The Australian Archives (AA) was established by the Archives
Act in 1983 as part of the Government's administrative law
reform package. Under the terms of the Act, we are the organisation
responsible for the broad management of the whole body of Commonwealth
records. Our primary objective is to ensure that records are readily
available to meet the proper interests of Government, Commonwealth
agencies and the public.
Part of our mandate, of course, concerns the records of Commonwealth
science and technology activities. In Australia, the majority
of science and technology research is sponsored by the Commonwealth
government, a proportion of which is directly undertaken by Commonwealth
agencies. These agencies include fairly high-profile organisations
such as the CSIRO and the Defence Science and Technology Organisation,
but also a myriad less well known agencies such as the National
Resource Information Centre, part of the Department of Primary
Industries and Energy and the Alligator Rivers Region Research
Institute, which is in the Department of the Environment, Sport
All of these agencies produce Commonwealth records and, whether
they are aware of it or not, are subject to the provisions of
the Archives Act. Under the terms of the Act, all Commonwealth
agencies have a responsibility towards their records. That responsibility
includes not disposing of any records without our permission;
assisting the Archives to decide on the disposal status of their
records (i.e. how long they should be kept); and, following that
decision, destroying valueless records as soon as possible and
transferring those with long term temporary or permanent value
to the Archives when they are no longer needed.
It follows then that the Australian Archives has a substantial
role to play in 'recovering' science in this country. We are uniquely
placed in the Australian context because we are responsible for
the records not only of research activities, but also for the
records of the policy decisions which create the impetus for much
of that research. If, in the future, a researcher wants to examine
the development of the Australian Government's policy on monitoring
the Greenhouse Effect and the implementation of that policy, or
the success or otherwise of the 'clever country', then the Australian
Archives will be the place to come for the records.
That is the theory; that is what we want to achieve. In practice,
of course, the situation is not nearly so clearly defined. At
the Archives we are faced with between 2500 and 3000 current Commonwealth
agencies, not to mention the thousands of now defunct or superseded
agencies, whose records we already hold. It is inevitable that
a number of agencies, for one reason or another, have slipped
through our net. Unfortunately, some agencies with science and
technology functions fall into this category. Before we can guarantee
that the researcher bent on knowing everything about the Commonwealth
Government and the Greenhouse Effect or the clever country will
be satisfied with our holdings, a number of things must happen
and happen quickly.
Recognising this fact was the first step, and I'm pleased to be
able to say that the Archives has moved substantially beyond that
point. I would like to discuss the progress we have made in dealing
with the Commonwealth's science and technology records.
What sets science and technology records apart?
Before coming to grips with selecting and preserving these records,
we need to determine whether science and technology records are
in fact any different from other Commonwealth records and thus
whether they need to be treated any differently. At the most fundamental
level, the answer to the first question is no. Any of the difficulties
we have with science and technolgy records apply equally to the
records of other Commonwealth activities; however, science and
technology records represent a microcosm of all the hurdles the
Australian Archives encounters. So, while the records are not
different, taken together they represent a challenge to the Archives
that requires a slightly different approach. The reasons for this
fall into two main groups:
- The number and diversity of science and technology agencies
and the apparent complexity of the records they create appears
more daunting to a non-specialist than other types of Commonwealth
- For one reason or another, science and technology agencies'
staff are less likely to be aware that they are creating Commonwealth
records which are the proper concern of the Australian Archives.
I will examine both of these issues briefly for it seems that
the crux of any perceived or real problems we have with science
and technology records lies within.
The diversity of science and technology agencies and their
The involvement of government in science and technology in this
country goes back to Federation and beyond, and there have always
been a number of different activities that have been sponsored
at any one time. It almost goes without saying, that the number
and complexity of science and technology functions that the Commonwealth
sponsors directly or indirectly has increased by an order of magnitude
since the Second World War. From 1945 until the present there
have been many more research fields, a greater degree of specialisation
and a greater complexity of research.1
This translates to many more agencies for the Archives to deal
with and records that cover more specialised subjects.
At the same time, there is no doubt that the number of records
produced by the Commonwealth has been increasing exponentially.
