Library and Archives Canada
Symbol of the Government of Canada

Institutional links


Librarian and Archivist of Canada
Ian E. Wilson*


The Fine Art of Destruction

Access Act Reform:

The Destruction of Records and Proposed Access Act Amendments

A One-Day Seminar and Training Session


Ottawa, Ontario
May 1, 2000



I wish to acknowledge the assistance of Richard Brown (Chief, Appraisal and Special Projects, Government Archives and Records Disposition Division, National Archives of Canada) for his fine work in the research and preparation of this paper.


As we meet today to discuss the rationale supporting the disposal of government records and potential reforms to the Access to Information Act, I think it is important to recognize that the issue of records destruction is not exactly new. In fact, records have long been the object of intentional destruction. Throughout history, for example, the deliberate destruction of records has been the constant companion of war, revolution, politics and social aggression. Consider the decision made by the first Qin Emperor of China, who unified the country and began construction of the Great Wall in 213 BC. To reinforce his position as the head of a "new world order", the Emperor ordered all previous historical writings destroyed upon his accession to power. Henceforth, he declared, "history" would begin with him.1 Or think about the events of the French Revolution following the storming of the Bastille in 1789, which witnessed a purge of all feudal documents by the revolutionary government - including its archivists - because they subjected "the feeble to the strong", and left a trail of razed châteaux in the countryside, most of them burnt down by the rural peasantry in order to destroy the manorial rolls and registers which detailed their seigneurial service obligations and taxes.2

During the twentieth century, we have witnessed many incidents of premeditated records destruction on a massive scale. Most recently, as noted in a number of news stories, the alleged "ethnic cleansing" of Kosovo Albanians by Bosnian Serbs included the systematic destruction of public documents. One Reuters news-clip which appeared in the Los Angeles Times ran as follows: "As the 20th Century draws to a close, the scene is chilling -- thousands of Kosovo Albanians are being massacred, driven from their country, their villages destroyed, their homes looted, their lives uprooted. Even archives housing their birth and marriage records are being burned by Yugoslav President Slobodan Milosevic’s Serb forces -- an act designed to deny that the Albanians ever existed".3

It is not difficult to understand why records have historically been targeted for deliberate and wanton destruction. To put it simply, they bear enormous "power". Records tell us who we are as people. They tell our stories, and they inform our culture and our history; they define political sovereignty, and our rights, privileges and obligations in society; they provide evidence of government policies, decisions, programs and services which have impact upon our lives as citizens; they sustain and facilitate access to judicial process at law. In essence, records constitute the very foundation of civilization.

There was a curious incident which occurred in East Germany about ten years ago as the Iron Curtain fell, which perfectly conveys, I think, the status and meaning which records can have within the community. In January 1990, a mob stormed the Berlin headquarters of Stasi, the East German secret police. The protestors broke up the furniture, scattered the agency’s surveillance files on the floor, and proceeded to stomp on them in what the New York Times rather temperately called "a show of popular frustration". Significantly, the mob did not actually destroy the files, but rather, realizing their potential importance as evidence, merely vented its pent-up anger against the old regime by subjecting the records to a form of symbolic erasure. In fact just one month earlier, the leaders of the East German democratic movement had acted to prevent the destruction of secret police files in the interests of identifying informers and documenting government abuse.4

Perhaps the "power" of the record is most thoughtfully illuminated by George Orwell in his novel 1984. In the negative utopia of his imaginary state of Oceania, the constant rewriting of history, the annihilation of old, outdated facts, and their replacement by new ones of changing orthodoxy, are all essential government monopolies. As the archivist in charge of preserving our nation’s historical memory, I find the archival twist in Orwell’s tale of an information society gone horribly wrong both enormously disturbing and chillingly insightful. Oceania’s national archival program -- which condones deliberate tampering with and "fixing history" in the Records Department of the Ministry of Truth under an official records destruction policy called "reality control" or "doublethink" -- is justified by the corporate slogan, "Who controls the past controls the future: who controls the present controls the past".

