New Arrivals    Books    Archival Products   Charts   Newsletters   Upcoming Events   Contact Us  

Popular Categories

   How-To - Genealogy Misc.
   How-To - Write & Publish
   How-To - Conservation
      - Acadie, Acadian
      - New Brunswick
      - Newfoundland & Lab.
      - Nova Scotia
      - Ontario
      - Prince Edward Island
      - Quebec
      - Western Canada
      - First Nations, Metis
      - Military - Before 1920
      - Loyalists / UEL
      - Pioneers' Stories
   British Home Children
   England & Wales
   Ireland & Northern Ireland
   United States
      - American Revolution
   more countries...

   Archival Products

   Genealogy Charts

   Gift Certificates

Popular Authors

   Thomas MacEntee
   Paul Milner
   Chris Paton
   Ron W. Shaw
   Gavin K. Watt

Popular Publishers

   Global Heritage Press
   MacDonald Research
   OGS - Ottawa Branch
   Unlock The Past

Search by topic, title, author or word:

News & How-To
Formerly branded as

Articles, press releases,and how-to information for everyone interested in genealogy and history

Subscribe to our free newsletter

Article Published June 30, 2000, Vol. IV No. 11

Gordon A. Watts POST-1901 CENSUS NEWS (Canada)
By: Gordon A. Watts,

Greetings Readers, and Members of Parliament

Keep Census data secret, public tells Ottawa?

I am on the road again, this time for a week or a little more. I will be gone at the end of the month but will be checking email and searching for the report of the Expert Panel on Access of Historic Census.

A recent report in the National Post of 20 June 2000 bore the headline Keep Census data secret, public tells Ottawa. The article was based on a focus group survey consisting of six groups totaling a maximum of 72 people. Hardly a number that could truly be said to represent the views of all Canadians. I wrote a letter to the editor regarding this article and they printed a somewhat edited version on 26 June 2000. This is a last minute addition to my column as I go out the door and I will have more to say about this in my next column.

Submission of Don Nisbet to the Expert Panel

As I expected to be on the road at this time, this column was prepared in advance. Not a difficult chore, really, as the column consists mainly of the submission of Don Nisbet to the Expert Panel. Don is a part of our ad-hoc Committee and has been an avid researcher on issues relating to our Census campaign. He is a frequent contributor to the Canada-Census-Campaign mail list, and, as is evident in his submission, is a pretty fair writer in addition to his research abilities. Without further ado, here is his submission.

Submission by
Donald A. Nisbet

to the Expert Panel on Access to Historical Census Records
April 24, 2000

"The law must be consonant with life. It cannot and should not ignore broad historical currents of history. Mankind is possessed of no greater urge than to try to understand the age?old question: "Who am I?" "Why am I?" Those emotions and anxieties that generate our thirst to know the past are not superficial and whimsical. They are real and they are "good cause" under the law of man and God."

Judge Wade S. Weatherford, Jr.
Seventh Judicial Circuit Court, South Carolina
Ruling on an adoptee's petition to gain access to
adoption records.

Dear Panel Members

I am one member of a cross-Canada committee of genealogists who support release of name-identified census records to public access after a reasonable period of time. As our group is an informal one with no official standing I am presenting this submission on my own behalf. Although it represents my personal views I believe it reflects in large the opinions of most Canadian genealogists.

Our group has made every effort from the outset to listen carefully to all opinions on this issue and where those differ from our own to support our views with careful research and attention to the facts and to express those views to the best of our abilities. We welcome the opportunity for a full and frank discussion of all aspects of this important issue and have full confidence in the expertise and impartiality of the panel members.

As our elected representatives are just beginning to learn we genealogists believe passionately that the ability to discover the heritage of our families and country is not a privilege but a right. A right that requires reasonable, balanced, and lawful access to personal information held by government including old census records. We also recognize that the legitimate and necessary right to personal privacy of all Canadians, including ourselves, must be part of that balance. No right, including the right to freedom of information, or the right to privacy, can be absolute and no civilized society can function without such balance between valid but competing rights.

The question for the panel is then this: is the release of historical census records to public access after 92 years a reasonable, balanced, and lawful answer to the important right and need of Canadians to know their personal and national pasts. We genealogists believe it is. It is lawful as access to personal information including census records is written into the statute and regulations of the Privacy Act. It is balanced because it recognizes and provides ample protection for the privacy of personal information supplied by citizens to government. It is reasonable because without such access an understanding of the past is impossible and important rights may be denied.

We believe that with the application of mature judgment to the facts it will be seen that the release of historical census records for legitimate research purposes after 92 years is a sound policy that carefully balances the rights of privacy with the equally important right of Canadians to access government information. There is no reason to tamper with this balance which was achieved after full debate in 1983 with the passage by Parliament of the Privacy Act. A simple amendment to the Statistics Act is all that is required to break this impasse and to bring that statute in line with the intentions of Parliament that Canadians should possess a reasonable right to access government information of all kinds. That right is a fundamental pillar of a democratic society.

In my submission I will touch on a number of issues raised in this debate: how genealogists make use of census records and their unique value; the role of census records in medical research; the claim that other records could easily substitute for the census; the intent of the Statistics Acts in force since 1905 and the claim of a promise of perpetual confidentiality; the contention that release of historical censuses harms participation in future enumerations; and, finally, I will comment on a number of issues raised by the Privacy Commissioner in his brief to the panel.

Genealogists believe that preserving Canada's census records and opening them to public access after 92 years strikes a fair balance between privacy and access rights and represents the views of most reasonable Canadians. We hope the Panel in its report to Mr. Manley will concur with that view.

Donald A. Nisbet


I will start with a few brief comments on the purposes and motivations of the millions of Canadians who undertake the sometimes difficult and expensive and always time consuming task of tracing their family roots and of its value to the country. The motivations of genealogists were the subject of a recent study by sociology Professor Ron Lambert of the University of Waterloo:

    Genealogists interest in the past is firmly rooted in a desire to know their ancestors as people and to bequeath this knowledge to future generations. They see themselves as a bridge or door between generations. There was little evidence of a religious motivation, a desire for illustrious ancestry, or a pursuit of the illusion of stability. Nor was there any evidence of a nostalgic and wistful yearning for an ethnically simpler time. For many genealogy provides a perspective on time which unites the past, present, and future that renews and enlarges the family, now extended in space and time. (Doing Family History, Families, February 1996)
That certainly reflects my intentions and those of the many genealogists with whom I am acquainted. We pursue genealogy out of personal interest of course but also as a contribution to our families . And despite what some may suppose genealogy is a profoundly social activity that connects the individual to the family and the family to the community and country. When I first took up genealogy 22 years ago I was a 6th generation British Columbian with no close family and but a few acquaintances or friends beyond my native province and with little knowledge or frankly little understanding of the rest of my country other than as an occasional tourist. That has changed in a profound way solely because of my genealogical pursuits. I now have a deeper respect and understanding for the constituent parts of this diverse country and a network of acquaintances, friends, and cousins from coast to coast that has made me feel a little less like a British Columbian and more like a Canadian without lessening my pride in my own province. That experience is not unique but commonplace among genealogists, it comes with the territory.

It is often what is least apparent that turns out to be of most value. What is not obvious to those yet to be swept up by an interest in family history is the intricate web of relationships spread across this country built up over time by genealogists and how important that network is to the unity of our country at a time when we seem more divided and parochial than ever. No stronger solvent for prejudice and misunderstanding exists than to come to know on a personal basis fellow Canadians from all walks of life and every part of the country who, despite the differences, turn out to be quite like ourselves. Ask those Canadians who have discovered often to their surprise that in their veins flows the blood of First Nations people or of one of the founders of old Acadia or New France or the Quebecois who finds in his family tree that Irish great-grandmother or English great-grandfather. Multiply all those individual discoveries by seven and half million (the estimated number of Canadians pursuing genealogy) and you only begin to grasp the positive role genealogy plays in this country. The notion that historical census records are a source of harm to individuals stems from an inability to separate the true sensitivity of current information from 92 year old personal data that can injure neither the living or the dead. Blocking access to historical censuses would be an ill-judged overreaction to privacy concerns that would harm the rights of the living for no reasonable or defensible purpose.


But are census records really that important to genealogists and can they not be replaced by other available records? These questions were I'm told raised by the Panel in the February 9th conference call with genealogical societies. I do not present myself as an expert but perhaps as a genealogist of 22 years experience I can answer these questions with some confidence. Genealogists do use a wide variety of records in researching their family trees and writing their family histories. However, the census is so unique in itself that it has no effective substitute and thus its importance to genealogy.

A census is a national snapshot of the entire Canadian family taken every 5 years. It is inclusive not exclusive: rich and poor, young and old, First Nations and immigrant (whether a descendant of New France or a recent refugee), the gifted and the ordinary, all of us. Census records are unique as they capture a wide range of personal information about nearly every Canadian at the same time and show them in the context of their families, neighbourhoods, and communities. It tells us how they earned their living and housed themselves, their religious beliefs, about their ethnic and national origins, family structure, and much more. As social historian Professor Donald DeBrats expressed it when lamenting the destruction of Australia's census records:

    [Our history] would be much more focused on the lives of ordinary people - men and women, black and white, immigrants and native-born - if the census of the past had been preserved. Not only would this be a different history but I believe it would be a history in which ordinary men and women ... would be engaged in and involved in, to a much greater degree than they presently are. They would see that the real history of their nation is not what happens in Canberra, on battlefields or in the diaries of famous men, but what happens in the ordinary lives of ordinary citizens each day. The census creates a people's history because the census is the only record of the people. It is the only record in which the people - all the people -- speak. (Saving our Census and Preserving our History, Report of the House of Representatives Standing Committee on Legal and Constitutional Affairs, 1998)
Perhaps some might question Professor DeBrats emphasis on the ordinary person in national history but to genealogists our focus is for the most part on just those millions of ordinary Canadians who make up our past, present, and future. Those of wealth and prominence who lead the country will always leave ample evidence of their accomplishments behind and will have their contributions well marked in the archives and history books while the average person will leave behind few traces of their life except perhaps a few scattered personal possessions or papers, a photograph or two, and perhaps an occasional anecdote passed down through the generations. As genealogists our job is to recall and honour in an honest and realistic way their modest contributions and to restore them as best we can to family memory. Census records are the one primary record set that most helps us do this.

