Article body

1. Introduction

In 2015, I called up a microfilm reel from the depths of the British Library. It contained several years of readers’ reports from the publisher Macmillan in the mid-nineteenth century. I had a very specific question that I was hoping it would help me answer: what did the reader or readers think of the manuscript of Maria Roscoe’s Vittoria Colonna: her life and poems (published by Macmillan in 1868)? Either I would find a report, or I would not, and that, in some ways, would be the end of it. I was doing additional research for an article that was based on my MA thesis; the thesis had focused on social context and textual analysis, and I wanted to add a more robust discussion of the reception of this work (Strowe 2018). Through some circuitous research path that I now cannot recreate, I had realized that there might be information on this initial type of reception (by the publisher) in the Macmillan archives, which are jointly held by the British Library and the University of Reading Special Collections Library. For the period I was interested in, the holdings were a little sparse, but there was a volume of readers’ reports listed that included 1868.

I found what I was looking for, and over the next few years even a little bit more (a few letters). But I also found that I was fascinated by the archive and what it contained. Another readers’ report, a few pages from the report on Roscoe’s book, for me encapsulates much of what I loved about the archive. In its entirety, this other report reads:

Translation of the Odyssey

 161. Leighton (G.O)

Is really too miserable for anything: it is below fifth form level. If publishing utter rubbish can do you any harm- avoid him. Just read the verses yourself: they’re quite grotesque[1].

Scores of questions arise. Who was G.O. Leighton? Was the translation really so terrible? What were the contents of the publisher’s letter back to Leighton? Did the publisher report this judgement in detail, or simply say, in a 19th century equivalent of what a scholar might now receive, “Thank you for your submission. Unfortunately at this time we are unable to undertake publication of this work.”? Did Leighton translate anything else? Was it all terrible? It is unlikely that the manuscript has survived, and I have not been able to find any record of a published Odyssey by a Leighton, G.O. or otherwise. It would not be possible therefore to conduct any kind of comparative translation analysis, or even the kind of habitus-oriented sociological analysis that has become more common with Bourdieusian approaches to translation.

I began to elaborate some actual research questions and research directions around some of these questions. In some ways what I was looking for were points of data that I could use to develop a narrative around the publication of translations in general, and about specific translators (Leighton, for example) or groups of translators (unpublished ones) in particular.

In this article, I look at some of the forms of loss and remediation that archives can offer, before turning to the notion of narrative and the framework of sociological (ontological) narrative theory in order to discuss some of them. The notion of the archive as narrative has appeared in a number of works, including the discussion of historiographical narratives in Barros, Carneiro, et al. (2018) or Richard Brown (1991) writing about Natalie Zemon Davis’ work on 16th century French archival materials. While the notion of narrative is discussed as significant in these works, it is not treated in such a way that makes it entirely clear what constitutes “narrative”–what features it may have, or at what levels it might operate. The ways in which Margaret Somers and Gloria Gibson (1994) articulate the characteristics and types of narrative, which have been brought into translation studies by Mona Baker (2006), can, I believe, provide a conceptual framework that facilitates deeper discussions of the constitution of narratives through, in, and around archives.

2. Remediating lack

Archival research, like any other, is about remediating some kind of lack of knowledge. We identify a gap in the literature, a set of research questions, a sense of objects or at least the kind of objects that we are hoping to find in the archive, and then we search for them. Perhaps the lack is broadly that we don’t know what the archive contains related to a particular area or topic. Perhaps it is more specific: particular knowledge that we hope (or are confident) will be contained in some item or items.

My archival work began with both: a general desire to know more about what the archives held related to translation, but also a more specific interest, partly raised by the unfortunate Leighton and the translation of the Odyssey, to see what could be uncovered specifically about aspiring translators whose work did not in fact end up being published, but of whom we might find traces in the archives. The materials from the middle of the 19th century are relatively sparse, so I focused my attention on the last decade of the century, and on the collections of letters in the George Bell and Sons Archive at the University of Reading, and the Macmillan and Co. Archive at the British Library. Preliminary research revealed that both the incoming letter collections from the period and the out-letter books in the George Bell Archive had references to translation, as did many of the readers’ reports in the Macmillan Archive. The Macmillan Archive also includes manuscript submission entry books, so there is systematic basic information even about manuscripts that were not accepted.

