Skip to content

Plaig! (non-Wegman edition)

Mark Vallen writes (link from here):

What initially disturbed me about the art of Shepard Fairey is that it displays none of the line, modeling and other idiosyncrasies that reveal an artist’s unique personal style. His imagery appears as though it’s xeroxed or run through some computer graphics program; that is to say, it is machine art that any second-rate art student could produce. . . .

Fairey’s Greetings from Iraq is not a direct scan or tracing of the FAP print, but it does indicate an over reliance on borrowing the design work of others. There was no political point or ironic statement to be made by expropriating the FAP print – it was simply the act of an artist too lazy to come up with an original artwork. . . .

Some supporters of Shepard Fairey like to toss around a long misunderstand quote by Pablo Picasso, “Good artists copy, great artists steal.” Aside from the ridiculous comparison of Fairey to Picasso, there’s little doubt that Picasso was referring to the “stealing” of aesthetic flourishes and stylings practiced by master artists, and not simply carting off their works and putting his signature to them.

A last ditch defense used by Fairey groupies is to acknowledge that their champion does indeed “borrow” the works of other artists both living and deceased, but it is argued that the plundered works are all in the “public domain”, and therefore the rights of artists have not been violated. There are those who say that artists should have the right to alter and otherwise modify already existing works in order to produce new ones or to make pertinent statements. Despite some reservations I generally agree with that viewpoint – provided that such a process is completely transparent. . . .

I’m reminded of George Orwell’s classic slam on lazy and dishonest writing:

Each of these passages has faults of its own, but, quite apart from avoidable ugliness, two qualities are common to all of them. The first is staleness of imagery; the other is lack of precision. The writer either has a meaning and cannot express it, or he inadvertently says something else, or he is almost indifferent as to whether his words mean anything or not. This mixture of vagueness and sheer incompetence . . .

Laziness and dishonesty go together, and that fits the stories of Shepard Fairey and Ed Wegman as well. You copy from someone else, and you have nothing of your own to add, so you hide your sources, and this sends you into a sort of spiral of lies. In which case, why do any work at all? In Fairey’s case, the work is all about promotion, not about the art itself. In Wegman’s case, the work all goes into lawsuits and backroom maneuvering, not into the statistics.

Once you’re hiding your sources, you might as well cut corners on the product, eh?


  1. numeric says:

    If imitation is the sincerest form of flattery, what is more sincere than taking ever larger and larger chunks of material from another individual, eventually putting your name to work completely produced by another? Of course, I didn’t think that up–I grabbed it from someone else, who I have forgotten. But they probably borrowed it from someone else.

    • Andrew says:


      An even higher form of flattery is to build upon the work of others to do something new. I’m pretty sure that Don Rubin is more flattered that I have built upon and extended his ideas, than if I’d merely copied his work!

  2. Rahul says:


    I don’t see the causal link between “hiding your sources” and “cut corners on the product”. Do you have any empirical evidence for this?

    I’m not sure what’s a good metric of “product quality” for art anyways but Fairey’s “product” seems to be doing well.

    Why not keep plagiarism as the moral / ethical offence that it really is? When you start making claims that it impairs the quality of product where’s the evidence for this?

    PS. In a related context: If Samsung makes phones by infringing on Apple’s Intellectual Property does that make their phones functionally inferior?

    • Rahul says:

      Alternatively, if 99% of chess-book readers don’t care about following up attributions to the original source, is the quality of a chess book significantly impaired because the author did not scrupulously cite his sources?

      My point is Plagiarism is strictly a moral & ethical problem. Even the law doesn’t criminalize plagiarism.

      The link to quality isn’t obvious. Nor the impact on utility derived by consumers of the work. It seems very possible to make a plagiarized product of a very high quality, perhaps even better, than all the obscure originals it is based upon.

      • Andrew says:


        We discussed this in the Chrissy blog post, but, yes, his book had lots of errors, and I think that had he cited the sources the errors wouldn’t been so glaring. There’s a big difference between the author of a book falsely writing “X happened” and the author honestly writing, “I read in book Z that X happened, but I didn’t bother to check it myself.”

        Regarding your hypothetical example of a higher quality plagiarized book: Sure, if you can do it, fine, but it doesn’t typically seem to happen that way. In the real world we get crap. Probably because the people who are willing to put in the effort to create high quality work also have no problem acknowledging their sources.

        As I’ve written in my article with Basbøll, the moral and ethical problem I’m concerned with regards the reader, who is not given the metadata necessary to follow up on the sources of the stories and learn more.

        P.S. I don’t know anything about Samsung and Apple; you’ll have to go for another blog for discussion about this. I’m talking about the transmission and garbling of information, which is a central topic of statistics, hence my phrase, “To throw away data: Plagiarism as a statistical crime.”

        • Rahul says:

          In the real world we get crap so many ways that sure, some of the crap is indeed plagiarized crap.

          But I don’t think you have any data to show that plagiarism leads to “crap” in any way more often than just non-plagiarized crap.

