Cryptography means, literally, ‘secret writing’, and it is the technology of making interpretation of a text impossible for those without a key, and trivial for those who have it. In the field of cryptography, a substitution cipher is one of the simplest methods of encoding a message. It works by substituting the letters of the alphabet of the original message (the plaintext), for letters of another alphabet (the ciphertext).
If the receiving party has the key, i.e. knows the algorithm by which the text is encoded, decoding it is a trivial task, but if the receiving party doesn’t know the algorithm, the message remains meaningless. While simple substitutions, with only modest modifications, can be difficult, if not impossible to crack, far more sophisticated methods of encryption exist.
Since all communication is symbolic, a kind of encryption is always in operation between people as they use signs, whether graphical (text, symbols, images), aural (sounds), or gestural (embodied) to stand in for, or to signify, ideas, relationships, and meanings.
To take this metaphor a bit further, miscommunication happens when the plaintext message is not decrypted properly, that is, when the receiving party does not assign the proper values to the given signs. This can cause the message to become compromised to some degree, either garbled or completely unintelligible. If I ask you to “hand me that wrench”, and the meaning I assign to “that wrench” doesn’t correspond with the meaning you assign to it, then you’ll probably hand me the wrong wrench.
A deceptive interface, in keeping with this analogy, deliberately introduces errors into the process of interpretation. Certain signs are encrypted incorrectly; their meaning does not correspond with the meaning they are supposed to have, and so even when the receiving party “decodes” the message using the proper algorithm, errors result. It’s as if I ask you to “hand me that wrench” while simultaneously pointing at a screwdriver or a different wrench.
Obviously, passing the wrong wrench isn’t a big deal, but this is the fundamental basis of all deception: one or a group of things that are being made to stand in for other things are not as they purport to be, in fact they are the opposite: imposters of meaning that can exist only because we are always already accepting that one thing stands in for another thing.
Communication depends on symbolic substitution, and deception depends on it doubly so, first for its surface meaning and second for its true deeper meaning. It is the ambiguity of substitution that provides cover for all deception: if the letter A can be the letter X, than why not Z? But it is only when the message is fully decoded to we realize the error, do we see the true meaning of the signs beyond their surface appearance. Dark patterns are the algorithms that encode faulty representations into the deceptive interfaces.
Understanding that dark patterns are informed by metis allows us to bypass the question of intent, where other theories of dark patterns have stymied. Since there is no way to know with certainty a designer’s intent, methods based on identifying dark patterns by their “intent to deceive” will never be reliable. But understanding metis allows us to set aside the question of intent. Is a given interface or statement deceptive? Is one thing pretending to be something it’s not? If so, then the interface is deceptive, regardless of whether the designer “indended it.”
It’s not about intent to deceive or about establishing the truth value of all statements, it’s about whether or not the interface exhibits the features of deceptive communication: subversion, disguise, camouflage, and the reversal. If it exhibits these things, we can call the interface deceptive, whether anyone intended it to be or not. In other words, its less about the intent of the designer, which we will never know, and more about the intent of the design itself.
By the same token, understanding metis allows us to bypass another problem: namely the weakness of logic to anti-logic. As Hane notes “It is one of fascism’s goals to monopolize our attention. It would like to shrink our imagination. … Fascism welcomes our attempts to play logical ‘gotcha’ with its inconsistencies because it knows we will lose — not because we won’t find a fallacy but because the fallacy won’t matter” (p.1). We cannot defend against dark patterns by identifying them one by one: we must use them to expand our imaginations.
What makes dark patterns dark then, which is the same as what allows deception to work, is this fundamental reliance on disguise. Instead of illuminating, it darkens; instead of revealing, it conceals. Dark patterns, like sophistic rhetoric, can be identified by the construction of their masks, by the place reality has gone askew.
When one reality is made to stand in for a different one, we know that we aren’t party to a good faith effort to persuade, but instead we’re being duped by a wily mind hoping to lull us into a false sense of security——for the better to rob us blind. But this darkness is not “bad”, or “evil”, it is the stuff of all communication, of all perception.
Dark patterns are indeed dark, but not because they are evil; they are dark because the darkness is necessary to effect their disguise—they are dark because their art is precisely to make their art invisible. Cunning works only when unseen. Dettiene and Vernant write, “the individual who is endowed with metis” must:
adopt an oblique course and make his intelligence sufficiently wily and supple to bend in every conceivable way and his gait so ‘askew’ that he can be ready to go in any direction. In other words, to use the Greek term, you could say that the task of the agkulometes one, who possesses twisting metis, is to devise the straightest way to achieve his end (p.6).
To get others to see something, you always must cloud their sight: to show it correctly you have to first draw it wrongly, as Arhheim writes, we make “things look right by doing them wrong” (p.115). This property of metis, its crookedness, is a property of all communication.
Detienne and Vernant write: “the only way to triumph over an adversary endowed with mētis is to turn its own weapons against it” (p.43), to be “even more multiple, more mobile, more polyvalent than his adversary” (p.5). We have learned the ploys of cunning from cunning itself: it effects a reversal, it seizes opportunity, it takes shifting form, and it connives with reality to dissimulate to seem opposite of what it is.
