Tom Kindt

Hans Walter Gabler

Thoughts on Scholarly Editing

A Review Article occasioned by

Paul Eggert, Securing the Past. Conservation in Art, Architecture and Literature. Cambridge: Cambridge University Press 2009. XII, 290 p. [Price: £ 17.99 (paperback); £ 45.00 (hardback)]. ISBN: 9780521725910; 9780521898089.

In Securing the Past, a monograph bracketing a set of analyses of conservation in architecture, art, and literature, Paul Eggert’s interest is in fields of force operative between the poles of origin as creative authorship, on the one hand, and of the cultural techniques of preservation, restoration, and editing, on the other hand. At bottom, he sees these activities as one common enterprise predicated on two essentials. One of them is ›agency‹, the term under which are subsumed and progressively theorized both originating authorship and the re-fashioning, even re-creation, of cultural objects in their historical descent. The other, and concurrent, essential is the materiality of these objects onto which the cultural techniques are expended. Being by profession himself a scholarly editor, it is the editorial predicament that shapes Eggert’s understanding and vision. He is aware of this: »I begin by recognising the categorical difference between editing and restoration. Scholarly editors do not physically alter […] original documents […]. In comparison, conservators of historic houses, paintings and sculptures make changes to the physical objects themselves.« (12) Nonetheless, his declared aim is »to bring the arts of restoration together to examine their linked, underlying philosophies« (9). This interdisciplinary approach, in so far as it applies combinatory thought to diverse practice, does stimulate fresh insights. Yet the book’s further reach towards abstractly theorizing the underlying philosophies is also a source of problems it ultimately leaves us with.

1. Architecture, Art, and Conservation: A Syncretistic Sweep

Chapters 2 and 3 deal with »The witness of historic buildings and the restoration of the churches« and »The new Ruskinians and the new aesthetes«, respectively. Chapter 4 focuses on »Forgery and authenticity: historical documents, literary works and paintings«, and chapter 5 problematizes »Conservators and agency: their role in the work«. Drawing as it does on the study’s antecedent, largely non-textual subject matter, this chapter especially underpins one of the centrally theorizable terms of the book’s over-all argument, ›agency‹. For the present reviewer, being like Eggert a literary critic and a textual scholar, it is hard to do justice to the sweep in these first five chapters of examples, observations and conclusions from the range of heterogeneous, even if comparable and mutually illuminating, cultural objects, as well as of activities over the past two hundred years or so in Europe or in Australia, in the service of securing the past. One feels an urge to bring together, say, a week-long intensive study seminar of restorers, conservators and conservation officials, museum curators, art historians, architects, local and regional politicians, sociologists, even criminologists, copyright lawyers and, indeed, creative artists, set them Eggert’s monograph through chapter 6 as their course text and, from their several vantage points of expertise, have them explore its implications. They would pick up from the book’s innumerable suggestive mentions of such matters as the correspondences between the Gothic revival and the restoration of churches in the 19th century, or the mirroring (or is it falsification) of the past in museums, or the vexed interchangeability of the authentic and the fake, or the perennial human tendency to shape the past in the image of the present, and ramify the book’s subject matter each from out of their specific expertise and knowledge. This could yield an in-depth assessment, beyond the present study’s valiant survey attempt, of how, and in what manifold ways, we, in our day and age, and at our point in history, conceive of securing the past.

Would we wish to have the textual scholar and editor in on such a seminar? It is a nice question. My instinct would be to have one, but to avoid having a sub-team of textual scholars; that is, to have Paul Eggert alone as preliminary key-note speaker and ask him to condense the second half of his book into at most an hour-long paper. This would bring to the fore just the generalisable and societally and culturally most relevant dimensions of textual scholarship and editing, such as they indeed share in the cultural pursuit of securing the past. Thus to engineer once again a judicious division of the realms that Eggert has comprehensively brought together would, owing to what he has in truth accomplished in his book, further contribute to deepening its achievement.

2. The Painter as Author Metaphoricized

So much for a flight of fancy triggered by the first half of Securing the Past. To turn again to the book as it stands. Its second five-chapter sweep sets in still outside the realm of texts. Chapter 6, entitled »Subtilising authorship: Rembrandt, scientific evidence and modern connoisseurship«, begins a trajectory that, ultimately in chapter 10, will culminate in a theorizing of the foundations of textual scholarship. Chapter 6 thematizes authorship in terms of creations in fine art, specifically of paintings by Rembrant/Rembrandt. Against the common-sense awareness of seeing them as painted, or authored, by the historical person whose real existence is amply witnessed and testified to, Eggert traces in valuable detail the activities of authentification and attribution carried out over two generations by the Rembrandt Research Project. To these, he proceeds to apply, by fleeting transfer (or, as must be recognised, by half-transfer), current thought from mainly literary theory towards defining authorship: »›Rembrandt‹ is not, then, the man who lived and painted. […] The term Rembrandt lives in its usages […] it has become an art-critical and curatorial abduction.« (122)

Thus, initially, the argument appears to run in analogy to the Barthes/Foucauldian theorizings of authorship that have become an essential ingredient of modern thought in literary criticism. [1] Roland Barthes’s title, »The Death of the Author«, is all too often (be it wilfully or ignorantly) taken, and misunderstood, literally. In truth, the theoretical position that Michel Foucault’s »What is an Author?« in particular designates is that texts, as works of art in language (written by live authors, of course!), and on account of the communicative vector inbuilt in language, generate an authorship-defining point of perspective from within themselves. This »author function« (in terms of analytical narratology, it might alternatively be called an »author-effect«) fundamentally generated out of language acquires structural as well as interpretative relevance for both the text’s composition and its potential for meaning. It is categorically distinct from the real author, who is and remains always outside the text’s autonomy. Such however, it turns out, is not what Eggert would want us to understand by ›Rembrandt‹ as »an art-critical and curatorial abduction«. For he goes on to claim that, as that abduction, the term »holds things together by its reference – factually, gesturally, wilfully – to the man who lived. The underlying appeal is to an integralness that reflects that of Rembrandt’s body.« (122)