Other kinds of records containing personal information add to but do not replace what can be garnered from census forms. Vital records for example only record one specific event for one family member at one time. A birth record might record, accurately or inaccurately, an ancestor's name, date and place of birth, and the names of her parents. Very valuable information but it will not show an ancestor and her family over a period of time. It will not record migration, schooling, siblings and their birth order, the loss of a parent, remarriage, the names of grandparents and other relatives, changes in occupation and financial status, and the names of neighbours and their children who will be family friends, schoolmates, and future marriage partners. Nor will those other records show as clearly the maintenance or loss over time of cultural identities that speak to persistence or assimilation: marriage across racial and ethnic lines; the survival or loss of a language or other attributes of distinctness such as traditional occupations, religions, or naming practices. All of this genealogically and historically valuable information, recorded on a regular basis every five or ten years, demonstrates the growth and development of Canada's families, ethnic groups, and communities in a way no other document can possibly do.

Census records do not tell the whole story nor are they error free but an experienced or creative genealogist can use them not only to solve genealogical problems directly but as a guide to formulate new questions that point to other record sources for future research. Genealogical method requires that we work back from the known to the unknown which is similar to the kind of reasoning a police detective would use. Often what is known in the present is not very detailed and sometimes contradictory: a death certificate might tell you your great-grandfather was born in Ireland while a family elder insists it was Nova Scotia and your cousin is adamant it was "somewhere in Ontario". Often on locating census records, assuming you ever figure out where the family lived at the time it was taken, you may find all three are partially correct: your great-grandfather's parents were from Ireland and his two eldest siblings were born there, he was born in Nova Scotia, and his younger brothers and sisters were born in Ontario. You will have encapsulated in one census the year and country of birth as well as the migration and emigration history of all family members. No other record past or present can accomplish that.

But there are numberless examples of how census records can be used to solve genealogical problems. Supplying clues to discover lost females lines - one of genealogy's perennial conundrums - would be one, as would their aid in identifying other families sharing common points-of-origin which is particularly helpful when research focus eventually turns to "the old country". Their value is inherent in the nature of the census records themselves: there is no substitute and the loss of, or denial of access to, these irreplaceable records would be a tragic sacrifice of our common heritage for no good purpose.


One of the most important uses of historical census records today is one our great-grandparents could not have conceived of when they replied to the enumerator's questions in 1906 or 1911 - their use by genealogists to build family trees important to scientists and medical doctors studying the causes, treatments, and potential cures of genetic diseases. That genealogy should be playing an important role in this medical revolution may be surprising to many, including some genealogists themselves, but genetic scientists are today melding the high technology of molecular biology with the classical approach of tracing genes through family trees into an effective tool to unlock the mysteries of these diseases which have plagued mankind for eons. Unquestionably the study of human genetics is revolutionizing the science and practice of medicine. Geneticists have already identified thousands of hereditary conditions including forms of common illnesses such as diabetes, heart disease, arthritis, cancer, and mental illness and genetic testing for some of these such as hereditary breast cancer are now available giving those carrying the faulty genes some control over the risk of developing the disease. And progress on promising treatments such as gene therapy and new generations of pharmaceuticals that target biochemical imbalances at the molecular level are at last bringing the science fiction of effective treatment and even cures for genetic diseases closer to scientific fact. For the first time in history there is hope for those millions of men, women, and children suffering the debilitating and often fatal effects of genetic disease.

Genealogy has thus assumed an important role in the health of Canadians. Increasingly knowledge of the health history of families and the patterns of transmission of hereditary diseases has become of great interest and value to individuals and the medical community alike. Family trees are used extensively by doctors, genetic counselors, and medical scientists to promote the health of Canadians: physicians use family trees to inform themselves of current medical problems of their patients and to monitor for future problems; genetic counselors use them to trace the inheritance patterns of genetic diseases and to advise their clients on life choices affecting their personal and family health; and genetic researchers use family trees (medical pedigrees) as a research tool in locating and identifying the particular gene or genes that play a role in specific hereditary diseases. As genetic counselor and author Carol Daus explains it:

    The genealogist and the geneticist both set out to accomplish a similar task: finding patterns in our family relationships. But aside from our most obvious hereditary traits, a closer look at your family history can reveal critical information about your present and future health. It has been proven that many of the illnesses and conditions that beset us can be predicted, diagnosed, and even prevented by studying our own family trees. As a result, increasing numbers of physicians and other health professionals recommend the need for developing thorough family health histories by tracing lineage back three or four generations.

    It's unfortunate that many of the worst diseases we suffer from are transmitted genetically. But the good news is that an accurate, detailed family tree is one of the best insurance policies you can have against illness - for yourself and your children. By knowing which illnesses you especially need to watch out for, you can take positive steps to prevent them or minimize their effect.

    There's another important reason for studying family health histories. If enough family trees became available for study, scientists could identify the genetic factors involved in the transmission of diseases like cancer and schizophrenia. Such a breakthrough could finally solve the mysteries of many illnesses and lead to effective new ways of preventing and treating them. In fact, mutations or changes in genes that bring about or increase the risk of certain diseases have already been identified, and in some cases genetic tests are now available that will tell you if you have that specific mutation. [Carol Daus,(Genetic Counsellor), Past Imperfect: How tracing your family medical history can save your life.]
Thus the quality of the monitoring, counseling, and research of all these health professionals ultimately depend on accurate and complete family trees supplied by family members. This trend will continue as the role of genetic medicine in health care assumes more and more importance and with it the need for accurate and complete medical family trees will grow. The basis for these family trees is first and foremost Canada's name-identified census records. Only from them can the large family groups so especially critical to genetic researchers be tracked and studied in a cost-effective manner. The loss of these records to genealogists will have such a major impact on the right of Canadians to know their family medical inheritances and so reduce their ability to control and protect the health of themselves and their children that this issue must be given very careful consideration by the panel and the government.

Certainly our current federal government has already recognized the critical importance of genetic research to the future health of Canadians and their economy and is now pouring hundreds of millions of tax dollars into this enterprise. Government is convinced it is money well spent and will in the long-term help prevent and treat diseases in so effective a manner it will reduce the burden of health care expenditures on the country. Those costs in health care dollars, lost productivity, and in human suffering are substantial: during their lifetime over 50% of Canadians will have a disease with a genetic component and over half of admissions to pediatric hospitals and over 12% to general hospitals are for genetic diseases. In response to these needs the Honourable Allan Rock, Minister of Health, has established a Population Health Fund that "focuses on the range of individual and collective factors that determine the health and well-being of Canadians" and recognizes that one of the key determinants is genetic endowment or those "inherited predispositions that influence the ways individuals are affected by particular diseases or health problems."( www.hc?

Finally, some remarks by the Privacy Commissioner in regard to the medical content of census records needs to addressed. Mr. Phillips has said that "genetic and other health information ... can be extracted from the census records". (Submission, p.4) Census records are filled with genealogical information but contain no genetic information and little if any useful medical information can be "extracted from the census records". I am uncertain how he defines these words and the usage of the term genetic puzzles me but if I take the widest possible meanings for these terms as referring to indications of health status then the number of questions that could remotely be termed as providing medical information from all censuses taken between 1906 and 1996 are confined to the two activity limitation/disability questions asked on each of the 1986 through 1996 long forms (1 of 5 households), one question in 1951 (Is this person permanently unable to work?), and four questions on physical and mental disability from the 1911 Census . Those are the only questions from all of Canada's twentieth century censuses that can be called medical or health questions and none of them identify specific diseases and would be useless to a medical practitioner.

The value of census records is not in providing direct medical information on disease but by providing a powerful research tool for genealogists to build accurate and complete family trees. Once accomplished other records containing specific facts or clues about medical conditions are consulted such as family knowledge and personal papers, or private papers such as medical or hospital records when such access is permitted, and public records such as death certificates or obituaries, and so on. This process converts the genealogical family tree into a medical family tree and this is converted in turn into a medical pedigree which codes the information into scientific symbolism and removes all references that identify a particular person. This pedigree can then safely function as a scientific tool for genetic research under strict ethical rules of use and conduct that apply to all medical research and provides for full protection of the privacy rights of individuals.

Mr. Phillips is not wrong to point to the ethical concerns that genetic research raises but to answer these by destroying records is foolish and smacks of fear-mongering. What is needed is a continuance and expansion of the ethical and legal rules controlling such research to ensure that it serves legitimate human and scientific purposes and that individual rights will not be injured. The expansion of human knowledge always brings with it potential for harm but the answer to this reality is not to follow the Luddites of the 19th century by attempting to stop medical progress but rather to intelligently control and regulate it. And finally, in these discussions the millions of men, women, and children who suffer the painful daily realities of these debilitating and life-threatening genetic diseases should never be forgotten - they deserve every opportunity that can be given them to live productive and healthy lives.


Some argue that census records need not be retained as historians and genealogists have now, and will have in future, access to vast amounts of government held records containing personal information on Canadians. The Privacy Commissioner's submission talks of "other available sources of the kinds of information historians and genealogists seek" and that "Canadians [are] well documented elsewhere in vital statistics ... and through the Social Insurance Number"(Submission, p8). This is misleading for two reasons:
  • Privacy laws are increasingly restricting the amount and kinds of personal information that can be collected, retained, or accessed in both the private and public sector

  • Current record management practices permit the destruction of a large percentage of government records and even more is lost unintentionally
The first reason seems self-evident as almost every government jurisdiction in Canada has put in place strict rules on what personal information in their files can be retained and accessed and provincial governments led by the example of Quebec are now looking at extending such laws into the private sector. As well the federal government currently has legislation in Parliament that will extend the authority of the Privacy Act into the private sector. Genealogists are well aware of this trend as they have seen more and more information sources removed or access restricted. This is not to argue against such laws but merely to point out that they have had a significant impact already on the pursuit of genealogy and no doubt that trend will continue.

The second reason is more complex and needs to be discussed in more detail. I will focus on two concerns: that large amounts of personal information held by government are routinely disposed of and will never be available to future researchers; and that through mishandling and chance much information that might be retained is lost.

