Introduction: Theory and Experience
In 1958 Ian Maclean, then Chief Archives Officer with the Archives Division of the Commonwealth National Library, toured North American and European archival institutions looking for best practices and suitable patterns for structuring archival services. It was a prelude to the creation of the Commonwealth Archives Office in 1961 as a separate entity. While there was much that Maclean gleaned from the overseas trip, there seemed to be little that made him doubt the validity of further developing the Division's own approach, which was to seek answers to the problems of twentieth century records through a pragmatic mix of Australian experience, an understanding of European archival theory measured against that experience, and a critical analysis of overseas practice. The result has been an Australian approach to recordkeeping broadly structured within concepts derived from Europe, but with significantly different principles more suited for records of the mid-twentieth century.
It has taken thirty years, and the advent of electronic records management considerations to the archival scene, for recordkeeping theory to receive the style of constant scrutiny and new expression that it had received in the Archives Division. Embedded in the new Office's approach to archival practice was Maclean's belief that if archivists had a separate professional identity then it was derived from a professional study which revolved around three elements:
Ian Maclean noted that such theory:
. . . in whatever form it may ultimately take, is also the essential groundwork of archival training. I say in whatever form because I am well aware that, even if its main principles continue to stand up to professional criticism, it needs much clarification and adjustment, not only in terms of logical argument but also in the light of the practical experience of archivists and records managers. 2<>It is the objective of this essay to carry forward the delayed process of clarification and adjustment by examining two articles written by Ian Maclean in which he communicated the Archives Division's recordkeeping experience to international audiences:
'Australian Experience in Records and Archives Management' in The American Archivist, vol.22, no.4, Oct. 1959, pp.383-418.
'An analysis of Jenkinson's "Manual of Archive Administration" in the light of Australian Experience' in Essays in Memory of Sir Hilary Jenkinson, edited for the Society of Archivists by Albert E.J. Hollaender, 1962, pp.128-152.
The first article, which Maclean wrote at the behest of archivists in North America, 3 dealt with the nature of the archival profession, the training of archivists, and the nexus between archives and records management. The greatest emphasis is upon registry practices which the Archives Division had helped revitalize during the 1950s. Their studies had stripped out time consuming correspondence registration activities and created a well controlled methodology for pre-action filing. Records management in the Commonwealth, with the full support of the Public Service Board had been upgraded in 1958 with the introduction of departmental Registrar's positions and the conduct of a Registrar's training program. The training program is set out in Maclean's article. Apart from the file management system, the Registrar's scheme also emphasized the survey and elimination of non registry material, with the archival institution monitoring the appropriateness of disposal schedules and having a power of veto.
The second article, again a requested article and shaped by its appearance in a festschrift for Sir Hilary Jenkinson, compared the practices of the Commonwealth Archives Office with the principles outlined in Jenkinson's Manual. In relation to Australian experience it dealt at length with the development of systems of control for the transmission of archives, arguing for an approach in which the documentation of records would accrete during their transmission along Jenkinson's chain of custody and indicating what progress had been made in this direction.
As the article citations and summaries indicate, they both have their own context in terms of the publications they appeared in, the audiences that would have read them, and the Australian programs they discuss. They also have an intellectual history in terms of the sources of the ideas within the work of the Archives Division with particular reference to the contributions of other staff members to those ideas. 4 Much of this context will be ignored in this chapter which is directed towards chipping away at the discourse Maclean's articles represent, with a view to finding the continuum within but never explicitly pulled together. Maclean's nominated areas of study do not sit idle in the articles themselves. They animate the development of practical approaches and refute the notion that archival theory is 'much ado about shelving'. A practical approach is revealed which is grounded in the characteristics of recorded information, focuses upon recordkeeping systems and gives conceptual dimensions to the processes involved in the ordering of records. As a discourse it did not gather much new momentum after the 1950s, but its encapsulation within the series system in the 1960s and its re-emergence within studies of electronic recordkeeping suggest that its elements are recursive and makes an essay on Maclean's articles overdue.
