català | castellano | english home   sitemap   avís legal   crèdits   contacte  
home home

R. Joseph Anderson, Center for History of Physics, American Institute of Physics

"Difficult to Document: Physics in Government and Industry"

The Center for History of Physics at the American Institute of Physics uses documentation research to develop collaborative strategies to identify and appraise historical records and papers in areas of science that are traditionally difficult for archivists and historians  to address and that, as a result, are often overlooked and eventually lost. The concept of archival documentation  research and documentation strategies was conceived at the AIP History Center and at a few other U.S. archives twenty years ago, and it remains an essential tool in work to preserve the history of modern physics, astronomy, geophysics, and related fields.[1]

Documentation strategies are planning tools that allow archivists to work together with their colleagues at other repositories, and with records creators and records users to develop coordinated, cooperative approaches to documenting the past.  They are especially useful in addressing under-documented and hard to document areas, which is often the case in the history of physics and other sciences.  The underlying principles are:

- Good knowledge of the universe of documentation that’s potentially available is needed before decisions about collecting can be made.

- Planning and analysis should precede efforts to collect records and papers

- Modern documentation is linked across institutional lines, so institutions must cooperate together to be effective in documenting shared areas of interest.

- All the stakeholders in archival records—the people who create them, the researchers who use, and the archivists who preserve them–should work together to decide what can and should be preserved and to develop appraisal guidelines.

A completed documentation strategy specifies who all are responsible for preserving records and papers, identifies by type the small percentage of records that are likely to be of permanent historical value, and describes institutional changes that are needed to implement the strategy.  It also provides for a means of sharing information on the records that are preserved.

The History Center completed its most ambitious documentation research project, a ten-year study of multi-institutional collaborations, in 2001, and we have recently started a new study to investigate how physicists at corporate R&D laboratories in the U.S. create and use records in the course of their research.  The Project to Document the History of Physicists in Industry entails field work at IBM, Xerox, General Electric, and 12 other major employers of physicists in industry, including interviews with approximately 100 bench scientists and R&D managers, meetings and interviews with information managers (e.g., archivists, heads of technical libraries, records managers), records surveys, and visits to major public and private repositories in the U.S. and Europe that preserve industrial records.  The findings will provide concrete guidelines and recommendations for preserving historically valuable R&D records.

The Center for History of Physics is a division of the American Institute of Physics, which, itself, is an umbrella organization representing the ten largest American professional societies in physics, astronomy, geophysics, and related fields. The Institute’s primary role is as a journal publisher, but it provides a wide variety of other services as well.  The AIP History Center’s role is to serve as a discipline center for the scientific fields covered by AIP, to act as the archives for the Institute and the Member Societies, and to try to insure that other science organizations are preserving the historically valuable records of their scientists and programs in physics and allied sciences.  Our goals in doing documentation research are to identify and develop appraisal criteria for records and papers of permanent historical value, to publish and advocate realistic strategies for preserving them, and to act as an advocate and support source in implementing the strategies.

To explain why documentation research is such an important part of the work that we do at the Center for History of Physics, it’s worthwhile to look briefly at the origins of modern science archives in the United States.  The history of modern science emerged as a viable separate field in the U.S. in the mid 1950s. [2]  At the time there were almost no archival sources for the history of modern physics and others sciences.  For example, in 1960 there was only one collection of personal papers of a 20th century physicist in an American archives, the papers of Enrico Fermi at the University of Chicago.  So the pioneering historians and archivists involved in this new field were concerned about developing archival sources.  At about the same time, some leading physicists began to express concern that the first generation of modern physicists were passing from the scene and that their memories and recollections were being lost with them.  Then as a precipitating factor, in 1958 the National Science Foundation created an office of social science to support projects in history and related disciplines.[3]

