A semiotic analysis of the canonical image macro meme

2.1 How a CIMM acquires its canonized meaning – first- and second-order signification

CIMMs are comprised of both image and word – both elements are necessary to the meme in order that it communicate. The visual component and the verbal component of a CIMM cannot be hierarchized, as both elements are equally important. Nevertheless, it is the image that is digested first by the meme user, priming the user to understand the meme in a specific way. I follow Gottfried Lessing’s distinction between the arts of time (literature) and the arts of space (plastic and visual) and argue that spatial art is able to make a more instantaneous impression upon its audience than the temporal (Lessing 1874 [1766]). The visual component of a CIMM informs its verbal component and furnishes the emotional landscape by which a CIMM will continue to be read.

Early semioticians such as Roland Barthes and Émile Benveniste held that while images were able to powerfully communicate meaning, they were not in possession of their own metalanguage. Benveniste referred to images’ lack of metalanguage as “liberty” (Benveniste 1974), and Barthes posited that the meaning-making of images could be reduced to “a message without code” (Barthes 1977: 36). Subsequent semiotic schools opposed Barthes and Benveniste on this point: the Greimasian school posited instead that “an image can … interpret another image without necessitating recourse to the system of verbal language” (Dondero 2020: 9), and Groupe μ maintained that the conception of a universal visual langue might be possible. Recently, Maria Giulia Dondero’s developments to the theory of the “uttered enunciation,” building on work by Louis Marin and Jacques Fontanille, have enriched and clarified the field of visual semiotics, facilitating a more rigorous interpretive model for reading images and allowing for the tentative proposition of a visual metalanguage contingent upon enunciation (Dondero 2020).

The development of plastic and visual semiotics beyond the formulations of denotation and connotation first proposed by Barthes in “Myth Today” (Barthes 2009 [1957]: 138) and Benveniste allows for a significantly more thorough reading of the images in memes. But most CIMMs do not require a sophisticated analytical approach to understand their meaning, as the image component is often chosen for its accessibility and its capacity for emotional communication – to borrow Dondero’s (2020) language, those where the “forces” [gestures, movement] in the “forms” are strongly evidenced and easily discernible. The prevalence of stock photos, cartoons, and snapshots from reality TV in most CIMM templates demonstrates that the most apt images for remix are easily “accessible” through obvious displays of emotion or discernible body language and gesture, which “many semioticians and linguistics … consider to be a more fundamental form of communication than vocal language” (Danesi 2004: 53). It is in the relationship between “a simulacrum of the ideal gaze” (Dondero 2020: 3), the embodied viewer, and the figures within the image where we see that the mechanisms for meaning-making in memes diverge from other visual images. A more clearly articulated visual semiotics will allow us to undertake a more nuanced reading of a CIMM’s image macro, but it will only be useful and relevant until the point at which a CIMM’s meaning becomes consolidated. Where a visual semiotic reading of an image interrogates the relationship between these three agents and maintains the tension between them, a semiotic reading of memes must understand that once the emotional landscape has been discerned from the figurative image, the figures – or signs – are no longer of much concern.

Though each individual CIMM template signifies something unique, the overarching pattern is that every CIMM template is only resonant – and therefore humorous – in that it has the capacity to connote a meaning beyond what it is actually capable of denoting, which makes it especially rich for remix. The humor in a successful CIMM remix is achieved by altering the original meme so as to accentuate all the more its salient point. In order to better understand the initial process of reading and inferring meaning from a CIMM, let us consider and interpret the “Distracted Boyfriend” meme (Figure 5).

Figure 5:

The original image macro template for the “Distracted Boyfriend” meme. Source: Don. 2017. “Distracted Boyfriend” meme. Know Your Meme. Original photo from iStockPhoto, “Disloyal man with his girlfriend looking at another girl.” https://www.istockphoto.com/photo/disloyal-man-with-his-girlfriend-looking-at-another-girl-gm493656728–77018851 and https://knowyourmeme.com/memes/distracted-boyfriend (accessed 8 January 2024).

