| Chapter 4 | |
| Is it even possible to get along? | |
| LetÕs start by asking a fundamental question: How can you get people with two different opinions to agree? If I hold opinion X, and you hold opinion Not X (and in the case where there is no obvious way to Òsplit the differenceÓ), how is it at all even possible to come to an agreement? | |
| The only a priori solutions might seem to be that either one or the other has to win out, based on who has the most power. Or that one of the participants has to change their mind, which is often difficult, because they may feel that betrays the values that caused them to take the position in the first place. | |
| The problem is that many people think of a negotiation as being a competitive event, like a prize fight. They think the object is to have your own point of view prevail, by a show of superior strength over the other party. The goal is to land the devastating punch that the other party has no effective counter for. | |
| In the competitive view, logical arguments are a weapon, like a clenched fist. Both parties are obliged to put up the front that the most logical argument will carry the day. But, in deciding what to say, they donÕt necessarily search for the most logical argument. | |
| Instead, like lawyers do, they cherry-pick the best arguments to present their side, and the ones they feel the other side will have the most difficulty responding to. In the worst case, they use tricks that exploit peopleÕs cognitive biases and limited rationality, such as stacking the deck, or crass appeals to strong emotions. | |
| If the participants are debating for the benefit of independent third parties who are trying to make up their minds, like politicians trying to inßuence voters, the third parties are just left with a game of LiarÕs Poker. They have little to trust or rely on to make up their minds. | |
| Goal stacks | |
| DonÕt stack the deckÑmerge your goal stacks. Imagine that your mindset in going into a discussion is not that itÕs Òus vs themÓ. Imagine that we donÕt even think that there are two sides. ThereÕs just us, and weÕre trying to collaborate to get whatÕs best for all of us. WeÕre all on the same side. (If you have to think of another side, because you think having another side will help motivate your team, letÕs imagine the ÒenemyÓ to be human suffering, poverty, unhappiness, etc. The disagreement itself is what weÕre all trying to fight.) | |
| Differing opinions are typically not the whole story. Opinions that people have are not isolated from one another. ThereÕs a structure to all the opinions that a single person or group might have at any given moment. | |
| ItÕs determined by peopleÕs values and goals. People have reasons for believing in one thing or another. Some are more important than others. And they are typically arranged in a hierarchy of importance, in what AI people call a goal stack. | |
| The goals on a stack are linked to each other by answers to the question ÒWhy?Ó. A small child might wear out their parents by asking, ÒWhy are we in the car, Daddy?Ó. Parent: ÒTo go to the airport.Ó ÒWhy?Ó ÒTo take a plane.Ó ÒWhy?Ó ÒTo go on vacation.Ó ÒWhy?Ó ÒTo have fun.Ó DonÕt get mad at the kid. TheyÕre trying to learn your goal stack. | |
| A common mistake in negotiation is to compare positions between the parties, notice where there are differences, and try to Òhammer outÓ the differences one by one. As with any hammer, thereÕs always the danger of hitting your thumb instead of the nail. | |
| Instead, negotiations should begin by trying to understand what values and goals the parties have in common, and try to expand that to as wide a range as possible. Only then can you work on resolving differences. Common goals become criteria which both sides can use for evaluating candidate solutions. | |
| The idea is to first work up the goal stack chain to the point where all parties share common goals, including the common goal of coming to terms with each other, and then work downwards to resolve differences (consistent with the shared goals) and secure mutual agreement. | |
| LetÕs say we have a labor-management dispute where the workers want higher salaries and the management wants lower salaries. If that were all there were to it, the side with the most power would simply win. At best, they could split the difference. | |
| But they both have a common interest in seeing a successful business with happy, healthy workers. Labor could make arguments that the business would lose the best workers to higher-paying companies. Or that the wages were difficult to live on, causing workers to take exhausting second jobs. Or that the company was doing well, and they deserve a fair share. | |
| Management, for its side, could open the books to labor so that they could independently verify that if the wage increase were implemented, it might jeopardize the competitiveness of the business or cause layoffs. Or it could show a competitive analysis showing that wages were comparable to competing companies. Or either side could propose a profit-sharing plan that would automatically adjust to market conditions. Or propose arbitration or some other ÒfairÓ way to decide, since both sides share an interest in a fair decision process. | |
| What if the goal stacks donÕt match? | |
| The reason to be optimistic that agreement can almost always be achieved is that, since weÕre all human beings, the topmost levels of our goal stacks are quite similar. | |
| Psychologist Abraham Maslow wrote about the Hierarchy of Needs [Maslow 1943] he felt were common to humanity. People first need their basic survival needs met: food, clothing, shelter. Then they need good social relationships, a feeling of security, and finally, what he called self-actualization: love, meaningful purpose to life and activity, and feeling of achievement. | |
| That finding common values is almost always possible is practically guaranteed by the Maslow hierarchy. So itÕs a question of working our way up to the point where the parties share common values, then collaborating on how we can best promote the values we share in common, in a particular situation. | |
| ItÕs very important to understand the relationship between concrete positions on issues, and the values that underlie them. These can often be elicited by simply asking the other party, ÒWhy do you hold that position?Ó. Sometimes it may take several iterations of ÒWhy?Ó before you get to what the other party really cares about. RosenbergÕs Nonviolent Communication [Rosenberg 2003] shows the importance of eliciting and being sympathetic to the other partyÕs needs, and presents concrete techniques for doing so. Eventually, you can build a conceptual model of the other sideÕs goal stack and priorities. | |
