The concept of what constitutes a derivative work seems to elude far too many courts, particularly in the photography context. In Ets-Hokin v. Skky Spirits, Inc., 225 F.3d 1068 (9th Cir. 2000), a photograph of a vodka bottle was taken for use in an advertisement. The district court had held that the bottle depicted in the photograph was a “preexisting work,” and thus the photograph was “based on” that work and was therefore a derivative work. Rather than dismiss this shockingly wrong conclusion on the basis of the plain words of the statute, the court of appeals went off on an irrelevant inquiry: Whether the object depicted in the photograph (the vodka bottle) was copyrightable. According to the court of appeals, if it was, the photograph was a derivative work; if not, the photograph was not a derivative work. In order to determine whether the bottle photographed was copyrightable, the court examined it as the design of a useful article, ultimately rejecting protection for the bottle and therefore upholding copyright in the photograph as a nonderivative work.
Photographs of other objects are not derivative works of those objects. First, a photograph of an object is not “based on” that object: It is a mere depiction of it. Second, even if one were to find that a photograph of an object is based on that “preexisting work” within the meaning of the definition of “derivative work” in Section 101, such a photograph must still “recast, transform, or adapt” the authorship in the preexisting work to be considered a derivative work. Such recasting, transformation, or adaptation does not occur in a photograph of an object, even copyrighted objects. What makes a derivative work a derivative work is the contribution of changes in the actual authorship of the preexisting work, not a mere depiction of that work.
Fortunately, Judge William H. Pauley III of the Southern District of New York set matters straight in SHL Imaging, Inc. v. Artisan House, Inc., 117 F. Supp. 2d 301 (S.D. N.Y. 2000). Judge Pauley rightly focused on the requirement of recasting, transformation, or adaptation, noting: “A photograph of Jeff Koons's ‘Puppy’ sculpture in Manhattan's Rockefeller Center merely depicts that sculpture; it does not recast, transform, or adapt Koons's sculptural authorship. In short, the authorship of the photographic work is entirely different and separate from the authorship of the sculpture
Now a district court in Illinois has returned to the Ets-Hokins approach,
Schrock v. Learning Curve Intern., Inc., 2008 WL 224280
(N.D.Ill., January 29, 2008). In Schrock, defendants hired plaintiff photographer to take pictures of hundreds of toys for use in marketing. There surprisingly was no work for hire or transfer, but instead a license with usage restrictions, which the photographer alleged had been violated. Defendants alleged the photographs were unauthorized derivative works of their toys, and regrettably the court agreed:
Schrock's photographs are product photographs-depictions of Thomas & Friends toys. Such depictions portray the three dimensional toy in two dimensions. In the words of Section 101, such photographs “recast, transform[ ] or adapt [ ]” the preexisting three dimensional toy into another medium, thus creating a derivative work “based upon” the preexisting work. Such photographs are no less derivative works than are three dimensional embodiments of two dimensional drawings (e.g., a guitar created to embody a written symbol, as in Pickett v. Prince, 207 F.3d 402 (7th Cir.2000), or porcelain dolls fashioned to embody Norman Rockwell illustrations, as in Saturday Evening Post, 816 F.2d at 1193). In sum, Schrock's photographs are derivative works of the copyrightable toys that they portray.
The court then widened its destructive trip through basic principles of copyright law by holding that there was no original derivative authorship based on the Seventh Circuit’s widely criticized Gracen opinion. In short it was a windy, blustery day in Chicago.
I'm entirely familiar with this area, so I'll refrain from making any judgments one way or the other. But, if a depiction of a copyrighted work does not recast, transform or adapt, is it merely a reproduction? If so, that's a heck of a lot easier result than making a DWs argument.
ReplyDeleteI also have to admit that the terms "recast, transform or adapt" would seem to include an awful lot of activities that you seem to dismiss--including making a copyrightable object fit for print.
Dear YFTL, yes I think a picture of a copyrighted object is a reproduction of that object. In the Shrock case, the owner of the object (toys) had given permission to take the photos, and was using the derivative work argument defensively -- to get out of a breach of contract that resulted in infringement.
ReplyDeleteOn the recasting point, I don't know what you mean by making a copyrightable object fit for print, but in the photography context, my view is that a photograph of an object does not recast the authorship of the object and that's what the statute requires. Sure enough a photograph of an object might lead us to "see" the object differently, but it doesn't actually recast the make up of the object.
