Skip to content

The Patron Saint of Superheroes

Chris Gavaler Explores the Multiverse of Comics, Pop Culture, and Politics

[Update: SCOTUS ruled on May 18, 2023, which I discuss here.]

Last week I looked at two pixilated versions of the same Miles Davis cover art photograph. I think the first likely violates copyright, and the second would fall under fair use.

Why? Because, unlike the Prince, Obama, and Soglin examples discussed two weeks ago, the differences between the mid-range pixilation image and the original involves more than just simplification. The remaining details are also distorted in relation to the corresponding areas in the original. The size of units in ratio to the size of the object that the units combine to represent may matter. Once the size of the pixel-like squares is greater than the regions from the original that they represent, the squares become the primary element of composition. The resulting effect combines simplification with exaggeration.

The squares in the image that triggered the lawsuit are akin to paint strokes. They are the micro-level units that combine to create the gestalt effect of the macro-level image. I suspect the nature of smaller units doesn’t matter matter legally because resemblance occurs at the macro-level. This image, for example, has no representational relationship to its source material:

But when viewed at a different size/distance and within a larger image context, it resembles and so represents my right eye:

If I were suing myself for copyright infringement of only my right eye, I suspect I would lose.

Micro-level units matter when their qualities become dominant, including the macro-level effects they create in combination. Look at three pixilated Miles Davis hands:

The first is from what I identified as the first likely fair-use version of Baio’s Davis series last week. The middle is from the image that triggered the lawsuit, and the last is from the source photo. All three are made of pixels, but only the first two are considered “pixel art,” which is a misleading term since each large square is made of multiple identically colored pixels arranged in the shape of a significantly larger square to produce the effect of an enlarged pixel.

I also suspect that the top hand would not resemble a hand out of context, while the other two would be more identifiable. That could matter legally. The pixels that combine to suggest the qualities of Prince’s, Obama’s, and Soglin’s faces are essentially identical in the source images and the adapted images.

Fairey’s HOPE does involve some alterations, but the effect does not alter a primary experience of resemblance between the original and the adaption. The more pixilated Davis image moves into different terrain.

It moves into the top right of my four-area self-portrait by combining both simplification and exaggeration. Though there may be cases when an image that only simplifies its source is protected by fair use, I suspect the region generally is in legally dicey waters.

That leaves one area, the top left: exaggerated but not simplified.

And that, not coincidentally, describes my Prince art:

This is also helpful for discussing process, which is often in focus during court cases. And process differentiates actions (performed by an artist) from effects (experienced by viewers). I suspect viewers experience the Prince image as though it were hand-painted by an artist who was looking at the Goldsmith photograph as a visual reference. It wasn’t. The process is similar to Baio’s pixilated Miles Davis series.

I began with a digital version of the photograph, slightly pixilated after enlarging. I then selected a mouse-scribbled jigsaw shape, copied it, and pasted it imperfectly over the original so as to duplicate content along certain edges and obscure other content at opposite edges. I did this multiple times with multiple jigsaw shapes, while periodically saving works-in-progress:

After I settled on a final version in Word Paint, I opened the document in Adobe Illustrator, and saturated the colors, performing a final copy and paste in Word Paint to combine areas.

Although the “raw material” is still entirely the original photo, the final effect is different from Warhol’s adaptation because the placement of the duplicated material exaggerates certain areas, creating caricatural-like facial features. Even though my authorial intent wasn’t initially parody, parody arguably emerged during the process. Regardless, I suspect the resulting macro-level resemblance between my image and its source is sufficiently distant to fall under fair use due to the range of distinguishing exaggerations.

There is still some gestalt resemblance, since the question of infringement wouldn’t come up otherwise, but despite a process that uses the source as digital raw material, the two images bear almost no similarities at the micro level. Looking again at only right eyes reveals fundamental dissimilarities and so presumably no plausible infringement:

Distinguishing between micro-level units (such as paint strokes and pixels) and macro-level effects (experienced only when viewing an entire canvas) is similar to the differences between words and paraphrasable content. Copying words verbatim is a form of plagiarism, but it is still possible to plagiarize without reproducing any words. Ideas are copyrighted, not just the words that constitute them. If I express essentially the same idea (or gestalt image effect) using entirely different words (or paint strokes), I’m still potentially plagiarizing.

Look at the pixilated “Kiss” again. It’s taken from the 2015 film Eadweard about the 19th century photographer Eadweard Muybridge. The film recreates many of Muybridge’s images, including the moment represented in my hyper-pixilated “Kiss.” Since the film uses entirely different actors, the micro-level “raw materials” are unrelated to the actual individuals who appear in Muybridge’ work, but were Muybridge’s photographs still copyrighted, the film recreation would likely be an infringement. My “Kiss” could potentially infringe on both the 2014 film and the original 19th-century images–except that the level of simplification and exaggeration is so extreme, resemblance is minimal. The image is much more about its style (because of the size and shape of micro-level units) than its representational content.

And look at my second self-portrait again. Although the emergent “new expression, meaning or message” seems less prone to parodic effects, I suspect the complete dissimilarity at the micro-level and the partial dissimilarity at the macro-level would place the second image within fair use:

But I could be wrong, since the courts have yet to establish any clear standards, and decisions keep establishing contradictory or ambiguous precedents. Although the 2nd Circuit of Appeals recently warned judges to “not assume the role of art critic,” until the judiciary develops the necessary expertise of art criticism, fair use will remain in legal chaos.

[Spoiler Alert: This somehow evolved into a four-part analysis (one, two, three, four), with an on-going artistic coda starting here.]

%d bloggers like this: