Some Thoughts on “The Mechanics of Bad Faith”
This piece was written as a response to the article A Short Introduction to the Mechanics of Bad Faith on Wonk Bridge.
In my colleague Oliver Cox’s thought-provoking essay, “A Short Introduction to the Mechanics of Bad Faith”, he describes bad faith as a rogue element in a communication system and proposes a set of new ways for dealing with bad faith. While I take his essay to be an ambitious, serious proposal for better, more civilized forms of online discourse, I rather think he’s shortchanging the concept of “bad faith.” In what follows, by drawing loosely on a few thinkers, I will outline an understanding of “bad faith” that treats the concept as a social structure of discursive engagement and not a quality belonging to individual expressions or actors. Along the way, I will layer example on example to draw out some of the payoffs of this understanding.
In defining “bad faith,” Oliver writes: “What is a bad-faith actor? What is bad faith, for that matter? Bad faith is to act in ways that spoof the normal modes of interaction, such as in debate, conversation, commerce, while actually pursuing hidden, selfish motives or even hoping to disrupt the operation of the system in which they operate.”
Here, my colleague is working with a definition of bad faith that is more or less interchangeable with deception. The “bad faith actor,” for all intents and purposes, is a liar who aims to “spoof” or “disrupt” normal modes of communication. I do not think I’m being reductive when I say that his essay adopts a kind of diplomat’s (or even a forum manager’s) point of view. In Oliver’s account, what matters when dealing with bad faith is reducing the friction and keeping the channels of communication open at all costs. Bad faith actors trouble us not because of what they are saying, but because they disrupt the system of communication. As always, the goal is to ensure that the system works properly.
In response, I’d like to point out that we inherit the concept of “bad faith” not from systems thinking, but from a tradition of Christian messianism and its requisite, secular counterpart — an atheistic humanism defined by radical social responsibility. In this vein, the call to attend to “bad faith” has generally emanated from self-proclaimed outsiders who have ethical commitments that supersede the normative. As such distinctions between good or bad faith is often the province of the Marxist radical and the Christian zealot — much more so than the diplomat and the systems architect. If we are to take the concept of “bad faith” seriously, we should also be brave enough to examine it on these terms.
To offer a provocative metaphor — “good faith” and “bad faith” work in the Matrix, but they’re not native to the Matrix. That is why they’re interesting to us — even if we’re secretly backing the Matrix. If bad faith disrupts the communication system, the noise is certainly worth listening to.
On Listening and Speaking
To distinguish between our two perspectives, I’d like to begin by challenging some of the assumptions my colleague makes about what exactly happens in a conversation between two people. As I see it, Oliver subordinates the social dimension of discussion to the machine metaphor he deploys. Oliver writes: “in essence, one should only transmit well-formed data but, if one receives malformed data that is nonetheless decipherable, one should parse it anyway.”
He rephrases the imperative this way: “(what you transmit) you should maximize obedience to the standard; in the domain you do not control (what others transmit) you should maximize the total information that you decipher.” Under this imperative, Oliver prescribes these rules of conduct: “When speaking: be clear”; “When listening: be charitable.”
By “well-formed data,” I think my colleague means transparent data or rather, data that presents limited interpretative possibilities. While I think this is the case for receivers and transmitters — I submit that mutual understanding between humans is a more complex and open-ended game. The “data” that matters most in a conversation is rarely the what is explicitly said. Often, it’s what is surfaced in the conversation’s interpretive subtext. The silences, the omissions, the evasions and equivocations — in other words the “data” that require interpretive latitude matter as much, if not more, than the face-value exchange on offer.
But let’s back up. Why, we might ask, is it worth listening to someone in the first place? After all, not all data is helpful or even relevant. Here, I submit that we listen because we want to know something in particular, even if we’re not sure what. I was lately reminded of Roland Barthes’ description of the phenomenology of waiting. The lover by the phone, having received no signs from anyone, nevertheless waits and waits as if having received an order not to move. Far from maximizing the information received, the absence of a transmission can absorb our interpretive faculties while excessive responses from the wrong one is unlikely to be processed.
The salesman knows that a “no” can mean “never” or “yes with time” — not because anyone is being deceitful, but because the nature of every conversation pertains to an inquiry, regardless of whether it is made explicit. This interpretive latitude vis-à-vis the question at hand is key, not because it guarantees a specific point of arrival, but because it underpins the wealth of interpretive possibilities available.
