The traditional view of lying holds that this phenomenon involves two central components: stating what one does not believe oneself and doing so with the intention to deceive. This view remained the generally accepted view of the nature of lying until very recently, with the intention-to-deceive requirement now coming under repeated attack. In this article, I argue that the tides have turned too quickly in the literature on lying. For while it is indeed true that there can be lies where there is no intention on the part of the speaker to deceive the hearer, this does not warrant severing the connection between lying and deception altogether. Thus, I defend the following account of lying: A lies to B if and only if (1) A states that p to B, (2) A believes that p is false, and (3) A intends to be deceptive to B in stating that p.
Lies and deception: an unhappy divorce
The traditional view of lying, with roots dating back at least to the work of Augustine in De mendacio, holds that this phenomenon involves two central components: stating what one does not believe oneself and doing so with the intention to deceive. More precisely
LIE-T: A lies to B if and only if (1) A states that p to B, (2) A believes that p is false and (3) A intends to deceive B by stating that p.1
LIE-T remained the generally accepted view of the nature of lying until very recently, with condition (3) now coming under repeated attack. The form of this challenge has been to produce clear instances of lying where there is no intention on the part of the speaker to deceive the hearer, thereby showing that (3) is not a necessary condition for lying. To this end, there are three central kinds of lies that are used as counterexamples: (i) bald-faced lies, (ii) knowledge-lies and (iii) coercion-lies.2 Each of (i)–(iii) is taken to show quite decisively that the traditional conception of lying has been radically misguided. In particular, it is concluded not only that lying does not require the intention to deceive but also that deception is not at all a part of what it is to lie.3 Thus, LIE-T has been replaced with a variety of competing accounts, none of which even makes mention of deception. If correct, this radical shift in our conception of lying has significant implications beyond the obvious ones involved in understanding the nature of this phenomenon. For one natural criticism that we might have of the liar is that she is engaged in intentional deception, where such deceit carries the weight of the prima facie moral wrongness of such acts. Divorcing lying from deception, however, also divorces it from this explanation of its prima facie moral wrongness.
In this article, I argue that the tides have turned too quickly in the literature on lying. For while it is indeed true that (i)–(iii) are lies and that there is no intention on the part of the speaker to deceive the hearer in such cases, this does not warrant severing the connection between lying and deception altogether. Thus, I replace LIE-T with the following:
LIE-L: A lies to B if and only if (i) A states that p to B,4 (ii) A believes that p is false and (iii) A intends to be deceptive to B in stating that p.
I show not only that LIE-L can capture all three kinds of lies that LIE-T cannot (i.e. (i)–(iii)) but also that non-deception accounts of lying wrongly count as lies classic cases of what I have elsewhere called selfless assertions. This reveals that, contrary to the currently wide-spread approach in philosophy, lying is indeed tied to deception as a matter of necessity.
1. Three counterexamples to the traditional view of lying
Let us begin with the first kind of counterexample to LIE-T: bald-faced lies. A bald-faced lie is an undisguised lie,5 one where a speaker states that p where she believes that p is false and it is common knowledge that what is being stated does not reflect what the speaker actually believes. For instance, suppose that a student is caught flagrantly cheating on an exam for the fourth time this term, all of the conclusive evidence for which is passed on to the Dean of Academic Affairs. Both the student and the Dean know that he cheated on the exam, and they each know that the other knows this, but the student is also aware of the fact that the Dean punishes students for academic dishonesty only when there is a confession. Given this, when the student is called to the Dean’s office, he states, ‘I did not cheat on the exam’.6
This is a classic bald-faced lie: the speaker states a proposition that he believes is false and both the speaker and the hearer know that this is the case and know that the other knows this. There is, then, no intention on the part of the speaker to deceive the hearer. In particular, in stating that he did not cheat on the exam, the student does not intend to bring about any false beliefs in the Dean, either about his cheating on the exam or about his beliefs regarding this event. Indeed, he may even wish for the Dean to believe that he did cheat on the exam, just to relish in the Dean’s spinelessness. Nonetheless, the student is clearly lying. This shows that condition (3) of LIE-T is false.7
The second kind of counterexample to LIE-T involves what Sorensen (2010) calls ‘knowledge-lies’. ‘An assertion that p is a knowledge-lie exactly if intended to prevent the addressee from knowing that p is untrue but is not intended to deceive the addressee into believing [that] p’ (Sorensen 2010: 610). For instance,
In Spartacus (Universal Pictures 1960), the victorious Roman general, Marcus Licinius Crassus, asks the recaptured slaves to identify Spartacus in exchange for leniency. Spartacus … rises to spare his comrades crucifixion. However, the slave on his right, Antoninus, springs to his feet and declares, ‘I am Spartacus!’ Then the slave on Spartacus’ left also stands and declares ‘I am Spartacus!’, then another slave, and another until the whole army of slaves is on their feet shouting, ‘I am Spartacus!’ (Sorensen 2010: 608).
