3.1 Argument from Consent

3.2 Argument from Self-Defence

Some have defended lying in business negotiations on the grounds of self-defence. For example, Dees and Cramton (Reference Dees and Cramton1991, 148) condone a policy of “defensive dishonesty,” whereas Strudler (Reference Strudler2005, 462) claims that “certain deceptive acts that occur in negotiation are morally acceptable because of their defensive function” (also Tenbrunsel Reference Tenbrunsel1998, 331).

Carson has been a particularly strong advocate of this approach with his “principle of self-defence,” or SD, as he abbreviates it. Carson’s SD states, “It is permissible to lie or attempt to deceive others about one’s negotiating position” provided that various necessary and jointly sufficient conditions are met. The real work is done by the first of these conditions, which is that “one’s negotiating partner is doing the same and is likely to harm one thereby.” Carson (Reference Carson, Brenkert and Beauchamp2010a, 356–57) emphasises that it must be “clear that the other person is misstating his intentions.” In the absence of such clarity, “the default position should be not to lie or deceive others” (Carson Reference Carson2010b, 194–95).

This conditionality severely restricts the application of SD. This is because you can sometimes know with a high degree of certainty that your counterparty is lying; and you can at other times be harmed by your counterparty lying; but it is hard to see how both can be the case at one and the same time. After all, you cannot be harmed by deception when you are not deceived. You may be harmed, of course, by a lack of information, but a lack of information is not itself a lie. A lie requires a false statement, and if you do not believe that false statement, it seems unlikely that it can harm you. It follows that, when self-defence would be justified, it is not actually needed. Indeed, in the absence of any harm to be averted, your counter-lies cease to be self-defence and become mere retaliation. Thus the conditions required to justify SD produce a “nil set.”

Carson anticipates this objection. In response, he argues that you can be harmed by deception even when you are not deceived—that is, that the set is not nil—because even failed attempts at deception can cause one to “reveal one’s own preferences” through “expressions of emotion” and “create uncertainty or fears that weaken one’s resolve” (Carson Reference Carson2010b, 197). Carson may be right. Perhaps there are emotional pressures that a liar can exert even when his lies are detected as such. For example, a bald-faced lie might demonstrate dominance and disrespect, placing the counterparty at a psychological disadvantage. But this feels like a marginal argument, and Carson admits as much in his concluding remark: “nonetheless, I concede that this objection supports the view that my SD principle seldom justifies lying or deception in negotiations” (Reference Carson2010b, 197; also 2010a, 358).

However, there is a rejoinder available to an advocate of SD. It is possible that you might know your counterparty is lying or will lie to you, without knowing for sure which statements of hers are lies. A clear case is when the counterparty makes two statements that are inconsistent: both cannot be true, but you may not know which is false. In theory at least, this is an instance when you can know the counterparty is lying and may be harmed by that lie. Again, though, this feels very much at the margin. After all, if you know for certain that your counterparty is lying regarding at least some of what she states, are you really likely (absent other supporting evidence) to attach great credence to anything she states, and sufficient credence to expose you to harm? It seems unlikely.

All the foregoing casts doubt on whether you can knowingly be harmed by a lie; but, even if you do know you are being harmed, it is far from obvious that lying in response would remove that harm. In the case of physical attack, most justifications of self-defence sanction actions that are designed to parry the attack, that is, to fend off the blows and prevent further blows. It is difficult to see how retaliatory lying would achieve anything equivalent.

The apologist could perhaps argue that retaliatory lying, though not removing the original harm, might nonetheless neutralise it in some way, through a balancing of the scales. In this vein, she might appeal to a legal distinction between justifiable conduct and excusable conduct. Joseph Heath (Reference Heath2018) characterises the latter as morally undesirable behaviour that has become habitual throughout a business domain and which all participants must emulate to remain competitive. Such behaviour is seen as excusable in that, although it is wrong, it is not considered blameworthy (523–26). The implication would be that, if everybody else is lying, then you cannot be blamed for lying yourself to “level up the playing field.” But Heath sets two conditions for such unjustifiable yet excusable conduct. First, the cost of not behaving in this way—in this case, of not lying—must be “unreasonably high” (529). This is rarely the case with lying, as I seek to demonstrate in section 3.5. Second, the party must not condone the practice as a whole but instead must take “positive action” (529) to curtail it. This would certainly rule out pre-emptive lying. And, as Heath goes on to warn, under this “ ‘everyone else is doing it’ argument” (520), there is a real danger that “we had no choice, our competitors were doing it” can all too easily slip “between ‘were doing it’ and ‘were going to do it,’ ‘were probably going to do it,’ or ‘might have done it’ ” (526).

What is clear is that Carson’s “principle of self-defence” at best justifies the apologist position in a very narrow group of cases. For example, it certainly could not justify Zach lying to Abe, as he does in our opening scenario. There is no evidence that Abe has lied first, or indeed ever lies, and therefore no case for self-defence as described by Carson.