This is partly a direct result of the increase in the number of
functions that fall under the Commonwealth umbrella, but it is
also a reflection of changes in technology that have made the
duplication of records and methods for collecting information
quick and easy. The photocopy machine has a lot to answer for
- world-wide an estimated 5000 million copies are being produced
today.2 Where records are concerned,
quantity breeds contempt - the more there are, the more difficult
it is to find the records that do have long term value, and conversely,
the easier it is to destroy them by mistake in the attempt to
'rationalise' a records storage problem (whether that problem
is shelves groaning with paper, or an overloaded computer memory).
We are dealing with growing quantities of records, but the formats
in which science and technology records are created and kept have
always been diverse and will continue to change and challenge
the Australian Archives as well. We can and will cope with a surprising
variety of record types. Notebooks, logbooks, photos, video and
audio recordings, maps, charts, models, paintings, magnetic tapes
and disks have all made their way to our repositories. Apart from
having some special requirements for storage and access, including
more closely controlled environmental conditions, they cause the
Archives few problems - except for the last, the electronic records.
These have proved difficult to deal with using traditional archival
practices and it is worth discussing the issues involved, if only
Science and technology researchers have been particularly innovative
in developing an array of computer applications to facilitate
data collection and manipulation. Therein lies a Pandora's Box
of archival concerns. Electronic records have changed the definition
of record from something that is obvious and immutable, to something
hard to grasp and categorise neatly. The laundry list of archival
woes could continue ad infinitum; however, I will make just three
Firstly, technology has affected the way records, and the information
they contain, are created, used and stored. Widespread use of
computers has meant that while it is certainly possible to store
much more information in a small physical space, it is also possible
to destroy or lose access to that information with awesome ease.
My second point is that technology has opened up avenues for data
collection and manipulation that were never a possibility when
paper was the only option. But the danger lies in letting the
collection of information be technology driven - collecting information
because we can, rather than because we have the increased capability
to make use of it. Think of readings that were once taken manually,
once a day, that can now be recorded every second. Think of the
terabytes of remotely sensed data downloaded each year from orbiting
satellites.3 Considerable resources
have had to be employed writing data compression programs to make
the data useable!
Finally, improved and expanded telecommunications mean that communication
between people is easier, but the record of that communication
is ephemeral. Letters are replaced by a telephone call or electronic
mail dialogue, or information is exchanged through electronic
bulletin boards - records that rarely make it into durable formats.
From the Archives' perspective, these developments impose extra
risks on records with long term value for research. Even in agencies
where we have tame records managers willing to undertake archival
work, electronic records have tended to be viewed as being outside
the agencies' record keeping system. They have been described
with imagination as 'islands of unmanaged information'4
- perhaps 'island continents' is a more realistic analogy! We
therefore have to rely more heavily on agency staff to help identify
the records with value. There are also problems of obsolescence,
where information becomes inaccessible because of technological
change, deletions are possible at the accidental or deliberate
touch of a button, and control of information lies much more firmly
in the hands of users. When we do determine that particular electronic
records should be kept, there are additional concerns about how
and where the records should be retained.
Science and technology agencies and the Australian Archives
Of course, the Archives only has to deal directly with these concerns
if we actually get our hands on the records. In this respect,
the Australian Archives is working from behind the eight ball.
Not only do we not have the resources to consult with all agencies
about their records, but archives work in Commonwealth agencies
tends to be assigned a very low priority, because there always
seems to be something more important to do with resources. Like
getting on with program activities that create more and more records.
In general, we can live with this. During the time since the Archives
Act was proclaimed, we have concentrated our efforts towards
firstly dealing with agencies that have substantial quantities
of largely temporary records and thus are likely to have storage
problems that can be resolved, because most of the records can
be destroyed sooner rather than later. These we euphemistically
called 'the big easies'. Secondly, we have directed our efforts
towards dealing with agencies that are likely to be creating the
most valuable records. In theory, this leaves the agencies with
more manageable quantities of records or with less valuable records
as the residue.
Science and technology agencies tend to be one of the groups left
behind. While they might meet the criterion of not holding overwhelming
quantities of records, in terms of the importance of their activities
in the Commonwealth scheme of things, they should have been high
on our list of agencies creating valuable records. With a few
notable exceptions, such as the CSIRO, and the Anglo Australian
Telescope, this was not so, and is an issue that needs addressing.
The reasons lie partly in the diversity of agencies and the specialisation
of their records that I discussed earlier. Why go searching for
trouble that you know exists, when there are easier tasks at hand?
