Articles on Law, Science, and Engineering

Publish and Perish

By Charles Walter and Edward P. Richards

Note: This article was originally published  in Volume 21 Number 4 IEEE Engineering in Medicine and Biology Magazine July/August 2002 at pg 127 in the "Engineering and the Law" series. After 18 years in the magazine, this is the last article in the E&L series appearing in IEEE/EMBM. The authors discontinued contributing articles when the publisher required assignment of their copyright as a condition for publishing further articles. Dr. Walter continues to own his copyrights for all of his articles in the E&L series, therefore access to this and his other articles from the E&L series are available to all free of charge on this website.

            Most of us are familiar with the pressures of "publish or perish." So what do you do if the journal in which you wish to publish your scientific articles tells you it won't publish your article unless you give it something extra of value?

            We are all familiar with the practice of "page charges" and the resulting disclaimer at the bottom of the title page identifying some research articles as paid advertisements. Two other charges which authors are now sometimes expected to pay to have their articles published are "warranties and indemnification" and "copyright transfer". Like page charges, these charges involve giving your publisher something of value as a condition of publication, but, unlike page charges, they cannot be funded by your research grant. They could come out of your own pocket, and they could be very expensive.

Warranties and Indemnification

            A number of publishers have begun requiring a laundry list of warranties and even transfers of valuable copyrights before publishing research articles. In addition, authors may have to indemnify and hold the publisher "harmless" from any damage or expense that may arise from a breach or alleged breach of any of the warranties. In addition to a warranty that the party submitting the article is the author of the submitted article, journals may require warranties that the article does not violate any proprietary right, any governmental disclosure restrictions, the Digital Millennium Copyright Act (DMCA), and the like.

            How can authors obtain the legal assistance necessary to understand what these warranties mean? Even then, how can they know for certain that someone will not claim that an article violates their proprietary rights? How can authors know that their article doesn't unlawfully disclose information that would aid Cuban farmers to raise sugarcane? How can authors know what the DMCA is, much less whether someone might claim that their article violate it?

            These are very complex legal questions. Paying a lawyer for opinions is expensive, and it probably will not even provide a definite answer to the bottom line question, "Should I make the warranty?" For example, the DMCA was enacted in 1998 to implement the World Intellectual Property Organization (WIPO) Copyright Treaty, the Performances and Phonograms (PAP) Treaty, and for other purposes. Title I contains five sections with numerous subsections that implement the WIPO Treaty. A sample of the complexity is the amended meaning of a "work made for hire" which now includes

            "The terms 'WTO Agreement' and 'WTO member country' have the meanings given those terms in paragraphs (9) and (10), respectively, of section 2 of the Uruguay Round Agreements Act."

            Title II of DMCA contains three sections addressing online copyright infringement. Title III contains two sections addressing computer maintenance or the repair copyright exemption. Title IV contains seven miscellaneous provisions. Title V contains five sections addressing protection of certain designs. A printout of DMCA from a commercial database is about fifty pages. Searching the Federal statutes in the same database for "export control" yields 263 hits. Intellectual property attorneys qualified to opine about your bottom line question charge $250 to $500 per hour. So analyzing the DMCA or these statutes in light of the factual nature of your specific research article is going to be prohibitively expensive.

            Thus, unless you are willing to pay an intellectual property attorney ten or twenty grand for a written opinion regarding these warranties, an author cannot possible know (a) what s/he is warranting, or (b) whether the warranty is true. Even then, you probably won't get an answer that protects you from lawsuits. Some of these publishing agreements contain promises that the author must indemnify and hold the publisher harmless, which means that the author must pay all expenses that arise from legal actions concerning the article, from publisher's attorney's fees to the cost of the staff time of the publisher's employees involed in the litigation.  Authors who sign such an agreement expose themselves to enormous litigation expenses which they may not recover even when they or the publisher wins the lawsuit. If they lose, or the publisher settles out of court, they might end up with debts that they cannot pay off in a lifetime. Since it makes no sense to risk your financial future on something you don't understand, the common sense solution is to either find a different publisher or just not publish the article at all.

