Anyone else surprised at how few memes are on know your meme?


4k confirmed memes and 20k submissions.

I expected there to be more.

A far more interesting question is what proportion of memes total do you think this represents? (Say a an image/video that has been seen by at least 1000 people)


If you go to the meme category, there are actually fewer:

KYM has articles for things that I wouldn’t really consider memes proper: notable people, events, websites, subcultures, etc. Nevertheless, I feel like that’s not too far away from a number I’d expect — or at least, the number I’d expect in their confirmed section, given that the criteria are a mix of article quality and also notability. There are a further 5.8k articles in the deadpool as well, and on any given day a number of these would probably meet the ‘notability’ criteria, but haven’t anyone there to take stewardship of the article.

I think it’s also important to note that the rate of meme proliferation has also increased ridiculously over the past few decades. The Net certainly has not been chugging along at today’s rates since the early 2000s. You can get a grasp for this by looking at some of the earlier memes on the site As far as I can tell, the first KYM article is on the Rickroll; even though that meme was from 2006 or 2007, it still has a lot of purchase today. From the same quick chronological search there, I think All Your Base is probably the one thing with the earliest-identifiable point of ‘crystallisation’ into a meme proper. If we’re thinking in terms of Advice Animals, early image macros, and rage comics, there were probably about a hundred or less distinct characters and templates (which is a lot, if you think about it), and these things dominated the scene for years and years. But really, all this is just to say that at a concept/template level, there has been a very steep growth curve over the past 8-12 years.


Without giving it much thought, I would say this just means that either the submission / reviewing method they have is ineffective compared to other websites such as, say, RateYourMusic, or that they don’t have many moderators reviewing their content, or maybe just that the meme field doesn’t come off as encyclopedic as music for the standard user, so not many people dedicate themselves to working on said meme database.

Thinking a little more about ‘notability’ —

If you look at the deadpool, you see a lot of entries like this:

I have no doubt that this image was seen by at least 1000 people, but I’m not convinced that it is a meme proper. At present, it really just seems to me like a mash-up or crossbreed; a member of one or both of its parent genera.

This doesn’t mean it couldn’t become a meme-unto-itself in the future, though. It is at least conceivable for “Ice Age Baby Mike Wazowski” to have gotten enough traction to take on life as a character of his own, sort of like Pepe or Doge (actually, I think the best example would be Lord Marquaad).

What this tells us is that there’s a certain level of ‘critical mass’ or ‘critical attention’ that a given memetic artifact needs to reach to metastasize beyond the level of one-off joke or reference, even if these artifacts are getting utilised in the practice of ‘memeing’. This attention also needs to be of the right, fertile type for ‘memeogenesis’ and continued survival — I still believe that it’s worthwhile to distinguish simple virality from mimesis (and memeogenesis), although they are related.

This is all pretty tangential, but I think my takeaway is basically that you need ‘notability’ criteria in an endeavour like this, and not only as a practical matter. Ideally these criteria would be trying to approximate criteria which are out there somewhere in collective consciousness(es)*, which are helping people (even at a n ad-hoc, subconscious level) to discriminate between classes and members.

* I specify that there might be multiple ‘consciousnesses’ because these subjective criteria could even differ from subculture to subculture; notability could be radically relative. You can have memes that totally blow up a small community that nevertheless have no impact on the mainstream. This is related to the ‘rate of proliferation’ point in various ways — most notably, I think, in that fact that there is no longer a mostly-singular, mostly-homogenous meme community anymore. Another way of putting it is that the memetic common-stock of behaviours and ideas has gotten comparatively smaller over the years. The Internet population grew while memetic-literacy barriers remained fairly rigid, or became antiquated; I think what has happened is that people who couldn’t or didn’t want to into that part of the culture just ended up forming their own communities, with fewer connections to what was previously considered the cultural commons. Here is an abominable and nonsensical diagram:


I have a friend that basically worked on KYM, and from what I’ve been told, KYM covers memes which source is known. And afaik the curator is not that many to begin with, so the submission process is kinda slow, moreover there will be evolution or advancement of existing meme and they also need to clarify it. Kinda like Wikipedia but more niche.


inb4 “KYM failed project”

curating a wiki project is a lot of work, and tracing the origin/development of memes often in the absence of documented histories/geneologies is really difficult

this becomes harder again when many memes are developed in closed FB groups and discord servers or high-volume imageboards, added with the high effort required to create wiki submissions without any tangible reward for editors


As memes have become totally mainstream, the act of studying memes has become more high prestige in the eyes of the general public than it was in the previous decade (or the one before that), if only for the fact that more people know about it. A lot more academics have started paying attention to memes over the past few years. KYM has also become far more popular and trusted as a consequence.