Prelude: Mid-Twentieth Century Records and the Branches of the Archival Profession
In relation to archival acquisition the 1950s had been a decade of salvage in Australia. It was also a decade in which archivists began to emerge in modest numbers and their ideas and practices became institutionalized in a manner which has left us with many legacies. 5 Within the Archives Division a network of repositories had been established and had been stocked using survey and collection approaches based on American experience with similar projects. Maclean tended to represent as an easy task the success of the Division's period as a hunter and gatherer of archives. The Commonwealth government had not been formed until 1901 and, although it inherited some nineteenth century records, in an international sense it had few 'old' records to deal with. By the time the Commonwealth Archives Office was created in 1961 it had already established a basic stock of historical material, its operations were confined to government records, and a fifty year access rule meant that few of the holdings of the Office were open to public research. 6
During the 1950s the Division had begun to focus strongly on analysing recordkeeping activities in a modern bureaucracy, reflecting the needs of an institution which in future would be dealing mainly with current records. This focus led Maclean to propose an archival profession consisting of three branches: those who worked in departments, those who worked in archival institutions but dealt with modern records, and those in archival institutions who worked on historical records. 7
Maclean was not impressed with the approach adopted by many archivists in the USA during the 1950s, which gave archival primacy to the third group, the historical archivist. He hoped for an easing of the tensions between archives and records management caused by the Hoover Commission report of 1949, arguing that with the report
the new era of records management was launched, in other countries as well as in the United States; and many archivists retreated to a defensive position in which they visualised themselves as historians serving historians, making only occasional sorties into the domain of records management with respect to the appraisal of records for archival purposes. 8
In private correspondence Maclean went as far as arguing that the true archivist was the records manager, equating such people with the first of his three branches. As he wrote in 1958 to Dr Grover of the US National Archives and Records Service,
It seems to me that the archival profession is undergoing significant change. The so-called records managers are really the archivists of the twentieth century. It is really necessary to start to think of archivists as we have known them to date as becoming what I might call, for want of a better term, historical archivists. I say this for two reasons. In the first case, the records of the mid-twentieth century are considerably different in form, quality and quantity to the records of the nineteenth century. In the second case, many archivists have ceased to be concerned primarily with collecting the records of the past for use by the present generation and are now concerned with organising the records of the present for use in the immediate future. 9
Maclean was greeting a false dawn. The successful systematization of registry processes diminished the need for records managers to understand the range of concepts that went into the development of the system. Records managers became participants in the processes of administrative recordkeeping and lumpers and carters of the physical record. They did not emerge as registrars, as managers of the recordkeeping processes, apart from the few exceptions to the general trend. 10
Grounding the Continuum: Characteristics of Records Materials
The Division accepted Jenkinson's concept of the archival document as its starting point, broadly agreeing with the definition Jenkinson offered in 1947 to students of the new course at University College London that
Archives are the Documents accumulated by a natural process in the course of the Conduct of Affairs of any kind, Public or Private, at any date; and preserved thereafter for Reference, in their own Custody, by the persons responsible for the Affairs in question or their successors. 11
Maclean himself, while not wishing to spend time on definitions of archives in his articles, proposed some interesting modifications to the Jenkinson definition of archival documents, including a rejection of the notion that they had to be preserved thereafter for reference. He also put forward a suggestion that it is not just their status as documents that have to be considered but also whether such documents achieve archive (or record) status only when they are made part of the record. For those alert to the philosophical nuances, Maclean can be read as suggesting that records creation is not a natural process, but involves a conscious decision to capture a record. 12
In making these suggestions Maclean stripped the definition back to its conceptual origins, which can be found in the text of the nineteenth century Dutch archivists, Muller, Feith and Fruin, and their definition of an archief, which in the English translation is inadequately described as an archival collection:
An archival collection is the whole of the written documents, drawings and printed matter, officially received or produced by an administrative body or one of its officials, in so far as these documents were intended to remain in the custody of that body or of that official. 13
An archive in the Dutch definition is a creation built up in the conduct of business, not a collection, as most people understand the term. Maclean's concept of incorporation into the official record is much closer to the original meaning. An archief of the type described in the Dutch manual is produced. It is never collected. The intention of keeping the record at the time of creation matters, which in a sense denies the naturalness of the process. Many archivists since the nineteenth century have argued that recordkeeping systems collect, rather than capture, records, or have tried along with Jenkinson to distinguish between archive making and archive keeping by giving a preservation twist to the latter but both arguments diverge from basic conceptual propositions about the nature of an archive in the sense outlined in Muller, Feith and Fruin's manual, whatever they may say about an archival collection. 14
Jenkinson had never accepted that age was a test for identifying the quality of archives and was aware of the artificiality of the archives/records distinction and the manner in which this was pushed too far by North American archivists. 15 In the Archives Division there was general agreement with Jenkinson's understanding that archives and records were, in their characteristics, much the same thing, and this had major ramifications in the manner in which Jenkinson's concepts of moral and physical defence and continuous custody were interpreted. Jenkinson acknowledged that archives quality is dependent upon custodians and assigned to them the task of setting aside material for preservation in official custody. His practical guidances for moral and physical defence were narrowly custodial in that the focus was on records that had been transferred to an archival repository. 16
For older records, Jenkinson could have his cake and eat it too. The evidential qualities of the material in custody depended upon proving 'an unblemished line of responsible custodians', but the best site for such material was an archives, where physical custody gave the archivist the chance to look towards both the moral and physical defence of the material. 17 Jenkinson developed his principles within this framework, this final custodial characteristic of the records material, but, as Maclean had noted to Grover, twentieth century government records were 'considerably different in form, quality, and quantity', and this required a different view of the custodial chain.