It is not easy to assess the usefulness of such an advancing and again retracting of a theoretical stance for the declared purpose of »subtilising authorship«. In fact, it is perhaps even unwise in the first place to attempt, as Eggert does, to re-theorize authorship at all on the basis of the art of painting. For is ›authorship‹ here not spoken of but metaphorically? It seems doubtful that the limners of paintings can be thought of as authors in the same way that the originators of works of art in language have throughout our cultural tradition been so designated. The categorical distinction between painters (say) and authors arises from the difference in nature of the materials out of which they work: out of line and colour the painter, out of language the author. Of these materials, language is inherently semantic, while line and colour do not bring with them innate meaning. The work of fine art – a painting – comes about by a willed arrangement of its material and sensual elements; and it is by this process rendered representative. By contrast, the work of art in language is brought about by harnessing – by yoking together – elements (words, phrases, structures of grammar and syntax) that always already have cores of meaning. The work in language is consequently at bottom predicated on a pre-existing semantic core and potential for communication in its material substratum and is thus, in essence, not so much representative as communicative.

The harnessing and yoking together of the language material is what we conventionally designate as writing. Empirically, it is true, acts of writing are commonly seen as acts of origination, which of course they are on account of the writer’s intellectual and creative input. Yet the view is indeed empirical, which means that it is not fully buttressed theoretically, since it leaves the innate semantics of language out of the reckoning. The potential of language to mean shapes writing as much as, reciprocally, it is instrumentalised and actualised by it. The origination of a piece of writing amounts therefore to a highly complex process of negotiation of meaning. All the more, it is true, we need (on the one hand) to lean on its empirical originator. For we not only wish to read the written, we also wish its content and meaning to be vouched for. Hence, we rely on the collocators of language, and accept them by convention and cultural agreement as the authors of any formed sets of writing. Yet if it is thus that in real life we gain our notion of ›author‹ and ›authorship‹, it is (on the other hand) also important to note that the designation is not just empirical. As concept, it has theoretical dimensions.

Conceptually, the empirically nameable and placeable originator of the writing whom we term author enacts a role in that triangled negotiation of meaning between him- or herself, the writing (as process and product: call it the text), and the recipient (vulgo, the reader). Being in this manner inscribed in a relational process of generating meaning is what essentially defines the author, and authorship. Empirically, the process constitutes a real-life condition of bringing forth works of art in language, which is something Eggert duly acknowledges at the opening of chapter 9, where pragmatically, by the run of his argument, the observation belongs. To recognise, however, that, with works of art in language whose medium, language, is innately communicative, author and authorship in turn are not just empirically and pragmatically, but in fact essentially inscribed in the generating of meaning, raises the definition of the terms to a systemic level. It is therefore that they can apply at most metaphorically, if at all, to the representative nature of works of fine art. Eggert’s »subtilising« of authorship, then, amounts (as suggested) to a metaphoricizing of the term. In the chapter context, this is useful rhetoric for discussing the problems – whether of a scholarly or a market-place nature – inherent in the authentification of paintings with the ›Rembrant/Rembrandt‹ signature. Without positing the empirical painter-authorship, Eggert would lose »the man who lived« as the originating ›agent‹, active on the same empirical level of reality as the securing agencies serving the ›Rembrandt signature‹ »by abduction« at their due historical stations as restorers, curators, evaluators and scholar art-historians. But Eggert’s retracting again the Barthes/Foucault stance on author/authorship that he yet briefly invokes is not sustainable in terms of theory.

Theoretical gain, by contrast, could be had from following up that fleeting invocation. Sustained (which would mean also: carried through to the book’s concluding theory chapter), it might have led to recognising fully that author/authorship are, conceptually and as terms, tied ineluctably to the realms of writing, and of works of art in language. Approaching writing in terms of its medium and mediality, Barthes/Foucault define author/authorship functionally. The ›author function‹ as inherent in texts, and springing as it does from the semantically communicative nature of language, is conceptualised from out of an ontological understanding of the medium. Thus radically understood, empirical authoring as issuing in writing and texts stands revealed as the real-life spin-off of authorship into the materiality of documents – but equally, we should add, into the immateriality of oral composition and transmission. Such considerations put yet further in doubt the feasibility (feasibility in terms of theory, that is) of applying the term ›authorship‹ to the bringing forth of fine art. The work of the sculptor or painter, and beyond (say) of the architect, is expressed by way of, and thereby always inseparably tied to, its material manifestation in the one unique original that is its outcome. In terms of its crafting by the hand of its originator, it is an autograph. The work of art in language, or indeed any meaningful language collocation, by contrast, does not in essence so exist. It is allographic. The term as coined and used refers, as we know, in the first instance again to the work’s material making, to its being scripted. What this implies, even just pragmatically, is that what is penned or printed in language is copyable without limit in any number of exemplars which all instantiate the work (that is, instantiate the work as text). Since we hardly ever think of works in language other than in scripted instantiation, this, to all appearances, ties ›allographic‹ to material media of reproduction.

But again, this is empirically, yet not theoretically sufficient. For in essence, any meaningful language collocation, and a fortiori any work of art in language, can exist without being recorded in writing, thus without instantiation in script. Were this not so, we would, for example, not be able to claim continuities from oral literature to literature in material transmission, or be able to interpret the full range of causes for the considerable variability of texts in transmissions from before the invention of printing. The circumstance that, in analogy to scripted language formations, oral collocations of language, too – be they laws or decrees, or proverbs, or works of art in language in any number of genres: poems, epics, plays, fables, fairy tales – can exist without script and be transmitted (as, for instance, recited from memory) in unlimited instantiations, helps to recognise that ›allographic‹ designates not merely an accidental attribute (i.e., the ›being scripted‹), but an essence. This distinguishes works in language fundamentally from works of architecture, sculpture, or painting. It means, moreover, (and does so perhaps even to the consternation of textual scholars and critics) that materiality must be thought of as accidental to works in language, and not as substantive and essential to them. From this follows as a further conclusion that for precisely this ontological reason the concepts of ›author‹ and ›authorship‹ must be posited specifically, and in theory exclusively, for application to works, and works of art, in language. Since, as works, they can in principle be instantiated materially or immaterially in unlimited replication, what brackets such allographic instantiation is the systemically functionalised concept of ›author‹ and ›authorship‹.