The self-defence argument might, though, be reframed by the apologist in the language of reciprocity: if your counterparty lies to you and thus fails to show you what Stephen Darwall (Reference Darwall1977, 40) calls “moral recognition respect,” surely you are then relieved of any obligation not to lie to her. But this line of argument sits most easily with acts that are supererogatory, where my uncalled-for gesture of help, kindness, or courtesy places you under an implicit obligation to reciprocate, and, to the contrary, where my failure or refusal to make such a gesture leaves you without any such obligation. It sits much less easily with circumstances involving harm. To take the extreme case, you would not typically view the obligation to refrain from murder as one imposed by the principle of reciprocity. If someone commits murder, you may have a right of self-defence, but not a right to reciprocate. In like fashion, the principle of reciprocity alone seems insufficient to justify inflicting harm through lying. Such a response would best be described not as reciprocation but rather as retaliation. Stripped of its defensive aspect, reciprocity of this kind becomes a case of “a wrong for a wrong,” one harm justified by another—the morality of the gangland and the blood feud.

This concept of negative reciprocity is similar in some respects to a more permissive version of the argument from self-defence advocated by Dees and Cramton (Reference Dees and Cramton1991, 144). They argue for a “notion of defensive fair-play,” based on their “mutual trust principle.” This formulation has gone through revision over time, and it remains unclear whether it should be seen first and foremost as a moral or a prudential maxim (Dees and Cramton Reference Dees and Cramton1995, 829–30). But it essentially states that the obligation to negotiate honestly is diminished, or indeed eliminated, when there is an absence of mutual trust between the parties. In effect, this inverts the burden of proof stipulated by Carson. To justify lying, you need no longer prove that your counterparty is lying. Instead, you need only satisfy yourself that you have no reason to trust that he is not lying.

While Carson’s SD seems too restrictive, Dees and Cramton’s mutual trust principle would appear too permissive. In many instances, it would seem to raise again Heath’s concern and justify what Roy Sorensen (Reference Sorensen2007, 262) calls a “pre-emptive lie,” with all the escalation effects that entails. It is, as the authors acknowledge, a principle suited to “a Hobbesian state of nature” and counter-intuitive to most of us (Dees and Cramton Reference Dees and Cramton1995, 828). Furthermore, it is vulnerable to extension ad absurdum. As Strudler (Reference Strudler1995, 807–8) points out, “the Principle seems to allow one to do anything to one’s opponent, no matter how horrible.” If I am permitted to lie to you in the absence of positive evidence that you will not lie to me, then presumably the same principle must apply to other forms of pre-emptive injury, such as unprovoked physical assault. In other words, in the absence of evidence that you are not going to attack me, it would appear morally permissible for me to attack you. Such a pre-emptive right of injury, whether it be through physical assault or lying, is untenable.

Advocates of these self-defence arguments face, then, a two-horned dilemma. On one hand, if Zach is required to be nearly certain that he is being lied to by Abe (in line with Carson’s SD), lying in self-defence is almost never justified, because Abe’s claimed deception cannot harm Zach when Zach is not deceived; and anyway, it is unclear how a counter-lie renders the original lie harmless. On the other hand, to the extent that this requirement is weakened (in line with Dees and Cramton’s mutual trust principle), then to that extent, Zach is permitted to lie pre-emptively, but also, by extension of the same principle, to commit all sorts of other unprovoked injuries—injuries that cannot possibly be justified. It appears that any apologist case based on self-defence is thus, at best, extremely limited.

It seems the apologist must still look elsewhere for a cogent defence of the special exemption. The foregoing debates rest on foundations that are essentially deontological in character. They assume implicitly that people have rights and, in particular, a right not to be lied to, but that this latter right can be waived through consent or forfeited if, by threatening others, you give them a right of self-defence. I will now turn to an apologist argument from a different ethical tradition, consequentialism, that is, an argument that appeals to the consequences of lying to assess its moral rightness or wrongness.

3.3 Argument from the “Greater Good”

Applbaum (Reference Applbaum1999, 6, 187–97), Jonathan Cohen (Reference Cohen2002, 118), and Nathaniel Davis (Reference Davis2016, para. 2) all see the “greater good” argument lurking behind many of the special exemption claims of the “adversary professions”—in business as well as in law and government. Each cites Adam Smith ([1776] Reference Smith2012) and his theorem of the invisible hand. They do not suggest that Smith himself advocated lying. But they do claim that behind many of the arguments marshalled in support of the permissibility of lying in negotiations, there rests an assumption that, through mechanisms similar to those Smith describes, society’s greater interest is served––specifically, that the self-interested pursuit of individual gain, in an adversarial negotiation, generates a clearing price for the market and ensures an efficient allocation of resources to the benefit of society as a whole; and crucially that, should part of that amoral, self-interested pursuit of individual gain involve the use of deceit, this is permissible as long as it is an essential element of the mechanism that serves that morally laudable objective, the greater good. Or such anyway is the apologist case. It is, for example, an argument that has been employed to defend “spoofing” (deceptive orders and order cancellations) in high-frequency trading (Cooper, Davis, and Van Vliet Reference Cooper, Davis and Van Vliet2016, 3, 10–13). But, as Applbaum (Reference Applbaum1999, 180) warns, “the appeal to the good ends of adversary institutions in equilibrium is notoriously underdemonstrated.” A closer analysis is needed. Let us break down the key claims into two components.