            Of course, publishers have legitimate concerns about publishing articles that violate the law. In the case of DMCA or export control violations, publishers may have to defend lawsuits claiming that they have assisted the circumvention of DVD safeguards or Cuban farmers growing sugar cane. But publishers have survived for many years even though they faced potential lawsuits for claiming libel, defamation, negligence, obscenity, and a long list of other unlawful acts. Publishers defend these lawsuits and usually they win on grounds that in this country freedom of speech is one of our most fundamental liberties. The difference here is that, instead of absorbing these inherent costs of doing business itself, the publisher now seeks to shift this cost to you, the author, by requiring your warranties and promises of indemnification.

            One consequence of this behavior is that publishers put themselves in the embarrassing position of stifling First Amendment rights. As described above, rather than be forced to take on legal expense and exposure they cannot afford or control, authors will not publish articles. This result is exactly the opposite of the constitutional and societal goal of advancing science, and it chills free speech.

            The bottom line on making warranties or promising indemnity is "Don't." As described above, this is especially true if you don't know exactly what you are warranting. But, if you are a university faculty or staff member, it is also true even with things as seeming simple and straightforward as warranting that you own the copyright of an article you wrote, or transferring title to the copyright to the publisher. As described in the next section, you may be warranting ownership of property you don't even own.

Copyright Transfer

Copyrights Are Valuable Property

            As mentioned above, a number of publishers have begun requiring transfers of valuable copyrights as a condition of publishing research articles.

            Question: "Why does a journal need all rights to the copyright to publish your article?"

            Answer: "It doesn't."

All it needs to publish your article is the right to do so as a contribution to the collective works in the journal (Section 201c Contribution). Since authors impliedly extend that Section 201c Contribution right when they submit a manuscript for publication, the journal doesn't need anything from the copyright owner to publish your article. It clearly doesn't need the copyright.

            Question: "If the journal wants the copyright anyway, why would it expect you to transfer your valuable property to it without fair compensation?"

            Answer: "Because it can."

A copyright is valuable property. It's part of our Constitution, with its purpose spelled out in Article I, Section 8:

            "To promote the progress of science ... by securing for limited times to authors ... the exclusive right to their ... writings...."

Requiring authors to give up their copyrights without compensation defeats our constitutional goal of "promoting the progress of science" by taking our constitutional incentive away from authors.

            In fact, the ownership of an author's original rights to his or her copyright are so highly regarded in our laws that a copyright is one of the few form of property that cannot be involuntarily transferred from its owner by any governmental body or other official or organization (except as provided in bankruptcy laws). In passing the statute, the U.S. House of Representatives said,

            "The purpose of this subsection is to reaffirm the basic principle that United States copyright of an individual author shall be secured to that author, and cannot be taken away by any involuntary transfer." H.R.Rep.No. 1476, 94th Cong., 2nd Sess. 123 (1976).

That means, for example, that even in a community property state a divorce court cannot award any interest to an original copyright to the spouse of the author.  (Although the court can allow the spouse to share in license fees generated by the copyright.)

            The actual value of a copyright is its market value, the price a willing buyer pays a willing seller. The market value of a research article copyright might not be apparent, but, clearly, it must be worth something to the publisher, else why would it pay its lawyers to draft the proposed transfer agreement?

            It is, of course, unseemly for a publisher to exploit its unfair advantage over academics and others who contribute articles in this "publish or perish" environment to extract something of value from them without paying them for it, just because it can. In fact, it is unfair. Given the relative positions of a publisher (which may even be an arm of a learned scientific society) versus the researchers who, for no fee, submit important articles so that their colleagues can share the results of their research (some of whom may even be members of the same learned society), it is clearly contrary to the interests of advancing science if the publisher attempts to extract something of value as a condition of publication. As described in the previous section, it also puts publishers in the untenable position of chilling free speech rights by discouraging authors from publishing articles or making publication prohibitively expensive.

Who Owns the Copyright?

            Obviously, you can't transfer ownership of property that you don't own. So the copyright transfer requirements imposed on university staff and faculty who submit research papers raises important and sometimes very complex questions of copyright ownership. Unless the university faculty or staff member who writes the paper owns the copyright, s/he cannot transfer it, and any warranty of copyright ownership and/or purported transfer (for compensation or without) might result in substantial financial exposure in litigation.