Of course, the vast majority of records destruction activities are not conceived for nefarious purposes. On the contrary, and I want to emphasize this point, the intentional destruction of records is a necessary and legitimate business activity, provided it is conducted rationally and in compliance with law. The notion that government should regularly dispose of its records in the interest of efficient public administration has a long and venerable history, tracing its modern roots back at least as far as the record-keeping regulations devised by the College of Notaries in Italy during the later Middle Ages, and the chancery of King Henry VII of England at the end of the fifteenth-century. For hundreds of years, governments have been routinely disposing of records no longer having further administrative use, typically following processes and procedures developed in the public interest by record-keepers and archivists. Historians may now bemoan the resulting gaps in the historical record, notably in reference to the survival of public documents prior to the nineteenth-century, but it must be remembered that many of these so-called "memory holes" are actually the product of intentional records destruction decisions made by records administrators for perfectly good and sound reasons, however unfortunate these decisions may be for historical research later on. And of course many records have been destroyed inadvertently by natural disasters. In Canada, the fires on the parliamentary precincts -- in the West Block in 1897 and in the Centre Block in 1916 -- both destroyed thousands of government files. This is one of the main reasons we now have first-class storage and laboratory facilities like the Gatineau Preservation Centre, where stringent conservation controls and fire protection measures assure the survival of government’s archival records.

Certainly since the turn of the 20th Century -- with the "professionalization" of records management practice in many countries -- the orderly retention and disposal of records has been considered a cornerstone of effective and efficient government administration. Notably, the development of more sophisticated methods to manage public records over the last century-- including their preservation in archives -- has not been a matter of mere bureaucratic exercise. As the Government of the United States discovered in the 1930s, when the files accumulated over many years of federal administration had reached the state of a virtually impenetrable mass of information, the absence of business routines to organize and dispose of records can inhibit government’s capacity to make policy, render decisions, and deliver programs and services. Quite literally, as this early example of the American experience indicates, government can either be overwhelmed by its own information, or alternatively placed in difficulty as the case may be, by ineffective access to documentation or even the absence of records altogether. It is, incidentally, the Americans to whom we must largely credit the development of modern records management techniques emerging from their efforts to fix their record-keeping problems before and after the Second War.

Here in Canada too, we have long recognized a fundamental requirement for government to have efficient and effective records administration. Over the course of the twentieth-century, beginning with the Report of the Royal Commission Inquiring into the State of the Records of the Public Departments of the Dominion of Canada in 1914, the federal government has developed processes and procedures to provide for an orderly destruction of records based on rational business principles. Records "scheduling", that is, the taking of official inventories in order to manage the life-cycle of records -- including the timing of their destruction by departments or their preservation by what was then the Public Archives -- was first introduced in 1924 with the general records disposal schedule for the administrative records of the "public departments", and the application of scheduling was subsequently extended to the operational records of government during the years 1936 to 1945 through a series of Treasury Board Minutes. Between 1945 and 1965, the destruction or archival preservation of records was authorized by the Public Records Committee. It is highly enlightening and revealing, I think, to note just how seriously the business of records destruction was considered at this time. In addition to the Dominion Archivist, the membership of the Public Records Committee was composed of some of government’s most senior bureaucrats, including the Secretary of the Treasury Board and the Comptroller of the Treasury, the Deputy Minister of Public Works, and the Under-Secretary of State for External Affairs. This Committee actually spent many hours in regular deliberation, pouring over the records inventory lists in order to make appropriate keep or destroy decisions.