With this in mind, while intended as a gesture of generosity, I rather think treating “bad faith” as the equivalent of “mal-formed data” reduces the question of good or bad faith to something that is incidental to the original inquiry. Often, when bad faith is at stake, its determination is essential to the conversation — it is the exact thing that we want to know.
Being “charitable” while listening is admirable and good manners, but it’s helpful to remember that charity can easily reflect back as feigned naivety or willful denial, a refusal to address an important subject because of a cynical calculus. Far from an attire for all occasions as my colleague suggests, the imperative to be generous takes as much interpretive liberty with the other as anything else.
Whose Bad Faith Is It Anyway?
For my part, I am less interested in the rigorous task of rooting out bad apples from good ones than the kind of interpretive game “bad faith” describes. Against the insistence that bad faith is a practice by an individual, I hope to show that “bad faith” is a distinctly social game and not a game with a clear, third-person objective (ball in hoop). We might take dating as a model — knowing what the other is thinking, even if it’s a mere guess, is the only way to “orient” ourselves.
Instead of rushing to “deal with” bad faith, let’s begin by amending our working definition of “bad faith.”
It might be evident already that I plan to route my understanding of “bad faith” through Jean-Paul Sartre. I realize that at a time when systemic and structural exploitation dominates critical discourse, Sartre and his emphasis on personal responsibility is a hard sell. When all we want is “better systems,” virtues like “authenticity” appear embarrassingly naïve at best and an exercise in mere style at worst. Yet Sartre still has something to say about the social structure of “bad faith” — or more precisely, about the interplay of “accuser” and “accused” when bad faith is in question.
In Being and Nothingness, Sartre is clear about what bad faith is not. Unlike legal and quotidian definitions, Sartrean “mauvaise foi” is never mere deception. To act in bad faith is never to lie. The liar knows the truth — he deceives the other in order to secure an outcome (he sells someone a bad car knowingly, he negotiates peace terms when he’s planning a surprise attack, and so on.) Acting in bad faith is also never a mistake — an error based on bad information.
Bad faith is first and foremost a form of self-deception [mensonge à soi]. To the observer, it takes the form of: I do not wish to know the truth of what I am. Oliver is right to point out that “bad faith” always involves an accusation of bad faith, but such an accusation is not always direct and never unidirectional. It’s important to note that to accuse (to demand that the other “come clean”), and to look away and feign ignorance (don’t ask don’t tell), both draw us into the game of self-deception.
Sartre’s own work admits how much the demand to know can itself be in bad faith. This is particularly evident in his controversial example of the secretive homosexual who feigns interest in women to his impatient friend. Here, the epistemology of bad faith overlaps with the epistemology of the closet. The friend might think, “we both know what you are — you can at least tell me, between us.” Yet it’s not clear whether the demand to own up to one’s sexuality can legitimately be made to someone else (or even to ourselves) — how are we sure that we are reading the right signs? What is it for one’s sexuality to be definitive? In a different vein, what are we demanding when “coming out of the closet” — for much of the twentieth century at least — was to accept a diminished life?
Bad faith speaks to the split within social discourse between an insistent demand to “out” the truth and the game of plausible interpretations that evade affirmation — the “evidence” that will never be quite right or quite enough. Here, the tug and pull between plausible deniability and indiscrete inquiry does not disappear by “being generous,” by pretending not to know the question. After all, is it better to pretend not to know? Don’t ask don’t tell certainly doesn’t eliminate curiosity as much as encourage speculation. Turning round and round with evasive maneuvers regarding what we really want to know about someone else — is an unavoidable aspect of the interpretive work that underlies bad faith.
Pascal’s Wager
Taking a step back from Sartre, let us look at an example where a question with more self-evident stakes. (Although I might point out that “publicly accepting one’s sexuality” has become, for better or worse, one of the more weighty political gestures of the past few decades.)
To do so — let’s go back to religious iterations of bad faith. In Christian tradition,“bad faith” describes the pilgrim who, before scripture, remains divided in mind. I have generally found Pascal’s Wager to be a helpful illustration of the problem. For the sake of brevity, I will take Pascal (perhaps a little too much) at face value. In his Pensées, Pascal muses that if one isn’t sure, it’s safer to bet on God’s existence. By choosing piety, you risk nothing, but what you stand to gain is infinite.
For critics of Pascal, the problem with the wager is that it enables a kind of “dodging of the question.” If one takes Pascal seriously, piety becomes, not so much a practice of belief, but more of an insurance policy for the soul. For the one who takes the wager, the nature of piety changes, opening up a pandora’s box of bad faith possibilities. As long as you behave as if you believe, keep the Sabbath, refrain from murder, etc. you can act piously without committing to the truth of what piety expresses — the truth of divine presence.
