We spoke with Dr. Jeremy Trevelyan Burman, PhD for this month’s Spotlight on Science Q&A. Dr. Burman was recently named to a tenure-track position supporting the new Reflecting on Psychology graduate program at the University of Groningen in the Netherlands. Here, he reflects on his article in Perspectives on Science: “The misunderstanding of memes: Biography of an unscientific object, 1976-1999.” This has been the journal’s top download almost continually since it was first published in 2012. Read the article for free on our SOS page.
What brought your attention to this subject matter, and when did you realize that the meme had such a complex history?
I wish I had a heroic story to tell. The truth is rather less exciting: the memes paper was written originally for a doctoral seminar when I was at York University in Toronto. Our focus there was on the Historiography of Psychology; the doing of history, rather than of its content. So this essay wasn’t what I set out originally to write. It was my Plan B (or maybe my Plan C). Now, though, it’s my most popular piece; my best-ever failure.
My original intent was to do something totally different: to build on my master’s training in the History of Biology and look at how 17th and 18th century French naturalists had catalogued different species of flowers according to their smells, rather than by visual inspection. I wanted to riff on themes presented in a wonderful book by Corbin that we had discussed; to show how something taken for granted today as a part of the disciplinary core of Biology—species groupings and scientific nomenclature—could have gone a very different way if the dominant sensory mode had been different. But it didn’t matter how I approached the project. The primary sources I found couldn’t support what I wanted to do. (Certain plants were “smelly,” and others not.) So I threw the whole thing out and started again.
My normal way of writing takes time. I look for a topic that interests me, and which relates in some way to my other work. Then I follow my intuition as I dive into the primary sources. Yet with my original project thrown out, this time, that was no longer possible. So I drew on an old interest to afford a new approach.
I had read Dawkins and Dennett when I was an undergrad at the University of Toronto, and then had fallen completely for Hofstadter—whom Dennett had talked about with great approbation—while reading the GEB to supplement a higher course in cognition. This then led me to read everything of theirs that I could get my hands on, including The Mind’s I (a collection that Dennett and Hofstadter had edited together).
This, as it turns out, was an important early book in terms of popularizing Dawkins’ ideas about memes. It was a bestseller well before his own works achieved the status or influence that they came to have. And it’s an early source of the view that the meme is something more than a metaphor. (I don’t know if it was the first bestseller to do this, but it is the earliest one of its stature that I’ve found.)
I remember noticing that the version of the meme presented in The Mind’s I seemed different from the one in the first edition of Dawkins’ book. The essay itself also included a series of asterisks to break up the text. But I didn’t think anything more about them at the time.
It wasn’t until I needed a new project for my seminar, many years later, that I returned to the asterisks. I carefully checked the two versions against each other. And this led to the discovery that the more popular version of the text had drawn from multiple different passages in the original without indicating clearly or transparently that it had done so. Instead, a footnote says only this: “Excerpt from The Selfish Gene by Richard Dawkins” [p. 124], as if the text presented in the popularization had appeared in the same form in the original. That’s what then led me to create the Table on page 83 of the article.
The importance of this table can’t be exaggerated. It provides direct and concrete evidence that changes were made in the re-presentation of the original argument. The result of those changes is then the appearance of a different sort of claim in Hofstadter’s and Dennett’s popularization than in Dawkins’ original: although the words are the same, their implications are different. This seemed like something worth following-up.
At the time, and I think mostly as a result of reading Hofstadter’s reflections on translating the GEB, I had become interested in the problem of translating “meaning” for “understanding.” (I’ve since published several other pieces about this problem, from several different perspectives, and my doctoral dissertation delved into it more deeply still.) This is also what got me interested in the “public understanding of science” as a discipline in itself. Applied to the question of the popularization and understanding of memes, these inspirations then seemed to fit with my goals and interests for the seminar paper.