Perhaps the one federal government department holding the most personal information on Canadians is Human Resources Development Canada. Here reside the Social Insurance files Mr. Phillips referred to as well as Employment Insurance, Old Age and Canada Pension files, and numerous other records. According to those arguing for closure or destruction of historical census records here lies a great cornucopia of personal information that will supply genealogists with all the information they could ever need. Apparently not. For the issue is not how much personal information is created but how much survives. According to the Department it appears very little will ever make it into the public domain:
    For the most part, client records are destroyed at the end of the authorized retention period. However, R&D authorities usually identify a small sample of client records (i.e. name identified files that contain personal information) for transfer to the National Archives at the end of the retention period for permanent preservation. These files would, therefore, be accessible to researchers in the future. However, there is an Access to Information and Privacy section at the NAC that applies the provisions of the Privacy Act to all access requests. No personal information is provided to the requester. The ATIP section reviews the files and personal information is restricted. (Email from Lynda Morrissey, A/Director Library & Information Mangagement, Human Resources Development Canada, 14 April 2000)
These policies are hardly unique to HRDC but rather prevail throughout federal government departments. It has been suggested for example that genealogists could use tax records in future for their research. This is untrue as Revenue Canada destroys all tax files and they never reach the National Archives. An examination of InfoSource, the catalogue of federal government records, suggests no clear and consistent policy exists for retaining files containing personal information. (Infosource, 1999/2000) Some files are destroyed after only a few months or years while others are held for periods as long as a 100 years but never transferred to the National Archives an example of which is the personnel files of Canada's civil servants present and past:
    Personnel records for Civil servants are retained for eighty years dating from the individual's birth date, after which most of them are destroyed. A few selected personnel files of former Civil Servants dating from 1885 have been retained. (InfoSource 1999-2000)
And a large proportion of those record sets which are targeted for archival preservation are never preserved in total but rather are merely selectively sampled reducing their value to historians and making them of minimal use to genealogists. It is apparent that much of the personal information collected by the federal government will not be preserved for future generations of researchers.

But widespread destruction of personal information held by the federal government has been commonplace in this country for decades. Much of our documentary heritage has already been lost. The deliberate and unauthorized destruction in 1919 of millions of files containing detailed information on Canadian men who registered for the draft during World War One is but one sad example. The destruction of all citizenship records prior to 1917 and the loss of a considerable portion of the records of the Royal Canadian Mounted Police, including service files, are but two more of a very long list.

The planned destruction of records is only half the problem for the chaotic record keeping practices of the federal government are seriously endangering the integrity of current government records and has probably already lead to the loss, deliberate and accidental, of a significant percent of them. But it doesn't end there. An even more serious challenge is that posed by the fragility of electronic data which form the bulk of records kept by government over the last 30 years including archived census forms. These concerns were recently outlined by the Honourable John Reid, Canada's Information Commissioner, who warned that if not remedied one of the cornerstones of our democracy, the ability to access government information, will be imperiled:
    Recent revelations about record-keeping shortcomings at HRDC have alerted all of us to the threat to good government which poor record-keeping poses. But, this is not a new problem. Every day in our work, we find instances where information is not professionally managed: records are not created when they should be; records are not properly indexed, filed or included in departmental record-keeping systems; records are not retained and are not disposed of in accordance with approved schedules. The management challenge posed by electronic records is enormous ... digital records are not proving to be as easy to store and retrieve as we all expected [and] five years or forever--whichever comes first--is considered the effective life of a digital document.

    The whole scheme of the Access to Information Act depends on records being created, properly indexed and filed, readily retrievable, appropriately archived and carefully assessed before destruction. If records about particular subjects are not created, the right of access is meaningless. Similarly, if records cannot be located and produced with ease, the right of access is of scant value. I cannot overstate the point. Information management in government is in crisis. The crisis not only threatens the viability of the right of access, it also threatens to undermine national archival requirements and the ability to deliver good government to the citizenry.(Remarks to Standing Committee on Human Resources Development and the Status of Persons with Disabilities. Ottawa, March 28, 2000)
The loss of digital data is a world-wide problem with serious future consequences as the National Archivist of the United States pointed out in a 1998 speech:
    What worries us and those who sue us alike is this: the Federal Government, along with everybody else in the age of digitized information, is losing unknown amounts of potentially valuable material. How much of the record material accumulating in the computers will survive deletion, deterioration, and the discarding of machinery that can read it? Society is courting disaster for historians will look back on this era and see a period of very little information. A 'digital gap' will span from the beginning of the wide-spread use of the computer until the time we eventually solve this problem. Let me tell you, there is no question about such a gap. The only question is, how big will it get before we learn how to preserve and provide ready access to vast quantities of electronic documents? (U.S. National Archivist, Speech, 2 March 1998)
The facts are that retention of microfilmed census records for future archival use is more important than ever as much other government information supposedly available to genealogists and historians will have been lost whether by plan or accident. It is true, as the Canadian Historical Association remarked to the panel, that "census records could become, by default, even more significant as a source of material on twentieth century Canada". There is no substitute for census records and we cannot afford to lose them. They must be preserved for future public access.


I will not touch on the specific legal issues in any great depth as other submissions will deal with those more exhaustively. Statistics Canada's position will be familiar to panel members. Essentially they view the Statistics Act as the sole authority for all issues relating to confidentiality and access to name-identified census records taken after 1901 and citing a 1985 legal opinion refuse to transfer custody of such records to the National Archives for subsequent public release as stipulated under the terms of the Privacy Act regulations and the National Archives Act. This position is also accepted by the Privacy Commissioner.

As I have no qualifications in law offering opinions on it to the distinguished legal minds on the panel is to say the least daunting. I can only approach the subject with a layman's view that Canada's statutes should be interpreted as speaking exclusively to those matters addressed in their wording and to no other and where they are silent on certain matters to look to other statutes for guidance. Perhaps this is a naive understanding of the law but it reflects a citizen's sense of justice that our laws should be circumscribed and the temptation to broaden their application resisted.

Statistics Canada asserts that the laws in place since 1905 forbid forever the release of name-identified information from all of Canada's censuses taken after 1901 including future census enumerations. Specifically they make two claims:
    1. Disclosure was explicitly forbidden in perpetuity 2. Permanent closure was a promise made to Canadians
I believe neither claim is supported by reference to the facts. I will examine each in order.


I have read all of the Statistics Acts in force since 1905 and most of the enumerator's instructions issued for each census during that same period. Nowhere, including in our present Act, is there any wording respecting confidentiality that expresses the intent of perpetuity. The Act is simply silent on any time limit. This silence is interpreted by Statistics Canada as mandating that census schedules may never be released to the public. But rather I would suggest it is reasonable to assume that statutes mean what they say and no more and had the intention been to specifically deny future access to census forms that the wording would have been found to express it. Statistics Canada does after all insist that this intent was clear and represented a major policy decision of government reversing past practice regarding the confidentiality of census information and was an "explicit guarantee of indefinite confidentiality that was promised to Canadians when the data was collected." (Statistics Canada, By Law, Census Records are not available after 1901) But the various Statistics Acts and their regulations have gone through numerous revisions and amendments since 1905 without inclusion of any wording connoting perpetual closure. If the intent was to promise perpetual secrecy of personal information in census returns I suggest that the Acts and Regulations would have been written in language that clearly and specifically articulated it. To accept Statistics Canada's position the synonym of "to be explicit" must be "to be silent".

I believe the absence of wording expressing perpetuity suggests that the intent of the confidentiality clauses of the various Statistics Acts was to protect personal information only when it is in the custody of Statistics Canada and focuses primarily on the administration of the census process. That is, the actual taking down of the personal information by the enumerators, it's handling and processing by employees of Statistics Canada, and its conversion into statistical, non-identifiable form. This administrative process takes a number of years to complete and while it is underway the confidentiality clause ensures Canadians that their personal information will not be accessed by other than Statistics Canada employees, shared with other government departments or agencies, or distributed in a format that would allow them to be identified. This protection extends of course as long as Statistics Canada has legal custody of the census forms.

But that custody is not expected to be forever and the census schedules like all government records are subject to evaluation by the National Archivist, under authority of the National Archives Act, of their worth as historical documents deserving of permanent preservation and future consultation by the public for legitimate research purposes according to the discretion of the archivist and (since 1983) the terms of the Privacy Act and Regulations. Canada's present archivist and his predecessors have all recognized the value of these records to our country and mandated their preservation. If the Statistics Acts intended perpetual closure of census records why was the National Archives Act not amended to exempt them from the authority of the archivist? And why, if census records were permanently closed to any access as early as 1905, would the Privacy Act regulations written in 1983 specifically permit public access to name-identified census records after 92 years without excluding post-1901 censuses? Further, this argument leads to the inevitable and illogical conclusion that the government would allow the National Archives to mandate the permanent preservation of such records without ever allowing anyone to view them and to expect our National Archives or Statistics Canada to store indefinitely a large and ever growing collection of documents taking up significant amounts of scarce and costly storage space for which Statistics Canada has no administrative use and which the National Archives is forbidden to make available to researchers. Statistics Canada's claim of perpetual closure defies both a reasonable reading of the law and common sense.

To accept Statistics Canada's position is to believe we arrived at this point by a remarkable comedy of errors committed by Parliament, civil servants, and legal staff from several departments of government when the National Archives and Privacy Acts were written and enacted: legislation was drafted, debated in Parliament by members of both Houses, studied in committee, passed into law, regulations were written, administrative procedures put in place and followed, all in contradiction to the Statistics Act and doomed to be set aside yet put into place apparently without comment from Statistics Canada.

And the assertion that nominative census records were closed in perpetuity as early as 1905 is contradicted by a reading of the enumerator's instructions issued by the Governor-in-Council in 1906 and which carried the force of law. Instruction number 33 reads:
    33. Clear and legible records. The enumerator is required to make all entries on the schedules in ink of good quality, and every name, word, figure or mark should be clear and legible. If a schedule cannot be read, or if the entries are made with a poor quality of ink, or in pencil, or if they are blurred or blotted, the work of the enumerator may be wholly wasted. The Census is intended to be a permanent record, and its schedules will be stored in the Archives of the Dominion. See Instruction No. 52
The same or similar instructions which identify census schedules as permanent records to be archived for future reference appear in the Orders-in-Council printed in the Canada Gazette for censuses from 1906 to as late as 1946. As the sole mandate of our National Archives is to preserve and make available to researchers Canada's documentary heritage the intent of this instruction is clear - census records are to be preserved and made available to future generations not closed in perpetuity.

The inclusion of Instruction #33 or its equivalent in the Orders-in-Council for many of the post-1901 censuses speaks clearly to one issue which Statistics Canada ignores: that from the very beginning it was acknowledged that the promise of confidentiality was not forever and that public access was sanctioned at some point in time not through the Statistics Act but rather through other more appropriate statutes addressing the preservation of government records and the protection of personal information contained in those records. Today those statutes are the National Archives Act and the Privacy Act. This is a long-held practice of most western countries including the United States and Great Britain where preservation and access to census records is addressed through statutes separate from their respective census laws. It may be useful to look at the acts regulating census confidentiality in those two countries.