Each slave in this case is offering a knowledge-lie; however, with the exception of Antoninus, none intends to deceive Crassus into believing that he is actually Spartacus. For once the second slave claims this identity, it is clear that he is instead aiming to prevent Crassus from learning who Spartacus is. Given that each slave seems to be lying, condition (3) of LIE-T is again shown to be false.
The third kind of counterexample to LIE-T involves what we might call coercion-lies. A coercion-lie occurs when a speaker believes that p is false, states that p, and does so, not with the intention to deceive, but because she is coerced or frightened into doing so. For instance, suppose that an innocent bystander witnesses the murder of a gang member by someone from a rival gang, but is threatened with death if she testifies against the murderer. Because of this, the bystander states on the stand at trial, ‘I did not witness the defendant murder the victim in question’.8 Here, the intention of the bystander is not to deceive the court into believing that she did not witness the murder; instead, her aim is to avoid retaliation from the defendant’s fellow gang members. Indeed, she may even desperately wish for the court to believe that she did witness the crime. That the court ends up being deceived by her statement is simply an unintended consequence of the action needed to achieve the aim of self-preservation. Despite this, the bystander clearly lies on the stand, evidenced at least in part by the fact that she could be found guilty of perjury. The intention to deceive is again shown not to be necessary for lying.
2. Non-deception accounts of lying
The combination of these three types of counterexamples provides a formidable challenge to the traditional view of lying and has prompted a flurry of alternative views. The three most prominent ones, offered by Fallis (2009), Carson (2010) and Sorensen (2007), respectively, are as follows:
LIE-F: A lies to B if and only if (1) A states that p to B, (2) A believes that p is false and (3) A believes that she makes this statement in a context where the following norm of conversation is in effect: Do not make statements that you believe to be false. (Fallis 2009: 34)
LIE-C: A lies to B if and only if (1) A states that p to B, (2) A believes that p is false or probably false (or, alternatively, A does not believe that p is true) and (3) A intends to warrant the truth of that p to B. (Carson 2010: 37)9
LIE-S: A lies to B if and only if (1) A asserts that p to B and (2) A does not believe that p. (Sorensen 2007: 256)
LIE-F, LIE-C and LIE-S are virtually identical in the first two conditions, with the latter accounts simply allowing classic cases of bullshit, where the speaker does not believe that p is false and also does not believe that p is true, to count as lies.10 They also all share the common feature of completely divorcing lies from deception. But merely stating what one believes to be false is not sufficient for lying, as speakers frequently say what they believe is false when being ironic, joking or reciting lines in a play and yet are not lying when they do so. For this reason, each proposal adds a further component to capture only those statements that are genuine lies. Fallis does so through requiring that the speaker believes that she is offering her statement in a context where the following norm of conversation is in effect: Do not make statements that you believe to be false. Since such a conversational norm is not believed to be in effect by a speaker who is being ironic, humorous or acting, LIE-F successfully rules them out as instances of lying. So, too, do LIE-C and LIE-S, the former because Carson understands intending to warrant the truth of a proposition as being a promise or guarantee that what one says is true11 and the latter because speakers typically do not offer flat-out assertions in cases of irony, jokes and acting.12
Given that all three accounts divorce lying from the intention to deceive, they can also capture bald-faced lies, knowledge-lies and coercion-lies, unlike LIE-T. When denying cheating to the Dean, the student certainly does not believe that the context is an ironic or humorous one, so he believes that the conversational norm is in effect: Do not make statements that you believe to be false; he intends to warrant the truth of what he says because he wishes to go on the record as denying having cheated on the exam; and he offers a flat-out assertion. The same is true of both the slaves claiming to be Spartacus and the innocent bystander denying having witnessed the murder in question. Neither the slaves nor the bystander thinks that there is anything about the contexts in which their statements are being offered that would prevent this usual conversational norm from being in effect; they all are inviting their hearers to trust them, even if the invitations are empty, and their statements are flat-out assertions.