The result was not a philosophical examination of the meme concept. Indeed, Dennett was quite dismissive of my approach. (I wrote to him and Hofstadter prior to publication because he had written elsewhere that this was the polite thing to do.) Of course, this response was very disappointing to me. But I still benefited: I rewrote some of the paper to take his feedback into account. And I suspect now that his reaction was the result of differences between our respective disciplines. Simply put, his preference is not for history. And I’m not a philosopher.
What do you consider the role of the internet to be in relation to creating, storing, or sharing human culture?
Tim Berners Lee is the 20th century’s Johannes Gutenberg. Inventing the World Wide Web was a very big deal. Still, though, without readers (and something worth reading), nothing happens. The internet is powerful as an enabler, but its existence is insufficient for ideas to move, develop, and evolve.
In other words, the internet is not in itself a “meme pool.” It’s a place where ideas can be reproduced, and through which they can be spread. But they do not do this by themselves. That reflects exactly the sort of misunderstanding that I had hoped to engage in my paper.
To be blunt, there’s no such thing as a meme. That we believe that there is a such a thing, however, is itself the result of an interesting process. And this is actually what we probably ought to be talking about, especially in considering the role of the internet. Getting to it won’t be easy, though. There’s a lot in the way.
To speak of ideas spreading virally, or jumping from brain-to-brain, was rhetorically useful in the context of Dawkins’ original purpose. But his argument was about genes and memes both being examples of “replicators,” and furthermore that it is the properties of these replicators that afford the material causes of evolution.
This original argument is part of what was lost during the book’s popularization. And that’s a shame. I think it was fantastically important. I hope Dawkins will return to it in the forthcoming extended edition of the book.
To paraphrase, the set of things with the right kinds of change-enabling properties—exemplified by genes, but not limited to genes—will show evolutionary change within their own respective domains under certain circumstances. Thus, evolutionary principles can be understood to apply outside the restricted disciplinary constraints afforded by the synthesis of Darwinian natural selection with Mendelian particulate inheritance.
If you accept this conclusion, the hard part is then to figure out how the relevant principles apply in each of the domains you want to study. That, though, is where the memes project really failed: it’s too beholden to Mendelian meta-theory (viz. particulate inheritance), and thus is not really flexible enough to accomplish what it seems to me that Dawkins had in mind when he wrote about the role of imitation in cultural evolution.
If you want to study culture from an evolutionary point of view, then you need to return to evolutionary first principles. What matters most—in terms of which principles apply in what domains—is then a philosophical question. (Dennett engages with this in Darwin’s Dangerous Idea.) Historically, though, it’s pretty clear: the meme was not the message of The Selfish Gene. It’s also clear that its subsequent popularity was not because the meme was so innately virulent and transmissible. That came as a result of an entirely different process that we’re blind-to if we adopt the meme’s point of view.
To riff on the work of a fellow Canadian, the original message was massaged. And of course, in this case, the medium is ultimately “people.”
Do you believe that it is possible to use the term meme as it was intended by Dawkins, metaphorically, in scientific writing now? (Or should its meaning be dictated by how it is generally understood?)
Not easily. This is because English isn’t policed in the same way as, say, French. We can’t dictate meanings in the same way. We can only be aware of them, and then act well or poorly.
As it happens, this is partly what it means to belong to a discipline. And that, in turn, is what I think Dennett didn’t like about the version of my paper that he saw ahead of print: I didn’t follow the norms that his discipline assumes. As a result, I treated the concept differently than he did. (But not, as he said, “unseriously.”)
Rather than interrogating a stable philosophical concept, I followed a changing historical object. Obviously, my doing this isn’t original. It’s just how we do history of science. But taking that approach also then afforded a big part of my objection to the philosophical concept: what we call “meaning” isn’t innate to the idea that appears to be moving around.
If meaning were encapsulated in a particulate unit that could self-replicate, as an overgeneralization of the Modern Synthesis would suggest, then—barring the rare mutation—everyone would always understand the same thing of the same words. This, however, usually only happens if the audience belongs to the same group (at the same time). What’s shared, then, isn’t so much the words themselves; it’s what they imply.