Admittedly, Eggert hardly intended, and certainly he did not in chapter 6 attempt, to delve into such ulterior theorizing around the terms and concepts of ›author‹ and ›authorship‹. His own already cited positioning of name and author (meaning at the same time ›name as author‹): »hold[ing] things together by […] reference – factually, gesturally, wilfully – to the man who lived«, and so vouchsafing an »integralness […] reflect[ing] that of Rembrandt’s body«, supports rather the chapter’s analysis of the role of scholarship in the service of »modern connoisseurship« (ch. 6, passim). With curatorial and art-historical expertise closely tied in real life to the monetary evaluation of works of art, what is clearly at work, and what Eggert illuminatingly analyses, is what might be termed applied scholarship (notionally analogous to applied science, which, as we know, enjoys both cultural and social acceptance, not least for its economic consequences). As applied scholarship (be it scientifically self-fashioned and autonomous, or else variously time-serving), art history in the twentieth century has assumed a task of mediating the material heritage of art to contemporary expectations and tastes in reception. The need, under market-place pressure, to authenticate Rembrandt paintings has however, as Eggert shows, at the same time, and in terms of knowledge, understanding and method, palpably advanced the scholarly discipline of art history, as well as the curatorial and restorational crafts. A lead might be taken from here to distinguish in future more explicitly between applied and pure humanities scholarship, and to elucidate their distinct agendas, as well as to observe them in interaction.

3. Textual Criticism: Laying the End-of-the-Twentieth-Century Land

With chapters 7 to 10, Eggert enters his native realm of textual criticism and scholarly editing. Chapters 7 to 9, progressively covering case analyses of exemplary editorial situations and modes, increasingly reflect also on their theoretical implications. These, and those similarly following comprehensively from the book’s coverage of subjects, are surveyed in the final chapter 10.

Shakespearean editing used traditionally to be where text-critical and editorial principles and paradigms were established in Anglo-American textual scholarship. This is acknowledged in chapter 7, with due reverence paid to bibliography and copy-text editing, the loadstars of Shakespearean textual criticism throughout most of the 20th century. Yet, headed »Materialist, performance or literary Shakespeare?« as it is, the chapter is nonetheless but tangential to this 20th-century mainstream of textual editing in Great Britain and the United States. It focuses, rather, on the fundamental end-of-the-century upheavals in the sub-discipline which altered, from within, its understanding of itself and which, from without, displaced it from its lead function in Anglo-American textual criticism at large. The displacement resulted from reformed thinking in literary criticism and theory and was, in this respect, energized from out of pure scholarship. While in their fuller scope, these fields of force are mapped out in chapters 8 and 9, the argument is set in motion with the survey in chapter 7 of some main factors that triggered renewed reflections on the textual situation for Shakespeare: sophisticated critical analysis of the plays as performance texts; increased awareness of the history of Shakespearean editing over the centuries as a history of adaptation in minutiae of language, style, or prosody; or the dependence of that history on its material substratum, by which Shakespearean textuality becomes amenable, for instance, both to being analysed in its material manifestations, and to being subjected to materialist literary theory. The emergence is recorded of the Oxford Shakespeare, the 20th century’s main Shakespeare edition worked from the ground up, which appeared in 1986 out of a vortex of all these cross-currents, and reflects them all. As a whole, admittedly, the chapter couldn’t claim to do comprehensive justice to the achievement of 20th-century Shakespearean textual criticism. As acknowledged, it serves mainly as a bridge into, and a preparation for the central argument beginning in chapter 8 around »Modes of editing literary works: conflicts in theory and practice«, and continuing in chapter 9 under the heading »Readers and editors: new directions in scholarly editing«.

To open chapter 8, the conflicting forces at work are panoramically named. They arise from orientations and re-orientations in terms both of understandings of culture and of movements of theory at the end of the 20th century and across the millennium threshold. These in turn affect, as Eggert sees it, concepts of editing as a cultural and scholarly task. Editorial scholarship finds itself under pressure to review its subject-matter as well as its methodologies, to re-justify what it is doing and achieving with, and on behalf of, the material objects it is dealing with (or immaterial objects, for that matter, considering that, for instance, the literary work behind its materially manifest texts may legitimately itself be defined as immaterial – as we have contended above, and shall more fully explicate below). A few general, yet pertinent, definitions of »What an editor does« (156-158) – very usefully containing also a roll-call of the many senses in which the term ›editor‹ is understood, in the first place – lead on to examples both from Australia and the US of how, and with what arguments, scholarly editing is both societally and culturally resisted.

What Eggert adroitly sees as perhaps the weightiest motivation for today’s (and certainly the late 20th-century’s) (Western) societies and cultural environments to resist scholarly editing as an imposition by specialists, is that it complicates straight consumptive reading by opposing the naïve assumption that texts be pure and stable, or the market-place expectation that editions be definitive. The endeavour of securing the past in the field of scholarly editing is, quite to the contrary, nowadays heading in distinctly new directions, with fresh strength gained through textual scholarship re-theorized and reformed. No longer (to pick up Eggert’s sporting-ground metaphor) is the editorial task defined (merely) as »tend[ing] the field properly« and then »let[ing] the [literary] critics get on with the main game.« To the irritation of the cultural as well as the literary critics, instead, »the editors [are now] wanting to expose the textual subsoil« (164) – that is, to reveal the process nature of texts, and thus the interplay of textual stability and instability. Since the notion of ›process‹ thus enters into defining the nature of texts, ›process‹ must pertain also to the nature of authorship – as we have already maintained above in emphasizing the authorial participation in the triangled negotiation of the meaning potential of language by which texts become texts. The answer-in-kind to this understanding of authorship and text must be ways and means for textual scholarship adequately to translate the processual nature of writing and of texts into processual modes of analytically unfolding and presenting texts in editions. This does not eliminate, nor in the day-to-day work of editing marginalise, the traditional task of editions to stay the corruption through error that ineluctably befall transmissions. Yet corruption is only a part-reason for the variability encountered in the materials documenting texts. For it is indeed of the very nature of texts to be variable; hence, their material documents of origin commonly testify amply to variation from processes of revision. Under today’s enlarged understanding of the nature of texts, consequently, it is incumbent on editors not just to establish texts by way of stabilising them against endogenous textual variation (that is, commonly, variation through textual error). A significant challenge arises further from the indigenous, text-immanent, variability and the demands it makes of editors to seek congenial forms of response to them in the shape and communicative potential of editions.

