D-Lib Magazine
February 2000

Volume 6 Number 2

ISSN 1082-9873

Lynn Pritcher
Duke University
lynn.pritcher@duke.edu.

Introduction

Creating access to digital images of historic documents, images, and maps is a vital service that many libraries and archives are providing for their patrons, local and international. And with most on-line collections, web visitors are able to see the items while using enhanced searching and browsing capabilities to find them. This capability expands the usability of the on-line collections in ways traditional finding aids and box lists cannot match. There is a wealth of interesting and helpful material in most libraries -- and as long as items are truly historic (19th Century and before) and without literary or copyright claims, the process of creating on-line collections is straightforward. Find the items, scan them, provide access: more primary resources on the web (though the process is a little more complicated than that). Creating an on-line resource for more recent historical documents (i.e., created or published less than 78 years ago) adds another large layer of difficulty to the task.

The Digital Scriptorium at Duke University creates on-line projects for a variety of collections, ranging from ancient papyri to 20th-Century photography. For the most part, everything that has been made available by the Scriptorium has either been out of copyright or, in the case of the photos, the library owns the copyright to the items themselves.

The Ad*Access project, however, presented an entirely new challenge. Ad*Access, an on-line database, gives researchers around the world images and information for more than 7,000 advertisements printed primarily in U.S. newspapers and magazines between 1911 and 1955. It is the collaborative effort of the Digital Scriptorium and the John W. Hartman Center for Sales, Advertising & Marketing History. Generously supported by the Duke Endowment "Library 2000" Fund, Ad*Access was created using materials in the J. Walter Thompson Company Competitive Advertisements Collection (ex. 1918 Rigaud ad at right).

The vast majority of the advertisements chosen for this project were created and published after 1923. Clearly beyond the copyright-free zone, the library had to answer the question of whether the pursuit of copyright permission was necessary, as the project was created by an educational institution for research and study. In other words, did the project fall within fair use?

This article describes how that decision was made and the implications of that decision. The author includes citations from copyright law and other resources dealing with intellectual property and fair use. The author, however, is NOT a lawyer, and does not purport to offer legal advice in any way. The emphasis of the article is to describe our project's efforts in resolving the question of copyright for Ad*Access.

Copyright ownership of an item grants rights to that owner. Abridging copyright through the reproduction, display or performance of items without the consent of the owner may result in litigation. The fair use clause (Section 107), along with Section 108 of United States Copyright Law,(1) provide librarians and archivists with the basic parameters for fair use.

Many of the arguments surrounding the inclusion of, and amendments to, the fair use clause in U.S. copyright law dealt with maintaining a balance between authors/publishers and educators/librarians, i.e., between restriction on use and expansive access to resources. The advent of networked environments and the Internet has further complicated the understanding and implementation of fair use within libraries and educational institutions.

To a far greater extent than photocopying, the digital technologies have eroded traditional roles and boundaries among authors and users…making it imperative to seek a balance between access to images and protection of authors' and publishers' economic and moral rights. (2)

Granted, Duke is an educational institution. Furthermore, the reason for the project was to provide researchers around the world access to selected advertisements in a format designed for targeted study of various issues identified through current scholarship as important. However, placing this resource on the Internet raised a variety of questions that the courts had not yet addressed.

The experience of how other libraries have dealt with this issue was limited and only minimally helpful. The majority of collections on the web at the beginning of the project [fall 1997] were created with pre-1920 materials, or the institutions held copyright to the items. Consultations with librarians at other institutions and discussions with Duke's University Counsel persuaded us to pursue identification of the copyright holders for the advertisements before allowing access to the images via the World Wide Web. We suspected that had we not sought to identify current copyright owners, any subsequent legal challenge over copyright to the Ad*Access project by a company would most likely have lost in a court of law, as well as the court of public opinion. While we were unaware of any legal challenges to Internet educational sites, the decision was made to err on the side of caution, and the copyright permission identification process began.

Through discussions and research, a strategy was developed based on the assumption that the company represented in the ad was the copyright holder. We worked with the mandate to document all research and attempts to contact every company, i.e., to make a good faith effort. Thus, while our extensive research would not exempt us from copyright claims, there would be documentation that we attempted to contact companies to determine their possible copyright ownership.