| Sometimes the positions can be quite different, but when you go up to the level of the values, it may turn out that both sides can agree on them. Sometimes the positions might seem unreasonable, but the values behind them are quite reasonable. Sometimes there might be another way of satisfying the values of both parties, that might involve concrete positions that will be more acceptable to the other side. Even in seemingly ÒintractableÓ situations, this can be the case. | |
| ItÕs always absurd to hear Arabs or Israelis insist that the conßict is so intractable that thereÕs no possible solution. Of course there is. Most people in the world live in places where they do feel secure, despite neighboring peoples with differing religions or ethnic backgrounds. What could possibly be so special about these two groups of people that would preclude them from living in peace as others do? | |
| There have been many other similar situations where longstanding conßicts have been resolved. Most dramatically, in Northern Ireland, where despite a centuries-old history of hostility, and terrorism in modern times, things now seem peaceful (August 2017). Similarly, in El Salvador, the Balkans, and other cases, peace was achieved. So why canÕt we solve this one? | |
| If your goal stacks differ, trade | |
| Some situations might be over-constrained, and the best you might be able to do in that situation might be a trade-off between mutually exclusive inßuences. But still, finding the best trade-off should be, a common value that both sides can work towards together. Here, the problem-solving mindset common in science and engineering, but sadly lacking in politics and business, is the best guarantee of success. | |
| Scientists and engineers donÕt get upset at the prospect of a trade-off, they just set out to figure out what the best trade-off is, where possible, trying to quantify it. They figure out how to change the situation to make trade-offs less necessary or less painful in the future. | |
| Uncovering differences in values and priorities can pave the way for creative solutions. If the parties value the same things differently, that can create the basis for a sensible trade. I give you the thing I care less about and you care more about, in exchange for getting from you the thing I care more about and you care less about. | |
| People can change their mind on concrete positions when they understand that there might be different ways of satisfying the values that underlie those positions. They donÕt have to feel like theyÕre surrendering or capitulating. | |
| Back to our labor-management dispute, say the workers held the salary raise as a high goal. But they were also willing to make productivity improvements that didnÕt seem too onerous to them, but would make the company more money. Then, the employer might not mind giving the raise. However, the employer might not believe that higher salaries would raise productivity, so they make a deal that the amount of the raise would be tied to the amount of the improvement in productivity. Since the employees believe that their salary raises will increase productivity, they agree. Since the employer doesnÕt have to pay unless the improvements are realized, he agrees. | |
| As in many of the situations we have been examining, this is another case where the principles of the Iterated PrisonersÕ Dilemma apply. A will to cooperate on the part of both players paves the way for a win-win outcome, where an attempt to compete, or to act on a fear of the other party, leads first to the danger, then the inevitability, of a poor outcome. | |
| Once our goal stacks align, we can collaborate on solutions | |
| LetÕs say a proposal is made (doesnÕt matter Òwhich sideÓ it comes from since weÕre all on the same side.) Now, we are going to discuss its merits and demerits. The very person who made this proposal should be able to say ÒThis disadvantage of this proposal is it will decrease the health of population XÓ. From the ÒcompetitiveÓ perspective this person is inconsistent, because they are shooting down their own proposal. If they were a politician, theyÕd be accused of Òßip-ßoppingÓ. | |
| But from the collaborative design perspective, they are helping everyone understand the implications of the proposal. Yes, they risk making it less likely that ÒtheirÓ proposal will be adopted wholesale. | |
| But imagine that someone else says ÒOh, now I get it and, yes, I see that that is a problem but we can mitigate it by doing YÓ. Now we have a better proposal than the initial one and the proposer should be delighted that the group has an even better solution. | |
| The original proposer should feel great that they started off the conversation with a basically good idea, and they were big enough to articulate a misgiving which lead to an even better proposal. The ÒopponentÓ should also feel good about themselves. They understood the basic proposal, and a ßaw, and then invented a fix for the ßaw. This is win-win. | |
| LetÕs imagine a variant of the above scenario. Instead of the original proposer coming up with the objection, someone else does. What happens? Well, often the original proponent will have a solution to this objection, they just failed to mention it in the original proposal. The critic did a great job. They helped the proposer better articulate the original vision. | |
| Our third scenario is the same as #2 except that the original proposer, upon hearing the objection says ÒWhoops, I didnÕt think of that. I donÕt know how to fix it.Ó. Had the critic not raised the objection, and the whole group adopted the original proposal, weÕd have an easy consensus. But weÕd only find out the ßaw upon implementation. | |
| Much better to find out earlier, before deployment. So the condition of having a plan with the known ßaw is better than the original plan. Our critic has added value, and should be commended, even by the original proposer. Maybe a second objection will help us see a pattern to the objections, and assist in discovering a solution to both. | |
| Reaching agreement by taming complexity | |
| Complexity undermines our ability to reason clearly. The human mind has a limited capacity for complex concepts. This capacity is easily exceeded, especially when weÕre operating competitively, which tends to restrict thinking to Òhow to winÓ. Thinking about collaboratively designing new alternatives helps expand the range of possible solutions. Furthermore, we can extend our cognitive capacity the same way we extend our physical limitations: with tools. | |
| Computers are tools that fundamentally help us manage complex knowledge. Decision support software can help record and communicate the rationale for, and dependencies between, questions, positions, reasons, proposals, and their relationship with human values. It can help analyze and optimize trade-offs, and compute the combinatorics of possible deals that rely on value differences. WeÕll further explore this possibility in the chapter Tools for Reasonocracy. | |
| By cooperating with each other and with our machines, we can end the prize fights, and keep our eyes on the prize |