Jenkinson's Manual briefly acknowledged that, for purposes of custodial continuity, the archivist might have to go out of his way to secure the custody of archives with which he is not primarily concerned. 18 For an archival institution with an interest in current recordkeeping, a brief acknowledgment can be interpreted as an open invitation. The Archives Division investigated custodial continuity for records of the present tied to an approach based on the characteristics of the archives themselves and current recordkeeping processes. Jenkinson's defence concepts were thereby linked to considerations of the evidential qualities of records in a current recordkeeping context, and Jenkinson's notion of continuous custody became a pointer to the need to preserve the authenticity and impartiality of archival documents through a recordkeeping continuum. 19 The approach that subsequently developed in the Commonwealth Archives Office is encapsulated well in the following comment written by Gerald Fischer in 1972:
In the present view of records management, scarcely any time is left for dust to settle on records before they are the subject of archival scrutiny, transfer and control. There is often no settling down process for records in a purely administrative way so that they find their own administrative historical level. More prying than ministers of the Crown were at royal births, archivists are on the scene from the very point of conception of records. This is unseemly to say the least and . . . may well produce some very wayward behaviour . . . 20<>In precept, Fischer is the Jenkinsonian reiterating rules of objectivity. Conceptually, however, the Commonwealth Archives Office's staff could claim to be the Jenkinsonians. 21 They were helping to establish recordkeeping processes which attended to the moral and physical defence of the archives and paid respect to notions of their continuous custody. Some believed that the dust would choke those government archival institutions which waited for it to settle on the records of the second half of the twentieth century.
Finding Provenance: the Comparative Study of Past and Present Recordkeeping Systems
The increase in bulk and complexity of government records is a global phenomenon, but reactions to it in terms of records creation processes have varied from country to country. Even within European theory, as David Bearman has explained, there are significant differences in approach varying from the tight registry approaches of Northern Europe to the form of record approach of the Italians and to the mixed approaches of the Dutch and the British. 22 In turn these variations are reflected in the programs and theories of National Archives. On the recordkeeping map Australia in the mid-twentieth century can be plotted alongside the Dutch and British.
In an article published in 1956 in Archives and Manuscripts, Maclean noted that twentieth century Commonwealth government records could be broadly separated into two main classes: file type records and form type records. While the impacts of the increased volume of records and the greater complexity of the business covered were obvious, Maclean also noted that these should not distract attention away from another major change relating to the greater systematic linking of records to business method, which involved the conscious manipulation of the records product and a movement away from the concept that records accumulation was a natural process. 23
The awareness that mid-twentieth century records involved greater systematization took the Division into a dual strategy for gaining control of current records within the aforementioned Registrar's project. First, registers of sets of all records held in departments should be established. A set was not synonymous with a series. It was a physical unit which might equate with the series or might be a portion of it. It could even be a physical representation of a number of series as a result of the amalgamation of records at times of administrative change. The set approach offered a fairly standard records management method for gaining control over forms of record and standardised series of records through the conduct of surveys.
Second, detailed attention should be paid to registries, not because they handled the bulk of records but because
. . . all departments have to deal with some complex transactions which cannot be subjected to set work procedures but which arise out of complex political, economic, social and administrative situations. These are the ones which create not only the most difficult organisational problems in the department but also the most difficult file-construction problems for the record staff. 24
In other words if the records are going to be consciously manipulated anyway archival institutions should help organisations to cope with the most difficult tasks within this manipulation. The Archives Division helped revitalise registry systems during the 1950s but their interest went deeper than this. In reverting to an emphasis on recordkeeping systems, the Division was revisiting the registraturprinzip of Northern Europe and taking a particular approach to provenance, grounding that notion in the internal structures for recordkeeping not the external structures in which records were created.