This shouldn’t be the case with memes. They shouldn’t mean different things to different people. Their meaning should be identical with the consequences of the particulate imparted. But that’s not what we find when we look.
Of course, the theory of genes on which the idea of memes is based has also changed. For example, we don’t believe anymore in Beadle’s and Tatum’s “one gene, one enzyme” hypothesis, or its subsequent variations (esp. one gene, one outcome). Adherents must therefore change their theory of memes to accommodate the change in the meta-theory of genes. But that’s not the sort of change we see occurring in the popularization. That’s something else.
To answer your question, therefore, it seems clear to me that we must distinguish between three very different uses of the word: the original proposal (replicator), the popularization (idea virus), and the contemporary (cat pictures with funny captions). Each can be examined rigorously on its own terms. How you then do that depends on your disciplinary allegiances.
Can you say more about these three different meanings?
Sure. The meme as introduced by Dawkins in the first edition of The Selfish Gene was part of a scientific argument. To wit: evolution is driven not by the gene, per se, but the set of evolution-things to which genes belong.
The main rhetorical flaw in this argument, Dawkins realized, was that he had only one example: genes. He needed a second replicator. He then asked a philosophical question: can we imagine another member of this proposed set that would have the properties that could cause something recognizable as evolutionary change? That’s when the original version of meme was introduced.
The original meme is an imaginary object, and not—strictly speaking—a scientific one. Note, though, that Dawkins presented his case for memes in a similar way as Darwin had done for the hypothetical “gemmules” which he suggested might be responsible for inheritance before anyone had heard of Mendel.
Of course, Darwin did this without knowing about genes. He did know, however, that inheritance was a problem that needed to be solved. So he suggested some possible causes. Gemmules were then later replaced by genes during the Modern Synthesis, and afterward were largely forgotten.
In the form discussed by Darwin, however, it’s clear that gemmules weren’t “real.” This then helps us to see that the meme wasn’t ever really real either. But that’s okay: neither proposal needed their object to be real for the purposes of the text in which they were presented. Indeed, in the original version of Dawkins’ argument—which is clearest in the first edition (and as he has since reiterated several times)—the take-away wasn’t supposed to be about memes at all. It was about replicators: things that have the properties of genes.
To move the metaphorical meme a further step toward a science of “memetics,” you have to ask another kind of question: is there such a thing? But, no.
There never was a Rosalind Franklin of memetics, although Dennett certainly came close. His version of the meme, however, isn’t the form that was popularized. It’s also very philosophically complex; more method than object.
The meaning of meme that people usually mean today, when they aren’t referring to cat pictures, was provided by Blackmore. (If Dennett is this story’s Franklin, then she’s its Watson or Crick.)
By citation count, Blackmore’s book is the primary text of the popularized meme; the expansion of the bricolage presented in The Mind’s I, in turn made possible—and then endorsed for its “courage”—by Dawkins himself. Indeed, its publication is why the date range in my article runs until 1999 when it could easily have extended further. The peer-reviewed Journal of Memetics was still actively publishing at that time, the secondary literature was growing by leaps and bounds, Dawkins’ The God Delusion and Dennett’s Breaking the Spell had not yet been published, etc.
Briefly put, though, I think a case can be made that the popularized view of memes satisfies the criteria for what Lakatos referred to as a degenerating research program. This is supported by a number of observations beyond the sequence of three different meanings. For example: the journal failed in 2005, there are no graduate programs now training students specifically in memetics, and the contemporary meaning of the term—relating to cat pictures with funny captions—is in no way related to the original source.
Dawkins is still name-dropped in meme-talk today, of course, but it’s clear that no one who speaks in this way has read him carefully. (Note that the student newspaper that reported in 2011 on Northwestern’s interdisciplinary memes major is satirical.)
Interdisciplinarity is regularly found as a scholarly ideal in academia nowadays; do you think that there is an increase in the risk of misunderstanding as a result?