3. Back to deception
Despite the virtues of these non-deception views of lying, I will now argue that they are misguided, as the divorce between lies and deception is an unhappy one.
The first point to notice is that there is a range of ways of being deceptive. Perhaps the most obvious is the one that is the focus of proponents of non-deception accounts of lying, where the aim is to bring about false beliefs in the victim of the deceit. But another, less explicit, form of deception is where the aim is to conceal information. According to the Oxford English Dictionary, deceit is ‘the action or practice of deceiving someone by concealing or misrepresenting the truth’. And Carson, despite endorsing a non-deception account of lying, claims that ‘[t]o conceal information is to do things to hide information from someone—to prevent someone from discovering it. Often, concealing information constitutes deception or attempted deception’ (2010: 57).13 Given this, I propose the following distinction, which will suffice for our purposes even if it does not fully capture all of the ways of being deceptive:
Deceit: A deceives B with respect to whether p if and only if A aims to bring about a false belief in B regarding whether p.
Deception: A is deceptive to B with respect to whether p if A aims to conceal information from B regarding whether p.
Concealing information regarding whether p can be understood broadly here, so that it subsumes, among other phenomena, concealing evidence regarding whether p. Moreover, notice that concealing information is importantly different from withholding information. To withhold information is to fail to provide it, rather than to hide or keep it secret. If I am trying to find a home for my challenging puppy, I withhold information about her lack of being housebroken if you do not ask me anything about it and I do not mention it. But if I frantically discard all of the training pads lying throughout my house before you come over, then I am concealing the information that she is not trained.14 Finally, notice that concealing information is sufficient, though not necessary, for being deceptive; thus, it is merely one instance of a more general phenomenon. Obviously, another way of being deceptive is to be deceitful, where one’s aim is to bring about a false belief in one’s hearer.
With this distinction in mind, let us return to the three counterexamples to LIE-T. In the case of the student’s bald-faced lie to the Dean, while he does not intend to deceive the Dean into falsely believing that he did not cheat, he does intend to conceal crucial evidence from the Dean that is needed for punishment from the university—namely, an admission of wrongdoing. Without the evidence provided by a confession, the Dean is paralyzed to take action against the student, and so concealment is the central aim of the student’s statement. According to our distinction earlier, then, the student does not intend to deceive the Dean, but he does intend to be deceptive to him.
In the case of the slaves’ knowledge-lies to Crassus, while it is clear that there is no intention to deceive Crassus into believing that they are all Spartacus, there is the intention to conceal the true identity of Spartacus. Each slave, besides Antoninus, aims to conceal the information that that person – that is, Spartacus – really is Spartacus and, in so doing, intends to be deceptive without deceiving.
Finally, in the case of the bystander’s coercion-lie to the court, while she does not intend for the court to believe that she did not witness the defendant murder the victim in question, she does aim to conceal the eyewitness testimony that can be used for a conviction. Otherwise put, the bystander is not aiming to prevent the court from convicting the defendant, but she is aiming to prevent the court from convicting the defendant on the basis of her testimony. Again, there is deception without the intention to deceive.
It is worth pointing out that one can be deceptive in the relevant sense, even if the information that one is aiming to conceal is common knowledge. In other words, ignorance of that which is being concealed is not necessary to be the victim of deception. To see this, notice that deception requires that A aims to conceal information from B, and A can certainly aim to do this even if A is ultimately, perhaps even inevitably, unsuccessful in achieving this. I can aim to win a marathon even if I know that I will ultimately fail to achieve this goal. In this sense, even if ‘conceal’ is a success term—that is, if A conceals x from B, A succeeds in hiding x from B—‘aiming to conceal’ is surely not—that is, A can aim to conceal x from B even if A fails to succeed in hiding x from B. It is, therefore, not available to the proponent of a non-deception view of lying to reject my analyses of the earlier cases by arguing that, because there is common knowledge of that which is being lied about,15 there is no concealment of information and, accordingly, no deception.