I was particularly interested in two things: 1) finding and quantifying evidence of translation work that was never actually published; 2) and finding material relating to the discussions taking place in the process of publication, from the location of texts to translate, or translators to work on them, to negotiations around pay and final conversations about details of the translations or the publication itself. The former area was spurred by the discovery of G.O. Leighton and curiosity about how much translation work was actually sent to Macmillan or George Bell and not published, and how that might compare with the rate of publication of other types of (non-translated) manuscripts sent to them. My interest is not so much in their translations themselves as in the evidence that these materials can provide about practices and practitioners of translation outside of the reward system of publishing, in order to expand our understanding of the broader, non-professional historical scope of translation. The latter area, which I hoped would also include communication with the un-published translators, was intended to make visible the figures and processes that underlie the publishing of translations. I wanted not only to find translators in various forms of invisibility, but also to resurrect their voices.

The idea of recuperating voices from the past is common in the literature around archives, particularly related to more marginalized voices. Heather Beattie (2009: 85) notes, for example, that “diaries written by women have been seen as particularly valuable, in part because in many cases women did not create the same kind of ‘public’ records as men.” Other types of archive are also seen as providing a kind of “voice” to figures from the past, through documents that surround the individual rather than documents that are produced directly by them. Prefacing his work on two enslaved Mauritian girls transported to Australia for murder, James Bradley describes his subjects as “historical ghosts, two of those many millions of figures who haunt the historian’s imagination” (J. Bradley 2011: 417), and goes on to suggest that part of the duty of the historian lies in “creating living voices for the deathly voiceless” (J. Bradley 2011: 418). His subjects are triply voiceless–female, enslaved, long-dead. His sources include census archives, court transcripts, state records, and letters about the pair and their situation, among other documents. In this case however, despite the rhetoric of “reclaim[ing] the voices of the voiceless,” James Bradley (2011: 416) notes that “the figures of Elizabeth and Constance [his subjects], never more than shadows, kept receding into the background.”

Also common, however, are critiques that focus on how this notion of recovery is problematic. Harriet Bradley observes that the voices in the archive firstly have already been partly heard and recovered simply through their inclusion in the archive: “only what has been stored can be located (the voices have been already selected and in a sense heard)” (H. Bradley 1999: 113). The truly lost voices are those that are irrecuperable even from the archive, not simply those that have been stored. Everything actually in the archive has already been judged, although perhaps not singly, as for example with the reader’s report about G.O. Leighton’s work, which was not selected individually, but which was one piece in a larger copybook that was selected for preservation. There is still, Harriet Bradley (1999: 113) acknowledges, the “intoxication of the moment of discovery when the historian experiences her first hearing of past voices,” the experience of the archive that revivifies the stored voices.

Even granting that “lost” voices in the archive are less lost than others, the notion of rediscovering voices has been critiqued on other grounds as well. Verne Harris writes against what he describes as positivist conceptions of the archive, in which the archive is seen as containing fixed reflections of reality (Harris 1997: 133). In addition to the complexity of the very notion of reality, Harris observes that even if we grant that somehow archives are able to “reflect reality,” the reflection is not clear or direct, but rather “deeply fractured and shifting” (Harris 1997: 135). Using the metaphor of voice, Harris writes that the archives “do not speak by themselves. They speak through many voices, including those of archivists” (Harris 1997: 135). There is no singular voice or set of singular voices to be “found” in the archive, no direct and unambiguous speech within the documents preserved.

The “discovery” of G.O. Leighton is a minor form of remediation through the archives. In addition to the very unflattering reader’s report that originally caught my attention, there is a letter in the Macmillan out-going letter carbon copy-book from Alexander Macmillan to G.O. Leighton[2], as well as an entry dated 25 March 1868 in the manuscript entry book for “A Translation of Homer,” from G.O. Leighton, Esq., with the address Milford House, Strand[3]. Leighton is unlikely to have been specifically and individually selected by an archivist; each of these documents is contained within a larger document that was preserved whole (the whole set of copied readers’ reports, the out-letter copy-book, the manuscript entry book). I am unaware of any other information about Leighton as a translator or about his translation of Homer. In this respect, then, his presence in the archive is certainly one of the less prominent presences, perhaps even undesired or accidental. His “voice,” even in a non-positivist sense, is still in many ways not present, as none of the documents were produced by him, despite being about his work. But it is not clear what his presence does in fact mean or what his voice would have said.