First, is business negotiation, in particular price negotiation, vital to the “greater good”? This is contestable, but I will concede to the apologist that, in many instances, the answer is yes—at least in a liberal, capitalist, market economy.

But the apologist must also prove a second contention: that lying is an essential element of that business negotiation. This contention is much more doubtful and, I argue, false. It is important to note that the question here is not about protecting the interests of individual protagonists (I address this in section 3.5); rather, it is about the successful maintenance of the system for the benefit of society as a whole. In this context, it is very hard to see how lying is beneficial, let alone essential. Seen from the perspective of society as a whole, negotiation is about communicating relative availability and need so that these can be better matched. Lies are false communications. They introduce an information asymmetry, where one party has more information than the other, that is, knows the truth while the other is deceived. To return to the example of spoofing, it has been argued that, because other traders are able to see only half the trade, they are left “misinformed about the real supply and demand for financial contracts” (Hersch Reference Hersch2020, 218). In economists’ terms, such misinformation, or asymmetry of information, represents a market failure. An efficient market requires properly matched information (Applbaum Reference Applbaum1999, 192; Heath Reference Heath2014, 37), and lying undermines this. This would suggest that, although a suitable process of negotiation may be crucial to delivering the greater good, that process should not include lies.

This conclusion, however, faces a potential objection. While the consequentialist case for lying in negotiations remains largely implicit, one particularly strident apologist, Strudler (Reference Strudler1995, 818), does make the case explicitly with his “mutual advantage view.” Posing the question “Is deceit wrong?” he insists that “no plausible case [can] be made even for the prima facie wrongness of deception” (812). Strudler argues that, on the contrary, deception is “a mutually advantageous tool” (813) that creates “a safe device for indirect communication” (818). He bases this, as we saw earlier, on the role of lying as a form of signalling. In other words, Strudler believes that lying actually improves the flow of information and, presumably therefore, enhances market efficiency and the general welfare (816–17).

Dees and Cramton (Reference Dees and Cramton1995, 824) have pointed out that there is one sense at least in which Strudler’s argument is self-defeating. If the purpose of deception is really to improve information transparency and to enhance the knowledge of the counterparty, then it is not deception. And in practice, we do not treat it so. Sometimes a comment is so very obviously figurative, light-hearted, or clearly intended to mean something beyond its face value that it cannot be interpreted literally. It is not intended to deceive, nor to disadvantage, and, consistent with our opening definition, does not therefore constitute a lie. But this does not characterise most lying in negotiations, especially when such lying is between parties who do not know each other well or relates to matters other than reservation prices. Most “deceit” is just that: intended to deceive (even if it actually fails to do so). It does not aim to improve information transparency or market efficiency.

It is possible, though, to put a different interpretation on Strudler’s mutual advantage view. One might accept that individual deception is self-interested and intended to deceive, while arguing that somehow these various deceptions interact with each other to create a mutually advantageous outcome. This would appear consistent with Strudler’s (Reference Strudler1995, 818) signalling thesis and with his unconventional claim that lying is the solution to the lack of trust between the parties, rather than its cause. But again, it is hard to see how Strudler’s logic would apply beyond disclosure of the reservation price, and even then, it would seem to fly in the face of empirical evidence that supports the more conventional direction of causality, that is, that it is lying that creates distrust, not the other way around (Boles, Croson, and Murnighan Reference Boles, Croson and Murnighan2000, 255; Tyler, Feldman, and Reichert Reference Tyler, Feldman and Reichert2006, 75; Schweitzer, Hershey, and Bradlow Reference Schweitzer, Hershey and Bradlow2006, 15, 17; Gaspar and Schweitzer Reference Gaspar and Schweitzer2013, 168). Indeed, Raymond Friedman (Reference Friedman1993), studying success and failure in labour negotiations, has concluded that “the one factor that is consistently most important is trust” (435) and that the crucial contributor to that trust during the bargaining process itself is “believing what the other side says” (451). It is by not lying, by instead developing a reputation for speaking honestly, by building confidence, that you earn the right to be believed, to have what you say taken at face value. It might seem naively simple, but the most effective communication is when someone who is trusted tells the truth.

The worst instances of the deceit that Strudler advocates can lead to a failure to transact at all. Strudler (Reference Strudler1995, 816) describes a personal experience which seems to illustrate just this point. He recounts an incident in Madrid when a taxi driver “feigned shock” at the sight of Strudler’s extensive luggage, declaring that there was simply too much to be accommodated in the car (a lie). Strudler took this at face value and chose to rent a car instead, leaving the taxi driver outraged that Strudler had not sought to negotiate further. Strudler seems to feel that this demonstrates the importance of correct signalling, but it appears an equally potent illustration of the adverse consequences of lying in a negotiation.