They've been expanding a lot, and have hired some major YouTubers and TikTokers to make meme explainer videos for them (viz. Know Your TikTok, owned by KYM; and Lessons In Meme Culture, who contributes videos to KYM).

But the way that they categorise and research memes is based primarily on giving a name to a meme and then collecting things that roughly seem to fall under the name, meaning that it's invariably morphological in terms of classification; the ultimate judge of whether two pics are the same meme on KYM is whether the audience would identify them as such. One alternative approach is TVTropes which creates categories based on tropes and then adds various works into lists that appear on the pages for the tropes.


@Seong, if you would call KYM's approach "morphological", what would you call TVTropes'? What other -logies are there when it comes to classifying memes?

I feel like this is the kind of thing you may have already discussed in an article.

1 Like

I'm not sure what we should call it, actually. Behavioural? Physiological? Functional? By morphological I mean something like formalism in art criticism (the view that the form, or the visible aspects of the art-object itself, is what the critic should focus on), but more lenient and less rigorous.

In biology, morphology focuses on the form and structure of an organism (cf. physiology, which focuses on the function of the organs). To further extend the analogy, formalism is like eidonomy, which is the morphological study of the external appearance of organisms; anatomy is its opposite insofar as it is the morphological study of the internals of an organism.

I'm not entirely sure whether it makes sense to analogise to this extent. What is the "anatomy" of a meme? Is it the literal internals of the file, like its metadata? Or maybe it's the metaphorical anatomy of the way that its diagrammatic structure performs a syntactic function of mapping one element in the picture to another. All of these methodologies seem very promising to me, and I think we ought to try them all out sooner or later. But specifically regarding the naming of the approaches, there's probably something better than only taking from the biological analogies.

I don't want to push the analogy with biological terms too hard without emphasising that it's a deliberate exercise in analogy, because meme studies has a long history of wasting time and effort on doing it naively. The primary benefit of analogising culture with biology isn't about identifying equivalents; the primary benefit is identifying analogies that are useful, and most analogies are not going to be useful, let alone make sense. In this case, it does seem like it leads to many interesting possibilities though.

1 Like

Sort of, yes. But as some people have explained, it's a bit of an inevitable consequence of the standards required to properly archive it, classify it, and present it as such, in their website.

Take the analogy for software, and add-ons (or games and mods). The main "product" is expected to be of a certain standard of quality, and every component of it is held against a certain measure. Against an add-on, which may do a specific thing you require, people may forego certain other issues or lack of quality that just "wouldn't run" if it was part of the main software.

A similar thing happens in what could be called a "formal and respected" meme archive, encyclopedia, database. It tries to uphold certain standards, and so if an object can't meet those standards (or, for a lack of resources, can't get to study enough of them in time), they simply won't be presented to the world.

We "know" lots, and lots, and lots, and lots of memes, there are "new memes" every day, but there isn't always a way to, or enough resources/time to provide all the required evidence to meet these encyclopedic standards. It's like "common knowledge" vs. scientific knowledge, or "facts".

But it also is probably a lack of efficiency in methods and resources, too.


Good points about analogy. All models are wrong, but some are useful, as they say.

Here is a tangential thought that might develop into something useful (or might not): instead of 'anatomy' and 'eidonomy', maybe we can think of genotype and phenotype. A artefact's 'genotype' is not its metadata per se, but the literal binary sequence of data that reproduces it. It's 'phenotype', like a phenotype in actual biology, is the collection of observable, human-identifiable traits that emerge in some realisation of this data.

Already I'm starting to feel like this analogy breaks down, lol. You can identify a an abstract Loss meme by 'phenotypical' characteristics, although my suspicion is that at the 'genotype' level any two Losslikes will probably be radically different. Maybe not, though — maybe, for a sufficiently sophisticated observer, patterns do emerge in the data that would allow you to correlate two Losslikes at the level of bits!


Regarding classification and expanding upon what Seong said, I think further insights from Biology would be helpful.