In the United States and Great Britain the laws equivalent to ours that govern census confidentiality are remarkably similar in phrasing to our Statistics Act, including clauses that like our Section 17 contain no mention of time limits or archival preservation, that forbid access to name-identified census forms by anyone but census employees, that limit use of census information except for statistical purposes, and that prohibit the release of data that could identify a particular person. I will address the American statute first.

Title 13 (Census) of the United States Code contains in Section 9 a clause respecting census confidentiality that mirrors closely Section 17 of our current Statistics Act. The relevant clauses read:
    Protection of Confidential Information -- Sections 9 and 214 of Title 13 of the United States Code

    Sec. 9. Information as confidential; exception

    (a) Neither the Secretary, nor any other officer or employee of the Department of Commerce or bureau or agency thereof, or local government census liaison may, except as provided in section 8 or 16 or chapter 10 of this title or section 210 of the Departments of Commerce, Justice, and State, the Judiciary, and Related Agencies Appropriations Act, 1998.(1)
    (1) use the information furnished under the provisions of this title for any purpose other than the statistical purposes for which it is supplied; or
    (2) make any publication whereby the data furnished by any particular establishment or individual under this title can be identified; or
    (3) permit anyone other than the sworn officers and employees of the Department or bureau or agency thereof to examine the individual reports. No department, bureau, agency, officer, or employee of the Government, except the Secretary in carrying out the purposes of this title, shall require, for any reason, copies of census reports which have been retained by any such establishment or individual. Copies of census reports which have been so retained shall be immune from legal process, and shall not, without the consent of the individual or establishment concerned, be admitted as evidence or used for any purpose in any action, suit, or other judicial or administrative proceeding.

    (b) The provisions of subsection (a) of this section relating to the confidential treatment of data for particular individuals and establishments, shall not apply to the censuses of governments provided for by subchapter III of chapter 5 of this title, nor to interim current data provided for by subchapter IV of chapter 5 of this title as to the subjects covered by censuses of governments, with respect to any information obtained therefor that is compiled from, or customarily provided in, public records.

    Sec. 214. Wrongful disclosure of information

    Whoever, being or having been an employee or staff member referred to in subchapter II of chapter 1 of this title, having taken and subscribed the oath of office, or having sworn to observe the limitations imposed by section 9 of this title, or whoever, being or having been a census liaison within the meaning of section 16(2) of this title, publishes or communicates any information, the disclosure of which is prohibited under the provisions of section 9 of this title, and which comes into his possession by reason of his being employed (or otherwise providing services) under the provisions of this title, shall be fined not more than $5,000 or imprisoned not more than 5 years, or both. Viewpoint
This statute is so similar to our legislation that from Statistics Canada's viewpoint this law can be read only one way: access to name-identified information from United States census forms is forbidden in perpetuity. They would be wrong. American census schedules are in fact released to public access after 72 years and have been for decades. The 1920 census was released in 1992 and the 1930 census will be open to all Americans on the 1st of April 2002.

The United States, despite the wording of Title 13, does not recognize that Statute as having authority over the disposal or future access to name-identified census records. Instead, that authority is exercised through the appropriate sections of Title 44 of the United States Code. Future access to personal schedules of American censuses is governed by Section 2108 which reads:


    Sec. 2108. Responsibility for custody, use, and withdrawal of records

    b) With regard to the census and survey records of the Bureau of the Census containing data identifying individuals enumerated in population censuses, any release pursuant to this section of such identifying information contained in such records shall be made by the Archivist pursuant to the specifications and agreements set forth in the exchange of correspondence on or about the date of October 10, 1952, between the Director of the Bureau of the Census and the Archivist of the United States, together with all amendments thereto, now or hereafter entered into between the Director of the Bureau of the Census and the Archivist of the United States. Such amendments, if any, shall be published in the Register.
The agreement referred to formalized an earlier arrangement establishing the 72 year rule for public release of American census schedules. This agreement was upheld by a ruling of the United States Department of Justice and applies to release of all future U.S. nominal censuses.

The British Law governing census confidentiality is the Census Confidentiality Act (1991). The relevant sections of the law reads as follows:
    Census (Confidentiality) Act 1991: Chapter 6

    An Act to make provision with respect to unlawful disclosure of information acquired in connection with the discharge of functions under the Census Act (1920); and for connected purposes.

    (2) If the Registrar-General for England and Wales or the Registrar-General for Scotland ("the Registrars") or any person who is:

    (a) under the control of either of the Registrars; or

    (b) a supplier of any services to either of them.

    discloses any personal census information to another person without lawful authority, he shall be guilty of an offence

    (3) If any person discloses to another person any personal census information which he knows has been disclosed in contravention of this Act, he shall be guilty of an offence.

    (5) A person guilty of an offence under subsection (2) or (3) shall be liable:

    (a) on summary conviction, to imprisonment for a term not exceeding six months or to a fine not exceeding the statutory maximum or both;

    (b) on conviction on indictment, to imprisonment for a term not exceeding two years or to a fine or to both.

    (7) In this section:

    "census information" means any information which is:

    (i) acquired by any person mentioned in subsection 2 above in the course of any work done by him in connection with the discharge of functions under section 2 or 4 of this Act;

    (ii) acquired by any such person in the course of working, for purposes of section 5 of this Act, with any information acquired as mentioned in sub-paragraph (i) above; or

    (iii) derived from any information so acquired

    "personal census information" means any census information which relates to an identifiable person or household."
Again, as in Canada and the United States, this law forbids release of census information identifying an individual, mentions no time limits to confidentiality, and is silent on preservation of census forms. The conclusion drawn by Statistics Canada must then be that release of census information in Britain is expressly and perpetually forbidden. Once more they would be wrong. For the British release their censuses 100 years after they are taken by the sole authority of the Public Records Act (1958), not the Census Act (1920) nor the Census (Confidentiality) Act 1991, as explained by the Public Record Office:
    The confidentiality of any individual's census information is an issue which the Census Office takes very seriously indeed. It is unlawful, for example, for the Census Office to pass any census information to other Government departments or any other organization except for the purposes of the Census Act itself or the Public Records Act 1958. Under this latter legislation, the Lord Chancellor has made an instrument [Instrument # 12] closing census returns to public inspection for 100 years. (
Statistics Canada's position presumes the Statistics Act is the only statute with authority over census records. While that may be true as long as the agency has custody and control of these records the idea that such custody extends until the end of time is absurd. Parliament passed the National Archives Act (1985) and the Privacy Act (1983) years after the Statistics Act (1971) came into force and fully understood that its provisions would be superceded upon transfer of these records to the National Archives. By refusing to turn over custody of post-1901 census schedules Statistics Canada is frustrating the intent and authority of the law and Parliament. This cannot continue or Statistics Canada risks losing the respect and cooperation of the citizenry on which they are utterly dependent during census enumerations.


According to Statistics Canada and the Privacy Commissioner the closure of future public access to nominal census forms was a promise of confidentiality in perpetuity or an explicit guarantee of indefinite confidentiality. This is sometimes referred to as the "Laurier promise" after our former Prime Minister who was leading the government at the time the 1905 Statistics Act was passed and the 1906 census was conducted and suggests that the government of the time made a political covenant with Canadians to bar forever future access to all nominal censuses. The Laurier promise is a fiction. No such promise was ever made by Sir Wilfred or his Ministers and no such promise is inherent in the legislation. This supposed promise is apparent only to those who would wish it into existence. I would like to examine it in detail.

On learning of this supposed promise by the government of Sir Wilfred Laurier our group decided to search it out. As logic demands that a promise to the public must necessarily be a public promise we set about looking for evidence of it. Of course the most obvious and appropriate place for a political leader to make a promise to the people is in Parliament so the debates of the House of Commons and Senate were read during the periods when the 1905 Census and Statistics Act, the 1906 Census and Statistics Act, and the 1918 Statistics Act were being considered. In both the House of Commons and Senate not a single reference respecting confidentiality of personal information provided by individuals was found in the 1905 or 1906 debates and no reference was located in the debates that the secrecy clause in the 1918 Statistics Act was to express perpetuity.

Noting Statistics Canada's statement that an "explicit guarantee of indefinite confidentiality was promised to Canadians when the data was collected" we assumed that this promise, by whomever and whenever made, must then have been conveyed to Canadians by Statistics Canada during the enumeration period of each census. Members of our group read all of the enumerator's instructions for the 1901 to the 1946 censuses, every population schedule for every post-1901 census, and all the instruction booklets for the self-enumerated censuses of 1971 through 1996 without finding a single occurrence of any statement that the confidentiality of census records would last forever. In fact it appears for the earlier censuses that there was direction to the enumerators not to inform citizens their information would be confidential at all unless they specifically asked the census taker.

Having exhausted the only legitimate methods by which government could have informed Canadians of this perpetual promise we were now left with only the unofficial forum of Canada's newspapers and began reading, page by page, rural and urban newspapers for the appropriate time periods. Focusing on the periods before, during, and after the 1906 and 1911 Census es were conducted, times when the census would be expected to interest both the press and the public at large, daily and weekly newspapers across Canada were examined. We looked for any story that addressed concerns about confidentiality of census records during the enumeration of both censuses and particularly for any expression that name-identified census information was to be closed to access in perpetuity. As the 1911 Census touched all Canadians and was thus of national interest newspapers across the country were consulted, while newspapers in the three prairie provinces were consulted for the period of the first quinquennial census in the summer of 1906.

The 1906 census was the first taken after the new census Act was passed in 1905. This is the law that supposedly reversed decades of past practice and introduced the new strengthened privacy and "perpetual closure" rules. You would think Statistics Canada would have been anxious to spread the good news to the citizenry but a reading of some western newspapers of that period suggests otherwise.

The Lethbridge Herald of June and July 1906 (the census was taken June 24th, 1906) contains only one mention of the census: "The enumerators begin taking the census on Monday. It will take some weeks to gather all the returns." The Regina Standard considered the census of such import that they refused to mention it at all while the Saskatoon Daily Phoenix informed its readers that a census was to take place and then promptly dropped the subject altogether and the Farmer's Advocate and Home Journal a national farm newspaper with a large circulation in western Canada devoted an entire six words to the 1906 census, "the census is now being taken". The near silence of the press is difficult to square with Statistics Canada's claim that Canadians were informed that an "explicit guarantee of indefinite confidentiality" now applied to name-identified census schedules.