As Leslie John (Reference John2016, 117) warns, far from being an essential element of negotiation, “lying … can be a real impediment to the creation of value in negotiation.” There is a danger that, as in Strudler’s case, false statements about the parties’ bargaining positions lead to a breakdown in negotiations, where actually a mutually beneficial transaction is possible (Dees and Cramton Reference Dees and Cramton1995, 825; Rogers et al. Reference Rogers, Zeckhauser, Gino and Norton2017, 468–69). This is particularly the case with reservation prices. For example, let us suppose that Abe is actually prepared to pay 90 per cent of list price, while Zach and his boss, pushed to the limit, would accept as little as 75 per cent. This is clearly a transaction that ought to be easily achievable and in the interest of all. After all, there is an overlap in acceptable outcomes (i.e., a cooperative surplus) equivalent to 15 per cent of list price. But if Zach insists (falsely) that his boss will not countenance any discount and Abe (equally falsely) claims he can pay no more than 70 per cent of list price, and they believe each other’s lies, then there is a danger that they will fail to reach agreement at all and walk away. A 15 per cent cooperative surplus has been transformed, through mutual lying, into an apparent 30 per cent gap, and a “deal that should happen” from a consequentialist perspective instead collapses.

Furthermore, there is a wider concern. As I have argued already, repeated lying undermines trust (Boles, Croson, and Murnighan Reference Boles, Croson and Murnighan2000, 255; Tyler, Feldman, and Reichert Reference Tyler, Feldman and Reichert2006, 75; Schweitzer, Hershey, and Bradlow Reference Schweitzer, Hershey and Bradlow2006, 15, 17; Gaspar and Schweitzer Reference Gaspar and Schweitzer2013, 168; Sunstein Reference Sunstein2021, 27). This in turn threatens the cooperative efforts that underpin social life (Lewicki Reference Lewicki, Bazerman and Lewicki1983, 74). For example, Dan Ariely (Reference Ariely2009, 215) reports a conversation with an Iranian student at MIT who complained that the lack of a platform of trust in his country meant that no one would pay in advance or offer credit and all hiring had to be done within the family. Normal commercial relations were not possible.

As Asha Rao and Stuart Schmidt (Reference Rao and Schmidt1998, 669) argue, trust “limits transaction costs,” because it encourages more informal exchange of information, without the need for exhaustive validation (also Arrow Reference Arrow1973, 314; Burr Reference Burr2001, 14), and conversely, lying increases those costs and can even lead to a collapse in the means of exchange. George Akerlof (Reference Akerlof1970) vividly illustrates this latter point with his seminal analysis of “lemons” (overpriced, poor quality used cars) in the automotive industry. Because of a lack of honest information, customers are only prepared to pay the price consistent with the average quality of all cars available in the market. Sellers of high-quality cars know their cars are worth more than this and therefore withdraw from the market. As a result, the average quality and consequent price of cars in the market declines, and more sellers withdraw. Before long, the only cars available for sale are the lemons, customer trust is undermined, and the market is threatened with collapse, thus highlighting “the major costs of dishonesty” (Akerlof Reference Akerlof1970, 495; also Arrow Reference Arrow1973, 307–8). As Roger Sullivan (Reference Sullivan1984, 2) warns, this “leads to the destruction of the system of value traded for fair value on which business as an enterprise is based” (also Hare Reference Hare1992, 11–12).

Communication and cooperation within a society rely on the presumption that most people are telling the truth most of the time. The survival of this presumption cannot be taken for granted. Empirical research reveals not only a tendency for lying to lead to more lying by the individual concerned (Barnes Reference Barnes1994, 44; Shell Reference Shell2018, 177; Gaspar, Methasani, and Schweitzer Reference Gaspar, Methasani and Schweitzer2019, 72); it also has an escalatory effect, where early use by one party encourages subsequent use by the other (Volkema, Fleck, and Hofmeister-Toth Reference Volkema, Fleck and Hofmeister-Toth2004, 333; Tyler, Feldman, and Reichert Reference Tyler, Feldman and Reichert2006, 75; Sobral and Islam Reference Sobral and Islam2013, 290–91; Olekalns, Kulik, and Chew Reference Olekalns, Kulik and Chew2014, 24). This “ethical fading,” as it has been called, weakens those very bonds that hold societies together and permit communal life (Tenbrunsel and Messick Reference Tenbrunsel and Messick2004, 224; Rees, Tenbrunsel, and Bazerman Reference Rees, Tenbrunsel and Bazerman2019, 28). But this may still understate the importance of trust. After all, for most of us, it has more than just an instrumental value: “trust is a basic human aspiration” (Banai et al. Reference Banai, Stefanidis, Shetach and Özbek2014, 673), something that meets one of our fundamental needs as social beings. It has a status that borders on a good in its own right—“a social good to be protected just as much as the air we breathe or the water we drink” (Bok Reference Bok1999, 26).

In short, the consequentialist case for negotiation is strong; but the consequentialist case for the permissibility of lying in negotiation is not. There is no necessary conflict between the search for truth and the search for agreement. Indeed, quite the reverse is true. And the costs of lying for society as a whole are considerable.

