essay. The significance of Ferguson for the officers of the 1787 fleet is discussed in Robert Dixon, The Course of Empire: Neo-Classical Culture in New South Wales, 1788–1860, Melbourne: Oxford University Press, 1986, pp. 7–24. 12. Joshua Reynolds, Discourses on Art, ed. Robert R. Wark, 2nd. edn, New Haven, Conn.: Yale University Press, 1975, p. 41.
5
William Godwin: Stories and families
1. Marilyn Butler, ‘Godwin, Burke and Caleb Williams’, Essays in Criticism, xxxii (July 1982), pp. 237–57, p. 242. 2. Gary Kelly, The English Jacobin Novel, 1780–1805, Oxford: Oxford University Press, 1976; Pamela Clemit, The Godwinian Novel: The Rational Fictions of Godwin, Brockden Brown, Mary Shelley, Oxford: Oxford University Press, 1993; Jon Klancher, ‘Godwin and the Genre Reformers: On Necessity and Contingency in Romantic Narrative Theory’, in Romanticism, History and the Possibilities of Genre: Reforming Literature, 1789–1837, ed. Tilottama Rajan and Julia M. Wright, Cambridge: Cambridge University Press, 1998, pp. 21–38. 3. Quoted by Raymond Williams in his entry on ‘family’ in Keywords: A Vocabulary of Culture and Society, London: Fontana/Croom Helm, 1976, revised edn, 1983, pp. 130–4. 4. Things as They Are; or, The Adventures of Caleb Williams, ed. P. Clemit, in Collected Novels and Memoirs of William Godwin, ed. Mark Philp, London: William Pickering, 1992, vol. 3, p. 279. Further references are to this edition, from the 1794 version of the novel unless otherwise stated. 5. Naomi Tadmor, ‘The Concept of the Household-Family in EighteenthCentury England’, Past and Present, 151, May 1996, pp. 111–40 (120); ‘ “Family” and “friend” in Pamela: A Case-Study in the History of the Family in Eighteenth Century England’, Social History, xiv, 1989, pp. 289–306. 6. Anna Kussmaul, Servants in Husbandry in Early Modern England, Cambridge: Cambridge University Press, 1981, p. 7. 7. In Caleb Williams, this sense is largely confined to descriptions of Emily Melvile’s association with the Tyrells where the word ‘family’ plays a particularly important role in helping to obscure the character of her position in the household. ‘The orphan daughter of Mr Tyrell’s paternal aunt’ (p. 282),
Notes 187 Emily is a part of the Tyrell family in two senses (she is a member of the household and a blood-relation of Mr Tyrell) but not in a third sense (she is not part of the Tyrell lineage, not a Tyrell). 8. Jon Klancher argues that ‘the notion of “necessity”, as it will under intense historical pressure, began to produce in Godwin’s revisions of Political Justice – and more fully in his project of cultural enquiry and criticism, The Enquirer – the complex and chaotic actions of “the contingent” ’ (‘Godwin and the Genre Reformers’, p. 27). I would only add that half of the ‘pressure’ came from within Political Justice, in the form of Godwin’s opposition to social contracts: the tension was latent in the conceptual structure of the book from the start. 9. Indeed, the critique of the master–servant relationship in the 1796 Political Justice quite closely echoes part of the critique of the husband–wife relationship in the 1793 edition: ‘It is absurd to expect that the inclinations and wishes of two human beings should coincide through any long period of time. To oblige them to act and to live together is to subject them to some inevitable portion of thwarting, bickering, and unhappiness’ (8. 761). 10. For an argument that the ‘narrative idea of life’ essentially depended on the co-ordination of service and marriage, see Gavin Edwards, ‘Narrative, Rites of Passage and the Early Modern Life-Cycle’, Trivium, 23, 1988, pp. 115–26.
6
Wordsworth’s moving accidents
1. See J. Hillis Miller, ‘The Ethics of Reading: Vast Gaps and Parting Hours’, in American Critcism in the Poststructuralist Age, ed. Ira Konigsberg, Ann Arbor: University of Michigan Press, 1981, pp. 19–41 (32). 2. Stephen Greenblatt, Renaissance Self-Fashioning: From More to Shakespeare, Chicago: Chicago University Press, 1980, pp. 222–54, 234. 3. Robert Graves, cited in Bate, Romantic Imagination, p. 87. 4. Daniel Watkins, A Materialist Critique of English Romantic Drama, Coral Gables: University of Florida Press, 1993, p. 12. 5. One of seven new stanzas, inserted after line 55, in the 1802 edition of Lyrical Ballads; see Lyrical Ballads, eds Brett and Jones, p. 181. 6. Aristotle, Poetics, ed. and trans. Stephen Halliwell, Loeb Classical Library, Cambridge, Mass. and London: Harvard University Press, 1995, p. 37. 7. Lines 85–7, 90–107. Text taken from William Wordsworth, Poems, in Two Volumes, and Other Poems, 1800–1807, ed. Jared Curtis, The Cornell Wordsworth, Ithaca, N.J.: Cornell University Press, 1983. 8. Bate, Romantic Imagination, p. 5. 9. The Prose Works of William Wordsworth, ed. W.J.B. Owen and J.W. Smysor, Harmondsworth: Penguin, 1988, p. 360. 10. James Averill, Wordsworth and the Poetry of Human Suffering, Ithaca: Cornell University Press, 1980, p. 202. Other useful discussions of Wordsworth and Othello include Peter J. Manning ‘Reading Wordsworth’s Revisions: Othello and the Drowned Man’, Studies in Romanticism 22, 1983, pp. 3–28; Susan Wolfson, ‘The Illusion of Mastery: Wordsworth’s Revisions of “The Drowned Man of Esthwaite”, 1799, 1805, 1850’, PMLA 99, 1984, pp. 917–35; Bate, Romantic Imagination, pp. 87–116.
188
Notes
11. Jonathan Wordsworth, Introduction to his edition of William Wordsworth, The Ruined Cottage, The Brothers, Michael, Cambridge: Cambridge University Press, 1985, pp. 1–2. 12. lll. V. 60–65. Text taken from William Wordsworth, The Borderers, ed. Robert Osborn, The Cornell Wordsworth, Ithaca and London: Cornell University Press, 1982. 13. See, from among a considerable literature, Geoffrey Hartman, Wordsworth’s Poetry 1787–1814, New Haven: Yale University Press, 1971; Stephen Parish, The Art of ‘Lyrical Ballads’, Cambridge: Harvard University Press, 1973. 14. The extensive critical discussions of Wordsworth’s interest in ‘accidents’ and his allusions to Othello are limited by a lack of interest in the meaning of the word ‘accident’ itself. Cynthia Chase noted a significant ‘shift of meaning’ between accident as ‘benign chance’ and as ‘chance disaster’, but did not address the broader and historically specific semantic complexities of the word: ‘The Accidents of Disfiguration: Limits to Literal and Rhetorical Reading in Book V of The Prelude’, Studies in Romanticism, 18, Winter 1979, pp. 545–65. Seamus Parry’s illuminating study of Wordsworth’s and Coleridge’s attitudes towards ‘accidence’ would likewise, it seems to me, benefit from a closer focus in the shifting meanings of the key words: ‘Coleridge and Wordsworth: Imagination, Accidence and Inevitability’, in 1800: The New Lyrical Ballads, ed. Seamus Parry and Nicola Trott, London: Palgrave, 2001, pp. 169–93. 15. William Wordsworth, Letters of William and Dorothy Wordsworth, Vol. I, The Early Years, ed. E. de Selincourt, 2nd edn, revised by Chester L. Shaver, Oxford: Oxford University Press, 1967, p. 154. All quotations from the letters are from this edition. 16. The Life and Correspondence of the Late Robert Southey, 6 vols, ed. Robert Cuthbert Southey, London: Longman, Brown, Green and Longman, 1850, vol. 4, p. 118; quoted in Philip Shaw, Waterloo and the Romantic Imagination, London: Palgrave, 2002, p. 96. 17. Johnson on Shakespeare, ed. Arthur Sherbo, The Yale edition of the Works of Samuel Johnson, vol. 7, New Haven and London: Yale University Press, 1958, pp. 65, 75. 18. Lines 264–71, 274–96. Text taken from William Wordsworth, ‘The Ruined Cottage’ and ‘The Pedlar ’, ed. James Butler, The Cornell Wordsworth, Ithaca and London: Cornell University Press, 1979. 19. In the 1800 Preface to Lyrical Ballads, Wordsworth wrote: ‘Aristotle, I have been told, has said, that Poetry is the most Philosophic of all writing.’ The current view is that Wordsworth had some familiarity with Aristotle’s ideas from conversation (probably with Coleridge) and that he had some first hand reading knowledge by 1804. See Duncan Wu, Wordsworth’s Reading, 1770–1799, Cambridge: Cambridge University Press, 1993, pp. 165, 171. See also, Richard W. Clancey, Wordsworth’s Classical Undersong: Education, Rhetoric and Poetic Truth, London: Macmillan, 2000. 20. Don Bialostosky, Making Tales: The Poetics of Wordsworth’s Narrative Experiments, Chicago: University of Chicago Press, 1984, p. 91. 21. First part, lines 258–88. Text taken from William Wordsworth, The Prelude, 1798–1799, ed. Stephen Parrish, The Cornell Wordsworth, Ithaca and London: Cornell University Press, 1977.