Tinbergen's levels of analysis, which is based on Aristotelian four causes, differentiates between four levels of analysis with which we could approach the observed phenomena or behaviour: Ontogeny, Phylogeny, Mechanism, and Function.
Let's look at each one and try to understand how we can classify memes based on that.
Starting with function, it is based on the final cause of Aristotle. It is related to the teleological nature of its object, its function or purpose. For example, the function of an eye is to see, with all the benefits that come with seeing. The appendix was an open question for many years with regards to its function. Similarly, certain animal behaviours, such as non-linear migration patterns, are still hotly debated with regards to their function.
A functional analysis of memes looks at how memes function, how they work. The classification between pre-ironic, ironic, post-ironic, and meta-ironic does exactly that. It classifies memes based on how they function, based on its purpose. Now teleology is much less controversial than biology when it comes to artificial objects like memes, since they obviously do have a purpose for their creation.

Another one is the mechanism. Mechanism corresponds to the formal cause of Aristotle. This looks at how the object or the behaviour achieves its function. What is the underlying structure, how does it work? We can ask this question about the eye, investigate it to realize that it works as a lens with millions of receptors that send signals to brain etc. Or we can ask this question about a behaviour, for example mating. The mechanism of mating is still an open question in many species, since we still don't know how it works.
Mechanism of a meme includes its use certain formal elements, its underlying structure and how it comes together to produce that effect. A mechanistic classification of memes classifies the memes based on their form. KYM is an example of that. They look at formal and structural similarities and classify the memes based of off that. Now, just like even though the function of an eye is consistent across different species, the mechanism varies wildly, the same is the case with memes. Hence, functional classification is built upon the types of memes that are achieved with a certain form. I think it is also important to note here the distinction of species and genera, species refers to the entities that are extremely similar and are self replicating, they are created based off of each other and hence they are linked by their family resemblances and genetic similarities. Thus, Loss is a species of meme, it is not a category of meme. We cannot distinguish Loss by asking certain fundamental questions about memes, nor can we classify memes based on being Loss or not being Loss. Loss is created as genetic self replications of its predecessors, hence are always interlinked to create a species that evolves.
Long story short, simply identifying species is not a mechanistic classification. A mechanistic classification identifies a certain formal or structural element based on which we can distinguish it. We can say that 4 panel-memes is a structure or a form of a meme. I think KYM discusses these formal elements, hence its analysis is mostly mechanistic. But its main purpose is identifying species. And everybody knows that merely identifying species without classifying them is a poor form of biology, and no analysis can go further with merely identification.

I think this answer is already pretty long, so I'll write the ontogenetic and phylogenetic types of classification later on another post. In the meantime, lmk what you think.


That's a fantastic and promising approach to engineering new concepts within meme studies. What do you think about more recent (Tinbergen's model is from 1963) iterations of the model, such as this:

(diagram from above paper.)

Or perhaps this one:

Tinbergen’s four questions for the study of behavior. a The traditional view of Tinbergen’s four questions has partitioned research into proximate or ultimate levels of analysis and historical and contemporary timescales of influence. b A synthetic approach to Tinbergen’s four questions considers both levels of analysis and multiple timescales. This yields additional questions that are important in the study of animal behavior

1 Like

KYM was invented back when there were like 3 memes per year per subculture, and they were all distinct and more or less independent. Now there is a constant flow of one idea or aesthetic morphing into the next, and nobody could possibly even name or individualise them all. Think about the wojakverse for example, it's impossible to document all of that, but at the same time lumping it all together feels linadequate

1 Like

Let me first finish the earlier point, then I'll respond to these newer approaches.

Now, the remaining levels of analysis are Phylogeny and Ontogeny.

Firstly, phylogeny: Now regarding the mechanistic levels of analysis, we've introduced the concept of species in memetics (albeit somewhat loosely) and tried to classify certain formal parts of a certain species in a mechanistic manner, hopefully to also find a purpose in this mechanism to connect it with functional analysis.

Phylogenetic analysis corresponds to the efficient cause in Aristotle's framework and its questions are about the evolutionary history of a behaviour.

Now, our previous analysis was static. For our dynamic analysis, we need to ask new questions. There are two questions we can ask in phylogenetic analysis: 1) how are new species formed? (i.e. how are memes born? How is a new meme created?) 2) What is the evolutionary explanation for a certain mechanism? (how do memes share certain characteristic formal elements?) The second part will be the one that is of most interest to us. An example from Biology: tree-formation was an open question for Evolutionary Biologists for a long time, mainly because the functional elements and mechanisms of trees are found in vastly differing and varying places in the plant world. It was eventually concluded that tree-formation was an example of convergent evolution. The birth of the mechanism of tree-formation was not due to an evolutionarily ancestor behaviour, rather, a lot of different plants independently developed similar mechanisms that fulfill similar functions. Previously, we've talked about eyes with regards to function and mechanism. Phylogenetically, eyes are also an example of convergent evolution.