3.4 Argument from Fiduciary Duty

The fourth apologist argument is somewhat different in character, or at least in its target. It does not need to contest that lying in business negotiations is pro tanto wrong. Instead, it argues that whether it is pro tanto wrong or not is largely irrelevant in practice, because any pro tanto wrongness is consistently outweighed by another consideration: management’s fiduciary obligation to its stakeholders. The argument runs that managers may be obliged to lie to optimise the outcome for the other stakeholders in the enterprise. This is their fiduciary duty, and it stands before any queasiness concerning falsehood.

There has of course been a long-standing debate as to which stakeholders are owed such an obligation. The traditional emphasis on owners is associated with Milton Friedman (Reference Friedman1962, Reference Friedman1970), whereas others, led by R. Edward Freeman (Reference Freeman1984), have advocated a more inclusive approach, extending to customers, suppliers, employees, and the community at large. A popular solution among practitioners has been to refer to a duty to protect and further the interests of the company, thus avoiding any specific or preferential allegiance, while retaining the notion of a fiduciary obligation to promote the greater interest of a wider constituency that still falls short of society as a whole.

This particular language is very much tied to that of business, but it has its parallels elsewhere. It is part of what the lawyer Charles Curtis (Reference Curtis1951, 16) calls “the special moral code which governs a man who is acting for another.” Applbaum has made a close study of the legal profession in this regard, examining the moral status of lawyers making statements in or out of court that they themselves believe to be false, based on an avowed obligation to promote the interests of their clients. This “argument of constituted description,” as Applbaum (Reference Applbaum1999, 89) characterises it, claims that a lawyer fulfils different roles as a lawyer and as an individual. In the former role, he has a duty to his client which places him beyond many of the strictures of any broader moral code. As Curtis (Reference Curtis1951, 9) puts it rather baldly and contentiously, “one of the functions of a lawyer is to lie for his client.”

A similar distinction is evident, too, in the British Civil Service. Sir Michael Quinlan (Reference Quinlan1993, 542), a past permanent secretary of the Ministry of Defence, describes the need occasionally to bow to ministerial wishes and defend “a square circle.” He does not countenance direct lying but does believe that unelected officials, once they have argued the case in confidence with ministers, are duty bound to promote it in public, whatever their private reservations (Applbaum Reference Applbaum1999, 61–74).

These different appeals to a fiduciary obligation—the business executive’s based on a duty to protect his company’s interest, the lawyer’s based on a similar duty to protect that of her client, and the civil servant’s based on the principle of democratic accountability—vary in degree, but they all point to a common distinction between principal and agent. The argument is conceptually similar in all three cases: in the event of conflict, the agent has a moral obligation—to protect and further the interests of the principal—that overrides other moral considerations. Accused of deceit or other wrongdoing, the agent may respond that they are merely fulfilling the duty to their principal that their role requires. If, in the case of a negotiation, protecting the interest of the principal entails lying or distortion, and that is the will of the principal, then such also is the obligation of the agent.

I believe this argument fails for four reasons. First, it is questionable whether lying really is in the interest—at least the long-term interest—of the principal. In the corporate context, we should ask whether it truly benefits the company to have its executives lie on its behalf. Most progressive companies put considerable store by their reputation for honest dealings, a reputation often promoted and defended at great expense. But I will leave this point aside for now and take it up again later, when I examine the argument from practicality in section 3.5.

Second, the weakness of the principal–agent argument comes into much clearer focus when we view the situation from the perspective of the victim, viz. the recipient of the lie. Through the actions of others and through no fault of their own, the victim’s interests have been set back; the victim has been wronged. The agent and principal may seek to shuffle between them the responsibility for this wrong, but such shuffling does not make it go away. It does not negate the wrong. The moral impermissibility of the lie is not extinguished (Applbaum Reference Applbaum1999, 11, 109).

Furthermore, the principal–agent argument can be extended ad absurdum, in much the same way as was the case with the permissive version of the self-defence argument. If the agent can justify lying in the interests of her principal, then why can she not similarly justify any manner of other wrongs, all the way from fraud to murder? This is no merely abstract concern. Many appalling crimes have been excused in just this manner. In particular, a common defence of war crimes has rested on a duty to carry out orders from higher authority—in other words, the duty of agents to fulfil the wishes of their principals. To take the very obvious example of the Nuremberg Trials, as the Holocaust Encyclopedia reminds us, “the defendants generally acknowledged that the crimes they were accused of occurred but denied that they were responsible, as they were following orders from a higher authority” (US Holocaust Memorial Museum 2021). I am not of course suggesting that lying can be equated with genocide, but the argument from fiduciary duty is essentially the same in both cases, revealing its structural weakness.

