UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A theoretical and historical analysis of pertinence- and provenance-based concepts of classification… Giroux, Alain 1998

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1998-0447.pdf [ 4.24MB ]
JSON: 831-1.0088686.json
JSON-LD: 831-1.0088686-ld.json
RDF/XML (Pretty): 831-1.0088686-rdf.xml
RDF/JSON: 831-1.0088686-rdf.json
Turtle: 831-1.0088686-turtle.txt
N-Triples: 831-1.0088686-rdf-ntriples.txt
Original Record: 831-1.0088686-source.json
Full Text

Full Text

A THEORETICAL AND HISTORICAL ANALYSIS OF PERTINENCE- AND PROVENANCE-BASED CONCEPTS OF CLASSIFICATION OF ARCHIVES by ALAIN GIROUX B.A., Universite de Montreal, 1989 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARCHIVAL STUDIES in THE FACULTY OF GRADUATE STUDIES SCHOOL OF LIBRARY, ARCHIVAL AND INFORMATION STUDIES We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August 1998 © Alain Giroux 1998 09/11/98 10:20 FAX 604 822 9587 SPECIAL COLLECTIONS g!003 in presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of School o f UteftW flgcH-l'tf/ll, A/v/W 3 - V F b f c / l A " U W SriOW'BS The University of British Columbia Vancouver, Canada Date 1L/ $Z?lEAW(l 11 ^ g DE-6 aim 0 9 - 1 1 - 9 8 1 3 : 2 1 RECEIVED FR0M:6Q4 822 9587 P.S3 ABSTEACT This thesis describes the main systems of classification used in public archives since the end of the eighteenth century. It provides an overview of the theory on which all archival classification is based. Grounded on the principles of this theory, pertinence-based and provenance-based classifications are described in the historical, juridical and intellectual context that framed their development and use. Each system of classification is examined to see how it structures archival material for both intellectual and physical control. 11 TABLE OF CONTENTS Abstract Table of Contents List of Figures List of Tables INTRODUCTION Chapter One General Principles for the Classification of Archival Material Chapter Two Pertinence-Based Classification Chapter Three Provenance-Based Classification Chapter Four Adapting Provenance-Based Classification to Modern Records Chapter Five Conclusion Bibliography LIST OF FIGURES Figure 1 Vertical View-of a Hierarchical 17 Classification Figure 2 Horizontal View of a Hierarchical 20 Classification LIST OF TABLES Table 1 List of Descriptive Data Elements from the 21 Ad Hoc Commission on Descriptive Norms Table 2 Commonwealth of Australia Government 6 6 Departments Rate of Change v INTRODUCTION "Le document d'archives n'existe que s'il est connu."1 The raison d'etre of archives is to link information that has been deemed worth preserving with information seekers. The preservation of documents in archives can only be justified by their use or expected use. Although the best way for a user to assess if a document fits his or her information need would be to view it directly, in most cases, this is not a viable option. No one could possibly want to look at all the holdings of an archival repository to determine which documents they need. One of the main tasks of archivists is, therefore, to create the links and establish the paths that will make archival documents known and allow users to discriminate only those with relevant information. There are three basic ways to organize information to ease access: by consecutive arrangement, indexing, and classification.2 Consecutive arrangements are a linear order such as alphabetical, numerical, or chronological. Consecutive arrangements maintain their integrity at any volume, but are virtually useless as keys to information content. They do not provide their subdivisions with categorical or subject significance. Consequently, they have little value for Manuel d'archivistique: Theorie et pratique des archives publiques en France (Paris: S.E.V.P.E.N., 1971), 153. 2 Thomas Wilds, "Information Retrieval," American Archivist 24, no. 3 (1961): 270. 1 information retrieval unless they are combined with indexing or classification. Indexing means the operation of creating an index for information retrieval. It involves the selection and assignment of terms to a documentary unit in order to indicate subjects or other features of the unit.3 Classification is the categorization and the placement of these units within categories. Of the three, it is the most significant means of structuring information. This thesis examines the systems of classification that have been developed to order archival material. It should be noted that in archival science the term "arrangement" is usually used instead of "classification". "Arrangement" has a broader meaning and is closely associated with the activity of description. In archival terminology the terms "arrangement and description" usually go hand in hand. Attention must be drawn to the fact that the term "arrangement" is often taken to mean both physical and intellectual arrangement. The latter would be more properly labeled "classification". There is, however, an unfortunate tendency in archival science for the word "arrangement" to displace the word "classification". This may be due to the fact that the physical processes of placing records into archival systems have dominated ways of thinking about our ordering processes.4 Arranging things James D. Anderson, wProposed American National Standard Guidelines for Indexes and Related Information Retrieval Devices, Draft 4.1," National Standards Organization, 1993. Photocopy. Frank Upward, "In Search of the Continuum: Ian Maclean's "Australian Experience" Essay on Recordkeeping," in The Records Continuum: Ian Maclean and 2 involves placing them in relation to each other according to some purpose or plan. The activity is a physical one. In arrangement, the placement is made to correspond to some potential order inherent either in the things themselves or in means of denoting those things, such as words or symbols. There is a fundamental difference between the two. "To arrange things is to place them in an order relative to each other...; to classify things is to allocate them to categories previously established."5 Classifying things involves assigning each of them to a class on the basis of the concept of that class being considered applicable. The activity is a mental one: the forming of an association between object and concept.6 Grouping things together physically for the purpose of holding a concept in mind or for other purposes of arrangement can follow so closely after formation or application of the concept that confusion of the mental operation with the physical operation is easy. The root of the confusion is the fact that classification, in archives, is often inseparable from the actual arrangement of things on the basis of similarities in characteristics. One only has to think of a file which is both the sum of the records related to a given subject and the container (folder) holding them physically together. In such circumstances, the initial formation of the classificatory concept may precede arrangement by an interval so short as to be Australian Archives First Fifty Years, edited by Sue McKemmish and Michael Piggott (Clayton, Australia: Ancora Press, 1994), 118. 5 Michael Cook, Archives Administration: A Manual for Intermediate and Smaller Organizations and for Local Government (Folkstone, UK: Wm Bowser and Sons Ltd, 1977), 103. 6Dallas Irvine, "Some Kinds of Classification," American Archivist 31 (January 1968), 13. 3 almost inappreciable. There is an interval, nevertheless, between the two steps.7 The first is concerned with intellectual control; the second with physical arrangement. Both are meant to ease access, but only the former is concerned with categorization. The focus of this thesis is the description of the different classification systems that have been developed to mediate users' access to information, not with arrangement. The focus is on how archival classification systems are structured to represent information contained in documents and to provide access to that information. It should also be noted that this study concentrates primarily on classification systems for archives transferred to public repositories (delegated archives). This means that it deals primarily with classifications aimed at documents selected for continuous preservation in archival repositories. Finally, it only examines how archival classifications were intended to function. It does not evaluate how these models actually performed or to which extent they reached their intended objectives. This thesis aims at describing and explaining the main classification systems encountered in public archives along two axes. The theory of classification is one. It remains constant over time for any archival classification. The context in which systems of classification developed is the other one. The Irvine, "Some Kinds of Classification," 13. 4 particular instances of archival classification based on this theory are framed and shaped by the time and place in which they were created. It is posited that archival classification systems can be better understood in the light of the basic concepts of classification theory and the political, juridical and historical context surrounding their development and implementation. There is, therefore, a need to analyze both the theory and methods behind various classification schemes within the framework of the spirit of the age in which they developed. Based on this premise, the structure of this thesis addresses both the constant and the variables to describe and explain archival systems of classification. Chapter One deals with the underlying principles and assumptions of all archival classification systems. It describes the concepts of categorization and the inferential process it entails. Chapter One provides the conceptual elements that are used in the subsequent chapters. Chapter Two describes the pertinence-based classification as it came into prominence in the nineteenth century following the turmoil of the French Revolution and the Napoleonic Wars and the establishment of public archives. Chapter Three describes the provenance-based classification and its corollary, the concept of fonds, as they became the 5 foundation of the emerging archival discourse in the late nineteenth century. Chapter Four describes the attempts made to adapt provenance classifications to the realities of sprawling modern administration and their open fonds. This chapter particularly reviews the solutions proposed by archivists to overcome the shortcomings of provenance and the concept of fonds in dealing with open fonds. The conclusion provides a brief overview of the findings from the individual chapters. 6 CHAPTER ONE: GENERAL PRINCIPLES FOR THE CLASSIFICATION OF ARCHIVAL MATERIAL The theory of classification is at the basis of all archival classification schemes. An overview of this theory's concepts is necessary to better understand the archival systems of classifications drawn from them. There is nothing more fundamental than classification to our thought process. It is an a priori assumption of human apprehension in general. We postulate that the world itself is orderly. We assume that the world comes as structured information rather than as arbitrary attributes. The world is perceived to possess a high correlational structure. Order is the guarantee of the intelligibility of the world. The fundamental step in ordering our perceptions of the world is simply to group together objects or concepts in categories by some system of relationships or association among them. Categorisation is, thus, the process of dividing the world of experience into groups whose members share one or more characteristics in common. It consists in the recognition of similarities across entities, such as documentary units, and the subsequent aggregation of similar entities into categories. Categorisation helps us discover order in complex systems that would otherwise consist of a multiplicity of unique and separate 7 entities. Such a mental operation is the foundation of our apprehension of the world. By assigning labels to groups of objects with similar characteristics, we create categories that allow us to organise and store information efficiently, and to retrieve stored information easily and effectively upon demand.9 This is in broad lines the theory of classification. Let's look in turn at the different elements that make up this theory and at how they apply to archival classifications. The first element, which underlies any classification, is the inferential process. Categorisation is based on inferences. We assume that the immediate, apparent characteristics of an object are the basis for inferences about its other attributes. This inferential process rests on the assumption that the world divides into natural kinds, and those stand in relations of greater and lesser resemblance to and distinction from one another. To infer means to draw a proposition from one or more propositions in which it is implicitly contained. Syllogisms are the most familiar form of inferred information: if A=B and B=C then A=C. The following example illustrates this inferential process: Elin K. Jacob, "Communication and Category Structure: The Communicative Process as a Constraint on the Semantic Representation of Information," in Advances in Classification Research: Proceedings of the 4th ASIS SIG/CR Classification Research Workshop, vol. 4, edited by Philip J. Smith et al. (Columbus: American Society for Information Science, 1993), 81. Jacob, "Communication and Category Structure," 81. 8 I f S o c r a t e s i s a man and I f a l l man a r e m o r t a l , t h e n S o c r a t e s i s m o r t a l . The i n f e r e n t i a l p r o c e s s t a k e s p l a c e when we move from t h e p r e m i s e s t o t h e c o n c l u s i o n s by b r e a k i n g them down i n t o c l a s s e s . The c l a s s " m o r t a l " i n c l u d e s t h e c l a s s "men" and , i n t u r n , t h e c l a s s "man" i n c l u d e s t h e c l a s s " S o c r a t e s " . Because " S o c r a t e s " b e l o n g s t o t h e second c l a s s , i t can be i n f e r r e d t h a t i t b e l o n g s a l s o t o t h e f i r s t c l a s s a s w e l l . As t h e example shows, t h e t h o u g h t p r o c e s s i s b a s e d on c l a s s o r c a t e g o r y c o n c e p t s and t h e i r e x t e n s i o n . The i n f e r e n t i a l p r o c e s s d e r i v e s from t h e f a c t t h a t t e r m s of t h e p r o p o s i t i o n s e x h i b i t a s e t of c l a s s - i n c l u s i o n p r o p e r t i e s . The c o n c e p t of c l a s s i s a t t h e h e a r t of t h e i n f e r e n t i a l p r o c e s s . C l a s s e s can be d e f i n e d a s g r o u p s of e n t i t i e s s h a r i n g , i n t h e eye of t h e c l a s s i f i e r , a t l e a s t one r e l e v a n t common c h a r a c t e r i s t i c . C o n s e q u e n t l y , a b a s i c a s s u m p t i o n i n d e f i n i n g c l a s s e s i s t h a t " i n f o r m a t i o n - r i c h b u n d l e s of p e r c e p t u a l and f u n c t i o n a l a t t r i b u t e s " 1 0 e x i s t t h a t .form n a t u r a l a g g r e g a t e s . Each c l a s s i s , t h u s , d e f i n e d by an " i n f o r m a t i o n - r i c h b u n d l e " of p r o p e r t i e s . 1 1 Each member of e a c h c l a s s h a s n e c e s s a r i l y e a c h of t h e p r o p e r t i e s c o n t a i n e d i n t h e b u n d l e of p r o p e r t i e s f o r t h a t 10 Eleanor Rosch, "Pr incip les of Categor iza t ion ," in Cognition and Categorization, edi ted by Eleanor Rosch and Barbara B. Lloyd (Hi l l sda le , N . J . : Lawrence Erblaren Associates , 1978), 31. The terms attributes, properties, characteristics, or features a r e used in te rchangeably in t h i s s tudy t o des igna t e t h a t which def ine c l a s s e s . 9 category. All the entities that have a given property or collection of properties in common form a class. In the first example above, "mortality" is the property that defines the class xxmen" and the membership of Socrates in that class. The same principle is used in archival classifications. For example, minutes have a certain format with defined properties such as date and people present, a medium, a creator, and so on. These properties are sufficient to define a class. Something exhibiting all of these properties would be deemed to belong to the class "minutes". However, not all these properties are necessary to define the class "minutes". Properties can be divided between essential and incidental. That is, among the properties that minutes have, some are essential; these properties are necessary to mark the membership of certain records in the class "minutes". From the example given above, the meeting date is an essential property of minutes. Other properties are incidental, that is, properties that entities happen to have but that do not define entities' membership in a given class. Minutes could be recorded on different media such as parchment, paper, microfilm or on a diskette. The medium upon which they are recorded is an incidental property to the membership to the class "minutes". The membership of an entity in a particular class is, thus, conditional on its possession of a set of essential properties for that class. 10 The intension, or definition, of a category is a statement of those essential features which are necessary and sufficient for determining category membership. It is a summary representation of an entire category of entities. The power of classification for intellectual control of archival material resides in this ability to constrain the membership of a category. It is assumed that boundaries between categories are made at discontinuities in attributes. Categories tend to be viewed as separate from each other and as clear-cut as possible. For example, the format of minutes is a property that ties the class "minutes" together. It distinguishes it from all other types of records that do not possess that property. Thus the critical assumption that intension equals extension not only determines the membership of a category but ensures stability of reference for a category label, such as "Minutes". If the intension of a category is not determinate and its extension is not fixed, the stability of reference is jeopardised. If intension does not equal extension, it follows that a category label cannot possess an invariant meaning that exists independent of context. Stability provides-the most important qualities from the user's point of view: consistency and reliability. The extension of a category identifies, or refers to, those entities which comprise the membership of the category. Because the intension of a category term defines the set of essential properties, each entity accorded membership within the category must exhibit the full complement of defining features. Intension 11 and extension overlap. Membership within a p a r t i c u l a r category (extension) e n t a i l s possession of an e s sen t i a l and defining character ( intension).1 2 For example, for a given documentary un i t , the in tent ion would be the data elements (propert ies) charac ter i s ing t h i s un i t , while the extension would be a l l the e n t i t i e s having those p roper t i e s . From the example above, the intension of the "Minutes" category might be "the o f f i c i a l record of the proceedings of a meeting"13 while i t s extension i s comprised of a l l the documentary un i t s f i t t i n g t h i s de f in i t ion . All members of a c lass can be iden t i f i ed by the label of t ha t c l a s s . Raising documentary un i t s to the level of e x p l i c i t representa t ion usually leads to giving them names.14 "The name i s the thing, and the t rue name i s the t rue thing.1 5 To speak the name i s to control the thing."1 6 Navigation through the c lasses of a c l a s s i f i c a t i on requires naming. In our example, "minutes" i s the label for the category "minutes". The label "Minutes" represents a l l the necessary and suf f ic ien t proper t ies t ha t Jacob, "Communication and Category S t r u c t u r e , " 83. Webster's Ninth New Collegiate Dictionary, s . v . "minutes", 1988. 14 Steven J . DeRose, "Navigation, Access, and Con t ro l , " in Proceedings of the Berkeley Finding Aid Conference, Apr i l 4-6, 1995. URL f i l e , h t t p : / / suns i t e . be rke l ey . edu /F ind ingAids /EAD/de rose .h tml . 15 This seems t o be p a r t i c u l a r l y t r u e in a rch ives where t h e name and what i t s tands for sometimes become i n d i s t i n g u i s h a b l e in a r c h i v i s t s ' minds. Carl Vincent , for example, noted t h a t "One of t h e mental b locks of a r c h i v i s t s must overcome i s t h ink ing t h a t t he var ious symbols used t o i n d i c a t e records groups and s e r i e s have an i n t r i n s i c va lue beyond i d e n t i f y i n g the s e c t i o n s of the f inding a ids and i n v e n t o r i e s provided by t h e a rch ives in which a d e s c r i p t i o n of the records and t h e i r p lace wi th in the government s t r u c t u r e can be found. . .The common use of a records group and/or s e r i e s number combined with a box or volume number as a l o c a t o r device t o enable r e t r i e v a l of the record wanted i s not r e l e v a n t t o the main po in t of de sc r ib ing t h e document." Carl Vincent , "The Records Group: A Concept in Evo lu t ion , " Archivaria 3, ( F a l l -Winter 1976-77), 13. Ursula LeGuin quoted in DeRose, "Navigation, Access, and C o n t r o l , " URL f i l e . 12 define membership in this category and it stands for all the documentary units that demonstrate these properties. The label "Minutes" represents the intension and the extension of the category. To help preserve the one-to-one relationship between intension and extension, each class needs to have a distinctive name, and only one name. Furthermore, the names, once applied, need to be permanent to ensure the stability of the classification. In reality, they often'have no predictability or consistency for retrieval purposes. An example would be a classification based on the hierarchical structure of an administration. The names of the administrative units in the structure may change while the class they represent remains the same, such as the "Ministry of External Affairs" becoming the "Ministry of Foreign Affairs". It is also apparent that, for this process to perform fully and consistently, category labels must be apprehended by all potential users. For example, what the category "dog" entails is known to all English-speakers. Conversely, the category "minutes" is comprehended by fewer people. And a category labelled "John Smith Fonds" can only be understood by a handful of people. Unrecognised categories defeat access. If the intension of the category "John Smith Fonds" is not known to its potential users, it follows that its extension will not be inferred. In archives, walking finding aids fill the gap between obscure categories and potential users of these categories. 13 Indeed, it is not necessary that any one user be able to identify a category's important properties if all members of a repository's constituency share the assumption that there exist at least some experts who possesses the ability to recognise these determining criteria. These experts are the walking finding aids. Consequently, the ability to identify a core concept is not necessary as long as users assume that such a set of essential features does exist and can be identified by expert knowledge.17 Users reconcile their inability to define the essential properties of a category such as "John Smith Fonds" by assuming 1- that these features were present, but hidden from view, and 2- that they could be ascertained by domain experts who possessed the specialised knowledge requisite to apprehend them.18 The expert knowledge of archivists is the knowledge of how they have structured information in their holdings. Or more precisely, how they have organised the documentary units bearing information. This is where the next element of the theory of classification comes into play.-Classifications do not only define membership in and boundaries of classes but also relationships between classes. Classes can be nested in other classes. This concept is familiar to archivists. From our example, the category "minutes" could be part of the category Jacob, "Communication and Category Structure," 85. Jacob, "Communication and Category Structure," 94. 14 "John Smith Fonds". Class i f i ca t ions with more than one level of breakdown has a form tha t i s said to be h i e ra rch ica l . Hierarchies are the usual means of s t ruc tur ing c lasses in archival c l a s s i f i c a t i o n schemes. Hierarchies are systematic frameworks for c l a s s i f i c a t i o n with a sequence of c lasses a t d i f ferent l e v e l s . They are intended to be exhaustive and to c lass a l l the documentary un i t s in a given domain in terms of t h e i r p roper t i e s . Hierarchies work along two axes: hor izontal and v e r t i c a l . Classes belonging to the same h ie ra rch ica l stratum, or l eve l , are discontinuous and mutually exclusive (see f ig . 1) . Each horizontal level of the hierarchy i s occupied by a f i n i t e number of c lasses without organisat ion among them. At each l eve l , the c lasses are non-overlapping. For example, each f i l e t ha t i s par t of a records se r i e s represents a c lass within the c lass "records s e r i e s " . On a given l eve l , proper t ies defining a c lass are necessar i ly d i f fe ren t from a l l other c lasses on tha t l eve l . Consequently, a document, whose place in a c lass was defined by i t s p roper t ies , cannot belong to two such c lasses simultaneously. For archival c l a s s i f i c a t i o n , t h i s means t ha t s e r i e s in a fonds are mutually exclusive and, stricto sensu, no one se r i e s has p r i o r i t y over other ser ies . 1 9 The word "ser ies" i t s e l f t yp i f i e s In theory , each h o r i z o n t a l l e v e l of a h ie ra rchy i s occupied by a f i n i t e number of c l a s s e s wi thout o rgan i za t i on among them. In p r a c t i c e , however, a r c h i v i s t s have ranked s e r i e s . Se r i e s a re given a r e l a t i v e va lue in term of t h e i r es t imated h i s t o r i c a l importance t o r e s e a r c h e r s and ranked accord ingly in f inding a i d s . Sche l lenberg , for example, advises such a method in t h e c l a s s i f i c a t i o n of s e r i e s in Modern Archives: Principles and Techniques (Chicago: Un ive r s i ty of Chicago Press , 1956), 169-70. 15 the aggregates of documentary units it labels: all aggregates labelled "series" are necessarily on the same hierarchical level. Figure 1 Horizontal View of a Hierarchical Classification. Conversely, in the vertical view of a hierarchy, any document is member to a string of classes because each class belongs respectively to each of the successive level of the hierarchy (see fig. 2). The elaboration of a hierarchical classification dictates that all classes are simultaneously included in a superior class and include one or more inferior class. The exceptions are for the class encompassing the entire domain, such as a fonds, and the class defined as elemental, such as a document. The properties defining lower level classes are necessarily properties defining higher level classes. Each lower level class is part of a higher level class in a defined vertical order. Each higher level class is a whole, with the immediate 16 lower class being its part. Each higher level class contains all of its lower level class. For example, a document is part of a file which is part of a series which is part of a fonds. The vertical integration is based on properties shared among all classes, such as provenance. Figure 2 Vertical View of a Hierarchical Classification These two axes represent two fundamental schemata of archival classification schemes. One could even talk of two gestalt. The schema for the horizontal axis is the container.20 Its basic structure consists of an interior, a boundary, and an exterior. The logic of its configuration is very simple: entities are either inside a container or out. Containers can be readily visualised in files, series, fonds, or records groups. The second schema, for the vertical axis, is that of the part-whole. Its basic structure consists of a whole, parts (class/containers), George Lakoff, Women, Fire, and Dangerous Things: What Categories Reveal about the Mind (Chicago: University of Chicago Press, 1987), 272. 17 and a configuration. For example, the parts of a file are the documents; the parts of a series are the files; and the parts of a fonds are the series. The whole exists only if the parts exist in a configuration. That configuration in archival classification is usually the hierarchy. This schema is asymmetric: if A is a part of B, then B is not a part of A. It is irreflexive: A is not a part of A. The two schemata are complementary. Each class is represented structurally by the container schema while the hierarchy of classes is represented structurally by the part-whole schema. The two schemata together are what guide archivists in structuring archival information and users in accessing this information. However, in themselves, these schemata cannot accomplish this. They posit certain principles of order but in themselves they do not structure information in a meaningful way. The schemata structuring the classes could be compared to the scales and diacritic signs on music sheets: it is only meaningful if you have the key in which it is to be played. Similarly, in a classification scheme, the key has to be supplied that will order the information in a certain way. This key and the order that will flow from it are supplied by the classifier. In other words, it may be assumed that information gathered from the perceived world is ordered but the order read is the order projected by the human mind, not an order inherent to it. 18 Classes and classifications are not built into nature, they are inventions. For sure, any definition of a class or classification distinguishes a set of real objective entities. However, the concepts of class and classification, the thing really present in the classifier's mind and named and referred to are invariably subjective. They have no real relationship to anything objective. If we return to the simple syllogisms given as examples above, it is clear that, in themselves, they were not explanatory. To be genuinely explanatory, the inferential process needs to exhibit more than simply a set of class-inclusion properties. "Explanation is intentional: How you pick out the explanatory factors matters to the propriety of the explanations."21 Thus, for an explanation of the fact that class "A" belongs to class "B", we require not only to pick out what we believe to be the widest class which includes the other classes. We also need to pick it out under a defined frame of reference. In the example of the syllogism about the mortality of Socrates, it was the "mortal" property that explained the inferential process and moved the syllogism to conclude that Socrates is mortal. Another property could have been selected as the explanatory factor. The selection of a different property yields different sub-class inclusion results. Priority on some basis or other is, therefore, necessary in hierarchical classification. This selection is the task of the classifier and represents the key of any classification scheme. The key property favoured by Jonathan Barnes, ed., The Cambridge Companion to Aristotle (Cambridge, U.K.: Cambridge University Press, 1995), 110-1. 19 the classifier will define the relationships between classes and, to a certain extent, their boundaries. It is therefore the most basic problem facing the classifier. This key property is selected from among the properties that characterise an entity. For archives, those entities are documentary units. The properties are construed as the data elements that make up the description of those documentary units. In systems of archival control, these descriptions stand for the documentary units they represent. When defining classes and classifications, classifiers do not deal directly with documentary units but with their descriptions. Descriptions are composed of the properties, essential and incidental, that will give an accurate picture of the documentary unit it represents. Classifications are based on descriptions because descriptions are made up of documentary units' characteristic properties. Classification is a matter of deciding which property or attribute making up a description is deemed to be the relevant feature for defining classes. Archival documentary units can be considered under a number of aspects. They are material objects with a content and a context of creation. Each of these aspects represents a number of possible attributes. Experience and practice have selected from each aspect a more or less well defined number of properties deemed relevant for describing and classifying archival documentary units. The core data elements (properties) used to 20 describe archival material have been established for over one hundred years. There is now a list of essential and incidental properties agreed upon by most archivists. The following are the data elements and sub-elements agreed upon by the ICA Ad Hoc Commission on Descriptive Standards22 and also found in the Canadian Rules for Archival Description: Table 1 List of Data Elements from the Ad Hoc Commission on Descriptive Norms IDENTITY STATEMENT AREA Reference code(s) Title Dates of creation of the material in the unit of description Level of description Extent of the unit of description (quantity, bulk, or size) CONTEXT AREA Name of creator Administrative / Biographical history Dates of accumulation of the unit of description Custodial history Immediate source of acquisition ELEMENTS OF DESCRIPTION CONTENT AND STRUCTURE AREA Scope and content / Abstract Appraisal, destruction and scheduling information Accruals System of a r r a n g e m e n t CONDITIONS OF ACCESS AND USE AREA Lega l s t a t u s Access c o n d i t i o n s C o p y r i g h t / C o n d i t i o n s g o v e r n i n g r e p r o d u c t i o n Language of m a t e r i a l P h y s i c a l c h a r a c t e r i s t i c s F i n d i n g a i d s In te rna t iona l Council of Archives, Ad Hoc Commission on Descript ive Standards, International Standards for Archival Description (General), URL f i l e , h t tp : / /www.arch ives .ca / ica /cg i -b in / ica?04_e . 21 APPLIED MATERIALS AREA Location of originals Existence of copies Related units of description Associated material Publication note NOTE AREA Note Normally, it is up to the classifier to decide for each class which properties to use, and among those which ones are essential properties and which ones are incidental. For Canadian archivists, the standard Rules for Archival Description23 already indicate which data elements are deemed necessary in descriptive records and which are optional, that is, which properties are essential and which ones are incidental. From this pool of properties, the key properties that will define the class/container (i.e., the aggregates of documentary units) are selected by the classifying archivist. It is at this point that the theory of classification comes to bear. To structure information is to categorise it. To categorise is to define the boundaries of categories. To define these boundaries is to define the key properties of these categories. Historically, out of all the properties that are used to describe Rules for Archival Description (Ottawa: Bureau of Canadian Archivists, Planning Committee on Descriptive Standards, 1990), 0-3, 0-9. 22 archival documentary un i t s , only two have been u t i l i s e d as key p rope r t i e s . There are two basic approaches t ha t wi l l determine which property wi l l be used as the key property. These approaches correspond to the two basic approaches used to c lass i fy and access information: what was the documentary uni t about? who was the documentary uni t created by?24 The f i r s t i s known in archival science as the p r inc ip le of per t inence, the second as the p r inc ip le of provenance. Hardenberg, reviewing archival p r inc ip les of c l a s s i f i c a t i on and arrangement, concluded tha t there was l i t t l e new under the sun in t h i s area. Early in t h i s century, he ref lec ted t ha t : "Meisner had himself dea l t with t h i s question by asser t ing tha t bas ica l ly there are only two pr inc ip les of arrangement, t ha t of provenance and tha t of pert inence, upon which a l l other forms of arrangement are variations..."25 The two are not mutually exclusive; however, for a system of c l a s s i f i c a t i o n to have consistency, one must be dominant over the other . One of the two must define the domain c lass tha t wi l l comprise a l l the other c l a s se s . The other main reason for se lec t ing a key property a r i s e s from the fact t ha t , to t h i s day, most archival r e t r i e v a l systems Subject Indexing for Archives: The Report of the Subject Indexing Working Group, p u b l i c a t i o n no. 4, (Ottawa: Bureau of Canadian A r c h i v i s t s , Planning Committee on Desc r ip t ive Standard, 1992), 2. 25 H. Hardenberg, "Some Ref lec t ion on t h e P r i n c i p l e for the Arrangement of Archives ," Der Archivar 16 (1963), reproduced in Modern Archives Administration and Records Management: A RAMP Reader, compiled by Pe te r Walne with the a s s i s t a n c e of a working group of the I n t e r n a t i o n a l Council on Archives, t r a n s l a t e d by F . J .C . Kavelaar ( P a r i s : General Information Programme and UNISIST, 1985), 111. 23 are presented to users with paper-based finding aids. Determining which is the key property of archival documents determines how they are to be organised and which paths are to be mapped out for access. Among the essential properties, the one selected as the key property will also act as the main access point. It must be selected because descriptions must be ordered. It will define the boundaries of the domain class. The descriptive records can help users find the documentary units they need only if they themselves can be found by users. They have to be arranged just as documents in a folder along alphabetical or chronological lines, depending on the data element selected by the arranger. The containers of descriptive records are catalogues or inventories but the principles are the same. The finding aid systems developed and used in archives are to a large extent the result of the form and limitation of paper-based finding aids. They are bound by an order of presentation and that order is predicated on the key property selected by the classifier. The next chapters endeavour to describe how these classification systems have been developed and used, particularly in public archives. It should be kept in mind that, to a large extent, issues in archival classification have to do with explaining the classes found in a culture and coded by that culture at that time. The classifier is as much defined by his time and place as his classification schemes are defined by him. A major influence on how properties are defined and attributed by classifiers is the information systems already in place in a 24 culture at a given time. Nineteenth century issues of historiography, evolutionism, and positivism played a major role in shaping archival classification systems. Twentieth century concerns over the management of current administration and the mass of records they produce were also unavoidable factors in defining archival classification. Therefore, in a very archival fashion, this study will not only look at the contents of archival classifications, but also at their context of development. 25 CHAPTER TWO: THE PERTINENCE MODEL A general definition of what constitutes archival classifications based on pertinence is that they classify document by their contents. They group documents on the basis of commonality of subject. Today, pertinence-based classifications are widely discredited in the archival world. It is seen as anathema to good archival practice. However, it certainly did not appear that way to archivists in the late eighteenth and early nineteenth centuries, when pertinence-based classification schemes were in their heyday in public archives. To group documents based on their subject content is a practice that seemed the most immediate and natural to archivists interested in easing access. It conformed so well with the mind set of the times that it was the norm rather than the exception. From the point of view of archivists, pertinence was the most appropriate type of classification to implement for effective intellectual control of their holdings. The inner workings, and past popularity of pertinence-based classifications are better understood in historical context. The Zeitgeist is as important to the understanding of classification at this time as are the principles of the theory of classification. Archivists were then at work when the tradition 26 of the Enlightenment's Encyclopedists still inspired efforts at universal codification systems. "It was the period of great scientific systems of classification."26 The age of Enlightenment had been preoccupied with classifying objects, in finding for them the right place in a taxonomy that would group like things together. Archivists formed in this school of thought conceived of the classification of archives as the distribution of documents into pre-determined classes. The common link, or key property, among documents was found within their contents, not outside in their context of creation. It corresponded to the view of the time that everything was to be clearly and logically defined. By definition, pertinence-based classifications are based on hierarchies and fit perfectly with the idea of schemes that could provide a canvas for structuring all knowledge. At its broadest, the domain class for pertinence-based classifications can be all human knowledge, or in the case of archival material, all human actions. Classifications based on pertinence were a perfect fit for the world view of the time. The tradition of the Enlightenment provides the general framework for an understanding of pertinence. However, if the spirit of the age made pertinence the preferred classification scheme, a number of circumstances made its implementation 26 Duchein, Michel. "Theoretical Principles and Practical Problems of Respect des Fonds in Archival Science." Archivaria 16 (Summer 1983): 72. 