Previous - Next
2. Records containing personal information: definitions, characteristics, and special categories
Introduction
Archival terminology used in this study
Personal
information defined
Characteristics of personal information records
Special category: essential personal information
records that must be preserved permanently
Special category: the appraisal of personnel
records of government employees
Special category: the "'politics of
appraisal," genealogy, and informational value
Notes
1. This chapter begins by defining the main archival terms used throughout the study. It is hoped that this will eliminate any confusion encountered by readers from archival traditions different from the author's. Thereafter, the nature of personal information itself is explained, and then various characteristics are described about records, or aggregations of records, containing such information. These characteristics are both physical and intellectual. The chapter ends by offering appraisal criteria for three special categories of personal information record: essential records which all archivists must retain; personnel files of government employees; and records involving political or other exceptional circumstances.
Archival terminology used in this study
2. The same archival terms can carry different meanings in various countries around the world. While the significance of these terms, as well as a fuller definition of "records containing personal information," will only be apparent at the end of this chapter, it is necessary to define some basic terms first in order to avoid misunderstanding in the rest of this study. The most important terms are defined below.
3. Case File: A case file is not itself a record, but an aggregation of records and documents brought together because conceptually they share certain common interrelationships. Physically, the case file may be one or many file folders or other filing entities (box, envelope, drawer, etc.). Conceptually, the case file almost always relates to a specific event, person, place, project, contract, or transaction. Case file is the generic term used throughout this study to cover the most common and voluminous format in which archivists and records managers encounter records containing personal information. Case files may also be referred to in different countries as project files, individual case files, transaction files, personal dossiers, and so on. Nevertheless, there are two quite distinct types of case files, depending on the conceptual function of the records gathered in them:
a. Particular Instance Case Files: Case files relating to one event or transaction only. These may sometimes be referred to as particular instance papers. Examples could include applications to obtain a position of employment, to win an architectural competition, or to obtain a grant or subsidy under a specific programme.
b. Continuing Events Case Files: Case files relating to several or many events or separate transactions over time, all relating to one individual. These may sometimes be referred to as continuing case files. Examples include medical health files, insurance records, employee or student files, and police records.
4. Subject Files: Files which contain documents relating to a specific subject or topical matter. Whereas 20,000 separate case files might deal with awarding or denying scholarships to 20,000 students each year, the related subject file (or files) would concern establishing general operational procedures, methods, and problems for dealing generically with all of them, general reports on the results of the programme, and recommending changes to it.
5. Policy Files: A sub-category of subject files which establishes or defines a programme activity at a senior level. In the example above, the policy file might deal with establishing the scholarships in the first place, allocating a global budget to the programme, major changes in eligibility rules, and deciding whether or not to continue the programme. While often policy and subject records are mixed together on the same subject file, under good records management practice they should be maintained as distinct entities within the subject file classification system (usually as a secondary file block under a primary file number).
6. Personnel Files: Continuing event case files maintained by an organization for each of its employees giving personal data relating to their employment history and service. In some countries, this is called a personal file or staff file.
7. Administrative or "Housekeeping" Files: Files which relate to a function (personnel, buildings, finance, equipment, etc.) that is common to all government departments, rather than unique to the operational (i.e., specific programme) mandate of each one.
8. Series: Files, bound volumes, registers, or documents organized or maintain as a single, organic system because they relate to a particular subject area or function, result from the same activities, have a common form, or are interrelated by the nature of their common creation, receipt, or use. Their interrelationship means that they should be appraised as a single entity.
9. Appraisal: This term, which is used in this study interchangeably with "selection," 1 is the process of determining which records should be retained permanently by an archives according to a series of values. These values are discussed in this study.
10. Sampling: The selection of items or files from a series in such a way so that the items or files chosen are either representative of the whole or reflective of some significant characteristic(s) of the whole from which they were taken. Various sampling methods and terms will be discussed in Chapter 4.