And even through the process of “discovering” a “lost” translator, the researcher encounters new forms of loss. The letter to Leighton, unfortunately, is illegible in places due to what is possibly water damage, but seems to be a very polite rejection of the manuscript. However, the letter appears to speak of Horace rather than Homer, suggesting one of two things: that the letter writer got it wrong and simply made an error, or that G.O. Leighton also sent in some translations from Horace. And which of these was the case, the archive may be unable to reveal.

3. Acquiring loss

There are losses that the archive bestows on us, which we cannot begin to comprehend until we begin to remedy what we had thought were the gaps to be filled. And these losses are in some ways more deeply felt, unless they can be resolved in the same archive. Because we encounter them in the archive, and because they may even arise from the archive, that archive, or the part of it that we have accessed, has already failed to fill in the gap. In some cases, it may be clear that no other archive is likely to fill it either.

The most obvious but also often smallest gaps arise from specific aspects of the documents themselves. The damage to the letter to G.O. Leighton in the Macmillan out-letter book means that while educated guesses can be made about contents, there is no reasonable way to get concretely past the smudging and running of the ink. The absence of any incoming letters from G.O. Leighton in the archive is unlikely to be remedied by any other archive. Both of these gaps arise out of the discovery of Leighton in the first place; without the initial reader’s report, the researcher would not know to look for the letters, and therefore not find them damaged or missing.

Within established patterns of correspondence with known participants, the contents of archival documents may point towards lacunae: the reference to a brief meeting in the office, the explicit mention of a previous letter not contained in the archive. In some cases, the contents or topics of those letters or meetings can be inferred from the rest of the available information. Such inferences, however, and the gaps that they fill, may range from the particulars of a known discussion (e.g. the exact price named in a conversation, which has turned out to be agreeable to the writer but is not repeated in the response) to the whole substance of what the recipient may have previously said about the quality or qualities of a translation.

In one telling example of some kind of misunderstanding, we see the kind of ambiguity that this type of lack can cause. L. Dora Schmitz was a German to English translator who worked on a number of publications for George Bell and Sons in the late 19th century, including the translation of the third volume on English literature by the Dutch scholar Bernhard ten Brink, written in German (ten Brink 1896). Within the correspondence between Schmitz and George Bell, one set of letters is of particular interest as regards losses discovered through archival research. After the publication of the volume, there are several letters exchanged regarding the quality of the translation, which result in Schmitz revising her translation for an eventual second edition. Of this exchange, the archive at the University of Reading appears to have only the letters from Schmitz to George Bell, without any of the letters from George Bell back to Schmitz that must have existed. Her letters refer to a particularly negative review of the volume that came out in The Nation, a news magazine from the United States, and of her reaction to it on both a personal and professional level:

I had not seen or heard of the review in the New York “Nation,” and need hardly tell you how vexed I feel at such a notice of the Ten Brink volume, for your sake as well as my own. The few notices I have seen have given me credit for having done the translation well, + thus I feel convinced this American reviewer must have some personal grudge against the volume, as you yourself suggest…

Any errors + defects I will gladly attend to before you publish a second edition… [February 27, 1897][4]

I will tomorrow send you the ten Brink volume + only hope you may not think I have made too many alterations. Translations are queer things + I believe we might go on for ever “improving” them. I hope I have avoided any “over-running.” That New York “Nation” review, made me determine to have all as right as possible for your second edition. [May 16, 1897][5]

I have been thinking over your note of the 18th, for I cannot remain satisfied that you should have the trouble of going over my many alterations in the ten Brink volume. Will you not return the book to me, + let me enter into a fresh copy such corrections only as are absolutelynecessary. If the volume is reset at some future time, the corrected book could then be made use of.