In 1966 -- emerging from the recommendations of the Glassco Royal Commission on Government Organization delivered five years earlier -- the Public Records Order introduced the basic foundation of our current records disposal system by ending the tenure of the Public Records Committee and delegating the responsibility of "records scheduling" to government departments in consultation with the Dominion Archivist. Since that time, beginning with the chapters devoted to record-keeping in Treasury Board’s Administrative Policy Manual (e.g., Chapter 460), the federal government has consistently adopted and followed an official records management process which incorporates intentional records destruction by institutions for reasons of business efficiency. This process was most recently confirmed with the passage of the Access to Information and Privacy Acts in 1983, the National Archives of Canada Act in 1987, and last but not least, the implementation of the Management of Government Information Holdings Policy by Treasury Board in 1989 -- the combined effect of which represents a legislative, regulatory and policy framework which condones records destruction by government subject to certain terms, conditions and limitations.

Today, there are dozens of records management manuals available on the market which provide guidance to government and business on the development of rational records destruction methods. In many of these text-books, one observes a number of common themes I would like to touch on in reference to current records destruction and preservation activities in the Government of Canada, three themes in particular:

(1) the destruction of certain records is essential to effective and efficient business administration;

(2) information is a valuable asset which needs to be managed with the same precision of business routine and rigour of accountability as is normally provided for financial, human or materiel resources; and

(3) the management of information is an exercise in business needs analysis.

Within the broad context of these themes, I particularly want to address two critical questions of obvious interest to us gathered here today, and to Canadians generally. What records should government keep? What records should government throw away?

It would be foolhardy to suppose that these are easy questions to answer. I have already suggested that records can have enormous social, political, economic and cultural benefit, in some instances a value beyond estimation. When I think of the records preserved by the National Archives, for example, we are rating government’s information assets on a potentially dizzying scale of measure. Exactly what value would one place on the hydrographic or topographic surveys which establish Canada’s sovereignty in the Arctic, or our treaties and agreements with First Nations peoples, or the documentation of the negotiations pertinent to a bilateral trade agreement of the status of NAFTA? How does one even begin to calculate value in this context?

Perhaps the conundrum of decision-making involved in the preservation or destruction of government records was best expressed by one of my predecessors, the late Dr. W. Kaye Lamb. An historian, archivist and record-keeper of great international stature, Dr. Lamb likened the destruction of records to a "fine art", a careful, well-considered and well-planned activity conducted "intelligently, reasonably, and with common sense". As he noted in an essay written forty years ago:

The sheer bulk of modern records makes destruction inescapable. The extent and cost of storage space in which to retain them all would be prohibitive. The difficulty is to decide wisely and well what shall be destroyed and what shall be retained. At the extremes are groups of documents about which no question need arise. It is obvious that great numbers of papers become superfluous after a time -- sometimes after a very short time -- and that there would be no justification for keeping them, even if it were feasible to do so. It is equally obvious that other papers belong to categories that must be retained permanently. But between these two extremes one finds a great mass of material, the interest and long-range value of which is a matter of opinion, and it is here that the most difficult decisions with regard to the destruction of records must be made.5

Dr. Lamb’s observations continue to have great resonance today, the more so as we begin to fully realize the implications of managing information in the electronic age. In the transition from a "mass" of paper in the early 1960s to the current digital superabundance of electronic records and data, the issue of records destruction is growing ever more complicated. We are now being inundated with computer-generated records to the extent that is becoming difficult to establish and maintain control over their identification, retrieval, accessibility, authenticity, and integrity. And the scale of the problem borders on the mind-boggling. For example, government now measures its electronic data holdings of information in thousands of terrabytes. To put this into more familiar terms of measure, one terrabyte of electronic data = 175,000 metres of records, or approximately 750 million pages of text. This does not even include the paper records of government, the mass of which is growing at an ever increasing rate despite the optimistic predictions of communications technology mavens heralding the advent of the "paperless office". In the case of government’s paper records, we are now talking about millions of files, and billions of documents. I recently visited several of the federal records centres managed by the National Archives on behalf of government, where many departments store their dormant files. On one occasion, standing on a mezzanine floor twenty feet in the air and peering down over the main concourse, I saw hundreds of rows of neatly arranged numbered boxes stretching off into the distance as far as the eye could see. I could only think of the closing scenes of the film "Raiders of the Lost Ark", with the camera panning back to reveal a vast warehouse storing thousands upon thousands of identical wooden containers.

But it is not merely the volume of records which is complicating the issue of their preservation or destruction, for there are other significant pressures being brought to bear which require attention. First, there are the rising expectations of citizens regarding access to government information. As government creates and accumulates ever more information about its policies, programs and services, it is a reasonable and legitimate expectation -- in the interest of open and transparent decision-making in a modern democracy such as ours -- that the public should be able to review the records which have impact upon their lives as Canadians or residents. The facilitation of records review by the public is, after all, one of the primary purposes of the Access to Information Act. In this regard, we should not under-estimate the recent influence of the Internet, which potentially affords greater access to government and corporate information than we could have possibly imagined even just ten years ago. As the new culture of the Internet rapidly expands in Canada, people expect, require and are demanding access to government records for a variety of reasons in unprecedented numbers. In effect, the Internet has created a new and highly ravenous generation of knowledge and information consumers -- as well as a techno-industrial complex to support and supply it -- in which information is treated as venture capital and communication infrastructures are traded as corporate futures in the stock markets. By turning information into a market commodity, notwithstanding the recent hiccups at Microsoft Corp and on the NASDAQ exchange, the Internet is coincidentally raising the level of public expectation regarding information access to heights previously unknown in the history of modern communication. With such emphasis being placed on information, and upon the platforms which provide access to information, it is becoming increasingly difficult to explain the necessity of destroying records or justifying impediments to information access. Given this new information culture, and growing public demand for access to information resources, it is critical that government has a comprehensive audit trail of decision-making which fully addresses these concerns, especially in relation to information disposal.

Perhaps the most important factor complicating the issue of records destruction is the ongoing adaptation by government -- as a necessary component of its participation in global and domestic information exchange portals -- of computer technology as the communications conduit to conduct the nation’s public business. Many of us here are already conducting our own personal business by computer using the electronic commerce environment. This is certainly the wave of the future, and it is potentially "tidal" in effect. We are all likely to alter our personal business habits, and to change the way we communicate with institutions -- both government and private business concerns -- over the course of the next several years. As government itself moves progressively towards full electronic record-keeping integration and e-service delivery, however, there are a number of problems we need to address, questions in fact that we never had to answer prior to the age of high technology. For example, we didn’t really have to define what a record was, largely because we always had access to a physical object: a paper file in a box, or a register, or a photograph, or a film. We didn’t really have to think about preserving records, largely because we had standards and conservation treatments to save physical recording media. Nor did we actually have to consider the necessity of creating records because documentation of most transactions and decisions eventually found their way into notes on a paper file. Today, with so much government business conducted by telephone through the medium of voicemail, we now have to think about voicemail communication as a significant government record which requires management and preservation.

What electronic communication has done is add a whole range of new intellectual concerns, some of them fairly complicated, to the traditional, physical business of record-keeping. Because we are no longer dealing exclusively with a physical medium of object, but often with signals and traces and bits and bytes -- with virtual records which live temporarily on a monitor screen according to the specifications of the record creator  -  we now have to consider the elements and components involved in record-making and records-preserving which provide records with authenticity, reliability and integrity. We have to rethink how we identify, manage and provide access to records. Will we be able to reproduce documents as they were originally conceived and drafted? Will we be able to we say exactly what occurred? Does an emulation or a representation of what transpired constitute evidence? Will we understand all of the nuances involved in decision-making and program and service delivery? Will we even be able to find the records we need to determine what happened? Answers to some of these questions are slowly beginning to emerge, as information managers and archivists around the world, working in close consultation with the legal and audit communities, begin to wrestle with the issues.