11. Personal information is any information about an identifiable individual that is recorded in any format. The Privacy Act of Canada, to cite one example for general guidance, gives an extended definition of personal information to include the following: information relating to race, national or ethnic origin, colour, religion, age, or marital status; the educational, medical, criminal, and employment history or status of the individual; financial transactions in which the individual is involved; any identifying number, symbol, or code assigned to the individual; address, fingerprints, or blood type of the individual; personal opinions and views of the individual (with certain narrow exceptions regarding grants and awards); correspondence sent to a government agency by the individual "that is implicitly or explicitly of a private or confidential nature" and replies to that correspondence which would reveal the contents of the original correspondence; the views and opinions of another person about the individual; and the name of the individual where it appears in a general context the disclosure of which would implicitly reveal information about the individual.2 From another, more general perspective, personal information includes inquiries or complaints or observations received from an individual about any government programme, information on law enforcement cases or about any transactions of the individual with the state for social or other benefit programmes, statistical or computerized information about the individuals and files on current or former government employees.3
12. The above personal information can appear in many types of records. These include applications, declarations, inquiries and complaints, appeals, requests, claims, reports, contracts, lists, registers, rolls, awards, subsidies, grants, invoices, certificates, loans, payments, examinations, questionnaires, hearings, agreements, wills, leases, licences and permits, patents' registrations, passports, allowances, and many, many others.4 These types of records are usually designed forms, but their function can be expressed in letters and memoranda as well, all of which are usually aggregated in case files. It is the role of the records manager to identify, describe, and protect personal information records in accordance with the privacy act of the jurisdiction involved and with good records management practices.
13. Personal information records are created in many contexts. While listing all these would not be useful, the following are the main categories identified in a RAMP study which dealt with the issue: civil status and affiliation (births, deaths, marriages, divorces, adoptions); health (doctors' records as well as related hospital, social security, home care, and drug-use records); wealth and income (taxation, banking, investments, wages and salaries); crime and punishment (while judgements are public, all case file information of process, proceedings, some types of evidence, prison duration, amnesties, and pardons are restricted for long periods in most countries); employment records (personnel files, also student and certain client records); personal opinions, especially those advanced under explicit or implicit promise of confidentiality; basic statistical information (censuses, surveys); and police records (in many forms and considered in most countries the most sensitive type of personal information).5
Characteristics of personal information records
14. Records containing personal information have many characteristics, relating both to their physical format and the circumstances or context of their creation. Taken together, these demonstrate the richness and diversity of such information. The paragraphs in this section describe some of these characteristics of records containing personal information. As will be seen later, understanding the complexity and interrelatedness of these factors is important for appraising such records.
15. Personal information is found in virtually every physical form of paper record or aggregation of records.6 An important qualifier for this study is that personal information records are defined only as those organized and retrieved by a personal identifier: name, social insurance or other identifying number, etc. For example, subject and policy files for the purposes of appraisal are not to be considered as personal information records. Although there may be occasional exceptions, the personal information found in such records is incidental to their main purpose and they must be appraised as part of the larger file registry systems and series to which they belong, using somewhat different (or at least modified) appraisal theory and criteria.7 Similarly, case files arranged, labelled, and retrieved by names of groups, associations, and companies, rather than of individuals, should also be appraised separately, even though these may (and often do) have many records containing personal information filed on them.8
16. Physical typology. With these exceptions, the physical typology of personal information records or aggregations of records (that is to say, the various types of records where identifiable personal information can be found) are listed below in ascending order of complexity. The listing is meant to be illustrative, rather than exhaustive, for there are, as noted in section 12 above, scores of different types of records containing personal information:
a. Forms. The form is a record, printed or otherwise produced, with spaces for inserting information or data. While some forms have open "comments" or "remarks" fields, most spaces on the form are narrowly defined and the range of responses predetermined. Usually one form relates to one individual ( and sometimes his or her family), and in this way is very similar to the particular instance case file. Forms on several or many individuals may be "batched" or gathered together into one physical filing entity or serial arrangement (financial vouchers are often maintained this way). This kind of personal information is very homogeneous and thus is increasingly transformed into computer records. It is important to remember that forms may be completed entirely by the citizen (questionnaires, modern census, some petitions), jointly by the citizen and the official (tax returns, mortgage approvals, many types of applications), or exclusively by the official (police records, candidate or client evaluations).