A corollary to the post-war diversity of science and technology
activities is their wide geographic distribution. Why go so far
from home to find trouble? There aren't many inducements unless
the agency in question happens to be on an island in the Whitsunday
The second part of the explanation lies in the attitudes of researchers
to their records and the record-keeping practices at the agencies.
Independence in research does not sit comfortably with conforming
to requirements for the proper management of the records produced,
when those requirements are set by a non-scientific body such
as the Australian Archives. The records created in these agencies
are far more likely to be maintained according to personal preference
and to be regarded as personal rather than Commonwealth property,
than are records anywhere else in the Commonwealth.
Personal records do have a habit of ending up mixed with official
records, and that can add to the confusion. It leads to a rather
ad hoc approach to their retention and destruction, and it is
as likely as not, that all of the records will go with an individual
researcher if he or she moves on.
Solving the problems
Well, it sounds grim, but the situation is redeemable. I'd like
to spend the last part of this paper discussing the various mechanisms
we are putting in place to turn this situation around, to have
science and technology agencies knocking on our door so that their
valuable records will be identified, preserved and made available
Selecting the valuable records
Firstly, the Australian Archives has to make sure that the right
records are selected and retained. This part of the process is
pivotal to all of the other archival concerns we might have. Our
track record in this respect sets no shining example, but we are
learning from past mistakes. The Australian Archives already holds
nearly five hundred shelf kilometres of records in its repositories.5
Nearly 40% of these are unevaluated and were brought into custody
in the days when space was at less of a premium and before we
had clearly defined criteria for selecting what should and should
not be kept. Proclamation of the Archives Act went some
way towards defining which records are our primary concern, and
since 1983 we have been introducing a new vigour to our disposal
program, so that we do not compound the storage problem in the
future. We have concentrated on refining our appraisal criteria
and have embarked on a program to reduce our unevaluated holdings.
Hindsight is proving a valuable asset in determining which of
these older records do have continuing value.
We make no apologies for being ruthless in selecting records for
retention. The Commonwealth, and its tax payers, cannot afford
the luxuries of the past - we cannot afford to keep more than
is absolutely necessary. The advantages of destroying the dross
are not only economic ones. Once the valueless records are destroyed,
the resources we do have can be much more usefully put to work
caring for and providing access to valuable records. There is
a substantial benefit for researchers.
However, no one who has appraised records to determine which should
be kept and which should be destroyed will tell you that it is
an easy task, particularly for modern records. There are always
some obviously valuable and equally obviously valueless records,
but by far the greatest proportion lies in the grey area between.
Each of those records might have its five minutes of research
glory, but this must be balanced against the costs of keeping
each of them indefinitely.
There is an abundance of archival theory that is a sound basis
for appraising records, but on its own it does not make it easy
to deal with the sheer quantity and variety of modern government
records, particularly science and technology records. As I mentioned
earlier, the agencies range from large organisations, to single
laboratories in remote locations. The records come in every conceivable
format, using every conceivable (and inconceivable) code to make
interpretation difficult. The information contained in the records
can be user friendly and within the realm of understanding of
a simple person, or can be so arcane as to defy understanding
even by other scientists in the same discipline, let alone the
average appraisal officer from the Australian Archives.
We felt that the appraisal solution lay in finding any common
threads that exist within this disparate group of organisations
and their records, that could be used to provide non-scientific
people with techniques to make the right decisions about the records.
As background, we looked at any dealings we had had with science
and technology agencies and our knowledge of their activities
and records, and we looked at the current literature, particularly
that from the United States.6
This review resulted in some fairly general observations about
the agencies and their records which formed a pattern that could
be used as a basis for an appraisal model. This was formalised
into a set of guidelines that our operational staff can use on
the records of science and technology agencies.
The model that provides the framework for the guidelines divides
science and technology functions into four major activities that
are likely to be undertaken in one form or another in each Commonwealth
agency carrying out science or technology based research. These
- Establishing the research program or project
- Designing and implementing the research methods
- Disseminating results
- Administering the program or project
By extension we can look at the sorts of records generated by
these activities and which of them are likely to have long term
value for research.