            Ordinarily, the owner of the original copyright in an article is the author who created the work. But Section 201(b) of the 1976 Copyright Act creates an important exception to this rule:

            "In the case of a work made for hire, the employer or other person for whom the work is prepared is considered the author ... and, unless the parties have expressly agreed otherwise in a written instrument signed by them, own all of the rights comprised in the copyright."

 

            In other words, if your article is a work made for hire, your employer owns the copyright unless you have an express, written agreement otherwise, signed by you and your employer.

            The 1976 Copyright Act defines a work made for hire by an employee as:

            "a work prepared by an employee within the scope of his or her employment."

            In a case called "Community for Creative Non-Violence v. Reid" [109 S.Ct. 2166 (1989)] the U.S. Supreme Court set forth the case-by-case factors for identifying who is an "employee" in the context of a work made for hire. These factors clearly identify university faculty and staff as "employees."

            So in order to decide whether a scientific research article is a "work made for hire," it is necessary only to ascertain whether or not it was prepared within the scope of employment.

            Some academic authors believe that there is a "teaching exception" to the commonlaw work made for hire doctrine. The existence of this so-called "exception" is based on the argument that since teachers are not required to reduce their lectures to notes, lecture notes are not prepared "within the scope of their employment" and therefore not prepared by an "employee," or not prepared "within the scope of employment" as that phrase is used in Section 201(b). In either case, lecture notes would not be a "work made for hire" as defined in Section 201(b). But if, in fact, lecture notes are not prepared by an "employee" or "within the scope of employment," they are not a "work made for hire" by definition, and the "teaching exception" is not really an exception at all. Believe it or not, but this is important, at least as far as lecture note copyrights are concerned!

            Section 301 of the 1976 Copyright Act specifically preempts rights in any copyright based on the common law or state law, and it doesn't mention any work made for hire exception for teachers. Therefore, if the "teaching exception" really was an exception to the commonlaw copyright meaning of a work made for hire, it was preempted by Section 301. Therefore, many intellectual property attorneys believe there is no "teaching exception" to the work made for hire doctrine in existence today, and that lecture note copyrights must belong to the employer. They are incorrect.

            This is because, if the "teaching exception" wasn't really an "exception," but stood on its own merits because the lecture notes are not part of the job, the notes were not created by an "employee" as that term is used in the definition of "work made for hire." Therefore, some intellectual property attorneys believe there never was a "teaching exception." If creating teaching notes are not part of a teacher's job, the copyrights belong to the teacher.

            But, either way, the argument for finding whether lecture notes might belong to a teacher wouldn't ordinarily apply to research articles written by university faculty and staff, because reducing research material to written form is an integral part of a researcher's job.

What is the Scope of Employment for Researchers?

            For purposes of the 1976 Copyright Act, the scope of employment for most university faculty is extremely broad. As an "employee," it usually involves long and often unscheduled hours doing at least four different activities: teaching, research, continuing education and community service. Where the faculty member has a doctorate in philosophy (Ph.D.) the underlying training and supposed qualifications for teaching and research are extremely broad. On the other hand, a faculty member with a doctorate in medicine (M.D.), science (Sc.D.), jurisprudence (J.D.), or the like (or even combinations of these more limited doctorates), have more narrow training and supposed qualifications for teaching and research. But, regardless of the subject matter covered in the classroom, teaching is like being a parent: there is no boundary to its scope. And good research also knows no boundaries. As to community service and continuing education, the scope of university faculty duties as an employee may be so diverse as to include working as a consultant in the "outside" world one day a week, or serving on the board of the local ballet company.

            As a practical matter, the scope of employment of university faculty and staff for purposes of the "work made for hire" definition in the Copyright Act may be even broader. Thus, most works created by a faculty or staff member working as an "employee" will be works made for hire. All in all, it is fair to say that the scope of employment for a university faculty member working as an employee is probably as broad as the university says it is.

            That is not to say that a faculty member cannot have hobbies that would lead to copyrightable works that are not works made for hire. They may, for example, be created during hours when the faculty member is not functioning as an "employee" as defined in Community for Creative Non-Violence v. Reid, the U.S. Supreme Court case mentioned above. One of the present authors composed music while employed at the University of Houston as a Professor of Chemical Engineering. Since the University didn't even know about the compositions, since the piano was at home, since university personnel, equipment and supplies were not used to prepare the works, and since the composition wasn't done on "university time," these musical works were probably not prepared as an "employee" as that term was defined by the Supreme Court for the 1976 Copyright Act definition of a work made for hire.