b. Lists and Registers. These are records in which essential core information, often from other records, is extracted and consolidated, rather than maintained in loose form. (A list or register is an aggregation of information whereas a file or volume is an aggregation of records.) Lists and registers, increasingly automated, are usually in chronological or numerical order. Examples are birth and death registers, militia rolls, electoral or polling lists, and payrolls. Such information may sometimes be recorded on cards. (Registers and lists can also be finding aids to other records; see "indexes" below.)
c. Letterbooks/Volumes. As with files, volumes are aggregations of records, where incoming and outgoing correspondence are accumulated, usually in alphabetical or chronological order. Incoming and outgoing correspondence, unlike on a file, is usually maintained in separate letterbooks. Unlike the subject file, however, correspondence in volumes is often retrieved directly, or through attached indexes, by a personal identifier. While this type of record is increasingly rare in modern bureaucracies, the automated office is in many ways its electronic successor.
d. Particular Instance Case Files. These files relate to one event or transaction, usually concerning one individual. The files tend to be relatively thin, very homogeneous in content, numerous, and cover a short time period (a few weeks or months, up to two years perhaps). As a result, they lend themselves easily for conversion to electronic format. Examples are files concerning grants, awards, and scholarships; vocational training allowances; immigration entry records; and many court records (The Queen v. John Smith, the bankruptcy of Jane Doe).
e. Continuing Events Case Files. These files document the interaction of the individual with the state in several related, but separate events, often covering several years or even decades. Compared to particular instance case files, these files thus tend to be thicker, have more varied contents, and cover a longer time period. Personnel files fall into this category as do police and medical records, various student records, etc.
f. Indexes. All of the above personal information records or aggregations of records can be served by one or more indexes. These finding aids may take the form of registers, indexes to registers, lists, card indexes, file history cards, or various types of computerized indexes. These indexes must also be considered records containing personal information.
While the above six formats where personal information is recorded or aggregated are, from one perspective, a physical typology of personal information records, it is important to remember that these are also conceptual or intellectual characteristics as well. The analysis, design, and standardization behind a form or the classification system for a file must be understood by the archivist, for these give the bare physical typology of the record its contextual significance. In these matters, through using a modern type of diplomatic analysis,9 the form and structure of the personal information records must be understood by the archivist, for these will often determine or at least influence the subject content, and thus the appraisal of the value of the records.
17. Several observations may be made concerning the above typology of records containing personal information. The first that records managers and archivists alike must recognize is the interconnections between these different formats of records containing personal information. Related personal information may be found in two or more physical locations: a particular instance case file on one individual may be paralleled with or actually turned into a subject file and then later into a policy file, depending on the importance of the issue and person involved; information from the original file may be extracted and entered in some kind of register; there will be indexes to some or all of these record formats; and much of the information may be available in (or input to) computer records, the printouts of which or the statistical reports from which may well be filed back on the original paper case file. These interconnections and partial (or complete) duplications of information must of course be identified physically and understood intellectually in the appraisal process.
18. Other media containing personal information may be found either physically on the case file, or a cross-reference to such media may be found on the file, even though the other media are physically stored separately because of difficulties caused by size or format or fragility. Examples are x-rays, sound recordings of telephone wiretaps, police mug-shot photographs, cartographic material, surveillance films, and so on. As well, associated artifacts (medical samples, weapons, clothing) are often connected with or cross-referenced to a case file, but these non-record, non-documentary items should not concern the archivist.
19. Finally, the physical arrangement of records containing personal information can have a large impact on their appraisal. Letters sent to protest some major, controversial government activity -- abortion legislation, trapping of furbearing animals, immigration policy, for example -- can be used to document the ebb and flow of public emotion, if the letters are arranged chronologically. If arranged alphabetically by the name of correspondent or geographically, however, then this chronological dimension would be very difficult to discern, both for the appraising archivist and the eventual researcher. Similarly, indexing methods can distort or hide as much information as they reveal.