Although I accept that the word is used, I’m not fully convinced that “interdisciplinarity” actually exists either. What we actually have is a kind of openness that has been institutionalized to a greater or lesser degree in individual departments, faculties, and schools. And that’s a good thing. But it’s not identical with the use of the word; it requires action, support, and sustained effort—by people.
One important driver of this institutionalized interdisciplinarity is a good and careful editor who is supported by good and careful reviewers. This makes it possible for scholars in more conservative places to produce texts—perhaps even on their own time—that are deemed by a recognized disciplinary institution (the journal) to be of sufficient quality to merit continued recognition and support. In this way, new niches can then be constructed.
Misunderstandings are inevitable when moving across boundaries, but openness can serve as a shield against this too. It simply requires good faith on all sides. And the careful appreciation of different perspectives.
If reviewers assume that something doesn’t make sense to them because it’s bad, then rejections will result (false negatives). This is the opposite of openness; risk aversion run amok. The danger, though, is on the other side: if an author acts in bad faith, and submits rubbish that nonetheless hits the right notes, then some material will be published that shouldn’t be (false positives).
I don’t know how to strike the right balance, unfortunately, except to do as my father has always suggested: “walk softly and carry a big committee.” If you ensure that you have access to a diversity of voices, and empower everyone so that they’re properly heard, then the odds are much better that someone will flag a potential problem before it becomes an embarrassment.
Is it possible that readers will learn to approach texts, particularly scientific ones, with a wider lens, outside of their own context?
In terms of important funding-relevant buzzwords, what we’re talking about here is kind of “cultural competence.” You can develop this by reading widely, and through direct immersion in other cultures. But the easiest way to teach it is through history, because the sources are in the same language.
The first step is simply the recognition of difference, rather than ignorance. The easy example that I usually give, in this connection, is the meaning of “genetic.”
This is obviously an important word in biology, but it’s also one that’s important to psychology. The meaning-difference is then most obvious when you realize the quite venerable and still-active Journal of Genetic Psychology is about development, not evolution.
This is an older meaning of “genetic,” not a disciplinary difference between biology and psychology. Darwin used it. So did G. Stanley Hall, James Mark Baldwin, and Jean Piaget. And it has nothing whatsoever to do with genes. Indeed, this is why Piaget’s research program—which he called “genetic epistemology”—is a “constructive” theory, rather than a “maturationist” one.
Accepting difference changes your outlook, no matter whether the cause is contextual, disciplinary, linguistic, or historical. Rather than focusing on how people are wrong to do something differently than you do, you start looking for the local norms and standards. Often, this then shows you things you couldn’t see. And then you discover that it was you who were ignorant, not they.
Fortunately, this is a skill that generalizes. I don’t think it’s just about approaching scientific texts, although obviously you can’t hold Aristotle’s writings on physics to the same standard you would apply to discussions of the Higgs Boson.
Aristotle lived in a different world. Our task, as historians of science, is to understand that world—not judge it.
Are there other examples that stand out to you where public understanding has inflected scientific meaning?
There’s are whole fields of study that are about exactly this. Where the phenomenon gets especially interesting is in the realm of transient mental illnesses. These are cases of real psychopathology that arise because what you understand about yourself affects how you experience yourself. (Although still metaphorical, this is where talk of an “idea virus” might possibly be more acceptable.)
I’ve actually put a syllabus together on this theme, and plan to submit it to the faculty here in Groningen for consideration as part of our MA program next year. Without knowing that I intended to do this, and unprompted to say anything about it specifically, every single one of our new students also just expressed an interest in the subject during orientation.
It’s an exciting time. I expect we’ll see a lot more reflection in coming years, and more critical thinking about how psychology can be stronger as a science. Clearing away some of the conflicting colonizations of psychological territory by other disciplines is just part of how we’ll do that. Considering the public understanding as being of different publics, rather than one single mass audience, will be another.
Stay tuned….