I think people get confused about what a photo is and isn't. I think if you photograph a two-dimensional work of art it will usually be a mere reproduction-- even if you have to do a bunch of work in Photoshop in order to make that photo look good.
ReplyDeleteIf you photograph a three-dimensional work of art, then it must be an adaptation, because you are converting from three dimensions to two. Creative decisions must be made. Or are you saying, Mr. Patry, that a photograph of a three-dimensional item is usually a mere reproduction? If that is the case, what happens when someone makes a photograph of a three-dimensional object that is in the public domain? Is the photographic copy in the public domain as well, and if not, why not?
Thanks Anon. I agree on your point about photographing a two dimensional object, but disagree on photographing a three-dimensional object. I don't think in the copyright sense, you are converting the three-dimensional object; you are not actually recasting or adapting it; you are depicting it. The object remains exactly as it is. I think to make a derivative work of a three-dimensional work you have to alter the authorship in that object and not, as is the case with photographs, create a new and different work of authorship.
ReplyDeleteAs for taking photos of pd works, that's what the Bridgeman case was about, and the answer is whether sufficient originality exists in the photograph: after all, a performance of the First Brahms Piano Concerto which i am listening to right now (Steven Hough, pf), is copyrightable even though the Brahms piece itself isn't.
I think there is at least a legitimate argument to be made that such photographs fit the definition of adapt ("to make fit (as for a specific or new use or situation) often by modification") even if they do not engage creatively with the underlying work, in the way that we tend to think of DWs. For example, a photograph of a commercial good allows the visual properties of the good to be reproduced cheaply and in a variety of media, giving it new marketing value.
ReplyDeleteAvi, sounds like you are taking the Ets-Hokins approach: a photo of a vodka bottle is not a derivative work because the bottle is not copyrightable; but, a photo of a Jeff Koons sculpture is. Under this theory, a video that shows a human being is not a derivative work, but a video that shows a Jeff Koons sculpture is.
ReplyDeleteI think none of the above are derivative works because I think that to be a derivative work it is not just enough to show another work in a new light -- after all a good piece of writing can do that too -- but rather there has to be a change IN the underlying work: how can one recast or adapt something that you only make a picture of?
Sort of. There is no statutory requirement that the underlying work be copyrightable. Certainly, one can create a derivative work based on something in the PD.
ReplyDeleteI guess the real object of my ire is the notion that derivative works should be held to a different standard of originality than other works; either a work embodies the creative contributions of its author or it does not. Accepting that, it would be unnecessary in these cases to mark the line between the reproduction right and the adaptation right, which seems difficult in the context of photography.
I'm puzzled about this. Isn't it true that if a sculptor were to sculpt a 3-dimensional work based on a photograph, painting, or other 2-D work, that sculpture would be a "re-casting", and therefore a derivative work... why then, should the analysis differ when a 3-D work is photographed (i.e., re-cast or transformed into a 2-D medium)?
ReplyDeleteDo I now start advising my entertainment industry clients differently when filming at locations where there are sculptural, architectural, and other works of art in view?
Gordon, excellent example. Yes I do think that making a sculptural work out of a photograph -- as Jeff Koons did with his Puppies -- work constitutes an infringing derivative work, although in that case it wouldn't have mattered if one sued for violation of the reproduction right. And a film that includes without permission 2D or 3d elements may be infringing depending on the facts (and section 120(a)) so I see no difference there.
ReplyDeleteThe difference with 2D to 3D is that there has been an actual recasting of the authorship of the 2D original, and in many ways. But a 2D photo of a 3D work doesn't recast anything.
If a photograph is a reproduction, what's the practical effect of saying it's not a derivative work? Either way the use falls within the scope of the copyright owner's rights.
ReplyDeleteGreg, I agree there is no difference in some of the examples given in the comments to the blog, but it did make a big difference in the case discussed in the blog. The photographer in that case had valid copyrights invalidated based on an erroneous opinion that they were derivative works that contained no separate expression from the "original" work they were supposed based on. Had the court taken the correct approach, it would have held that the photos were not derivative works, and that they had been infringed.
ReplyDeleteBill Patry wrote: "I think that to be a derivative work it is not just enough to show another work in a new light -- after all a good piece of writing can do that too -- but rather there has to be a change IN the underlying work: how can one recast or adapt something that you only make a picture of?"