But I want to come back to the main challenge posed by Dr. Lamb’s "fine art of destruction", which expresses the need for government to undertake a progressive reduction of records by determining their short-term, medium-term and long-term values. Regardless of all the hardware and software implications involved in managing computer-produced records, or the intricacies of such complicated intellectual matters as the creation of industry standards to determine the admissibility of electronic records in court proceedings, there is a very important and fundamental bottom-line to the modern administration of public records. One must have a rational business process and criteria in place which allows government institutions  -  and more importantly, public servants  -  to "distinguish unerringly ", as Dr. Lamb advised, "between ephemeral material and significant papers of permanent interest and value".6 In other words, there must be a business process susceptible to scrutiny and audit which explains how government makes decisions about the status of records and why it assigns value to certain records and not to others, and why certain records are retained and others are destroyed; there must be a rational records evaluation process which lends context, continuity and support to the destruction or preservation of public records. It is time that we begin to think about establishing information industry standards relating to the retention and disposal of records in particular business domains and at particular levels of decision-making, and further -- considering the transitory nature of some information -- to consider establishing criteria and standards to determine what records need to be created to document an action or a decision by government linked to a clear indication as to how long these records will be kept.

At this point, however, before we begin to look at ways to refine the processes and procedures of records destruction by government, I think it would be useful to briefly review the current legislative and policy framework. It is relatively straightforward, and I am not in the habit of quoting legislation and policy; nevertheless, in the context of our discussions today, some elements are worth repeating.

First, under the National Archives of Canada Act, Section 5(1), "no record under the control of a government institution and no ministerial record", whether or not it is surplus property, can be destroyed or disposed of without the consent of the National Archivist. In essence, any destruction of records regardless of medium -- including their alienation from the control of the Government of Canada -- without the permission of the National Archivist  -  normally conveyed to institutions in the form of a Records Disposition Authority  -  is illegal. Now I want to put some additional context around this section of statute, so as to avoid any misunderstanding of its purpose or meaning. The authority provided by the National Archivist in reference to records destruction is basically an enabling mechanism, that is, it permits government institutions to implement their internal records disposal mechanisms. In other words, when I authorize government institutions to carry out their records disposal plans -- after I have decided which records must be transferred to the National Archives -- I am not ordering the destruction of the remaining records. Rather, I am indicating that the National Archives of Canada has no interest in preserving these records because they do not meet our selection criteria conceived within the context of national archival or historic importance. In fact, the whole purpose of Section 5 is to allow the National Archives to preserve national archival and historic memory by providing an opportunity for us to intervene directly in government’s records destruction process. This reverses the traditional course of records destruction events during the scheduling years of the Public Records Committee and before, when the archives more-or-less picked the records "left-overs"once the Deputy Heads had decided what they needed to keep for continuity of business and reference purposes and what they thought should be destroyed. As such, this clause is of enormous benefit to the preservation of national history for Canadians.

Second, and this refers directly to the accountability of government institutions for records destruction, when I provide a Records Disposition Authority to an institution under Section 5, the Authority does not relieve the institution of any obligations, responsibilities or liabilities associated with the delivery of programs and services to Canadians supported and documented by the creation and retention of records. Ultimately, the decision to destroy or otherwise dispose of records which do not have archival or historic value rests with the Deputy Head of a government institution, more specifically as directed by Treasury Board under the Management of Government Information Holdings Policy, with the Senior Official nominated by the Deputy Head to manage the institution’s records. The scheduling of records and the timing of their destruction is an institutional responsibility, subject to any archival terms and conditions I impose.

Another clause in the archival legislation of interest to our discussion is Section 6, which enables the National Archivist to require government institutions to transfer records to the care and control of the National Archives under the terms and conditions of agreements. With the Records Disposition Authority provided under Section 5, the agreements provide the framework to articulate the results of the intellectual and business processes engaged by the National Archives to identify and preserve records of archival or historic value, the strategic details of which we will be hearing about later this morning. Our approach requires a very detailed explanation and documentation of the procedures undertaken by the National Archives to establish the archival value of government records, not only to assure that the rationale for every archival decision is clear and recorded, but also to assure that the reasons supporting this decision-making form part of the future archival record of government itself. At the moment, and this may come as a bit of a surprise to some people, our appraisal strategy is aimed at a target in which the National Archives preserves about 1% of the records created by government.