By 1961 three history of science undertakings, a national conference and two projects, had gotten underway in the United States with NSF funding. The conference, called ‘The Conference on Science Manuscripts,’ was held in Washington D.C. in 1960 and was concerned with documenting the history of science as a whole.  The two projects addressed the history of physics specifically.  The first project was organized in 1961 by my parent organization, the American Institute of Physics, and it represents the origins of the Center for History of Physics, which was created in its current form in 1965.  The other project was undertaken as a joint venture by the American Physical Society and the American Philosophical Society, and it created the Archives for the History of Quantum Physics, which remains one of the most valuable resources in the history of science and is available on microfilm at depository libraries in Asia, Australia, Europe, the Middle East, and North America.[4]

The papers presented at the ‘The Conference on Science Manuscripts’ were published in ISIS, and the striking thing, reading the articles 40 years later, is how modern they are in tone and outlook.  Until fairly recently in the U.S., archival and manuscript collecting usually consisted of one or a few repositories staking a claim to a historical period or subject area, and acquiring all the materials that were available.  Most archivists at the time put a premium on custodial work and collecting.  Competition among repositories, albeit usually low key, was common, and active cooperation was rare or non-existent.  While cooperation may still be more often practiced in theory than in reality, it represents an ideal today that many archivists would subscribe to.  That wasn’t true forty years ago, but the conference speakers recommended a distinctly cooperative agenda for documenting the history of science.  For example, they argued for creating a network of collecting repositories, including academic archives, public manuscript libraries like the Library of Congress, private collections like the American Philosophical Society, and the U.S. National Archives to take responsibility for this new mission of documenting the history of science.  The participants also specified that the network of repositories would need to share information on their holdings.  The conference speakers, themselves, represented a mix of records creators, records users, and those who preserve records - that, is scientists, historians, and archivists - and they agreed that all of these stakeholders had a role to play in preserving the history of the field.  And at least one speaker suggested that professional science organizations should also get into the act, serving as advocates for the preservation of the records and papers of modern science and coordinators of the work that needed to be accomplished.[5]

The two projects in the history of physics that I’ve mentioned developed separately from the conference, but they both put into practice the principles of cooperation that the Conference advocated.  I think the reason for this almost unique emphasis forty years ago on archival cooperation stems from the nature of the field that they were working to document and from the large number of physicists who played an active role in the early efforts, and especially in initiating the two projects.  Scientists are often competitive, of course, but by 1960 much important science was conducted using cooperative strategies and information sharing.  So these attributes were grafted onto efforts to document the history of science in the U.S. from the beginning.

Cooperation is an admirable goal, but it took the AIP History Center some time to figure out how to actually implement cooperation as an effective preservation strategy, and it still remains to be worked out in a number of hard-to-document areas.  The AIP History Center’s first documentation strategy involved learning how to work with scientists and archivists at colleges and universities.  Fortunately, the Center got underway when the prestige of physics on college campuses was especially high, which means that it was easier to make a case for preserving the history of physics.  It was also at a time when many American universities were developing professional archives programs for the first time.  One of the Center’s early actions was to develop a list of the most productive American physicists, most of whom were at universities.  We then began to contact university archivists about preserving the papers of important scientists at their institutions.  By the1970s we had started a regular program of contacting academic archivists when important physicists died or when their papers became available for other reasons.

Because the archival records of physics and allied fields like astronomy and geophysics are too vast for any one organization to preserve, learning how to work with college archivists and to persuade them to preserve the records of their science programs and the papers of important scientists was critical for us.  I should add that the History Century has no effective alternative to cooperation and persuasion in order to accomplish our work, since we don’t have the facilities to house archival collections much beyond the records of the American Institute of Physics and its ten Member Societies. Our own collecting is restricted—for both philosophical and practical reasons - to AIP and Member Society records, oral history interviews, photographs, ephemera, print material, and the occasional orphaned collection of personal papers or institutional records that we can’t find another home for.  The advantage of this necessity is that we won't be tempted - or accused - of competing with the repositories with whom we cooperate.