To date no report appearing in the newspapers examined has stipulated that census confidentiality would be in effect forever nor have any advertisements by Statistics Canada explaining the census procedures and rules been found. The general impression from the newspapers is that the census was not a story of great interest to the readers but rather a routine matter. Most 1911 newspapers carried at least a few stories on the census but they were generally not featured prominently. Typically one mention would be made a few days previous to the census date that it was imminent and another on the day it began. During the 1911 Census period occasional stories would appear usually along the following themes:
  • speculation on the increase of population for the community or province over the last census
  • how and why certain individuals or groups were or were not co-operating fully with the census officials and usually including comments on the value of the census to the community
  • that the enumerators were sworn not to divulge information to others including neighbours or share it with other government departments, and that citizens were legally required to answer the questions
  • and finally, mention of the successful completion of the census and when the official counts would be released.
In other words, pick up a paper from May of 1996 or no doubt May 2001 and you will have a good idea how the census was perceived and reported in the press early in this century. Little seems to have changed in nearly 100 years.

Stories dealing with census confidentiality were of course of primary interest to us and were carefully read. Not one such story I have seen to date stated that confidentiality would last indefinitely. Without exception every story focused on the possible current uses of personal information from the census by government: concerns of taxation, school attendance, fears of deportation or conscription, and so on. Perhaps a few quotes from newspapers of 1911 might interest the panel. For example, all Vancouver newspapers were examined from the 1st of May to the 30th of June of 1911 looking for items about the 1911 Census . The ones summarized below are representative:
    Vancouver Province, 31 May 1911, page 17

    Census Enumerators Ready To Start: Enumeration begins at 9 o'clock, enumerators have all signed pledge. "Everything will be done in absolute secrecy, the enumerators will not know anything regarding the facts obtained from residents. All the data will be forwarded to the government". Work completed in two or three weeks, enumerators paid five cents for every name and translators get five dollars per day.

    Vancouver Daily World, 2 June 1911, page 2

    Chinese Give Census Takers Flat Refusal, Little Trouble Experienced Among White Population: Chinatown residents and merchants refuse to answer questions until Chinese consul-general intervenes, "fear being deported"; Irate housewife refuses to answer questions, gives enumerator "a bitter tongue lashing for his inquisitiveness", may be prosecuted. Good progress made, most citizens prepared for census taker, "apart from the Chinese incident we have had little trouble".

    New Westminster British Columbian, 21 June 1911, p.1

    No great difficulties encountered. Some people were averse to parting with information of a private nature until explained to them that the enumerators, the Commissioner, and the officials in Ottawa were all sworn to secrecy, and that court proceedings could be taken for "conscientious objectors" which had the desired effect in all cases.
Further examples are taken from newspapers in other parts of Canada:
    Manitoba Free Press, Wednesday, May 31, p.5

    Census Coming. Census is not taken "for vainglorious purposes", nor for military purposes, nor to increase taxation.

    Ottawa Evening Journal, Wednesday, May 31, p.5

    Canada's Census Day is tomorrow. "The hardest proposition that we are up against", remarked a [census] official, "is to get information from these foreign people. Fleeing, many of them, from a land of oppression, they have a haunting suspicion that we want to bleed them with taxes, or make conscript soldiers of them, or something of the kind".

    Ottawa Evening Journal, Wednesday, May 31, p.8

    Unlucky jail birds. "Those sensitive individuals who do not wish to have their names preserved in the archives for a century as having been an inmate of jail, prison, etc. on June 1st.1911, will be well advised to postpone any intended escapade".

    Toronto Globe, Friday, June 2, p.9

    Many people take a wrong view of Census men. "A false impression seems to exist as to the object the Government has in view. This is especially true in the district inhabited by people of foreign extraction. The information obtained is private and will not be used for any of the reasons anticipated by persons unfamiliar with the English language or customs". One Italian man fears that if he admits to the full size of his family it might result in an increase in his water rates. There was a particular reluctance on the part of local businessmen to give information concerning their salaries or insurance coverage. "The Census Commissioners wish to express through the newspapers that there need be no reluctance to answer questions in this respect, or in regard to the size of families etc., as no municipality can get any of the information secured in the Federal Census for taxation or any other purpose."

    Toronto Globe, Monday June 5, p.8

    Census men in Ward grow grey with worry. Summonses to be issued if foreigners do not give information desired. The enumerators in "The Ward" got a poor reception in this district, which contains a lot of new immigrants. One lady said that there were no knives or revolvers in the house, and not a single keg of beer was concealed! Another person was afraid of conscription. It is known that there are many illegal immigrants living in that area, and they would probably be afraid to be enumerated.
The theme here is that yes, for at least a small part of the population, there was then, is today, and will probably remain in future, some concern about privacy and data confidentiality during census enumerations. But that concern was not in 1906 and 1911, nor 1996, and will not be in 2001 or likely 2096 a fear about release of censuses 92 years in the future but rather present-day concerns about how this current personal information might be accessed and utilized by other departments of government or in today's environment by private companies. Early this century the fears focused on taxation of property and income, conscription, deportation, and a suspicion of government among immigrant groups from countries where authorities were people rightfully feared. A paid advertisement our statistical agency placed in newspapers across the country in May 1931 makes that clear:
    No person can be harmed in any way by furnishing the information required. The census has nothing to do with taxation, with military service, with the compulsion of school attendance, with immigration, or the enforcement of any Dominion, Provincial, or Municipal law or regulation. (Seventh Census of Canada, 1931, Volume One, p.53)
The evidence is overwhelming that no promise of perpetual secrecy of census records was ever made to Canadians. It is not in the law itself, it was never communicated to citizens by the enumerators, nor was it ever conveyed unofficially through the agency of Canada's newspapers. It is a fiction.

And finally, as a Canadian adult head-of-household I have willingly participated in every census since 1971. I can tell the panel categorically that Statistics Canada has never informed me that my personal information would be held confidential in perpetuity. I have been a genealogist since 1978 and aware since before that date that Canada's census records had been released to public access and it was always my expectation that they would continue to be and that my personal information would in due course be accessible to others. I neither fear that occurrence nor has knowledge of this eventual disclosure affected my willingness to participate in a census enumeration nor would destruction or permanent closure of census records in future encourage me to provide accurate answers to census questions. In fact, as a genealogist who believes the release of historical census records for research purposes is an important and legitimate use of the census, destruction of census forms would discourage, not encourage my participation.


The conjecture that release of historical name-identified census records creates concerns about privacy and data confidentiality and lessens the public's willingness to respond to census surveys has never been established in Canada or elsewhere. When the Chief Statistician spoke to the Panel in January he commented that "Statistics Canada is very concerned that response rates and the quality of respondent's would decline if the absolute guarantee of confidentiality was to be weakened".(Speaking Notes, Mr. Fellegi, p.9) In his brief the Privacy Commissioner suggests such a connection and implies that this is shown by comparing American and Canadian census response rates:
    [the] guarantee of the confidentiality of the process is seen by its authors [Statistics Canada] as essential to obtaining a satisfactory level of response. The response rate is extremely good. In 1991 and 1996, more than 97% of census respondents complied when summoned to answer the intrusive questions sent to them by Statistics Canada. This compares favourably with the U.S., where the response rate for the last census, in 1990, was only 63 percent.(Submission of Privacy Commissioner, page 5)
In other words, Mr. Phillips is intimating that Americans in significant numbers are not responding to American census enumerations over concerns that the United States releases name-identified census schedules to public access after 72 years and, further, that Canadians are significantly more cooperative with our census takers because Statistics Canada forbids such access. He offers no proof for either claim and bases it on an inaccurate and misleading comparison.

First, the figures of 63 and 97 percent do not measure the same thing and therefore cannot be compared. And second there is no evidence from the United States that data confidentiality or privacy concerns are a primary cause of lower census response rates there let alone specific concern with release of historical census records. I would like to address this in some detail.

The high figure of 97 percent is incorrectly identified by the Privacy Commissioner as Canada's response rate when it actually represents the Coverage Rate which is the percentage of the estimated population of Canada on May 14th,1996 that was counted by Statistics Canada. The actual figure was 97.4 percent and the concomitant Undercoverage Rate of 2.6% represents the number of Canadians Statistics Canada believes were missed during the census enumeration. The properly comparable Coverage Rate number for the United States is not 63 percent but 98.4 percent, higher than the Statistics Canada figure. So where does the 63% figure originate?

The figure of 63 percent used by Mr. Phillips is not correct. The true number is 65 percent ( and represents, in the terminology of the United States Census Bureau, the Mail Return Rate and "relates to housing unit addresses and is defined as the proportion of addresses (including vacant) which return census forms" while the Mail Response Rate "relates to households (occupied housing units) and is defined as the proportion of households in the 1990 Mail universe which completed their 1990 census form without the aid of an enumerator." (Who Responds/Who Doesn't, David Word, U.S. Bureau of the Census). Obviously vacant dwellings not returning census forms is irrelevant to this discussion and thus the 65 percent rate cannot be used to examine the willingness of Americans to fill out census forms for whatever reason. It is only the Mail Response Rate which has any relevance and that U.S figure is actually not 63 or 65 percent but 74.1 percent and the comparable number for Canada is 85.5 percent:
    In the period allotted for respondent returned questionnaires for the 1996 Census, prior to the commencement of the field follow-up exercise used to complete the collection phase, 85.5% of households had either mailed their information back to Statistics Canada or provided it to us in some other fashion (i.e. via the Census Help Line). The corresponding return rate for the Census of Agriculture was 77.3%. (Email from Pamela White, Statistics Canada, 4 April 2000)
The difference in Mail Response Rates between the two countries then is not the 34% suggested by Mr. Phillips but rather 11.4 percent. It should also be noted that failing to return a completed census form by mail before the designated due date does not mean that a household will fail to provide a completed form to the statistical agency, only that the form was not returned by the designated day, or was returned when due but contained incomplete or possibly inaccurate information requiring follow-up contact by a census enumerator to complete it. This may have been as significant as having no form returned or as mundane as a person's stated age being in conflict with their birth year or as commonplace as one or two questions left unanswered. Certainly nonresponse as defined by the statistical agencies does not equal refusal nor necessarily even reticence to co-operate with census enumeration.

No doubt concerns about privacy and data confidentiality must play at least some role in census nonresponse. So, in the context of the definitions of nonresponse, what are the factors that affect a household's response rate and how large a role do privacy and confidentiality concerns play in it? And is there any specific evidence that releasing historical census records to public access impacts negatively on response rates? There is a paucity of Canadian studies relating to these issues from either academics or Statistics Canada. In the United States however both social scientists and the United States Census Bureau have thoroughly examined the problem of nonresponse to census surveys with special attention to the 1990 census.