Now, the application of this concept to memes should not be difficult. Once we understand the mechanisms of a certain behaviour as inherited or evolved for a certain need, we can ask questions about 4-panel meme formation. There are a vast number of different 4-panel memes out there. Is the reason these different memes all share the same form and mechanism that they evolved from similar types of memes that utilize 4-panel forms? Or did the form of 4-panel memes develop independently, because the mechanism responds to a specific function that many memes want to utilize?
I don't think that these kinds of questions are silly to ask. Yes, memes always are shaped by the outside world and take influences from it, but so do species, they respond to their environment in a certain way and develop a certain mechanism. We can ask for example, if the ancestor of 4-panel memes was comic books and strips; if these memes evolved from webcomics, which in turn evolved from comic books, or if only a group of these memes have their origin in comic books and others have a different origin.

This kind of analysis, though hard, may be fruitful. At the very least, by analysing phylogenetic history, we possibly can discover an example of convergent evolution! We can find that two different forms of memes, say one from 4chan and other from Instagram, evolved independently but share similar structural elements. I think such a discovery would be beautiful.

Secondly, ontogenesis: Ontogenesis corresponds to the material cause. It asks a different dynamic question about the behaviour or the organ. How is that organ or that behaviour created in the particular species? What is the developmentary history of that organ or behaviour in the development a member of a particular species? For example, when it comes to mating, a question about when a species becomes sexually active, or what happens in their body when they become sexually active, are ontogenetic questions about mating behaviour. We can also ask similar questions about eyes: at what point does a human develop an eye in the fetal development? Or does the eye stop developing after birth or do our eyes continue to be developed afterwards as well, if so how and to what point? To connect this analysis to Aristotle better, we can ask a question about bone-formation. Trying to understand what bones are, what they are made of, and how animals are able to produce these massive rigid shapes in their body, was an interesting ontogenetic question, that involved both the developmental history and the material analysis of bones.

I think this level of analysis is the hardest to apply to memetics at the moment. But I think this is mainly because we don't have the vocabulary or the conceptual framework to talk about the ontogeny of a meme. We rarely talk about how memes are made, how certain processes occur in the development of a meme or how applying a certain process changes the content of the meme. The closest ontogenetic analysis I can find about memes is distortion and the introduction of noise in memes. Certain memes are heavily distorted and the images have a lot of noise. Nowadays, it is a style, a mechanism that is used to fulfill a spesific function. Phylogenetically, its origins had a different ontogeny than the contemporary ones. Previously, it was due to the residue noise from the image being compressed each time it was saved and uploaded again and again. Nowadays, people use programs to artificially introduce the noise, and it has evolved to become a graphic effect.

Finally, as you can see, we can analyse the phenomenon of distorted memes on these 4 levels of analysis:
1-) What is their function?
2-) What kind of a form or structure do they have?
3-) What is their Phylogenetic history?
4-) How are they created or formed?

These points are not clear in my head tbh, I hope that they make sense to you. I really believe that this kind of a framework will fill certain gaps in our analysis of memes. Lmk what you think!

Again this response became needlessly long :smiley: Seong I think I'll respond to you next time.


A fascinating theory and analysis. To shore up your point about phylogeny a bit: it interested me that over in this thread, Seong described one of his memes in terms of panels —

I think this is intuitively quite different from the panels we see in various kinds of 4koma:

The latter examples are literally comic strips, so their heritage is pretty clear, but Seong's meme has more in common with Disapproving Drake or something like that. I think it's safe to say that there isn't really a common ancestor behind Seong's Nietzsche meme and Poptepipic, but they nevertheless are independently displaying similar features.

Beyond a merely superficial convergence, I would suggest there's a mechanical one at play here, too. I get a sense that there is something resembling cross-pollination, even, because a meme like Seong's ends up relying (to some degree) on the inherent temporality of the panel arrangement to deliver the punchline — even though earlier memes in that genre didn't really care much for this function.

My suspicion is that this sort of cross-pollination process happens frequently on the bleeding edge of memecultures. Prolific and successful memers (and memecultures) are prolific and successful, I think, by virtue of their ability to recognise, appropriate, and adapt memetic devices and forms from adjacent genera and/or species in their own work. In this capacity memes and memecultures can be kind of hypersexual,1 if their reproduction happens at a higher-than-usual rate and can transcend 'species' boundaries in a way that we typically wouldn't expect of meatspace sexual reproduction.

1 Maybe even orgiastic, if you're cynical.