Scrivener's Error: Warped Weft

Mistaken Federalism

originally posted Dec 2006–Oct 2008

General Framework

{Dec 2006} This ongoing essay is inspired by, but not limited to, a real matter that is currently being litigated. Do not assume that any fact stated in this essay — including minutiae like the sex of any person — represents that specific matter.

Famous Author (you've seen Famous Author's works in school libraries if you grew up in the US) is estranged from his family, and has lived nearly a century. Never having been married or otherwise had children, Famous Author has no obvious intestate heirs under the law of the state in which he lived for the last decade prior to his death. (It is even probable that whatever living relatives he has do not even know of his death, although the announcement made several national top-end newspapers.)

Famous Author had a local attorney establish a literary holding company, for both tax and other reasons, about fifteen years before death. Famous Author's works registered after that date are credited in the Copyright Office's records to "Famous Author, Inc., by Famous Author as a work made for hire." Those publishing contracts were between Publisher and Famous Author, Inc., not directly with Famous Author. He did not, however, transfer previously registered copyrights to Famous Author, Inc. Just to make things even more fun:

  • Famous Author provided creative expression to several series of fictional works. Some of these series were entirely completed before the founding of Famous Author, Inc.; a few of them were entirely completed only after the founding of Famous Author, Inc.; but many of them have split histories, and copyright records show early books as "acFamous Author" and later books as "cFamous Author, Inc., by aFamous Author as a work made for hire."
  • As implied by the previous item, Famous Author often — particularly toward the end of life — worked in collaboration with several Less Famous Authors. Some of these collaborators produced works that fall into each of the three categories mentioned in the previous item. To make things even more interesting, some of the collaborations involved more than two natural persons.

Famous Author had another local attorney write a will for him. The will divides his estate among three separate interests:

  • Certain low-value specific bequests, and the contents of Famous Author's reference library, go to a trust whose purpose is to provide working space and research materials for authors writing in Famous Author's field. No party objects to these provisions.
  • The copyrights in Famous Author's works that are "published posthumously" are to go to Miss X, a nonrelated former business associate. Miss X had no role in creating any of Famous Author's works, nor did Miss X ever have any role with Famous Author, Inc.
  • The residuary is to go to Mr Y, a nonrelated person. Mr Y had no role in creating any of Famous Author's works, nor did Mr Y ever have any role with Famous Author, Inc.

As noted above, Famous Author dies. His will therefore must go to probate. And now the fun begins, involving federalism, bad will drafting, bad record-keeping, bad statutory drafting... and probably bad faith. But this is just another pathetic cliffhanger.

In the meantime, though, you should read this entry on writers' wills on Neil Gaiman's blog. (I don't endorse everything in detail — particularly not in the sample document he refers to — but that is as much a matter of opinion and hairsplitting as anything else, and stiffs-and-gifts isn't my primary area anyway.)

The Real Case: Andre Norton

{Oct 2008} A while back, I started to describe a hypothetical instance based on a pending probate matter. Before I could get very far, though, I was consulted by a claimant in a way that — thanks to the procedural posture of the matter — made further comment at that time inappropriate. Now, however, the Tennessee Court of Appeals has issued its appellate opinion (PDF), which reaches part of the right result... and utterly fails to consider the underlying question that I was starting to reach for in that long-ago post.

First, a little background.1 The author in question was Andre Norton. Unfortunately, her health was not the best during her last years; neither was her choice of caretakers or advisors. This resulted in significant problems with overreaching by at least one caretaker — based on what I have learned, best described as elder abuse — and a will that was drafted less competently than I would expect of a first-year law student doing his/her first legal writing project. And things only got worse when the estate was probated; although the Court of Appeals does not make its criticism very explicit, there was a lot of professional incompetence (and possibly misconduct) in the trial court, which becomes clear only when one reads the trial court's opinion.2 In any event, the will in question purported to send "copyrights" one place (with a few exceptions) and "royalties" to another place... but not using consistent language, or structure, or anything else. Further, there are substantial questions as to whether the will is consistent with the now-deceased author's intent under any interpretation of the will, as demonstrated by both other documents and accounts of oral communications. And, just to make things even messier, Ms Norton cowrote a number of works with others, some of whom are not even US citizens or residents!