27 poss ib le . They were, p r inc ipa l ly , the delegation of archives to public r epos i to r i e s a f te r the French Revolution and the Napoleonic wars, the replacement of a rch iv i s t s / admin i s t r a to r s with a r c h i v i s t s / h i s t o r i a n s and the autonomy and s e l f - r e f e r e n t i a l nature of documents/monuments. The l a t t e r was the most immediate factor t ha t made c l a s s i f i c a t i o n based on pert inence poss ib le . At the time of the Enlightenment, how documents were perceived was a function of t h e i r speci f ic j u r i d i c a l context. In the eye of the law, there were two categories of records: the j u r i d i c a l l y re levant document ( ju r id i ca l instrument) and the j u r i d i c a l l y i r r e l evan t monument (h i s t o r i ca l a r t i f a c t ) . 2 7 The dual notion of document and monument was based on legal considera t ions . The s t a tus of the document was a function of i t s j u r i d i c a l value as evidence in a court of law. Authentic acts28 epitomized t h i s type of document. I t was autonomous and au tho r i t a t i ve by i t s e l f . I t was the memory of a fact or event and bore witness to i t . The document contained in i t s e l f the potency to perform whatever action i t was created to perform. So t h a t , " . . . i t was possible to go d i r ec t l y from the document to the e n t i r e fact generating i t . . . there was a b i l a t e r a l 27 Amedee Outrey, "La not ion t r a d i t i o n n e l l e de t i t r e s e t l e s o r i g i n e s de l a l e g i s l a t i o n r e v o l u t i o n n a i r e sur l e s a r c h i v e s , " Revue historique de droit frangais et Stranger 32, s e r . 4 (1955), 438-9. 28 An a u t h e n t i c a c t i s defined as "an a c t which has been executed before a notary or pub l i c o f f i c e r au thor ized t o execute such func t ions , or which i s t e s t i f i e d by a pub l i c s e a l , or has been rendered pub l i c by t h e a u t h o r i t y of a competent m a g i s t r a t e or which i s c e r t i f i e d as being a copy of a pub l i c r e g i s t e r . " Black Law Dictionary, 4th ed. r e v . , s . v . " au then t i c a c t " , 168. 28 re la t ionsh ip between each document and the fact i t was about, so tha t i f a fact , (A), was manifested in wri t ten form, the document r e su l t ing from i t (B), wi l l guide us d i r ec t l y to the fac t : A--> B--> A."29 The document did not merely represent a fact ; i t embodied i t . I t was the vessel keeping present an action tha t was to survive the passage of time. The document kept in an archives30 was, therefore , a r e l i a b l e memory and a r e l i a b l e witness. An act ion had occurred only as long as i t s memory was maintained.31 Consequently, the j u r i d i c a l l y re levant document insured the durab i l i t y of the event i t embodied. In t h i s function, the document had taken over from ora l testimony. Oral t r a d i t i o n had provided a chain of people to keep a l ive the memory of s ign i f i can t facts and events. Each l ink kept the memory a l ive and could bear witness to i t . Gradually, however, the wri t ten document came to be recognized as a t rus ted witness and to possess the same j u r i d i c a l q u a l i t i e s as the eyewitness. Documents Luciana Duran t i , "Diplomatics: New Uses for an Old Science (Par t I I ) , " Arch ivar ia 29 (Winter 1989-90), 15. 30 This unders tanding of document had d i r e c t consequences on t h e not ion of a r c h i v e s , as guardian of t h e i r r e l i a b i l i t y . The d e f i n i t i o n of a rch ives was a c o r o l l a r y of t h a t of document. Archival r e p o s i t o r i e s were not t h e r e t o ensure the p r e s e r v a t i o n of the contex t of c r e a t i o n but the l e g a l i t y of the a c t i o n s born by the documents. By secur ing t h e custody of t h e documents, a r ch ives were ensur ing , in the eye of t h e law, the v a l i d i t y of the document they kept . This unders tanding of a rch ives showed how i t was i n t i m a t e l y l inked with the r u l e s of evidence in law. Archives aimed a t ensur ing t h a t documents were t h e t r u e w i l l of the a u t h o r i t y t h a t was purpor ted t o have w r i t t e n them. Archives p h y s i c a l l y ensured a u t h e n t i c i t y . While t h e document was s e l f - s u f f i c i e n t in bear ing wi tnes s , t he a rch ives ensured t h a t i t was a t ru s twor thy w i t n e s s . 31 For example, Michel Duchein remarked t h a t one of t h e reasons behind t h e mass d e s t r u c t i o n of documents a f t e r t h e French Revolution was not only t h e need t o e rase the memory of t h e feudal system but a l s o t h e a c t u a l d e s t r u c t i o n of i t s r i g h t s ("La r evo lu t i on f r a n c a i s e e t l e s a r c h i v e s : La memoire e t I ' o u b l i dans 1 ' imagina i re r e p u b l i c a i n . " In Etudes d'archivistique, 1957-1992, P a r i s : Assoc ia t ion des a r c h i v i s t e s f r a n c a i s , 1992, 61-4) . 29 could "bear witness on their own."32 Each act contained in itself its context of creation.33 Since they were autonomous entities, documents did not need to be grouped into classes. They were usually simply arranged chronologically. This type of arrangement showed the succession of events along the linear passage of time. The notion of monument was complementary to that of document. Monuments were documents that were deemed to have lost their juridical weight. They usually consisted of obsolete acts or copies of acts kept for their historical interest. While legal value was the primary reason for keeping documents, historical value became the primary reason for keeping monuments. However, the fact that they were now preserved for their historical value rather than their legal value did not change their nature in the eye of contemporaries. They were still autonomous documents. The autonomy of the document meant that they were still treated as discrete items. There was no need to make note of their context of creation. As former documents, they embodied that context and were still self-referential. In this respect, Michel Duchein noted how the value of a document changed to transform it into a monument while maintaining its autonomy as a reliable witness of the events it represented. Duranti, "Diplomatics (Part II)," 4. 33 Duchein remarked in "La Revolution frangaise et les archives," 63, that the very name of the Ecole des chartes, created in 1821 to train diplomatists, shows how much this understanding of document was still predominant at the beginning of the nineteenth century. Archivists were still thinking in terms of individual documents(chartes), not in terms of aggregates such as series or fonds. 30 "...the archival document was considered to possess an interest by itself, independent of its context, in the same way that, in archaeological digs...one was interested in the objects of art as collector's items without preserving them in the context of their discovery."34 Consequently, arrangement imposed on documents by creators did not need to be maintained by archivists because it was not perceived as being meaningful in itself. Contextual information was not viewed as enriching the informational value of monuments. If anything, it hindered retrieval for historical research. This introduced not only the possibility but the need for re-classification. Archivists would group monuments around any property they deemed meaningful. That property was usually content or subject as it seemed to answer best the perceived needs of researchers. This perceived double nature of records as document/monument would be pivotal when the French Revolution and the Napoleonic wars brought changes of an unprecedented scale to the role and nature of public archives. In themselves these historical events did not affect or influence pertinence-based classification but they made its applications possible in a whole new sphere. The method of classifying archival material remained the same. In this sense, the marrying of old methodology with new circumstances would have a radical impact on archival classification. 34 Duchein, "Theoretical Principles and Practical Problems," 65-6. 31 With the exception of Great-Britain and Russia, the Revolution brought about the disappearance of the old governmental, administrative, and juridical structures of Europe. By 1815, public archives not only in France, but also in Italy, Germany, Spain and the Netherlands had a new face.35 The fundamental change these events created was the abolition or nationalization of most of the institutions of the Ancien Regime. The archives of the abolished institutions were now placed under the responsibility of the new governments. Through nationalization or appropriation these archives now belonged to the State. The Revolution and the Napoleonic wars made public archival institutions the custodian of a large number of these delegated archives. This meant the transfer to government repositories of all those records now the property of the State. The result was a sudden influx of documents rendered obsolete by the abolition of the authority from which they had emanated. With the abolition of the rights of the Ancien Regime, there was no need for the new administration to keep all the documents that had been preserved by the old administrations to protect and enforce rights established under it. Even after the destruction of large quantities of records for political or economical reasons, there remained still a vast number of records kept mostly for historical reasons. In the process, public archives were transformed from passive preserver of documents into Ernst Posner, "Some Aspects of Archival Development Since the French Revolution." In Archives and the Public Interest, edited by Ken Munden, 23-35. Washington: Public Affairs Press, 1967, 26. 32 cu l tu ra l i n s t i t u t i o n s ac t ive ly co l lec t ing monuments. In France, for example, the new l e g i s l a t i o n regarding archives introduced a f te r the Revolution acknowledged the r e spons ib i l i t y of the s t a t e in caring for i t s documentary heri tage.3 6 I t meant tha t more and more records acquired the digni ty of nat ional monuments and would need to be w r e -c l a s s i f i ed" . With the ro le of archives now changing from adminis t ra t ive to h i s t o r i c a l arsenal , so did the ro le and preoccupation of a r c h i v i s t s . While documents needed the care of an a r c h i v i s t / j u r i s t , monuments needed the care of an a r c h i v i s t / h i s t o r i a n . Before the Revolution, a r ch iv i s t s were adminis t ra t ive or j u r i d i c a l o f f i c i a l s who supported and defended the claims of t h e i r sponsors. By the end of the eighteenth century, a school of h i s t o r i ans , influenced by the work of Jean Mabillon on diplomatics, had found a new i n t e r e s t in the study of o r ig ina l documents.37 When p o l i t i c a l events demanded the appraisal and c l a s s i f i c a t i on of masses of old documents, i t ra ised the demand for a r ch iv i s t s who were h i s t o r i c a l scholars ra ther than administrators.3 8 The a r ch iv i s t s t ra ined in administrat ion or I n s t i t u t i o n s car ing for "documentary h e r i t a g e " p re -da t ed t h e Revolut ion. The d i p l o m a t i s t Tess ie r noted in La Diplomatique, 3rd ed. r ev . Co l l ec t ion "Que s a i s - j e ? " , no. 536 ( P a r i s : Presses u n i v e r s i t a i r e de France, 1966), 29, t h a t "en 1762, l e con t ro l eu r genera l Ber t in p rovoqua . . .une dec i s ion de 1 ' a u t o r i t e roya le en vue de c e n t r a l i s e r dans un depot c e n t r a l , d i t Cabinet des c h a r t e s , l e s copies de documents dont une enquete uniquement subordonnee a des f ins h i s t o r i q u e s e t methodiquement pour su iv ie s dans tous l e s c h a r t r i e r s du royaume s i g n a l e r a i t l ' i n t e r e t . " The new S t a t e s continued t h i s p r a c t i c e , a l b e i t on an unprecedented s c a l e . 37 Maynard Bichford, "The Provenance of Provenance in Germanic Areas , " Provenance 7, no. 2 (1989), 56. 38 Brichford, "Provenance of Provenance in Germanic Areas," 62. 33 registry work were gradually replaced by archivists trained in history or librarianship. In the new state archives, archivists and users had the same preoccupation. It also meant that the new archivists were now chosen from the rank of archives' users. Both archivist/historians and users were in tune with the Romantic trends in historiography at the beginning of the nineteenth century. As Posner noted: "The needs of scholarly investigation and research work were held so preponderantly important that it seemed obvious that records should be arranged and catalogued in a manner that would facilitate every kind of scholarly use."39 The state of mind of historian/archivists was that of collectors or antiquarians. This change in value was to have important consequences not only for what was to be kept but also for how it was to be kept. Classification and retrieval systems able to meet the needs of historical researchers became a preoccupation of the new archivists.40 Consequently, European archivists quite consciously classified and described archives for the purpose of historical research. To archivists of the time, the grouping of documents by creating bodies imposed old and superfluous divisions that impeded research. Researchers had to repeat their search on a particular subject as many times as there were fonds.41 To be exhaustive, a search on a given topic 39Posner, "Aspects of Archival Development," 30. 40 Michel Duchein, "L'Histoire des archives europeennes et 1'evolution du metier d'archiviste en Europe," in Etudes d'archivistique, 1957-1992 (Paris: Association des archivistes francais, 1992), 70. On the distinction between "archives" and "fonds", A. Matilla noted that 34 had to be conducted for each delegated fonds. Furthermore, users' information requests were usually content oriented, targeting a specific subject, person or event, not a records creator. Classifying all documents according to a uniform plan could only facilitate research.42 Archivists believed that the best and proven way to provide access to the records under their care was by re-classifying these records according to the subjects most likely to interest historical researchers. By so doing, the archivists of the time were not so much superimposing a new scheme on original classifications as replacing classifications that were deemed irrelevant to them or their clients. The original classification or arrangement had been meant to serve the needs of the administration; a new classification was needed to serve the needs of historical researchers. Consequently, systematic classification schemes were developed, usually covering the whole of a repository, and the records were marshaled into them. Pertinence not only facilitated access for researchers, it also helped archivists manage their holdings. The problem with the multiplicity of fonds was a scattering of information, both physically and intellectually. Since pertinence-based classification also implied physical arrangement, it solved both "le mot fonds est la denomination que l'on substitue au terme archives quand les archives d'un organisme viennent faire partie d'autres archives plus vastes. Quoted in Manuel d'archivistique frangaise, 188, fn 1. 42 Robert-Henri Bautier, "La phase cruciale de l'histoire des archives: la constitution des depots d'archives et la naissance de 1'archivistique." Archivum 18 (1968): 147. 35 problems. Pertinence classification not only brought intellectually together monuments with a common subject but also arranged them physically together in the same sequence. Pertinence implied the analysis of each individual document or dossier in order to find its place in a preconceived classification scheme. This.was a labour-intensive task. Fortunately for users, the archivist/historians were willing to devote a lot of their time to classifying or re-classifying the records under their care. The advantages of such a system for both archivists and researchers were obvious to them. It cut down search time for users and retrieval work for archivists: all relevant documents were arranged together and they could all be retrieved as one unit. Using pertinence to classify archival material was in keeping with the tradition of past archivists or librarians. It is true that the autonomy and self-referential nature of monuments allowed archives to be treated very much like books in a library.43 However, pertinence had been the way to organize records in administrative archives. Archivists simply transposed their past practices to public archives. The context of those past practices was that an administrative archives had usually held the records of one and only one administration. Archivists had always dealt with the holdings of a repository as a whole. 43 Brichford, for example, has posited that the pertinence model used in archives was directly inspired by librarianship theory and practices in "Provenance of Provenance in Germanic Areas," 69. 36 Since administrative archives represented one fonds, that had not been a problem: the classification and arrangement scheme was applied uniformly across that fonds. The sudden influx of records from different quarters did not change this practice. Delegated archives were re-classified because archivists did not adapt to this new reality. It could be said that the change came from the lack of change in classificatory practices. Archivists applied an old solution to a new problem. Classificatorily, pertinence presented users with a comprehensive scheme. Theoretically, the intension of the domain class for a pertinence-based classification was human activity. In practice, of course, the domain class was limited by the extent of the holdings and the collecting bias of archivists. If all of human activities can be fodder to historical research, not all of them can be expected to be represented in the holdings of a repository. Of all the human activities, only some would be deemed worth documenting by archivists. These factors would necessarily constrict the extent of the domain class. The same restrictions would apply to the extension of this class. Its virtual extension would be all archival monuments. The actual extension was the extent of a repository's holdings. Sub-classes were formed within the class domain around what an archives actually had acquired. To ensure retrieval consistency and accuracy, the pertinence model assumed an in-depth knowledge of the classification scheme on the part of the classifier. This is the Achilles' heel of the pertinence model. The essential 37 feature that determined category membership and constituted the definition of a sub-class was subject, as defined by a classifier. Consequently, the membership in a sub-class was not fixed by a set of context-independent criteria, but was dependent solely upon immediate context, goals, and individual experience of the classifier. As a consequence, the stability of the reference was undermined and "the possibility of effective communication between classifier and users was impoverished."44 It was extremely difficult, if not impossible, to maintain a consistent classification over a whole holding and over time. The users of pertinence-based classification perceived this weakness to be mainly of an administrative nature and not an inherent flaw. In the context of the time, classification by pertinence worked for archivists and researchers. It met their needs for both intellectual and physical control. There came a time when it was felt that classifications that had been made from the point of view of pertinence had to be drastically revisited or even replaced to meet the needs of another point of view. A changing context brought new needs and expectations. The practice of historical research was moving away from Romanticism and its proclivity for things Gothic. Archivists, the faithful servant of historians would follow suit. What had been qualities became faults. Pertinence, for example, did not keep track of the source 44 Jacob, "Communication and Category Structure," 87. 38 or context of creation of records. This became a liability of this type of classification. When historiography shifted away from Romanticism, it became apparent to users and archivists that this type of retrieval system meant the loss of the archival bond and the loss of context of creation. A new type of classification was needed that would take into account these elements. The new point of view was that: "[Pertinence] systems were appropriate for the records maintained and used by the [records] creator, but their superimposition on inactive records...with their own original arrangement given by their creator could only cause havoc. "45 45 Luciana Duranti, "The Odyssey of Records Managers" in Canadian Archival Studies and the Rediscovery of Provenance, edited by Tom Nesmith (Metuchen, N.J.: Society of American Archivists and Association of Canadian Archivists in association with the Scarecrow Press, 1993), 51. 