 I am glad you think some of the alterations superfluous + I corrected only with the view to make the translation more literal, as I thought that was what you wished. [May 30, 1897][6]

The researcher feels the lack of letters back from George Bell to Schmitz on both a general and a specific level. In general, we understand that there would have been correspondence. More specifically, however, we are teased by Schmitz’s phrases “as you yourself suggest” and “your note of the 18th.” The sense of particular loss is deepened by knowing exactly what is missing. The final lines in particular– “I am glad you think some of the alterations superfluous + I corrected only with the view to make the translation more literal, as I thought that was what you wished” –drive home the potential significance of these missing letters for understanding how publisher and translator would have communicated about quality and the characteristics of the translated text. What in the letters from George Bell would have led Schmitz to this conclusion? What might those letters have said about why certain changes were “superfluous”? What kind of discourse characterized the discussion of translation quality within the specific relationship between these agents? Given the tone of all of their communications, it is unlikely to be as blunt as either the earlier reader’s report on G.O. Leighton’s work or as the review in The Nation (which praises the source text, but describes the writing in the translation as “confused, harsh, and repellant,” announces that “it is a sin to put so badly written a book into the hands of students” and gives examples of what the reviewer characterizes as “the translator’s offences against both German and English, and–we must add–against common sense as well”; Anonymous 1897: 94).

We might call the missing letters “ghost letters.” Their existence is borne out only by references and inference, by the traces left in other documents or sources, and their absence haunts and informs any understanding of the remaining archival material. In this respect, they join the references to meetings (or in later archives perhaps to phone calls): events leaving no document to be placed in the archive, and identifiable only through reference. These all combine with damaged materials to form part of the corpus of lost items and information whose loss only becomes apparent through gaining information that is obliquely about them, whether in the form of those references or in the smudged ink that indicates an intent that has lost legibility.

4. Archive and narrative

Translation itself may be a telling metaphor for the archive. Translation, that is, not in its most widespread acceptation, but in the way that we know it in translation studies, involving multiple processes of decoding, recoding, understanding, contextualizing, decontextualizing, etc. depending on our perspective. It is possible to bring to bear on the archive all of the frameworks through which we understand translation: meaning, grammar, systems, logics, power, representation, and indeed many of these are also linked to the notion of narrative and the processes of narration and renarration.

The issue of multiple potential interpretations, depending on contexts, agents, etc., is one that arises in both translation studies and archival work (see Ketelaar 2001 and 2012). The lack of fixed or stable meanings that might emanate from the archives is the basis of a postmodern understanding of the archive (see Cook 2001) that requires a focus on the processes through which meaning is constituted by the producers of materials, archivists, researchers, and any others who come into contact with the materials. Ketelaar (2012: 23) describes a relevant experiment conducted by historian Philippe Artières, in which Artières and four other historians studied the same dossier of around 80 19th and 20th century documents having to do with a particular event, and came up with five completely different interpretations. Similar exercises with translation show similar variety in the understandings and interpretations of the “same” words (see for example Tymoczko 1999: 51-55 and 2010: 266-276; Weinberger and Paz 1987). The process of understanding and interpretation of words or documents involve determinations of focus, relative importance of different aspects, and choices between various options for interpretation or re-rendering.

Baker’s work on narrative theory suggests one way of approaching this multiplicity and understanding both its origins and some of its constitutive features. Baker (2006; 2014) draws on the work of Somers and Gibson (1994) to emphasize the constructedness of our understandings of the world and ourselves. This construction is accomplished through the “stories we elaborate in order to make meaning of our lives and to both guide and justify our actions” (Harding 2012: 287), ranging from our own sense of self (personal or ontological narratives) to how we understand what might be called the grand sweep of history (meta-narratives). Narratives are described by Somers and Gibson as “constellations of relationships (connected parts) embedded in time and space, constituted by causal emplotment” (Somers and Gibson 1994: 59, emphasis in the original). Drawing on other work in the social sciences, Somers and Gibson situate their understanding of narrative at the heart of lived experience, asserting that “social life is itself storied and the narrative is an ontological condition of social life” (Somers and Gibson 1994: 38, emphasis in the original). They list ways in which new research highlights the importance of this narrative basis for experience:

…research is showing us that stories guide action; that people construct identities (however multiple and changing) by locating themselves or being located within a repertoire of emplotted stories; that “experience” is constituted through narratives; that people make sense of what has happened and is happening to them by attempting to assemble or in some way to integrate these happenings within one or more narratives; and that people are guided to act in certain ways, and no others, on the basis of the projections, expectations, and memories derived from a multiplicity but ultimately limited repertoire of available social, public, and cultural narratives.