Finally, the fiduciary duty defence of lying fundamentally misinterprets the relationship between principal and agent. If the negotiator is, to employ Friedman’s (Reference Friedman1970, 211) language, “the agent of the individuals who own the corporation” or, for that matter, of the stakeholders more broadly, then it follows that the manager’s obligation must be to stand in the owners’, or stakeholders’, shoes. Crucially, though, this implies that what would be morally impermissible for the owners must also be morally impermissible for the agent. The obligation of truthfulness is replicated, not exempted. As most commentators agree, “what is wrong for principals to do directly does not become right when delegated to their agents” (Cohen Reference Cohen2002, 119) and “I cannot (ethically) hire done on my behalf what I would not (ethically) do myself” (Goodpaster Reference Goodpaster1991, 68; also DeGeorge Reference DeGeorge, Bowie and Freeman1992, 65–66; Bowie and Freeman Reference Bowie, Freeman, Bowie and Freeman1992, 9; Quinn and Jones Reference Quinn and Jones1995, 35; Heath Reference Heath2014, 276, 315–16).

The apologist might rejoin that, whereas the owners are free to choose their own moral position, the manager as agent is bound by undertakings and, specifically, by an implicit promise to maximise profits whatever that takes. But such a promise would itself be morally impermissible and void. Applbaum (Reference Applbaum1999, 8) captures the point well: “The answer cannot simply be that the professional has promised. Whether the promissor is a contract killer or a contract liar, a promise to wrong another has no moral force” (also Quinn and Jones Reference Quinn and Jones1995, 35–36; Carson Reference Carson2010b, 173; Heath Reference Heath2014, 275).

The fiduciary duty, or principal–agent, defence of lying fails, then, on multiple grounds. Its basic premise that the interests of the principal are best served by lying is highly questionable (see further discussion in section 3.5); its shuffling of responsibility between offending parties cannot erase the wrong suffered by the victim; and its line of argument by extension permits any number of other wrongs that are clearly impermissible. Above all, though, it essentially misinterprets the relationship between principal and agent, a relationship that cannot turn wrong into right.

3.5 Argument from Practicality

The final apologist argument is different in character again. It rests not on a question of principle but rather on one of practicality. It insists that lying in a negotiation must be permissible, because a prohibition on such lying is simply infeasible, placing an unworkable burden on the negotiator. The typical test case is the reservation price. Surely, the argument goes, a negotiator cannot be expected to reveal her reservation price. That would be to unreasonably weaken her hand. And yet, the argument continues, if she is asked “is that your best offer?” and does not reply in the affirmative—even if that is a lie—then she has done just that. She has undermined her own negotiating position. This pragmatic defence of lying is, as far as I can tell, absent from the philosophical literature, but it is regularly cited by practitioners actually involved in negotiation. When challenged as to why they have claimed dishonestly that their offer was “best” or “final,” they are prone to reply along the lines of “well, what else was I supposed to say? If I’d said anything different, the other side would have known I had more in my pocket!” I will offer two responses to this argument from practicality: a principled response and a pragmatic response.

The principled response might be briefly characterised as “so what?”: doing the right thing often is difficult and disadvantageous, but that doesn’t mean it’s not the right thing. After all, there are plenty of parallel instances where we would not consider potential disadvantage an adequate justification for morally impermissible behaviour. For example, few would accept that because not participating in fraud, insider trading, or corporate espionage disadvantages your position, such disadvantage justifies, or even excuses, you committing such acts. Furthermore, it is quite normal for companies to incur significant costs pursuing corporate citizenship/CSR agendas that go well beyond the requirements of the law (for an extensive list, see Norman Reference Norman2011, 49). Adopting a similar logic, perhaps we should conclude that, however disadvantageous a prohibition on lying may be, that is the necessary price demanded by morally permissible conduct.

But I accept that there is something unsatisfactory about this principled response—at least when unsupported by a practical course of action. As Badaracco (Reference Badaracco2002, 39) says, “no one manages to climb even part way up the greasy pole without paying close attention to his or her interests.” For a business manager to act consistently against their self-interest would be a recipe for Darwinian extinction or, as Badaracco puts it, “martyrdom and self-immolation” (52). A moral requirement that they do so would be ultimately self-defeating. There needs to be a more practical alternative (Provis Reference Provis2000b, 10–11).

This is a challenging area, but experienced negotiators have found a practical way to meet this challenge—at least to a significant degree. They tend to avoid statements about their reservation price, including references to “best” or “final” offers. Instead, they seek to establish a consistent pattern of behaviour that means that a refusal to engage in such statements does not signal one thing or the other. They focus on the offer they are prepared to put forward at any given moment, reflecting the other terms and conditions and the knowledge then in their possession. They have a tendency to say less rather than more, to focus on the facts rather than the rhetoric. They certainly do not volunteer the “whole truth.” Instead, they withhold any comment on their reservation price and simply engage in a series of offers and counteroffers, none of which constitutes lying (Boatright Reference Boatright1992, 72; Carson Reference Carson2005, 402). This approach is not dissimilar to that Shell (Reference Shell2018, 185–86, 189) advocates in his book on negotiating strategy, Bargaining for Advantage.