The notions of internal and external structure are confused ones within archival theory, and Chris Hurley, elsewhere in this publication, offers a much better construct when he writes about context areas and recordkeeping areas. 25 In the 1950s, however, this was another unresolved contradiction that had been wrought from nineteenth century theory. The Division's interest in current recordkeeping processes again took them into the middle of the controversy. In this case the tension was caused not by a confusion between an archief and an archival collection, but by the difference between the French principle of respect des fonds and the germanic provenienzprinzip. Maclean's articles show the Division was much more influenced by the germanic view, with its emphasis on registration order. 26
Respect des fonds was a principle developed in France relevant to the preparation of inventories and the physical grouping of records in repositories. It was a classification system for recordkeeping entities pitched more at what Chris Hurley has described as the 'ambient' level. In its original conception Michel Duchein has noted that it had little to say on the internal dimensions of structure relating to the process of records creation. 27 As a principle its functionality can be found in the manner in which it can deal with the juridical (or legal) entity at a broad level. The fonds is the framework in which records are created, but it is not the creating agent itself. It is a useful fiction. 28
The provenienzprinzip, on the other hand had as its companion the registraturprinzip, which imposed control on registry records, which were the body of official records, registered before action in accordance with a classification scheme based on the administrative objects of the organisation itself. In a very strong sense germanic provenance gives expression to provenance at the time of records creation. Its functionality as a principle is derived from the registration order, which reflects the activities of the entities creating the records. Records have their internal structure imposed upon them as a result of current recordkeeping processes based on the activities of the organisation that created them.
These principles seem to have been brought together near the end of the nineteenth century by Muller, Feith and Fruin within their approach to the ordering of an archive. 29 Both principles had relevance to the records encountered in the Dutch situation. In bringing the principles together a third view of provenance was created, which amalgamated the other views and had its own features. The Dutch contribution was to place between the other approaches the notion of a dorsal spine of internal organisational structure to which different types of records could be attached, be they registered items or form type records. That principle was that 'the system of classification should be based on the original organisation of the archive which corresponds in its main outline with the organisation of the administration which produced it.' 30
Although it has only recently become apparent to non-Dutch archivists, Muller, Feith and Fruin's principle has no relevance to twentieth century records. 31 The Archives Division discovered its irrelevance experientially. By the 1950s the structural bones of government recordkeeping were more complex than any vertebrate could be. The Archives Division rejected the physical laying out of archives in the traditional manner as an impossibility for twentieth century records and not worth pursuing for any records anyhow, since what mattered for archival storage was location control, not the creation of browsable clumps of records. 32 More importantly, however, the Division began to develop practices which extended upon aspects of nineteenth century European archival theory and marked the beginning of new ways of thinking about archival tasks. 33 Maclean discussed these practical ideas within the traditionally prestigious archival term 'classification', a term which in North America and Australia has continually bumped into alternative usage by groups such as librarians and records managers and may well be falling into disrepute. The term, however, has very particular manifestations within archival theory and practice which need to be considered before deciding whether to let go of it.
Classification: the Ordering of Recordkeeping Processes
Classification can be viewed as consisting of three ordering processes. First there is the development of concepts and precepts (general notions and practical rules) by which the process will be carried out, second a plan or schema is needed for implementation, and third implementation occurs. The object of the exercise is categorization, and the placement of entities within the categories. This is a fairly obvious approach which particular applications of the process tend to obscure. Within professional thought classification has tended to take on particular meanings. Thus in the nineteenth century, with great clarity and some originality, Muller, Feith and Fruin produced the manual, which followed standard rational thought of the time about classing. It had concepts and rules, and as discussed above it proposed a schemata for preserving or re-establishing an archief's dorsal spine. It was based on administrative objects as reflected in the structures of records creating entities, was physically implemented within repositories, and was mirrored in the inventories of records.