Somers and Gibson 1994: 38-39

Somers and Gibson identify four levels of narrative at work: ontological narratives, public narratives, conceptual narratives, and meta-narratives. Ontological narratives are at work in the individual’s construction of their own sense of identity (Baker also calls these “personal” narratives); public narratives deal with communities or institutions above the individual level. Conceptual narratives relate to the discursive practices of researchers themselves–for Somers and Gibson, sociological researchers in particular, which Baker broadens and clarifies slightly by reframing them as “disciplinary narratives” (Baker 2006: 39). Finally, meta-narratives are narratives that operate at the highest level, often known as “master-narratives” within which all other narratives are inscribed; Somers and Gibson give as examples concepts like “the Individual versus Society” or “Liberalism and the triumph of Liberty” (Somers and Gibson 1994: 63). Somers and Gibson (1994: 59-60) also identify four features of narrative that are key to interpretation and analysis: the fact that any piece of information or event must be understood with relation to other elements in a larger configuration (relationality), the causes or progressions that we understand to exist among the items in a given grouping (causal emplotment), the ways in which we select particular events to form part of our understandings (selective appropriation), and the location of the narratives within temporal and spatial contexts (temporality).

Baker (2006; 2014) develops Somers and Gibson’s work in translation studies as a way to understand the kinds of shifts that take place in translation, by conceptualizing translation as a form of re-narration in which the relationships, selections, and contexts of a source narrative are reconstructed by the translator (Baker 2014: 159). These shifts can involve anything from re-arrangements of events that can prompt new patterns of perceived causality, to the relationship between particular texts and the broad cultural narratives in which they are located or re-located (Baker 2006; 2014).

In some ways, narrative theory has both a strong version and a weaker version, much like linguistic relativism (the Sapir-Whorf hypothesis), which I would argue tries to draw similar connections between expression and language use, and the experience of reality. The strong version would say that all experience is constructed through narrative, and nothing can be experienced that is not narrative. Baker observes that “in social and communication theory, as well as in the work of some historians such as Hayden White, narrative tends on the whole to be treated as the principal and inescapable mode by which we experience the world” (Baker 2006: 9, Baker’s emphasis), stating that both Somers (1992) and Baker herself follow this understanding of narrative. Baker contrasts this with an approach to narrative deriving from literary studies, where, Baker quotes Labov as saying, it is “one method of recapitulating past experience” (Labov, quoted in Baker 2006: 8, Baker’s emphasis). As Baker observes, the actual object of study in these two cases is not entirely the same: in literary study, narrative is viewed more narrowly as encompassing something like a genre or mode of written or spoken expression, while in social theory it is seen as something that “underpins all modes of communication.” I would argue that the strong version does not need to be true in order for narrative to be a useful framework, or for the tools and categories of narrative theory to be useful in exploring a topic. At the same time, however, I am referring to “narrative” in the broader way that it is used in social theory rather than as a genre of writing or utterances.

I want to focus here on selective appropriation, relationality, and causal emplotment in relation to the archive. There are, as I see it, three primary “moments” of relevant narrative involved in the archive. The first is the initial inscription or creation of documents, which represent in some way a trace of the creator’s understanding of the world, of what is important to write or produce, of any narratives embedded in the actual content in the case of textual documents, and of other documents or items that are part of the creator’s narrative context and thus the constellation of relationality. The second is the construction of the archive, which represents the archivist’s narrative situation rather than the creator’s (even if the archivist is the same person as the creator, the situation is different). The third is the researcher’s use of the archive, which, while it may be partly an attempt to recreate the narratives of others (for example to recover the voices of the creators of the materials), is also strongly conditioned by their own narrative position, including the very discourses about archives and archival research. Each of these moments involves all of the elements of narrative, although particular elements may be more salient depending on the interests and focus of the observer.

Taking the letters written by L. Dora Schmitz, it is possible to see these features at work in her writing and her self-positioning, both individually as a translator and in relation to George Bell. This work is related to translation in two main ways. First, it is often about translation, regarding her perception of her own translations, her notions of what constitutes a good translation or what might constitute improvements, and her ideas of how other people might respond to translations. These are themselves partly personal narrative: her sense of self is inscribed through statements about her “vexation” at the criticism, her perception of her own translation, and her relationship with George Bell. Her professional identity is constructed, for example, through the linking and interpreting of information from several reviews, her approach to them and to the potential second edition, and the sense of responsibility she expresses towards her publisher.