The result of the Court of Appeals's opinion looks something like this (points numbered for convenience in later reference, and not parallel to the opinion):

  1. The Executrix must be replaced, as she also holds a challenged claim in the estate that is the subject of the appeal. This is a no-brainer; the will itself should have provided for a better chain of contingent executors, and probably a separate literary executor. Either way, though, this is clearly the right result and should have been voluntarily achieved in the trial court at a much earlier stage. This is not unique to either IP or estates law: Conflicts of interest should be avoided wherever possible.
  2. The will was ambiguous in a fashion requiring reference to external materials.
  3. Those external materials include the "technical meaning" of some terms from the publishing industry and copyright law.

    Having previously disposed of the full copyrights to her co-authors and without expressly reserving any rights incident to those copyrights, we, therefore, cannot infer that Ms. Norton intended to give the royalties generated from those collaborative works to Dr. Horadam. Based on the evidence in the record and the Will as a whole, we find Ms. Norton clearly intended to award two different assets when she referred to "copyrights" and "royalties" in her Will. Therefore, we reverse the trial court's determination that Ms. Norton used the terms "copyrights" and "royalties" interchangeably. In order to properly interpret the meaning of Ms. Norton's bequests, we must examine what, if any, technical meaning is assigned to the terms "copyright", "royalty", and "posthumous publication".

    Horadam v. Stewart (In re Norton's Estate), No. M2007-00046-COA-R3-CV (TN App. 06 Oct. 2008), slip op. at 13 (citations omitted; some punctuation corrected).

  4. "Royalties" and "copyrights" are distinct terms, and distinct property interests, disposed of separately by the will. Contrary to the trial court's interpretation, Dr Horadam receives only the royalty stream from "posthumous publications" — not the copyrights, and particularly not any interest in anything devised or otherwise transferred to coauthors.

Bluntly, the Court of Appeals reached the right result... for some of the wrong reasons, and without the authority to reach any result on the critical issue (the fourth one). In fact, the trial court below also lacked authority to reach that result. And it's all the lawyers' fault. I do, however, need to offer one quick correction to a much more famous and eminent blogger: This is not the end of the controversy over Andre Norton's will. Dr Horadam could appeal this opinion from the Tennessee Court of Appeals to the Tennessee Supreme Court... and then, because it raises substantial issues of federal law, to the US Supreme Court. Conversely, absent an appeal there is still significant work to be done in the trial court. The one thing that one can say at the moment is that, until there is a different ruling, Dr Horadam can neither block publication of any of Andre Norton's works, nor authorize publication or other exploitation of any of Andre Norton's works. The effect of this ruling is that he owns only an interest in certain royalty streams, presumably limited to publication (although that issue was not tested).


I'm going to break down the opinion a bit farther, trying to extract the issues most relevant to authors and other creators of IP interests... whether deceased yet or not. One cannot emphasize enough that, as Mr Gaiman put it in greater detail (including links to sample language that will generally do the job in the US, even though it's not the way I would do it), y'all need a valid will. Do it. Now. And make certain that the person(s) you choose as executor know(s) how to ask for advice on managing the IP portions of the estate, and from whom!

The Court of Appeals disposed of point 1 pretty well. It did a much less satisfactory job on point 2: That there is an ambiguity in the will. Whether a written document is ambiguous is ordinarily a pure question of law. One should always be suspicious when an appellate opinion cites Black's Law Dictionary; the level of suspicion should increase geometrically with each additional citation. Bluntly, Black's Law Dictionary is a pretty worthless hunk of outdated, inelegantly-written opinions of what various legal terms "mean" — and that's a particular problem when the underlying question is whether something is ambiguous! Since the Court of Appeals cites Black's Law Dictionary three times, we should be only 1/8 as confident in its reasoning as we might otherwise be... and the Court's footnote 3 only reinforces that conclusion.

Some jurisdictions make no distinction between latent and patent ambiguities when 3 admitting parol evidence to identify, explain, or define the subject matter imperfectly described in a writing. See, e.g., Battle v. Wolfe, 283 S.W. 1073 (Tex. App. 1926) (noting express repudiation of the rule distinguishing patent and latent ambiguity if extrinsic evidence is necessary to identify subject matter); Cumberledge v. Brooks, 85 N.E. 197, 199 (Ill. 1908) (holding the admissibility of parol proof not dependent on distinction between patent and latent ambiguity); Armistead v. Armistead, 32 Ga. 597 (1861) (criticizing distinction as “wholly unphilosophical”).

Horadam, slip op. at 7. In this instance, the Supreme Court of Georgia had it right a century and a half ago: There is no philosophical — and, indeed, no real legal — distinction between the two; it is only a question of whether one characterizes outside sources used to try to resolve an ambiguity as "authority" or "evidence," which is a case-by-case determination.