The Dutch word used in the title of the manual, ordinen, can be variously translated. 'Ordering', 'arrangement' and 'classement' are some of the terms which have been used. 34 There is a tendency, however, for 'arrangement' to displace classification in archival terminology and for the description of records to become the focus for the development of schema by which records are placed. 35 The physical processes of placing records into archival systems have dominated ways of thinking about our ordering processes. The Archives Division's thought about the characteristics of twentieth century records and its study of recordkeeping systems, suggested that ordering or classificatory processes are multifaceted. In Jenkinson's Manual, classificatory processes are widely discussed, using terms like differentiation, and the placing of records. As Maclean pointed out, this refusal to face up to the complexities involved in classifying documents is apparent in the writing of T.R. Schellenberg. 36 In a records management context Schellenberg claimed that 'Classification, as applied to public records, means the arrangement of them according to a plan designed to make them available for current use', a narrow use of the term which loses contact with notions of placement and differentiation.
In the Archives Division, however, a more classical notion of classing was developed. Maclean was uncomfortable with the professional meanings that were developing around the term and defined it broadly and with imprecision. He saw Schellenberg's definition cited above as a threat to the proper service provision of archivists because it de-emphasized the evidential status of records and made use the raison d'etre. 37 It changed recordkeeping, which involves ongoing classificatory processes, into records management. Records management classification, from a utilitarian point of view, serves a narrow range of purposes. When it is viewed, as the Division viewed it, as a key component in the ordering of the record the focus of its practice is greatly expanded. 38
Alternative terms for the meaning that classification had within the Divisions's discourse include the Jenkinsonian terms 'placing' and 'differentiation'. A more modern interpretation of the term, as it was used, might be 'the establishment of processes for document discrimination' and 'the placement of documents within recordkeeping systems'. Thus Maclean's American Archivist article discussed classing in relation to differentiation into series of records, to file-making, to mail-handling, and to disposition, with an overriding emphasis upon classifying the activities which produced the records.
This aspect of the Australian recordkeeping continuum brings it in to alignment with the service based version developed by the Canadian archivist, Jay Atherton some twenty-five years later. Atherton's continuum was a revision of the life history model based, on a recognition that the parallels between archives and records management were revealed if you switched from thinking about the physical tasks involved in managing the record and thought about the management tasks within a service focus. He proposed a four stage model: creation or receipt, classification within some predetermined system, scheduling, and lastly maintenance and use. The service focus that Atherton proposed was based on one of the fundamental characteristics identified by Ian Maclean and the Archives Division that 'records are created to serve an administrative purpose, usually to document a transaction or decision'. 39
Atherton's use of the term 'classification' is typical of the narrow use which has brought the term into some disrepute and, linguistically, the focus is still the physical record rather than the processes of recordkeeping, but the elements are similarly 'Australian' in that characteristics, systems, and placement (classification) are brought together in an interrelated fashion and applied to records in general, not records in archival repositories alone. In Maclean's articles classification is deployed across all recordkeeping processes. The Archives Division's service continuum was influenced by Jenkinson's Manual and the extensive treatment it gave to 'Archive Making'. To understand the difference between a continuum of the Australian style and Atherton's, one only has to look at the creation and receipt stage. In Atherton this is a separate stage from classification, whereas in Maclean's writing a genuine continuum is put in train. The records are transmitted along Jenkinson's continuous chain. The process of transmission is grounded in the characteristics of records, and their provenance can be found in recordkeeping systems or forms of documentation. It is given effect by the classificatory processes involved in recordkeeping and derived from the processes of differentiation involved in placing records according to their nature and the available systems options.
The first task, during the creation or receipt stages, was to differentiate series of records, which in the Australian governmental context usually meant differentiating between registry material and standard or form type material which was held in separate sets throughout the workplace. Working from Jenkinson's classic input-process-output model of recordkeeping, 40 an understanding of how documents became part of the record was built up which was simple enough but took account of the complexity of twentieth century records which could be managed through classification processes. In going about their business agencies accumulate documents as records in various ways, the documents themselves take many different forms, and the types of transactions producing records were tremendously varied. Classing brings some order from the chaos. 41
In the average Commonwealth government agency, only about ten percent of the documents created or received ended up on registry files, but they involved documents that had a naturally deeper level of complexity since they could not be readily hived off into simple series. They also served a broader range of administrative objects. Within the registry, files were constructed using a classification structure based on the transactions of the creating agency and were carefully titled in terms of the action they would cover, the topic and objective of the action, and any client focus they might have. This enabled the documents to be placed on a file before action in such a way that the sequence of action was captured within its covers without registries having to go through the time consuming process of registering individual documents. The Division's first rule for filing documents was that the sequence of administrative action should be strictly observed. 42
The Division was aware that not all material in Australian registries consisted of documentation of organisational activity, but any other purpose of subject classification (other than the action as subject) was clearly subordinate. Maclean's second rule for filing was that there be a clear line of demarcation between files that have the purpose of 'recording the sequence of action for particular pieces of business' and those created with the purpose of 'gathering in one place information about particular subjects'. Information files may have been used in the background when actions were being taken, or decisions made, but were not products of that action. In case we miss the point of the special feature of good registry classification systems, Maclean added that 'in a sense, what the records staff are classing are actions rather than written pieces of information'. 43 Classification in this registry/set type of approach means the appropriate placement of the documents into series and systems, and subject classing means placement of the documents into categories of activity. Within the Archives Division's approach to records management, the evidential status of records, in the ideal, was protected by attention to form and series outside of the registry, by systematic capture of sequences of action in the context of a transaction based classing system within the registry, and by an emphasis on controlled and authorised disposition classing across all recordkeeping.