          My point is that the argument “Plagiarism leads to a significant utility loss for the readers” is bogus. At least I don’t see you show any data to support your claim.

          Plagiarism is pure and simple an ethical offence. Just because we, as a society, think that using without attribution is unfair to the original creator. And because plagiarism interferes with our systems of academic reward allocation. To bring quality into this argument seems a stretch.

          • jrkrideau says:

            Plagiarism is pure and simple an ethical offence.

            No, Plagiarishm is purely and simply theft. Of course, it is a moral and ethical offence as well.

            Steal someone’s idea and get the job
            Steal a song and get rich
            Copy Hill & Gelman and publish under your name.

            No, plagiarism is not just …anything

            • James Annan says:

              No, theft is the taking of personal property without permission, with the intent to permanently deprive the owner of it. Neither plagiarism nor copyright infringement are theft. Copying is not taking, information is not property and the owner is not deprived of it, even temporarily.

              • Rahul says:

                Exactly right.

                If anyone is so confident that plagiarism is theft, how come in the history of plagiarism none have been penalized by the courts?

                Even petty theft goes punished but never plagiarism.

                Plagiarism is not theft. It never has been under US law, nor under any other legal system that I know of.

          • Corey says:

            Plagiarism by definition erases the provenance of the copied text or ideas. This harms readers who care about said provenance for whatever reason, e.g., historical interest, finding work by the same (original) author, tracking down the source of errors in the record, etc. This isn’t just about giving credit where it’s due.

            • Andrew says:


              Exactly. It’s an important but not always recognized principle of statistics that you don’t just need the data, you also need to know where the data came from and how they arose, so as to be able to account for sampling error, measurement error, selection bias, etc. More generally, there’s a big, big difference between “Here’s a story” and “Here’s a story and here’s where it came from.”

              • This morning my six-year-old and I were discussing whether a monkey that steals a bag of chips knows that it’s doing something wrong. That doesn’t really illuminate this discussion, but it does seem like Andrew & others are making a strong distinction between stealing objects and stealing ideas. If someone steals my bike, I don’t object because of some vague sense that other people might suffer from not knowing the proper provenance of the bike, but rather because it’s *my* bike. I’m being denied the use of it, and I’m being caused pain because of its absence. None of us, I think, have any trouble with the idea that theft of objects is bad (unethical). Most of us, I would have thought, have similar notions about the theft of ideas.

                Yes, it is the case that plagiarism robs data of their context and origin. But this is not the only reason that plagiarism is bad. If a colleague comes up with a new analysis algorithm, and I steal it and present it as mine, I’m certainly denying the reader the opportunity to ask my knowledgeable colleague about the background of the idea. But perhaps my colleague writes awful papers, devoid of historical context and clear writing, and is moreover a misanthrope who doesn’t answer questions, and so the net benefit to the “audience” is clearly positive due to my theft — would this justify plagiarism? I certainly hope not.

              • Corey says:

                Your bike is what economists call a rivalrous good, which makes its theft a poor analogue of plagiarism. And giving credit where it’s due is clearly important, perhaps the most important thing plagiarism destroys. It’s just not the only thing.

              • Corey: I agree with your statement: “…giving credit where it’s due is clearly important, perhaps the most important thing plagiarism destroys. It’s just not the only thing.” My main motivation for writing my comment was to poke Andrew for greater clarity, since I can easily read his statements as implying that the really important thing about plagiarism is its obfuscation of the source of data and ideas — for example the above, “Regarding your hypothetical example of a higher quality plagiarized book: Sure, if you can do it, fine, but it doesn’t typically seem to happen that way.” This confusion existed in the earlier post & its comments on “The plagiarist next door” as well.

                If one wants to argue that the main problem with plagiarism is that it’s theft, and that as an added drawback it removes the reader’s ability to assess the origins of data, that’s one view (that I agree with). If one wants to argue that the main problem with plagiarism is the issue of the reader being separated from the origins of the idea, that’s a different view. I don’t actually think Andrew is pushing the second view, but I’m not certain. I’m clearly not the only confused person, as evidenced by more than one comment here and earlier. Does it matter? Probably not, but the different views would lead to different things we tell students, for example, about lifting text in assignments, journal articles, etc.

                I would also find the second view hard to explain to my six-year old, but whether that’s important is debatable…

              • Andrew says:


                There are many problems with plagiarism, which is no surprise given the vehemence by which plagiarists deny, even when they’ve been caught red-handed. As a statistician and social scientist, I’m particularly interested in the corruption-of-metadata angle.

                By the way, in your hypothetical example of stealing someone’s work and presenting it as your own, when you say the net benefit to the “audience” is clearly positive . . . OK, but the net benefit would be even more positive if you cited the source and explained where your idea came from. And, it’s funny, the people who plagiarize typically don’t present the work better. It always seems the opposite, as when Ed Wegman copied from Wikipedia but couldn’t be bothered to fix the exponentials—and then his journal tried to charge people a couple thousand dollars in subscription fees to see the copied work.