39 CHAPTER THREE: PROVENANCE-BASED CLASSIFICATIONS A general definition of what constitutes archival classifications based on provenance is that they classify documents by their source. Provenance dictates that the records of each agency or person be kept together, or, if they are dispersed, that they be reunited. Provenance is generally considered to cover two principles of classification: respect des fonds and respect de l'ordre original. An examination of respect des fonds and respect de 1'ordre original in context is necessary to show the importance and limitation of these concepts for mid-nineteenth century archivists at work when the tradition of the Enlightenment's Encyclopedists still inspired efforts at universal codification systems. The changed world view that turned the advantages of classification by pertinence into shortcomings was slow to evolve. The solution was also slow in its development and implementation. Respect des fonds and respect de l'ordre original were devised at different times in response to different concerns. The first one to be developed was respect des fonds. Most histories of archives mark 1841 as a turning point in archival theory on classification when Natalis De Wailly, of the French Ministry of the Interior, issued a circular entitled Instructions pour la mise en ordre et le classement des archives 40 departementales.46 The c i r cu l a r of 1841 i s held as a tangible break with the pert inence model of c l a s s i f i c a t i o n by preconceived categories for documents of d i f fe ren t o r ig ins . Respect des fonds i s seen as the new scheme introduced to replace systems of c l a s s i f i c a t i o n inher i ted from the eighteenth century. The c i r cu la r to French archives departementales explained how to c lass i fy records by provenance as follows: "II convient de fo rmule r . . . l e pr incipe e t l e s elements de la methode a suivre dans l e classement: "1° rassembler les d i f fe ren ts documents par fonds, c ' e s t - a - d i r e former co l lec t ion de tous les t i t r e s qui proviennent d'un corps, d'un etablissement, d'une famille ou d'un individu, e t de disposer d 'apres un cer ta in ordre l e s d i f fe ren ts fonds; 2° c lasser dans chaque fonds les documents suivant les matieres, en assignant a chacun un rang p a r t i c u l i e r ; 3° co-ordonner l e s matieres , selon les cas, d 'apres 1'ordre chronologique, topographique ou simplement alphabetique. "47 The t rue innovation of the c i r cu l a r was i t s c lear formulation of the p r inc ip le of provenance as a solut ion to growing problems with the i n t e l l e c t u a l and physical control of archival holdings. However, i t did not introduce the concept of provenance to public archives . Respect des fonds had been formulated before 1841 in cer ta in c i t i e s and s t a t e s of Europe such as Halle (Saxony) in 1777, Denmark in 1791, Naples in 1812, The c i r c u l a r was aimed s o l e l y a t archives departementales, not t he Archives nationales. During the f i r s t four decades of the n ine t een th cen tury , su rv iv ing p r o v i n c i a l records of t h e ancien Regime remained untouched in l o c a l q u a r t e r s . De Wail ly, who himself had worked a t t he Archives n a t i o n a l e s , knew t h e problems c rea ted t h e r e by c l a s s i f i c a t i o n by p e r t i n e n c e . The c i r c u l a r aimed a t p revent ing a r epea t of t h i s s i t u a t i o n in archives departementales. 47 " I n s t r u c t i o n s pour l a mise en o rd re e t l e classement des a rch ives depar tementa les , 1841," quoted in Manuel d'archivistigue frangaise, 208. 41 or the Netherlands in 1826.48 This evidence suggests that the acceptance of the principle of provenance was a slow process and not the sudden result of a regulation. The circular was not so much a break with the past as the end point of a slowly evolving processr Provenance or organization according to source had been a normal practice. Archivists had kept records according to their origin in chancelleries or treasuries for centuries. Provenance had been a significant factor in the authentication, appraisal, and description of archives as well as their classification.49 The preservation of provenance, as an essential property of documents, however, had not been a preoccupation before. This would change as a number of new factors came to bear on public archives. To use a hackneyed expression, public archives underwent a "paradigm shift" in the first half of the nineteenth century. They experienced a shift in mission from administrative to cultural institutions. The transfer of government records had been neglected and the link between archives and administration had rapidly vanished. The profound changes of the years 1789-1815 had cut that link. In effect, public archives did not have to worry about caring for current records until almost the middle of the nineteenth century. For example, in Spain, the Archivo Historico Nacional, created in 1866, dealt exclusively with the 48 Antonia Heredia Herrera, Archivistica general: teoria y practica, 5th ed. (Sevilla: Publicaciones de la Exma, Diputacion provincial de Sevilla, 1991), 33. 49 Brichford, "Provenance of Provenance in Germanic Areas," 55. 42 records of defunct organizations. In England, the Public Record Office, created in 1838, went decades without regularly acquiring new records.50 In France, under the new Napoleonic regulations, records were to be kept in administration for 40 years before being considered for transfer to archives. This attitude towards "modern" records changed as changes in historiography put these records in a new light. The gradual acceptance of historical growth as an important concept in historiography helped reestablish the link between public archives and government administrations. The fault line between documents and monuments was moving ever more closely to the present. It became, therefore, more and more evident that in order to insure the preservation of the government's historical records, public archives would have to play an active role. The preservation of these records was too important to be left to chance or to administrators. It was realized that, in order for archival repositories to be historical institutions, they had to take in the records produced by current agencies and administrations. This meant an ever mounting quantity of records to process. Pertinence-based classification did not cope well with this new reality because it only processed records on an individual basis. Provenance was to answer these very practical problems facing public archives. The objectives of the circular Duchein, "L'Histoire des archives europeennes," 72. 43 were to guide departmental archivists towards a standardized practice of classification and description. To ensure that these objectives be met, the Ministry of the Interior approved, one month after the publication of the circular, the creation of a Commission des archives departementales et communales to oversee its implementation. Natalis de Wailly, author of the circular, was also one of its members. When, at an early session of the Commission, objections were made to the type of classification adopted under the circular, de Wailly spelled out what the objectives he hoped to achieve by using provenance: "...le classement general par fonds et par matieres est le seul vraiment propre a assurer le prompt accomplissement d'un ordre regulier et uniforme. II offre plusieurs genres d'avantages; avant tout il est plus facile qu'aucune autre methode a mettre en pratique: car il ne consiste d'abord que dans un simple rapprochement de pieces dont il s'agit uniquement de discerner l'origine; puis lorsqu'un premier triage est fait, la division par matieres s'opere naturellement et sans peine, puisque la quantite des pieces de chaque fonds est limitee. Ce classement, dans un grand nombre de cas, est d'autant plus facile qu'etant la reproduction de celui des anciens chartriers, il peut etre opere d'apres les anciens inventaires, et des lors il suffit de faire au moyen des cotes un recolement des documents inventories pour les retablir dans leur ordre primitifs. Si au lieu de cette methode, qu'on peut dire fondee sur la nature des choses, on propose un ordre theorique, tous ces avantages seront perdus: les archivistes n'auront pas seulement a faire, comme dans le classement par fonds, une reconnaissance en grande partie materielle, ils devront apprecier la valeur des documents, resoudre une foule de problemes que doit faire naitre a chaque instant 1'application d'un principe systematique, et enfin dans le cours d'une operation qui embrassera non les papiers d'un seul fonds mais la masse des fonds, ne perdre jamais de vue le but de la methode...Alors, si 1'on vient a commettre des erreurs graves, les archives tomberont dans le 44 desordre auquel il sera difficile de remedier...Les recherches d'erudition ayant ete jadis faites dans les archives d'etablissements distincts les uns des autres et par consequent classes par fonds, un nouveaux classement romprait la tradition qu'il est si desirable de conserver entre les travaux de 1'erudition ancienne et ceux de 1'erudition contemporaine...Il faut enfin observer que le classement par fonds n'empeche pas d'obtenir tous les resultats qu'on espere d'un classement d'un ordre plus eleve. II ne s'agit pour cela que d'obtenir des inventaires bien faits. Au moyen de ces inventaires, on pourra faire des releves dans toutes especes de systemes et rediger des tables de toute nature."51 Clearly, for de Wailly and the Commission, there were two groups of practical advantages to the replacement of the pertinence model with provenance-based classification: intellectual control and physical control. The first group of advantages brought by provenance related to improved access. Provenance allowed archivists to avoid some of the pitfalls of pertinence when it came to ordering archival material. Pertinence was as much about physical arrangement as it was about classification. The two were inextricably entwined in archival practice: to classify was to arrange, and to arrange was to classify. Consequently, the depth of indexing was necessarily limited to a single heading. A document could be physically filed in only one place. This place was determined by what archivists considered was the document's main subject. This was one of the weaknesses of pertinence-based classifications. If the archivists content analysis was wrong, a document could be misclassified Quoted by Gustave Desjardins in Le service des archives departementales: conferences faites aux Sieves de 1'Ecole des chartes, les 10, 18, 25 et 30 juin 1890 (Paris: E. Bourloton, 1890), 30-1. 45 and, consequently, misf i led. If the a rch iv i s t was not cons is tent in his ana lys i s , documents with a common main subject could be dispersed in a reposi tory among di f ferent headings and d i f fe ren t loca t ions . Provenance a l l ev ia t ed t h i s problem by moving the level of c l a s s i f i c a t i o n down from the level of an a rchives ' holdings to the level of the fonds. Now, content analysis did not have to be consis tent across the whole of a r epos i to ry ' s holdings, but only in r e l a t i on to a given fonds. The chance for misf i les or dispersion was diminished and access improved. Despite the adoption of provenance, the ideal of across-holdings indexing was not abandoned. The best means to access information was s t i l l by c lass i fy ing documents along a uniform scheme.52 However, t h i s was now done in the confines of the fonds, in r e l a t i on to the a c t i v i t i e s of the i n s t i t u t i o n or administrat ion tha t created the records. The c i r cu l a r did not t o t a l l y r e j ec t pert inence but only cer ta in aspects of i t . I t seemed apparent from the object ives quoted above tha t provenance was to supplement per t inence, not replace i t . Provenance allowed a rch iv i s t s to sh i f t the leve l where they would implement c l a s s i f i c a t i on by per t inence. Instead of c lass i fying documents by pert inence across the whole of a r epos i to ry ' s holdings, they would l imi t i t to the boundaries of each fonds. Below the fonds l eve l , pertinence was s t i l l viewed as In regard t o r e spec t for o r i g i n a l o rder , Baut ie r in "Phase c r u c i a l e de l ' h i s t o i r e des a r c h i v e s , " 147, has drawn a t t e n t i o n t o the fundamentally d i f f e r e n t a d m i n i s t r a t i v e p r a c t i c e s of Roman and Germanic c o u n t r i e s . Archival t r a d i t i o n s a re of n e c e s s i t y grounded in these p r a c t i c e s . Since t h e seventeenth century , t he work of French, I t a l i a n or Spanish a r c h i v i s t s had been t o put o rder in fonds made up of d o s s i e r s accumulated in ca r ry ing out bus ines s wi thout the b e n e f i t s of a c l a s s i f i c a t i o n p lan . Transfers were made t o a r c h i v a l r e p o s i t o r i e s wi thout apparent o rde r . I t i s up t o the a r c h i v i s t s t o organize and c l a s s i f y the f i l e s . Respect for o r i g i n a l order was more e a s i l y app l i ed in c o u n t r i e s using a Registratursystem. 46 a good classification scheme that yielded excellent returns for historical research. It would appear that there were no perceived faults in classification by pertinence, only in the ability of archivists implement it properly. The means of classification recommended, below the fonds level was still by matieres, or subject. This view may be due to the fact that, for de Wailly and the Commission, classification by provenance did not change their view of documents as autonomous entities. The term "deeds" clearly indicated that they were still dealing with and classifying individual documents. Moreover, since documents/monuments were seen as autonomous and self-referential entities, there was no context to preserve beyond that linking the records to the records creator. At this point, the archival bond was not yet a concern. For these reasons, provenance was not so much displacing the pertinence model as easing its implementation, at least, at the sub-class level. De Wailly mentioned that finding aids could eventually be created that would pull together documents related to a given subject. Of course, these "releves" or "tables" would only be done on paper and would not affect the classification or the arrangement of the documents. A further advantage of provenance was that it helped maintain the link with past scholarship. The old call numbers and numbering schemes, used by the records creator, had also been used as references to source material in publications. 47 Classification by provenance preserved this link and also saved archivists some work. The second group of advantages brought about by provenance were related to economic benefits. Certain economic advantages could be derived from recycling the information and the work put into the old inventories already existing for some of the fonds transferred to archival repositories. Where possible, archivists were not to waste time in analyzing documents for their classification but to follow existing ordering schemes. Provenance was believed to be easier to use and, if de Wailly did not write that less skilled people would be able to implement it, he certainly hinted that the learning curve would be less steep. Furthermore, this system was intended for archivists in archives departementales where they would be under little or no direct supervision.53 In these conditions, it was preferable to use a classification scheme that was less prone to interpretation than the pertinence model. Archivists needed only to follow the three easy steps outlined in the Circular. A further advantage was that it sped up the processing of large quantities of records. Accruals from a records creator could be simply linked with previous accessions from that creator under the umbrella of fonds. Bartlett, "Respect des fonds," 110. 48 On the surface, provenance appears rather straightforward. Brenneke, for example, remarked, "the new principle of respect was not scientific but rather practical."54 By this, he meant that the circular aimed at answering immediate needs of the administration and was not intended to present a profound rethinking of archival theory. The immediate purpose of the circular seemed in line with this interpretation. Underneath the surface, however, the introduction of the concept of fonds constitutes a pivotal change for archival classification. It represents a profound remodeling of the intellectual classification of documents in public archival repositories. The fonds introduces a new domain class on which to build a classification. With pertinence, the domain class had been the sum of all and any archival monuments deemed worth preserving. The essential property that defines these domain classes is also radically different. With pertinence, the essential property was based on documents1 contents. The sub-classes nested in that domain class were defined by the perceived interests of historical researchers: militia, biographica, ecclesiastica, etc. The fonds brought a new domain class defined by a new property: they all came from a common records creator, a common origin. The act of grouping documents on the basis of origin implies an understanding of what an "originator of records" was. The limits of this new domain class had to be defined to be 54 Adolf Brenneke, Ein Beitrag zur Theorie und Geschichte des Europaischen Archiwesens (Leipzig: Koehler & Amelang, 1953), 62. 49 respected. Unfortunately, the nineteenth century did not have a clear definition of what constituted a records creator. It apparently seemed so self-evident to contemporaries that it did not warrant an explanation. At best, as in the principles quoted above; examples of types of fonds are given: corps, etablissement, famille, individu. We are offered a number of possible containers to define what a fonds is. Of course, a list cannot work as a definition. What is missing is the quality that made them these particular instances of the class "originators of records". However, from rule number one, it can be inferred that a fonds is synonymous with the authority to create records. Competence is, thus, the defining quality of a records creator. It is at the core of the intension or definition of any fonds. The boundaries of a fonds are defined by the boundaries of its area of jurisdiction. Since the intension and extension of a fonds overlap, it follows that this feature will also determine the membership of documentary units in a fonds. What defines a domain class defines the entities that make up that class. By relying on the concept of fonds, the foundation of the principle of provenance appears to be the records creator as juridical entity. The concept of fonds not only brought profound changes to the intellectual control of archival material but to its physical management as well. In this area, provenance provides archivists 50 with aggregates of records of a functional scale. With classification by pertinence, the holdings of an archival repository had represented one unbroken whole to classifiers. The boundaries for the class domain were those of the repository itself. Under provenance, the boundaries of the class domain are those of the fonds.. This allows archivists to work on smaller, manageable units (corps, etablissement, famille, individu). The more restrained boundaries facilitate the physical arrangement of items. As rule number one of the circular stipulates, the contents of a fonds is defined as the sum of all the records originating from a records creator. The grouping of monuments is now based on their belonging to a single source or origin. Only documents sharing a common origin can be part of the same fonds. It also implies that records not sharing that property are not part of that fonds. The fonds provides, thus, a means of physical control. A fonds is a container to be "respected". The properties of "competence" and "origin" that respectively define records creator and records created are linked through the concept of fonds. The . fonds embodies this causal relationship. For something to be, something else must be. The relationship of premises to conclusion is a relationship of material causation. The fonds as records created flows from the fonds as records creator. The relation between records creator and records created is one of cause-effect. The concept of causation is one of the most fundamental of human concepts and one of predilection in historical thinking. It can be inferred 51 that person(s) or corporate bodies create records because they can do so. Or, more precisely, because they have the authority or jurisdiction to create them. The fonds as a container can be deduced from the fonds as contents. The fonds as contents equals the sum of the records created. The fonds represents the physical, evidence of the authority to create these records. The properties that define the fonds as records creator (container) and the fonds as the records created (contents) are different but because they are linked through material causation, they both have the same class boundaries.55 It is this link that provided archivists with a good tool for both the physical and intellectual control of their holdings. It is also this link that allowed the fonds to provide access at the domain class of a classification scheme. Provenance, as an access tool, is based on the belief that queries also follow a causation path from records creator to records created. The fonds is the physical clustering of objects recording the transactions of an agent: the records creator. There is an immediate causal link between agent and object. The two are associated through the name of the records creator. For access purposes, that name functions as the label for a fonds (as domain class). The fonds brings together records with a common provenance and, from the label representing these records, infers 55 Terry Cook, "The Concept of the Archival Fonds: Theory, Description, and Provenance in the Post-Custodial Era," in The Archival Fonds: from Theory to Practice, edited by Terry Eastwood (Ottawa: Bureau of Canadian Archivists, 1992), 40. 