From its outlining of the innovative stance in textual criticism and scholarly editing, the chapter leads on to an in-depth discussion of »Gabler’s Ulysses« (164-168; 173-179), i.e., the Critical and Synoptic Edition of James Joyce’s novel I prepared in the late 1970s and early 1980s and published in three volumes in 1984 (touching it up with a few amendments in 1986, the year that also saw the commercial publication of its reading text only). Eggert’s understanding of the edition’s over-all conception is thorough, and his survey of the debates it sparked is both comprehensive and fair. Following from here through the remainder of chapter 8, and into chapter 9, not only is Eggert’s own highly relevant editorial experience from his participation in the Cambridge UP D.H. Lawrence edition and, above all, his leading role in the manifold activities of literary editing in Australia infused into the discussion. Further samples, too, from recent editorial history are investigated in themselves and in the context of debates they elicited, such as James L. W. West III’s edition of Theodore Dreiser’s Sister Carrie, or J.C.C. Mays’s edition of Coleridge’s Poetical Works – editions, in other words, that were enacted outside, or at most but tangentially to, the Shakespeare-and-Renaissance-engendered editorial paradigm (that is, the Greg-Bowers paradigm, or theory, of copy-text editing).

Eggert knows the ropes of scholarly editing and possesses all the experience and skill needed to file into shape and tighten the requisite nuts and bolts. At the same time, moreover, he opens horizons from which to gain enlightening perspectives on the specialised craft of scholarly editing. These are in one respect theoretical, such as when, for the purpose of exploring the text-constitutive role of reading – that is, the constitutive role of reception for both editors and readers; as, indeed, for authors in respect of their own texts-in-process – the factor of textual meaning is brought into play to buttress the significance of scholarly editing for securing the past. In another respect, the horizon is enlarged in directions of methodology. Here, in particular, the »German Encounter« (203-212) is focussed on in chapter 9 and the unaccustomed elements, even alternative systematics, of German textual scholarship in contrast to the customary paradigms in Anglo-American text-critical thought and practice are laid out at length. In terms of the book’s disposition, this follows from its highlighting of both »Gabler’s Ulysses« and J.C.C. Mays’s edition of Coleridge’s Poetical Works that in different, and in a sense complementary, ways result from a fusion of Anglo-American and German editorial thinking. The German way in textual criticism and scholarly editing is thus impressively critiqued – a feat nowhere that I know of accomplished in English so comprehensively and with such understanding as here.

4. Implicating Meaning

Chapter 10 attempts to draw the theorizable sum of the preceding chapter discussions. Headed »The editorial gaze and the nature of the work«, and following on from the intense engagement with scholarly text editing through chapters 7 to 9, this concluding chapter patently contends that all active investment into securing the past, whether in architecture, or the fine arts, or the wide (and, indeed, much variegated) areas of textual transmission may be, and should be subsumed under the common denominator of ›editing‹. To enhance the chapter’s claim to anchoring the monograph as a whole in theory, Eggert begins by citing René Wellek and Austin Warren’s Theory of Literature and its (mainly) inner-American responses. These however (e.g., E.D. Hirsch’s Validity in Interpretation) would be incompletely understood without their backgrounds in European thought. Therefore, the chapter proceeds to draw in, successively, philosophical positions from Europe of the 1930s, Edmund Husserl and Roman Ingarden (phenomenology and the notion of the ideal text), as well as Max Heidegger (»The Origin of the Work of Art«), to which Jacques Derrida and French post-structuralism in turn can be identified as having reacted in the post-war period. Thence, an »Anglo-American Editorial Scene« (227-231), hovering between pragmatism and theory (and tied here to the names of John McLaverty and Peter Shillingsburg), is briefly sketched out before the survey of phil­osophical positions is rounded off with a scenario for future orientation in editorial thinking, decisively at the same time tied back to the philosophies of C.S. Peirce and Theodor W. Adorno. Taken together, the positions in philosophy cited serve to theorize the concept of ›the work‹. What the chapter is made to bear out, and what the book as a whole claims, is that it is the work (from the past) that centrally demands securing. To this end, so the argument goes, the work must be subjected to the »editorial gaze«. For this concluding theory chapter, furthermore, the editorial gaze is now insistently trained on the work in terms of what it (and, with regard to the work in language, what its text – or is it: its texts?) mean. The philosophical positions adduced are all concerned with questions of meaning – and, overwhelmingly so, with the meaning of artefacts (works) in language. And here lies the rub.

Eggert gains a heuristic definition of ›work‹ from setting the lexical term in English (identical as noun and as verb) against its apparent equivalents in German, French, Italian, Spanish, and Russian (where the respective terms are nouns only). »Getting a grip on the concept is notoriously difficult in whatever language.« So he contends in arguing the need to test the concept of ›work‹ against his philosophical tour-d’horizon, for the benefit of editors and conservators engaged in »cultural heritage conservation or scholarly editing.« (214) So centred in text-critical and editorial thinking is the ensuing discussion that it seems justified to meet it on the same ground.

To contend that an editor edits a work appears plausible enough, on the face of it. A closer look into the usages across languages, however, will soon reveal that in German, for instance, to edit ein Werk, while it may indicate the editing of a single work, yet conventionally signifies editing the works, that is: the oeuvre, of an author. The Scandinavian languages, taking this notion one step further, speak of editing ett författarskap (the Swedish variant of the term), that is ›an authorship‹, i.e., roughly again an oeuvre. So made aware, we recall of course immediately that, in the anglophone environment, one will quite commonly speak of editing Shakespeare, or Milton, or Keats, or Wordsworth – or D.H. Lawrence. The two-fold potential of signification of the noun ›work‹ as ›individual work‹ or ›oeuvre‹, or the metonymic exchangeability of work and author, are thus not absent from English, either.