20. Context Typology. The context characteristics of a file with records containing personal information -- quite aside from its general subject content or thematic programme area (immigration, taxation, etc.) -- include its general focus (scope or coverage) and the circumstances and location of its creation. There are many issues to be considered here:
a. Focus (Scope and Coverage).
-- Is the principal focus of the case file series on the actions and thoughts of the government employee administering the programme, the person (or citizen in this study) receiving or interacting with a programme, a third party, or a combination?
-- Is the coverage universal for the entire (or at least entire adult) population of the country or limited to special groups (armed forces, immigrants, various tribes), classes or occupations (miners' health files, pilots' licences), genders (family allowances), ages (pension claims or educational grants), or regions (agricultural subsidies, fisheries allowances)?
-- Does the coverage include the rejected and unsuccessful cases as well as the accepted and successful ones?
-- Within the coverage, are all cases available and documented in a standard way or are some consciously cut out (no file created), lost, or overlooked?
-- Do the files collectively cover the same time span as the programme under which they were created, or just a portion of it?
-- Are the files related to other series and other programmes?
b. Circumstance of Creation.
-- Is the documentation created directly by the employee and/or citizen, or indirectly by a. third party (lawyer, translator, referee, reporter)?
-- Is participation in the programme required by law or voluntary?
-- Has the documentation remained in raw, unprocessed form or has it been replaced or supplemented by aggregated data?
-- Is the information in the paper record transformed in the normal course of business into microfilm or electronic format?
-- Is there one file or several files created on an individual in any one programme?
-- Are exceptional, controversial, and precedent-setting cases created or maintained separately from the routine and regular cases, by different file jackets, numbers, or colours or by indexing or abstracting the relevant information or by creating additional files (perhaps at an appeals or special hearing level)?
-- Is there significant discretion by the employee or citizen in creating the documentation or applying and interacting with the programme?
-- Is the main emphasis of the programme and thus its case files a single function (judicial, regulatory, investigative, licensing, taxation, subsidy or grant or award, social service, etc.) or multi-functional?
-- Despite a possible apparent homogeneity of a file series' content, format, and physical appearance over time, were there significant changes in the administrative structure, implementing personnel, mandates and policies, or even legislation of the programme, that may have affected the files' content?
-- Were the files used for more than one purpose by the creating or perhaps later administrating unit of the government?
c. Location
-- Are there various levels in the bureaucracy -headquarters, region, field -- that created documentation on the individual's interaction with the programme, whether on one central file or on several files kept in each office of the administrative hierarchy?
-- If there are several files and/or several levels of bureaucracy interacting with the individual, are there formal or informal linkages of the resulting information?
21. It is important that archivists identify and assess both the physical and context characteristics of case files, for as will become apparent in the next two chapters these have a direct bearing on determining their potential permanent value within a theoretical appraisal model. These characteristics, however, are often not obvious, but subtle and buried. They can only be unearthed by the archivist undertaking thorough research into the nature and history of the records and their creators.
Special category: essential personal information records that must be preserved permanently
22. There are certain categories or classes of records containing personal information which should be preserved by archivists around the world. These records may not necessarily be stored physically in the national archives and may be created by state or provincial and local or municipal governments. Nevertheless, their importance to research by providing the core demographic profile of the nation, to individuals' legal rights, and to government administration is incontestable. Archivists should collectively ensure that these categories of personal information records are safeguarded:
a. Records Proving Civil Status. Records, usually forms and registers, recording births, deaths, and marriages should be preserved, as well as records of divorce and adoption, and citizenship and naturalization.
b. Land Registry Records. Records documenting the surveying and disposition and transactions of land and other fixed property.
c. Records Falling Under Statutory Requirements for Permanent Preservation. In certain countries, court records must by law be kept permanently, as must wills and decisions of many types of regulatory and semi-judicial tribunals and boards and agencies The permanent preservation of all court records would present insuperable problems; archivists should verify which records must be kept permanently (on this issue of selectivity, see section 23 below).
d. Census of the Population. The national census is the single most essential personal information record in terms both of research for many disciplines and for genealogists, and of providing the core demographic information vital to the design, delivery, and modification by the government of its own major programmes. Census data is of course very sensitive, and archivists must safeguard it from public disclosure until such time as the sensitivity has disappeared.