These same aspects can guide us to defending against deceptive interfaces: after all, Athena’s most powerful technology was not her sword, but her shield Aegis, which transformed the beholder to stone and ended all possibility of movement. The gorgon on Athena’s shield is the dissolution of metis: it fixes it, allows it to be observed unchanging.
Being turned to stone is like being caught in the most powerful snare, total and never to be escaped. To fix metis in our gaze is to undo its power, and to do this our mastery must encircle metis as metis would encircle its adversary. We must trap it in its own net, reach through its illusions and grasp its core, and hold on fiercely whatever might come.
As we’ve come to learn, the key to defeating deceptive interfaces lies in understanding the nature of their cunning: a reliance on disguise and a corresponding weakness to observation. In each of the historical examples of online deception, for example the piercing of the disguise marks the end of the ploy and the defeat of the malicious actor.
From calling out trolls on social media to outing catfishers (or spear-phishers for that matter), from detecting trojan horses to mapping the membership of bot-nets: neutralizing the threat of deception relies on identifying and piercing the disguise. So the question becomes: by what means can we reliably identify and pierce such disguises?
The answer, in a word, is rhetoric. When trolls get called out, when catfishers, like Joan, are exposed, when fake news is fact-checked, members of the audience are able to detect duplicity in the discourse of the deceiver through analysis that is fundamentally rhetorical in character. The same is true of deceptive and misleading charts and graphics: when the method of their disguise is recognized, their true character is revealed.
And the same is ultimately true of deceptive interfaces composed according to dark patterns: once we pierce their disguise, they’re powerless. Once we understand how the disguise has been constructed through language, we can see the way to its undoing. This means we must understand the technical mechanism by which the disguise is effected, and here is where dark patterns are particularly instructive. Dark patterns provide a series of isolated examples that illustrate the different ways disguises are constructed.
Importantly, metis does more than underlie all deception. Ultimately, metis is revealed as the generative force that animates not just deception but all perception itself. Metis is the source and seat of rhetorical agency, the font of all techne, the true name of rhetorical kairos and true mother of phronesis, prudence and decorum. At the heart of the deception proves to be a fundamental truth: that deception is the truth, all reality is composed of illusion.
Perhaps, as Baudrillard despaired, the precession of simulacra has begun and all signification is liquified yet, still, what is liquefied can be made solid again and if deception is at the heart of this thing we call reality, what of it? We have no other and indeed it is all the more marvelous (thaumata) and amazing (in the original sense of the word) for being so comprised.
The implications of this understanding for digital rhetorics, rhetorical activism, and rhetorics of science, technology, mathematics, and medicine prove to be extensive. The concept of metis as generative could contribute discussions of agency both in terms of human and nonhuman actants, and in terms of activism and hermeneutics. Metis in visual design could contribute to discussions of the misuse of data.
Such understandings of metis could also also contribute to discussion of fake news, privacy jacking, memes, and the rhetoric of the so-called Alt-Right movement evident in, for example, the tweets of Donald Trump, which are themselves an archive of metis in their own right.
For all of us, as writers, designers, makers, and authors of all kinds-primarily as teachers-reclaiming mētis means recommitting ourselves to civic and disciplinary obligations that have their very real expressions in the work produced by us and by our students both seen and unseen, now and into the future.
Reclaiming mētis means acknowledging the explicitly moral character contemporary rhetoric has come to assume and questioning why we made this choice and what it costs us as a discipline. And perhaps it means revisiting this choice so as to better face new threats without compromising the virtue our allied rhetorical disciplines have rightly come to value.
For rhetoricians of science, technology, and medicine especially, it means revisiting our hermeneutic turn and the role rhetoric plays in culture, while for rhetoricians of feminism, multiculturalism, and activism, it means revisiting the question of agency, what it means, and how it works in complex real-world situations characterized by (always) asymmetrical power relationships.
Ultimately, reclaiming mētis means that we become more cunning than our most cunning adversaries, that we become more wise and more wary than the hunter, that we develop our own disguises, our own ploys, so that we may deceive the deceivers and secure the safety of all.
After all, as a discipline, we’ve long understood the best rhetoric hides itself, obscures its traces, makes itself invisible. Dettiene and Vernant draw our attention to the moment when Odysseus prepares to address the Trojans:
Consider the most subtle and most dangerous orator of Greece preparing, before the assembled Trojans, to weave the glittering web of his words: there he is, standing awkwardly with his eyes fixed on the ground, not raising his head; he holds the staff quite still as if he did not know what to do with it. He looks like a tongue-tied yokel or even a witless man (aphrona). At the moment when he is about to speak the master of tricks, the magician of words pretends to have lost his tongue, as if he were unskilled in the rudiments of oratory. (p.23)
Odysseus, himself sees deception not as a good or evil, but as a tool to be used as the situation warrants. But more fool him were his disguise be detected, for all further attempts at persuasion would be sure to fail without a shifting tack.
In the end, dark patterns are poor rhetoric not because they are immoral, but because they are so trivial to defeat: you just have to recognize them. And once you’ve done so, their patent hostility makes them forever unpersuasive. A better rhetoric would not allow the user to know how the deception was effected. It would happen entirely out of the user’s sight and without anyone noticing.