The situation this points to is analogous to, and in a sense repeats what we discussed above with respect to author/authorship. Neither these terms, nor the term ›work‹ can – pace Eggert – be applied with identical signification and coincident implications to restoration in the fine arts, or architecture, on the one hand, and to the editing of transmissions in language on the other hand. A fundamental distinction instead must be made, one that Eggert does not consider: in restoring works of the fine arts, or architecture, there can never be any going-behind their material existence and presence, meaning also: their existence as presence. Editing works (of art) in language, by contrast, can never be accomplished without a preliminary, yet foundational going behind the extant textual materials.

If there has been one constant fundamental to editing throughout its history since antiquity, it has been both the need and the practice to go behind the texts witnessed in material documents in order to elicit edited texts. Materially extant texts have ever been deemed flawed. The cultural technique of editing was consequently invented to mend their deficiency, and the main goal with edited texts has been to invest them with, and in, a new materiality differing from that of all antecedent text materialisations, on the basis of which they could be, and were, established. Great efforts, indeed, were undertaken to contain the extant instantiations of texts-to-be-edited in a systematized methodology supporting the assumption that, and defining the ways in which they related. Going behind the materially extant instantiations, too, into their lost, hence no longer material ancestry, led by dint of method to such logical constructs as archetypes, if not indeed to original originals, or urtexts. These were similarly posited by combining imagination, or divination, with methodologically controlled analytical procedures.

The venturing behind the materially extant textual manifestations relied on four apriori assumptions: one, that the variation between both extant and lost instantiations of a given text was due to errors of transmission, and errors of transmission alone; two (concomitantly), that there was at the source of a given transmission only one stable text; three, that it was the task of a scholarly edition to collapse the manifest instantiations of the given text into one invariant text; and four, that to unveil that text as the recaptured text of the lost source (or, to recover a text as close as at all attainable to that source) was tantamount to securing the pristine work. It should be observed in passing, moreover, that under these methodological conditions texts and their material instantiations, that is: texts and the documents (extant or lost) that carried them, were always thought of in conjunction, and viewed as inseparable; ›text‹ and ›document‹ tended to be metonymically exchangeable. This habitual attitude may, in part, explain Eggert’s ease in arguing for restoration and scholarly editing as conceptual equivalents. The true flaw in the methodology as a whole, however, was and is the equation of text and work. It is a logical flaw, yet assuredly Eggert is not to be made answerable for it. It is in fact even to this day deeply ingrained in our cultural assumptions. Hence, Eggert builds on it. It is his doing so, however, that involves him in the particular intricacies of buttressing the argument for the mutual dependence of work and meaning that the monograph’s concluding chapter develops. There can be no doubt, of course, that we perceive a work as what it is, and that we are able to relate to it only by way of a hermeneutical exploration of its meaning(s). Yet just how this relates, in turn, to securing the work for the past by editing its text(s) is, or would have been, for this book the pertinent question.

We maintained above that works (and works of art) in language can be instantiated both materially and immaterially, and can in principle be replicated without limit. The instantiations are textual, and as texts – whether materialised in documents, or replicated orally – they are always (by default, as one might say) variant. The variation may be transmissional, as foregrounded by traditional textual criticism and scholarly editing. It may be compositional and revisional, as evidenced in drafts, working papers, and successive publication in revised authors’ editions. Or it may be oral, as when any one recitation of the work’s text from memory is never literatim identical with any antecedent or succeeding one. Any one text, whether it has come down derivatively through transmission, or in a manuscript layered in revisions, or by way of oral performance, instantiates the work. It follows, conversely (as already posited), that the work exists but immaterially, even as it constitutes the energizing centre of its textual representations. Some would hold that this amounts to theorizing the work platonically, as an ideal. Suffice it to maintain that the notion of ›work‹ as an immaterial entity is the pre-condition for seeing the ›work‹ endowed with an energy to hold together its instantiations as texts.

5. Texts and Work under the Editorial Gaze

What editors edit are not works, but texts. Leaving aside the new options for multi-text editions that re-conceptualising ›work‹ in the preceding manner opens, it is of course perfectly conceivable, and fundamentally indeed highly desirable, that among the work’s many textual instantiations an edited text should be the one optimally representing the work (rivalled at most, perhaps, by a first-edition text or the text of an authorial manuscript). Such an edited text may well be the best result achievable from historically aware and textually critical efforts to secure the work, as a creation in language, from the past. Nonetheless, an edited text, even while it may in quality surpass all other extant textual instantiations of the work, is never more – though neither is it commonly less – than one (considered) textual representation of the work. Yet not the rivalry among instantiations is at issue here. The decisive point to be made is that they all (by whatever degree, which textual scholarship makes it its business to determine) represent the work. Under guidance of Paul Eggert’s book, therefore, the question becomes just how securing the past is accomplished through scholarly editing. How do texts hold up under the editorial gaze?

In the first instance, the editorial gaze is not directed at the compass of complexities or depths of meaning of the work (which are ultimately what define the work as by nature immaterial). It is trained on the material minutiae of the text revealed through comparison of its multiple instantiations. To the largest degree – at least in scripted records of transmission – these instantiations will be identical: the invariant substance from the multiplicity of text materialisations in documents goes a long way towards establishing the material edited text as a valid simulacrum of the (immaterial) work. Taking the invariance as given, what the editorial gaze will fasten on as matter for editorial concern is the variation distinguishing the individual instantiations from one another. It is here, indeed, that linguistics, hermeneutics and theory impinge on editorial procedure and editorial decisions. Is a reading possible in terms of the lexis, grammar or rules of syntax of the language employed to text the work? Is a word or phrase, a grammatical or syntactic construction meaningful in itself, and in immediate or wider contexts of the work’s material instantiations under scrutiny, as well as of the edited text under construction? Are, moreover, textual alternatives (variants) to be adjudicated as mutually exclusive, or complementary to one another? It is under this latter question, especially, that heterogeneous positions of literary and text theory get adduced, precisely for their divergence on principles, to support and justify even opposing stances and solutions of editorial pragmatics. Orthodox editing aimed at eliminating error, on the one hand, will produce edited texts as stable and closed. Modes of editing, on the other hand, developed from a notion that variants are integral to a work’s textual spectrum will be geared to accommodating this perception and endeavour to represent texts as by nature progressive and open.