Notes 189 22. See Raymond Williams, Modern Tragedy, London: Chatto and Windus, 1966, pp. 13–15. Williams’s critique of conventional thinking about tragedy is very close to Wordsworth’s at this point. 23. As Jonathan Culler says of Oedipus Rex: ‘When the shepherd reveals that Oedipus is in fact the son of Laius, Oedipus leaps to the conclusion, and every reader leaps with him, that he is in fact the murderer of Laius. His conclusion is based not on any new evidence concerning a past deed but on the force of meaning, the interweaving of prophesies and the demands of narrative coherence.’ The Pursuit of Signs: Semiotics, Literature and Deconstruction, London: Routledge and Kegan Paul, 1981, pp. 169–87 (174). 24. Book 4, lines 263, 247–50. Text taken from William Wordsworth, The Prelude, A Parallel Text, ed. J.C. Maxwell, Harmondsworth: Penguin, 1972. 25. Jonathan Arac, review of Averill, Studies in Romanticism, 22, 1, Spring 1983, pp. 136–46 (144).
7
Crabbe’s Parables
1. Some of Crabbe’s sermons are collected in Posthumous Sermons of George Crabbe, ed. J.D. Hastings, London: Hatchard, 1850. 2. Quoted from the Chicago University Library collection of Crabbe’s manuscript sermons, in Peter New, George Crabbe’s Poetry, London: Macmillan, 1976, p. 35. 3. L.J. Swingle, ‘Late Crabbe in Relation to the Augustans and Romantics: The Temporal Labyrinth of his Tales in Verse, 1812,’ English Literary History, 42, 1975, pp. 580–94. 4. Gavin Edwards, George Crabbe’s Poetry on Border Land, pp. 7–18, 77–144. 5. The most relevant discussions of biblical parables, from a considerable literature on the topic, are John Drury, The Parables in the Gospels: History and Allegory, London: SPCK, 1985; J. Hillis Miller, Tropes, Parables, Performatives: Essays in Twentieth Century Literature, Durham, N.C.: Duke University Press, 1991; David Lawton, Faith, Text and History: The Bible in English, Charlottesville: University of Virginia Press, 1991; and Edmund Leach, ‘Against Genres: Are Parables Lights Set in Candlesticks or Put under a Bushell?’, in Structuralist Interpretations of Biblical Myth, eds Edmund Leach and D. Alan Aycock, Cambridge: Cambridge University Press, 1983, pp. 89–112. My reading of the Samaritan proverb is particularly indebted to Lawton, p. 150. 6. For an effect of joining and separation in another of Crabbe’s 1812 tales, ‘The Parting Hour’, see J. Hillis Miller, ‘The Ethics of Reading: Vast Gaps and Parting Hours’, in American Criticism in the Poststructuralist Age, ed. Ira Konigsberg, Ann Arbor: University of Michigan Press, 1981, pp. 19–41. 7. Christopher Ricks, ‘The Music of Time’, Sunday Times, May 23, 1975. 8. Raymond Williams, Keywords, p. 66. 9. The Poetical Works of the Rev. George Crabbe: With his Letters and Journals, and His Life, by His Son, 8 vols, London: John Murray, 1834, vol. 1, p. 198.
8
Relations: Mary Wollstonecraft and Mary Shelley
1. ‘The issues posed by [the narrative structure of Frankenstein] may most of all concern relation, or how narrative relation relates to intersubjective relation,
190
2. 3.
4. 5.
6. 7. 8. 9.
10.
11. 12.
13.
Notes and the relation of relation, in both these senses, to language as the medium of telling and listening, as the medium of transmission, transaction, and transference.’ Peter Brooks, Body Work: Objects of Desire in Modern Narrative, Cambridge and London: Harvard University Press, 1993, p. 200. Gary Kelly, English Fiction of the Romantic Period, London: Longman, 1989, p. 41. William Godwin, Preface, Memoirs of the Author of ‘A Vindication of the Rights of Woman’, London: Johnson, 1798, in Mary Wollstonecraft and William Godwin, ‘A Short residence in Sweden’ and ‘Memoirs of the Author of A Vindication of the Rights of Woman’, Harmondsworth: Penguin, 1987, p. 204. Mary Wollstonecraft, ‘Mary’ and ‘The Wrongs of Woman; or, Maria’, ed. Gary Kelly, Oxford: Worlds Classics, 1976, p. 176. Claudia Johnson, ‘Mary Wollstonecraft’s novels’, in The Cambridge Companion to Mary Wollstonecraft, Cambridge: Cambridge University Press, 2002, p. 207. Barbara Taylor, Mary Wollstonecraft and the Feminist Imagination, Cambridge: Cambridge University Press, 2003, p. 244. Beth Newman, ‘Narratives of Seduction and the Seductions of Narrative: The Frame Structure of Frankenstein’, English Literary History, 1986, pp. 141–64. Chris Baldick, In Frankenstein’s Shadow: Myth, Monstrosity, and NineteenthCentury Writing, Oxford: Clarendon Press, 1987, p. 10. Mary Shelley, Frankenstein; or, The Modern Prometheus (the1818 text), ed. Marilyn Butler, Oxford: Oxford University Press, 1994, Introduction, p. xliv. All quotations from the novel are taken from this edition. It is possible to see the transformation of the metaphoric into the literal as itself obeying a Frankensteinian logic, with metaphors dying and subsequently acquiring a dangerous life of their own by being taken literally. Only Blake among the writers of this period made this connection strongly, for instance in ‘A Poison Tree’ (1794) or in the following passage from The Marriage of Heaven and Hell (1793): ‘The ancient poets animated all sensible objects with gods or geniuses, calling them by the names, and adorning them with the properties, of woods, rivers, mountains, lakes, cities, nations, and whatever their enlarged and numerous senses could perceive. / And particularly they studied the genius of each city and country, placing it under its mental deity. / Till a system was formed, which some took advantage of and enslaved the vulgar by attempting to realise or abstract the mental deities from their objects. Thus began priesthood – choosing forms of worship from poetic tales. / And at length they pronounced that the gods had ordered such things. / Thus men forget that all deities reside in the human breast.’ The Poems of William Blake, ed. W.H. Stevenson and David Erdman, London: Longman, 1971, pp. 111, 212. Shelley’s Prose, ed. D.L. Clark, Albuquerque: University of New Mexico Press, 1986, pp. 185–6. See Frankenstein, ed. Butler, p. xvii. For an analysis of ‘London’ in these terms, see Gavin Edwards, ‘Repeating the Same Dull Round’, in Unnam’d Forms: Blake and Textuality, eds Nelson Hilton and Thomas Vogler, Berkeley: University of California Press, 1986, pp. 26–48. For Wollstonecraft’s earlier life and work as a possible source for Blake’s Visions, see Nelson Hilton, ‘An Original Story’, in Unnam’d Forms, eds Hilton and Vogler, pp. 69–104. Though they occupied overlapping milieux, the
Notes 191 extent of Wollstonecraft’s knowledge of Blake’s poetry has not been established. Nevertheless, the linking, by Oothoon in Visions, of the reductive making of equivalencies by commerce and abstract reasoning (‘Wilt thou take the ape / For thy counsellor, or the dog for a schoolmaster to thy children’) seems to combine in Maria’s mind with the faces seen in London’s ‘charterd streets’ after her husband’s attempt to persuade her to prostitute herself to Mr S: ‘They [his “sophisticated sentiments”] had excited sensations similar to those I have felt, in viewing the squalid inhabitants of some of the lanes and back streets of the metropolis, mortified at being compelled to consider them as my fellow-creatures, as if an ape had claimed kindred with me’ (p. 168).