Despite this weakness in the opinion, it does point out one preventive measure that holders of IP rights must take when creating an estate plan: Make certain that the will, trust, or transfer to a holding entity is excrutiatingly clear on what is being transferred and what is not, and is at minimum internally consistent. If possible, identify interests by a registration number and/or date of creation; identify the statutory source of the interests (e.g., the Copyright Act); identify the nature of the underlying work that creates the interest (e.g., unpublished manuscript, screenplay, sculpture); and, perhaps most important, keep the identifications up to date. The best way to do this is to include an appendix to the will that gets added to periodically, at least once each year, to include new works. The exact mechanism varies from state to state; in some states, you can merely refer to an authoritative index, in others you must formally add a codicil (amendment) to the will, in still others you can merely update an exhibit to the will without replacing the earlier versions.


My point 3 is where — with all due respect — the Tennessee Court of Appeals's opinion begins to break down into the sorts of meaningless, or in any event insupportable, distinctions that tend to give law a bad name. It is fairly clear, and self-evident to anyone with half a brain, that not every term found in Andre Norton's will has an internal definition. If it did — even if it defined "black" as "the presence of all colors" — this opinion would be entirely unnecessary; one could merely apply the definitions found in the will to all of the terms and be done with it. That's not to say that there would be no dispute, as any reasonably creative lawyer (if that's not an oxymoron) can find plenty of nondefinitional problems with this situation; it's only to say that it wouldn't be this one.

Instead, the real question is what will constitute a satisfactory source for definitions. Here, I am afraid, the Court of Appeals goes badly, badly wrong. This is partly forced upon it by stare decisis — respect for authority of the past. Yes, I know that it's bad Latin, but no worse than my sarcastic postmodern version: stagnans decisis. In this instance, the Tennessee Court of Appeals — an intermediate appellate court — did not believe that it had the authority to do away with the silly distinction between "patent" and "latent" ambiguities in wills. That, in turn, masked the remainder of the problem; the Court bloody well knew that, however it characterized the undoubted ambiguities, it would have to reach outside of the document itself to resolve them.

And here is where stare decisis, and similar legal doctrines, turn around to bite everyone in the ass. At a fundamental level, how do we define an outside source of data? Legal doctrine makes a huge distinction between "authority" and "evidence"; however, that's all based upon a formalistic distinction from purely syllogistic reasoning. A better logic model, such as Toulmin's general model of argument, makes clear that when we're trying to discern the meaning of an ambiguous claim, we're dealing not with a distinction between "evidence" and "authority," but between "warrant" (assumption and method of reasoning) and "data" (everything else). In this conception, both legal/technical definitions of "copyright", "royalties", and "posthumous publications" — the foundation of the ambiguity that is immediately at issue — and use of those terms by both the testator (maker of the will) and in the relevant industry are mere "data" used to support the claim.

That is surely correct... and foreshadows my last point. Much of the "testimony" recounted by the Court of Appeals (and, for that matter, the trial court) is from a law professor who is essentially restating the contents of Nimmer on Copyright. Leaving aside for the moment whether Nimmer really is as authoritative as claimed — it isn't, and particularly not in the area of succession of copyright interests, as anyone who actually read the damned thing and its preface would understand — this demonstrates exactly why the distinction between "evidence" and "authority" is so meaningless in this context. One can characterize that "testimony" as either a summary of authority or expert opinion evidence; the way it is presented to the court does not, or should not, matter — only its actual value as data, in the Toulmin model, for evaluating the claim.3 This is one of the many instances in which legal procedure and labelling is outcome-determinative.

In any event, the Tennessee Court of Appeals eventually had to answer two questions:

  • What is a "posthumous publication"?
  • Are "copyrights" and "royalties" the same thing?

The Court did, ultimately, reach the correct result on these questions. It held that a "posthumous publication" is a work that was not published at all during the lifetime of the author (which, unfortunately, leaves the question of whether an unfinished collaborative work would be "posthumous" after the death of the first author, but before the death of the collaborating author, and makes things even more dicey when considering the messed-up corporate structure in place). It also held that "copyrights" and "royalties" are not the same thing. The final issue, though, is whether the Tennessee Court of Appeals even had the right and power to answer these questions... which I'll get into over the next few days. The short answer is that it did not.


A court must inquire into its subject-matter jurisdiction at every step — that is, whether the nature of the lawsuit is one within its power to hear. The US system divides that authority between the states and the federal (national) system. Two particular doctrines are relevant to this dispute. On the one hand, there is a judge-made "probate exception" that keeps probate matters in state court, and denies federal courts subject-matter jurisdiction over them.4 As the Supreme Court has recently made excrutiatingly clear (or, at least, excrutiating), though, this is a very, very narrow exception.