The problem with bad faith is that it both circumvents evidence (revelation) and ups its stakes — all evidence of divine presence becomes evidence of my wager (my clever choice). Unlike the divine injunction to“see,” Pascal’s wager recasts “seeing” as betting on one’s belief. (Here, it’s helpful to remember that our early empirical methods of inquiry derive partly from the faith-based orthodoxy of “seeing is believing.” One converts, quite simply, because one has seen. Blood-hungry crusades aside, the Christian tradition sets a surprisingly low bar for conversion — one does not need incontrovertible proof — one simply needs enough proof to believe. After all, what the pilgrim witnesses is not always extraordinary. Visions in conversion narratives — if they aren’t elided entirely — can range from the banal, to the repetitive, to the cryptic. In theory at least, “good faith” seeing balances the affirmation of having received enough revelation, and the private, personal choice that no further proof is needed.)
In a modern vernacular, we might say that Pascal presents conversion as a question of risk. The pilgrim adjudicates between risk avoidance — hedge your bets and believe — and an absurd exposure to risk — continuing to doubt. Yet as anyone who deals with “risk management” can attest, avoiding risk is not a one-time deal but an endless commitment. For the pilgrim that takes Pascal’s wager seriously, practicing “piety” becomes a strenuous responsibility. Going to church, giving alms, praying at dinner, are part of a never-ending dance of demonstrating that one believes, all the while begging and dodging the question of whether one really sees.
The Public Wants to Know
Here, I think it’s helpful to lay out the stakes of our disagreement. As I understand it, Oliver understands the game of discourse as more or less a marketplace of ideas. When players present their ideas to a forum, they can either play fair, thereby enabling the best idea to emerge, or play dirty, thereby undermining public trust in the marketplace.
According to Oliver, if we accuse someone of bad faith — we risk alienating our opponents and undermining our trust in the exchange of ideas: “the assumption of good faith is essentially the assumption that, despite our differences, we are all trying to seek the truth. It is, metaphorically, the phones in JFK and Khrushchev’s offices. It’s a fragile line of communication — and when one side introduces bad faith as a talking point, the other may be tempted to cut the cord.”
Let’s set aside for now whether a phone line in Khrushchev’s office is “good faith comms” or more likely as not, a piece of hardware that holds both parties hostage. By this logic, JFK and Khrushchev are representatives of two opposing sides in a debate. (You can cut the cake however you’d like: communism vs democracy, communism vs free-market capitalism, etc. etc.) In order for any debate to take place, we need what my colleague calls an “interoperable” language. As I understand it, by this metaphor, he means a language that works like an universal adaptor — a peg for every hole. To participate in the conversation, everyone must come equipped to play — pegs and holes at attention.
For our purposes here, I will not go into detail on why I think language is not, in fact, interoperable. As decades of postcolonial and critical race studies have demonstrated, how well the adaptor works depends very much on who we think is wielding it, for whom, and against whom. For my part, I hope I have shown that language is a game of epistemic power, a medium that for better or worse, offers enough elasticity and latitude for evasion, defenses, and flipping the tables.
So far, I have also tried to show that what something means (praying, negotiating, displaying interest) depends very much on 1) the demand to know (the “call-out, if you will) — regardless of whether it’s spoken or implicit — and 2) the game of deferral, avoidance, and misinterpretation in play. We mistrust someone if they lie, but “bad faith” draws us into the game of self-deception. To take another Sartrean example, the woman flirting in a café has no intention of giving the game away. Here, expressing interest a way of dodging the question of whether she — or her suitor — is actually interested. Flirting can leave the door open to changing one’s mind — but crucially, it avoids “outing” what neither party truly wants to know. No gesture, no promise, no act, can clinch the game. Yet only the sourest of suitors would cry “deceit” in all seriousness — after all, who has not deceived themselves in pursuit of a lover?
Where the stakes of bad faith are raised, I put forth, depends on the urgency and legitimacy of the demand to know publicly and to know together. To take my colleague’s example, let us remember that the Cold War was not a clash of ideas, clever mind games, or JFK and Khrushchev talking. Less of a “string” of communications, it was a string of wars with real victims and casualties (millions of whom fall beneath the notice of our trusted leaders.) Again, it may seem obvious to note here, but none of the victims of “ideological” wars had phone lines to Kennedy, and none of them were consulted about the violence visited upon them.