9
The still unravished bride of Lammermoor
1. Georg Lukacs, The Historical Novel, trans. Hannah and Stanley Mitchell, London: Merlin Press, 1962, p. 34. 2. James Chandler, England in 1819: The Politics of Literary Culture and the Case of Romantic Historicism, Chicago and London: University of Chicago Press, 1998, p. 322. 3. Walter Scott, The Bride of Lammermoor, ed. Fiona Robertson, Oxford: Oxford World’s Classics, 1991, Introduction, pp. viii–ix. 4. Peter Garside, ‘Union and the Bride of Lammermoor’, Studies in Scottish Literature, xix (1984), pp. 72–93. 5. A. Van Gennep, The Rites of Passage, trans. M.B. Vizedom and G.L. Caffee, Chicago: University of Chicago Press, 1960. See also, Gavin Edwards, ‘Narrative, rites of passage and the early modern life-cycle’, Trivium, 23, 1988, pp. 115–26; Edmund Leach, Culture and Communication: The Logic by which Symbols Are Connected, Cambridge: Cambridge University Press, 1976; Roy Rappaport, Ecology, Meaning and Religion, 2nd edn, California: California University Press, 1979; Victor Turner, ‘Social Dramas and Stories about Them’, Critical Inquiry, 7, I, Autumn 1980, pp. 137–64. 6. T.M. Kelley, ‘Keats, Ekphrasis, and History’, in Keats and History, ed. Nicholas Roe, Cambridge: Cambridge University Press, 1995, p. 226. 7. The concept of ‘artisan painter’ has been used by Peter Lord in the context of Welsh art-history. While it describes a broad European phenomenon, Lord argues that it was only in Wales, where there was no indigenous academic painting tradition, that artisan painting become an important means for the creation of a distinctive popular-national consciousness. Peter Lord, The Visual Culture of Wales: Imagining the Nation, Cardiff: University of Wales Press, 2000, pp. 168–244. See also, ‘Welsh Artisan Painters’, in An Oxford Companion to the Romantic Age, pp. 752–4. 8. See Ernst Gombrich, The Image and the Eye: Further Studies in the Psychology of Pictorial Representation, Oxford: Clarendon Press, 1982, pp. 40–62, 78–104. 9. James A.W. Heffernan, ‘Ekphrasis and Representation’, NLH, 2, Spring 1991, pp. 297–316. See also, Lucien Dallenbach, The Mirror in the Text, Cambridge: Polity Press, 1989. 10. ‘If the poet reads the urn as a silent Grecian “historian”, the questioning of his rhyme provides a particularly co-operative voice, for historia is the
192
11.
12.
13. 14. 15. 16.
17.
Notes Grecian method of learning by inquiry. Yet the attempt to double the historians on this occasion produces an ironic counterplay in the language of poetic enquiry. Far from recovering the mysterious legend presumably known to the “sylvan historian”, the speaker’s “rhyme” doubles back on itself to reflect his own perplexities. . . . Keats elaborates the play of rhyme throughout the ode, presenting a speaker in pursuit of interpretation through rhyme that primarily expresses the ardour of the pursuer’. Susan Wolfson, The Questioning Presence: Wordsworth, Keats and the Interogatory Mode in Romantic Poetry, Ithaca and London: Cornell University Press, 1986, p. 319. Ann Curthoys and John Docker, ‘Time, Eternity, Truth, and Death: History as Allegory’, Humanities Research, 1, 1999, pp. 5–25 (p. 13). See also the same authors’ ‘The Two Histories: Metaphor in English Historiographical Writing’, Rethinking History, 1: 3, 1997, pp. 259–73, and Bonnie Smith, ‘Gender and the Practices of Scientific History: The Seminar and Archival Research in the Nineteenth Century’, American Historical Review, 100, 1995, pp. 150–76. In Smith’s account it is the historical archive that is identified as female, whereas Curthoys and Docker focus on historiographical writing in which the female is the past itself. Walter Scott, The Bride of Lammermoor, ed. J.H. Alexander, Edinburgh and New York: Edinburgh University Press and Columbia University Press, 1995, p. 10. All quotations from the novel are taken from this edition, which is based on the first edition of 1819. Reynolds, Discourses, ed. Wark, 1975, pp. 57–8. The Letters of Sir Walter Scott, vol. 5, 1817–1819, ed. H.J.C. Grierson, London: Constable, 1933, p. 350. See Richard C. Sha, The Visual and Verbal Sketch in British Romanticism, Philadelphia, Pa.: University of California Press, 1998. For helpful observations on these aspects of the narrative, see Jina Politi, ‘Narrative and Historical Transformation in The Bride of Lammermoor’, Scottish Literary Journal 15, 1988, pp. 70–81; Jane Millgate, Walter Scott: The Making of the Novelist, Edinburgh: Edinburgh University Press, 1984. On the thematic implications of the differences between the 1819 edition and the 1830 Magnum Opus edition, see Gardside, ‘Union and the Bride of Lammermoor’.
Bibliography
Primary sources Aristotle. Poetics, ed. and trans. Stephen Halliwell, Loeb Classical Library, Cambridge, Mass. and London: Harvard University Press, 1995. Blackstone, William. Commentaries on the Laws of England, Oxford: Clarendon Press, 4 vols, 1765–9. Blake, William. The Poems of William Blake, ed. W.H. Stevenson and David Erdman, London: Longman, 1971. Boswell, James. The Life of Samuel Johnson LLD, ed. R.W. Chapman, London: Oxford University Press, 1961. Burke, Edmund. The Correspondence of Edmund Burke, 10 vols, ed. T.W. Copeland, Chicago: University of Chicago Press, 1958–78. ——. Reflections on the Revolution in France, ed. C.C. O’Brien, Harmondsworth: Penguin, 1969. Crabbe, George. The Poetical Works of the Rev. George Crabbe: With His Letters and Journals, and His Life, by His Son, 8 vols, London: John Murray, 1834. ——. Posthumous Sermons of George Crabbe, ed. J.D. Hastings, London: Hatchard, 1850. ——. George Crabbe: The Complete Poetical Works, eds Norma Dalrymple-Champneys and Arthur Pollard, 3 vols, Oxford: Clarendon Press, 1988. Ferguson, Adam. An Essay on the History of Civil Society, with an Introduction by Louis Schneider, New Brunswick and London: Transaction Publishers, 1980. Godwin, William. Collected Novels and Memoirs of William Godwin, ed. Mark Philp, London: William Pickering, 1992. ——. Political and Philosophical Writings of William Godwin, 7 vols, ed. Mark Philp, London: William Pickering, 1993. Hawksworth, John. An Account of the Voyages undertaken by order of the present Monarch for making Discoveries in the Southern Hemisphere, successively performed by Commodore Byron, Captain Wallis, Captain Cartaret, and Captain Cook in the Dolphin, the Swallow, and the Endeavour, drawn up from the Journals which were kept by the several Commanders, and from the papers of Joseph Banks Esq by John Hawksworth, LLD., in 3 vols. Illustrated by cuts, and a great variety of Charts and Maps relating to Countries now first discovered, or hitherto but imperfectly known. London: Strahan and Cadell, 1773. Hobbes, Thomas. Leviathan; or, The Matter, Forme, and Power of a Commonwealth Ecclesiastical and Civil (1651), ed. Michael Oakshott, New York: Collier, 1966. Johnson, Samuel. The Rambler, eds W.J. Bate and Albrecht B. Strauss, The Yale edition of the Works of Samuel Johnson, vols 3–5, New Haven and London: Yale University Press, 1958. ——. Johnson on Shakespeare, ed. Arthur Sherbo, The Yale edition of the Works of Samuel Johnson, vols 7 and 8, New Haven and London: Yale University Press, 1958. ——. An Account of the Life of Mr. Richard Savage, ed. Clarence Tracy, Oxford: Oxford University Press, 1971. 193
194
Bibliography
Johnson, Samuel and Boswell, James. Samuel Johnson, ‘A Journey to the Western Islands of Scotland’, and James Boswell, ‘The Journal of a Tour to the Hebrides’, ed. R.W. Chapman, Oxford: Oxford University Press, 1970. Keats, John. The Complete Poems, ed. John Barnard, Harmondsworth: Penguin, 1973. Knox, Vicesimus. Elegant Extracts in Prose and Verse, ed. Vicesimus Knox, 3 vols, London, 1791. Paine, Thomas. Rights of Man, ed. Henry Collins, Harmondsworth: Penguin, 1969. Pigott, Charles. A Political Dictionary: Explaining the True Meaning of Words, London: D.I. Eaton, 1795. Reynolds, Joshua. Discourses on Art, ed. Robert R. Wark, 2nd edn, New Haven, Conn.: Yale University Press, 1975. Scott, Walter. The Letters of Sir Walter Scott, 12 vols, ed. H.J.C. Grierson, London: Constable, 1932–57. ——. The Bride of Lammermoor, ed. Fiona Robertson, Oxford: Oxford World’s Classics, 1991. ——. The Bride of Lammermoor, ed. J.H. Alexander, Edinburgh and New York: Edinburgh University Press and Columbia University Press, 1995. Shelley, Mary. Frankenstein; or, The Modern Prometheus (1818 text), ed. Marilyn Butler, Oxford: Oxford University Press, 1994. Shelley, Percy Byssche. Shelley’s Prose, ed. D.L. Clark, Albuquerque: University of New Mexico Press, 1986. Shelley, Percy Byssche. The Poems of Shelley, vol. 2, 1817–1819, eds Kelvin Everest and Goeffrey Matthews, London: Longman, 2000. Southey, Robert. The Life and Correspondence of the Late Robert Southey, 6 vols, ed. Robert Cuthbert Southey, London: Longman, Brown, Green and Longman, 1850. Tench, Watkin. Sydney’s First Four Years, being a reprint of ‘A Narrative of the Expedition to Botany Bay’ and ‘An Account of the Settlement at Port Jackson’, by Captain Watkin Tench, ed. L.F. Fitzhardinge, Sydney: Library of Australian History, 1979. ——. 1788, ed. Tim Flannery, Melbourne: Text Publishing Company, 1996. ——. Letters from Revolutionary France, ed. Gavin Edwards, Cardiff: University of Wales Press, 2001. Voltaire. Les Oeuvres completes de Voltaire/The Complete Works of Voltaire, vol. 4, Histoire de Charles XII, ed. Gunnar von Proschwitz, Oxford: Voltaire Foundation, 1996. Wollstonecraft, Mary. ‘Mary’ and ‘The Wrongs of Woman; or, Maria’, ed. Gary Kelly, Oxford: Worlds Classics, 1976. ——. The Works of Mary Wollstonecraft, eds M. Butler and J. Todd, London: Pickering, 1991. Wollstonecraft, Mary and Godwin, William. Mary Wollstonecraft, ‘A Short residence in Sweden’ and William Godwin, ‘Memoirs of the Author of A Vindication of the Rights of Woman’, Harmondsworth: Penguin, 1987. Wordsworth, William. The Prelude, A Parallel Text, ed. J.C. Maxwell. Harmondsworth: Penguin, 1972. ——. The Prelude, 1798–1799, ed. Stephen Parrish, The Cornell Wordsworth, Ithaca and London: Cornell University Press, 1977. ——. ‘The Ruined Cottage’ and ‘The Pedlar’, ed. James Butler, The Cornell Wordsworth, Ithaca and London: Cornell University Press, 1979. ——. The Borderers, ed. Robert Osborn, The Cornell Wordsworth, Ithaca and London: Cornell University Press, 1982.