Over time the Center’s staff became more assured in carrying out cooperative work to preserve the papers of academic scientists, and they became active in a variety of other areas as well.  One such area was to expand its work to share information on collections, first in the United States, and by the late 1960s in Europe by developing a union catalog of manuscript collections in our field.  The catalog - called the International Catalog of Sources for the History of Physics and Allied Sciences or ICOS - has been online free-of-charge since 1997 (, and it contains approximately 7,400 catalog records for personal papers, archival collections, oral history interviews and other primary resources from about 500 repositories in the U.S. and abroad.  We also expanded our work beyond physics to include the related fields of astronomy and geophysics early on.

A major turning point in the Center’s work came in the 1980s when we contracted with the U.S. Department of Energy (DOE) to study and create recommendations for managing and preserving records at the national accelerator labs that DOE manages.  This entailed extensive field work at Brookhaven, Argonne, Lawrence Berkeley and other major federal contract laboratories.  The project, which extended over several years and produced detailed reports and recommendations, got us involved directly for the first time with the challenges of trying to document big science at government labs, and we found that it presented a whole new set of problems.  Very briefly, the records retention schedules devised by the U.S. National Archives and Records Administration, which is responsible for scheduling and preserving records at the DOE labs, focused on the records created by the top administrators in the labs and largely ignored the records generated by scientists.  These, of course, are the kinds of materials that are the most important for documenting the history of science, and these records were being routinely destroyed.  At the end of the project we:

published a handbook for secretaries on how to file laboratory records and how to work with records managers, identified core laboratory records that should be preserved permanently, persuaded one laboratory, Lawrence Berkeley, to hire an archivist to work as a liaison with the National Archives,  and recommended that the National Archives radically revise its appraisal and scheduling procedures for DOE laboratory records.[6]

During the course of the project, History Center staff began to realize that a special organizational approach to high energy physics had developed at the DOE laboratories in the United States and at CERN in Europe.  In order to conduct major experiments in high energy physics, scientists from many different institutions, including a variety of universities and government labs in the U.S. and abroad, and occasionally corporate labs as well, were joining together in temporary collaborations that might last a few years or more.  Their work represented cutting edge science, but these multi-institutional collaborations were largely ignored in the historical literature of the time, and they presented special problems for archivists concerned with the records of modern science because of the large number of participants from different organizations and the lack of a permanent institutional home.  Once the experiment ended, the records were dispersed or destroyed - if they hadn’t already been lost along the way - and the only remains would be the published literature that the collaboration produced.

The AIP History Center decided that the problems presented by preserving the history of multi-institutional collaborations were too important to ignore, and in the late 1980s we began planning a systematic documentation research project.  What was to become a three phase, ten-year project got underway in 1989 with funding from the National Science Foundation, the Andrew Mellon Foundation, and other sources.  During the first three-year phase of the project, we conducted several hundred interviews with scientists in Europe and America who were members of collaborations in high energy physics.  In the two subsequent phases we extended the work to multi-institutional collaborations in space science, geophysics, and five smaller fields.  Overall, during the ten-year study we:

- Conducted 650 interviews with scientists in selected collaborations and made scores of site visits in the U.S. and abroad.  We also met with many records managers and archivists.

- Published ten reports that outline strategies for identifying and preserving historically valuable records produced by multi-institutional collaborations.  The reports list the most valuable types of records that collaborations create and include other recommendations for preserving them.  To briefly highlight the recommendations:

a.  Collaborative projects should appoint a team member who is responsible for coordinating collaboration-wide records keeping.

b.  The U.S. National Archives should redraw their R&D record schedules to reflect modern big science and especially to recognize collaborative structures.

c.  Government agencies that fund the collaborations should support a marginal increase in their contract funds to support academic archives and allow them to collect records of collaborations.

d.  Government science agencies like the National Aeronautics and Space Administration and the Department of Energy should develop adequate records management programs to identify and preserve their own records

e.  Academic archives should save the papers of faculty who lead or participate in major collaborations, and expand their collecting policies to include records of collaborations housed within the university.[7]