52 the general topic of all the records it brought together. Schellenberg outlined this notion when he wrote that: "A certain amount of meaningful information can be provided about public records simply by identifying the government agency that produced them. The more precise the information on their producer, the more precise is the information on their content. If records were produced by a government agency, a mere identification of the agency will reveal, to a degree at least, their content or nature."56 In other words, since the records creator is the fonds and the fonds acts as a domain class for the records created by the records creator, one can be inferred from the other. The fonds is usually labeled with the name of the records creator to facilitate this process. This inferential process, when using the concept of fonds for classfication, works thus: 1- the label (name of the records creator) implies a field of competence; 2-this field of competence implies that records were created to carry it out; and 3- that these records can only be part of this fonds because membership in a fonds is based on origin from the same records creator. The records creator stands for the fonds and its name stands as a class label for the whole of the records it subsumes. The access process consists in moving from the records creator to the records created. It is possible to move from one to the other because intension and extension overlap. These advantages derived from the implementation of provenance-based c l a s s i f i c a t i on and the concept of fonds were 56 Schel lenberg , Management of Archives, 133-4. 53 gradually recognized and adopted by public archives elsewhere. However, respect des fonds was meant to be a solution to the management of an archives' holdings by breaking it down into fonds. It still relied on pertinence for the organization of material within the fonds. As the nineteenth century progressed, this became increasingly unsatisfactory. Respect des fonds became a partial solution to growing theoretical concerns over changes in historiography. Historicism was becoming the prominent view of users of public archives. Classification by pertinence within a fonds did not meet the needs of this new approach. What distinguished the new historical outlook from major Enlightenment patterns of thought was its rejection of a mechanistic world view. From the viewpoint of historicism, the stuff of history could not be mechanically and rationally broken into universal categories to exemplify some static intellectual order. Scientific history became the dominant discourse of historiography. From the point of view of historicism, records creators were living organisms, subject to birth, growth, and death. A fonds would follow the same movement in a symbiotic relationship. In this respect, Posner noted that: "Under the influence of the organic conception of history and the idea of organic growth in general, the registry came to be thought of as an organism paralleling in its development that of the record-54 creating agency and expanding and differentiating in strict concordance with the changes of the latter."57 The notion of fonds as a natural entity, as opposed to an artificial construct, was to be one of the most cogent ideas in archival theory. Under its influence, the fonds of extinct records creators were likened to fossils. Records were the solidified remains of a fleeting and ephemeral corporate body. Fonds were dead organisms to be reconstructed the way a paleontologist would reconstruct a dinosaur or an archaeologist a ruined temple. In the Dutch Manual, this idea is voiced thus: "A paleontologist analyses each bone and fits it, on the basis of his knowledge of its structure and of the functions it served, into its proper place in a skeleton. An archivist proceeds in a somewhat similar manner. He generally deals with skeletal remains of a number of organic bodies. He fits together individual documents and series on the basis of his knowledge of the structure and functioning of the body that produced them. "58 Following this idea, the fonds had an underlying structure that was sui generis. Archivists could not impose a new structure because an organism could have one and only one possible structure. To use the Rankian formula, a fonds was also to be reconstructed wie es eigentlich gewesen ist. It was, therefore, crucial to preserve records in a repository in the same order as they were kept by the records creator. By preserving or in some cases, where possible, reconstituting, in the archival repository, the organic, historical processes that created Ernst Posner, "The Role of Records in German Administration," Archives and the Public Interest, edited by Ken Munden (Washington: Public Affairs Press, 1967), 91. 58 Schel lenberg, Management of Archives, 87. 55 records, the archivist/historians sought to preserve the historical quality of public records.59 This new philosophy also had a new methodology. "Historicism emphasized the organic evolution of the state and the understanding of that development through an intuitive, inside view of human institutions as they actually existed. It follows that the records of the state, to be properly understood, should be kept as an organic unity. The archivist and historian could then correctly interpret the records by grasping the unique internal coherence of the records and the various human [i.e., those imposed by human activity] relationships embodied in them."60 This concern for the preservation of the way in which material had accumulated in archives also reflected changes in the nature of documents and in administrative practices. What constituted a document and how it was accessed had also changed. In current administrative practices, the document as juridical instrument was losing some of its autonomy. "With the diffusion of education, the growing accessibility of writing instruments and materials, the development of communication systems, the increase of business activity, and the rise of complex bureaucracies, two things happened. First, people began to create documents for the purpose of communicating facts, feelings and thoughts, asking for and providing opinions, preserving memories, elaborating data, and so on. Therefore, an ever decreasing proportion of written documents came to originate from juridical acta and Friederick Meinecke, for example, noted that by applying the principle of provenance to the State Archives, "the registry of every single agency...now became a living organism of its own with its peculiar principle of life, and the different persons with their individual traditions and impulses now came to light." Quoted in Posner, "Genesis of the Principle of Provenance," 42-3. 60 Richard Klumpenhouwer, "Concepts of Value in the Archival Appraisal Literature: An Historical and Critical Analysis," (Master's Thesis, University of British Columbia, 1988), 39. 56 present a required wri t ten form...Secondly, j u r i d i c a l ac t s , and spec i f ica l ly those defined as t r ansac t ions , began to r e s u l t from a combination of re la ted ac t s , j u r i d i c a l and non- jur id ica l , each of which produced documents. As a corol lary , many documents came to re fer to the same act."6 1 Consequently, documents re la ted to the same business were grouped together in dossiers.6 2 Records included more than documents which were au tho r i t a t i ve by themselves but a lso a l l documents of a solemn character created by persons authorized to do so. I t a lso comprised documents t ha t were not "authentic" in the diplomatic sense of the word but which could be t rus ted because of t h e i r context of creat ion and t h e i r custody in an archives.6 3 The j u r i d i c a l instrument was loosing i t s autonomy as a s e l f - r e f e r e n t i a l en t i t y to the doss ie r . The author i ty and value of each document in a dossier was co-dependent on a l l the other documents in tha t dossier.64 The notion of context had moved to the dossier l eve l . The ideal was to preserve or reconstruct the o r ig ina l order of the f i l e s and to assign i t s proper place to each document. The conjunction of the changing nature of records with h is tor ic ism would have a considerable impact on archival prac t ices and theory. 6 1 Durant i , "Diplomatics (Par t I I ) , " 8. 6 2 Bau t i e r , "Phase c r u c i a l e de l ' h i s t o i r e des a r c h i v e s , " 146. 6 3Outrey, "Notion d ' a r c h i v e s en France ," 283. 64The s i m i l a r i t i e s between the d o s s i e r and t h e fonds a re a l s o s t r i k i n g . The Lexicon of Archives Terminology def ines d o s s i e r as " the ensemble of documents rece ived or c rea ted by a person(s ) or co rpora te body for the conduct of a s p e c i f i c b u s i n e s s . " Both the fonds and t h e d o s s i e r a re def ined along the same elements of record c r e a t o r and bus iness a c t i v i t y . Laroche noted in "Que s i g n i f i e l e r e spec t des fonds? Esquisse d 'une a r c h i v i s t i q u e s t r u c t u r a l e , " Gazette des Archives 73, supplement (1971), 14, t h a t " le d o s s i e r p o u r r a i t e t r e cons idere comme l e p lus p e t i t fonds, d e f i n i s s a b l e , e t pour a i n s i d i r e I'unite de fonds." 57 As the historian V.H. Galbraith put it: "the archivist [was now doing] his work in a world whose fundamental conceptions of what history is had been revolutionized...."65 This new trend and methodology in historical research influenced to a great extent the formulation of the principle of respect de .1'ordre original as an integral part of classification by provenance. This meant the application of respect for historical growth to the sources of historical research that had come into existence in the course of historical events.66 Now, provenance required not only that records be organized according to their source, but also in the order records creators had kept them. Pertinence was not acceptable anymore for the classification of records below the fonds level. In this context, archivists gradually came to believe that one of their main duties was to present a historically neutral view of the records in their custody. This neutral historical view, that would not obscure any other possible historical view or research, was the history of the records creators themselves. This context was best preserved by communicating an understanding of the administrative history of the records creator. In turn, this administrative history was best preserved by characterizing V.H. Galbraith, An Introduction to the Use of the Public Records (London: Oxford University Press, 1934), 11. 66 Ernst Posner, "Max Lehmann and the Genesis of the Principle of Provenance," Archives and the Public Interest, edited by Ken Munden (Washington: Public Affairs Press, 1967), 40-1. 58 the natural growth and accumulation of documents. This idea is still present in archives today. Carlo Laroche has remarked that: "L'histoire des activites d'un sujet [records creator] est une histoire specialisee parmi d'autres et n'a aucun titre a les englober. Mais parmi les histoires specialisees elle est peut-etre celle qui contient le noyau commun a toutes, c'est-a-dire l'histoire toute nue, sous les vetements qui sont 1'unique objet de nos descriptions historiques."" Reestablishing the original relationship of the records, and thus showing the functioning of discontinued offices, made possible the investigation of their history.68 The origin, or provenance, provided an understanding of the records, and, in turn, the records provided an avenue for understanding historical events. Archivists, thus, strive to recreate or preserve the administrative history of a records creator. In Jenkinson's words, "the only correct basis of Arrangement [is] exposition of the Administrative objects which the Archives originally served."69 This belief favoured classification that used the administrative structure of records creators. This type of classification uses the administrative units of an organization as classes. The administrative hierarchy becomes a hierachical classif iction. The Verwaltungstruktur Prinzip70 became prevalent for the classification of records below the fonds level because Laroche, "Que signifie le respect des fonds?," 49. 68Posner, "Some Aspects of Archival Development," 30-2. 69 Hillary Jenkmson, A Manual of Archives Administration: Including the Problems of War Archives and Archive Making. Oxford: Clarendon Press, 1922, 80 70Johannes Papritz, "Methodes modernes de classement d'archives: documentation ecrite d'avant 1800." In Conseil international des archives. V congres international des archives, Bruxelles, 1-5 septembre 1964, 1-140, Bruxelles: Archives generales du royaume, 1964, 12. 59 the implementation of respect for original order could so easily be grafted onto it. The preference for the administrative history as the neutral history, and provenance and fonds as the means to insure the preservation of this particular history, spread this type of classification in public archives. These methods also answered the requirements of historians. This idea of organic growth supported by the principle of respect de 1'ordre original would gradually gain acceptance in all public archives. Respect for organic growth slowly found a place in archival practices.71 It was formally expressed by Bonaini, in Italy, with the development of the metodo storico, a telling name in itself. In 1867, he stated that documents were to remain in the fonds in which they were created and, within that fonds, in their place of origin. The placement of documents was not the result of artificial creations but the result of natural accumulation. The metodo storico was made State policy and applied to the whole of Italy in 1875.72 In Prussia, the Privy State Archives, under the historian Sybel, adopted a similar regulation in 1881 making registraturprinzip a corollary principle to provenienzprinzip.73 In the late 1890's, the Dutch archivists association would also make recommendations to their government for the standardization of archival practice in public 71 Bartlett, "Respect des fonds," 111. 72Herrera, Archivistica General, 33. 73Britchford, "Provenance of Provenance in Germanic Areas," 63. 60 archives, putting original order as one of its founding principles .74 The principle of provenance and the concept of fonds were a response to changing conditions in the nineteenth century. In a simplified chronology, we have three markers for that evolution in three important texts, where classification by provenance is codified. The circular of 1841 marks the first introduction on a large scale of the provenance model in public archives. In 1898, the Dutch Manual of Muller et al.75 provides rules for the application of provenance to old records. Over time, solutions to problems had been recorded, systematized and fixed into rules. By the end of the nineteenth century, archivists had developed through their practice a coherent body of principles. All possible case scenarios were thought to have been encountered and dealt with satisfactorily. The Manual was the new summa compiling archives administration into 100 rules acquired through trial and error. Its rapid translation in French, German, Italian proved how readily archivists in Europe identified their own practices in what the Manual was proposing. In 1910, provenance was recognized at the Brussels Conference of Librarians and Archivists as the core principle of the archival discipline. The Actes of the Conference show that fonds and provenance were finally adopted unanimously by archivists from all over Europe. 74Marjorie Rabe Barritt, "Coming to America: Dutch Archivistiek and American Archival Practice," Archival Issues 18, no. 1 (1993), 45-6. 75 Muller, S., J.A. Feith and R. Fruin. Manual for the Arrangement and Description of Archives. 2nd ed., translated by arthur H. Leavitt. New York: H.W. Wilson co., 1968. 61 The recognition of this common ground was consecrated with the adoption of the following resolution: "M. le President. -- Voici le voeu libelle par M. Wiersum: Le principe de provenance est le meilleur systeme a adopter pour classer et inventorier un fonds d'archives, non seulement au point de vue du classement logique des pieces mais aussi dans 1'interet bien compris des etudes historiques. (Voeu adoptee a 1' unanimite) . "76 Archivists had produced a body of rules and methods to guide them in their classification work. The general acceptance of provenance and fonds was to provide the archival community with a unique identity. Unfortunately, they also contained all the flaws that would plague their application, especially to open fonds. The main problems with classification by provenance rest with the concept of fonds. In archival classifications, the concept of fonds pulls double duty. There is the fonds as records creator and the fonds as aggregate of records. Unfortunately, these two definitions of fonds are at odd because each relies on a different key property. For physical control purposes, it is the property of "origin" that determines membership in a fonds. On the other hand, "competence" determines membership in a fonds, for intellectual control, that is, for retrieval purposes. As key properties, origin and competence provide different boundaries to the same category: the fonds. Each implies a different domain 76 Congres de Bruxelles, 1910. Actes (Bruxelles: Siege de la Commission permanente des Congres internationaux des archivistes et des bibliothecaires, 1912), 635. 62 class with its respective intension and extension. Of course, in cases where the domain class is not divided into sub-classes, as for the fonds of person(s), or for closed fonds in general, the two can be reconciled. However, for open fonds this can be problematic. The resolution of the contradictions at the heart of the concept of fonds were to form an important chapter in the history of archival theory. 63 CHAPTER FOUR: ADAPTING PROVENANCE-BASED CLASSIFICATION TO MODERN RECORDS The core of provenance is the concept of fonds. For classification and access, the power of the concept of fonds as a domain class resides in its ability to constrain the membership of documentary units. Membership in a fonds ensures stability of reference. It means that a user can expect a domain class named "John Smith Fonds" to contain the records of John Smith and only the records of John Smith. This stability of reference relies on a constant relation between three elements: the fonds label, the fonds intension and the fonds extension. These elements and their relation are captured in the definition of fonds itself where a fonds is: "The total body of records/archives accumulated by a particular individual, institution or organisation in the exercise of its activities and functions"77 "The total body of records/archives" is the extension of the fonds. "Accumulated...in the exercise of its activities and functions" is the intension of the fonds. "By a particular individual, institution or organisation", where "particular" might be "John Smith", is the fonds label. The fonds label "John Smith Fonds" points to the intension of the John Smith fonds, 77 Peter Walne, ed., Dictionary of Archival Terminology / Dictionnaire de terminologie archivistique. English and French with Equivalent in Dutch, German, Italian, Russian and Spanish. 2nd ed., rev. (Munich: K.G. Saur, 1988), 72. 