The Series System: Building Block for the Continuum
The driving force behind the Division's discourse, judging from the Maclean articles, was its emphasis on transmission along the custodial chain and the services that are offered along that chain through a records management role in a registry, in documenting and controlling the disposition of records outside the registry, or through the removal of material from systems towards to a final arrangement of closed groups within the archives. The archivist's contact with current recordkeeping issues needed to be coordinated with archival acquisition processes. Accessioning practices had to be worked out afresh in accordance with a Jenkinsonian respect for the characteristics of the materials being dealt with. The Division worked on the concepts present in European theory in the faithful belief that if one understood the characteristics of records, the recordkeeping systems and the placement of records subject to archival transmission, new and more satisfactory procedures would emerge. Also present was an understanding of the need to attach basic arrangement and description activities to the transmission of archives in ways which supported departmental use. In this way the complexities of administrative change and the shuffling of records for storage purposes could be managed on an ongoing basis. The goal of the Division was a transmission process that built upon the stages of transmission and ended with proper documentation of closed groups of records. 44
Maclean and his staff, however, had given up trying to solve the problems involved in accessioning within the records group approaches of the earlier texts and current archival practices. The records group approach was rendered impotent by the internal structures for mid-twentieth century recordkeeping. As Cornelis Dekker has commented, trying to arrange such records in accordance with 'dorsal spine' approaches is like trying to find a needle in a haystack. 45
In handling the complexities they faced, the Division saw no reason in overseas practice not to strive for a provenance grouping for records creators which operated at a single layer. Maclean's overseas study tour had suggested to him that, in purely mechanical details of documentation, fonds and sous-fonds and groups and sub-groups were much the same thing. While the presence of sub-groups meant a hierarchical form of inventorying could be imposed, and while the North Americans might express some philosophical differences in their approach to grouping, documentation at the different levels and in different countries was essentially the same and the concepts behind the group variations were substantially based on identifying a convenient level for the whole of a group in order to set in train the preparation of the inventories in which records were described. 46
The trip also showed him that the group approach was linked to the preparation of guides and inventories in ways that transferred massive listing tasks to the archival authority. The order of magnitude problem so tellingly identified by David Bearman was well understood by Maclean. While there were backlogs for arrangement and description within the Archives Division, they were not the same size as those that existed in the USA, for example, where the size of the problem boggled Maclean's mind. 47 It was perceived that Jenkinson's custodial interpretation of moral and physical defence within a record group strategy made most sense within its own concern with 'dead' groups, especially whenever the archival repository held most of the records from the group that would be received in custody. It was even practical when only small quantities of material were being dealt with. It was apparent to Maclean, however, that even his own office's small backlogs set up impossible work program loads. 48
Although later publicity has made it appear that Australian Archives' major contribution to archival systems has been in handling administrative (structural) change, the Division's original contribution to records system study can be found in its examination of the way archives could be fragmented in their internal structure by that change or by the systematised storage practices of mid-twentieth century recordkeeping. 49 Staff members analysed the effect on the structure of record series of the movement of records from one series of records to another because of administrative change, or the removal of sets of records from their current context in accordance with the need to trim inactive records for the sake of a recordkeeping system's efficiency, a process which included transfer to archival repositories. 'Sets' of records came to be viewed as different from series, sometimes being a subclass created by the structured removal of records from a series and sometimes being a superior class, consisting of the amalgamation of records from different series, and this view had actually replaced the series view until Scott turned the attention back to series with a methodology which handled this form of complexity by series registration, using cross references to previous and subsequent series. 50
The most notable result of the Division's exploration of archival concepts was Scott's system of control for the transmission of records, a system which could be integrated with, and incorporated into, current recordkeeping processes. Scott's discovery came after the creation of the Commonwealth Archives Office, at a time when, according to Ian Maclean, morale was sinking partly as a result of the enormity of the tasks ahead of them and the failure to gain additional resources. 51 Ian Maclean, in a retrospective comment, claimed that as Chief Archivist it took him 'not a minute' to consider the proposition for the new system proposed by Peter Scott when it was presented to him by his deputy, Keith Penny. For Maclean it provided the final building block for the integration of current and so-called intermediate records management and 'the foundation for the comprehensive range of activities and coverage that distinguishes Australian Archives'. 52
Part of the discovery's origins lay in the Office's understanding of registry control mechanisms, a point made clear in Chris Hurley's article elsewhere in this publication. Scott's agency registration processes provided a measure of stability to methods for documenting administrative change. The series registration approach gave a way of linking records to record creating entities and of linking related series of records, including the listing of the component series within a recordkeeping system. Maclean, however, also notes that it was a product of the longstanding pursuit of ideal classificatory techniques which he and Keith Penny had been pursuing. 53 The Division's study of records transmission within and through the various stages of custodianship paid a special dividend.