Perhaps the three most important consequences of the Department of Energy records project and the Study of Multi-Institutional Collaborations, along with other collateral work, are:

The U.S. National Archives suspended its original and inadequate schedule for research and development records in the early 1990s, and it has since been working with individual government science agencies to design and implement new R&D schedules that will preserve historically valuable records of individual scientists and of collaborations.  The first agency to adopt a new R&D schedule was the Department of Energy in 1998, and others are following suit. Some of the DOE national labs now have archivists on staff and all have improved their records management programs.
The need to include the records of big science at government labs and multi-institutional collaborations is widely recognized today, and is a regular theme at archival meetings and other gatherings in the U.S.

These changes are cumulative over time, and they’re not attributable to the work of the History Center alone.  However, our documentation research work has provided a knowledge base in this new area of archives, suggested strategies for attacking some of the major issues, and has acted as a means of calling attention to and addressing some of the most important problems that science archivists face.  And the Center for History of Physics continues to be an active advocate for implementing new strategies for preserving the records of modern physics and allied fields.

In the same way that the Center’s work at the Department of Energy national labs in the 1980s led to recognizing the problems entailed in preserving the records of multi-institutional collaborations, the collabs study - where we found industrial partners in many of the teams we studied - made us more aware of the problems of preserving records at industrial R&D laboratories.  For the past hundred years America’s economy has been characterized by sustained innovation and technological change based on increasingly sophisticated industrial research and development.  Today about one-third of the Ph.D. physicists in the U.S. are employed in industry.  However, very few programs are endeavoring to document the research that corporate physicists accomplish.  The lack of attention that this very important area of science has received and the dearth of significant and readily accessible documentary sources has limited the ability of historians, other scholars, and policy makers to explore one of the most vital and productive sectors of America’s industrial economy.  The absence of preservation strategies also means that scientists often don’t have access to their own earlier work and that of their colleagues.  Finally it means that corporate managers and policy makers can’t profit from their companies usable past.

Up till now the bulk of the History Center's contacts have been with academic and government scientists, and in these areas archival programs exist that are intended to cover nearly all academic and federal government institutions.  Our challenge has been to get academic archives and the National Archives and Records Administration to focus on and actively collect records that document the history of science.  This remains a big challenge, especially in government labs and in academic institutions with weak archival programs, but at least we usually have an archival infrastructure to work with.  In contrast, most companies in the U.S. do not maintain their own in-house archives.  What’s more, it’s very rare for academic and public archives to collect and preserve the history of industry in the United States.    A recent national symposium on ‘The Records of American Business’ concluded that comparatively little is known about American industrial institutions because ‘there is very little documentation with which to work.’  And documentation is even more scarce for industrial R&D than for industry as a whole.[8]  So industry presents a new set of problems for us, and the History Center has recently started a three-year project to create a framework for identifying and preserving the records of physicists in industry on an ongoing basis.

Our new Project to Document the History of Physicists in Industry represents the first systematic investigation of records-keeping practices and needs in America’s high-technology industries.  The project will use some of the methods that we developed in the Center's earlier projects, but industrial R&D represents a new set of problems for archivists and requires new approaches as well.  Key activities include 1) records surveys and questionnaire-based interviews with records creators, records users and information professionals at 15 of the largest companies which employ physicists in American industry; 2) longer career-length oral history interviews with 15-20 leading industrial physicists; 3) identification and cataloging of extant corporate records, laboratory notebooks and other sources; and 4) a study of existing public and private archival programs that document industry in the U.S. In addition we will study some of the major European industrial archives.  It appears that more effective programs for preserving business records exist in the United Kingdom, Denmark, and some other European countries, although we don’t yet know if these programs cover research and development in addition to other corporate operations.  Overall, however, European archival practices, especially British models, may offer potentially valuable insights for documenting American high-tech industry.  The tradition among public and academic archives in the United Kingdom was similar to the U.S. pattern of neglect until the 1960s, but since then several developments, especially a newly energetic Business Archives Council, have significantly improved conditions.  The Council, a registered charity, has promoted and helped to underwrite national surveys of corporate records, and it has worked to place records in local and regional repositories and at the Business Records Centre at the University of Glasgow and the Modern Records Centre at the University of Warwick, collections that specialize in corporate records. [9]