64 that is, the property defining membership in this fonds. That property is determined by the activity conducted by John Smith. Documents are members of this fonds if they possess the property that defines the fonds' intension. If the intension of a fonds is determined, it follows that its label possesses an invariant meaning.78 The stability of classification by provenance is based on that premise. A fonds (domain class) is all of the records (class extension) created by a records creator (class label) out of the conduct of its business (class intension). These elements are inter-dependent. A change in any of these terms would compromise a classification by provenance. When the concept of fonds was first formulated in the nineteenth century, this did not present a problem. The definition of fonds fit what archivists were dealing with. Records creators were usually blessed with a well defined intension, a stable name and a stable administrative structure to support it. They provided archivists with a solid base on which to build reliable classifications. It could be safely assumed by researchers that a records creator had the sole purview of a field of competence, to the exclusion of all other records creators. Under its name (class label), all the records that were created out of the exercise of that field of competence could be expected to be found. For such fonds, the link between a creator and records was constant and could be inferred straightforwardly 78 Jacob, "Communication and Category Structure," 87. 65 because the relation between intension, extension and level remained constant. In modern times, however, this scenario seems to be the exception rather than the rule. Intension, extension and label are constantly subject to change. Modern administrations have often been in mutation and experienced growing pains. The twentieth century has seen an increasing rate of governments' reshuffling, creation and abolition of agencies. Consequently, public archives have been particularly subjected to this problem. Taking Australia as an example, P.J. Scott compiled the rate of change in government agencies in this century (table 1). Table 2 Commonwealth of A u s t r a l i a Government Departments Rate of Change79 P e r i o d Number of Changes A v e r a g e R a t e of A v e r a g e Number ( C r e a t i o n , a b o l i t i o n ) Change ( d e p a r t m e n t s of D e p a r t m e n t s p e r y e a r ) E x i s t i n g _ ___ 0 . 7 9 . 0 1.6 1 2 . 0 2 . 2 1 3 . 9 3 . 1 2 5 . 3 1.3 2 4 . 2 0 . 9 2 5 . 6 1 2 . 3 2 8 . 7 As the t ab le ind ica tes , a c h a r a c t e r i s t i c of modern bureaucracies has been the growing r a t e and frequency a t which they have been subjected to adminis t ra t ive changes. The r e l a t i v e 1900-1911-1921-1931-1941-1951-1961-1971--1910 -1920 -1930 -1940 -1950 -1960 -1970 -1980 7 7 16 22 31 13 9 123 79 T a b l e r e p r o d u c e d from P e t e r J . S c o t t and G. F i n l a y , " A r c h i v e s and A d m i n i s t r a t i v e Change: Some Methods and Approaches ( P a r t V ) , " Archives and Manuscripts 7 (August 1 9 7 8 ) , 1 5 . 66 I. instability of modern government agencies has entailed the instability of provenance-based classification. The more unstable the administrations, the more problematic intellectual control became. For example, let's assume that a Ministry of Communications is responsible for the attribution and regulation of all airwaves. Based on this situation, there is a well-defined domain class label (the Ministry of Communication Fonds) and class intension (the Ministry's field of competence) in a stable relationship. Based on this, researchers can infer which records are contained within this fonds. This inferential process would be broken if the name of the Ministry were changed to "telecommunications" while the Ministry of Culture became the Ministry of Communications. A similar problem would occur if the Ministry was subject to changes in its field of competence and saw the control of maritime airwaves moved to the Ministry of Ocean and Fisheries while keeping the control of commercial airwaves. In such cases, the label of a records creator cannot imply a constant field of competence (intension) and represent all of the records created out of the exercise of that competence (extension). With modern administrations and their open fonds, archivists are more and more often confronted with such cases. In this respect, the classification of modern administration records has been a challenge to archivists. Classification, based on a definition of fonds framed by the realities of nineteenth century 67 archives, has had difficulties coping with the realities of modern administration and modern records. This situation has been further complicated by two factors. The first one is an issue of physical.control. The classification of modern records often means the organization of huge quantities of records. The second factor is the reliance of archivists on administrative structures for classification purposes (Verwaltungstruktur Prinzip). In line with their vitalist view of archives and history, they have usually used the administrative structure of records creators as templates for the sub-classes of their classification schemes. It means that the problem described above in the example of the Ministry of Communications can be true for any administrative unit used as a sub-class within the "Ministry of Communications Fonds". Archivists had first to deal with the physical control of open fonds. The immediate problem posed by modern records was their sheer quantity, not how to provide access to them. The increase in the size of administrations witnessed a proportional increase in the amount of records created. Archivists had first to gain physical control over these records before thinking of intellectual control. To face this more pressing problem, the concept of fonds provided an excellent means of physical control of records in the custody of a repository. Indeed, one of the main practical advantages of the concept of fonds is that it allows archivists to deal with large quantities of records. The 68 extension of a fonds (domain class) can expand without causing any problem with classification. It is not uncommon in repositories to have records sit on shelves for years before the resources are found to process them. During that time, the concept of fonds provides an easy way of physically controlling -those records. Archivists simply need to link them to the records creator that originated them. Unfortunately, the fonds used to control the records physically is not necessarily the same as the fonds used for intellectual control. These two aspects of the fonds are defined by two different understandings of provenance. For physical control, the key property defining the fonds as domain class is "origin of the records". For intellectual control, however, this key property is the authority or competence that called for the creation of records. Each of these properties define a different fonds as domain class. Any change in the records creator's label or intension will break that synchronicity between the two fonds. Records accruals may be added to the fonds defined by "origin" although they do not fully fit into the original fonds defined by "competence". This is one of the problem facing classification by provenance for open fonds. The second impediment to the classification of modern records is the Verwaltungstruktur Prinzip. This method uses the administrative structure as a skeleton to which documentary units are attached. This principle of classification assumes that functions correspond to units in the administrative structure 69 used by a records creator to conduct its business. Classifications patterned on administrative structures use the administrative units as sub-classes. For archivists using administrative structure for classification, the structure is the classification. The various administrative units are not elements in a classification based on subject or function; the administrative unit is the class and the unit name is the class name. Archivists assumed a direct relationship between administrative structures and competence. The rationale is that each administrative unit that makes up the structure implies a portion of the field of competence of a records creator. In turn, this area of competence is used to infer the records created by that administrative unit. Theoretically, at each level of the administration, a user would infer, based on the name of that administrative unit, whether the records it would have created are relevant to his search. The rungs of that retrieval ladder are the names of administrative units. The names of administrative units act as labels to their mandates, competence, functions or administrative activities. To work, this inferential access process depends on a stable relation between the two for a univocal link. It is, therefore, imperative that this ladder leading to the actual records remain stable. A change in the administrative structure would break the chain leading to the information.80 For older organizations, it was a reasonable 80 There is, therefore, a need to put in place a mechanism to insure the stability of such classifications. The second part of the Rules for Archival Description is an example of this approach. Based on AACR2R conventions, inclusive classes and including classes are listed consecutively, starting with the domain class and ending with the targeted class. For example, 70 assumption to make tha t t h e i r adminis t ra t ive s t ruc tu re would remain s t ab le , but i t i s not necessar i ly so for modern ones. Their f luid adminis t ra t ive s t ruc tu re makes i t d i f f i c u l t to adapt t r a d i t i o n a l provenance-based c l a s s i f i c a t i o n . Changes in the s t ruc tu re t r a n s l a t e in to changes in the c l a s s i f i c a t i o n and changes in the c l a s s i f i c a t i o n disrupt the i n f e r en t i a l process on which r e t r i e v a l i s based. This problem i s compounded by the type of codif icat ion a r ch iv i s t s have attached to these schemes. These codif icat ions often t r i e d to embody the subordinate re la t ionsh ips of the adminis t ra t ive un i t s / c l a s se s of a c l a s s i f i c a t i o n . Despite t h i s disadvantage, i t makes sense to use the adminis t ra t ive s t ruc tu re for c l a s s i f i ca to ry purposes. An adminis t ra t ive un i t necessarily c rea tes records r e l a t ed to the functions for which i t was es tab l i shed . From a synchronic point of view, t h i s i s always t r u e . At any point in time, the two wi l l coincide. This i s why t h i s approach worked so well for closed fonds; by se lec t ing the l a s t adminis t ra t ive s t ruc tu re in use a t the time the fonds was closed, a r ch iv i s t s had in hand such a synchronic moment. A records creator may have gone through a number of changes during i t s existence but, once i t s fonds i s closed, i t wi l l not undergo any further changes. The fonds label "Ministry of Communications. Divis ion of Cu l tu re . Cen t ra l Branch. Disbursement Of f i ce . " The formulat ion of name a u t h o r i t i e s mimic t h e h i e r a r c h i c a l s t r u c t u r e of an admin i s t r a t i on in a l i n e a r r e p r e s e n t a t i o n . The boundary of each con ta ine r i s r ep resen ted by the per iod , very much l i k e sen tences in a t e x t . In theory , users a re i n v i t e d t o follow t h i s path down an a d m i n i s t r a t i v e h i e r a r chy . Each u n i t has i t s mandate or funct ion or a c t i v i t y desc r ibed by i t s name. The u s e r ' s re levancy judgments a re based on t h e d e s c r i p t i v e na tu re of the name. 71 will not change. The fonds intension will not change. The relation between the two will not change. Its records can now be organized in a stable classification. The fact that the fonds is closed seals that stability. The problems with open fonds is that there is no such final synchronic moment. With open fonds, a classification will have to take into account a diachronic view of the records creator. Classifications based on the concept of fonds have difficulty accommodating this type of view. For example, for the same class label (name of administrative unit/records creator), there could be more that one function assigned over time. In such cases, all the records it created would have the same "origin" property but different "competence" property. Likewise, two different administrative units (class labels) could have been responsible for the same function at different times. The records they created would have the same "competence" property but different "origin" property. These problems plaguing the classification of records in open fonds did not appear overnight. At first archivists saw no fundamental difference between the records they had dealt with in the past and modern records. At a time when the management of modern records in archival repositories was becoming more problematic, the concepts of fonds and provenance were just finally establishing themselves as the dominant theory in archival science. It is fair to assume that having just established and agreed on a common classificatory model, archivists were probably not ready to overhaul it to satisfy the 72 requirements of modern records. As far as archivists were concerned, the problem was not with provenance but in its application. Since the concept of fonds had been successfully applied to the classification of closed fonds, it was also applied to open fonds. Archivists tried to solve new problems with tried solutions. The first strategy of archivists was, therefore, to try to fit modern records into the accepted definition of fonds. The solution entailed waiting for administrations to become stable or to artificially create such stable conditions. This solution bypassed the classification problem of open fonds by waiting for them to become closed. In this scenario, archivists accept accruals from records creator, but do not process the records while waiting for the originating agency to die out. For example, Duchein explained how the French public archives use this method: "As a general rule, we may admit that a chronological section can receive treatment when all documents which compose it reached the public domain and are no longer likely to receive additions or suffer withdrawals or eliminations. "81 The natural death of an agency, however, can be long in coming, and sudden deaths brought by historic breaks, such as revolutions, are few and far between. When Duchein wrote this, in the 1970's, the historical break used by the French National Duchein, "Theoretical Principles and Practical Problems," 80. 73 Archives was the end of the Third Republic. There was no decided "arrangement" for records created after 10 July 1940. Another variation used in implementing this option is to wait for significant changes to take place in administrative structures in lieu of significant constitutional or historical landmarks. In the absence of History with a capital "H", archivists settle for history with a lower case to frame their classification schemes. This option, used to alleviate the problem of open fonds, consists in a division by time periods set arbitrarily. A fonds would be sliced up in time periods and be classified according to each time frame. For example, Muller et al. recommend classifying records after periods of 25 years.82 With this approach, the fonds becomes a succession of frozen pictures over time. (This approach should not be confused with chronological arrangement of documents.) This type of classification was not a new idea: periodization is a basic tool of historiography. It is not unique to archival practices either. In the library world, as early as 1664, it had been proposed that, along with a subject classification, a chronological classification be used for the ordering of collections. Several similar schemes were proposed during the nineteenth century.83 Archivists were simply following this lead in their practices. Rule 14 in Muller, S., J.A. Feith and R. Fruin. Manual for the Arrangement and Description of Archives, p. 44-7. 83 -* Eric de Grolier, Theorie et pratique des classification documentaires (Paris: Editions documentaires industrielles et techniques, 1956), 63. 74 ^ The underlying principle in this option was to treat open fonds as a string of smaller closed fonds, one for each time period. What these various solutions have in common is the perceived necessity to have or to create artificially the condition of a closed fonds to achieve a stable classification by provenance. These strategies were in line with the metaphor of records creators as living organisms and archives as fossilized remains of administrative activity. The analogy with fossils worked well for closed fonds. A fossil has one structure, immutably imbedded in stone. The administrative structure is akin to the skeletal structure of a fossil. This analogy, however, did not work so well for open fonds. Unfortunately for archivists, open fonds are works in progress, so to speak, and their final, fossilized form cannot be known until they are closed. Not only can open fonds take a long time to die, they are also very likely to mutate before they do die. Like the field of biology from which it borrowed the idea, archival science has had to deal with evolutionary organisms. In the field of biology, Darwin discovered that organisms evolved into their present forms. They did not have the same form from the day of Creation. When this idea found credence, biology made the transition from creationism to Darwinism. The taxonomies of organisms moved from static classifications to evolutionary taxonomies that provided room for such changes over time. The solutions described above show a view of the world where the only possible fonds is a static closed fonds. Archival science has had difficulty making a smooth 75 transition from closed fonds to fluid open fonds. Archival science needed a solution similar to Darwinism where the classification of open fonds could accommodate such changes. Archivists devised other classification solutions, based on provenance and the concept of fonds, that would take into account the fact that open fonds are subject to changes. The solution would have to maintain the advantages of the concept of fonds for physical control while providing users with a stable classification despite the changes in label and intension to which open fonds are subject. Provenance and the fonds concept in public repositories have had to serve the needs of two masters: users who count on this system for access, and archivists who rely on it for the physical control of their holdings. The objective was to come up with a fonds where label and intension are in a stable relation while still physically managing the amount of records flooding into archival repositories. The solution was to revise the definition of the concept of fonds itself and the notion of records creator on which it is centred. The proposed solution was to re-define the domain class as a records aggregate of a scale that would answer both needs optimally. What archivists needed was to delineate a chunk of archival material of a scale that was meaningful for access and manageable for control. Archivists have tried to define a domain class of a scale that was "just right" for this double duty. For some, this meant re-defining the fonds as domain class, for others, a different entity all together was chosen. 76 Despite what the definition of fonds implies, what a fonds is, or any other documentary unit, is always the result of a conscious decision. A fonds is not a "natural aggregate". The definition of fonds is open to interpretation. As Senecal has rightly observed: "[L]e fonds est un decoupage intellectual d'une realite documentaire cree pour les besoins precis de la description en archivistique, plutot qu'etre une entite physique particuliere. Ni le choix du fonds, ni sa nature, ni meme le choix de 1' entite a. deer ire (serie, sous-serie, dossier, piece) ne sont des elements "objectifs", donnes intrinsequement. "84 In other words, the material world has fixed scalar properties but the scale of observation comes from the observer's decisions.85 For example, what has constituted archival material over time can illustrate this point. Records are time and space dependent. However, the scale of observation defining what is a records comes from the observer's decisions and can be based on medium, provenance, contents, and so on. Archivists have established a set of criteria to distinguish archival material from other types of material, in other words, what is worthy of their attention. For example, paper-based documents were for a time refused admittance in archival repositories. Archival documents were so defined as to admit only parchment-based documents. Only when the definition of archival document was modified were paper-based documents admitted. Similarly, in the 84 Senecal, "Reflexion sur le concept de fonds d'archives," 45. 85 Valerie Ahl and T.F.H. Allen, Hierarchy Theory: A Vision, Vocabulary, and Epistemology (New York: Columbia University Press, 1996), 50. 