Numerous American studies have examined the motivations or causes of census nonresponse. The following is only a partial list of factors found to negatively impact response rates for the 1990 U.S. decennial census:
  • 1 race and ethnicity
  • 2 dwelling type and tenure
  • 3 family composition
  • 4 citizenship status
  • 5 involvement in criminal behaviour
  • 6 volume and handling of household mail
  • 7 increases in the number of American households
  • 8 literacy and English competency
  • 9 age
  • 10 poverty
  • 11 lack of knowledge about the census
  • 12 poor design and wording of questionnaires
  • 13 errors by census employees
  • 14 an increasingly mobile population
Research conducted following the 1980 Census found two other important factors contributed to low response: failure to receive the form and never starting to fill it out. These reasons continued to hold true for 1990:
    As the results of several independent public opinion polls indicated, failure to return the Questionnaire was due not to a philosophical objection to the Census, but simply that people were too busy to do so.
Many of these factors are easily understandable and only some of them could be related to deliberate nonresponse to enumeration. Over 18% of Americans change addresses every year and some will inevitably receive but not return their census forms, while an inability to read or write in English, a household composed of unrelated individuals, illness, social alienation resulting from poverty, illiteracy, census questionnaires difficult to read or understand, assumptions by some that not returning the form by the stated date (Census Day) meant it was too late to turn it in, or simple time pressures as people cope with work and family responsibilities, are all easily understood as reasons for missing a deadline for returning census forms. Other factors point to intentional noncompliance. In 1990 there were an estimated 3.3 million undocumented aliens living illegally in the United States and over one million criminals and fugitives from justice few of whom would wish to fill out forms for government officials. Remove the tiny proportion of citizens who resist census enumeration based on philosophical or political motivations and we have catalogued the largest proportion of the nonrespondent population, a group whose failure to respond has it appears little to do with general concerns about privacy or data confidentiality and nothing to do with release of historical census records.

Many of these factors of course apply to our country as well but lower levels of criminality, illegal immigration, poverty, and racial inequality, plus higher rates of literacy and greater acceptance and experience of government involvement in the lives of citizens, when combined with Statistics Canada's practices of sending an enumerator to personally deliver each census form and of prosecuting those who don't co-operate, a strategy abandoned by the Americans, would easily account for most of that 11.4 percent difference in the Mail Response Rate between the two countries. Perhaps it is instructive here that Statistics Canada's greatest admitted failure in the 1991 and 1996 censuses was their inability to properly enumerate the part of our population where inequality, illiteracy, racial tension, and poverty are most prevalent - among Canada's First Nations people. (Statistics Canada, Coverage Reports, 1991 and 1996 Censuses)

Concerns of privacy and confidentiality are then but one part of the bigger picture of what leads to census nonresponse. But how big a part? This question was investigated in depth by the United States Census Bureau by means of a study which focused on the effects of those concerns on the Mail Response Rate to the 1990 decennial census. The authors first summarized the results of earlier studies examining the connection between privacy concerns and survey nonresponse:
    It is an article of faith among survey researchers that people will not respond to surveys, or at any rate not respond honestly, unless they are assured of the confidentiality of their replies. The evidence for this belief is at best modest. While some early field experiments suggest modest effects on either survey response rates or response quality, or both, we know little about the mechanisms by which these effects come about. In fact, recent experiments suggests that under some circumstances, an elaborate assurance of confidentiality may be counterproductive, increasing respondents concerns and their reluctance to respond. (Singer, Impact of Privacy and Confidentiality Concerns on Survey Participation: The Case of the 1990 Census)
The purpose of the study was to examine the assumption "that concern about the confidentiality of the data reduces the likelihood of response, and that an assurance of confidentiality reduces this concern". Although some survey participants "identified concerns about privacy and confidentiality as factors in self-reported participation" the study concluded that "the effect of privacy and confidentiality is not large". (Singer, Privacy and the 1990 U.S. Census)

The Australian Bureau of the Census uses a Drop-off/Pickup method for census data collection in which enumerators hand deliver each census schedule and arrange a pick-up appointment day and time with each householder. No envelope is provided meaning that when picked up the personal information is not concealed from the enumerator. If a householder specifically requests it the enumerator will provide a "Gold Privacy envelope" allowing the information to be concealed at pick-up. Mail return is not permitted with two exceptions:
  • where multiple attempts have failed to contact anyone at a dwelling, or;
  • only if it will convert/avoid a potential Refusal situation and only after the householder has been offered a Gold Privacy envelope
Only three percent of Australia's 1991 census forms were marked for the mailback return method but as the Australian Bureau of the Census has a low regard for this method of data collection from concerns it may "have a significant and detrimental impact on data quality" a study was undertaken to examine its impact on the 1991 census. The findings are relevant to this discussion.

First, the authors presumed that the use of the mailback method likely indicated those households both least likely to cooperate in enumeration and with the highest degree of concerns about privacy and data confidentiality. A not unreasonable assumption. Yet the study found that the Mail Response Rate for these forms returned by mail though below the Leave/Pickup rate employed by Australian census takers was still 70 percent. This return rate was "approximately in line with return rates for overseas (mailback) Censuses" and the study noted that the "return rate before follow-up (but after an initial request) in the US in 1990 was only 65% [and] that in Canada in 1986 intensive follow-up was needed to achieve an 87% return rate." In other words if Australia despite its vaunted policy of maximizing privacy through census destruction had used the mailback return method for their 1991 census the Mail Response Rate would be expected to drop into the range found in Canada and the United States, a drop that would be difficult to explain as public reaction to census privacy concerns in general and that would have no possible connection to release of historical census records as the Australian practice of destruction applied to 1991 schedules. (ABS.1991 Census Data Quality: A Study of Posted-in Forms)

The Australian study also analyzed response rates for individual questions focusing on those thought to relate most strongly to privacy concerns of which the most significant was judged to be Question number two- Name. The study concluded:
    Households returning Census forms by mail were largely co-operative. Overall non-response to 'Name' (4.1%) again appears to be quite low. While non-response to Question 2 ('Name') is apparently higher for posted-in forms, 95% or more of records on these [mailed in] forms still contain names. This may indicate that privacy concerns of people mailing their Census form were not as important an issue as previously thought.
This tells us only one thing: that Australian Mailback Response Rates had much to do with the mechanics of data collection methods, little to do with privacy or confidentiality concerns, and nothing to do with historical census release.

So if even the wider concerns of privacy and data confidentiality appear to have but only a small effect on census nonresponse what can be said about the main issue addressed here: the effect of the release of historical censuses into the public domain after the passage of long periods of time, namely 92 years in Canada and 72 years in the United States, on people's willingness to participate in census enumerations. Apparently there is no known effect at all. In my survey of the literature concerning these issues I have not found a single article from social scientists or the research branches of the statistical agencies of Canada, the United States, Australia, or Great Britain that identified this as a subject of interest let alone concern nor has the Privacy Commissioner or Statistics Canada been able to provide any proof beyond wishful conjecture that such a relationship exists.

The subject of census nonresponse has been extensively researched by the U.S. Census Bureau and has been a high-profile matter to both American politicians and census officials since the 1990 decennial census. This concern is not as some might suppose driven by alarm that Americans are refusing to participate in their census but rather by the reality that the Mailout/Mailback census method is the cheapest to administer. Enumerating those who don't respond by the designated date is labour intensive and expensive but it does get the job done: the United States Census Bureau was more effective in enumerating Americans in 1990 than was Statistics Canada in enumerating Canadians in 1996.

If any connection had been discovered between nonresponse rates in the United States and release of historical censuses it would have been made a public issue. It has not. And the recent extension of the 1952 agreement (the 72 year rule) between the U.S. Census Bureau and the National Archives to include Survey records held by the Bureau as well as the decennial censuses is proof that such concerns have not arisen. (Email from Craig Best, U.S. Census Bureau, April 17, 2000) It appears the current view of the United States Census Bureau towards release of historical census records has not changed since a Bureau representative testified before a Committee of the United States House of Representatives in 1978:
    We support the permanent preservation of population census records for legitimate future uses. The Department of Commerce is keenly aware of the research interests in archival census records. In terms of volume of activity, those interests are primarily genealogical. (Committee on Government Operations, House of Representatives, Washington, 10 July 1978)
The final report of the Congressional Committee itself recognized the importance of permitting access to America's historical census records:
    Because documentation contained in census records is invaluable to the Nation and our society, we feel very strongly that the census records should be open to the citizens of this country after the passage of a reasonable period of time. Researchers who use census records come from every part of the United States and represent a wide range of scholarly interests. Researchers are continuously seeking answers to new and complex questions which were not even considered when the census occurred. Their work contributes significantly to a better understanding of the development of American society and of the United States as a country. (Committee, ibid, p.6)
And the Committee was unequivocal in their rejection of the notion that releasing historical census records damages census participation:
    To the best of our knowledge, this concern has never been demonstrated. No evidence has ever been presented that knowledge of the eventual availability of census schedules, a fact which has been well publicized by archival, historical, genealogical, and religious organizations for over 20 years, affects the willingness of citizens to cooperate and furnish information. Since 1941, over one-half million rolls of census microfilm have been sold to universities, city libraries, historical societies, state archives and libraries, and individuals, making knowledge of the availability of census records and their use very widespread. Despite this widespread knowledge of their availability, the National Archives has never received a complaint about census records being open for research. To the extent that distrust of government is a hindrance to census takers, it is the fear that the Census Bureau will disclose the information to other government authorities (e.g. the IRS or FBI) now, not that the information will be disclosed to historical and genealogical researchers in the distant future.
If the supposition that release of historical census records harms participation rates is a theory then where are the facts supporting it? There is however ample evidence to refute it for we have decades of experience (not speculation) from countries around the world who release name-identified census records into the public domain after varying periods of time. The experience of our own country, Great Britain, and the United States alone speak volumes.

Between 1917 and 1993 six Canadian censuses have been transferred to the National Archives and opened to public access after closure periods ranging from 57 (the 1861 census) to 104 (1881 census) years. In fact three censuses were released within the short period of 8 years from 1985 to 1993 under the new Privacy Act. If Canadians were concerned their privacy was being violated by this action, as privacy advocates and census officials claim, then you would expect to see that concern reflected in their response to the censuses that followed this release, 1986 through 1996. No such concern was registered according to Statistics Canada as "the return rate over the past several Censuses has remained fairly constant, in the range of 85%. There has been no significant decline in this rate over this four census period [1981 through 1996]." (Email from Pamela White, Statistics Canada, 14 April 2000)

These six nominal censuses (1851-1901) exposed the personal information of the 24 million Canadians enumerated in them for periods as long as 82 years. Over an equivalent period British censuses from 1851 through 1891 have been opened after a 100 year closure releasing the personal details of 116 million Britons. And the United States has after a closure of only 72 years released their nominal censuses of 1850 through 1920 containing information on 481 million Americans. A grand total of 620 million individuals between the three countries. And the result? Not a single demonstrated incident of harm to persons living or dead and not a single complaint received by the national record keepers, census offices, or privacy officials of any of the three countries, not one. The Privacy Commissioner's and Chief Statistician's theory has been put to the test 620 million times and found wanting on every occasion. Releasing historical census records after a reasonable period of time has no adverse impact on census participation and harms neither the living or the dead.