The visual component of this meme shows a simple and easily legible scene: the Girlfriend looks at the Boyfriend with disbelief as he ogles the Girl in Red. This latter figure, though occupying a key role in the action depicted in this image, merely functions as a catalyst for the interaction between the Boyfriend and the Girlfriend. Her blurred-out face and her telling position with her back to the couple indicate the insignificance of her own emotions and her obliviousness to the scene playing out behind her. The image component of the meme, in which all interactions between the figures are read by the meme user, signifies the “turning of one’s head”: distraction from long-term happiness (staid and steady like the Girlfriend in blue) by the thrill of the new (fleeting and pleasing to the eye like the Girl in Red). The image macro captures the gesture with clear directionality and legible (to the point of farcical) body language. In this way, the meme’s visual component depicts a situation that most people are likely to understand, and one with which the image will, over time, become inextricably linked.

In all situations of remix, the Distracted Boyfriend has the opportunity to achieve healthy contentment by the option offered by his Girlfriend, the woman in blue. However, he is always tempted by the option furnished by the Girl in Red, who is always representative of the easier option, a kind of temporary joy unlike the sustained happiness offered by the Girlfriend. Thus, the distinctive image macro for the “Distracted Boyfriend” meme comes to signify the subject being led astray from long-term satisfaction by the thrill of the short-term solution. Remixed “Distracted Boyfriend” memes will have to ensure that this specific connotation is appropriate to the situation represented in the remix. The emotions that the meme user is meant to understand from the image in all cases of the “Distracted Boyfriend” meme are indignation on behalf of the overlooked Girlfriend, and admonishment (but simultaneously wry understanding) for the Boyfriend, whose head has been turned by the Girl in Red. More often than not, the remixed meme will invite the meme user to occupy the position and assume the role of the Distracted Boyfriend – as such, this particular CIMM takes on something of a confessional function in the internet landscape.

With CIMMs, unlike with other visual images, the positionality of the assumed viewer is dynamic. The image macro of a CIMM only invites a meme user to behold the image insofar as its appropriate emotions are gleaned; at this point, the meme user is invited into the image to inhabit it emotionally. As such, the figurative image of a CIMM can be understood as not the locus but the conduit of meaning. If a CIMM image template undergoes two simultaneous readings, as seen in the case of the “Distracted Boyfriend” meme, then meme users will not empathize with the denotational aspect of the meme (i.e., a meme user will not empathize with the Girlfriend per se) but rather with the connotational aspect (i.e., with the Girlfriend’s outrage). Memes may seem realistic, but they are not meant to be read mimetically. Rather, the recognizable figures in the first-order semiological system come to be synechdocal for the emotions or reactions that they demonstrate. It is these emotions or reactions that are then applied in situations of remix.

2.2 Signification and remix – image

To become a CIMM, the original image template must provide ample potential for remix (Figure 6). For that to happen, the image component on its own must have an adequate signification on its own that allows for its application onto various hypothetical scenarios, to be furnished by the CIMM’s verbal component. It is only in this case that an image macro meme will attain canonical status. A successful CIMM (one which attains the greatest resonance with its user and therefore becomes widely spread) is never remixed beyond recognition. Only specific parts of a CIMM are pasted over or changed out, but the formula by which the CIMM is remixed stays consistent.

Figure 6:

The original image macro template for the “Distracted Boyfriend” meme. Source: Don. 2017. “Distracted Boyfriend” meme. Know Your Meme. Original photo from iStockPhoto, “Disloyal man with his girlfriend looking at another girl.” https://www.istockphoto.com/photo/disloyal-man-with-his-girlfriend-looking-at-another-girl-gm493656728–77018851 and https://knowyourmeme.com/memes/distracted-boyfriend (accessed 8 January 2024).

Let us consider the “Distracted Boyfriend” meme in various remixed states to understand which parts of the original image template must stay the same for it to retain signification (Figure 7).

Figure 7:

Nostalgic doodle remix. Source: remixed by the author via Microsoft Word.

Here, the Boyfriend is labelled as “an 11-year-old,” the Girlfriend as a normal “s,” and the woman in red as a Superman “s.” We see the meme-ification of a situation with which most people from a specific generation will be familiar – as children, they were more drawn to the thrill of sketching out a Superman “s” in place of a regular “s.” All parts of the original image template are discernible in this remix. The superimposed phrase adds a new layer that does not obscure the base template or its emotionality, but builds upon it, relying on the image template’s pre-ascertained signification to inform the memetic text.