ReplyDeleteBill, what about lighting design and the critical element of lighting in the authorship of photographs? If the lighting is "on" the copyrighted object and alters the object significantly from its intended appearance, has it been recast in the resulting reproduction? Just saying that it is a reproduction avoids the issue of its status as a potential derivative work because all derivative works tend to be reproductions of the original - - at least in parts.
Josh, not surprisingly I disagree with your lighting hypo; to me such lighting is an element of originality in the photo and has zero to do with altering the authorship in the underlying work, which remains the same when the lights are turned off, and the audience goes home.
ReplyDeleteThen are we to assume that Ansel Adams’s work are derivatives, and are in some part mountain, and thus are unprotectable. Or that they are scenes affair because the subject is certainly in the public domain, and thus not afforded copyright protection??
ReplyDeleteLet me tell you I spent two years in a very intense photography program, where we were each given the same art directive, and compelled to shoot the same subject, based on the same angle, camera lenses, and lighting plan, every day of the week due on the same day the following week. Then each student was required to mount on museum board their assignment, and put it on the critique-board, where upon each image was then critiqued by every student and the professor.
Now we each had to set the shot from scratch, not just walk up and load the 4X5 with a piece of film. The outcome….out of 30 images…. I never once saw the same image…they were all, the same subject…but each photographer added to the assignment his interpolation of the art directive. We all made distinct choices, in the exact angle of the camera and the various angles of the lighting sets, within the approximate angles that were assigned. As well as color and composition choices with in the set of the subject, and any elements that were added in as supporting cast. Now we each had to then process and print our film in the school dark rooms, where we again made additional choices as to the push/pull, temp-controls that change the density and contrast of the base film, as well as make the final print, using controls that we use under the enlarger….burning & dodging, contrast controls, masking, and paper choices.
My last Q”s
1 Did in fact the toy company register their copyright in each of the toys that were filmed?
2 Did the Judge scrutinize the original subjects(toys) to the same degree as the protect ability as the images?
3 Did the photographer add any other elements to the shot, as supporting cast?
4 Didn’t the license they mutually agreed infer the intent of the photographer to seek copyright in the images?
5 Are not the images as transformative as thumbnails in and regarding search engines, are we to assume that the images/advertisments will transplant the toys and then be chosen over the toys to be played with?
A copyright protects only the expression of ideas, not the ideas
ReplyDeletethemselves. See Wildlife Express Corp. v. Carol Wright Sales, Inc., 18 F.3d 502, 507 (7th Cir.
1994); 17 U.S.C. § 102(b) (“In no case does copyright protection for an original work of
authorship extend to any idea, procedure, process, system, method of operation, concept,
principle, or discovery, regardless of the form in which it is described, explained, illustrated, or
embodied in such work.”). Courts have long held that the primary objective of the Copyright
Act is the promotion of arts and sciences and that this objective “is accomplished by assur[ing]
authors the right to their original expression, but also by encourag[ing] others to build freely
upon the ideas and information conveyed by a work.” Wildlife Express, 18 F.3d at 507 (internal
quotations and citations omitted).
I'm unclear on what you mean when you say that "a photograph of an object does not recast the authorship of the object." What does the phrase "recast the authorship of the object" mean? My understanding of derivative works is that they must "recast, transform or adapt" the original work, not the "authorship" of the original.
ReplyDeleteYou also say that a derivative work has to "recast the make up of the object." You clearly don't mean that it has to alter the original work in some way (though you say that "there has to be a change IN the underlying work"; the vast majority of derivative works leave the original in its original state. So what does that mean?
Anon, what I meant is that for a derivative work to be a derivative work, it has to recast the authorship of the work it is allegedly based on. If it doesn't do that, it can't be a derivative work. So I absolutely do mean you have to alter the original in some way. The easiest way to do that is to physically alter the earlier work in the process of creating your work, and that usually means altering the object (copy) in which the original is embodied. So, in the case of a photograph, one could create a derivative photograph by cropping the original and by adding to or deleting material from the original. (Say, deleting one figure in the shot and replacing it with another). But a photo of a copyrighted lamp can't be a derivative work of the lamp since the authorship of the lamp hasn't been changed.
ReplyDeleteIf one doesn't alter the original, how I ask you can you say that a derivative work has been created?
I confess that I'm quite confused by that answer. Isn't a sequel to a novel a derivative work? Isn't Koons' puppy sculpture a derivative work? (They're also reproductions, of course.) The original novel and the original photograph are unchanged - a changed copy of the original has been created.