Bibliography
195
——. Poems, in Two Volumes, and Other Poems, 1800–1807, ed. Jared Curtis, The Cornell Wordsworth, Ithaca: Cornell University Press, 1983. ——. The Ruined Cottage, The Brothers, Michael, ed. Jonathan Wordsworth, Cambridge: Cambridge University Press 1985. ——. The Prose Works of William Wordsworth, ed. W.J.B. Owen and J.W. Smysor, Harmondsworth: Penguin, 1988. Wordsworth, William and Coleridge, Samuel Taylor. ‘Lyrical Ballads’, the Text of the 1798 Edition with the Additional 1800 Poems and the Prefaces, eds R.L. Brett and A.R. Jones, London: Methuen, 1963. Wordsworth, William and Wordsworth, Dorothy. Letters of William and Dorothy Wordsworth, Vol. I: The Early Years, ed. E. de Selincourt, 2nd edn, revised by Chester L. Shaver, Oxford: Oxford University Press, 1967.
Secondary sources Anderson, Perry. ‘Origins of the Present Crisis’, New Left Review, 23, Jan–Feb, 1964, pp. 26–53. Arac, Jonathan. Review of Averill, ‘Wordsworth and the Poetry of Human Suffering’, Studies in Romanticism, 22, 1, Spring 1983, pp. 136–46. Atkinson, Alan. The Europeans in Australia: A History, Oxford: Oxford University Press, 1997. Ault, Donald. Narrative Unbound: Re-Visioning William Blake’s ‘The Four Zoas’, New York: Station Hill Press, 1987. Austin, J.L. How to Do Things with Words, Oxford: Clarendon Press, 1962. Averill, James. Wordsworth and the Poetry of Human Suffering, Ithaca: Cornell University Press, 1980. Bainbridge, Simon. Napoleon and English Romanticism, Cambridge: Cambridge University Press, 1995. Baldick, Chris. In Frankenstein’s Shadow: Myth, Monstrosity, and Nineteenth-Century Writing, Oxford: Clarendon Press, 1987. Barrell, John. English Literature in History, 1730–80: An Equal, Wide Survey, London: Hutchinson, 1983. ——. The Political Theory of Painting from Reynolds to Hazltt: ‘The Body of the Public’, New Haven, Conn.: Yale University Press, 1986. ——. Imagining the King’s Death: Figurative Treason, Fantasies of Regicide, 1793–6, Oxford: Oxford University Press, 2000. Bate, Jonathan. Shakespeare and the English Romantic Imagination, Oxford: Clarendon Press, 1986. Benjamin, Walter. Illuminations, ed. Hannah Arendt, London: Cape, 1977. Benveniste, Emile. Problems in General Linguistics, trans. Mary Meek, Coral Gables, Fla.: University of Miami Press, 1971. Bhabha, Homi K. Nation and Narration, London and New York: Routledge, 1990. Bialostosky, Don. Making Tales: The Poetics of Wordsworth’s Narrative Experiments, Chicago: University of Chicago Press, 1984. Blakemore, Steven. Burke and the Fall of Language: The French Revolution as Linguistic Event, Hanover and London: University Press of New England, 1988. Borges, Jorge-Luis. Borges: A Reader, ed. E.R. Monegal and A. Reid, New York: Dutton, 1981.
196
Bibliography
Boyd, Julian and Boyd, Zelda. ‘The Perfect of Experience’, Studies in Romanticism, 16, 1, Winter 1977, pp. 3–13. Brooks, Peter. Reading for the Plot: Design and Intention in Narrative, Oxford: Oxford University Press, 1984. ——. Body Work: Objects of Desire in Modern Narrative, Boston: Harvard University Press, 1993. Butler, Marilyn. Romantics, Rebels and Reactionaries, Oxford: Oxford University Press, 1981. ——. ‘Godwin, Burke and Caleb Williams’, Essays in Criticism, xxxii, July 1982, 237–57. ——. Burke, Paine, Godwin, and the Revolution Controversy, ed. Marilyn Butler, Cambridge: Cambridge University Press, 1984. ——. ‘Revolving in Deep Time: The French Revolution as Narrative’, in The French Revolution and English Romanticism: Politics and Rhetoric, eds Keith Hanley and Raman Selden, London: Harvester, 1990, pp. 1–22. Carter, Paul. The Road to Botany Bay: A Study in Spatial History, London: Faber, 1986. Chambers, Ross. Story and Situation: Narrative Seduction and the Power of Fiction, Minneapolis: Minnesota University Press, 1984. Chandler, James. England in 1819: The Politics of Literary Culture and the Case of Romantic Historicism, Chicago and London: University of Chicago Press, 1998. Chase, Cynthia. ‘The Accidents of Disfiguration: Limits to Literal and Rhetorical Reading in Book V of The Prelude’, Studies in Romanticism, 18, Winter 1979, pp. 545–65. Clancey, Richard W. Wordsworth’s Classical Undersong: Education, Rhetoric and Poetic Truth, London: Macmillan, 2000. Clark, Jonathan. Samuel Johnson: Literature, Religion and English Cultural Politics from the Restoration to Romanticism, Cambridge: Cambridge University Press, 1994. Clemit, Pamela. The Godwinian Novel: The Rational Fictions of Godwin, Brockden Brown, Mary Shelley, Oxford: Oxford University Press, 1993. Coleman, Deirdre. ‘Chronicler of Two New Societies’, Southerly, 63, 1, 2003, pp. 206–10. Connolly, Claire. ‘Reflections on the Act of Union’, in Burke’s ‘Reflections on the Revolution in France’: New Interdisciplinary Essays, ed. John Whale, Manchester: Manchester University Press, 2000, pp. 168–92. Culler, Jonathan. The Pursuit of Signs: Semiotics, Literature and Deconstruction, London: Routledge and Kegan Paul, 1981. Curthoys, Ann and Docker, John. ‘The Two Histories: Metaphor in English Historiographical Writing’, Rethinking History, 1, 3, 1997, pp. 259–73. ——. ‘Time, Eternity, Truth, and Death: History as Allegory’, Humanities Research, 1, 1999, pp. 5–25. Dallenbach, Lucien. The Mirror in the Text, Cambridge: Polity Press, 1989. Davie, Donald. The Late Augustans: Longer Poems of the Later Eighteenth Century, ed. Donald Davie, London: Heineman, 1958. Deane, Seamus. Strange Country: Modernity and Nationhood in Irish Writing since 1790, Oxford: Clarendon Press, 1997. Dixon, Robert. The Course of Empire: Neo-Classical Culture in New South Wales, 1788–1860, Melbourne: Oxford University Press, 1986. Drury, John. The Parables in the Gospels: History and Allegory, London: SPCK, 1985. Dummett, Michael. Truth and Other Enigmas, London: Duckworth, 1978.