As a government archives we are interested in keeping those records
that demonstrate the development of government policy at the highest
levels and the implementation of those policies in the significant
programs. These are the records that show evidence of how government
functions. Within our guidelines, this is the first stage of the
model encompassing the establishment of the research program.
As a general rule, our appraisal staff do not have substantial
problems identifying the records that will be most valuable for
these purposes because they do not differ dramatically from the
policy records of any other Commonwealth function. A researcher
interested in looking at the development of government science
and technology policy should be well served by the records we
hold or will hold in the future.
Similarly, the records that document the administration of particular
programs or projects are likely to present little by way of special
appraisal problems. Indeed, many of them already have disposal
coverage in the form of our General Disposal Authorities, which
we have prepared to cover the records common to most Commonwealth
As a government archives, we are also interested in keeping those
records that show the implementation of government policy through
its significant programs. Within our guidelines for science and
technology records, these activities are the second two phases
of the model. Our hypothetical researcher may be interested in
continuing a particular piece of scientific research and would
want access to the core research data as a means of building upon
or revising the scientific knowledge base. These records are the
ones most likely to cause non-specialist appraisal staff headaches.
They come in the widest range of formats and lend themselves least
to common understanding. The data itself is difficult, but there
is also confusion over what to keep because of the overlap between
published and unpublished material. The advice contained in the
guidelines is mainly directed towards providing a means of dealing
with these core research records. The most important piece of
advice is for the Archives to involve agency staff wherever possible.
The guidelines include a broad classification of science and technology
records that identifies likely retention periods for the most
commonly occurring types of information. These can be adapted
for the records of specific agencies to form disposal classes
that can be authorised by the Archives.
We have also developed a number of complementary guidelines that
give general definitions of the appraisal criteria of the Archives,
and others that document procedures for dealing with older records
and records in a variety of non-paper formats, such as sound and
moving image records and photographic records.
Considerable resources have been devoted to developing guidelines
for appraising electronic records. They have special significance
for the records of science and technology agencies, simply because
of the volume of research information kept in electronic format
and its likely long-term value. An indication of the continuing
value of these records is demonstrated by a recent survey of databases
in the USA which suggested that out of 300 scientific databases,
130 (or about 40%) of them should be kept. This compares with
keeping an average of only 9% when databases from a wider range
of subject areas are considered.7
Our guidelines are primarily about ensuring that electronic records
with these sorts of continuing values are identified as soon as
possible, before data is lost or software and hardware become
obsolete. The best approach is for these decisions to be made
during system development. That way arrangements can be made for
the valuable records at the outset. They can then be backed up
in a systematic fashion and migrated with system upgrades.
These guidelines are currently in the Disposal Manual which is
for internal distribution to our operational staff. They have
already proved valuable in appraising the records of science and
technology agencies, including for example, the Australian National
Parks and Wildlife Service. Of course, their value is limited
to assisting appraisal of the records of agencies that come to
us for advice, or agencies that we target for disposal work. These
have tended to be the larger and more centralised agencies. Small,
specialised agencies are now high on our list of priorities and
much of the work we undertake in the next few years will be directed
towards developing techniques to provide them with appropriate
archival advice and services.
The next step is to work out exactly the dimensions of this task;
how many agencies are we talking about, what sorts of functions
are they responsible for, how receptive will they be to our concerns
about their records. By developing profiles of information on
the agencies we will be in a much better position to give them
appropriate and timely advice. We already produce a range of guides
for agency staff to use in dealing with records. These include
appraisal guidelines, advice on planning disposal programs in
agencies, information on transferring records and storing records
with the Archives. In the future we hope to develop the science
and technology appraisal guidelines further and make them available
to staff in the agencies, and then target those agencies for archival
work. By getting agencies involved we can tap into a rich source
of subject expertise that we can never hope to duplicate ourselves.
Making the records we hold available for research
All of these measures are directed towards selecting and keeping
valuable records that are held by agencies, are being created
now or will be created in the future. This begs the question of
what we are doing to make our current holdings available for researchers.
Under the Archives Act, most Commonwealth records are available
for public access at 30 years, although there is provision for
extended closure of sensitive records and accelerated access to
some other records in certain circumstances.
The records in our custody are controlled by the Australian Archives
Records Information Service database, more commonly known as RINSE.