How might this play out in our opening scenario? How might Zach negotiate effectively without lying? Well, one could imagine a conversation that goes something like this:

Zach has not lied, nor has he undermined his negotiating position. Yes, he has signalled some preparedness to compromise. But Abe almost certainly knew he would get some discount, and Zach has really given nothing else away. Zach has also managed to put Abe into the position of being obliged to make the next move, a key strategy favoured by honed negotiators. Furthermore, Zach has reinforced his personal standing and integrity with Abe, which may be significant in future dealings. By contrast, in the original scenario, Zach makes a categorical statement that his boss will not permit a discount, only an hour later to reveal by his own action in granting a discount that he was previously lying to Abe. Such behaviour, repeated over time, will lead Abe to discount much of what Zach says, even when he is telling the truth.

The preceding case illustrates an important point: lying might confer a short-term advantage, but it also incurs long-term costs for the liar that include loss of trust, retribution, and diminished reputation (Shapiro and Bies Reference Shapiro and Bies1994, 18, 29–31; Friedman and Shapiro Reference Friedman and Shapiro1995, 251; O’Connor and Carnevale Reference O’Connor and Carnevale1997, 513–14; Boles, Croson, and Murnighan Reference Boles, Croson and Murnighan2000, 254–55; Croson, Boles, and Murnighan Reference Croson, Boles and Murnighan2003, 157–58; Tyler, Feldman, and Reichert Reference Tyler, Feldman and Reichert2006, 75; Boero et al. Reference Boero, Bravo, Castellani and Squazzoni2009, 876; Rogers et al. Reference Rogers, Zeckhauser, Gino and Norton2017, 465–66). Maurice Schweitzer, John Hershey, and Eric Bradlow (Reference Schweitzer, Hershey and Bradlow2006) have researched the impact of lying on trust in particular. They find that while trust can generally be restored following a period of untrustworthy behaviour, “trust harmed by deception never fully recovers” (15).

More broadly, Joseph Reitz, James Wall, and Mary Sue Love (Reference Reitz, Wall and Love1998, 13) enumerate four significant costs in lying: 1) “rigidity in future negotiations”—the inability to alter one’s offer, having initially adopted a dishonest position; 2) “a damaged relationship with the opponent” once falsity is uncovered; 3) “a sullied reputation” in the wider community; and, in their view most important of all, 4) “lost opportunities”—the failure to identify mutually beneficial trade-offs because of false signalling around the value attached to different elements being negotiated.

This fourth component, the ability through open and honest debate to find mutually beneficial outcomes, has received particular attention. Such mutual gains bargaining (MGB), as it has been termed, was first popularised by Roger Fisher and William Ury in 1981 (Fisher, Ury, and Patton Reference Fisher, Ury and Patton2011) and has since become the subject of an extensive literature, with a notable focus on labour negotiations. Beyond simply increasing the resource to be divided, Dean Pruitt (Reference Pruitt, Bazerman and Lewicki1983) identifies four strategies for delivering MGB—or “integrative agreements,” as he prefers to call them. The detail of these strategies is unimportant here. The crucial thing common to them all is that they rely on an open exchange of information, for example, “information about the nature of the two parties’ interests and their priorities among these interests” (41). It is exactly this sort of information exchange, and the consequent value created, that the liar risks forgoing.

The apologist’s appeal to practical necessity rests on an erroneous assumption that those who fare best in negotiations are what Andrea Schneider (Reference Schneider2000, 24–28) labels “ ‘Rambo’ negotiators.” In her study of that perhaps most adversarial of all professions, the law, Schneider found the exact opposite was the case. In her survey of 690 lawyers in Milwaukee and Chicago, 53 per cent of “adversarial” (i.e., “Rambo”) lawyers were rated by their peers as “ineffective,” while the equivalent for “problem solving” (i.e., non-“Rambo”) lawyers was under 4 per cent. Aggressive negotiating tactics, including lying, are not necessarily outcome maximising, even for the aggressor (Burr Reference Burr2001, 10; Cohen Reference Cohen2002, 118).

A similar conclusion emerges from a simulated market entry exercise conducted by Rebecca Guidice, G. Stoney Alder, and Steven Phelan (Reference Guidice, Alder and Phelan2009) among 112 undergraduate and MBA students. In successive rounds, students declared their intent (or not) to enter the market and then acted accordingly (or not). Points were awarded for entering an under-supplied market and deducted for entering an over-supplied one. In effect, the students were negotiating with each other for the right to participate in the market. A declaration of intent that was different from the subsequent action was rated a “bluff” (in this case, a knowingly false statement intended to deceive, i.e., a lie). Guidice, Alder, and Phelan found that such “bluffing had a significant negative effect on performance” and concluded that “while it is often assumed that bluffing is advantageous, it actually is not” (547).