Bibliography
197
Eagleton, Terry. Heathcliffe and the Great Hunger: Studies in Irish Culture, London: Verso, 1995. Edwards, Gavin. ‘Mind-Forg’d Manacles: A Contribution to the Reading of Blake’s “London” ’, Literature and History, 5, i, Spring 1979, pp. 87–105. ——. ‘Why Are Human Wishes Vain? On reading Samuel Johnson’s Vanity of Human Wishes’, Proceedings of the English Association North, vol. 2, Liverpool: Liverpool University Press, 1986, pp. 52–62. ——. ‘Repeating the Same Dull Round’, in Unnam’d Forms: Blake and Textuality, eds Nelson Hilton and Thomas Vogler, Berkeley: University of California Press, 1986, pp. 26–48. ——. ‘Narrative, Rites of Passage and the Early Modern Life-Cycle’, Trivium, 23, 1988, pp. 115–26. ——. George Crabbe’s Poetry on Border Land, Lampeter and Lewiston: Edwin Mellen Press, 1990. ——. ‘From Chester to Quimper via Sydney: Watkin Tench in Revolutionary France’, Literature and History, 11, 2, 2002, pp. 1–18. English, Barbara and Saville, John. Strict Settlement: A Guide for Historians, Hull: University of Hull Press, 1983. Erskine-Hill, Howard. ‘Literature and the Jacobite Cause: Was There a Rhetoric of Jacobitism?’, in Literature and Jacobitism, ed. Evelyn Cruickshanks, Edinburgh: Edinburgh University Press, 1982. Esterhammer, Angela. The Romantic Performative: Language and Action in British and German Romanticism, Stanford: Stanford University Press, 2000. Everest, Kelvin. Coleridge’s Secret Ministry: The Context of the Conversation Poems, Brighton: Harvester, 1979. Freud, Sigmund. Collected Papers, trans. James Strachey, vol. 28, London: Hogarth Press, 1937. Garside, Peter. ‘Union and the Bride of Lammermoor’, Studies in Scottish Literature, xix, 1984, 72–93. Genette, Gerard. Figures of Literary Discourse, trans. Alan Sheridan, New York: Columbia University Press, 1982. Green, H.M. A History of Australian Literature, Pure and Applied, vol. 1: 1789– 1923, 2nd edn, revised by Dorothy Green, Sydney: Angus and Robertson, 1984. Greenblatt, Stephen. Renaissance Self-Fashioning: From More to Shakespeare, Chicago: Chicago University Press, 1980. Gombrich, Ernst. The Image and the Eye: Further Studies in the Psychology of Pictorial Representation, Oxford: Clarendon Press, 1982. Hartman, Geoffrey. Wordsworth’s Poetry 1787–1814, New Haven: Yale University Press, 1971. Heffernan, James A.W. ‘Ekphrasis and Representation’, NLH, 2, Spring 1991, pp. 297–316. Higgins, Ian. Swift’s Politics: A Study in Disaffection, Oxford: Oxford University Press, 1994. Hilton, Nelson. ‘An Original Story’, in Unnam’d Forms: Blake and Textuality, eds Nelson Hilton and Tom Vogler, Berkeley: University of California Press, 1986, pp. 69–104. Hobsbawm, Eric. Interesting Times: A Twentieth-Century Life, Harmondsworth: Allan Lane, 2002.
198
Bibliography
Hobsbawm, Eric and Ranger, Terence (eds). The Invention of Tradition, Cambridge: Cambridge University Press, 1983. Holmes, Richard. Dr Johnson and Mr Savage, London: Hodder and Stoughton, 1993. Hunt, Lynn. The Family Romance of the French Revolution, Berkeley: University of California Press, 1989. Johnson, Claudia (ed.). ‘Mary Wollstonecraft’s Novels’, in The Cambridge Companion to Mary Wollstonecraft, Cambridge: Cambridge University Press, 2002. Kelly, Gary. The English Jacobin Novel, 1780–1805, Oxford: Oxford University Press, 1976. ——. English Fiction of the Romantic Period, London: Longman, 1989. Kelley, T.M. ‘Keats, Ekphrasis, and History’, in Keats and History, ed. Nicholas Roe, Cambridge: Cambridge University Press, 1995, pp. 212–37. Kermode, Frank. The Sense of an Ending, Oxford: Oxford University Press, 1967. Klancher, Jon. ‘Godwin and the Genre Reformers: On Necessity and Contingency in Romantic Narrative Theory’, in Romanticism, History and the Possibilities of Genre: Reforming Literature, 1789–1837, eds Tilottama Rajan and Julia M. Wright, Cambridge: Cambridge University Press, 1998, pp. 21–38. Kussmaul, Anna. Servants in Husbandry in Early Modern England, Cambridge: Cambridge University Press, 1981. Lacan, Jacques. Ecrits: A Selection, trans. Alan Sheridan, London: Tavistock, 1977. Lawton, David. Faith, Text and History: The Bible in English, Charlottesville: University of Virginia Press, 1991. Leach, Edmund. ‘The Legitimacy of Soloman’, in Genesis as Myth and Other Essays, London: Cape, 1969. ——. Culture and Communication: The Logic by Which Symbols Are Connected, Cambridge: Cambridge University Press, 1976. ——. ‘Against Genres: Are Parables Lights Set in Candlesticks or Put under a Bushell?’, in Structuralist Interpretations of Biblical Myth, eds Edmund Leach and Alan Aycock, Cambridge: Cambridge University Press, 1983, pp. 89–112. Lord, Peter, ‘Welsh Artisan Painters’, in An Oxford Companion to the Romantic Age: British Culture 1776–1832, ed. Iain McCalman, Oxford: Oxford University Press, 1999, pp. 752–4. ——. The Visual Culture of Wales: Imaging the Nation, Cardiff: University of Wales Press, 2000. Lukacs, Georg. The Historical Novel, trans. Hannah and Stanley Mitchell, London: Merlin Press, 1962. Lynch, Deirdre. The Economy of Character: Novels, Market Culture and the Business of Inner Meaning, Chicago: University of Chicago Press, 1998. MacIntyre, Alasdair. After Virtue: A Study in Moral Theory, Notre Dame: University of Notre Dame Press, 1981. McCalman, Iain (ed.). An Oxford Companion to the Romantic Age: British Culture 1776–1832, Oxford: Oxford University Press, 1999. McGann, Jerome. The Beauty of Inflections: Literary Investigations in Historical Method and Theory, Oxford: Clarendon Press, 1985. ——. ‘George Crabbe: Poetry and Truth’, London Review of Books, March 16, 1989, pp. 16–17. Manning, Peter J. ‘Reading Wordsworth’s Revisions: Othello and the Drowned Man’, Studies in Romanticism, 22, 1983, pp. 3–28.