No one needs me to recount the bad faith involved in the Johnson administration continuing to prop up “containment” as a legitimate policy, or the decision to frame the Vietnam War as a “moral battle against communism” when the death tolls were climbing publicly everyday. Here, it’s not that capitalism is an inferior system or that communism is superior. It’s that at one point in the US, throwing your chips behind “free-market capitalism,” engaging in debates about capitalism versus communism as if they were the point — became less about “mere theorizing” but more a case of dodging the reality at hand.
I want to be clear that I am not trying to challenge the project of civilized discourse that my colleague proposes. Nor am I disparaging the democratizing potential of technology (about this I remain cautiously optimistic). I bring this up to re-focus why accusations of “bad faith” — even as they draw us into making indiscrete demands and abandoning politeness and propriety can be productive or even necessary. If nothing else, let us remember that “bad faith” is never about two parties discussing their ideas across a table — communism good / capitalism bad or vice versa. It is about the connection between the discursive games we rehearse and the questions they both pose and evade, the immensely costly self-deceptions in which our games can entrap us.
After all — to take another one of Oliver’s examples — the problem with cult leaders isn’t that their ideas are bad (let’s be honest, cultish ideas typically range from practical self-help to cheap messianism.) It’s that ideas are so often a form of doublespeak that conceals the all too predictable abuses that take place behind closed doors.
Just Who Is Accusing Whom?
Although Sartre does not reprise the concept explicitly in his later works, he returns to “bad faith” in a different form throughout the 1940s and 1950s through the questions of commitment and responsibility. In fact, the post-war popularity of existentialism owes much to its insistence on joining personal responsibility and historical realities at a time when so much of the world simply chose to “move on.” In the American and European imagination, the 1940s still persists as an unparalleled example of the abdication of responsibility by an entire people and civilization.
As our education teaches us, to be neutral in 1939 is to be guilty of a particularly unforgivable kind of cowardice. That such cowardice came, not through the violation of bourgeois, law-abiding behavior but through its embrace makes it a fertile testing ground for bad faith. To an extent, bourgeois life has always involved a degree of bad faith, of turning away from and talking over the uglier realities of poverty and prejudice. Yet the 1940s revealed just how much it was possible for people to deceive themselves.
I realize that no one wants to talk about Nazis, and it sounds like I’m making a moralizing digression. Yet it’s important to note the extent to which our narratives about the Second World War have served as the crucible through which we have developed our secular vernacular of bad faith — one that turns on “accusation,” “complicity,” and “collusion.”
In an oft quoted passage from Simone Weil’s “Reflections on War,” Weil reminds us that under extraordinary duress, we are still bound by forms of agency: “But the impotence one feels today — an impotence we should never consider permanent — does not excuse one from remaining true to one’s self, nor does it excuse capitulation to the enemy, whatever mask he may wear.”
Weil’s diagnosis of “impotence” is often read as an empowering rallying cry for activists, but it’s helpful to note that the possibility of agency also implicates our own powerlessness. Feelings of “impotence,” Weil cautions, are not admissible in the court of divine or historical judgment. One can be wrong about feeling impotent, and often, such feelings enable the slippage from neutrality to “capitulation” — from not wishing to know because it makes no difference to refusing to act.
We need only to look to our cinematic love affair with “good Nazis” to see how the interplay between “needing to know what it’s like, to understand” and “mere excuse-making” renders the Nazi era such a fertile staging ground for embarrassingly uncomfortable narratives. Films and stories like Jojo Rabbit, Good, The Reader etc. etc. tell us again and again how many Nazis were merely average people going about their business, feeding their pets and buying their groceries in a manner indistinguishable from us.
Yet this very relatability and familiarity, even as it implies plausible deniability — the mere accident of having been born at the wrong place, the wrong time — can’t help but feel disingenuous in a new way. Is it that they were powerless or that we are culpable by our intimate rehearsals of their powerlessness? Claims of German victimhood — of having been swindled by a charismatic speaker and caught up in the torrent of history — begin to sound like an eerie denial of guilt and a counter-accusation to the tune of “Cell Block Tango”:
“…if you’d’ve been there, if you’d’ve seen it — then you would have done the same.”
Living In Dark Times
In his poem “To Those Who Follow in Our Wake,” Bertolt Brecht reflects on how “bad faith” can become oddly environmental — something acutely inflected in every action, however banal or seemingly gentle:
Truly, I live in dark times!
An artless word is foolish. A smooth forehead
Points to insensitivity. He who laughs
Has not yet received
The terrible news.
What times are these, in which
A conversation about trees is almost a crime
For in doing so we maintain our silence about so much wrongdoing!
And he who walks quietly across the street,
Passes out of the reach of his friends
Who are in danger.