The Re-emergence of Recordkeeping Theory
The series system was a product of the Archives Division's discourse, which in turn was based upon archival concepts that had also given rise to an alternative discourse. Within the alternative discourse, which has been the dominant one in the USA, the archief became the archival collection, provenance became the organisational structures in which records were created, classification became arrangement, the intention to keep a record became confused with notions of permanent retention, and one essential characteristic of archival materials stood out – they were things in special repositories (or at least designated for such a location). Today the alternative discourses could probably be categorized as the 'documenting' approach and the collecting approach respectively, although neither exists uninfluenced by the other.
The conceptual base for the 'documenting' approach was not, however, an antipodean freak, and in the mid twentieth century was well known, and well understood, in the USA. Within it, the fonds was the entire government record, provenance related to records creation processes and the custodial chain through which records were transmitted, and the essential characteristic of archives was their connection to the activities which gave rise to them. Ernst Posner, noticing the emergence of elements of this strand of archival thought in 1940 pointed in advance to the Archives Division's discourse when he concluded an article with the following passage:
If all the public records of a nation are one sole undivided fonds, the agencies that are destined to receive and keep them ultimately will be justified in claiming the right to give their advice as to how the files of government offices should be organized and kept from the beginning so as to insure a satisfactory original arrangement that will also be suitable for retention by the archives agencies. We may assume that gradually the archivists will become the nation's experts who must be consulted in all questions of public record making and record keeping and likewise become the trustees who will safeguard the written monuments of the past, of the present day, and of the future. 54
Thus, Maclean, in the 1950s, was not being irrational when he proclaimed his office's ideas as the ideas of the future and the basis for the development of an archival profession. In the 1950s articles on recordkeeping theory and practice were not uncommon in archival literature and even the US National Archives and Records Service was internally debating its own role in relation to current recordkeeping, in ways which made some connection to the new discourse. 55 The concepts are not alien to the archival profession and have had various influences on archival practice in most countries. What was alien to most archival cultures was the strategic force with which the Archives Division and then the Commonwealth Archives Office pursued its understandings within a wholehearted focus upon mid-twentieth century records.
Accordingly, the Commonwealth Archives Office for several decades appeared to be something of a maverick institution on the global scene, not really in touch with archival theory. It was a disconcerting presence because of its own satisfaction with its system and with its strategies during a period when so many other archival institutions were creating the order of magnitude problems mentioned earlier. 56
On the one hand the Commonwealth Archives Office was addressing issues in ways that are becoming more commonplace in the 1990s, and a study of its practices has a new relevance outside of the framework of the discourse. On the other hand, its discourse is conceptually traditional and its achievements were a product of chance. The approach had been formed in the mid-twentieth century when the new concept was the proactive government archival institution with an interest in current recordkeeping. It had a charter (and was given resources) to carry out the custodial role for records up to and including World War II. It was able to use recently developed survey tactics to make a large dint in the corpus of records for which it was responsible. In a professional manner it then set about analysing mid-twentieth century records and was doing so well in advance of other archival institutions. It was, however, a product of its times, and one can go too far in labelling it the first 'postcustodial' institution. With this qualification, the links through to new models for archival thought are quite striking. In two cases, the 'characteristics of recorded information' and 'recordkeeping systems' the new models are developing and extending the concepts. In the third instance, classification, as I will try to show below, there is still a lot we can learn from the Archives Division's discourse, if we care to do so.