            These two concepts of "within the scope of employment"--one for defining an "employee" and the other for further defining a "work made for hire" in the 1976 Copyright Act--are distinct. True, judges and attorneys not seasoned in the nuances of intellectual property law get them mixed up, but in principle, they are sufficiently separate that an "employee" working off-time "within the scope of employment" as that phrase is used in the 1976 Copyright Act definition of the work made for hire does not create a work made for hire. For example, a faculty member may be employed to teach nonlinear dynamics and conduct research on developing Lyapunov Functions for a class of chemical reactors. While in the bathtub at one in the morning, the faculty member may think up a particularly interesting Lyapunov Function that he has studied in his research and even discussed in his lectures to advanced students. After drying off, s/he may then write it down, even write an article about it. If the written work was created during off-hours, it was not created by an "employee" so it cannot be a "work made for hire" even if it is within the "scope of employment." The university may accuse the faculty member of misappropriating ideas that were developed in part at its expense, but it cannot claim the copyright as a "work made for hire" unless it is created by an employee during the employment period.

            In this regard, research notes or even a research article may be like a teacher's lecture notes. Credit here belongs to the physicist, biophysicist and "scientist of conscience", Leo Szilard and the great mathematician and biomathematician, Stanislaw Ulam, who, with Edward Teller, designed the two-stage radiation implosion design that led to the creation of modern thermonuclear weapons. My wife and I (CFW) were at dinner with Ulam and his wife Francoise while they were visiting mutual friends in Los Alamos in the late 70's. Ulam, an extremely verbal person, was full of stories, quotations, and tales. One is that, during development of the atomic bomb, Szilard, lying in his bathtub in Los Alamos, wondered to himself what the American public would think if they knew they were paying him to lie in his tub and think. I wondered if Szilard would lose his copyright if he reduced his thoughts to research notes after drying off. Indeed, the scope and hours of employment for theoreticians may be unbounded.

            Obviously, works authored by employees when they are not functioning as employees or outside the scope of employment as that phrase is used in the 1976 Copyright Act must not be done on university time. In addition, they must be done away from the place of employment and strictly without university resources. Thus a theoretical insight on one's own time at home would still be a work for hire if it was based on data developed during university research.  Otherwise, the university has a claim for reimbursement and restitution of resources used by a nonemployee or by an employee outside the scope of employment.

            In any case, it seems clear that the vast majority of research articles prepared by university faculty or staff worthy of publication in a good journal are probably works made for hire.

What is the Agreement with your Employer?

            Section 201(b) requires that if there is no express written agreement signed by both the university and the faculty or staff member, the copyright to works made for hire belongs to the university.

            On the other hand, if there is such an agreement, and it is lawful, the copyright belongs to the faculty or staff member.

            However, if the contract is with a state employee and it is not lawful, it is probably void and the copyright probably belongs to the state employer. This illustrates an important difference between employment by state versus private institutions. Contracts by private institutions are not controlled by laws limiting transfers of state property, whereas contracts by state institutions may be.

            Most states have criminal laws that prohibit individuals from taking state property. Taking more valuable state property is usually a felony.

            If the laws of a state merely prohibit the transfer of state property to state employees, a state university or college may be able to lawfully agree that initial title to a copyright in a work made for hire vests in an employee who creates it, but it would not be able to transfer title to an employee of a copyright vested in it as a work made for hire.

            On the other hand, if the laws of a state prohibits any arrangement wherein a state employee receives title to what would otherwise be state property, an employee could not own the copyright to a work made for hire even if the university attempted to contract otherwise.

            Most of the copyright arrangements between academic institutions and their faculty and staff fall between the two extremes of no contract satisfying the requirements of Section 201(b) and a lawful contract that does. Many university have "copyright policies" which the parties incorporate into their written employment agreements. These incorporated policies can satisfy 201(b) and copyright ownership can depend on whether the work falls into a specific category spelled out in the policy. For example, a lawful, incorporated policy may state that the copyright shall belong to the employee unless it is a computer program. Then the copyright to a research article belongs to the faculty or staff member who authors it (except, of course, for computer programs incorporated in the article).