Bibliography
199
Mee, Jon. Romanticism, Enthusiasm, and Regulation: Poetics and the Policing of Culture in the Romantic Period, Oxford: Oxford University Press, 2003. Miller, J. Hillis. ‘The Ethics of Reading: Vast Gaps and Parting Hours’, in American Critcism in the Poststructuralist Age, ed. Ira Konigsberg, Ann Arbor: University of Michigan Press, 1981, pp. 19–41. ——. Tropes, Parables, Performatives: Essays in Twentieth Century Literature, Durham, N.C.: Duke University Press, 1991. Millgate, Jane. Walter Scott: The Making of the Novelist, Edinburgh: Edinburgh University Press, 1984. Mink, Louis. ‘History and Fiction as Modes of Comprehension’, New Literary History, 1, 1970, pp. 541–58. Mitchell, Adrian. ‘Watkin Tench’s Sentimental Enclosures’, Australia and New Zealand Studies in Canada, 3, 1998, pp. 20–35. New, Peter. George Crabbe’s Poetry, London: Macmillan, 1976. Newman, Beth. ‘Narratives of Seduction and the Seductions of Narrative: The Frame Structure of Frankenstein’, English Literary History, 1986, pp. 141–64. Nietzsche, Friedrich. ‘The Birth of Tragedy’ and ‘The Genealogy of Morals’, trans. F. Golffing, N.Y.: Doubleday, 1956. Nussbaum, Felicity A. Torrid Zones: Maternity, Sexuality, and Empire in Eighteenth Century English Narratives, Baltimore and London: Johns Hopkins University Press, 1995. Parish, Stephen. The Art of ‘Lyrical Ballads’, Cambridge: Harvard University Press, 1973. Parry, Seamus. ‘Coleridge and Wordsworth: Imagination, Accidence and Inevitability’, in 1800: The New Lyrical Ballads, eds Seamus Parry and Nicola Trott, London: Palgrave, 2001, pp. 169–93. Pateman, Carole. The Sexual Contract, Cambridge: Polity Press, 1989. Philp, Mark. ‘English Republicanism in the 1790s’, Journal of Political Philosophy, 6, 3, September 1998, pp. 235–62. ——. ‘Revolution’, in An Oxford Companion to the Romantic Age: British Culture, 1776–1832, ed. Iain McCalman, Oxford: Oxford University Press, 1999, pp. 17–25. Pocock, J.G.A. ‘Post-Puritan England and the Problem of Enlightenment’, in Culture and Politics from Puritanism to the Enlightenment, ed. P. Zagorin, Berkeley: University of California Press, 1980, pp. 91–111. ——. ‘Conservative Enlightenment and Democratic Revolutions: The American and French Cases in British Perspective’, Government and Opposition, xxiv, 1989, pp. 81–105. Politi, Jina. ‘Narrative and Historical Transformation in The Bride of Lammermoor’, Scottish Literary Journal, 15, 1988, pp. 70–81. Rappaport Roy. Ecology, Meaning and Religion, 2nd edn, California: California University Press, 1979. Rée, Jonathan. ‘Bound to Be in the Wrong’, London Review of Books, January 20, 2005, pp. 20–2. Ricks, Christopher. ‘Allusion: The Poet as Heir’, in Studies in the Eighteenth Century III, Papers presented at the Third David Nichol Smith Memorial Seminar, Canberra 1973, eds R.F. Brissenden and J.C. Eade, Canberra: Australian National University Press, 1979, pp. 209–40. ——. ‘The Music of Time’, Sunday Times, May 23, 1975.
200
Bibliography
Sartre, Jean-Paul. Nausea, trans. Robert Baldrick, Harmondsworth: Penguin, 1965. Sha, Richard C. The Visual and Verbal Sketch in British Romanticism, Philadelphia, Pa.: University of California Press, 1998. Shaw, Philip. Waterloo and the Romantic Imagination, London: Palgrave, 2002. Simpson, David. The Academic Postmodern and the Rule of Literature: A Report on Half-Knowledge, Chicago and London: University of Chicago Press, 1995. Smith, Bonnie. ‘Gender and the Practices of Scientific History: The Seminar and Archival Research in the Nineteenth Century’, American Historical Review, 100, 1995, pp. 150–76. Strawson, Galen. ‘Tales of the Unexpected’, Guardian Review, January 10, 2004, p. 15. Swingle, L.J. ‘Late Crabbe in Relation to the Augustans and Romantics: The Temporal Labyrinth of his Tales in Verse, 1812’, English Literary History, 42, 1975, pp. 580–94. Tadmor, Naomi. ‘ “Family” and “friend” in Pamela: A Case-Study in the History of the Family in Eighteenth Century England’, Social History, xiv, 1989, pp. 289–306. ——. ‘The Concept of the Household-Family in Eighteenth Century England’, Past and Present, 151, May 1996, pp. 111–40. Taylor, Barbara. Mary Wollstonecraft and the Feminist Imagination, Cambridge: Cambridge University Press, 2003. Tracy, Clarence. The Artificial Bastard: A Biography of Richard Savage, Toronto: University of Toronto Press, 1955. Turner, Victor. ‘Social Dramas and Stories About Them’, Critical Inquiry, 7, I, Autumn 1980, pp. 137–64. Van Gennep, A. The Rites of Passage, trans. M.B. Vizedom and G.L. Caffee, Chicago: University of Chicago Press, 1960. Wagner, Corinna. ‘Symbolic Performances, Public Contests: Negotiating Politics in the 1790s’, British Association of Romantic Studies Bulletin and Review, 23, March 2003, pp. 11–14. Watkins, Daniel. A Materialist Critique of English Romantic Drama, Coral Gables: University of Florida Press, 1993. White, Hayden. The Content of the Form: Narrative Discourse and Historical Representation, Baltimore: Johns Hopkins University Press, 1987. Williams, Raymond. Modern Tragedy, London: Chatto and Windus, 1966. ——. Keywords: A Vocabulary of Culture and Society, London: Fontana/Croom Helm, revised edn, 1983. Wolfson, Susan. ‘The Illusion of Mastery: Wordsworth’s Revisions of “The Drowned Man of Esthwaite”, 1799, 1805, 1850’, PMLA 99, 1984, pp. 917–35. ——. The Questioning Presence: Wordsworth, Keats and the Interogatory Mode in Romantic Poetry, Ithaca and London: Cornell University Press, 1986. Wu, Duncan. Wordsworth’s Reading, 1770–1799, Cambridge: Cambridge University Press, 1993.
Index
accident, 5, 18, 58, 101, 102, 105, 106, 107–22, 127 Act of Union (1707), 160, 173, 177 Allen, William, 167–8 American War of Independence, 11 analepsis, 21, 22, 26 see also prolepsis Anderson, Perry, 36, 40 Anglican Church, 19, 35, 39, 57 Annals of the Fine Arts, 162 annulment, 28–9, 30, 35, 52, 53 anticipated retrospection, 9, 10, 20, 36, 51, 94, 96, 170 see also anticipation; retrospection anticipation, 9, 10, 16, 17, 36, 50, 51, 96, 170 see also anticipated retrospection; retrospection Arac, Jonathan, 122 Aristotle, 102, 104, 105, 107, 110–11, 113, 114, 120, 121, 167 Poetics, 110, 111, 114 Atkinson, Alan, 60, 74 Europeans in Australia, The, 60 Austen, Jane, 175–6 Emma, 162, 175–6 Pride and Prejudice, 44 Austin, J.L., 126 How to do things with words, 19 ‘speech act’ theory, 19; see also performatives, performativity Australia, 11, 56 autobiography, 7, 14, 17, 21, 23–4, 64, 99, 102, 139, 140, 141–2, 150–2, 158 Averill, James, 106, 107, 109, 115 Baldick, Chris, 144, 147, 148 Baruel, Abbe, 147 Bate, Jonathan, 105, 106, 116, 122
