Last major update 10-26-01



L.H.O.O.Q.--Internet-Related Derivative Works

Use NS 6.1+ or I.E. 5+, 800 x 600 screen recom.


Larger (270 x 414) LHOOQ graphic

This is the most famous derivative work in the world. For years it was believed to be, as Dadaist Marcel Duchamp (1887-1968) asserted, just a cheap, mass-marketed postcard that he casually purchased in Paris in 1919 and on which he then penciled in a moustache and goatee and the letters L.H.O.O.Q. (In French slang at the time, "Elle a chaud au cul.") More recent scholarship suggests that Duchamp laboriously altered the postcard before adding the moustache, including merging his own portrait with that of Mona Lisa. This work is sometimes referred to as "Mona Lisa With a Moustache" and sometimes as "L.H.O.O.Q." This Web page will use the latter, shorter designation most of the time.

Why is adding the moustache, beard, and L.H.O.O.Q. the preparation of a derivative work? Why isn't that so insubstantial a juxtaposition of material that it amounts to no more than painting a hobby horse blue or moving a picture from one wall of a room to another?

To answer that, we should begin by asking, "What is a derivative work?" The two principal references to the term in the 1976 Copyright Act (in which it first appeared as a separate category of work) are in sections 101 and 106(2). Section 101 of the Copyright Act defines a derivative work in these terms:

A "derivative work" is a work based upon one or more preexisting works, such as a translation, musical arrangement, dramatization, fictionalization, motion picture version, sound recording, art reproduction, abridgment, condensation, or any other form in which a work may be recast, transformed, or adapted. A work consisting of editorial revisions, annotations, elaborations, or other modifications, which, as a whole, represent an original work of authorship, is a "derivative work".

Section 106(2) gives copyright owners the exclusive right "to prepare derivative works based upon the copyrighted work" and to authorize their preparation. Preparation of a derivative work does not imply embodiment in a tangible copy, such as a paper writing or recordation in a magnetic disk. The legislative history of the 1976 Act indicates Congress' odd concern to protect owners of copyright in pantomimes and similar works, which might be the basis for a derivative work without recordation. In legislating to protect pantomimes, dumb shows, and choreography, Congress deliberately omitted the requirement of fixation in a copy, characteristic of reproduction and distribution under clauses (1) and (3) of section 106. For no particular reason, however, Congress expanded derivative-work protection against infringement well beyond dumb shows and similar works; it dropped any requirement of a copy for works derivative of books and other literary works, despite the absence of any expression of felt need for such expansion. Thus, unauthorized preparation of a derivative work is infringement regardless of whether the defendant stores the derivative work in a permanent memory.

The two sentences of the definition appear to be independent of one another. They may even be at cross purposes, for a work consisting of editorial revisions, annotations, elaborations, or other modifications is unlikely to be a translation, musical arrangement, dramatization, fictionalization, motion picture version, sound recording, or art reproduction (it might be an abridgement or condensation but such works are hardly important enough to deserve mention). The first sentence is more of a definition in terms of characteristics: a derivative work is a transormation of another work, but then the first sentence closes with a catalog of known forms of transformation.

Does the use of the word ''work'' here imply that there must be a work of authorship in the constitutional sense? That is, must a derivative work have some minimum of originality so that its creator is an author per Article I, section 8, clause 8? The Batlin case holds that it does--for copyright protection to attach to a derivative work. But Annie Lee says that it is unresolved whether the same rule applies for a claim of infringement to prevail against a supposed derivative work, and reasons on the basis of the assumption that a set of non-original derivative works may exist.

The second sentence of the definition simply enumerates a few examples of editorial modifications, and then ends the catalog with the statement that these and other editorial modifications qualify as derivative works if they are original. But saying that these examples qualify if they are original does not mean that others (such as the items listed in the first sentence) cannot qualify unless also original. About all that one can get out of this mess is that a subsequent work based on an earlier work is a derivative work if it is a transformation of the first work or is an editorial modification that possesses enough originality for it to be a work of authorship itself.

Probably, painting the hobby horse War Cloud blue and renaming it Thunder does not qualify under either sentence, either under whatever standard applied in the 1960s when it occurred or under the present statute. But festooning Mona Lisa with a moustache and the letters L.H.O.O.Q. is another matter, at least for purposes of the first sentence of the present U.S. statute.

The creation of L.H.O.O.Q. profoundly transformed the perception of La Joconde (what the French call the painting, in contrast with the Americans and Germans, who call it the Mona Lisa). In 1919 the cult of Jocondisme was practically a secular religion of the French bourgeoisie and an important part of their self image as patrons of the arts. They regarded the painting with reverence, and Duchamp's salacious comment and defacement was a major stroke of epater le bourgeois ("freaking out" or substantially offending the bourgeois). The transformation of a cult icon into an object of ridicule by adding a small quantum of additional material can readily be deemed preparation of a derivative work. Furthermore, Duchamp was able to make his point with considerably less apparent addition of his own material than 2 Live Crew needed in the Big Hairy Woman case.