In the Figure 8 remix, each figure’s facial expression is obscured by the remixed object labels, and therefore their respective body language, which informs how to read the situation at hand, cannot be identified. A meme user whose first experience with this CIMM was this specific iteration of the meme might not understand what renders the CIMM funny, as they cannot understand its connotations – the body language, which makes the situation resonant and therefore humorous, is not readable. However, meme users already familiar with this specific image template will be able to cognitively “fill in the gaps” – already understanding the context of the image, they will find the meme humorous despite not being able to discern the figures’ facial expressions. Thus, when one is already familiar with an image macro meme template, as long as the template is recognizable in the CIMM’s remixed state, one will be able to understand the meme’s message.

Figure 8:

Nostalgic doodle remix covering faces. Source: remixed by the author via Microsoft Word.

As shown in Figure 9, in the case that all three figures are totally obscured to the point that the original template cannot be ascertained from the remix, the meme ceases to be intelligible and therefore also ceases to be humorous.

Figure 9:

Nostalgic doodle remix to the point of template obfuscation. Source: remixed by the author via Microsoft Word.

Within each image macro meme, the immediately digestible visual component acts as a prompt towards its viewer, signaling the emotion that the verbal component and subsequently the text is meant to elicit. Consider the difference in signification between Figures 7 and 10. Each text is necessarily read differently, due to their differing visual components. While the image macro in Figure 7 fits with the situation at hand and adequately depicts an eleven-year-old choosing the enticing Superman “s” over the staid, normal “s,” the image macro in Figure 10 – a template used to evoke a situation in which three equally responsible parties try and heap their portion of the blame on others – is less appropriate for this situation, and therefore the resultant meme is less humorous. Thus, we can see a meme user’s reaction to a remixed CIMM will be firstly informed by its visual component, which acts as the immediate signal for how the meme is to be read, and the humor that is to be derived from it; the verbal component is contingent upon the emotional landscape that the visual component has provided. Therefore, any CIMM can only be remixed insofar as the connotations that the template image evokes are still applicable to the new situation posited by the remix.

Figure 10:

“Spiderman pointing at Spiderman” nostalgic doodle meme. Source: Spiderman pointing at Spiderman template found at imgflip.com. https://imgflip.com/memetemplate/206014687/3-Spidermen-Pointing-at-each-other (accessed 8 January 2024). Remixed by author.

2.3 Signification and remix – verbal component

If the visual component of a CIMM is what furnishes its emotional digestibility, then the verbal component can be said to fully situate the meme reader, creating the context in which the emotion of the image is to be elicited. CIMMs are funniest and most resonant when the remixed verbal component exposes something both niche and widely relatable: following the suggestion that new media has turned hidden and mundane practices into highly visible public culture (Burgess and Green 2018), one could argue that a meme has the highest ludic potential when it exposes specific but common behaviors that individuals have largely considered inconsequential, vaguely taboo and, well, individual. The verbal component of a CIMM is the part of the meme that does the network positioning by which memes are often understood: “are you in or are you out?” If a meme user finds the verbal component of a remixed meme funny, they signal that they can be counted among a specific community to which the meme relates. This urge is also rewarded by the medium of social media, which encourages a meme user to share the meme with others to demonstrate one’s belonging to a network and to extend that network to others, incentivizing meme users to relate themselves to memes.

In the case of image labelling memes, the remixed verbal component is only constrained in that each image label must correspond with the correct figure and reveal the equivalent tensions between figures posited by the original image template. However, other CIMMs have text components that are more distinctive and provide the primary means of remix. In these cases, I will show that the verbal component, despite its greater capacity for remix, can also only mutate up to a fixed point before it loses its significance. As with its image template counterpart, there are formulae that must be followed in instances of remix.

Let us consider variations on the “One Does Not Simply Walk into Mordor” meme (Figure 11) to understand the extent to which the verbal component of a CIMM can be remixed before it loses relevance.

Figure 11:

Original “One Does Not Simply Walk into Mordor” meme. Source: Captain Blubber. 2014. One Does Not Simply Walk Into Mordor. Know Your Meme. https://knowyourmeme.com/memes/one-does-not-simply-walk-into-mordor (accessed 8 January 2024).