It is perfectly true that scholarly editing happens, and is enacted, or should happen and be enacted, in awareness of its wider critical and theoretical implications. Yet at the same time there is of course no escaping the fact that scholarly editing is a pragmatic endeavour. We maintained above that editing works (of art) in language cannot be accomplished without (first) going behind the extant textual materials, and we have shown how this may be understood, and has in fact been realised throughout the history of editing. At the level of strict editorial pragmatics, however, it is a work’s irreducibly material text(s) that become tangibly and inescapably the practicing editor’s concern. This is where editorial adjudication and decisions are called for. How comprehensively these are guided, let alone determined, by the broad approaches of hermeneutics, philosophy, or stances of theory, to the work, is a moot question. Or how they could be so determined or guided, considering the vast predominance of invariance over variation in the extant instantiations of material text for the work. At the pragmatic level, the scholarly editor can do no more towards securing the past for works (of art) in language through his craft than to mend, or touch up, or lay open the work’s extant textual record at its every point of indeterminacy – meaning simply, its every point of non-identity in the total compass of that record. (Jerome J. McGann once pointed out very perceptively that the textual record extant for a work will always frame such indeterminacy within its own material determinacy. [2]) We should also recognise that every textual instantiation of a work as edited text distinctly involves, too, a modicum of critical, and therefore creative input on the part of the scholarly editor. An edited text, while it is a material instantiation of the work, is at the same time decidedly the editor’s text, which confers a responsibility the editor need neither shirk, nor hide by denying it.

In a curious way, though, as it happens, the Anglo-American rulings in the editorial field have, since the second half of the 20th century, made it incumbent on editors to hide behind the author. The golden rule for scholarly editing since the 1950s has been to fulfil the author’s intention. The rule’s essential implication is that the editor is empowered not just, as by an older dispensation of textual editing, to adjudicate from specialised skill the readings from the extant material record of texts for a given work. The editor is now invested, too, with a hermeneutic dominance over the work. To determine teleologically the meaning of the work – the author’s final intentions determining ultimate meaning – is defined as an obligation to be fulfilled in the establishing itself of the work’s single instantiation as edited text.

When and how this assimilation of hermeneutics to the very practice and acts of textual editing happened, marks an interesting moment in the development of literary studies and theory, and therefore, too, in the intellectual history of the 20th century; and it is fascinating to observe both how the assimilation was decreed, and how in the aftermath oblivion set in that a momentous shift had indeed occurred. The rule in question proceeded, as is well known, to become the foundation of the Anglo-American ›theory of copy-text editing‹, or the ›Greg-Bowers theory of copy-text editing‹, as it is commonly designated. Greg and Bowers, however, should be kept strictly apart in the matter, for it is precisely at the point of transition from Greg to Bowers that the shift occurred.

W.W. Greg was a textual scholar rooted in classical and medievalist methodologies of textual criticism. He saw the extant earliest printings of Shakespeare’s texts as derivative of lost manuscripts (which of course they are). Perceiving them thus analogously to the late derivations, as they survive in scribal manuscripts, of long-lost original text inscriptions of works from Classical Antiquity or the Middle Ages, he recognised at the same time that, in contrast, the manuscript originals of Shakespeare lay buried very closely under the surface of the first edition printings. Additionally familiar with Elizabethan printing practices, Greg fused his expertise into a pragmatic ruling by which edited texts from the first editions as copy-texts could be achieved that would approximate closely the inscription in the manuscript printer’s copies for those editions, or with luck lay bare (namely where play texts could be assumed to have been printed directly from autograph) the material substance of Shakespeare’s own penning of his texts. In brief: W.W. Greg’s rationale of copy-text was substantively text-directed, and only accidentally geared towards the author. It was aimed at achieving the most authentic edited text from the extant first-edition records whose textual authority was materially evidenced or inferable. But it was not intention-directed. It was Fredson Bowers who not only saw, but capitalised on the intentionalist implications of Greg’s recommendations for attaining authentic edited texts. The admirably creative as well as power-conscious critic, textual scholar and editor that Bowers was, his was something of a coup-d’êtat. At the intellectual moment when New Criticism culminated in literary theory of the Wellek-Warren persuasion which resoundingly proclaimed the intentional fallacy, Bowers defined the fulfilling of the author’s intention as the finest flower of scholarly editing.

This lode-star conception remains apparently unquestioned to this day in main-stream Anglo-American textual criticism and editing. Methodologically, Paul Eggert certainly seems thoroughly imbued with it; which may be succinctly illustrated. In discussing (in chapter 9) »The German Encounter«, he cites literatim Hans Zeller’s stand on the question of intention: »A principle such as authorial intention cannot serve as a central criterion for the constitution of text [because it] remains a mere idea of the author on the part of the editor, and as such cannot be established reliably.« (206-7) Amazingly, and to me amusingly, Eggert makes no connection when pronouncing, with respect to »Gabler’s Ulysses«: »Gabler’s reading text aimed to capture the novel, as he stated, at its highest point of compositional development. This was not the traditional way of expressing the idea of a text of final authorial intention, but in truth the aim was deeply traditional.« (173) The first sentence I fully subscribe to: I did indeed so wish to capture the novel, or more precisely: the novel’s text. But the second sentence, while I do not object to the label of ›traditional‹ it confers, is yet an assessment prejudiced by the conception that copy-text editing cannot but imply realising »a text of final authorial intention« (134).