Preliminary work, including selecting, scanning, and entering data information for each ad in a category, was necessary to identify the companies we would need to contact for inclusion in the project. Advertisements for the last two categories included in the project, Beauty & Hygiene and Transportation, were selected after we had gained some experience with the permission process. The difficulties encountered in finding copyright information affected the selection procedure for these two categories, leading to the decision to choose only those companies that had a large number of ads or were well known, recognizable names. (Table 1 below outlines the amount of advertisements in each category within the five decades covered by Ad*Access.) These criteria would (it was hoped) prevent our having to track down a multitude of companies in order to include only one or two ads.

Once the company, product, and date information were assembled, the process of researching the companies began. Databases accessible through Duke's electronic resources and on-line resources, such as Yahoo!, WebCrawler and SavvySearch, contained a great deal of the company information we needed.

Older print sources proved to be invaluable in locating companies we could not find in the databases or on-line. A number of company names were discovered by using contemporary sources to track corporate mergers, acquisitions, and buy-outs. Much of this work was done through foot-soldier researching -- following up on references in bibliographies and footnotes to additional materials.

Mergers, acquisitions and buy-outs were also the most challenging aspect of pursuing copyright permissions. We found throughout the copyright-pursuit process that companies that are in the business of acquisitions do not necessarily keep easily-accessible records of these transactions.

Railroads proved to be particularly troublesome in this area. In the early part of the twentieth century, there were numerous regional and local railroad companies, some owning only a mile or so of track. Eventually, railroads began to consolidate, buying up smaller companies and merging with larger companies. History clubs, successor companies, and other unexpected organizations own the copyright, trade and/or service marks of many now defunct railroads (e.g., Maytag owning rights to Rock Island Railroad). The few viable national railroad companies own most of the others. While researching these companies, Norfolk Southern and CSX Transportation were in the middle of negotiating a buy-out of parts of Conrail, causing a delay of over three months in our copyright pursuit.

Norfolk Southern alone is the product of more than 300 acquisitions, mergers, reorganizations, and consolidations dating back to the 1830s. However, they have no records relating to the disposition of the majority of those earlier companies. So our inquiring as to their ownership of certain railroad companies led them to state they were not likely to have exclusive rights to the marks of some railroads we were missing. They did not think that, even with the transaction with CSX, they would have the authority to give permission for use of some railroad ads. (The advertisement at left represents one railroad company for which we were unable to clarify current copyright ownership.)

Another example of the difficulties with buy-outs is the Stewart Warner Radio and Television Company. Bought and sold a few times, at the time at which we tried to find the company, Stewart was supposedly owned by BTR, a British conglomerate whose business is built on buying and selling companies, with a pretty quick turnaround if possible. The contact person at BTR had no recollection of Stewart Warner and stated he had no paperwork available about the company. We faxed them the pages from the print source we used (Capital Changes Reporter, 1993 edition), showing their acquisition of Stewart. BTR agreed to allow participation in the project, including a note that they gave permission only to what was actually theirs to give.

That raises the issue of whether the permission was the company's to give in the first place. The assumption underpinning our pursuit of copyright was based on the ads falling under the "works made for hire" concept. As described by a circular from the U.S. Copyright Office:

[w]hether or not a particular work is made for hire is determined by the relationship between the parties. This determination may be difficult, because the statutory definition of a work made for hire is complex and not always easily applied. (3)

This explains in part the lack of certainty displayed in the permission letters we received from many companies. A work made for hire is described as:

[1] a work prepared by an employee within the scope of his or her employment; or
[2] a work specially ordered or commissioned for use as a contribution to a collective work, as a part of a motion picture or other audiovisual work, as a translation, as a supplementary work, as a compilation, as an instructional text, as a test, as answer material for a test, or as an atlas, if the parties expressly agree in a written instrument signed by them that the work shall be considered a work made for hire. (4)

Litigation in 1989, involving the Community for Creative Non-Violence v. Reid decision, clarified further "works made for hire." The decision stated that unless a work is made by a formal employee or agent of the party commissioning the work and falls into any one of the categories listed above in [2], then the "work cannot be a work made for hire even if there is an agreement in writing stating that it is one." (5)