Secondly, the letters are related to translation in the broader sense that Baker often refers to in her work on narrative, in that the texts themselves are the representation, the single textual manifestation, of her broader thoughts on these and other topics, and as such convey parts of her own understanding of events in such a way that is intended to allow the reader at George Bell and Sons to understand these events and her interpretations of them, as well as her own sense of her identity and responsibilities. For this reason, the lack of replies is not simply a minor gap in information, which can sometimes be remedied by educated guesses based on Schmitz’s responses, but a missing opportunity to see in what ways those narratives were understood and received, and within what other narrative contexts the editors at George Bell may have understood events. The loss acquired by being aware of specific absent materials can be viewed as the loss of an entire narrative position, which cannot necessarily be made up for by the triangulation of events or facts from other archival or non-archival sources.

At the same time, the construction of the archive as archive (rather than the creation of materials) follows its own narrative logics, which not only select individual items or sets of items for inclusion, but also arrange them in structures that are made meaningful by the narrative logics of both the archive itself and of the archivist’s perception of the narratives of the subjects. This is similar to Ketelaar’s (2001: 133; see also 1999) notion of “archivalization,” which he defines as “the conscious or unconscious choice (determined by social and cultural factors) to consider something worth archiving.” It is perhaps impossible to distinguish between various levels of selection and even happenstance; in my work, for example, it seems unlikely that I would be able to determine whether particular letters were discarded by an employee at George Bell a day, a week, a year after they were received, or whether they were weeded out by a later employee or archivist in the preparation of the George Bell and Sons Archive for donation to the University of Reading Archive of British Publishers and Printers. The selection of materials, then, is a combination of intent in which it is often impossible to describe a single operative principle of selective appropriation or even a single agent, or to distinguish the operations of these from items that may have been preserved simply because they were in the same file or document, or without any particular consideration apart from their existence as letters.

Similarly, the logics of the archives being examined here are mixed between the logic of a working publishing house with limited storage space, whose clerks decide what correspondence is worth keeping, and the logics of an archive determined to preserve materials, but also potentially to try to preserve certain forms of logic from the archival subject. The “worth” of materials, and the relations between them, are not fixed or even inherent to the objects; they are in the first instance, determined in part by utility (one imagines that letters from authors or translators who were to be published would take priority over those whose work was dismissed immediately), but also by social contexts and the networks of correspondence within which the agents of the publishing house would have understood its social role. Their relationality might be made up not only of the other letters or archival documents, that is, but also of the eventual published works, non-professional social relationships with the editors, and even the interests or curiosity of individuals. The worth of materials and the relationships that might be considered relevant by an archivist, however, do not necessarily map onto these same social contexts, and indeed, even between archivists responsible for organizing collections within the same library, relative ideas of worth, other objects within the narrative “constellation,” and logic shift both in relation to wider disciplinary narratives about the value and meaning of archives themselves as well as in relation to specific understandings of the particular archive. Ketelaar refers to the “tacit narratives” that “are hidden in categorization, codification, and labeling” (Ketelaar 2001: 135).

The researcher then adds to this a further set of principles for selection and organization of materials, creating essentially a sub-archive, a constellation of selected documents within the wider archive, which are perceived through the disciplinary and meta-narratives in circulation in the researcher’s time and place, as well as their understanding and interpretation of all foregoing narratives, set into a configuration with new or different additional elements, and perceived only through the “reading” of the archive, its materials, and its logics.

The presence of G.O. Leighton in the Macmillan Archive, for example, touches off for me as a researcher a desire to explore the “lost voices” of translation through the archives, “lost voices” being a disciplinary narrative through which I can understand materials that I have selected partly for a lack of success that I infer through the lack of particular materials in the constellation (a published work) and which I hoped to draw into related and perhaps even individually emplotted understandings of particular cases. This strategy for selective appropriation, however, and this logic of relationality are essentially in conflict with at least the narrative positioning of the publishing house (if not of the archivist) and the choices of the agents within it, who would have no reason to preserve individually the letters of an unsuccessful and unskilled translator, even though some documents might be preserved as part of a larger documentary item (an out-letter book).