            Some university copyright policies are vague or ambiguous. These may require interpretation in litigation or they may not satisfy the "expressly" requirement in 201(b). Some policies are not expressly incorporated into the employment agreement, and, perhaps depending on the vagaries or acceptability of the extrinsic evidence rules of the different states (and a number of other factors), may or may not satisfy 201(b). Some policies are not signed by both parties and may not satisfy 201(b). While it is unclear why the employee should have to sign an agreement giving him or her title to the copyright, it is clear that the statute requires both the signatures of both the employer and the employee.

Untangling the Mess

            To begin, all authors should find out what their employer's copyright policy is.  If there is no written policy, or the policy does not grant the employer the copyright in research works, then the author should encourage the employer to change the policy.  Until there is a general policy giving the author the copyright, the employer can give the author permission to sign copyright agreements and submit articles on behalf of the institution.

            Rather than having authors sign overreaching copyright and warranty agreements, publishers should set forth written guidelines and standards for authors submitting articles, and authors should be willing to state that they have read the guidelines and standards, and, to the best of their knowledge, their article does not violate them.  These standards should recite commonly accepted academic expections such as that the work being submitted by the authors is the work of the authors, and that quoted work be properly identified and referenced.

            Some journals may elect to avoid articles that may attract litigation (for example, articles about decoding commercial encryption schemes), or otherwise decide what risks it wants to take. Hopefully, most journals would take a more principled stand and publish the article under rights enshrined in our First Amendment. There is clearly no excuse for the present course some journals seem to be taking, which is to chill First Amendment rights by trying to shift responsibility to authors. Authors are in no position to make a decision without spending a great deal of money, and they are not going to do that. If journal publishers insist on making their contributors lightening rods for litigation, the symbiotic relationship that exists today between authors and academic publishers is ended.

            If free speech is to continue in the United States, journals have to continue to take modest risks that they might get sued. They should concern themselves with preventive measures such as guidelines and standards as suggested above, or other measures that do not conflict with their basic mission of distributing scientific information to the research community.

            Protecting against claims like copyright infringement, defamation, DMCA violations, violation of trade secret laws, and violation of export laws by demanding warranties from authors is nearsighted over-lawyering. These lawyers are just doing their job, which does not include the advancement of science or protecting First Amendment rights of their clients' authors. But the price it too high for the benefit expected:

            --Copyright infringement claims might, in the worst case, result in a journal issue being recalled.

            --Defamation issues have always been a concern of publishers, but are extremely unlikely under U.S. law in the context of academic research articles.

            --The DMCA suits are frightening to all civil libertarians, which hopefully includes every citizen. But they seem more of a threat to our First Amendment rights than to journal solvency. Unless the journal is actually distributing offending software code, it is very likely to be protected by its First Amendment rights.

            --Claims for misappropriating trade secrets are likely to fail unless the journal knows that the author misappropriated them (in which case the journal should decline to publish the article).

None of these consequences are worth the harm threatened by requiring warranties and indemnification from authors.

            With respect to copyright ownership, this is not a situation where the publisher has a large investment in the work that it must recoup to make its profit. The facts are just the opposite: The university and the author have invested their time, effort, resources and money to create the work. They should be rewarded with the copyright precisely as contemplated in Article I, Section 8 of our constitution. And, in the case of state employers, it may be unlawful to transfer the copyright without adequate payment to the taxpayers who really own it.

            All the journal needs to publish the article is the Section 201c Contribution right.  Many scholarly journals are also published on the Internet and are otherwise available electronically.  In general, such distribution is important to authors and it is worth giving the publisher a non-exclusive license so that the work can be more widely available.  The author and his employer should be willing to grant these rights, in writing if that's what the journal wants. But, as described above, the Section 201c Contribution right is probably implied by the submission of the article to the journal for publication, in disciplines that do not allow articles to be submitted to more than one journal at a time.  In disciplines that allow multiple submission, the journal's Section 201c Contribution right would be evidenced by accepting an offer of publication.

            Authors should be willing to state that, by submitting an article to the journal, they represent that they wrote the article and collected the data, unless specifically stated otherwise in the article. And, as outlined above, that they believe that they have read the journal's written guidelines and standards and believe that they have complied with them.

            If the journal wants more, it should pay for it. In the next article in this series, we will discuss how journals can commission articles and ensure ownership of the copyright as a work made for hire.