beginnings, 8, 9, 10, 14, 15, 16, 17, 18, 35, 42–3, 45, 46, 47–8, 51, 52, 54, 60, 64, 67, 96, 111, 148, 149, 157, 158, 160, 161 Bialostosky, Don, 116 biography, 7, 15, 16, 17, 21, 22–3, 66, 89, 140, 152 birth, 6, 8, 9, 17, 21, 23, 25, 27–8, 42, 54, 89, 140, 148 Blackstone, William, 47 Commentaries on the Laws of England, 47 Blake, William, 9, 57, 79, 154, 155, 157 ‘London’, 154 Visions of the Daughters of Albion, 79, 154 Borges, Jorge-Luis, 30, 31 ‘Other Death, The’, 29, 30, 53 Boswell, James, 21, 22, 25, 82, 88, 89 Life of Samuel Johnson, The, 182 Botany Bay, 11, 60, 61, 62, 80 Brest, 11, 13, 57 Britain, 8, 10–11, 56, 57, 58, 65, 72, 73, 117, 162 Britain, Battle of, 11 Brittany, 11, 57 Brontë, Emily Wuthering Heights, 21, 23 Brooks, Peter, 10, 94, 142, 144 Buccleugh, Duke of, 167 Burger, Gottfried, 116–17 ‘Wilde Jager, Der’, 116 Burke, Edmund, 8, 9–10, 11, 13, 15, 35–55, 56, 57, 67, 81, 86, 95–6, 122, 148, 149, 150, 161, 165, 178 Reflections on the Revolution in France, 8–9, 10, 11, 13, 15, 35–55, 57, 67, 86, 95–6, 148, 149, 159, 165, 178 Burke, Richard, 41 Burney, Charles, 40 201
202
Index
Burney, Fanny, 40 Butler, Marilyn, 10, 45, 144, 151
Culloden, Battle of, 8, 11, 31 Curthois, Anne, 163
Caesar, Gaius Julius, 58, 62, 102 Gallic Wars, 12, 62 Carter, Paul, 17, 60–2, 65, 68, 70, 80 Road to Botany Bay, The: A Study in Spatial History, 17, 60–2 Chambers, Ross, 19, 102, 126 Story and Situation: Narrative Seduction and the Power of Fiction, 18–19, 102 character, 5, 7, 22, 26, 82, 88–9, 90, 91, 92–3, 94, 95, 97–8, 105, 107, 120, 137, 140, 142, 150, 151, 157, 163 Charles I, 53, 164 chronology, chronological order, 21, 22, 23, 24–5, 59, 62, 67, 79 Civil War, 36, 37, 38–9, 40, 41, 53, 55 class, 11, 36, 37, 38, 39, 79, 87, 89, 90, 98, 128, 129, 130–2, 136–7, 138, 144, 145, 147, 163, 177 Clemit, Pamela, 81 Coleman, Deirdre, 13 Coleridge, Samuel Taylor, 51, 113, 114, 116, 122, 178 colony, colonization, 11, 12, 47, 56, 59, 60, 65, 71, 72, 73, 75, 78, 145 see also New South Wales contracts, 5–6, 11, 15, 18, 19, 28, 29, 36, 45–6, 49, 52, 95–7, 98, 99, 102, 103, 105, 142–3, 150–1, 152–6, 159, 160, 176 Cook, James, 62 Journals, 17 counter-revolutionary wars, 10 Crabbe, George, 18, 19–20, 101–2, 123–38, 144, 147, 152 ‘Ellen Orford’, 19–20, 127 ‘Parish Register, The’, 19, 126–7 ‘Parting Hour, The’, 101–2 ‘Procrastination’, 124, 125, 128, 131, 133–6, 137–8 Tales (1812), 124 ‘Widow’s Tale, The’, 124, 126, 127, 128–32, 133, 134, 137, 147, 152 Cromwell, Oliver, 41, 47, 53 Crosby, Benjmain, 84
Davie, Donald, 22 death, 6, 8, 9, 17, 20, 21, 23, 25, 29, 42, 89, 98, 140, 148 deferred action, see Freud, Sigmund, nachträglichkeit Dickens, Charles Great Expectations, 27 dissenters, 37, 38, 39, 41, 49, 51–2 Docker, John, 163 Eagleton, Terry, 48, 50, 52 Edinburgh, 162 ekphrasis, 162 Eliot, George Felix Holt, the Radical, 84 empire, 60, 61, 79 see also narrative, cause-and-effect empirical (imperial) narrative ends, endings, 8, 9, 10, 14, 15, 16, 17, 18, 35, 42–3, 46, 47, 51, 52, 64, 65, 66, 67, 94–5, 96, 111, 148, 149, 157, 158, 160, 161 England, 36, 37, 47, 59, 65, 125 enlightenment, 6, 7, 9, 22, 57, 73, 147 entail, 9, 10, 43–5, 49, 50, 51, 96 enthusiasm, 8, 37, 40, 41 fabula, 23, 24, 26, 27 see also sjuzhet family, 5, 42, 43, 44, 48, 81–99, 139, 143, 144, 145, 149 see also household; lineage; patrilineage family romance, see Freud, Sigmund Ferguson, Adam, 73–4, 76, 78 Essay on the History of Civil Society, 73–4, 78 ‘First Fleet’, 60, 61 Fitzhardinge, L.F., 56, 62, 70 Sydney’s First Four Years, 62, 70 Flannery, Tim, 70 France, 10, 11, 12, 36, 37, 38, 40, 44, 57, 103, 113 French revolution, 8–9, 10–11, 12, 15, 35, 36–7, 38–42, 44, 46, 48–9, 51, 52–3, 57, 58, 59, 60, 72–3, 122
Index 203 Freud, Sigmund, 25, 47–8 Beyond the Pleasure Principle, 25 ‘family romance, the’, 27 nachträglichkeit, 52–3 psychoanalysis, 122 Genesis, 124, 150, 151 Genette, Gerard, 167 gentle, 4, 5, 161 gentleman, gentlewoman, 4, 37, 74–5, 76 Gibbon, Edward, 40 Glorious Revolution (1688), 35, 36, 37, 38–9, 49, 50, 51, 54 Godwin, William, 4, 5, 14, 18, 81–99, 105, 107, 110, 128, 139–41, 143, 152, 155, 157 Enquiry Concerning Political Justice, An, 4, 81, 82–3, 85, 86, 93–4, 95–7, 137 Memoirs of the Author of the Vindication of the Rights of Woman, A, 140 Things as They Are; or, the Adventures of Caleb Williams, 14, 81, 82–3, 84–5, 86–8, 89–93, 94–5, 97, 98–9, 101, 137, 139, 140, 141, 142–3 Goethe, Johann von Sorrows of Werther, The, 146 Gordon Riots (1780), 40, 52 Goslar, 116 Grasmere, 117 Green, H.M., 56 History of Australian Literature, A, 56 Greenblatt, Stephen, 102, 104 narrative self-fashioning, 102 Hanoverian line, 31, 32 Hawksworth, Sir John, 62 Account of the Voyages Undertaken . . . in the Southern Hemisphere, 62 Heffernan, James, 162 Hegel, Georg, 143 histoire, 167, 169 historical novels, 15, 67, 159–62, 163–7, 169–70 historicism, 15, 118, 159
history, 7, 15, 20, 21, 22, 23, 50, 52, 53, 66, 82, 90, 92, 118, 127, 128, 151, 161, 162–3, 165, 167 history painting, 160, 161–2, 163–9, 170, 174 Hobbes, Thomas, 31 Leviathan, 31 Hobsbawm, Eric, 17 household, 44, 85–6, 87, 88, 96, 143 see also family illegitimacy, illegitimation, 29–30, 52 see also legitimacy, legitimate illocution, illocutionary, 19, 52, 125, 127, 139, 140, 141–2, 152 imitation, 104, 105, 111, 120, 144 see also mimesis ‘invention of tradition, the’, 16–17 Ireland, 8, 11, 37, 38, 41, 47, 53 jacobins, 94, 147 jacobites, jacobitism, 8, 31–2, 39, 53, 54, 55, 172 James I, 47 James II, 38, 54 Johnson, Claudia, 141 Johnson, Joseph, 57 Johnson, Samuel, 7–9, 21–32, 35, 36, 39, 50, 52, 53, 82, 88, 105, 107, 109, 111, 140, 144 Account of the Life of Mr Richard Savage, 7–8, 21, 22–8, 29–32, 35, 39, 52, 53, 57, 66 Dictionary of the English Language, 85, 86, 107, 109 Preface to Shakespeare, 111 Tour of the Western Islands of Scotland, 62 journals, 12, 46, 56, 57, 58, 60, 61, 62–4, 66–9, 73, 77, 78, 79, 85, 161, 168 Keats, John, 51, 159, 160, 162–3, 176 ‘Ode on a Grecian Urn’, 159, 160, 162–3, 176 Kelley, T.M., 160, 161 Kelly, Gary, 81, 94, 139–40, 141, 149–50, 152 Kermode, Frank, 15
204
Index
Kierkegaard, Søren, 18 Klancher, Jon, 81 Knox, Vicesimus Elegant Extracts in Verse and Prose, 101, 102, 105–6, 120 Kussmaul, Anna, 85 Leach, Edmund, 42 legitimacy, legitimate, 29–31, 48–9, 53, 54 see also illegitimacy, illegitimation Lessing, G.E. Laoccoon, 161 letters, 11–13, 37, 57, 58, 59, 61, 168 Levi-Strauss, Claude, 42, 44 liberal arts, 73, 74–6, 78, 128 life, 15–20, 22, 47–8, 89, 92, 101, 105, 110, 111, 127, 141, 146, 150 life-stories, 15, 124, 127, 128, 129, 142, 150–1, 153, 157 lineage, 44, 82, 86, 96 see also family; patrilineage Locke, John, 31 London, 12, 13, 56, 57, 59, 61, 64, 65, 162 lyric, 15, 46, 106, 107, 159, 162–3 McCalman, Iain, 40 Macclesfield, Anne Countess of, 24, 25, 26, 27–8, 30, 31, 52, 53 Macclesfield, Earl of, 24, 27–8, 30 MacIntyre, Alasdair, 6–7, 17, 22, 88–90, 140 Magna Carta, 43, 45, 46–7 Malthus, Thomas, 44, 86 Essay on the Principle of Population, An, 44, 86 master/servant relationship, 82, 85, 86–8, 97, 98, 99, 136, 137, 143 Mathews, William, 107, 119 mechanical arts, 72, 73–4, 75, 76, 77, 78 Mee, Jon, 41 memoir, 14, 90, 99, 155, 157, 158 men, 85, 86, 99, 143, 144, 156, 158, 175 men and women, 11, 99, 143, 144, 148, 150, 153–4, 158, 175–6, 177 see also women Mercer, Captain Thomas, 47, 48