In short, we comprehend the interference language in Markham as essentially a reiteration of the general principle that, when one court is exercising in rem jurisdiction over a res, a second court will not assume in rem jurisdiction over the same res. Thus, the probate exception reserves to state probate courts the probate or annulment of a will and the administration of a decedent's estate; it also precludes federal courts from endeavoring to dispose of property that is in the custody of a state probate court. But it does not bar federal courts from adjudicating matters outside those confines and otherwise within federal jurisdiction.

Marshall v. Marshall, 547 U.S. ____ (2006) (citations omitted).5

On the other hand, federal courts have exclusive jurisdiction over copyright claims — and the meaning of the Copyright Act — under both the Copyright Act itself (17 U.S.C. § 301) and the federal jurisdictional statutes (28 U.S.C. §§ 1331, 1338(a)). As the Tennessee Court of Appeals's decision makes clear — and, again, excrutiating, but not excrutiatingly clear — interpreting Andre Norton's badly drafted will requires answering the two questions posed above. Both of those questions, in turn, have their answers in the Copyright Act, implying that answering them falls within the exclusive jurisdiction of the federal courts. Perhaps the biggest hint is that every single nontestimonial source cited on these questions — and much of the testimony — is federal.

We thus have a choice-of-jurisdiction question that parallels the classic choice-of-law conundrums of renvoi and dépeçage, which try to describe the circular logic of these kinds of cross-references (in a confusing way intimated by the use of bad French translations). The key point, though, is that at the level of subject-matter jurisdiction, the "right result" from a court that has no subject-matter jurisdiction is not wrong, but meaningless, and cannot bind the parties (or anyone else), nor serve as precedent for others. In this context, that means that the matter should have been either dismissed in its entirety or stayed while the parties were sent to the United States District Court for the Middle District of Tennessee to obtain a declaratory judgment on these two questions. And the result, under pretty clear federal precedent, would have been this one; but it would have been clearer, shorter, and meaningful.

  • Although the Copyright Act does not use the term "posthumous," it does contain an extensive discussion of works first published after the death of the author, used in calculating the term of copyright. The Tennessee Court of Appeals does not once refer to these provisions. Similarly, a large body of case law in the Second Circuit, which has been adopted by implication in the Sixth Circuit, reaches the same conclusion... and does use the word "posthumous."
  • At its core, though, this dispute requires deciding whether "copyrights" and "royalties" are coextensive, either in practice or in theory. In turn, that requires discerning the meaning of "copyright"... a question exclusively within the scope of the Copyright Act. The Tennessee Court of Appeals's confusion appears to come from its (and Nimmer's) inelegant mixture of cases decided under the 1909 Act — under which the various rights that form part of a copyright were "indivisible" — and the 1976 Act — which explicitly made the various rights that form part of a copyright divisible, including the technical probate sense of "devise".6 A federal court would not have been confused by these issues; or, at least, one hopes not.

This error means that the dispute between Dr Horadam and Sue Stewart is not over, and in fact will never be over until it is ruled upon by a federal court. That is, there will always remain some uncertainty, because a third party claiming an interest in the underlying properties — hypothetically, a film producer claiming that Dr Horadam had licensed him the rights to film one of the novels during the period prior to the Tennessee Court of Appeals's decision — could file its own declaratory judgment action in federal court. (It would arguably implicate legal malpractice if it did not!).


Andre Norton's estate will continue to provide amusement for those of us with particularly eclectic (and sick) senses of humor for quite some time. The situation includes a catalog of "don't do this with your own estate!" problems.

  • If you have a business entity set up to handle your intellectual property, don't refer to its possessions in your will! Ms Norton did, but it wasn't set up properly or managed properly; this entire brouhaha could have been avoided by using her holding corporation correctly, and in fact it would have been vastly easier to set up partial distributions from corporate assets.
  • If you do not have a business entity set up to handle your intellectual property, ensure that your lawyer uses the correct terms of art (and includes specific definitions in the will itself, if there are terms unique to your kind of intellectual property interests — in this instance, a definition of "royalties" would have been useful). In turn, that means that a "general practice" attorney, or even one who specializes in trusts and estates, must consult with an intellectual property attorney who understands the substance of what is at issue, not just the trusts-and-estates-imposed procedure.
  • If there is a challenge to your will (or you are challenging a will or intestacy proceeding for someone who died without a will), make sure that the challenge is being heard in the right court. Again, this refers back to the specialization of attorneys problem; as the client, though, you are entitled to make your lawyer explain why you are in a particular court.