77 twentieth century, it took a further modification of the definition of archival documents for films and magnetically-stored documents to be admitted in some archival repositories.86 Archivists also distinguished between public and private records. For example, the French National Archives, considered for a time the records it acquired from defunct institutions such as hospitals or religious congregations as public records. They later modified their definition of public records to exclude these types of records.87 The changing definition of documents and monuments in the eyes of archivists is another example of how the definition of what constitutes archival material is time and space bound. At one time a document was considered to be self-referential; at another, this quality was attributed only to the interrelated documents of a file. Now, with electronic records, documentation of the recordkeeping system itself is necessary to achieve what an instrument used to be able to do on its own. Similarly, what constitutes and defines a fonds is not a given but always a conscious decision. Finding the right scale for the class label and intention was a matter of finding the right definition of fonds that would meet both objectives of intellectual and physical control of open fonds. The pivotal element in this process has been a re-definition of "records creator" or class label. The changing nature of records and government has forced archives to adopt a different definition of what may constitute a records creator. In this area, three 86 Robert-Henri Bautier, "Phase cruciale de l'histoire des archives," 150. 87 • Jean Favier, ed. , La pratique archivistique frangaise (Paris: Archives Nationales, 1993), 70. 78 approaches have met with some success amongst archivists: functional analysis, the records group concept, and the records series system. The first approach is based on the premise that an administrative structure is a representation of a records creator's functions. In itself, this was not a new idea. This is basically what the Verwaltungstruktur Prinzip is about. Muller et al. had already expressed it formally in their Manual.8" What was new was a re-definition of the domain class intension that would fit administrative units as representatives of a single function as opposed to all the functions of a records creator. With this type of approach, there is a shift in the intension of the domain class. Instead of "competence" or "authority" as its defining notion, this approach relies on "function". It considers administrative units that embody a "function" to be sufficiently autonomous to be the domain class of a hierarchical classification. This approach pushed the idea of using administrative structures as classificatory frameworks to its logical consequences. The schemes proposed by Margaret Norton Cross and Michel Duchein are examples of this approach. Margaret Cross Norton, state archivist for Illinois, proposed her scheme in the early 1940s. It was based on the American public administrative structure and followed a hierarchy Muller et al., Manual for the Arrangement and Description of Archives, 52, ssq 79 of Department - Division - Branch - Office. She remarked that Departments, the highest hierarchical level, were too often subject to re-organization and could not therefore provide a stable base for classification. She argued that Divisions were more stable and embodied actual functions while Departments were artificial aggregations of such functions. On the other hand, she felt that branch and office were too small and not sufficiently significant. Her study of the nature of departmental reorganizations revealed that the basis for changes was always a shifting of functions and that, for the most part, those functions were fairly well expressed by division or sub-department headings. Therefore, a classification which would keep as a domain class the records of the division would permit shifts between major departments without ruining the classification system.89 Her solution was to define competence in terms of autonomous functions and then to scale down the domain class to the levels of those units embodying those functions. Norton's objective was to scale the domain class to the level of the division as a means to circumvent changes in administrative structure and provide classification based on administrative hierarchy with some stability. The problem with this idea was that it assumed that the administrative structure below the department level would remain stable and thus preserve 89 Margaret Cross Norton, Norton on Archives: The Writings of Margaret Cross Norton on Archival and Records Management, (ed. by Thornton W. Mitchell, Carbondale and Edwardsville (Illinois): Southern Illinois University Press, 1975), p. 112-113. 80 the classification's stability. It also assumed that an administrative unit can be equated with a specific and unambiguous function to provide intellectual control. These were assumptions that experience often does not bear out. Michel Duchein of the French Archives Nationales came up with a different approach to functional analysis. He realized that, as it stood, the criteria defining fonds d'archives could apply as well to a ministry as to an office of that ministry. While the definition of fonds clearly links records creator and competence, it does not define these terms. More precisely, it does not distinguish between authority (competence) and delegated authority. Looking at governments, one approach is to consider that all authority is necessarily delegated from the head of State or from the People. In this category, we have the case of the former USSR where the whole of the state apparatus is considered to generate one single fonds. Starting from the same premise that all authority is delegated, the other approach considers where that delegated authority is sufficiently autonomous to stand on its own. This "where" is open to interpretation. Duchein drew attention to this fact when he wrote that: "...we can adopt two different intellectual positions, which we shall qualify respectively as "maximalist" and as "minimalist". The former position consists in defining fonds at the highest level, by considering that the true unity of function... was situated at the top...By contrast, the minimalist position consists in reducing the fonds to the level of the smallest possible functional cell, by considering that the true 81 organic "whole" of archives results from the work of this small cell."90 Any administrative unit could fit the criteria of "functional cell" that defined a fonds. Under the accepted definition of fonds as domain class, there was no constraint of linking a fonds to the highest hierarchical level. Archivists had an enormous leeway as to how they could apply the name of fonds to any aggregate of records. There lay the danger of an atomization of documentary units where the distinction between competence (function) and delegated authority (functional unit) is not clearly made. Archivists have used this "loophole" of the fonds concept to breathe some stability into classification that relies on administrative structure. By endowing the smaller, more stable, administration units with autonomous competence, they could still rely on administrative structure as classification structure. Duchein recognized the advantages of this practice but also its weaknesses. Duchein wanted to be able to scale down the fonds while at the same time maintaining a contextual link between this manageable fonds and the larger administrative structure it came from. While retaining the advantages of scalability, he still wanted it to display the hierarchical structure of an administration. To Duchein, the danger of calling smaller administrative units fonds was to lose sight of the "big picture". The "big picture" was the fonds defined in terms of the 90 Duchein, "Theoretical Principles and Practical Problems," 69. 82 larger administrations. Associating the fonds with the large hierarchical structure of an administration allowed archivists to use this structure for a larger, more inclusive classification. As explained above, in such schemes administrative units are treated as classes. The unit at the top of the hierarchy represents the broader domain class while subordinate units represent sub-classes. A user can use the administrative hierarchy as a hierarchy to navigate from the general to the specific realm of activity he is looking for. Scaling down the domain class below the top of the hierarchy threatens the rigorous use of the administrative structure as hierarchy of activity from which to draw inferences about records content. Duchein's solution to this problem was to make the scaled down fonds interdependent. He makes this point when writing that: "... it appears inevitable to introduce into archival science a new notion: that of a hierarchy of fonds corresponding to the hierarchy of creating agencies, involving the subordination of certain fonds in relation to others."91 With this "new notion" classifications based on functional analysis can be given free reign. It allows the concept of fonds to be applied to any level of an administrative hierarchy to provide units of a manageable scale, as long as the administrative context of each unit is recorded, that is, the various structural and functional relationship among units is made clear to the researcher. 91 Duchein, "Theoretical Principles and Practical Problems," 71. 83 The second type of approach is the record group concept. The American National Archives is credited with its introduction. Created in 1934, the Archives immediately faced the problem of processing huge quantities of dormant records. The record group concept was the response the American National Archives devised, in 1941, to deal with the problem of modern administration and modern records. The record group was defined as meaning: "a major archival unit established somewhat arbitrarily with due regard to the principle of provenance and to the desirability of making the unit of convenient size and character for work of arrangement and description and for the publication of inventories."92 The record group was created along the same lines as the fonds: it correlated intension, label, and extension for the definition of records group as domain class. The difference was, as Fenyo put it, that the record group concept "was conceived by a sincere archivist"93 meaning that it clearly embodied the objective of a unit of manageable scale. Grover, a senior archivist with the American National Archives, made this even clearer when he wrote in the staff circular on The Control of Records at the Records Group Level that "no statement [criteria] can be made as to precisely what should be the size of a record group."94 The record group was to be something that should be flexible. The preeminent concern in defining a record group was 92 Wayne, C. Grover, The Control of Records at the Records group Level, in Staff Information Circular no 15 (Washington: National Archives, 1950), 2. 93 Mario D. Fenyo, "The Records group Concept: A Critique," American Archivist 29, no. 2 (1966), 232. 94 Grover, Control of Records, 3. 84 the manageability of its scale to accommodate the massive amount of records that came under the control of the newly created National Archives. American archivists were determined not to let theoretical consideration get in the way of pragmatism. The record group put a heavy emphasis on physical control. With the record group it is almost the domain class extension that dictated the definition of the class intension. The functional analysis model of the fonds concept and the records group concept both deal with the establishment of classifications at the domain class level. However, there is still the matter of establishing and nesting sub-classes in these domain classes. Duchein dealt partly with this problem with his notion of a "hierarchy of fonds". This solution, however, is mainly realized in what he called the "minimalist" approach. For those favouring the "maximalist" approach, this problem remained to be addressed. Terry Cook explained it thus: "The maximalist position of identifying the fonds with the department also means that there are many levels unaccounted for between it and the series of files created in one of its smaller internal divisions."95 For the ordering of material in sub-classes, North American and European archivists have developed the similar concept of sub-group or sous-fonds. Instead of spinning off smaller units as autonomous fonds, the sub-fonds concept allows the embedding of those smaller units within the larger fonds as Terry Cook, "Concept of the Archival Fonds," 56. 85 semi-autonomous aggregates. The concept of sub-grouping is linked to the administrative structure of an organization for intellectual control purposes, but it also allows archivists to break down fonds or record groups down into more convenient size for physical control. The last approach to the problem of archival classification is the Australian records series system. While Europeans and North Americans were tinkering with the concept of fonds, the Australians were moving in a unique direction to solve the problems of classifying modern records. They developed, in the early 1960s, a system that shifted the domain class from the fonds or records group to the records series. It should be stressed that, in this new approach, provenance was still the criteria governing classification. What the Australians proposed was a fundamental change in the classification of archival material. Two main problems had prompted earlier revisions of the concept of fonds for classification: the quantity of modern records and the instability of modern administration. The Australian scheme side-stepped the problem of records quantity. In theory,96 the size of a records series is not a concern for its classification. As for the use of administrative structure for classification, the Australian scheme moved away from this premise as well. The 96 In theory only, because the definition of what constitutes a records series is not very precise and often left to interpretation. Quantity has been known to be a determining factor in determining what constitutes a records series. 86 records series system does not use administrative structures as a basis for classification. It separates the description of the records created from the description of the records creator. It posits that preserving context as information pertinent to understand documents and as a basis for classification are two very different things. They resolutely moved away from the causal link between function and administrative units that are at the core of classification using concepts of fonds or record group. It could be said that, in an effort to resolve the old problem of stability of reference, the Australian system simply moved down one notch from the lowest sub-class to the level of the records series. This system, however, has a fundamentally different basis. In the records series system, "records creators" are moved out of the equation defining the domain class. They simply become the agents creating records. Each series becomes a domain class. The class label of this new domain class can now be simply a description of the series. The main advantage of the series system is its potential to reconcile physical and intellectual control by keeping them separate. Peter Scott made clear that the decision to move the scale of the manageable level of aggregate down to the series was motivated primarily by a need for improved intellectual control.97 This scheme does not try to reconcile the class domains for the 97 Peter J. Scott, "The Records group Concept: A Case for Abandonment." American Archivist 29, no. 4 (1966): 4 93. 87 key properties of "origin" and "competence". It eliminates the domain class based on "origin" all together. This does not mean the elimination of information about "origin" from the descriptive records, simply its elimination as a key property for classification. Descriptions of records and descriptions of agencies are maintained separately. This liberated the series system from the constraint of competing domain classes: one for physical control, one for intellectual control. The records series system dealt with problems of open fonds by eliminating the concept of fonds. "The focal point of the Australian system is the identification of series that are then registered and subsequently controlled as distinct entities."98 With this approach, provenance may not be the same thing as respect des fonds. The first attempts at solving the problem of open fonds had tried to force open fonds to behave like closed fonds. The second attempts had tried to re-define the concept of fonds, particularly its intension, to reconcile physical and intellectual control. The last solution did away with the concept of fonds altogether, and with it transformed the problem of classification into a problem of information retrieval, and so re-thought the application of the principle of provenance itself. If classifications based on provenance could not adapt to open fonds, the Australian system showed that maybe it did not have to 98 Mark Wagland and Russel Kelly, "The Series System - A Revolution in archival Control," In The Records Continuum: Ian Maclean and Australian Archives First Fifty Years, edited by Sue McKemminsh and Michael Piggott, 131, Clayton (Australia): Ancora Press, 1994. 88 adapt after all, that maybe another way altogether could be found to resolve perennial difficulties of controlling archives both physically and intellectually. Although the jury is still out on the success of this new approach, it has definitely stirred archivists to think in new terms that are congenial to such problems in the information age of computerized control of documentation. 89 CHAPTER FIVE: CONCLUSION Access is all. This is, or should be, the a priori basis of all archival work. It is pointless to maintain information if it is not accessible to its potential users. For the large information holdings found in archives, classification is the most efficient means of providing access. In the documentary world, the object of classification schemes is to mediate users' approach to a given body of documentation. They aim at helping users identify documents whose contents match a defined query. Documentary units are grouped together on the basis of a commonality of properties. The choice of these properties is based on what is believed will present users with an accurate picture of the documentary units for relevancy judgments. Relevancy judgments are based on the conditional probability that a document is in a particular class given its possession of some properties or features. For documentary units to be nested in a hierarchical classification, they must share a key property. Archival classification has used two key properties to organize documentary units: pertinence and provenance. Until the mid-nineteenth century, archivists favoured the development of pertinence-based classification in public archives. Chief amongst the reasons for this was the taxonomic 90 mind-set inherited from the Enlightenment. At the end of the eighteenth century, the preoccupation with mapping a canvas of all human knowledge led archivists to favour a classification system whose domain class would be human knowledge itself. The sub-classes were subjects where particular documents could be fitted. Other factors led archivists to develop pertinence classifications, such as the centralization of records in public repositories, after the French Revolution; the juridical nature of documents/monuments; and the belief that pertinence was the most efficient means of answering users' queries. This approach is now largely discredited. Pertinence-based classifications usually replaced the order established by a records creator. In theory, or course, pertinence doesn't have to imply re-arrangement. Pertinence could be applied while respecting original order. Unfortunately, it has been inextricably linked with physical arrangement. Re-classification by pertinence usually meant re-arrangement. Several factors influenced the change from pertinence-based to provenance-based classification. Historical science and historians had a significant impact on the emergence of provenance and respect for original order as the cardinal principles of archival classification. The belief, shared by historians and archivists, in the organic nature of institutions and the organic accumulation of records was the fundamental factor in the development of the concept of provenance and the concept of fonds. This belief in the organic nature of archives 91 favoured the use of administrative structures as the framework for the classification of records because it was believed to provide a "neutral" or naturalistic means of establishing classes. Other factors were also instrumental in making the implementation of provenance and fonds possible. They include the shifting role of archives from administrative to cultural institutions and the change in the juridical nature of records from self-referential to inter-dependent, from instruments to files. Provenance and fonds also proved to be excellent tools for the physical control of records. However, there were perennial difficulties in applying the concepts of provenance and fonds in changing administrative structures, difficulties which grew more pronounced in fluid modern administrations. This instability made it particularly difficult to reconcile the fonds defined by the property of "origin" for physical control with the fonds defined by "competence" for intellectual control. The dual definition of fonds d'archives as records creator and as an aggregate of records stems from this double duty. The solutions proposed by archivists to resolve this problem fall into three groups: the first proposed to treat open fonds as closed fonds; the second proposed to revise the definition of records creators and fonds and to scale them to the needs to archivists; the third, the Australian Records Series system, eliminated the concept of fonds altogether. 