In Figure 12, we see that deviations from the standard verbal presentation of memes – white characters outlined in black, usually in the font “impact” and most frequently with all words capitalized – detract from an image macro meme’s immediate identification as a meme, and are less widely propagated, such that we see few, if any, mainstream memes that deviate from these norms.^[3]

Figure 12:

Original verbal component, original syntax, but different visual representation. Source: remixed by author via imgflip.com.

As shown in Figure 13, in the case of this CIMM, it is the specific words and syntax of “one does not simply” that must remain the same in each remixed CIMM for the text to achieve signification.

Figure 13:

Remixed text keeping “one does not simply” part of original verbal component. Source: gaioaa. 2013. Image #339,386. Know Your Meme. https://knowyourmeme.com/photos/339386-one-does-not-simply-walk-into-mordor (accessed 8 January 2024).

When the phrase’s meaning is paraphrased (Figure 14), the resulting meme is less recognizable and therefore less resonant. Although the words “walk into Mordor” remain in this verbal component, we understand that the meme’s signification does not hinge on this part of the phrase. More than the content of the words, it is their arrangement in CIMMs that renders the verbal aspect of the meme amusing and recognizable.

Figure 14:

Remixed verbal component keeping meaning of original meme but not syntax. Source: remixed by author via imgflip.com.

Even when the syntax and font are used appropriately (Figure 15), the absence of capitalization also renders the meme less recognizable and less resonant.

Figure 15:

Remixed verbal component keeping syntax but not capitalization. Source: jiblington. 2013. Image #380,790. Know Your Meme. https://knowyourmeme.com/photos/380790-one-does-not-simply-walk-into-mordor (accessed 8 January 2024).

Finally, verbal components cannot be mixed and matched between different CIMMs (Figure 16). Here, the verbal component is notably identifiable with the “Is This a Pigeon” CIMM precisely through its specific syntax, structure, and word choice. As such, its application to the image component of “One Does Not Simply Walk into Mordor” does not succeed in creating a resonant meme.

Figure 16:

Remixed meme using verbal template from different CIMM. Source: remixed by author via imgflip.com.

Almost all, if not all, of the above limitations apply to CIMMs that have a distinctive verbal component: see, for example “Condescending Wonka” (see Figure 17); memes whose verbal components are even more grammatically elaborate, such as “Mocking SpongeBob” (see Figure 18), which uses alternating uppercase and lowercase letters to produce the readerly effect of said mocking; or even “Doge” (see Figure 19)^[4] and “LOLcats” (see Figure 20)^[5] and their associated linguistic modes.

Figure 17:

Example of “Condescending Wonka” meme, where rhetorical question and passive aggressive response makes up verbal component of CIMM. Source: DireClownFail. 2012. Image #266,562. Know Your Meme. https://knowyourmeme.com/photos/266562-condescending-wonka-creepy-wonka (accessed 8 January 2024).

Figure 18:

Example of “Mocking SpongeBob” meme, where alternating capitalization denotes scorn. Source: Matt. 2018. “Dont use that weird spongebob mocking meme” Me: DonT uSe thAt WeIrd SpoNgEboB MoCkinG Meme. Know Your Meme. https://knowyourmeme.com/photos/1253025-mocking-spongebob (accessed 8 January 2024).

Figure 19:

Example of Doge meme (regarding education), in which the multi-colored, sans-serif font is combined with grammatically incorrect structures to denote earnest stupidity. Source: chairmanfrank. 2015. Thesis Doge. Know Your Meme. https://knowyourmeme.com/photos/746779-doge (accessed 8 Jan 2024).

Figure 20:

The original LOLCats meme, featuring the recognizable “I Can Has Cheezburger” phrase. This meme spawned its own linguistic mode of “lolspeak.” Source: Jason Skimmer. 2011. Image #48,689. Know Your Meme. https://knowyourmeme.com/photos/48689-cheezburger (accessed 8 January 2024).