In choosing the 15 companies that we plan to include in the study, we selected a judgment sample from the 27 largest employers of Ph.D. physicists in industry.  The 27 companies contain the largest concentration of our target population, employing about half of all physicists who work for American corporations, and they typically include at least some basic research within their R&D operations.[10] Basic research is of interest for us because it represents the most advanced physics.  We used company size (concentrating predominantly but not exclusively on the largest corporations among the 27), representation of all industry sectors, a variety of product mixes, ownership structure, and other criteria to select the 15 companies that we will invite to participate in the study, and we intentionally included a few companies that have existing archives programs.  The 15 companies are IBM, Xerox, Lucent Technologies, Raytheon, Lockheed Martin, Boeing, Exxon Mobil, Honeywell, Eastman Kodak, Corning, General Atomics, 3M, General Electric, Hewlett-Packard, and Texas Instruments.  We believe that our findings will apply generally to the records of advanced corporate research beyond the  companies participating in the study.  However, if the first three-year project raises important new questions, we may consider additional studies.       

The project got underway in November 2002, field work began in March 2003, and we have now completed interviews at four companies.  While it is too early too draw any concrete findings, the initial interviews have been informative and the interview subjects have been frank and forthcoming.  One significant change that we’ve found is the apparent breakdown in the way that scientists in corporate labs document their research.  Physicists at three of the four companies that we’ve studied so far report that they don’t use lab notebooks, either paper or electronic, any longer.  In the past formal notebooks were a common feature of industrial labs, and it appears that computers may have helped to eliminate paper notebooks without providing a viable electronic alternative.  It’s interesting to note that the decline in the maintenance of lab notebooks was identified as a problem in the investigation of the 2002 Schön incident, in which it was found that a young Lucent Technologies Bell Labs physicist, J.H. Schön, had intentionally published fraudulent findings.[11]  In another development, some of the records managers that we’ve talked with say that PowerPoint presentations, along with traditional technical reports, are becoming an effective form of documenting research.  And we have also seen a new emphasis on instituting effective records management programs in one of the four laboratories that we’ve visited.  In the later stages of the project we will analyze all the interviews and records surveys to determine patterns of records creation, use, and retention in high-tech industry, and we will evaluate current records-keeping practices for both paper and electronic records.  We'll also visit and evaluate programs at major non-profit and public American and European repositories that document the history of industrial R&D.

The end products of the project will be published recommendations and appraisal guidelines that provide strategies for documenting industrial R&D, along with oral history interviews and records surveys that will be made available to researchers. The recommendations will be based on our findings, including the needs of users and an evaluation of best practices in the United States and abroad.  Being realistic, we don't expect to convince companies that don't have archival programs to start them.  Instead, we intend to emphasize cost-effective approaches that incorporate existing records-keeping programs (e.g., records management programs, technical libraries) and public-private partnerships. The recommendations will also reflect the new opportunities provided by maturing electronic records systems and increasingly stringent contractual requirements for documenting corporate contracts in the U.S.  As remains true of our earlier documentation research in academic institutions and government laboratories, the effects of the Project to Document the History of Physicists in Industry will be cumulative.  By bringing attention to poorly documented areas in the history of physics and allied sciences, identifying specific problems, and developing recommendations and collaborative strategies for addressing the problems, the AIP History Center is continuing to expand the knowledge base that scientists, archivists, and historians can use to preserve the full history of scientific research. 