92 Since the end of the eighteenth century, changing circumstances have provided either the impulse or motivation for changing ideas about classification. In answer to changing conditions, archivists are constantly re-examining their classification schemes. Some, like pertinence, are now disavowed in archival science, although it is still being used in repositories to supplement classifications by provenance. Classification by pertinence is used each time subject headings are used to improve access to records arranged by provenance. Provenance-based classification is now the preferred method to organize archival material. Unfortunately, the concern for maintaining physical control over archival holdings has too often hobbled this type of classification. In this regard, the greatest achievement of the Australian system has been to show that archival classification doesn't have, and shouldn't have, to serve both purposes. It recognizes that "arrangement" is not synonymous with "classification". 93 BIBLIOGRAPHY Congres de Bruxelles 1910, Actes. Bruxelles; Siege de la Commission permanente des Congres internationaux des archivistes et des bibliothecaires, 1910. Elsevier's Lexicon of Archive Terminology: French, English, German, Spanish, Italian, Dutch. Amsterdam: Elsevier Publilshing Company, 1964. Normes et procedures archivistiques des Archives nationales du Quebec. Quebec: Archives nationales du Quebec, 1988. Subject Indexing for Archives: The Report of the Subject Indexing Working Group. Ottawa: Bureau of Canadian Archivists, Planning Committee on Descriptive Standard, Publication no. 4, 1992. Toward Descriptive Standards: Report and Recommendations of the Canadian Working Group on Archival Descriptive Standards. Ottawa: Bureau of Canadian Archivists, Working Group on Archival Descriptive Standards, 1986. Abraham, Terry. "Oliver W. Holmes Revisited: Five Levels of Arrangement and Description Practice." American Archivist 54, no. 3 (Summer 1991): 370-7. Ahl, Valerie and T.F.H. Allen. Hierarchy Theory: A Vision, Vocabulary, and Epistemology. New York: Columbia University Press, 1996. Barnes, Jonathan, ed. The Cambridge Companion to Aristotle. Cambridge, U.K.: Cambridge University Press, 1995. Barr, Debra. "The Fonds Concept in the Working Group on Archival Description Standards Report." Archivaria 25 (Winter 1987-88): 163-70. Barr, Debra. "Protecting Provenance: Response to the Report of the Working Group on Description at the Fonds Level." Archivaria 28 (Summer 1990): 141-5. Bartlett, Nancy. "Respect des Fonds: the Origin of the Modern Archival Principle of Provenance." In Bibliographical Foundations of French Historical Studies, edited by Lawrence J. McCrank, 107-15. New York: Haworth Press, 1992. 94 Bautier, Robert-Henri. "Les archives." In L'histoire et ses methodes, edited by Charles Samaran, 1120-61. Bibliotheque de la Pleiade. Paris: NRF, 1961. Bautier, Robert-Henri. "La phase cruciale de l'histoire des archives: la constitution des depots d'archives et la naissance de 1'archivistique." Archivum 18 (1968): 139-56. Bearman, David A. and Richard H. Lytle. "The Power of the Principle of Provenance." Archivaria 21 (Winter 1985-1986): 14-27. Bearman, David. "Documenting Documentation." Archivaria 34 (Summer 1992): 33-49. Bearman, David. "Diplomatics, Weberian Bureaucracy, and the Management of Electronic Records in Europe and America." American Archivist 55, no. 3 (1992): 168-80. Bearman, David. "Archival Data Management to Achieve Organizational Accountability for Electronic Records." In Archival Documents: Providing Accountability Through Recordkeeping, edited by Sue McKemmish and Frank Upward, 215-28. Melbourne: Ancora Press, 1993. Berner, Richard C. "Perspective on the Records group Concept." Georgia Archives 4 (1976): 48-55. Berner, Richard C. Archival Theory and Practice in the United States: A Historical Analysis. Seattle: University of Washington Press, 1983. Boles, Frank. "Disrespecting Original Order." American Archivist 45 (Winter 1982): 46-52 Brenneke, Adolf. Ein Beitrag zur Theorie und Geschichte des Europaischen Archiwesens. Leipzig: Koehler & Amelang, 1953. Brichford, Maynard. "The Provenance of Provenance in Germanic Areas." Provenance 7, no. 2 (1989): 54-70. Campbell, E.G. "Fonctional Classification of Archival Material." American Archivist 11, no. 4 (1941): 431-41. Cappon, L.J. "Historical Manuscripts as Archives: Some Definitions and their Applications." American Archivist 19, no. 2 (1956): 101-10. Celentano, Augusto, Maria Grazia Fugini and Silvano Pozzi. "Knowledge Based Retrieval of Office Documents." In 13th International Conference on Research and 95 Development in Information Retrieval, Universite Libre de Bruxelles, 5-7 September 1990, 241-63. Bruxelles: Presses universitaires de Bruxelles, 1990. Chouinard, Denys. "Respect des fonds et la vie des documents: principe pertinent et actuel?" In Actes: XIXs Congres de 1'Association des archivistes du Quebec, 5-8. Montreal: Association des Archivistes du Quebec, 1990. Colin, Hubert. "La mise en oeuvre du principe de respect des fonds: reflexions sur 1'etat actuel." Archives: Le bulletin des archivistes du Quebec 22, no 2 (1987): 3-8. Cook, Michael. Archives Administration: A Manual for Intermediate and Smaller Organizations and for Local Government. Folkstone (UK): Wm Bowser and Sons Ltd, 1977. Cook, Terry. "The Concept of the Archival Fonds: Theory, Description, and Provenance in the Post-Custodial Era." In The Archival Fonds: from Theory to Practice, edited by Terry Eastwood, 31-86. Ottawa: Bureau of Canadian Archivists, 1992. Couture, Carol and Jean-Yves Rousseau. Les archives au XXs siecle: une reponse aux besoins de 1'administration et de la recherche. Montreal: Presses de 1'Universite de Montreal, 1982. Delmas, Bruno. "Bilan et perspective de 1'archivistique frangaise au seuil du troisieme millenaire." In Archival Science on the Threshold of the Year 2000: Proceedings of the International Conference, Macerata, 3-8 September 1990, edited by Oddo Bucci, 81-110. Macerata: Universita degli Studi di Macerata, 1992. Deker, Cornelius. "La bible archivistique neerlandaise et ce qu'il en est advenu." In Archival Science on the Threshold of the Year 2000: Proceedings of the International Conference, Macerata, 3-8 September 1990, edited by Oddo Bucci, 69-80. Macerata: Universita degli Studi di Macerata, 1992. DeRose, Steven J. "Navigation, Access, and Control" Proceedings of the Berkeley Finding Aid Conference, April 4-6, 1995, URL file. Desjardins, Gustave. Le service des archives departementales: conferences faites aux eleves de 1'Ecole des chartes, les 10, 18, 25 et 30 juin 1890. Paris: E. Bourloton, 1890: 25-39. Dollar, Charles M. "Archival Theory and Practices and Informatics: Some Considerations." In Archival 96 Science on the Threshold of the Year 2000: Proceedings of the International Conference, Macerata, 3-8 September 1990, edited by Oddo Bucci, 311-28. Macerata: Universita degli Studi di Macerata, 1992. Doyle, Murielle. "Le respect des fonds et la vie des documents." In Actes: XIXs Congres de 1 'Association des archivistes du Quebec, 9-11. Montreal: Association des Archivistes du Quebec, 1990. Duchein, Michel. "Theoretical Principles and Practical Problems of Respect des Fonds in Archival Science." Archivaria 16 (Summer 1983): 64-82. Duchein, Michel. "La revolution frangaise et les archives: La memoire et l'oubli dans 1'imaginaire republicain." In Etudes d'archivistique, 1957-1992, 59-66. Paris: Association des archivistes francais, 1992. Duchein, Michel. "L'histoire des archives europeennes et 1'evolution du metier d'archiviste en Europe." In Etudes d'archivistique, 1957-1992. 67-80. Paris: Association des archivistes francais, 1992. Durance, Cybthia J., comp. Interpretation of the Concepts of Fonds, Collection, and Item Description of Archival Holdings: A Position Paper. National Archives of Canda, Archives and Government Records Branch, November 1993. Photocopy. Duranti, Luciana. "Diplomatics: New Uses for an Old Science (Part II)." Archivaria 29 (Winter 1989-1990): 4-17 Duranti, Luciana. "Diplomatics: New Uses for an Old Science (Part IV)." Archivaria 31 (Winter 1990-1991): 10-25 Duranti, Luciana. "The Odyssey of Records Managers." In Canadian Archival Studies and the Rediscovery of Provenance, edited by Tom Nesmith, 29-60. Metuchen, N.J. and London: Society of American Archivists and Association of Canadian Archivists in association with the Scarecrow Press, 1993. Duranti, Luciana. "Origin and Development of the Concept of Archival Description." Archivaria 35 (Spring 1993): 47-54. Evans, Max J. "Authority Control: An Alternative to the Records group Concept." American Archivist 49, no.2 (1986): 249-61. Favier, Jean, ed. La pratique archivistique frangaise. Paris: Archives nationales, 1993. 97 Fenyo, Mario D. "The Records Group Concept: a Critique." American Archivist 29, no. 2 (1966): 229-39. Fischer, Gerald L. xvLetting the Archival Dust Settle: Some Remarks on the Records group Concept." Journal of the Society of Archivists 4, no. 8 (1973): 640-5. Galbraith, V.H. An Introduction to the Use of the Public Records. London: Oxford University Press, 1934. Garon, Robert. "Le principe de provenance." Archives: Le bulletin des archivistes du Quebec 1, no 2 (1969): 12-9. Gavrel, Katharine. Conceptual Problems Posed by Electronic Records" A Ramp Study. Paris: General Information Programme and UNISYST, 1990. Gracy, David B. Archives and Manuscripts: Arrangement and Description. Basic Manual Series, series I, vol. 2, Chicago: Society of American Archivists, 1977. Grolier, Eric de. Theorie et pratique des classification documentaires. Paris: Editions documentaires industrielles et techniques, 1956. Grover, Wayne, C. The Control of Records at the Records group Level. Staff Information Circular No 15, Washington: National Archives, July 1950. Hardenberg, H. "Some Reflection on the Principle for the Arrangement of Archives." Der Archivar 16 (1963): 113-8, reproduced in Modern Archives Administration and Records Management: A RAMP Reader, compiled by Peter Walne with the assistance of a working group of the International Council on Archives, translated by F.J.C. Kavelaar, 111-4. Paris: General Information Programme and UNISIST, 1985. Herrera, Antonia Heredia. Archivistica general: Teoria y practica. 5th ed., rev. and enl. Sevilla: Publicaciones de la Excma. Diputacion Provincial de Sevilla, 1991. Holmes, Oliver W. "Archival Arrangement: Five Different Operations at Five Different Levels." American Archivist 27 (January 1964): 21-42. Holmes, Oliver W. "History and Theory of Archival Pratice." In University Archives: Papers Presented at an Institute Conducted by the University of Illinois Graduate School of Library Science, November 1-4, 1964, edited by Rolland 98 E. Stevens, 1-20. Champaign (111.): Distributed by the Illinois Bookstore, 1965. Hurley, Chris. "The Australian ("Series") System: An Exposition." In The Records Continuum: Ian Maclean and Australian Archives First Fifty Years, edited by Sue McKemminsh and Michael Piggott, 150-72. Clayton (Australia): Ancora Press, 1994. Hurley, Chris. "Problems with Provenance." Archives and Manuscripts 23, no. 2 (1995): 234-59. Irvine, Dallas. "Some Kinds of Classification." American Archivist 31 (January 1968): 13-22. Jacob, Elin K. "Communication and Category Structure: The Communicative Process as a Constraint on the Semantic Representation of Information." In Advances in Classification Research: Proceedings of the 4th ASIS SIG/CR Classification Research Workshop, vol. 4, edited by Philip J. Smith et al., 81-99. Columbus: American Society for Information Science, 1993. Jenkinson, Hillary. A Manual of Archives Administration: Including the Problems of War Archives and Archive Making. London: Clarendon Press, 1922. Kimball, Gregg D. "The Burke-Cappon Debate: Some Further Criticisms and Considerations for Archival Theory." American Archivist 48 (Fall 1985): 369-76. Klumpenhouwer, Richard. "Concepts of Value in the Archival Appraisal Literature: An Historical and Critical Analysis." Master's Thesis, University of British Columbia, 1988. Lakoff, George. Women, Fire, and Dangerous Things: What Categories Reveal about the Mind. Chicago: University of Chicago Press, 1987. Laroche, Carlo. "Que signifie le respect des fonds? Esquisse d'une archivistique structurale." Gazette des archives 73, supplement (1971). Lewinson, Paul "Problems of Archives Classification." American Archivist 2, no. 3 (1939): 179-90. Lytle, Richard H. "Subject Retrieval in Archives: A Comparison of the Provenance and Content Indexing Methods." Ph.D. diss., University of Maryland, 1979. 99 Lytle, Richard H. "Intellectual Access to Archives, I: Provenance and Content Indexing Methods of Subject Retrieval." American Archivist 43 (Winter 1980): 64-75. Lytle, Richard H. "Intellectual Access to Archives, II: Report of an Experiment Comparing Provenance and Content Indexing Methods of Subject Retrieval." American Archivist 43 (Spring 1980): 191-207. MacLean, Ian. "Australian Experience in Record and Archives Management." American Archivist 22, no. 4 (1959): 387-418. McKemmish, Sue. "Are Records Ever Actual?" In The Records Continuum: Ian Maclean and Australian Archives First Fifty Years, edited by Sue McKemminsh and Michael Piggott, 187-203. Clayton, (Australia): Ancora Press, 1994. Miller, Frederic M. Arranging and Describing Archives and Manuscripts. Archival Fundamental Series. Chicago: Society of American Archivists, 1990. Muller, S., J.A. Feith and R. Fruin. Manual for the Arrangement and Description of Archives. 2nd ed., translated by arthur H. Leavitt. New York: H.W. Wilson Co., 1968. Nesmith, Tom. "The Archival Studies in English-Speaking Canada and the North American Rediscovery of Provenance." In Canadian Archival Studies and the Rediscovery of Provenance, edited by Tom Nesmith, 1-28. Metuchen, N.J. and London: Society of American Archivists and Association of Canadian Archivists in association with The Scarecrow Press, 1993. Norton, Margaret Cross. Norton on Archives: The Writings of Margaret Cross Norton on Archival and Records Management. Edited by Thornton W. Mitchell. Carbondale and Edwardsville (Illinois): Southern Illinois University Press, 1975. Outrey, Amedee. "Sur la notion d'archives en France a la fin du XVIIIeme siecle." Revue historique de droit frangais et Stranger 30, ser. 4 (1953): 277-86. Outrey, Amedee. "La notion traditionnelle de titres et les origines de la legislation revolutionnaire sur les archives." Revue historique de droit frangais et Stranger 32, ser. 4 (1955) : 438-63. Papritz, Johannes. "Methodes modernes de classement d1archives: documentation ecrite d'avant 1800." In Conseil international des archives. V congres international des 100 archives, Bruxelles, 1-5 septembre 1964, 1-140. Bruxelles: Archives generales du royaume, 1964. Perotin, Yves. "Le concept d'archives et les frontieres de 1'archivistique." In Actes de la septieme conference de la Table ronde internationale des archives, 9-13. Paris: Imprimerie nationale, 1963. Pistolese, Serafino. "Les archives europeennes du onzieme siecle a nos jours: essai historique et juridique." In Annales Institutorum, 3-50. Rome: Biblioteca D'Arte Editrice, 1934. Posner, Ernst. "Max Lehmann and the Genesis of the Principle of Provenance." In Archives and the Public Interest, edited by Ken Munden, 36-44. Washington: Public Affairs Press, 1967. Posner, Ernst. "The Role of Records in German Administration." In Archives and the Public Interest, edited by Ken Munden, 87-97. Washington: Public Affairs Press, 1967. Posner, Ernst. "Some Aspects of Archival Development Since the French Revolution." In Archives and the Public Interest, edited by Ken Munden, 23-35. Washington: Public Affairs Press, 1967. Rabe Barritt, Marjorie. "Coming to America: Dutch Archivistiek and American Archival Practice." Archival Issues 18, no. 1, (1993): 42-54. Roe, Kathleen. "Enhanced Authority Control: Is it Time?" Archivaria 35 (Spring 1993): 119-29. Rosch, Eleanor. "Principles of Categorization." In Cognition and Categorization, edited by Eleanor Rosch and Barbara B. Lloyd, 27-48. Hillsdale, N.J.: Lawrence Erblaren Associates, Publishers, 1978. Rousseau, Jean-Yves. "Le respect des fonds: La quintessence de la discipline archivistique." Archives: Le bulletin des archivistes du Quebec 22, no. 2, (1990): 9-14. Schellenberg, Theodore R. European Archival Practices in Arranging Records. Staff Information Circulars, no. 5, Washington: National Archives, July 1939. Schellenberg, Theodore R. Principles of Arrangement. Saff Information Papers, no. 18, Washington: National Archives, June 1951. Schellenberg, Theodore R. Modern Archives: Principles and Techniques. Chicago: University of Chicago Press, 1956. 101 Schellenberg, Theodore R. "The Principle of Provenance and Modern Records in the United States." American Archivist 28 (January 1965): 39-41. Schellenberg, Theodore R. The Management of Archives. New York: Columbia University Press, 1965. Peter J. "The Records group Concept: A Case for Abandonment." American Archivist 29, no. 4 (1966): 493-504. Peter J.. "Facing the Reality of Administrative Change: Some Further Remarks on the Records Group Concept." Journal of the Society of Archivists 5, no. 1 (1974): 94-100. Peter J. and G. Finlay. "Archives and Administrative Change: Some Methods and Approaches (Part I)." Archives and Manuscripts 7, no. 3 (1978): 115-27. Peter J., C D . Smith and G. Finlay. "Archives and Administrative Change: Some Methods and Approaches (Part 2)." Archives and Manuscripts 7, no. 4 (1979): 151-65. Peter J., C D . Smith and G. Finlay. "Archives and Administrative Change: Some Methods and Approaches (Part 3)." Archives and Manuscripts 8, no. 1 (1980): 41-54. Peter J., C D . Smith and G. Finlay. "Archives and Administrative Change: Some Methods and Approaches (Part 4)." Archives and Manuscripts 8, no. 2 (1980): 51-69. Peter J. "Archives and Administrative Change: Some Methods and Approaches (Part 5)." Archives and Manuscripts 9, no. 1 (1981): 3-18. Senecal, Sylvain. "Une reflexion sur le concept de fonds d'archives: Comment tenir compte du principe de provenance dans un contexte organisationnel dynamique." Archives 22, no. 3 (1991): 41-51. Simpson, George Gaylord. Principles of Animal Taxonomy. New York: Columbia University Press, 1962. Smith, Colin. "A Case for Abandonment of "Respect." Archives and Manuscripts 14, no. 2 (1986): 155-68. Smith, Colin. "A Case for Abandonment of "Respect," Part II." Archives and Manuscripts 15, no. 1 (1987): 20-8. Scott, Scott, Scott, Scott, Scott, Scott, Scott, 102 Stibbe, Hugo. "Implementing the Concept of Fonds: Primary Access Point, Multilevel Description and Authority Control." Archivaria 34 (Summer 1992): 109-37. Szedo, Antal. "Les Methodes de classement d'archives: documents posterieurs a 1800." Archivum 14 (1964): 57-67. Tessier, Georges. La diplomatique. 3rd ed. rev. Collection "Que sais-je?", no. 536. Paris: Presses universitaires de France, 1966. Thibodeau, Sharon Gibbs. "Archival Arrangement and Description." In Managing Archives and Archival Institutions, edited by James Gregory Bradsher, 67-77. Chicago: University of Chicago Press, 1988. Upward, Frank. "Institutionalizing the Archival Document: Some Theoretical Perspectives on Terry Eastwood's Challenge." In Archival Documents: Providing Accountability Through Recordkeeping, edited by Sue McKemmish and Frank Upward, 41-54. Melbourne: Ancora Press, 1993. Upward, Frank. "In Search of the Continuum: Ian Maclean's "Australian Experience" Essay on Recordkeeping." In The Records Continuum: Ian Maclean and Australian Archives First Fifty Years, edited by Sue McKemminsh and Michael Piggott, 110-30. Clayton (Australia): Ancora Press, 1994. Wagland, Mark and Russell Kelly, "The Series System - A Revolution in archival Control." In The Records Continuum: Ian Maclean and Australian Archives First Fifty Years, edited by Sue McKemminsh and Michael Piggott, 131-49. Clayton (Australia): Ancora Press, 1994 Walne, Peter, ed. Dictionary of Archival Terminology / Dictionnaire de terminologie archivistique. English and French with Equivalent in Dutch, German, Italian, Russian and Spanish. 2nd ed., rev. Munich: K.G. Saur, 1988. Wilds, Thomas. "Information Retrieval." American Archivist 24, no. 3 (1961): 269-79. Wright, Alman R. "Archival classification." American Archivist 3, no. 3 (1940): 173-86. 103 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items