Archives have been destroyed and damaged and will continue to suffer this fate as result of carelessness, accidental fires, arson, cyclones, pillage, shelling and air attacks, external and in-house flooding and so on. Archives have been destroyed and damaged and will continue to suffer this fate, by archivists and users, by mould and termites, but also by enemy-action and by partisans and liberators, by revolutionaries and counter-revolutionaries. Archives have been destroyed and damaged and will continue to suffer this fate due to the inherent instability of the materials they are made of, due to poor storage facilities, due to lack of training or lack of staff discipline, but also due to lack of interest from peers, administrators, etc.
Continuing acts of terrorism, ethnic cleansing and related archival cleansing and other acts of barbarism will add many more record groups to the list. Some of the disasters resulted from brutal violence by agents of the dominant political system, others from similar action by their opponents. To give a few recent examples, so far repositories and archives have been completely destroyed in Liberia, Burundi and Rwanda. The same has been reported about the territory of former Yugoslavia. Continuing attacks by humidity, heat and termites will result in the annihilation of archives in several countries in the tropics in the African, Asian, Pacific and South American regions in the next decades. Reality forces one to state that, without massive assistance, parts of Africa, the Pacific region and South and Central America will be bereft, not only of their oral tradition, but also of their archives.
Several colleagues provided data on the fate not only of public and official records, but also of private archives and special collections, like photographic and audio-visual archives. The outcome of this information is equally appalling. In many regions private archives and audio-visual materials will just vanish together with large sections of public and official records.
The loss of archives is as serious as the loss of memory in a human being; societies simply cannot function properly without the collective memory of their archives. That is why it is so vital to take action to stem the losses which have been revealed in this survey. There are things we can do.
This report may be the result of my hands, but I have received a lot of information and advice from several ICA-officers and other colleagues, especially Mr Ingmar Fröjd, Mr Björn Lindh, Mr George Mackenzie, Mr Michael Roper, Mr Atique Zafar Sheikh, Ms Soemartini, Ms Comfort Ukwu and Ms Zakiah Hanum Nor who discussed my ideas or sensitized me to other perspectives. Mr Ken Hall volunteered, as usual, as conscript language editor. However, most thanks go to the colleagues who collected and commented on all data, some of them in adverse circumstances.
2 INTRODUCTION By their very nature archives are unique both as individual documents and as documents in context. Lost archives are irreplaceable, any loss is final, reconstruction is impossible. Most record groups have been subject to a well defined appraisal process and have been selected for permanent retention because of their legal, informational or cultural value. Even the loss of parts of record groups selected this way for whatever cause, devalues legal and informational worth of the remainder. Archives are threatened by both internal and external factors, such as quality of component materials, rodents, mould, acidity, fire, users, etc. Regrettably we have to add external factors of another kind, such as political systems, shelling, arson and cleansing. In this report mainly neutral - generic - terms like fire, water, dust, use, will be used, whatever their cause. Archives are generally considered to form the skeleton of the "Memory of the World", by containing not only factual information but also the informational context in which other elements of life, for example paintings and sculptures, wars and discoveries, can be placed and better understood.
However, by using the generic term `archives' one implicitly accepts its limitations: `archives' are part of a European concept, based on Roman law1, a concept that was imposed on modern societies all over the world. Many societies outside Europe had developed advanced writing systems and preservation practices long before European colonists arrived with their own record-keeping systems based on European paper. Such paper does not survive well outside temperate climates.2 Climatically well proven systems for `memorising' data have been put aside as not suitable for `European' administrations. In some cultures both systems `co-habitated', the European one providing core data, `facts', the indigenous one providing circumstantial evidence of some importance for understanding local traditions relating, for example, to religion or to culture, or providing other kinds of information.
In essence the information system embodied in `European' `archives' was created to deal with property. In other cultures it dealt mainly with different kinds of data, like locations of fresh water (e.g. Australia), movement of herds (e.g. North-America) or the relationship between deities and man. Under the assumption that script for storing data was introduced in accordance with local needs, one should keep in mind that even in highly literate cultures elements of oral and other traditions are still used. There are many good reasons to reconsider the validity of `European' definitions of history and pre-history and to accept `data' transmitted via other traditions as part of the corpus of historic data. One might also reconsider the validity of `European' archival definitions for their applicability in non-European societies. This report, however, will restrict itself to records and archives according to the established European tradition. Before doing so, a few lines on the relativity of archives for the knowledge of the `history of man', by relating them to the voyage of human species in time. According to many scientists, just after the last Ice Age, `Modern Man' started about 100,000 years ago to domesticate animals and to adopt a sedentary life. Modern Man added script to his utensils for preserving the `Memory of Man' only about 5,000 years ago. The earliest recordings of his writing, even official records, are to be found in museums, not in archives.