It is true that the Ulysses edition, through its phase of becoming a critical reading text, was established from a copy-text. This copy-text however was, in the first place, a virtual construct. It was and is not a text to be found inscribed throughout in one material document. Rather, it was constituted as the aggregate of James Joyce’s scripted text for the novel as it progressed materially through a sequence of documents of drafting, fair-copying, additional composition and successive revision. This copy-text, therefore, while assembled from multiple documents, was and is yet in its entirety without a direct material document basis of its own. (Be it also mentioned in passing that it thus applies, in its way, the strategy of logically divorcing text and document that Fredson Bowers was the first [to my knowledge] to devise and practice in his editing of Henry Fielding’s Tom Jones.) Leaving aside further detailing of the nature of the copy-text for the reading text of Ulysses, the operations through which it came about, and the manner of its heuristic deployment, what simply needs to be emphasized is that Eggert is mistaken in assuming the copy-text editing phase for Ulysses to have been a moment of realising »the idea of a text of final authorial intention« (173), let alone one of constructing an edited text that would fulfil that intention. All that the Ulysses edition claims for its (right-hand page) reading text is that it represents the work, as a text, in as close an editorial approximation as possible to what James Joyce wrote. The copy-text editing invoked and practised in establishing the edited text was therefore decidedly of the Gregian persuasion. It followed Greg’s pragmatic, text-directed recommendations and rules as they antedated their being re-interpreted as foundation for an intentionalist methodology, devised and decreed by Fredson Bowers, and dogmatised by Thomas Tanselle.

Thus: to posit, as Eggert does, an editorial gaze taking in all the complexities and depths of meaning of a work so as to accomplish the editing of one specific textual instantiation of it, appears both to over-estimate and over-tax the editorial role. Admittedly, the editor as editor, when setting out to engage with the work in the tangible materiality of its text(s), must make sense of it, and so read the work across the range of textual representations available to be considered as basis for the editing. To such a degree, the editor does engage as a reader with the meaning of the work. But even if this is so: the editor’s engagement with the meaning of the work has nevertheless only a minor, if not indeed a marginal, effect on the editorial engagement with, and the establishing of, an envisaged edition’s edited text. The proof of editorial skill arises only rarely from interpretation. What editing requires in bulk is adjudicating and adjusting minutiae in the material textual record under scrutiny – minutiae, that is, in terms of a work’s over-all complexities of meaning.

Beyond the editor as editor and reader, however, there is the reader as reader of the work and the edition – or indeed: of the work through the edition – to be considered. It is here that all questions and problems of meaning come fully into their own. For that product of criticism and humanities scholarship, the scholarly edition, the central question arises how it could, or should, relate to the reader’s quest for the meaning of a work in and through a text. The questions and problems of meaning, it is true, are adumbrated throughout Eggert’s tenth and final chapter. Positing that there is a relationship between the scholarly edition and the reader’s quest, the chapter goes to great lengths to discourse, in impressive diversity, how a work’s meaning(s) might be construed for an edition’s, or an editor’s, or a reader’s benefit. But the survey disposition of the argument turns out, in the end, to have little bearing on the specifics of conceptualising as well as of practising scholarly editing. What the chapter does not truly face, let alone solve, is the problem of how the search for, and the construction of meaning can, or might, be built and structured into a scholarly edition. The simple reason for this lack is that the chapter, as well as the book in its entirety, does not conceive the scholarly edition otherwise than as a text edition. Its all but unreserved adherence to the postulate of fulfilling authorial intention, notably, carries with it, as we have seen, the implication that such fulfilling supposedly also fulfils every hermeneutic requirement to be made of a scholarly edition.

Yet to secure a work (of art) in language as the inheritance from the past that it is, it is not enough to establish for it an edited text. A text edition only does not suffice to satisfy the needs of readers and users that it has been traditional to expect editions to meet and to fulfil. Over and above seeing editions as critically considered instantiations of the text of given works, it has therefore in our culture also been customary to regard them as the proper scholarly tools for mediating works of the past in terms of their content and meaning to the present of the editions’ own time. This used to be accomplished through annotation and commentary. Such discoursing of the work in natural language within the edition centered on the work’s text fell progressively into disuse, however, in the course of the 20th century. The rigors of formalisation of the textual apparatus won absolute ascendancy over the natural-language mode of the commentary discourses. The shadow of New Criticism, too, descended on the products of textual scholarship. The edited text standing in for the work gained absolute self-sufficiency over against all manner of historical or biographical or political or social ramifications that might be adduced to explore its meanings and interpret it – authorial intention excepted; for, as said above, even while the author’s intention was new-critically banned as a fallacy, it was simultaneously rescued for editorial scholarship by becoming text-itself.

Hence: where thus the real-world referents fell by the wayside that had been customarily resorted to for elucidating a work, or that, reciprocally, the work had contributed to shedding light on, the significance of the commentary as one set of the traditional scholarly edition’s discourses dwindled. Clinching with apparent finality the argument for marginalising, if not outright eliminating, the discoursing of editions through annotation and commentary, moreover, was the belief, seriously and optimistically held, that the critical texts realised by modern textual scholarship were definitive, and would never need to be done again. (As time went by, the optimism was somewhat dampened: perhaps, one distant day, texts might, after all, need to be re-edited. The modern scholarly editions however would definitely as editions remain definitive: for did they not assemble all material evidence required to establish critical texts?) Commentaries, on the other hand (so it was held), were inevitably short-lived; as ephemera of editorial scholarship, they would need to be redone at briefest intervals.

6. From Material to Medial Securing: The Scholarly-Edition-to-Come

With so much said, it still remains true, as Paul Eggert’s book Securing the Past posits, that to secure the past for a work (of art) in language – for a work of literature – scholarly editing is the cultural technique required. Yet the technique should be deployed comprehensively. It is not sufficient to realise it only in part by establishing a critical edition text alone. Admittedly, the range the monograph has set itself, encompassing art, architecture and literature, goes some way towards justifying that, in terms of literature, it largely confines its discussion to aspects of text editing. From the complexity of components making up the scholarly edition, it is texts that are directly bound to agency and materials, and it is foremost on their grounds that the conservation and restoration of works of art and architecture, on the one hand, and, on the other hand, the securing and bringing to life of the cultural heritage in language, and of works of literature specifically, are compatible for comparison at all. To have attempted the comparison has brought out the compatibilities as well as the incompatibilities. As we have seen, Eggert has in his concluding chapters guided us towards considering, or re-considering, that, or whether, beyond fulfilling its task of establishing an edition text, a scholarly edition could, or should, mediate (as scholarly editions did of old) the content and meaning of a work of literature, and thus engage hermeneutically with it.