23. It should be noted that not all records connected to these four essential categories must be preserved. For example, while all the raw census data is essential, the questionnaires on which it was collected, various forms and process documents, and related subject files are not essential, and should be appraised by other criteria. Similarly, for court records, it may be that only the actual judgements and case precedents are vital, whereas the actual court operational records, transcripts, and case files are not. In short, for these few categories of essential personal information records, whose retention is not in question, the problems they present to archivists do not concern appraisal, but rather preservation (storage cost and space, privacy and use, and their possible conversion to microfilm, electronic media, or optical disk). Such practical and technical issues, however, can affect the ultimate decision of how to acquire the records and thus do form an element, albeit a secondary one, in the appraisal process (see Chapter 4 for more on this distinction between appraisal and preservation).
Special category: the appraisal of personnel records of government employees
24. The personnel files of the government's own employees (including two subgroups: members of its armed forces and police) are a special category of personal information records. As mentioned before, personnel files are continuing event case files. However, for convenience, they are handled separately here. Personnel records are usually the only category of "housekeeping" or administrative records that contain personal information of significance. The retention of such records for their period of primary administrative, legal, and fiscal use -- usually the employee's active career and pensionable retirement -- is not an issue of archival appraisal, even though this period may extend to seventy years or more. The permanent value of such records primarily rests on their quantitative and statistical information, and will come from manipulating the electronic version of the record. Sociological and historical research into worker absenteeism, patterns of illness, gender and age issues, wages, human resource policy, and career mobility would require such data.10
25. In addition to this raw quantitative or aggregated information, which is usually best preserved in electronic format and not in the bulky original paper files, there are qualitative reasons for keeping some categories of civilian (i.e., non-military) personnel records in the original paper format. Noted individuals serve the state in all countries, and for the purposes of biography and the history of administration, as well as of the subject fields in which the public servants were involved (science, agriculture, transportation, and so on), their files should be preserved. There can often be found on such files, in addition to the paper version of information extracted to the electronic media, press clippings of noteworthy accomplishments, key speeches and publications, notices of awards, congratulatory letters, curriculum vitae, courses attended, bibliographies, and so on. It is to be noted, however, that the value of this qualitative information tends to be much lower in the 1980s than in previous decades for two reasons. First, various privacy laws now severely restrict the amount and type of personal information which employers may create or retain on their employees, and also require the culling of files and destruction of certain types of information (work performance appraisals, for example) after short periods of time. Secondly, most basic demographic information in the paper file for more recent years is now available in electronic format.
26. Nevertheless, some personnel files, especially for earlier years, do contain valuable information. What categories of public servants' personnel files to keep permanently in archives is difficult to determine. In this regard, the Canadian model may be suggestive, although it is presented as a guideline only, with the recognition that circumstances will vary from nation to nation. In Canada, the following categories of civilian personnel files are kept: the top four levels of the government administrative hierarchy for all departments (the file on any person reaching at any point in his or her career the rank of deputy minister, assistant or associate deputy minister, director general, and director); the top level for all boards and commissions (chairperson, commissioner); all ambassadors, consuls general, and heads of missions in the foreign service; the top four levels (down to inspector) of the national police force; captains of all government vessels; surveyors and engineers involved in significant public works; all federally appointed judges; and senior officials connected with the head of state (Governor General), the legislative function (Parliament), and the judicial function (Supreme Court and Federal Court).
27. Other significant files to be kept, although more difficult to identify, are those of Canadians, in addition to the above, who did not reach these formal senior positions in federal government employment, but who in their careers either inside or outside the public service were recognized leaders in academic, scientific, business, medical, legal, or other fields; sports figures achieving national or international stature; recipients of major national and international civilian or military awards or decorations; individuals who served as Members of Parliament or of provincial legislatures, or who were mayors of large cities; and all civilian government employees killed while on duty (prison guards, surveyors, hostages). Furthermore, the files of employees who occupied unique positions in the government or who had unusually long careers may also be considered for archival retention, as well as the caveat category of those of "any individual involved in any act or event of major historical interest." 11 There are also other appraisal considerations for series of personnel records, in terms of treating them in common with all other personal information records, which will be addressed in the next two chapters. This section only addresses particular aspects of appraisal relating to these records.