[1] One of the first articles describing the concept of documentation strategies was coauthored by Joan Warnow-Blewett, then the AIP History Center’s Associate Director, and described how it had developed and was used by the Center:  Larry J. Hackman and Joan Warnow-Blewett, ‘The Documentation Strategy Process:  A Model and a Case Study, 12-47, American Archivist, Vol. 50, No. 1 (Winter 1987), 12-47.   Warnow-Blewett has published extensively on documentation research; see for example Joan Warnow-Blewett, ‘Documenting Recent Science:  Progress and Needs,’ 267-298, Osiris, Vol. 7 (1992).

[2] Preface to ‘Constructing Knowledge in the History of Science,’ ed. by Arnold Thackray, vii, Osiris, Vol 10, 1995.

[3] George T. Mazuzan, ‘The National Science Foundation:  A Brief History,’ File:  nsf8816, July 15, 1994 (available on the Web at

[4] Warnow-Blewett, 267-270, Spencer Weart, ‘Preserving the Heritage of Discovery,’ Physics Today, vol 55, no. 1 (January 2002), 28-29

[5] ‘The Conference on Science Manuscripts,’ Isis, vol. 53, part 1, no. 171 (March 1962).  See especially remarks of Ernst Posner in a discussion session, 50-51, and Luther Evans, ‘Some Proposals for Action,’ 101-105.

[6] The reports produced by the DOE study listed below are available free from the AIP History Center:  Joan Warnow, et al., A Study of Preservation of Documents at Department of Energy Laboratories, 33 pp., January 1982. Joan Warnow and the AIP Advisory Committee on the Documentation of Postwar Science, Guidelines for Records Appraisal at Major Research Facilities; Selection of Permanent Records of DOE Laboratories: Institutional Management and Policy, and Physics Research, 31 pp., 1982, revised 1985.  Jane Wolff, Files Maintenance and Records Disposition: A Handbook for Secretaries at Department of Energy Laboratories, 21 pp., 1982, revised 1985.

[7] The AIP Study of Multi-Institutional Collaborations published reports at the completion of each phase of the study:  Phase I:  High Energy Physics, 4 reports, 1993; Phase II:  Space Science and Geophysics, 2 reports, 1995; and Phase III:  Ground-Based Astronomy, Materials Science, Heavy-Ion and Nuclear Physics, Medical Physics, and Computer-Mediated Collaborations, 1999.  In addition final reports consisting of Highlights and Project Recommendations and the more complete Documenting Multi-Institutional Collaborations were published in 2001.  All 10 reports are available online at and are also available free in paper.

[8] Elizabeth W. Adkins, ‘The Development of Business Archives in the United States: An Overview and a Personal Perspective,’ The American Archivist, Vol. 60, No. 1 (Winter 1997), 10-11; Francis X. Blouin, Jr., introduction to James O’Toole, ed., The Records of American Business (Society of American Archivists, 1997), 1.  Bruce H. Bruemmer and Sheldon Hochheiser, The High-Technology Company; A Historical Research and Archival Guide (Charles Babbage Institute, 1989), 1-6, point out that ‘the inadequacy of information about business, its structure, and its documentation is a common theme in recent archival literature . . .’ and add that ‘the large, high-technology company is among the most difficult of institutions to document.’

[9] Michael S. Moss and Lesley M. Richmond, ‘Business Records:  The Prospect from the Global Village,’ in James M. O’Toole, ed., The Records of American Business (Chicago:  Society of American Archivists, 1997): 369-390, and Henrik Fode and Jørgen Fink, ‘The Business Records of a Nation:  The Case of Denmark,’ American Archivist 60 (Winter 1997): 72-86.

[10] The 27 companies were identified as the leading corporate employers of physicists in the past three AIP Education and Employment Statistics Division’s biannual surveys of the ten AIP Member Societies.

[11] A former Bell Labs (Lucent Technologies) director reported that Schön’s fabricated publications had gone undetected initially in part because the once ‘ubiquitous lab notebook’ system that had been used to document laboratory research had apparently been ‘forgotten or abandoned.’  Barbara Goss Levi, ‘Investigation Finds that One Lucent Physicist Engaged in Scientific Misconduct,’ Physics Today, Vol. 55, No. 11 (November 2002), 17.