This database is available on-line to the public in each of our
Regional Offices. There are a number of different ways to interrogate
the system to find information on agencies, functions, and record
series. RINSE also supplies the links between agencies and functions
and their records, which allows researchers to make the connections
that exist between policy matters and researchers throughout the
network of committees, councils, funding channels and the budget
This is supplemented by the Australian National Guide to Archival
Material I - ANGAM I - which is available on microfiche and describes
agencies of the Commonwealth Government that have existed since
Federation and the records those agencies created and controlled.
ANGAM II is a database listing all records that have had access
decisions made about them. This database is also available on-line
to the public and can be searched using keywords.
Despite the comprehensive nature of our control over the holdings,
we are still concerned that the listings of some records are inadequate
to obtain access, unless the searcher knows very specifically
what he or she is looking for. There is scope for developing other
sorts of finding aids, including subject guides. We have issued
a guide to family history sources for genealogists and a guide
to lighthouses in Australia. Work is continuing on other subject
guides, and in 1993 we will be undertaking a joint pilot project
with the Australian Science Archives Project to test the feasibility
of producing a guide to the Australian Archives' holdings of science
and technology records. I suspect that this will be the most ambitious
subject guide yet attempted, and perhaps the most rewarding.
The project to reduce our unevaluated holdings will also contribute
to improving our knowledge of what is actually held. There are
many treasures waiting to see the light of day but, equally, much
rubbish that should be destroyed to improve the chances of survival
of those records with lasting value.
Information is also held in RINSE about personal records collections
held by the Archives. These include the records of some individual
scientists who have been closely involved with the Commonwealth
during at least part of their careers. Such collections provide
the Archives with the opportunity to retrieve many Commonwealth
records that have ended up mixed with the personal records of
individuals, while maintaining the integrity of those record collections.
There is scope for us to make sure researchers employed by the
Commonwealth on significant projects are aware of this service,
although it becomes more difficult to identify the most significant
players as research projects have become more and more team or
cooperative efforts between the Commonwealth and other organisations.
This leads me on to my final point, that the current science and
technology research environment within the Commonwealth which
encourages joint ventures with private organisations or other
governments will have a significant impact on aspects of the Archives'
work in the future. Who will own the records? We have been dealing
with this problem in another guise over the past few years, as
the Government has steadily privatised and corporatised a number
of its functions. We have requested a string of legal opinions
from the Attorney-General's Department to determine who owns the
records after privatisation has occurred. If records are somehow
removed from Commonwealth ownership, then the Archives has but
a slim chance of ensuring that a complete record of significant
Commonwealth activities is retained.
Similar confusion over records' ownership will arise from the
current crop of Cooperative Research Centres and joint ventures
unless specific steps are taken to determine who has responsibility
for the records and where they are to be retained. As with electronic
records, this should be done at the earliest opportunity so that
valuable records are not lost before they are identified and arrangements
made for their proper retention or disposal.
At the Archives we understand that the ways in which science and
technology are carried out, and the ways in which research is
communicated and recorded are changing. Only a dynamic and flexible
approach that includes the cooperation of researchers themselves
will ensure that the records of science are 'recovered', that
they remain in the public arena, and are accessible in the future.
1 R.W. Home, 'Science
on Service, 1939-1945', Australian Science in the Making,
ed. R.W. Home, Melbourne, 1988, pp. 220-251.
2 Denis Muller in the Age,
29 October, 1992, p. 10.
3 David Bearman, 'Electronic
Records of Science', Archives and Museum Informatics, vol.
6, no. 1, 1992, pp. 14-15.
4 Dagmar Parer, 'Towards Better
Management of Commonwealth Records', Informaa Quarterly,
vol. 8, no. 3, 1992, pp. 9-10.
5 Australian Archives Annual
Report 1991-92, Canberra, 1992, p.36.
6 Particularly Joan K. Haas,
Helen Willa Samuels and Barbara Trippel Simmons, Appraising
the Records of Modern Science and Technology: a Guide, Massachusetts,
1985; and Clark A. Elliott ed.,Understanding Progress as Process:
Documentation of the History of Post-War Science and Technology
in the United States: Final Report of the Joint Committee on Archives
of Science and Technology, Chicago, 1983.
7 National Academy of Public
Administration, The Archives of the Future: Archival Strategies
for the Treatment of Electronic Databases: a Study of Major Agencies
in the United States Government, Washington, 1991.