Liedzeit

Abstract

There is no universally accepted definition of lying. Most definitions focus on the falsity of statements, or the false belief in the truth of statements and on the deception motive. While Wittgenstein’s concept of family resemblance seems to suggest that a definition is neither desirable nor helpful I offer a new definition that is not based on any belief in truth or falsity but only on data available to the liar and not to the addressee of the lie. This, I hope, should shed some light on the concept of truth. In a second part Wittgenstein’s view on the language game of lying is discussed. What is the prerequisite for lies to occur in primitive language-games and why does the language game of lying have to be learned? And how do children in real-life situations first start lying?

1. What is a Lie?

If philosophers think about lies and lying, it is in most cases from an ethical point of view. Why do people lie? Why is it wrong to lie? Or is it? Am I allowed to utter a noble lie? There has also been interest from a logical point of view, judging from the number of books on the Cretan Liar paradox and its variants. But curiously in theories of Truth the concept of lying is more or less ignored, lying for example is not even mentioned in the Stanford Encyclopedia entry on Truth. And yet if Pavese’s intuition is right, and we know what “the lie is” and the lying is in some sense at least connected to truth, then the lie should help us understand the phenomenon of truth. What then is a lie? Whatever the definition might be, surely it must involve saying something that is not true and very likely the motive of lying must be to deceive others in order to gain some kind of benefit. The best (short) definition I am aware of is the one by Jeremy Bentham who said that the lie is “a wilful falsehood, uttered for an evil purpose”. (Bentham 1990, 204) This definition has the advantage of being short but since there are many different types of lies, bald-faced, white lies, black lies, blue lies, half-truths, honest lies, jocose lies and many more, it is no wonder that there is no universally accepted definition. Most people e.g. would not regard telling a joke as a lie although false statements are made. A polite lie, e.g. faking delight on receiving an unwanted gift, may be technically called a lie as there are false statements and there is an intention to deceive, (but it would be seen as a lesser sin). Or as no sin at all since there is no “evil purpose”. According to a Stanford entry on “The Definition of Lying and Deception” the “most widely accepted definition of lying” is the one by Arnold Isenberg: “A lie is a statement made by one who does not believe it with the intention that someone else shall be led to believe it.” (Mahon, 2016)

There are then four conditions to lying, the statement condition, the untruthfulness condition, the addressee condition (there is someone who is lied to) and the intention-to-deceive-condition.
Not surprisingly all of these conditions have been questioned. Aren’t there lies of omission? - No statement. Can one not make a truthful statement with the intent to deceive? As in Freud’s favourite joke: “Where are you going?” – “To Pinsk”. –“Liar, you want me to believe that you are going to Minsk, whereas you really want to go to Pinsk!” – No Untruthfulness. Can I not lie to a dog or a robot? – No person addressee. Must a lie succeed, in order to be a lie? If the addressee does not believe the false statement is it not still a lie? – No deception. (See Mason, 2016 for a detailed discussion.) Of course, this should not come as a surprise. As with any not totally trivial concept, a definition easily gets rampant or it does not cover all aspects.
Fortunately, Wittgenstein provides us with two tools to help clarify things, family resemblance and the language-game. The concept of family resemblance prepares us for the fact, that not everything that we reference by a certain word must have “something” in common. Just as in the case of Wittgenstein’s game example there is a family resemblance between statements we call lies. Internal lies, palter, double bluff etc. do indeed have something in common but not every trait is shared by all members of the family. So the joke is maybe only a distant cousin.
Does this not mean that defining lying is as unnecessary as it is futile? Isn’t it downright frivolous to even try to define lying? I think there is a case to be made for definitions.
First, even though a definition will not hold for all cases that people consider lies, it would be nice if a definition covered much ground. Then it could maybe be shown that the cases not handled are somehow derivative. Second, in the example of game it would be convenient to have a definition that would for example clearly differentiate game activity from sports activity. A club that offers its members sports activity has tax benefits compared to games clubs. That is why Chess or Go or Bridge clubs would love to see their “game” defined as sports. Without a clear cut definition a judge would be required to rule from case to case on the basis of her linguistic intuition alone. (Like the famous judge who said he did not know what pornography was, but could recognize it in a specific case.) And thirdly it seems to me that defining a concept like lying might be considered a language game in itself. It is something that philosophers do. So, with this in mind, I make the following move:

That is the addressee is expected to do something that he would not do if he had access to the same (or equivalent) data that the liar has access to.
A simple case would be: A kid says (s1) “The cat knocked over the vase.”
Say statement s1 is false, the kid in fact did the damage. But the lie would succeed if the parent refrains from penalizing the kid. The truth of s1 does not matter and neither does the belief of the liar or the addressee concerning the truth of s1. It is only the action, the penalty or absence of penalty that is important for the lie.
If the parent had access to the data, i.e. she had seen the deed herself or if there were camera evidence or circumstantial evidence (like blood on the kid’s hand and on the vase) the lie would not succeed. Imagine the kid replies to the accusation of having broken the vase (s2): “Of course, I knocked down the vase. Every time something gets broken in this household, it is always me!”
Let us say s2 is true. Clearly, s2 is not an admission of guilt. On the contrary, it is designed to make the parent feel guilty for making unfair accusations. The point though is, that the truth of s1 or s2 is not important and neither is the belief of the parent. She may very well strongly suspect the kid to be guilty in both cases. As long as there is no penalty from the perspective of the kid the lie has been a success. The “evil purpose”, avoiding a well-deserved punishment, has been achieved.
The advantage of definition L2 is that the actual truth or falsity of the statement is irrelevant and that the belief of liar and addressee as to the truth-value is also of no (real) consequence. Of course, one could say that the statement is designed to make the addressee infer wrongly as to what actually happened and that at the end it is truth or falsity that is in question. In fact it is the action (the punishment or non-punishment) that constitutes what it is that the addressee believes. And it is the belief that constitutes what is regarded true or false. Truth has then in this case a pragmatic, or even behaviouristic flavour.
In any case, definition L2 confirms Pavese’s dictum that we can know what a lie is (and we know how to lie) although we do not necessarily know what the truth in a particular case is and even less what “the Truth” is.

2. The Language Game of Lying

In PI 249 Wittgenstein asserts that: “Lying is a language-game that needs to be learnt like any other one.” The remark is in brackets, presumably meaning that this is self-evident, and hardly needs not to be stated. Nevertheless John Searle disagrees:

Searle seems to be of the opinion that truth telling is a language-game on its own and that lying consists in violating rules learnt in this game. This seems rather odd especially in the context, as Searle writes about the logical status of fictional recourse. And lying is in many cases not just a denial of truth but the construction of a new fictional reality. Searle has been severely attacked by Dale Jacquette but curiously he seems to agree with Searle that there is a “truth-telling” game. Lying is “derivatively a specific language game different from the truth-telling language game” (Jacquette 2004, 173)
But if one considers Wittgenstein’s examples of language games: giving orders, and obeying them, reporting an event, forming and testing a hypotheses, guessing riddles etc. (PI 23) it is apparent that a language game is a subset of language that highlights a specific aspect. What would be the unique feature of the truth-telling game, except that it is not lying? Jacquette starts by reintroducing us to the famous first example of a language game where a builder A gives orders to an assistant B about some building stones. A orders e.g. “Slab” and B delivers a slab. This is supposed to be a complete language game. (PI 2)
Now he asks us “for the sake of argument” to assume A to have lied. This is followed by a kind of wild goose chase where he fears that Wittgenstein’s remark “might be worse off than even Searle suspects” because “the lie appears to involve the inner psychological process of intending to deceive other language users by deliberately imparting false information.” (Jacquette 2004, 164) And do inner processes not suggest a private language? After some discussion Jacquette eventually assures us that a special feeling is not what is necessary but that lying is “a matter of communicating falsehoods with a particular motive” (Jacquette 2004, 169)
“What is characteristic of lying is a motive.” And a motive needs a “huge surrounding”. (PGL 1988, 213f.) Builder A could not have lied, because it is inconceivable that he could have a motive.
In fact, we should have rejected the assumption even for the sake of argument. The whole point of this language game is to show that at the communication is possible without thinking, without “inner processes”. Once the assistant is successfully trained he will bring the right building block. There is no room for error, or for doubt and much less for deception. The question then is how do motives occur, how do situations that allow lying arise?
To answer this we first need to answer Wittgenstein’s question “How is doubt introduced into the language-game?” (OC 458) In On Certainty Wittgenstein revisits and updates his builder game. Now the number of building blocks carried is given sometimes by counting sometimes by guessing. (OC 564)
If a member guesses the number of slabs on a heap this is where vagueness and doubt enter the language game. Similarly, with “‘I don’t know’... a new element enters in our language-games.” (OC 593)
Imagine A orders “Bring me as many slabs as you can carry.” This is obviously vague. B does not know how many slabs he can carry. So it might appear to him that it is a good idea to err on the safe side and to carry just a few. Questioned, if this really was all he could carry he answers “yes” knowing that A could not possibly know better. The situation is: people have access to different data. The motive in this case: laziness.
Actually, lying is quite complex. In real life children first start learning the lying game with a simpler variant of lying, denial. In a situation that might be called the transgression language game they do something that they might or might not know is “wrong” or forbidden. When asked “Did you do beat your sister?” a 3-year old will deny it even though the parent was present. (It might very well be that it is the tone of the question that leads the child to the denial, because it thinks this is what the parent wants to hear, rather than a fear of punishment.) A 6-year old will not make the same mistake. He will have developed what is called a “second-order mental state understanding” (Talwar & Lee 2008, 868), that is he will know not only what the other (the parent e.g.) can know about the state of affairs the lie was about, but also about consequences following from it. He will be able to maintain a lie in follow-up questions, what is called “semantic leakage control”. (Talwar & Lee 2008, 879) In experiments following the “temptation resistance paradigm” children are explicitly told by a researcher not to peek at or play with a toy when left alone. Not only does the majority of children peek anyway (82%), they also lie about it (64% of those who peeked). (Talwar & Lee 2008, 872f.) The interesting part is, that they now have to somehow conceal the knowledge gained in order to maintain the lie, although the peeking happened in the first place to gain an advantage in the game played by the illegal extra information. According to Talwar & Lee only with increased age children “begin to differentiate antisocial lies from honest mistakes, guesses, exaggerations, and eventually sarcasm and irony.” And only in early adolescence “children’s conceptual and moral understanding of lying and truth-telling becomes comparable to adults.” (Talwar & Lee 2008, 869)
It seems then that it takes many years to really master the Lying Game.

3. Conclusion

Wittgenstein’s remark about the necessity to learn the lying game is made in the context of the ironical question whether we are not too hasty in assuming the smile of the child is not pretence. The baby cannot lie and neither can the dog. Not because a dog is too honest, but because it is incapable of having a motive for lying. “You can’t teach a dog to tell his motives any more that you can teach a pig to retrieve.” (PGL 1988, 315)
The ability to develop motives is something that seems to be unique to humans and something that has to be learned. But a motive it not something that is necessary to tell the truth. There is no motive in “Slab”. First doubt has to enter the language game. And that arises through the possibility of errors of honest mistakes and misunderstandings and finally – lying. Only with the background of “untruths” it makes sense to call a statement true. Truth is derivative of error and lying, not the other way round. And because of this: “To convince someone of the truth, it is not enough to state it, but rather one must find the path from error to truth.” (GB 1993, 119)

References

Bentham, Jeremy (1990) “Judicial Fictions”, in: Philip Kerr (ed.) The Penguin Book of Lies, London: Penguin, 204-209.
Jaquette, Dale (2004), “Wittgenstein on Lying as a Language Game”, in: Danièle Moyal-Sharrock (ed.) The Third Wittgenstein, Aldershot: Ashgate, 159-176.
Mahon, James Edwin (2016) “The Definition of Lying and Deception”, The Stanford Encyclopedia of Philosophy (Winter 2016 Edition), Edward N. Zalta (ed.), URL = https://plato.stanford.edu/archives/win2016/entries/lying-definition/.
Searle, John R. (1979), Expression and Meaning: Studies in the Theory of Speech Acts, Cambridge: University Press.
Talwar, Victoria; Lee, Kang (2008), “Social and Cognitive Correlates of Children’s Lying Behavior” in: Child Development 79, 866-881.