ReplyDeleteIt sounds like you're saying that only works that actually incorporate the original work (Nagel prints on tiles, etc.) are derivative works. But you obviously aren't saying that.
So what is the distinction between a sculpture that recreates an original and a photograph that that recreates an original? Is it different if it is a painting of an original? A silk screen? Why?
Anon, I don't mean there is a physical incorporation requirement in the sense that once physically cuts out part of a novel and pastes it into a sequel. Koon's Puppy was a derivative work of the Rogers' photograph for a number of reasons: it took a flat black and white work that you could only see the front of and added a number of elements that did not exist in the original: the entire back of the piece, the perspectives were radically changed, wild colors and flowers were added to name only a few. These were changes in the authorship of the photo.
ReplyDeleteLet me ask you (again, gently, nicely, politely), if you don't have to alter the authorship of the original to constitute a derivative work, what is a derivative work?
(Sorry if I didn't seem to be being polite. My expression of confusion was genuine and not intended to be pointed or rude.)
ReplyDeleteI guess my confusion stems from my not understanding what you mean by "authorship," exactly. Obviously Koons made numerous changes and additions to the puppy picture when he made a sculpture. You said that those were "changes in the authorship of the photo," but I really don't know what that means (which was really my initial question). How is that a change in authorship (aside, of course, from the fact that it's the product of a different author)?
A second question is the basic one of whether a photograph can ever change the original enough (change the authorship, as you put it) to be a derivative work. I would think that it can if the author does enough to it (staging, props, cropping parts of it out, etc. - a photograph of a Barbie in a blender is a derivative work).
So, if I'm understanding you correctly, the amount of change (the originality?) is somehow connected to the question of changing the authorship. I'm not sure that I follow that, and I would love it if you explained it further because I fear it may be something basic that I'm misunderstanding.
Anon, you were very polite, not to worry! I was referring to my repeated question to you as hopefully not being construes as impolite, and you still haven't answered: tell me what you think a derivative work is if it doesn't have to alter anything from the original.
ReplyDeleteAs for your question to me, the answer is straightforward: by authorship I mean the contribution of original creative material and in the derivative context, this means the derivative author has to contribute authorship of his or her own that changes the original authorship.
Oh, I think we fundamentally agree about what the nature of a derivative work is - a new work that is based on an original work but is different from (recasts, transforms, or adapts) the original (loosely speaking). I guess our only confusion is in whether a photograph of an original work can be a derivative.
ReplyDeleteAm I correct to think that what you mean when you say that a photograph is not a derivative work is that a photograph that does nothing to change the original is merely a reproduction? If so, what about things like Barbie in a blender (unblended)? Presumably a photograph of that (two objects in conjunction with one another, one of which is copyrighted) is a derivative work, right? If so, then where does one draw the line? how much originality is required? A backdrop? Lighting? The angle of the shot? Cropping? Is it a different level of originality than we require for copyrightability (i.e. minimal)? Why?
Anon, I think we do agree. The Barbie example is a tough one because it may not involve actual changes to Barbie, but instead a contextual presentation of her. Nice hypo
ReplyDeleteA drivative work is also an "abridgment" which requires no additive component other than the decision of detraction and leave the original as it were (if not as it was).
ReplyDeleteIndeed, abridgments are a form of derivative work, but they still require that you alter the authorship of the original by taking away some of it. There are a number of very old English and American cases on abridgments, all of which recognize that there is authorship in an abridgment where the abridger exercises creativity in which parts to leave in, emphasize, de-emphasize, etc. Just copying chunks of the original was considered an infringing reproduction.
ReplyDeleteDon't want to kick-start the entire "altering the original" argument again, but I'm very curious about your specific use of the "direction" of 2D->3D in your 3:11pm post. I can understand the idea of 'adding a back' to a 2D object to make a 3D sculpture, but what about making a hologram of an existing 3D work?
ReplyDeleteThere's nothing in law that distinguishes the nature of a reproduction, derivation or infringement thereby based on the number of dimensions being used, and with greatest respect I don't understand how you can infer it matters. IMO the issue is not with the 'recast transform adapt' phrase (I love T17.. so few words; so many dictionary entries...) but with the fundamental copyright concepts of originality and creativity. Irrespective of the dimensions or means of reproduction my argument would be:-
1) Did your work use the *creative elements* of the original as a fundamental starting point and not simply use it as 'an object' with no care about the creative intent behind it?