28. A special category of personnel records relates to those of the armed forces of the nation. Some countries are inclined to preserve all such records generated during periods of active warfare, as a memorial to those who served (and perhaps died for) their country and as files likely to have more intrinsic research interest. For peace-time military personnel files, in addition to preserving a profile of all armed forces personnel through various electronic records in data bases, a similar approach is recommended to that adopted for civilian personnel records in the preceding paragraphs: all files of all personnel reaching certain ranks (which will vary from country to country) in the armed forces should be preserved, as well as those of military personnel achieving particular distinction, awards, prominence, and so on.
Special category: the "'politics of appraisal," genealogy, and informational value
29. As noted in Chapter 1.10 above, appraisal is sometimes taken out of the archivist's hands by political pressure. Whether case records from the Nazi period in Germany or military personnel files in the United States or financial records concerning Canadian Indians, certain national traditions make destroying some categories of records very difficult for the archivist. Here it is not appraisal criteria that are brought to bear, but rather resource questions of how far the archives can accommodate these pressures. The most obvious example concerns genealogy which, in recent years, has enjoyed an enormous revival and placed a concomitant pressure on archives to preserve records for this purpose. In simple terms of archival appraisal, guidance may be clearly given: in a national archives, case files of most operational programmes should not be kept for purely genealogical value or reasons. That does not mean genealogy is not important or that genealogists will not find useful information in samples or selections from case file series kept for other reasons, but it does mean that very few such series are preserved purely for genealogical reasons. Otherwise, archives around the world would keep every case file permanently for, by definition, they all have genealogical value. Just how many files and series archivists do keep will depend on the amount of public pressure to which they are subjected. Of course, genealogical information on a large scale is preserved in the four essential categories of records outlined in section 22 above. Such records, combined with printed and local sources (city directories, tax rolls, and so on), would provide basic genealogical information on all citizens. Any further detail on individual lives from specific programme case files, kept solely for genealogical value, may with a few exceptions be a luxury archives can ill afford.
30. There are also series of personal information case files where the archival concern is primarily (and sometimes exclusively) the informational value they contain on specific individuals, places, or events. This is in contrast to the evidential value the files may have in shedding light on the nature of the government institution which created them or the collective value they may have in how they reflect the whole to which they are a part. Informational value as a concept exists in and of itself, without regard to what it evidences about government operations. Census returns or immigrant landing records are classic examples.12 Two other illustrations of informational value might be small series of files concerning art acquisitions by a national gallery or senior research grants to leading scientists; these files would be primarily important for the information they contain on each donor and painting, or on each scientist and his or her project, without denying that both series (or a sample of each) would also contain evidence of the operations of government agencies. Conversely, a series of files documenting the subsidies for vocational training paid to hundreds of thousands of workers annually is not valuable for the information it contains on any one worker, but rather collectively (through a small sample, most likely) for the evidence it provides of how the programme functioned and was received. The appraisal of personal information records for the informational value they contain on specific individuals is straight-forward and follows the criteria of appraisal generally. Such criteria will be outlined in Chapter 4 (and have already been alluded to in the discussion of personnel records).
31. This study, however, focuses on the more difficult problem of appraising records containing personal information for their collective or evidential values. To understand better the archivist's task in appraising such series of records, which of course form the vast majority of personal information records, it is necessary to place them into a theoretical context concerning the way society creates and uses these kinds of records.
1. I recognize that this equivalency is not technically correct. In theory, archivists would decide which groups or fonds of records to target through an acquisition or documentation strategy of some sort. That would be followed by appraisal, which is the choosing of series of records, or sometimes blocks of files within series, from those targetted fonds for transfer to the archives on a temporary basis. Selection occurs later on in the archives as the archivist arranges and describes the records, and selects out (and destroys) those without value while designating the rest for permanent retention. In working reality, there is (until very recently) no strategic targetting at the front end and very lisle in-house selection at the back end; there is simply not the luxury of double appraisal, especially when dealing generically with case files, and thus "appraisal" and "selection" occur simultaneously.