Writing from his Danish exile in 1939, Brecht reflects on how his own precarious safety abandonment of his colleagues and friends. Here it is not that one’s actions — eating, going about one’s business, having a conversation — have changed in themselves, but they cannot but reflect that the world has fundamentally changed. Actions that seemed neutral and safe are now a form of “silence” about wrongdoing.
Brecht shows us the strange, conspiratorial knowledge that unwitting collaborators under fascism share. In 1939 Denmark, everyone knows that the world has changed. But having been spared (if by geography alone), some people can go on eating, drinking, and tending to their gardens just as before. But going on as before is also to deny the very knowledge that enables a kind of mutual recognition — the knowledge that everything has changed.
Even though small gestures of normality are gestures of denial, Brecht shows us “bad faith” is double-edged — that denial is also complicity, that going on as before is also to tacitly testify to the fact that things have changed. Staying “silent” is to choose to walk “quietly across the street” and “pass out of reach” of your friends in need (something we may be inadvertently doing now as more lockdown measures loom). By behaving no differently as before, such actions reflect a choice not to know, and conspirators recognize each other through this theatre of dissembling. Gestures have lost their middle ground — as if to say, everything we now do, we do under the searing judgment of posterity.
Telling a Story Everybody Knows
What does this mean for communication? From our conversations in text and in person, I rather think that Oliver and I agree that there is something unproductive about public discourse as it stands. No doubt, we often talk at cross-purposes, and greater generosity is probably a good thing in our conduct toward others. Yet I rather disagree that a “falsifiability test” is what’s needed, and I am baffled by the assertion that machines are “faithful, diligent, and disinterested” teachers of good faith.
To repeat Sartre’s assertion — bad faith is not about lying to others. Like the “closeted” man who sends “signals,” all of them suggestive but none of them determining — the game is not quite one of “lies” and “facts,” but a game of mutual self-deception. Demanding to know — and to know together — is always a risk. We risk deceiving ourselves about the legitimacy of our demand and breaking the seal on the depth of our own culpability. Yet when the demand sticks, there is usually a story that needs to be told.
For instance — and I won’t linger on this because others have written about it with so much more eloquence — a simple message like “Black Lives Matter” tends to invite a myriad of bad faith responses in the manner of “all lives matter” or “blue lives matter.” Here, my colleague’s “falsifiability test” would, I venture, be useless if not insulting. In a case like this, it’s not about the verifiability of police brutality. As countless court cases have demonstrated, the nature of evidence has always been contingent on what the law chooses to know and not know. Yet, while statistics render visible sheer scale of police brutality and the prison industrial system, even setting them aside, if you live in the US, on some level, you already know. In its simplest formulation — this is what the charge of “bad faith” has to say: there is no way for you not to know, and I know that you know.
To be honest, I don’t know well enough what it means for a machine to be “faithful” or “diligent.” But if a machine is “disinterested,” this would, in my mind, rule it out as a participant. As I have tried to show, the relation between “accuser” and “accused” is a tricky one, but being interested is an inherent part of the so-called mechanics of bad faith. “My” bad faith is always a game I’m playing with someone else — someone who can either demand that I know what I do not wish to know (try to kill the game), or who colludes with me not to know (play the game). Either way — they cannot avoid becoming an interested party.
I am not sure my lengthy response helps us deal with “bad faith,” but I hope it redirects us to the problems in discourse that to my mind, are unlikely to be solved by better rules of engagement. Finally, I’ll leave my reader with an example in the Tom McCarthy film Spotlight — an efficient portrayal of The Boston Globe journalists uncovering the story of child abuse hidden in plain sight. Yet the trope of uncovering — of getting liars to ‘fess up — gives way to another story of widespread culpability, plausible deniability, and indiscrete questions.
At one point, while explaining why he does what he does, Stanley Tucci as the lawyer Mitchell Garabedian practically turns to the camera to inform us: “If it takes a village to raise a child, it takes a village to abuse one.” After a confrontation with the administrators at Boston High, Rachel McAdams as Sacha Pfeiffer tells her editor — probably talking about the audience as much as everyone in the film — “It’s like everyone already knows the story.”
By the end, the righteous satisfaction of uncovering the truth is somewhat unsettled by the fact that, as it turns out, everyone already knew — including the Globe itself. In the final scene, instead of encountering outrage at what is uncovered, the Globe is flooded with testimonials of abuse. The story about the church coverup turns out to be the story of a community that needed to tell itself the story it always knew — the story that everyone, it turned out, had been telling in one form or another.