Thus, none of the counterexamples facing LIE-T succeeds in showing that the broader notion of deception is not necessary for lying.16 Moreover, that there is such a necessary relationship between lying and deception can be supported by considering a case of what I have elsewhere called selfless assertion.17 There are three central components to this phenomenon: first, a subject, for purely non-epistemic reasons, does not believe that p; second, despite this lack of belief, the subject is aware that p is very well supported by all of the available evidence and, third, because of this, the subject asserts that p without believing that p. Here is an instance of selfless assertion:
creationist teacher: Stella is a devoutly Christian fourth-grade teacher, and her religious beliefs are grounded in a personal relationship with God that she takes herself to have had since she was a very young child. This relationship grounds her belief in the truth of creationism and, accordingly, a belief in the falsity of evolutionary theory. Despite this, Stella fully recognizes that there is an overwhelming amount of scientific evidence against both of these beliefs. Indeed, she readily admits that she is not basing her own commitment to creationism on evidence at all but, rather, on the personal faith that she has in an all-powerful Creator. Because of this, Stella thinks that her religious beliefs are irrelevant to her duties as a teacher; accordingly, she regards her obligation as a teacher to include presenting material that is best supported by the available evidence, which clearly includes the truth of evolutionary theory. As a result, while presenting her biology lesson today, Stella asserts to her students, ‘Modern day Homo sapiens evolved from Homo erectus’, though she herself does not believe this proposition.18
Despite the fact that Stella’s statement that Homo sapiens evolved from Homo erectus satisfies all three non-deception accounts of lying above, Stella does not seem to be lying to her students. Why not? My answer to this question is that it is precisely because Stella does not intend to be deceptive to her students.19
To see this, the first point to notice is that Stella’s statement that Homo sapiens evolved from Homo erectus clearly satisfies the conditions put forth in LIE-F, LIE-C and LIE-S. She offers this statement to her students, where she herself believes that it is false. Moreover, as she clearly does not regard the context of her classroom as an ironic, humorous or theatrical one, she does so while believing that the following norm of conversation is in effect: Do not make statements that you believe to be false. The reason that she violates this norm is that she believes it is overridden or defeated by the duty to state what the scientific evidence best supports when teaching her biology lesson.20 Stella also intends to warrant the truth of the proposition that Homo sapiens evolved from Homo erectus because she is promising her students that what she says is true, just as she does when she states what she herself believes.21 And, finally, there is nothing about her statement or the context that prevents her statement from qualifying as an assertion.
The second point to notice is that Stella does not in any way aim to be deceptive to her students in stating that Homo sapiens evolved from Homo erectus. For though she does not herself believe this, she regards her own personal beliefs regarding religion—particularly those that are grounded in her relationship with God—as irrelevant to the information she conveys during her biology lesson. Reporting to her students what her religious beliefs are about the origin of humans would, for Stella, be comparable to sharing with them what her favourite aspect of evolutionary theory is. Both are irrelevant to her biology lesson. Given this, when Stella states to her students a proposition that she believes is false, her aim is not to bring about a false belief in her students or to conceal her own beliefs on the matter. In fact, we can imagine that she would willingly share her own views about evolutionary theory with her students, were they to ask her. Instead, Stella’s aim is to convey to her students the theories that are best supported by the current scientific evidence, which include evolutionary theory but not creationism.
The final point to make is that Stella is not lying to her students. Beyond the intuitiveness of this conclusion, it can be further supported by considering a slightly modified version of creationist teacher. Suppose that everything about the case remains the same, except that Stella states to her students that Homo sapiens evolved from Homo erectus, not because she regards her religious beliefs on the matter as irrelevant to her biology lesson, but because she will get fired from her teaching job if she reveals such beliefs to her students. In such a case, the aim of Stella reporting what she herself does not believe is to conceal her own religiously grounded beliefs on the topic, and thus she intends to be deceptive to her students. Corresponding to this, Stella’s statement also seems to be a lie.22
We have seen, then, that LIE-F, LIE-C and LIE-S all count as lies assertions that are clearly not, thereby showing that such non-deception accounts of lying fail to provide sufficient conditions for lying. Should we then simply add an intention-to-be-deceptive requirement to these accounts? No, since the alternative requirements found in these views (i.e. conditions (3) in LIE-F and LIE-C) also fail to provide necessary conditions for lying.23 In particular, such accounts fail to count as lies assertions that clearly are. To see this, consider first the case below:
deceptive anthropologist: Shawn is an anthropologist who visits a highly isolated tribal community living in the Amazon rainforest. He does not have any beliefs at all about the norms governing the conversations in their interactions. Nevertheless, he wishes to gain their trust quickly, and so he states to them, ‘My grandmother was an anthropologist who lived with members of your tribe decades ago, and so I feel as though I already know you’. Not only does Shawn believe that this is false, he also states this with the intention to deceive the tribe members into believing that he has a personal connection with their ancestors.