Script is nowadays a reliable way for transferring information. How `reliable' will it be in future? How to convey a message to homo sapiens over a period of 50,000 years? For instance, a message like: `keep out, radiation zone', put on top of underground nuclear waste belts? What kind of `sign' will be understood 5,000, 25,000 or 50,000 years from now, as a warning not to drill in the ground because of the danger of radiation? What material should one choose for preserving any sign for such a long period: paper, wood-blocks, parchment, microfilm, clay-tablets, palm leaves, solid rock, computer-tape or diskettes, acoustic systems? Will there be any institution keeping records as over 50,000 years old? Will records of that age be more likely to be kept in museums, as happens nowadays with records of 5,000 years ago? What equipment will people have by then to decipher messages - computers, or only brains and reading glasses? Such questions are not easily answered. As a native Australian proverb goes, `rocks vanish, words remain'.
These questions open a domain of professional relevance: durability of `data carriers', like paper, computer-diskettes, movie films, clay-tablets, of `data', like script of any kind or graphics, of the chemical and physical fixation techniques that make `data-carriers' and `data' stick together (water in ink; magnetism; heat); and of instruments and `brains' that make `data' understandable and thereby turn data into `information' (several early scripts are still awaiting deciphering).3 Little is known about the expected life span of specific `data-carriers' apart from rock, of the `sticking-material-technique' and of the `equipment-brain-span' that make information out of data (or even identify possible data as such).
Here is an example for the sake of argument. In modern archival literature one can read a lot about acidity and the ageing of paper. However, how much has been published on ageing of paper as such? How much on life expectancy of a specific make of paper of a given era, exposed to a continuously high relative humidity, or a cyclical high and low relative humidity, or a continuously low relative humidity, combined with temperatures high, low, moderate or cyclical, combined with dust, exposure to sunlight, folders, boxes, administrators, archivists or users? Is this data available? Is data available on the ageing of paper in thick-walled, heavily insulated repositories in a variety of climates? Is there data on what happens to paper in thin-walled repositories fitted with cooling equipment that functions a few hours per day only? Is any information available on what happens to records stored in properly conditioned repositories and consulted or listed in hot and humid searchrooms or office blocks? Do we have any idea of the factual relationship between storage conditions and chemical and physical decay of paper, photographic materials etc.? Do we have any data for any formula that will enable us to make reliable estimates on the return on our investments in staff-time or in money? Do we have any data that can be used for risk calculation or for setting priorities?
Here are some postulates. In tropical climates, as has been established, it may take records, even if of long-lasting paper, only 100 to 200 years to become dust. Before that, they cross the no-use line (identical to a no-research line) and, shortly after, the no-touch line (identical to no-reformatting line or past-lamination line). In moderate climate zones, the no-research line may be crossed after 1,000 years and the no-reformatting line after 1,500 years. Special problems are posed by newspapers when printed on unstable paper of low quality. In some countries, this kind of paper is also used for stationary. The no-research line of unstable paper will be crossed in the tropics within 100 years, in more favourable climates within 400 years.
However, long before record has become dust, the data may have faded away. For example, some makes of ink fade easily, other kinds `eat' paper. Some kinds of photocopies do not stand up under sunlight, other kinds can, if not properly processed, be wiped out easily. Some kinds of stencil seem to lose contrast, etc. Poor quality of ink, of magnetism - submitted to chemical and physical processes as they are - will increase the speed of decay of carriers and their data even further, even when, by comparison, kept under stable conditions. One may conclude that, according to the materials used and their environmental and office and repository conditions, the life span of carriers and data may vary in the tropics from a few years for some materials to twice the life span of man for other materials and in moderate climate zones from one or more decades to 5-20 times the life span of man. If one adds variables like fire, wind, water and war, a similarity with Russian roulette becomes apparent, as will be shown later. The report presents the scale of the problem we face; the challenge is to develop strategies to deal with it.