middles, mediations, 35, 42–3, 111, 160 Mill, James, 82, 87 Milton, John, 105, 175 Paradise Lost, 146–7, 150, 151, 168, 175 mimesis, 104, 105, 111 see also imitation Mink, Louis, 16, 17, 18 nachträglichkeit, see Freud, Sigmund Napoleonic wars, 10, 11 narrative cause-and-effect empirical (imperial) narrative, 61–2, 70, 71, 79 frame narrative, 78, 142, 152, 160 historical narrative, 64, 67, 71, 79, 92, 118 narrative and family, 81–99, 143, 147–50 narrative and letters, 12–13, 168 narrative idea of life, 6–8, 9, 21, 22, 57, 88–9, 94, 97, 105, 110, 140, 148, 149, 159 narrative order, 5, 6–10, 11, 12, 14–15, 19, 22, 29, 32, 35, 46, 57, 58, 96, 159, 160, 169, 171, 172 narrative painting, 161–78; see also history painting narrative power, 18–19, 90, 101–4, 102, 129–30, 139, 141–2, 144, 152 narrative scepticism, 4–5, 8–9, 19–20, 71, 128 and passim; see also autobiography; biography; historical novels; memoir; parables nation and narration, 17, 60 necessity, 93–4, 95, 97, 110, 111, 120, 153 New Holland, 56 New South Wales, 9, 11, 12, 56–7, 58–9, 60, 62, 64, 65, 66, 72, 73, 76, 79 Newman, Beth, 142 Nietzsche, Friedrich, 48 Nordhaven, 113 ‘order of Time, the’, 7, 21–32, 35, 52 Orwell, George, 26
Index 205 Paine, Thomas, 150, 154 Rights of Man, 4 Palestine, 125 parables, 123–38, 146 Paton, Samuel, 63, 64 patrilineage, 31, 44, 48, 150 see also family; lineage performatives, performativity, 19, 28, 29, 49, 102, 126, 127, 154, 155, 157 see also Austin, J.L., ‘speech act’ theory peripeteia, 27, 66, 71 Peters, Reverend Hugh, 39, 41 Pigott, Charles Political Dictionary, 5 plot, 10, 20, 66, 94, 127, 128, 160 Plutarch Lives’, 146 Poe, Edgar Allan, 44 Pope, Alexander Dunciad, 31 Port Jackson, 60, 61 Portsmouth, 59 prescription, prescriptionism, 48–9, 50–1, 52, 53–4 Price, Dr Richard, 38–9, 41, 49, 53 Discourse on the Love of Our Country, A, 38 prolepsis, 21 see also analepsis promises, 5–6, 13, 17–18, 28, 36, 41, 46, 49, 50–1, 95–7, 142–3, 152–3, 157, 158, 159, 160, 176 Quarterly, 162 Quimper, 13, 57 Racedown, 107 Rée, Jonathan, 16 regicide, 8, 39, 40, 41, 42, 159 Restoration (1660), 31 retrospection, 9, 10, 36, 51, 71, 78, 94, 96, 170 see also anticipation; anticipated retrospection revolution, 5, 8, 9–15, 35–42, 44, 49, 53, 54, 57, 72 see also French revolution; Glorious Revolution (1688)
Revolution Society, 38, 39, 51–2 revolutionary wars, 11, 57, 78 Reynolds, Joshua, 75–6, 78, 79, 80, 161, 162, 165–7, 168–9, 174, 176 Discourses on Art, 75–6, 78, 79, 161, 162, 165–6, 168–9 Ricks, Christopher, 135 rites (of passage), ritual, 19, 105, 126, 154, 160, 162, 169, 170–1, 172–4, 177–8 Rivers, Earl, 24, 27, 28 Robertson, Fiona, 160 Roland, Marie Jeanne, 149–50 romanticism, 106, 168 Russian formalism, 23 Sartre, Jean-Paul, 15–16, 17–18, 23, 48, 50, 60 Nausea, 15–16, 17, 23–4, 60 Savage, Richard, 21, 23, 24, 25–6, 27, 30, 31, 52 see also Johnson, Samuel, Account of the Life of Mr Richard Savage Scotland, 118, 160, 166 Scott, Sir Walter, 5, 10, 11, 31, 67, 117, 118, 120, 121, 159–78 Bride of Lammermoor, The, 15, 159–62, 163–78 Old Mortality, 26, 167 Waverley, 67–8, 118 Shakespeare, William, 18, 19, 105, 106, 107, 109, 110, 111, 112, 113, 114, 116, 117, 118, 119–20, 121, 122 As You Like It, 105 Hamlet, 105, 116, 174 Othello, 12, 18–19, 58, 89, 100–3, 104, 105, 107, 109, 111, 112, 113, 114, 116, 117, 118, 120, 122, 125, 127, 152 Tempest, The, 48, 111 Shelley, Mary, née Wollstonecraft (previous name), 14, 15, 139–58 Frankenstein, or the Modern Prometheus, 15, 139, 140, 141–7, 148, 150–1 Shelley, Percy Bysche, 51, 151 Treatise on Morals, A, 151
206
Index
Simpson, David, 18 sjuzhet, 23, 24, 26, 27 see also fabula sketches, 69, 161, 164–5, 166, 167, 168–9, 170, 172, 174–5, 178 Smith, Bonnie, 163 Sophocles Oedipus Rex, 27, 114 Southey, Robert, 109 Speech Acts, 18, 19, 49–50, 52, 160 Steele, Sir Richard, 25 Sterne, Laurence Tristram Shandy, 158 story, 4, 5–6, 7, 15–20, 22, 81–99, 101, 102, 103, 111, 120, 124, 125, 126, 127, 128, 129, 132, 133, 139, 141, 144, 158, 161–2, 178 storytelling, 4, 10, 11–12, 16, 18–19, 58, 90–2, 102–3, 104, 105, 107, 120, 124, 127, 128, 130, 139, 141, 142, 144, 149, 150, 152, 160, 166 Strawson, Galen, 16 Stuart line, 31, 39, 55 Swift, Jonathan, 39 Tale of Tub, A, 158 Swingle, L.J., 124 Sydney, 12 Tadmor, Naomi, 85, 86 tales, 3–4, 100, 101, 124, 126, 127, 128, 129, 130, 132, 137, 141, 142, 165 Taylor, Barbara, 141 Tench, Watkin, 9, 11–14, 18, 46, 56–80, 102, 105, 140, 161, 167 Account of the Settlement at Port Jackson, An, 9, 12, 14, 57, 58–60, 62–3, 65, 66–7, 69–73, 74, 76–7, 78, 168 Letters Written in France to a Friend in London, 11, 12–13, 57, 58, 59, 62, 102 Narrative of the Expedition to Botany Bay, A, 11, 12, 14, 17, 56–7, 58, 61, 62–6, 67, 69, 70
see also Fitzhardinge, L.F., Sydney’s First Four Years tenses future, 28 past simple, 28, 29–30, 31, 46, 63, 64, 65, 67, 77 present, 28, 29, 65, 67, 77 present perfect, 46, 48, 65, 67, 68, 77, 78 Tories, 8, 31, 32, 39, 172 Trafalgar, Battle of, 118 tragedy, 103, 106, 107, 110–11, 116, 120, 121 Vallon, Annette, 103 Volney, Constantin Ruins of Empire, 146, 150 Voltaire, 73 History of Charles XII, 72 Watt, Isaac Logick, 107 West Indies, 11, 77, 78, 79 whigs, 39, 172 Williams, Raymond, 137 Wilson, John, 108 Wolfson, Susan, 163 Wollstonecraft, Mary, 12, 14, 15, 17, 98, 99, 139–58 Vindication of the Rights of Woman, A, 98 Wrongs of Women, The; or, Maria, 14–15, 17, 84–5, 86, 99, 139–40, 141–3, 147–8, 149–51, 153–58 women, 14, 31, 44, 84, 98, 99, 139, 141, 142, 143, 147–8, 150, 175, 177 see also men and women Wordsworth, Dorothy, 51, 107 Wordsworth, Jonathan, 106–7 Wordsworth, William, 3–6, 9, 10, 18, 19, 46, 51, 58, 67, 91, 100–22, 125, 144, 167 Borderers, The, 103, 106 Essay Upon Epitaphs, 105 ‘Hart-Leap Well’, 106, 115–17 Lyrical Ballads, 3, 4, 5–6, 91, 103, 106, 107, 112–13, 115
Index 207 ‘Ode: Intimations of Immortality’, 104–5 Prelude, The, (1798), 10, 21, 118–19, 121–2, 125 Ruined Cottage, The, 106, 112–13, 114, 115 ‘Ruth’, 103–4
‘Salisbury Plain’, 113–14 ‘Simon Lee’, 3–5, 8, 91, 110 ‘Strange fits of passion I have known’, 46 ‘Thorn, The’, 91 ‘Tintern Abbey’, 10, 46, 51