2. Canada, Privacy Act, 29-30-31 Elizabeth II, c. 111, section 3. Curiously, with the exception of the Canadian act, most privacy legislation at the national level does not give a thorough definition of personal or private information: on this point, see Michel Duchein, Obstacles to the Access. Use. and Transfer of Information from Archives: A RAMP Study (Paris, 1983), p. 19.
3. Canada, Minister of Supply and Services, Personal Information Index 1988 (Ottawa, 1988), p. iii.
4. This partial listing is taken from the Canadian Personal Information Index, pp. SI-1 to SI-33, and passim. For a very full list of the many different categories of individual records, see Maynard J. Brichford, Archives & Manuscripts: Appraisal and Accessioning (Chicago, 1977), pp. 22-23.
5. The categories are from Duchein, Obstacles to the Access. Use and Transfer of Information from Archives, pp. 19-23. The Canadian Personal Information Index lists more than 4,000 separate series of personal information records, from agricultural subsidies to veterans' treatment, from housing loans to postal contracts, from student summer employment applications to immigrant medical forms, from bankruptcy trustees to old age pensions. This is an excellent source for records managers and archivists wishing to conceive a way of identifying, describing, and controlling records containing personal information.
6. The following typology mixes together records per se (such as a form or index card) and aggregations of records (such as files and volumes). Having made the distinction several times in the text and again now, for ease of reference and not to burden the text with excessive qualifiers at each point, I shall refer to both generically as "records," even while recognizing that this common usage is not technically correct.
7. Concerning subject and policy files, personal information is often found in the individual memoranda, forms, reports, letters, and other documents on such files. The personal information is usually qualitative and subjective rather than quantitative and statistical, although useful reports of statistical summaries and aggregations of personal information may also be placed on such files. Often individual cases which have precedent-setting value at the level of procedures and practices are transformed by records managers from a particular instance case file into a subject file, or at least the individual subject of the case file is referred to in the documents on the subject file. Such personal information records found on subject and, rarer, policy files are not homogeneous, often scattered and episodic, and not retrievable by a personal identifier (name, identity number, etc.).
8. Such scattered personal information on subject files is, of course, still personal information in terms of being treated with proper security and confidentiality by records managers and archivists and in terms of falling under the provision of privacy legislation.
9. Diplomatics, which is still generally unknown in North American archival circles, studies the actual form of documents in order to ascertain their validity and value as historical sources. Where archival science focuses on the provenancial context of records creators and of the records themselves and their informational content, diplomatics concentrates much more narrowly on the form of individual documents to elucidate their authenticity, status (original or copy), formal authorship, character as private or public document, and so on. For more details on the European origin of diplomatics, its nature, and possible applicability to modern records issues, see Luciana Duranti, "Diplomatics: New Uses for an Old Science," Archivaria 28 (Summer 1988), pp. 7-27, which is the first of a series of six articles. See as well Don C. Skemer, "Diplomatics and Archives," American Archivist 52 (Summer 1989), pp. 376-82.
10. For electronic personnel records and their appraisal, see Harold Naugler, The Archival Appraisal of Machine-Readable Records: A RAMP Study With Guidelines (Paris, 1984), pp. 42-43.
11. These categories are taken from the operational guidelines that accompany the General Records Disposal Schedule of the Government of Canada, Schedule 5: Personnel, which is issued periodically by the National Archives of Canada. The guidelines are in National Archives of Canada, operational file 9440-5, memorandum of Federal Archives Division to National Personnel Records Centre, 10 May 1982. The Personnel Records Centre is responsible for applying these appraisal criteria, developed by archivists in the Federal (now Government) Archives Division, to the personnel records of all government departments and agencies.
12. While conceptually this statement is true, in reality there is no such clear-cut distinction, even in the most obvious cases. For example, with the national census, the nature and design of the questionnaire form, the different types of data collected over time, and the markings of the census officers are all examples of evidential value. However, if these things were the only value of the census records, a small sample would be sufficient for archival purposes.