3 COLLECTION OF DATA
The `Memory of the World' Programme was launched by UNESCO in 1992. It is meant to preserve endangered documentary heritage as well as to democratize access to it and ensure a wider diffusion. The programme intends to sensitize governments to the importance of protecting their documentary heritage.
ICA was contracted by UNESCO to collect basic data on archives as part of the documentary heritage. These data should give an insight into the hazards archives have been, and still are, exposed to in the 20th Century. ICA was also to prepare a list of archives that have been destroyed or damaged as result of natural or man-made disasters (appendix 2).
A questionnaire (appendix 1) was prepared by ICA and agreed upon by representatives of IFLA and UNESCO. The questionnaire was sent in August 1994 to all Category A members of ICA and to those members of other categories that have suffered losses relevant to the purpose of the report. The organisers of the Pan-African Conference on Archival Policies and Programmes in Africa and of the "Memory of the World" Experts' Meeting of the Asia Pacific Region circulated the questionnaire also among non-members of ICA. In total about 225 questionnaires in 156 countries have been mailed. When applicable, Category A members received a list of other ICA members in their country who received the questionnaire separately. As requested in the cover letter many members circulated the questionnaire, resulting giving information on about 6,250 repositories in 105 countries; some 6,000 repositories reporting losses (appendix 2).
Table 1995/1 (form A) Repositories covered
Table 1995/1 already confronts us with a statistical problem, namely both `under-response' and `over-response', `under-representation' and `over-representation'. North-American archives suffered from very few disasters of any kind. The Chinese archival authorities reported in general terms on 3,000 repositories. The Pacific countries supplied few answers. The Russian archival authorities provided in broad terms information without specifying the number of repositories involved. The Spanish civil-war resulted in the total or partial destruction of over 1,700 repositories. Italian archival authorities provided detailed information on over 600 repositories.4 Many respondents reported on one event causing destruction or severe damage, however the great majority reported multiple occurrences of losses.
This spread of answers does affect the statistical consistency of the findings. Extrapolations have to be carefully handled, especially in case of the data presented in the columns `North America' and `Pacific'. However, the findings do present a good overview of causes of destruction and damage and resulted in a long list of examples of destroyed and damaged record groups. Professional archivists all over the world will be able to interpret the findings in accordance with local, national and regional circumstances and to inform administrative authorities accordingly.
Country reports presented at the Pan-African Conference on Archival Policies and Programmes in Africa (Abuja, Nigeria 1994) and at the Memory of the World Programme's Experts' Meeting of the Asia-Pacific Meeting (Kuala Lumpur-Malaysia 1994) have been of great help for a better understanding of the complexity of the subject. Both meetings provided a perfect occasion for studying both the country reports and the completed forms with the authors.
Addressees were requested to take into account that the questionnaire intended to deal with all archival holdings (including audiovisual archives) that had been selected for permanent retention. In some cases it was apparently difficult, impossible or, given national legislation, irrelevant to make such a distinction. Several respondents provided additional information to clarify such cases. Addressees were also invited to indicate, for all archives involved, the survival of finding aids or of printed or other reproductions (in transcribed or in other form e.g. facsimile or microforms) of parts of the archives involved. They were also invited to indicate in shelf metres the amount of documents that have been destroyed or heavily damaged. This kind of information has been provided fragmentarily and will not be presented in a table of its own.
The majority of returns were received by February 1995, including information up to events as late as the 1995 earthquake damaging the Kobe region in Japan. Several respondents considered in their cover letters that data gathering was a stimulus: several institutions never collected this kind of data systematically before. Other correspondents apologized for their incapability to provide comprehensive answers, the explanation being a dramatic one: losses - always caused by war - being unquantifiable. One of the respondents suggested a text providing some examples of annihilation of archive repositories "instead of a comprehensive answer needing a truck for carrying thousands of questionnaires that had to be completed otherwise."
Special attention was requested by respondents for systematic removal of archives by occupying forces - a removal possibly resulting in destruction of some if not all archive series involved, in order to remove or destroy proof of evidence, or simply for reasons of `archival' or cultural `cleansing'. Some respondents asked for anonymous presentation in the report, as did some other respondents providing data on e.g. neglect by national or local authorities.