Advertising agencies work with artists, photographers, etc. in creating advertisements for particular companies. In the Advertising Compliance Handbook, advertisers are admonished to carefully word agreements with artists. It is noted, however, that "even if the artist has transferred the copyrights to his works to the advertiser, the artist may recapture his copyrights after thirty-five years, notwithstanding language in the agreement to transfer or assign the right to his work." (6)

Only one company contacted viewed "works made for hire" to mean that the permission for use of their advertisements would need to come from their current advertising agency. The ad agency was concerned that the ownership of some of the images could still be held by the models, photographers, and the advertising agencies responsible for the original ad. When provided with copies of all ads [only a few companies requested this], the agency approved the use of the ads. With a few other companies, we were asked to sign letters agreeing to indemnify them (7) against any future litigation arising out of our use of the ads. After consulting with University Counsel as to the risk of signing and agreeing to these terms, we were advised to sign the letters, keeping these and all other records of communication in a file.

Apart from the more legalistic concerns of some companies, a number of companies had reservations about participating in our project simply because these historic ads would now be available over the Internet. An example of this is represented by our interaction with a family-owned business that made radios in the 1920s. Initial discussions with the company archivist directed us to write a formal letter of request to the board. According to the archivist, the board -- made up of children of the original founder -- were reluctant to have any images of their products on the web, due to uncertainty about this new entity little known to them. A brief web search by this author revealed a half dozen sites with unauthorized images of their radios. We tactfully referred to the existence of these sites in our letter, to show that we were trying to be both respectful and legal. We received permission to include the company in our project.

Companies were also concerned about the potential for users of our site to copy and then digitally manipulate the advertisements. Each company was informed (through the copyright permissions letter) that the items were scanned at 150 dpi -- a good level for viewing but not high enough for quality printouts. The Digital Scriptorium placed a copyright statement in the HTML code at the top of each web page, in addition to a blanket copyright statement on each page with both database information and an image, to address these concerns. The fact remains that the advertisements in Ad*Access were printed in popular magazines and papers, and many are easily found today in a library, a flea market, or on e-bay. The images are available for people to use as they might, copyright notwithstanding.

Though not a copyright issue, there was the additional consideration of potentially negative reflections on the companies. For example, the presentation of women and minorities in advertising is vastly different today than 50+ years ago. The possibility exists, though slim, that when some companies see the entire run of their advertisements they may want to pull a few of the ads due to the imagery in the illustrations. One company asked for, and received, this option when initially contacted. The database information remains for any image removed since facts cannot be copyrighted, and the database simply lists factual information from each advertisement. Public and corporate reactions to the site have been overwhelmingly positive. The handful of companies who chose not to participate in the project will be notified of these responses, in hopes that they may decide to allow their images to be viewed.

Initiated in the spring of 1998, the pursuit of copyright permissions for Ad*Access did not get fully underway until the summer of 1998, for reasons noted above. Then, two forms of delayed responses from the contacted companies extended the time expected to obtain permissions. First, discovering and speaking with the person who could definitively answer our questions regarding the copyright ownership of historic Company X by Corporation Y took a good deal of time. Unanswered phone messages required re-calling individuals, only to learn in many instances that a different person at the company was the correct contact, requiring this action to be repeated. Secondly, once contacted, Corporation Y then had to process the request through the necessary channels on their end, thereby adding additional days or weeks to their response time. A six-month lag between initial company contact and resolution of the copyright question was not uncommon.

Through calculating the personnel (including over 440 student hours) and the phone, fax, mailing and copying costs, the amount per image for the copyright process was about $1.43. These figures represent per-ad costs for all 7,307 ads. The companies actually found and contacted, however, represent a good deal less than one-half of all companies included in the project. Therefore, this figure is somewhat deceptive. It is sufficient to say that the cost and time requirements of the copyright segment of Ad*Access were unknown at the beginning of the project, and the amount of hours finally needed for this one aspect was a surprise to all involved. However, the experience gained by the staff (and the students) is applicable to both electronic and print projects at the library, including one in progress now.