In relation to Renaissance Florentine archives, Steve Milner (1999) discusses this in terms of the “partiality” of both the archive and the researcher, each as well in both senses of the word “partial”:

The records contained within the archives are by nature fragmentary; vestiges of the past, they are also partial in the sense of being subjective, testimonies to past relationships either between individuals or between individuals and institutions – social or political. Likewise, the readings of historians are partial both in the sense that the historian’s research is focused upon particular parts of the archive and in his or her subjectivity as an historian.

Milner 1999: 89

The first partiality in each case, the selectivity of the archive or of the researcher in assembling either the materials seen as “relevant” or as worth keeping, and in choosing the materials on which to focus in a research project, is relatively clear. The second, what Milner calls subjectivity, is, I would argue, not only the researcher’s subjectivity, but the wider framework of intersecting narratives within which the researcher necessarily operates. This does include personal narratives or public narratives that may have been used to construct the “subject” of the archive or of the researcher, but it can also include broader public, disciplinary, or even meta-narratives. And this second partiality is more closely related to issues of relationality and causal emplotment than to selective appropriation. Milner observes (1999: 96) that “the ‘source’ says nothing until the principles of selectivity underlying one’s research are applied,” but this idea of “principles of selectivity” is, I think, less about the selection than it is about the configurations within which the items are to be set and the relationships understood or imposed in order to create narrative, and which themselves constitute potential motives for selective appropriation.

The conflict between narrative positionings of the various agents (creators, archivists, researchers) thus means that materials may have been eliminated from the eventual archive that would be necessary or at least desirable elements of the researcher’s constellation, but it also means that the logic of the organization of the existing materials in the archive will often be dictated by divergent concerns and interests. Some organization arises naturally from the objects themselves, for example a numbered run of out-letter books, within each of which the letters are “organized” simply by the order in which they were sent out or transcribed by the clerk. Others, however, come from subsequent institutional decisions, whether from the publishing house or library archivists. The incoming letters for George Bell and Sons in this decade are loosely arranged by date and then by name; each group contains letters from a particular period, usually between roughly six months and two years, within which the letters are usually approximately arranged in alphabetical order by the name of the correspondent. The letters from L. Dora Schmitz mentioned above, for example, are from the group MS1640/325, which is one of two main groups for the year 1897[7]. This organization facilitates the kind of initial research that I was interested in, which involved looking chronologically through materials and finding references to translation. The Macmillan incoming letters, however, which are also held at the University of Reading, are arranged first alphabetically by author, facilitating research on materials related to specific, known individuals, but not broad, chronologically-based research logics (there is apparently no letter from G.O. Leighton among them).

These forms of organization facilitate not only different research aims, but the narratives associated with them. The Macmillan letters, with their focus on the individual correspondents but not the publishing house itself, lend themselves to attempts to discover events and relationships having to do with single individuals, although seen through the selective lens of later narratives of conservation and archiving. The George Bell letters, arranged more or less chronologically, do allow relatively easily the tracing of short-term personal stories, but given a greater or lesser lack of rigor in the alphabetization depending on the set, and the overall chronological organization, they are much better at demonstrating something about the publishing house itself and its activities.

All of this means that a framework of sociological narrative theory can help explain not only some losses of information (e.g. letters not saved) but also potential losses of the logics that governed the selection or organization of materials. These forms of loss are visible not in the materials themselves, but through the narrative constructions that we interpret or imagine around them and the narrative positionings of the agents involved, again, as interpreted or imagined by the researcher. Selective appropriation as a feature of narrative comes to bear on some kinds of potential loss or lack of information, while the relationality of narratives allows for the theorization of losses of coherence or logic among materials.

5. Conclusion

Work like that of Ketelaar (2001) and Cook (2001) stresses the multiplicity of the archive from a deconstructionist point of view, in the context of understanding the narrative of the archive itself (rather than the narratives contained within it). The application of Somers and Gibson’s (1994) notion of narrative, however, provides a framework for analysis of the multiple narratives not only of the archive but in the archive, thus allowing for specific analysis of the constitution of the narratives but also for exploration of conflicts and contradictions between them. We can perceive as narrative (in this sense) the contents of archival materials, their preservation and organization into the archive itself, and the processes of research and the creation of knowledge from an archive. As a result, we should not only engage in forms of self-reflection and reflection on the archive as an institution, but also analyze shifts and differences in narrative positioning across the various roles involved and, importantly, break down more clearly what actually is seen as constituting and characterizing the narratives involved.