L.H.O.O.Q. is particularly interesting in providing a test model for Internet-related works. It illustrates one of two principal categories of derivative work: derivative works where the old and new elements can be juxtaposed without physical commingling. In a translation or dramatization of a literary work, for example, the old and new elements are intimately commingled; they become very difficult to dissect apart for purposes of analysis. But L.H.O.O.Q. illustrates a paradigm in which the old elements may be regarded as a first layer, over which a second layer is superimposed. The second layer is transparent in the main, but is opaque and obscures the original layer in some places (for example, where the facial hair is located). Thus, the figure at the right corresponds to the first layer (Mona Lisa). The small figure at the left top of this paragraph corresponds to the second layer (moustache). In addition, we need an opaque overlay containing the letters L.H.O.O.Q..

L.H.O.O.Q.
 
When the first layer is suitably obscured by the other material, the result is Duchamp's derivative work. The combination is shown here at the left, together with the addition of some other layers to show you what Duchamp might be doing on his Internet Website if he were now a thriving centenarian.

The layers paradigm is significant in a computer-related or Internet context because it readily describes a system in which the person ultimately responsible for creating the composite (here, corresponding to Duchamp) does not make a physical copy of the original work in the sense of storing it in permanent form (fixed as a copy) distributed to the end user. Rather, the person distributes only the material of the subsequent layers, the aggrieved copyright owner (here, corresponding to Leonardo da Vinci) distributes the material of the underlying layer, and the end user's system receives both. The end user's system then causes a temporary combination, in its computer RAM and the user's brain. The combination is a composite of the layers. Framing and superimposition of popup windows exemplify this paradigm.

Comparable layered arrangements occur in other electronics contexts. Add-on programs, in which a second program (say, an enhancement of a word processor or database manager, or a video game speed-up kit) co-acts with the program that it enhances without changing the permanent code of the latter. The modifications appear only transiently in RAM. Manufacture of integrated circuit semiconductor chips is carried out by a layering process. Material is deposited on and etched away from successive layers of the product, forming a semiconductor sandwich with properties dependent on the shape (topography) and chemical makeup of the various resulting layers. It formerly was customary to represent the design of a semiconductor chip product by a set of plastic overlays illustrating each layer. Netscape 3 used a layers concept which has since given way to the Cascading Style Sheets of HTML 4, which both NS 6+ and IE 5+ now support.

Past literary and audiovisual derivative works have tended to follow the other paradigm, that of commingling. Shown on a separate page are some other ways to make derivative works based on Mona Lisa, where any layering is solely in time (not space). In some instances the layering in time appears to create a single continuously morphing Mona Lisa, rather than just a series of separate effects. In all of these further examples, however, the person responsible for creation of the derivative work (assuming that it is a derivative work) commingles in one image file Leonardo's content and that of the subsequent artist. The taking of the original material thus requires the preparer of the derivative work to reproduce a copy of the underlying work (see 17 U.S.C. § 106(1)), rather than exploit it by layering over a copy of the underlying work that its author or licensee provides. Typically, that makes it easier to establish infringement liability.

In general, the case-law on derivative works does not plumb the depths of legal analysis. Annie Lee, in rejecting picture-framing as a preparation of a derivative work, is as conclusory and uninstructive in its way as Mirage is in reaching the contrary result. The Artic decision on speed-up kits for video games is simply fatuous in its rationale of "someone would like to lay his hands on that revenue." No development of law has occurred for add-on programs, although ten years ago they seemed controversial. (Apparently one of two things occurred for all of them. Either the proprietor of the original program decided that the add-on was a positive benefit or at least no commercial threat. Or else the proprietor bought up the rights to the add-on or just infringingly incorporated it into the main program as a feature, for example, as Microsoft did with "Stacker" file compression.)

Thus far, framing of Web sites has not produced any worthwhile jurisprudence, although surely that is a next legal frontier. The layers paradigm seems particularly apt for understanding such framing. Thus, in iframing one (the iframer) prepares one's own material as what we may call HTML Layer 1, for combination with a framed content provider's HTML Layer 0. The process of combining the layers is shown on an accompanying page. At least superficially, the addition of layers in this manner seems to be very much like what Duchamp did when he overlaid a moustache on the Mona Lisa. In the alert box and flying popup window demonstrations, we clearly saw the superimposition of one layer containing a window with its own content on another's prior layer of different content.

What does the layers paradigm tell us about developing a model or structure for analyzing whether framing content is preparation of a derivative work? Or perhaps more important, an infringing derivative work? Or at another level, a preparation of a derivative work causing liability for copyright infringement to attach to the conduct of the actor?

Those are not necessarily the same things. A second work might be a derivative work based on an earlier work, but it might not be infringing if the complained-of elements that it took from the earlier work were not copyright-protected elements (for example, facts, idea, or mise-en-scene). Even if the taking is infringement, the taking might be licensed, fair use, or otherwise excused. Arguably, the three things -- derivative work, infringing derivative work, copyright-liable derivative work -- are successive subsets.

- End of first page of LHOOQ -

[Proceed to next page of LHOOQ for additional material]