An alternative to the process we followed in searching out the companies would have been to check with the Copyright Office to see if any of the ads had been registered when first created. The copyright law of 1909, under which the ads in the project originally fell, required that a notice of copyright be affixed to each copy (or forfeit copyright), and that the item be registered with the Register of Copyrights (noncompliance possibly causing a fine or the voiding of copyright). (8) Every magazine and newspaper in which the ads were printed most likely carried a copyright notice; this notice, however, fails to cover the advertisements not originating from the magazine or newspaper itself. The Berne Convention Implementation Act, signed by the US, does not require a copyright notice to be affixed on anything created after March 1, 1989. This aspect of the law is not retroactive; therefore, lack of a copyright notice on any of the ads may mean lack of copyright. (9) The 1976 Copyright Law, however, noted that lack of a copyright notice on an item could be remedied through certain procedures. (10)

The fact is, the advertisements in Ad*Access may never have been covered under copyright. The items may have been viewed by their creators as ephemeral and not worth registering with the Copyright Office; therefore, the pursuit of permissions for this project may not have been entirely legally necessary. The goodwill and positive responses generated by our communications with the many companies, however, was worth the time and effort of the process. The Hartman Center interacts with hundreds of companies which use the Center's resources. Engendering a negative name within this community would be counterproductive. Additionally:

The two fundamental exclusive rights [of copyright], i.e., the reproduction right and the representation right, remain operative; yet the way in which they now interact and accumulate in a networked environment is undeniably uncharted legal territory.(11)

The development of copyright law in relation to the Internet and image archives is still that: developing. If and when the courts are presented with test cases in this area, the clarified understanding of fair use in connection with web-based archival collections will hopefully allow an increase in the exhibits of items still within copyright. Until that point, the experience gained from the Ad*Access project assists in the consideration of other on-line exhibits to be produced by the Digital Scriptorium at Duke University.

Notes

(1), (4) United States Copyright Office, Library of Congress, "Title 17, Chapter 1," Copyright Law of the United States, (http://www.loc.gov/copyright/title17/), Section 108; Section 101.

(2) Barbara Hoffman, "A Picture is Worth a Thousand Words," in Growing Pains: Adapting Copyright for Libraries, Education, and Society, ed. Laura N. Gasaway, (Littleton, Colorado: Fred B. Rothman & Co., 1997), 395-440.

(3) United States Copyright Office, Library of Congress, Circular 9: Works Made for Hire under the 1976 Copyright Act June 1999, (http://lcweb.loc.gov/copyright/circs/circ09.pdf).

(5), (6), (9) Kenneth A. Plevan and Miriam L. Siroky, Advertising Compliance Handbook. 2nd ed., (New York: Practising Law Institute, 1991), 351-381.

(7) An example of part of an indemnity letter: The Library agrees to indemnify and hold harmless Company X and its employees, agents, representatives, associates, affiliates, predecessors, successors and assigns, parent and subsidiary corporations, and its officers and directors and each and all of them from and against any claims and expenses (including reasonable attorney's fees) arising out of, or in connection with, the Library's use of the advertisements…

(8), (10) Arlene Bielefield and Lawrence Cheeseman, Technology and Copyright Law: A Guidebook for the Library, Research, and Teaching Professions, (New York: Neal-Schuman Publishers, Inc., 1997), 23; 25.

(11) Frédéric Pollaud-Dulan, ed., The Internet and Authors' Rights. Perspectives on Intellectual Property, Volume 5. (London: Sweet & Maxwell Limited, 1999), vi.

Additional Sources

G. Peter Albert, Jr. and Laff, Whitesel and Saret, Ltd., Intellectual Property Law in Cyberspace, (Washington, D.C.: The Bureau of National Affairs, Inc., 1999).

Ronald V. Bettig, Copyrighting Culture: The Political Economy of Intellectual Property, (Boulder, Colorado: Westview Press, 1996).

Henry M. Gladney, "Digital Dilemma: Intellectual Property," D-Lib Magazine, December 1999, (http://www.dlib.org/dlib/december99/12gladney.html).

P. Bernt Hugenholtz, ed., The Future of Copyright in a Digital Environment, Information Law Series 4, (The Hague: Kluwer Law International, 1996).

William F. Patry, The Fair Use Privilege in Copyright Law, 2nd. ed., (Washington, D.C.: The Bureau of National Affairs, 1995).

Copyright © 2000 Lynn Pritcher

D-Lib Magazine Access Terms and Conditions

DOI: 10.1045/february2000-pritcher