Mr. Phillips' opinions on access to historical censuses will be well-known to the panel, the government, and the public through his submission to the panel, annual reports, and of course his numerous speeches and media interviews. Mr. Phillips accepts the position of Statistics Canada that the Statistics Act forbids forever any release of personal information from census records, including those older than 92 years. He opposes release to the public for research purposes of Canada's historical census records and advocates the destruction of census records on the grounds that they violate the privacy interests of those enumerated and suggests that genealogists and historians will have access to other records that will be effective substitutes for census schedules. He also proposes that if census records are to be retained future censuses include a check box where Canadians can choose whether their personal information will be archived or destroyed. And further, he has stated his belief that the dead as well as the living possess the right to privacy and that his opposition represents support of that principle as well. Finally, he suggests that censuses have become increasingly intrusive over time. I will address each of these issues raised by Mr. Phillips in order.


First, the destruction of census records is not in the purview of Statistics Canada or the Privacy Commissioner, but is the sole and proper authority of the National Archivist as granted under the National Archives Act. The National Archives has recognized Canada's census records as a national treasure which must be preserved so future generations of Canadians may discover and better understand the history of our country and people. Our recently appointed National Archivist, Mr. Ian Wilson, has reiterated this view by placing a moratorium on the destruction of any census records and by mandating the preservation of the 2001 census. This long-standing policy is opposed by our Privacy Commissioner who points to Australia where, unique among western nations, they have destroyed census records for nearly 100 years. He suggests that this is the road to take in future arguing that destruction provides perfect privacy for census respondents. Should Canadians take Mr. Phillips advice and implement the "Australian solution"? That question can best be answered by looking to recent developments in Australia.

In 1997 the issues of preservation and public access to census records was placed before a Committee of the Australian House of Representatives. Over a period of one year the Committee travelled throughout the country in an exhaustive investigation of the issue, hearing from all sides of the debate including government, interested private groups, and the public at large. In 1998 a final report was issued which concluded:
    In drawing together all the arguments about data quality, privacy, confidentiality and research value, the Committee considers that saving name-identified census records rather than destroying them would make a significant contribution towards preserving Australia's history. Accordingly, the Committee has concluded that name-identified information contained in census forms should be retained. (House of Representatives Standing Committee on Legal and Constitutional Affairs. Committee Report, Saving Our Census, Section 7.34)
In response to this report the Australian government is now drafting legislation that will, for the first time in nearly 100 years, allow for the retention and public access after 99 years of Australia's name-identified census records beginning in the year 2001. By destroying census records, or by prolonging our current impasse, Canada would be following a path now beginning to be abandoned by the only western country to practice or even advocate such a policy.

Certainly many other western nations neither destroy nor refuse access to their historical census records. The European Union recognizes genealogical and historical research as legitimate uses of census records that do not violate the principles of privacy protection. In Great Britain name-identified census records are released into the public domain 100 years after they are taken and have been for many years. Indeed the 1901 British census will not merely be released, it is now being scanned into digital format and will be placed in its entirety on the Internet accessible to everyone world-wide with an interest in it. This is being done in part to meet what the British government sees as an important commitment to its citizens to make government information more accessible to the population and to encourage interest in genealogy and local history so Britons may gain a deeper understanding of who they are as a people. And in the United States name-identified census records have been released for decades after a closure period of only 72 years, considerably less than the 92 period mandated by our Privacy Act regulations. Canada, in deciding the fate of its census records, cannot ignore the practice of other western nations who are our chief allies and trading partners in a world where international standards and agreements are increasingly dictating or modifying in a growing number of areas the practices of national governments.

And the denial of legitimate rights and the prevention of redress for past government and individual actions are inevitable results of destroying government records of all kinds including census records. Our Privacy Commissioner is strangely silent on this serious threat to citizen's rights that will flow from the loss of large parts of our documentary heritage. This fact was eloquently stated by the National Archivist of the United States in a recent speech and deserves to be quoted at length:
    [our] mission statement commits us to ensure ready access to essential evidence, documenting the rights, identities, and entitlements of citizens; the actions for which Federal officials are responsible; and the historical experience of our nation. And if you really want to know how important that is, ask the ethnic Albanians now returning to their ruined communities in Kosovo [where] the horrors of ethnic cleansing had come to include the deliberate destruction of public records as a means of erasing the identities and culture of Albanian Kosovars [and] Saddam Hussein certainly understood that getting rid of records, or otherwise denying access to them, makes it a lot easier to end a nation's independence and subjugate its citizens. The Serbs in the Bosnian War understood that, too, when they shelled the National Library in Sarajevo along with other cultural institutions.

    Not the least of the importance of records in a democracy is their indispensability for justice. The government and the public both need records to defend themselves in courts that don't deny access to evidence but allow people to present evidence so that justice can prevail. You've all seen or heard reports in the press about efforts to trace "Nazi Gold" looted from victims of the Holocaust. Much [of that] research already has been undertaken by the National Archives. The Interior Department, and the Bureau of Indian Affairs have come in for intense criticism in Congress and in court for allegedly being unable to produce records documenting entitlements to millions of dollars that may be owed to Native Americans [and] a story from the Washington Post reported a charge by a Pentagon inspector general that certain chemical weapons logs from the Persian Gulf War were missing - records that some veteran's representatives speculated might contain information on harmful chemical exposures.

    Our democracy depends on [the] millions of records in the care of government archives at all levels. These records document all our identities, establish all our entitlements, and enable us to hold accountable those to whom we entrust office, federal, state, and local. Keeping the right records affects people's money, their health, and their history. Records do matter in the life our democratic nation. (Speech by John Carlin, National Archivist of the United States, July 12, 1999)
Many South African's would agree with Mr. Carlin for they experienced first-hand the harm done to individuals and the nation from a willful destruction of a country's archival heritage. In their final report South Africa's Truth and Reconciliation Commission devoted an entire chapter to the impact the destruction of records documenting decades of government enforcement of the racial policy of apartheid has had on the people of South Africa. They concluded that:
    Any attempt to reconstruct the past must involve the recovery of memory - much of it contained in countless documentary records. The mass destruction of records has had a severe impact on South Africa's social memory. Swathes of official documentary memory have been obliterated. Moreover, the apparent complete destruction of all records confiscated from individuals and organizations by the Security Branch of the South African Police has removed from our heritage the country's richest accumulation of records documenting the struggle against apartheid. Numerous investigations of gross human rights violations were hampered by absence of documentation. Ultimately all South Africans have suffered the consequences - all are victims of the apartheid state's attempted imposition of a selective amnesia. An inclusive remembering of the painful truths about the past is crucial to the creation of national unity and transcending the divisions of the past. (
These thoughts are as pertinent to Canada as to the United States and South Africa. We need only recall our First Nations peoples long struggle for their treaty rights, Japanese Canadians wrongfully imprisoned during World War Two, Canada's peacekeepers in Bosnia whose medical files were altered, the Dionne quintuplets or Quebec's "Duplessis orphans", and the tens of thousands of adults abused as children in residential schools or state institutions. All are persons whose struggle for justice and compensation for past wrongs would have failed without access to records preserved by government. The loss of access to or destruction of records, including our censuses, will inevitably lead to rights denied and injustices done.


Barring government acceptance of the Privacy Commissioner's idea of destroying all census records Mr. Phillip's recommends for future censuses adoption of a plan which sees "every member of a responding household have the choice to consent, on an 'opt-in' basis, to have their returns retained (in whole or in part) and released to the National Archives".(Submission of the Privacy Commissioner, p.9) Those who leave the check box blank are to be considered as directing the government to destroy their census form when no longer needed for statistical purposes. The Privacy Commissioner supports this as meeting an important principle of privacy protection: giving individuals the right to control future uses of their personal information. Aside from the apparent unfairness and probable illegality of providing only a "negative billing" option for those who wish their personal information retained, the problems with such a seemingly simple solution are myriad. The practical and legal difficulties for Statistics Canada alone are easily apparent.

First, Statistics Canada will have to make major revisions in how they take a census, changes that will mean considerable increases in the cost burdens of conducting each census. To start with not one but two check boxes will have to be provided so that each respondent communicates unequivocally their wishes of retention or destruction without either option being made under the unacceptable "negative billing" method. After all the importance Mr. Phillips places on individuals controlling their own personal information surely means he grants equal control to those who would say yes as to those who would choose no. Further, given the important legal right to choose which Mr. Phillips believes he would be establishing within the census process it would be necessary to take certain minimal precautions to ensure that this right is properly safeguarded. A declaration printed on the form beside each individuals check boxes establishing that the person has marked the box in their own hand and that they clearly and fully understand its purpose and potential consequences (informed consent), a declaration that of course must be signed and dated, would be the acceptable minimum. For the inevitable number of persons who check neither box, or both, or who fail to sign their declaration, it will be a necessity that enumerators personally contact each person, not household, and have the deficiencies corrected. Not to do so would create a form that could be neither retained or destroyed by Statistics Canada as to do either would potentially violate an important legal right.

This doesn't even begin to address the complexities introduced by this concept. Certainly the traditional use by Statistics Canada of the household form to enumerate most dwellings will have to be abandoned and replaced by an individual form for every Canadian. The household form presupposes that one member of each household obtains and reports the personal information requested for all members of that household - a potential serious violation of privacy rights in itself you would think from the Privacy Commissioner's viewpoint. I have the 1996 Form 2A and 2B (Short and Long Forms) in front of me and personal information for six persons is laid out in a double page spread that makes separating each person's information impossible. As well one person must respond to some questions on behalf of the whole household. Only a household consisting of one person would present no difficulty and even a household made up of a legally married couple with no children is problematic as neither spouse has the right to decide how the personal information of the other is used meaning each person must make a separate decision on the matter. If both decide on the same option then no problem arises and the form may simply be destroyed or retained accordingly. But if they choose separate options then the form could not be destroyed or retained unless a way is found of dividing it without losing some of the information common to both such as address, form identification number, enumeration district, the housing survey, and so forth.