Each of these CIMMs’ verbal component follows its own specific rules for modulation, outside of which it must not err if it is to retain its original signification. The verbal component, like the visual component, is recognizable outside of its corresponding image macro meme format (for example, the “Mocking SpongeBob” meme’s grammatical formula is often used outside of the meme context to denote scorn) – however, it is only recognizable as a direct consequence of having formed part of the original CIMM template. The signification of the verbal component of a meme, whether it hinges on the syntax, grammar, or lexis of the CIMM, informs the second-order semiological system of image macro memes to a significant degree.

Despite their ludic potential and capacity for remix, CIMMs are by and large a normative genre which must follow specific, hitherto unspoken rules in order to achieve resonance and signification. Both the visual and the verbal component of a CIMM must be crafted according to meme-specific constraints, as detailed above. The verbal component of a CIMM has a greater capacity for remix than its visual component and can undergo greater change before it is rendered unintelligible, irrelevant, and therefore no longer funny. In all cases, the visual component and the verbal component of a remixed CIMM must work in tandem for the new text to be maximally resonant.

2.4 Consolidation of a CIMM

A complete CIMM goes on to consolidate its specific signification by a symbiotic process – “funnier” memes, or in other words, remixed CIMMs whose verbal component more accurately or resonantly demonstrates the visual component at hand, are more widely spread. Remixed CIMMs whose verbal component fit less accurately or resonantly with the visual component (or vice versa) are not spread. The meme user becomes the point at which varying references come together (Milner 2016: 29), or “the space on which all the quotations [references or significations] that make up a writing [meme] are inscribed without any of them being lost” (Barthes 1977: 148). Popular memes are those which resonate with the greatest number of meme users, whose overlapping significations are both niche and understandable to the largest number of people. The more opportunity there is for remix within the template of a meme, the more that specific image and its verbal counterpart will be spread, thereby establishing an image macro as a CIMM. Meme users will also understand a certain verbal grammatical structure to be associated with the image macro at hand, which will become just as inextricably associated with the image and its emotions.

Memes, with the immediacy of their visual components and their recognizable, accessible, and abbreviated text components, are the perfect evolution of literary material for generations living in late-stage capitalism. “Writing has never been capitalism’s thing. Capitalism is profoundly illiterate,” say Deleuze and Guattari (1983 [1977]: 240). It follows, then, that memes have acquired such primacy as texts in the modern day. Fisher notes that “[t]eenagers process capital’s image-dense data very effectively without having to read” (2009: 25). This process of consolidation is further enhanced by the fact that the meme is also propagated as an exact replica of itself, with no variation outside of the primary remix (forwarded as a direct message [DM], tags, retweeted, etc.). In this way, memes are proliferated in the same way and with as little deviation as works of literature in the post-Gutenberg era, only sped up through the means of instant messaging. Modern-age social media is the prime grounds for this type of proliferation. In an attention economy, memes, with the immediacy of their visual components and their accessible and highly abbreviated verbal components, constitute the perfect texts for communication governed by algorithm.^[6] Since remix changes the least it can, the lack of superfluous variation in the image itself leads to a much quicker and more entrenched association of the exact image pertaining to a particular CIMM with the set of emotions that it comes to signify. Through constant reinforcement, the “myth” of a meme – the unquestioned association between a CIMM and its signification – becomes naturalized and consolidates itself in the meme’s own metalanguage, or second-order semiological system.

It is a knowledge of the metalanguage of memes that renders a meme user meme-literate or not. A literate meme-user will understand the significance of individual CIMMs without recourse to interpretation of the image macro, and will have the ability to wield the appropriate CIMM to fit the situation at hand. Those who encounter the Distracted Boyfriend meme, for example, will not need to undertake a visual reading of the image every time they encounter a remix before they understand the indignation experienced by the Girlfriend and the temptation experienced by the Boyfriend, and are able to apply these emotions to the situation created by the meme’s verbal component. Once an image macro meme becomes canonical, a meme user will largely bypass the interpretive reading of the image and skip directly to the emotions that the image is meant and known to elicit. As such, the image and the grammatical structure of the accompanying text become a shorthand for a highly specific and complex set of interacting emotions. Moreover, since memes now have their own rich history, a literate meme user will also have a generalized knowledge of past CIMMs, and understand how they have influenced the development of contemporary ones. With a greater knowledge of the meme landscape, meme remixers and users are able to create and enjoy memes that are increasingly interreferential.