But all of that pales next to the need to have a will in the first place. It is absolutely essential, and not just for older authors/composers/artists: You could be hit by a falling meteorite at any time. With a will, at least there's something to fight over; without it, there is no legal indication of what you wanted done. And keep it up to date; if you have children, or own real property of any kind, or jointly own any intellectual property, your will should be reviewed at least every three years.

If you jointly own intellectual property with anyone else — either as the creator, an inheritor, or whatever — you should ensure that your coowner(s) know both the location of your will and who you have designated as executor. Some deaths are more untimely than others; if there is a time-critical negotiation for rights in that IP going on at the moment of death, it's essential to know who can conclude the negotiations.

Finally, you should give some thought to giving special instructions to your executor regarding your intellectual property, or even appointing a literary executor whose only duties are to handle the intellectual property. This can be extremely important; it's definitely something to discuss with the attorney who is creating your estate plan. Not every estate plan needs an attorney, but if any of this six-part essay is not crystal-clear and potentially applies to you, yours probably does. Too often, the expense of going to a lawyer for good preventive law — like a good estate plan that actually covers all of the issues — is treated as excessive. Instead, it's like changing your oil filter; one must wonder how badly depleted and/or damaged Andre Norton's literary estate will be after what has already been three years of litigation (and, as I've noted above, it's not done yet!).


  1. I do not pretend that this summary of facts includes all of the nuances, or is entirely fair to everyone involved. Suffice it to say that there were a lot of mistakes, including significant questions about legal malpractice, and that this ugly situation is an excellent example of why authors and other owners of inheritable intellectual property interests cannot rely upon "general practitioner" attorneys to structure their estates. Once again, I call on the organized bar to admit that attorneys are, in fact, specialists, even if we don't get certification as such. In this particular instance, the legal malpractice appears to have resulted from more than one attorney's twin assumptions that copyright interests are indivisible (not true after 01 January 1978) and that they follow all of the rules applicable to any other personal property, such as a car, when managing an estate (never been true).
  2. Technically, I suppose one could object that it wasn't really a "trial court," but a chancery proceeding before a judge. However, as that lower-court opinion makes clear, there was virtually a trial. I am using "trial court" more as shorthand here than a technical description, because in many other proceedings — indeed, in many other states — probate is handled in merely a separate division of the general-jurisdiction civil court, not a separate court.
  3. Some "speech and communication" types — such as the ones who created the reference to Toulmin's model embedded above — tend to place expert opinion as part of the warrant. That may be easier to teach, but it distorts Toulmin's own writings, in which he treats what we would otherwise call "expert opinion" on facts as data; only when "expert opinion" goes to the very method of evaluating the data — that is, the warrant — is it part of the warrant in Toulmin's own conception. In this context, expert opinion on what the definition of any of the ambiguous terms is would be data, which is what the testimony in the trial court (and the bulk of the Court of Appeals's opinion) concerned; expert opinion on the grammatical function of the terms, and how those definitions influence evaluation of the entire statement, falls inside the warrant.
  4. See, e.g., the extensive discussion of the exception in Markham v. Allen, 326 U.S. 490 (1946), which makes clear that this is purely a judge-made doctrine to prevent interference with the supposed exclusive jurisdiction over the state-law-unique aspects of administering an estate. Id. at 494.
  5. Thanks to the torpid pace at the overworked Clerk's Office of the Supreme Court (and the anal-retentive citation system imposed on cases by the Four Evil Law Reviews), this is only an unofficial version of the opinion over two years after it was issued, and will remain so for probably another year.
  6. Cf., e.g., DeSylva v. Ballentine, 351 U.S. 570 (1956) (cross-referencing state definition of "child" in contest over copyright renewal rights).

Law and reality in publishing (seldom the same thing!) from the author's side of the slush pile, with occasional forays into military affairs, censorship and the First Amendment, legal theory, and anything else that strikes me as interesting.

Ritual disclaimer: This blog contains legal commentary, but it is only general commentary. It does not constitute legal advice for your situation. It does not create an attorney-client relationship or any other expectation of confidentiality, nor is it an offer of representation.

All material © 2003–11 except where otherwise indicated. All rights reserved. N.B. This blawg does not use the Creative Commons License, although I'm usually pretty good-natured about permissions for attributed reuse.


Warped Weft

Essays in Warped Weft are consolidated from Scrivener's Error with only formatting corrections (and the occasional typographical correction).