It is widely believed that it is the more experienced, “street-wise” negotiators who are most inclined to lie, but again, this is the inverse of the truth. Attitudinal studies, across several cultures, have reported a greater reluctance to lie among older negotiators (Volkema Reference Volkema2004, 75; Morse and Cohen Reference Morse and Cohen2019, 15). Deepak Malhotra and Max Bazerman (Reference Malhotra and Bazerman2007, 213), too, have found that their more seasoned executive students are much less likely to condone lying in negotiations than their younger MBA students. They attribute this to the fact that their executive students “have many more years of negotiation experience” and “have witnessed and experienced the consequences of negotiating in bad faith.” Asked about lying, Malhotra and Bazerman’s advice is crystal clear: “Don’t do it. Don’t ever lie” (Malhotra and Bazerman Reference Malhotra and Bazerman2007, 218; also Nyerges Reference Nyerges1987, 21, 24; Shell Reference Shell2018, 194).

However, to conclude that lying is invariably to the disadvantage of the liar—that “honesty always pays”—would be unduly sanguine. There is evidence that lying can confer at least short-term advantage (O’Connor and Carnevale Reference O’Connor and Carnevale1997, 504; Aquino Reference Aquino1998, 208; Croson, Boles, and Murnighan Reference Croson, Boles and Murnighan2003, 156–57; Gaspar and Schweitzer Reference Gaspar and Schweitzer2013, 161; Rogers et al. Reference Rogers, Zeckhauser, Gino and Norton2017, 464–65; Gunia and Levine Reference Gunia and Levine2019, 134; Gaspar, Methasani, and Schweitzer Reference Gaspar, Methasani and Schweitzer2019, 64, 74). It often goes undetected (Ekman and O’Sullivan Reference Ekman and O’Sullivan1991, 916; Bond and DePaulo Reference Bond and DePaulo2006, 222, 230; Adler Reference Adler2007, 70). And there are clear cases of negotiation that are inherently single-shot, such as the purchase or sale of a used car or a transaction in a tourist bazaar. In these cases, we must accept that lying may give some partial benefit to the liar. But this does not characterise most business negotiations. In the main, business negotiations are held between parties in the same or related industries or communities. Such negotiations are not by their nature single-shot but instead form part of an extended relationship or at least involve parties who know of each other through reputation. Here the long-term costs of lying are much greater.

In summary, there are methods available to protect one’s negotiating position while refraining from lying. And lying does not come free; it carries costs of its own, especially in the longer term. Experienced negotiators, like the economist and merchant banker John Rutledge (Reference Rutledge1996, 78), know that, far from being systematically disadvantaged, “nice guys often finish first.” And the inverse of that holds too. As another financier, the Goldman Sachs whistle-blower Greg Smith (Reference Smith2012), declared in the New York Times, “if clients don’t trust you they will eventually stop doing business with you. It doesn’t matter how smart you are.”

But the apologist may raise a further objection. The practical strategy, outlined herein, could lead to the accusation of inconsistency. The apologist might argue that it relies on an untenable moral distinction between active lying and passive withholding of information. Why, she may ask, is passively withholding information from the other side morally any different from actively lying to them? Both can prove seriously misleading. And clearly the consequences for the counterparty in each case can be equally damaging.

This is a complex area. Some radical consequentialists, like Peter Singer (Reference Singer1972, 229–43; Reference Singer2009, 15–16), are reluctant to accept that there is any morally significant difference between doing (acts of commission) and merely allowing (acts of omission). And yet, if we examine the fundamental objections to lying, whether they be Kantian, consequentialist, or aretaic, it does seem reasonable to draw such a distinction. From a Kantian perspective, lying uses others as a mere means, while withholding information only represents a refusal to be useful to others, that is, a refusal to subordinate our ends to their ends—something that is arguably supererogatory. Equally from a consequentialist perspective, withholding information, except in the most egregious cases, does not engender the same broad mistrust as does lying. It does not fundamentally threaten social cooperation and cohesion. Indeed, it seems that some degree of withholding of information is essential to that social cohesion (Sullivan Reference Sullivan1984, 11; Sunstein Reference Sunstein2021, 33). For example, it is hard to imagine a functioning society where people volunteered all their thoughts about each other all the time—from sneering contempt to unbridled lust. The film Liar, Liar provides a humorous illustration of just that (Shadyac Reference Shadyac1997). Virtue ethicists in their turn might argue that, whereas a refusal to lie is part of the virtue of honesty, willingly divulging one’s reservation price would betray an excessively open character, bordering on naivety. In Aristotelian terms, a simple refusal to lie, but to say no more, might reasonably be seen to sit at the golden mean between opposing vices of manipulative lying, on one hand (a deficiency of truthfulness), and a mindless and naive refusal to withhold information which one is not obliged to disclose, on the other (an excess of truthfulness). Furthermore, this ethical distinction between lying and simply withholding information draws some support from the law. As Chris Provis (Reference Provis2000a, 155) points out, “if questioned by the police, I may not lie to them, but I have the right to remain silent, and fundamentally the same distinction is available to negotiators.”

In summary, passively withholding information is morally different from actively lying. Not lying, but at the same time refusing to reveal, for example, one’s reservation price, is therefore a morally viable option. The apologist’s argument from practicality is neutralised, because there are means available to the ethical negotiator that allow her to protect her position. And there are many other practical benefits that accrue from refusing to lie.