2) Did your work add creative originality of your own?
if (1) is yes, it's a derivative or a reproduction based on the answer to (2). If (1) is no, such as a photograph of a sculpture where the photographer was 'creatively' interested in the shadows and leaves blowing in the wind rather than the "artistic meaning" of the sculpture, then IMO it's neither.
Taking the book argument - if I photograph the cover of a book because I specifically wish to use the artwork, typefaces etc in a composition of my own, IMO it's derivative as the original artist's creative output is what I "take" for use. If I photograph a book simply as a stack of paper to depict the concept "this website sells books" without caring which book I use, it's not.
Can something be a derivative work but not infringe the original? In other words, can you have a "derivative work" (according to the definition of the Copyright Act) that does not otherwise meet the test for infringement? Or does the designation of something as a derivative work automatically mean it's infringing?
ReplyDeleteThanks...
The term derivative work is used in two senses: a work that qualifies for protection, and as an infringement. Although Nimmer has tried to collapse the two, I think they are different and should be analyzed differently, although many courts do use the term as synonymous for infringing.
ReplyDeleteHello William! I have a specific scenario , hopefully for you to please decipher for me. In the 80s & 90s my partner and i did a lot of staged hair productions where we designed hair, clothes ,choreography , specific music or sounds for the show and even designed the stage. we during these shows video recorded them for our promotion.directly after these shows we would march the models around the corner to a photo studio where we agreed to let the photographer photograph the models , in exchange he would get to keep copies of the photographs to build his photographic portfolio and we would keep the color slides for our library and our promotion. During these photo shoots we adjusted the models hair etc. in order to portray our work as we wanted it to look,as we are the hair artist and the photos were a representation of our services and an extension of the particular shows.( we also have a service mark registered for our particular logo). well many years later I am planning to use these works as promotion , as was the original plan, but in going through our library of slide film , out of like 4,000 photos 100 or so are stamped with the photographers name, as we will still use them and of course give credit where it is due,who owns these slides?
ReplyDeleteI have a pretty good handle on what constitutes a derivative with my personal art, however I am still unclear about how a derivative designation would be applied in digital 3d modeling. If I start with a basic mesh(structure) and my end product is substantially different through the use of topography, shading ect...is it still considered a derivative when it bears no resemblence to the originating mesh?
ReplyDeleteBlame for some courts' confusion, I hereby joke, shall be placed squarely on the shoulders of the American photographer Walker Evans (1903-1975). Evans recognized that nearly all photographs function as documents, but Evans had esthetic ambitions, *pictorial* ambitions: creating photographs in what he called "documentary style" and attempting to make those photographs "literate, authoritative, transcendent." His work aggravated the easy confusion between the thing pictured and the picture itself. He did this on purpose, confusing people into thinking the pictures were nothing more than "plain facts." That confusion -- a confusion that long pre-dates the internet -- leads to the everyday language that speaks of the photograph as if it *is* the thing pictured.
ReplyDeleteAgreed with the above. There is nothing factual about a photograph. A photograph is like a window, but you can see the glass.
ReplyDeleteSince when is anything about reality based on what film or a digital chip can capture.
This was directly at issue in the Ty case in the 7th Circuit (which you argued). Why did PIL concede that photographs of beanie babies are derivative works of the dolls?
ReplyDeleteHi Julie, I didn't concede that; I argued the opposite, and Judge Posner just got it wrong in his opinion.
ReplyDeleteFirst off, thank you for putting this info on your blog, it is VERY helpful!
ReplyDeleteOK So if I understand this right, if you take a photo of someone else's work it is just a depiction of the object?
I have a kind of an odd question and situation that nobody seems to be able to answer.
What if you take a photo of part of an object and then take that photo, alter it so it looks like a drawing or sketch, and use it as a logo and name for your company?
To clarify, here's the situation. I took a photo of a part of a nacho cheese dispenser that had a button on it that said "Push For Cheese".
A group of us thought it was funny so we used that button and named our company "Push For Cheese".
Our original logo was the photo of the button on the nacho cheese dispenser, which had been altered to look sort of like a sketch drawing.
We started offering merchandise like t-shirts and such through CafePress with that logo but we never sold anything. Someone shortly thereafter mentioned that it could be considered copyright infringement and the makers of the nacho cheese dispenser could potentially sue us for that.
Since then, I have changed the logo to a different one but it's not as good looking as the old one with the altered photograph.
Is this even an issue or is this something that could potentially cause problems?