Lest anyone think that 1% of the government record sounds awfully small, let me give you some idea what this amounts to just in terms of the paper records, the extent of our holdings now being about 100,000 metres. The size of this "collection" is staggering and an impression is very difficult to convey, but think about driving from the Westin along the Queensway out to the west end, taking Highway 416 down to Prescott-Johnstown at the 401, then crossing the International Bridge over the St. Lawrence to Ogdensburg, New York. Now imagine your route entirely lined along the road shoulder with boxes about 1-foot square each containing approximately 2500 pages of text! We also have government photographs which number in the millions, multi-thousands of hours of audio and visual recordings, and hundreds of pieces of documentary art, not to mention about 2 ½ terrabytes of electronic data, and we are only just on the cusp of the electronic archive.

Of course, it is not the magnitude of the collection which is so impressive but rather the quality, value and significance of the records. Inside those boxes alongside the highway are Cabinet Conclusions; operational records of our royal commissions dating from 1873; immigration records of the 19th and 20th centuries including the passenger lists of the ships which brought many of our ancestors to this country; records of the men and women who have defended us since the late 18th century and the documentation of Canada’s participation in the global conflicts of the 20th century; our treaties with First Nations Peoples; the plans, drawings and specifications for the construction of our railways from 1836; the geological surveys of our land-mass; the accident reports of shipwrecks, train-wrecks and air crashes; the documentation of our diplomatic missions and international treaties. We have records with the capacity to touch directly and inform the lives of ordinary Canadians, as anyone who saw the recent story on Stephen Truscott broadcast on CBC’s program "The Fifth Estate" will know. Our goal is not merely to facilitate historical research, but to serve and protect all citizens by documenting government business and preserving government information to enable people -- many of whom would not normally think they would ever need to use the archives -- to prove citizenship, establish entitlements to pension, settle land claims, or to document other rights, privileges and obligations. The National Archives is, and I say this with the greatest pride, a remarkable institution. I invite you all to pay us a visit and revel in your historical heritage.

I began my remarks this morning with some references to mobs of people storming government buildings to destroy records. Such events were a commonplace in the early modern period. In one -- normally calm -- Italian city (Ferrara) during the Renaissance, there were no less than twenty "records riots" in the space of 40 years. Here, so endemic to the culture was records destruction by mob riot, that the authorities eventually had to sanction the practice by creating official "records burning ceremonies" to maintain civil order. Ironically in the 21st century, and we see this with the mob invading Stasi headquarters in East Berlin a decade ago, people seem to be more interested in preserving government records than destroying them. They understand the "power" of the record, especially in the democratic context of government accountability. They are insisting not only upon the preservation of records, but upon gaining access to them. In this light, I have to believe that the first Qin Emperor of China, were he ruling today, might well reconsider his decision to erase his country’s national memory.

Thank you very much.



1 Marc Drogin, Biblioclasm: The Mythical Origins, Magic Powers, and Perishabillity of the Written Word (London, 1989), 82-83, cited in James M. O,Toole, "The Symbolic Significance of Archives", American Archivist, vol. 56, no. 2 (spring 1993), 254.

2 Ernst Posner, "Some Aspects of Archival Development Since the French Revolution", American Archivist, vol. 3 (July 1940), 161-162; and George Rudé, Revolutionary Europe, 1783-1815 (London, 1964), 99.

3 Arthur Spiegelman, "Experts Debate Whether Serb Offensive is Genocide", Los Angeles Times, April 1, 1999.

4 This story from the New York Times is described in James M. O’Toole, "The Symbolic Significance of Archives", American Archivist, vol. 56, no. 2 (spring 1993), 254.

5 W. Kaye Lamb, "The Fine Art of Destruction", in Albert E.J. Hollaender (ed), Essays in the Memory of Sir Hilary Jenkinson (Chichester: Printed for the Society of Archivists by Moore and Tillyer, 1962), 50-51.

6 Ibid., 51.


Top of page