While David Bearman is the only present day archivist consistently developing ideas across all three areas of study nominated by Maclean, 57 a number of writers in recent years have added to our understanding of archival documents. While the terminology may vary, such documents are increasingly being seen as those records which can be identified by their nature as evidence of social and organizational activity, wherever they are located and whenever they are created – which within electronic systems requires an intention to keep a record in the time-honoured sense of Muller, Feith and Fruin's definition cited above. Such records are the subject of recordkeeping theory, which, as my colleague Sue McKemmish has noted, is increasingly focusing upon:
The renewed understanding of the authoritative and informational nature of records of activity is leading to a new approach to provenance which is as much functional as structural, emphasizing the functionality of records at the time of their creation and aspiring to link that functionality to the functions of an archival institution and the retention of a social and organisational account of events through the transmission of the
The 1950s interest in recordkeeping systems has been relatively dormant until revived by David Bearman, at least in terms of the published literature. 59 Within Maclean's writing the recordkeeping system is the fulcrum for the other major areas of study, the point which gives effect to an understanding of the characteristics of records and the classificatory processes that are needed to cope with the plurality of recorded information. There is more than a hint in Maclean's writings that he believed only the pre-action style of control of a registry system created levels of evidentiality sufficient to qualify it as a recordkeeping system in comparison with other records systems. Bearman gives all this a twist when he argues that it is the recordkeeping system which is the locus of provenance. There are parallels through to Maclean in Bearman's important observation that information systems are not necessarily recordkeeping systems. Maclean, within registry systems, argued for distinctions between background information files (which of course could be used in a transactional context), and 'sequence of action' files, which were purely transactional, although of course they could subsequently serve informational purposes based on their evidentiality. 60
In relation to classificatory processes it is Bearman who is less explicit in drawing together practical recordkeeping routines within a theoretical framework. Nevertheless his writings are filled with references to processes of placement and differentiation, including explanations of when and where to capture records. These processes take us into the systematic recognition of categories of business which was the basic grounding for the Archives Division's contribution to Australian registry systems. 61
The placement of records into a recordkeeping continuum depends upon organisational analysis in the writings of both Maclean and Bearman. Once the continuum is set in train, organisational analysis continues to be relevant to other aspects of placement including
The continuum model provides the structure for the placement of records through current and future activities and systems. The characteristics of the records, the recordkeeping system, and the ordering processes operate in an interconnected manner across space and through time. If I seem to be labouring the point, it is because the recordkeeping continuum is liable to be misinterpreted as a version of the life cycle concept simply because both encompass current recordkeeping processes. A continuum approach is very different conceptually, and in practical terms its impact upon electronic recordkeeping practices can be substantial. Approaches based on creation, maintenance and disposition refer to only some, not all, of the ordering processes involved in recordkeeping, and do not make up a continuum.
Maclean has given a recent description of the origins of the continuum, the development of which he traced to the records reduction campaign carried out by the Archives Division from 1950:
Thus began a transition in Australia from British/Continental archives administration in the traditional sense, with the primarily 'cultural' orientation, to what is nowadays sometimes called 'the continuum' of (public) records administration, with its emphasis both on administrative efficiency and also the safe-keeping of a cultural end-product . . . 62
One can question whether the continuum approach developed in the Archives Division adequately served social goals, and ask whether a national archives institution should also serve cultural goals outside the continuum. However the Commonwealth Archives Office in time may come to be seen as the first national archival institution to take non-custodial aspects present in archival concepts and use them to get out of the 'historical shunt'. The development of a 'documenting' approach within the Archives Division took it beneath the physical world of shelving and the management of records. Aspects of European theory were applied to mid-twentieth century recordkeeping processes and in the course of doing so concepts were articulated which are re-emerging as the profession, at a global level, has begun to gear up a new discourse on electronic recordkeeping. As the new theory develops it will need to gather new strategic force and strive to achieve credibility in the Networked Age. It will also need to be grounded in understandings of the characteristics of records, focused upon knowledge of recordkeeping systems, and dependent upon the classificatory processes of recordkeeping.
© 1998. All Rights Reserved. Licence: Limited to on-line viewing and the making of one (1) printout for off-line reading purposes only.