This question opens vistas distinctly beyond the limits of Eggert’s monograph. We can here no more than hint at some perspectives implied. As a matter of fact, though, Securing the Past itself hides in its bibliography the link to a key term by which the scholarly edition of the future might find its bearings for a return to the depth and scope of its own ancestry in the realm of humanities scholarship. It lists Peter Shillingsburg, From Gutenberg to Google: Electronic Representations of Literary Texts, of 2006. Understandably, the potential of electronic representations of literary texts is not developed in Eggert’s argument, so predicated as it is on the materiality of the past-to-be-secured, including that of literary works perceived materially, since perceived as texts in documents. However, if we accept the contention developed above, namely that texts in their multiplicity (and variance) are but instantiations, materially documented representations, of the work that, as a work (of art) in language, stands outside the realm of the material, then to conceive of texts as equally, or alternatively, instantiated materially or electronically should present no difficulty. Every instantiation, whether on paper or as a digitized record, implies conceptually, as well as materially and in terms of agency, the divorcing of a text from one (antecedent) text carrier, followed by its inscription on a succeeding one. A text, if so re-inscribed digitally, may hence become, and be editorially formed as, the nucleus of a scholarly edition living no longer on paper, as all its ancestry of text instantiations of the work of necessity did, but in the digital medium. This I have argued before, and drawn conclusions from, elsewhere. [3] The buzz-word for how to build around a digital edition text a digital scholarly edition genuinely answering to the demands to be made of the scholarly edition (as a genre of humanities scholarship) comes from Peter Shillingsburg. The term he has given currency to in From Gutenberg to Google (having, importantly, observed since around the beginning of the new millennium, both in the US and in Europe, the envisioning and incipient emergence of digital research sites for the future) is the ›knowledge site‹. The bearing this has on the scholarly edition is that it provides an opening for re-conceptualising and innovatively re-shaping the erstwhile unity of text edition, apparatus, annotations, and commentary.

Since at least the 18th century, securing the past for works of literature through scholarly editions has been most comprehensively accomplished by means of the so-called Variorum Edition (editio cum notis variorum: ›edition with the notes of many‹). In the New Variorum Shakespeare, for instance, initiated in the latter half of the 19th century and still in progress, the tradition, amazingly, is still going strong. The format is compilational. Reference information collected from a wide variety of sources (lexical, linguistic, critical, historical, and in all other manner of ways factual) is gathered and linked by lemma reference to the text, say, of a given Shakespearean play as it advances as a text through its speech directions and speeches, scenes, and acts. Indexes will of course help users to find their way about and across the information material gathered; but the backbone along which the materials are principally organised is still the text’s consecutive seriality – which, within the material two-dimensionality between the covers of a book, could hardly be otherwise. By and large, such is the matrix throughout of orthodox commentary. Positivist by conception, in the first place, commentaries of the traditional school might be termed ›information sites‹.

There can be no belittling the usefulness of the information sites we are familiar with, and rely on, in books. However, the digital medium opens up the possibility, by contrast, of building knowledge sites. What, as we would suggest, here distinguishes ›knowledge‹ from ›information‹ is that knowledge, and the building of knowledge, grows out of, as well as initiates, creatively participatory intelligence. In simple terms, the combining of information with information, and/or with content and perceived meaning of a text instantiating a work (in the case, that is, of knowledge sites organised around the hub of text editions of works of literature), heightens the level and increases the range of knowledge. A knowledge site is thus relational, whereas information sites – even with indexes to offset the handicap – are by nature, and cannot mutate beyond being, serially arranged compilations.

The distinction is at bottom also a medial one. While what exists between the covers of books are information sites, the digital medium provides structural design potential and scope to accommodate knowledge sites. This, from the technical point of view, is simply because the digital medium can be programmed to organise, and to allow access to, its contents relationally. Given a technical infrastructuring (a software design) that permits data input as well as data access by relational patterns, new-generation digital scholarly editions may again be realised as akin to their erstwhile ancestors in books, and be offered as unified wholes of text edition, apparatus, annotations, and commentary. Relational by conception, they will, in terms of organisation, have shed the fetters of their positivist heritage. They ought, moreover, not be given to the world as finished products. The relational combination of their text-and-information content should provide nodes of knowledge to engage with. But then, the engagement cannot but generate enhanced knowledge. The knowledge site must consequently open up to enlargements of content and a deepening of hermeneutic understanding. That is, it should mutate further so as to become a genuine research site. Here, as we may recognise in conclusion, the scholarly edition, as a technique to secure from the past essentially immaterial works of literature, becomes (in the most positive way) thoroughly incompatible with anything one could even imagine being undertaken and achieved to secure from the past works of fine art or of architecture through conserving and restoring them in their irreducible materiality.

Across the disciplines, however, that Securing the Past brackets, it can still appropriately be said, as Paul Eggert does in summing up the vision that led him to write the book, that »the work […], as being constantly involved in a negative dialectic of material medium […] and meaningful experience […], and as being constituted by an unrolling semiosis across time, [is] necessarily interwoven in the lives of all who create it, gaze at it or read it […].« (237) It has for me, as these pages testify, been stimulating to engage with the book’s ideas and contentions, and to allow them to trigger insight and to generate understanding that, even while diverging time and again from Eggert’s argument, would without this reading experience have remained elusive.

Hans Walter Gabler

Munich, Germany


[1] Roland Barthes, The Death of the Author (1967/1968); Michel Foucault, What is an Author? (1969). [zurück]

[2] Jerome J. McGann, Ulysses as a Postmodern Text: The Gabler Edition. Criticism 27 (1984–85), 283–306; »Coda«. [zurück]

[3] Hans Walter Gabler, Theorizing the Digital Scholarly Edition, [Online in:] Literature Compass 7/2 (2010), (special issue Scholarly Editing in the Twenty-First Century), 43-56. Downloadable at:; or via (15.2.2011). [zurück]


JLTonline ISSN 1862-8990

This work may be copied for non-profit educational use if proper credit is given to the author and JLTonline.

For other permission, please contact JLTonline.

How to cite this item:

Hans Walter Gabler, Thoughts on Scholarly Editing. (Review of: Paul Eggert, Securing the Past. Conservation in Art, Architecture and Literature, Cambridge: Cambridge University Press 2009.)

In: JLTonline (03.03.2011)


A Persistent Identifier can be found in the PDF-Version of this article.