  3. John Mashey says:

    To me, it seems 2 different kinds of plagiarism are represented here.
    1) The sort where work is presented as creative/original, but may not be, from laziness, or whatever.

    2) Some of the Wegman/Said cases seem like that, but the Wegman Report and the retracted Social Network Analysis paper seem different.
    a) The WR plagiarized (and sometimes falsified) paleoclimate text from (expert) Ray Bradley
    b) It plagiarized text of the key papers, sometimes with changes of meaning.
    c) It used plagiarized text on Social Netowrk Analysis from 2 textbooks and Wikipedia. That was also in the retracted paper.
    d) And there were some smaller chunks.

    From the introduced errors and misinterpretations, it was fairly clear they had little or no knowledge of paleoclimate (a+b) or SNA (c), but wished to make strong statements related to them. Hence, the purpose seemed to be more to provide that material to claim nonexistent expertise, as opposed to original material. A similar effect appeared in McShane and Wyner(2010).

    Maybe that was a form of laziness … but there was not way Wegman’s team were going toe become paleo and SNA experts in the short time

    Of course, they *could* have just redone MBH99’s analysis with different PCA centering … but would have discovered it made no significant difference … not the answer Rep. Barton wanted.
    (Most of this is in Strange Scholarship in the Wegman report.

    Anyway, I do think there is a difference between
    a) An essay or art or something that is supposed to be original content, and isn’t.

    b) Plagiarized text not claimed to be original, but whose antecedents are hidden to hide lack of expertise. That seems related to bibliography-padding to claim false familiarity with the relevant literature.

    The 2009/2011 plagiarisms in their Wiley WIREs:CS journal (including the one that converted 2^n to 2n, to Andrew’s glee), seemed more like laziness. They were able to do better when allowed to quietly rewrite, but I don’t think they were trying to create and illusion of expertise to support contentious analysis.

    That’s the mess that eventually led to the lawsuits that Andrew alluded to.

  4. Somehow I doubt anyone here cares to rehash old points so long after the fact, and I’m almost positive nobody wants to read John Mashey’s 250 page document (I feel sorry for anyone else who has actually forced themselves to do so), but I can’t let this comment go unanswered. While it is unquestionably true work Wegman was responsible for contained significant amounts of plagiarism, Mashey says things like:

    a) The WR plagiarized (and sometimes falsified) paleoclimate text from (expert) Ray Bradley

    Which are far less true. Back during a humorous time when the USA Today gave credit to Mashey for some work he didn’t do (in addition to the work he did do) in exposing the plagiarism, which somehow didn’t bother anybody, I pointed out Mashey had misrepresented Ray Bradley’s work to claim the Wegman Report inverted its conclusions as the only example of the report supposedly misrepresenting a source in a way which inverted its conclusions. I’ve pointed it out a number of times since, but despite even participating in that first discussion, Mashey has never addressed my criticism, and to this day he continues to repeat his false claim. Similarly, he says:

    Of course, they *could* have just redone MBH99’s analysis with different PCA centering … but would have discovered it made no significant difference … not the answer Rep. Barton wanted.

    But this is also false, on multiple counts. Primarily, it is beyond dispute Wegman did in fact redo the calculations Mashey claims he didn’t do. Mashey has long tried to pretend Wegman only copied the results output by other people, but he does that by only looking at one figure. That figure happened to be generated from an archived data set rather than a freshly run set of results (the names for the two were very similar, so it would be easy to mix them up). Other figures in the Wegman Report, however, were clearly generated from newly run results, a fact Mashey has long ignored.

    As for the claim it would have made no significant difference, this rests on the notion the displayed results were “cherry-picked.” The originally displayed results were chosen and labeled as non-random examples. Had Wegman realized the mistake about which data set his results were being taken from, he may well have not used “cherry-picked” results. And had he not, he would have still gotten a figure which showed the inappropriate PCA methodology introduces significant biases.

    There’s a lot more which could be said, including responses to several points I can predict Mashey will make in response to this comment. This comment is already long enough though, and I don’t think people here really care. So I’ll just leave it at that. Mashey isn’t just wrong on things; he’s wrong in ways that makes it feel like I’d be picking on a child to point them out.

    But who knows? Maybe one day I’ll show up in one of Mashey’s 200+ page documents about conspiracies. It’s always been a dream of mine!

  5. John Mashey says:

    1) If you haven’t seen it, German researchers have been doing some good work on software tools, as in Visual Assessment of Alleged Plagiarism Cases. I had some nice email discussions with Martin Potthast there. Of course, Debora Weber=Wulff’s False Feathers: A Perspective on Academic Plagiarism is well wroth reading, including their community efforts in finding and documenting plagiarism.

    2) Over time, there may well be progress in being better able to mechanically detect graphical/image plagiarism of the sort you describe. At Hot Chips we just had a tutorial on “deep Learning”, including some of the image-processing issues, an area seeing great progress given the cheap parallel hardware found in GPUs.

Leave a Reply