Given the increasing complexity of household make up in this country with its tangled mix of 'traditional' and 'non-traditional' groupings, such as blended families, unrelated persons living together, same-sex couples, and the increasing numbers of the elderly living in care homes, the household form becomes unworkable under the conditions Mr. Phillips proposes. Clearly the 11,000,000 household forms distributed by Statistics Canada in 1996 will have to be replaced by some 31,000,000 individual forms in 2001 if the Privacy Commissioner's proposal is adopted meaning a vast increase in printing, handling, enumeration, processing, and storage costs that will place an enormous burden on Statistics Canada's personnel and budget.

With larger facilities, new technologies, increased staffing, more equipment, and a significant growth in expenditures, some of these problems might be overcome, assuming government is prepared to budget for and taxpayers are willing to bear the burden of, these increased costs. But the issue of informed consent is one so serious it is difficult to see how the census could be conducted at all. For who will mark the check boxes for those deemed, or thought to be, incapable of making that choice: minor children, the sick or incapacitated, those suffering mental illness, and the hundreds of thousands of our elderly suffering from various forms and levels of dementia. On what legal and moral authority will their rights be assumed by others and how will that authority be verified and documented by Statistics Canada? In 1996 my mother was resident in an extended care hospital and was incompetent to manage her own affairs, a right delegated to me via a lawful Power of Attorney. I was not contacted during the 1996 census as her personal information was obtained by a Statistics Canada enumerator through use of the hospital's administrative records which is the usual practice when institutions are enumerated. Under current law that is fine. But if Mr. Phillips' check box rule had been in effect in 1996 I would never have accepted that anyone but I had the right to decide which box would be checked. That would require the enumerator to locate me, verify my legal authority, and have me personally sign and date her census form and anything less would be both unethical and unlawful. You cannot establish a right and then treat it lightly in the name of administrative convenience.

The questions that arise from even a cursory examination of this proposal suggests it was given little serious thought. For example, who owns the personal information of children and do parents have the authority to destroy it? If not, then Statistics Canada will have to retain all forms for minor children and when they reach the age of majority locate them and receive their permission to continue to retain it or to destroy it. What of a person who is mentally ill at census time but who may recover their reason in future? If their form was destroyed through someone else's choice do they have grounds to sue for future damages resulting from the loss of that personal information, or if it was retained how will their permission now be obtained? And what of those hundreds of thousands of elderly Canadians we are all aware of every time we visit a care home: those suffering various degrees of mental impairment that place them in the grey area between mental competence and incompetence. On what evidence and by what authority will they be allowed, or refused, the right to make their own check mark? I do not know the answers to any of these questions and I doubt our Privacy Commissioner does either. That they would be answered by designing a form that presumes destruction as a choice if no mark is made is contrary to the very principles Mr. Phillips claims he is defending.


Our Privacy Commissioner argues that the dead have privacy rights equal to those of the living and his opposition to release of historical census records is in part a defence of those rights. Further he proposes major amendments to Canada's Privacy Act so that there would be "no limitation of a person's right to privacy, whether they are living or dead". (Submission, p.3)

I believe the lack of any such rights of the deceased within our legal tradition stems from the inherent and eminent common sense of our common law. Nor have our legislators established such rights in our constitution or charter of rights. A theoretical concept that would place the rights of the dead ahead of the needs of the living is as impracticable as it is profoundly undemocratic. A right granted by law is a power to be exercised over others, in this case it gives power to the dead to direct or restrict the actions of the living. Pretending to limit the ability of future generations to govern themselves according to the particular challenges of their own times by granting ourselves rights of privacy or whatever is nothing but a dangerous conceit. Let us grant to future generations at least what we demand for ourselves: the right to make our own laws suitable to today's circumstances that meet the needs of the living.

And genealogists are not asking for an "absolute" law on access to historical census records. The present law granting access under the Privacy Act to census forms has worked well for nearly 20 years and there is no demand from the public or Parliament to change it in the way Mr. Phillips wishes. That may not always be so and if circumstances and attitudes change over time then future Parliaments may choose to amend the law. There is ample time for minds to change as the only census affected by a restoration of current rules is the 1906 immediately and the 1911 Census in 2003. The 1916 would not become eligible for public access until 2008, the 1951 in 2043, and the 1996 not until the year 2088. We will have many more generations, Parliaments, and Privacy Commissioners in the meantime. Genealogists are not pretending they can speak for the dead or the unborn: the dead have no claim on the present and the unborn will in good time speak for themselves. In the meantime the decision on what access, if any, Canadians will have to historical census records belongs rightfully and solely to our present Parliament and to no other.


In his submission to the panel the Privacy Commissioner remarked that the "questions asked in censuses have become increasingly intrusive". It is not my intention to defend Statistics Canada on this issue but at the same time Mr. Phillips comments are not entirely accurate at least as to the number of questions Statistics Canada has asked Canadians to respond to over this century. The graph below tracks the numbers of questions asked of all householders from 1901 to 1966 and for eighty percent of households from 1971 to 1996 and demonstrates a clear lessening of the burden on census respondents over that period.

(Graph not included here)
The long form introduced in 1971 asks more questions than previous censuses but it applies to only 1 of 5 households and when you compare the 39 questions of 1911 to the 7 questions of 1996 it is clear that for all households to 1966 and for 80 percent of households from 1971 intrusiveness has decreased not increased. Nor is there any certainty the long form will be used in future censuses:
    [our] research has included examining alternatives to the current approach of short/long form, with a 20 percent sample for the latter, but no decisions to move away from this format have been taken at this time. (Email from Pamela White, Statistics Canada, 14 April 2000)
The United States Census Bureau will abandon the long form after the current 2000 census and only a short form asking six or seven questions will be distributed to Americans in 2010. Given the tendency of Canadians to follow American trends it seems unlikely the long form has much of a future in Canada. Of course this addresses the quantity not the quality of the questions asked although obviously the two are very closely tied. Perhaps Canadians should have more public input into what questions are asked as in the United States where Congress must approve not just the census but each individual question as well. And no doubt Canadians have a limit on what they will answer and what they will not and Statistics Canada must be keenly aware of that.


When Mr. Phillips commented in his submission that "to know things about our ancestors is sometimes to know things about ourselves" he captured the very raison d'etre of genealogy and history alike. We offer no apologies nor do we fear the self-knowledge and wider understanding of our society and country such study can bring. As for the state of perpetual amnesia Mr. Phillips would see imposed on ourselves, our children, and our nation I can only wonder if he has ever witnessed first hand the terrible withering of the human spirit that loss of memory wreaks. Perhaps not. But one thing is certain. We will have a history, of our families and our country, the only question is will it be accurate and truthful. That will be answered by how carefully this generation gathers and conserves its records.

David Irving has long tried to peddle his alternate history of the Holocaust and failed miserably. Not because he is not a clever or persuasive man but because the historical documents of that terrible event and the memories of survivors exposed him for the liar he is. What lies about us and our country will rise from the ashes of our documentary heritage thrown by misguided privacy zealots into their bonfire of perfect and perpetual privacy.


This debate has never been a dispute between those who value privacy and those who do not. Genealogists and historians fully understand the need for rules respecting confidentiality of census information and are not seeking a diminishment of privacy rights by weakening the Privacy Act or the role of the Privacy Commissioner. The debate is about the proper balance between the fundamental democratic right of Canadians to access government records for legitimate purposes and the right to have their personal information treated confidentially. But even as important a right as privacy is subject to the reasonable limits that recognition of other rights or important social values and goals place on them. The right to access historical census records as granted under authority of the Privacy Act regulations and the National Archives Act represents both the intent and will of Parliament and should be recognized. We are not asking for increased privileges of access nor that the closure period be reduced but only that present law be respected. Nothing has occurred to upset the balance between access and privacy rights in place since 1983 that requires changing the 92 year closure period and there is no demand from Canadians to destroy census records or deny access to them forever. In fact this controversy has been initiated from within the civil service for their own perceived purposes and is not coming from government, parliament, or the population at large.

And there is no evidence releasing historical census records to public access after a long closure period violates anyone's right to privacy, affects the willingness of Canadians to participate in census enumerations, or harms individuals living or dead. For decades Canada, the United States, and Great Britain have released their census records without public complaint or injury to citizens. And neither the Privacy Commissioner or Chief Statistician has presented any credible proof to support their views nor have they accepted responsibility for the serious consequences that will inevitably arise from such a widespread destruction or closure of public records. This debate has dragged on long enough and needs to be settled, and soon, by Parliament. There is real concern that if this issue is unresolved by the time of the 2001 census it may effect the willingness of some Canadians to participate as some genealogists are already talking openly of a census boycott. The government must recognize the seriousness of this issue and act quickly to resolve it.

If our current government doesn't move soon to end this impasse it will be brought before the courts. And that will serve no one well for it is not in the public interest to have the confidentiality of Canadian's personal information held by government exposed to the vagary of a court decision. Only the government can prevent this by legislating in a timely manner an amendment to the Statistics Act that will make clear the rules for public access to historical census records. We would urge the panel to recommend to Mr. Manley that he do just that at the very earliest opportunity by restoring public access to Canada's twentieth century and future censuses after a 92 year closure. But irrespective of the final position of this panel, or government, on access to historical census records that position must be reflected in new legislation and voted on by parliament. The status quo is no longer an option.

Request for Submissions to the Expert Panel

Have you made a major submission to the Expert Panel on Access to Historic Census Records - on behalf of a genealogical or historical society, or yourself? If so, please send me a file(s) containing that submission so that I might include it in future columns, and thus inform our readers.

Canada Census Campaign mail list

The Canada-Census-Campaign-L mail list was set up to provide a forum for those interested in obtaining release of Historic Census Records in Canada.  It is not for look-ups or individual queries.  Your comments and questions relating to release of Post 1901 Census records are welcome.  Subscribe to the list by sending an e-mail to

With only the word subscribe in the subject line and the body of the message.  Do not include any other text or signature files in the body of the message.  To subscribe in Digest mode, change the ‘L’ in the address to a ‘D’.

Post 1901 Census Project website

If you have not already done so, please send me any responses you have received from your Member of Parliament regarding the Census problem.  You are invited to view the website at

Until next time.  Happy Hunting.

Gordon A. WATTS

© Inc. 1992-2018
Sign up for our free newsletter!   |   Unsubscribe from our newsletter

Norway Bay United & Anglican Cemetery
(Pontiac County, Quebec)

The Merivale Cemeteries
(Protestant - Ottawa area)