Shawn does not satisfy condition (3) of LIE-F because he fails to believe that the following conversational norm is in effect: Do not make statements that you believe to be false. But surely this lack of belief does not prevent him from lying. For not only does Shawn state what he believes to be false, he does so with the explicit aim to deceive the tribe members into believing this falsehood. That Shawn’s statement is a lie is further supported by noting that a committee investigating a supposed ethics violation involving his research methods would hardly think the matter resolved when he says, ‘Look, I didn’t lie to the tribe members since I had no idea what conversational norms were operative in their community’.
Consider, now, the following case:
sabotaging friend: Fran wants to sabotage the relationship between Sam and Betty, but does not want to be held responsible for their break-up. So she tells Sam, ‘Betty is cheating on you, but don’t take my word for it’. Fran not only believes that it is false that Betty is being unfaithful, she also offers her assertion with the deliberate intention to deceive Sam.
Fran does not satisfy condition (3) of LIE-C because she does not intend to warrant the truth of the proposition that Betty is cheating on Sam. Specifically, because she does not want to shoulder the responsibility of Betty and Sam breaking up, she makes it explicit that she is not promising or guaranteeing that what she says is true. Despite this, Fran is clearly lying to Sam, a verdict supported by the similarity the situation bears to the paradigmatic case of Iago lying to Othello about Desdemona’s fidelity.24
Given this, I propose the following account of lying:
LIE-L: A lies to B if and only if (1) A states that p to B, (2) A believes that p is false and (3) A intends to be deceptive to B in stating that p.
LIE-L avoids all of the problems afflicting rival views. It delivers the correct result in the three kinds of counterexamples to LIE-T: since the aim of the speaker is to be deceptive in bald-faced lies, knowledge-lies and coercion-lies, my view counts all three as lies. It also provides the right verdict in creationist teacher and its modified version: Stella does not lie to her students in the original scenario because her aim is to report what current scientific evidence supports regarding evolutionary theory, but she does lie in the modified version because her intention is to conceal her own religiously grounded beliefs about creationism to avoid termination. Still further, it provides the correct verdict in the counterexamples to conditions (3) of LIE-F and LIE-C: because there is the intention to be deceptive in Shawn’s statement to the tribal community in deceptive anthropologist and in Fran’s statement to Sam in sabotaging friend, LIE-L correctly regards both as lies. Finally, my account distinguishes lying from irony, joking and acting since the intention to be deceptive is present in the former statements, but not in the latter. Although there may be reason, then, to sever the connection between lying and the intention to deceive, lying nonetheless remains fundamentally tied to the intention to be deceptive.25
It is unclear what sort of ‘exemption’ could be added to Fallis’s account of lying to respond to this counterexample that would not simply be ad hoc. Moreover, by regarding this case as requiring only such a slight modification, it seems to have blinded Fallis to the far deeper point that there is a necessary connection between lying and deception. The positive view that I defend respects this point, provides a unified account of lying and does not need to resort to such ad hoc moves.
Norms of conversation can ‘clash’ with each other as well as with other interests that we have (Grice, op. cit., p. 30). For example when a teacher who believes in creationism has to give a lesson on evolution, Grice’s first maxim of quality comes into conflict with Grice’s second maxim of quality. If the teacher violates the norms against saying what she believes to be false solely to obey the norm against saying that for which she lacks adequate evidence, some (for example, Lackey 2007: 602) might want to say that she is not lying. In order to accommodate that intuition, my definition might be modified to include an exemption for such cases (Fallis 2009: 52, n. 74).