An analysis of the answers shows several important disparities; some reporters refer to repositories of archive services as such, some refer also to records temporarily moved to and destroyed or damaged in auxiliary repositories, others include records that should have been transferred to an archive repository, a few reporters did not discriminate between records kept in archive repositories and records kept by creating agencies, even if not yet selected for permanent retention. These disparities do not influence the spectrum of answers substantially. If the amount of destroyed and damaged archives increases, the causes of their destruction or damage do not change.
From a theoretical point of view, it could have been of interest to make cross-tabulations, like the number of collections destroyed as result of fire, floods, war, etc. From a statistical point of view, cross-tabulations are not always very helpful in analysing the problems one is researching and they would certainly not justify the additional workload. Apart from this, from the point of view of the user, loss of information is the most important factor. Intentionally therefore, the arrangement of the questionnaire did not foresee cross-tabulations. However, some respondents kindly arranged their answers allowing some cross-tabulations. An analysis of these forms demonstrates, not surprisingly, a cause-effect relation: fires quite often resulting in the installation or improvement of fire alarms and fire-fighting equipment, floods leading to the installation of water alarms or the transfer of records to safer repositories, and leakage generally to a better maintenance of the building.
Many respondents reported a lack of knowledge of the full history of their (previous) collections, many institutions having been established only after 1945 or having professional staff even more recently. Two of the cover letters illustrate in a few lines the impact of what has happened in far too many cases, all over the world:
Some of our repositories only completed form A of the questionnaire, since they were founded after 1945 and suffered no losses since. All other repositories suffered great losses. During this century, especially during the Second World War many repositories were completely destroyed. It is still impossible to estimate the total damage as all finding aids were destroyed together with the collections themselves. Therefore, most repositories could provide estimates only.
Currently, the most serious dangers are posed by the level of pollution of the environment, by the bad quality of paper used for records and by the lack of cost-effective conservation methods. An overall threat is posed by financial constraints, limiting the use of acid-free storage materials and the provision of conservation workshops with proper equipment.
We do, however our best to protect our holdings against fire and theft. We managed to secure the information in the most important records by producing microforms and by making diazo-copies available to the public.5 Regrettably we cannot provide all details as far as the destruction of archives of the fascist period is concerned, since civil servants - members of the fascist party - wantonly destroyed records in order to dissolve their traces.6 At the Gardone Riviera Round Table on Archives of 1987 (`Policies for the preservation of the archival heritage'), heads of national archives, chairs of professional associations and representatives of IFLA discussed the `state of the art' of preservation in archives and libraries. Papers had been prepared by Mr D.W.G. Clements and Ms Marie Allen, based on a questionnaire conducted in 1986 jointly by IFLA (550 libraries, 194 responded = 35%) and by ICA (300 archive services, 217 responded = 72%), providing data on a total of 263 archive repositories.
Three publications7 present together a more or less complete survey of all papers that were submitted to the Gardone Riviera Round Table and of the discussions of the meeting. The tables presented hereafter are derived from the original hand-out `reporting forms' presenting `database tabulations from ica/ifla questionnaire on conservation'. The CITRA-publication carried a summary outline of these forms only. A synthesis for archives and libraries per geographical area, based on the complete set of reporting forms, has been published in the Nederlands Archievenblad.
The data, as presented on the basis of the 1986 questionnaire, can be considered to be representative of the state of preservation and conservation in archives all over the world. From a statistical point of view it is not advisable to deduce `fixed conclusions' from any column based on less than 20 answers. However, smaller figures may be used as an indication of the archival situation in those geographical areas.
Analysis of the data presented provides some understanding of the archival habitat. A little confusing may be that some of the 1986 tables present data on 217 archive institutions and other tables data on 263 archive repositories: some institutions provided data on more than one repository.
Comparing the outcome of both questionnaires one gets a feeling of déjà vu: tropical and sub-tropical climate zones are hard on materials used for records, bindings, microforms, etc.. Archive services based in countries with a better climate are usually better off and better equipped to preserve archives of any kind. Many archive services based in areas plagued by war in this century lost essential sections of their holdings, containing unique information on local, national, regional and even global history.