In response to your discussion of the Principal-Agent problem:
Voss covers this in his book, calling the principal the "man behind the table." One of your tasks is to find out if there is a man behind the table and what he wants.
I worry that this is a reason as memetic immune disorder type of situation. That in your journey towards becoming a more skilled negotiator, effectiveness will sometimes go down instead of up.
For example, when I imagine myself saying something like "it sounds like you're very happy with the way this turned out", it feels like it comes out very awkwardly, and that they can tell it is me trying to game them rather than acting authentically, which would backfire and make me less effective rather than more effective via that first principle of acting in good faith.
Good point! Voss talks a bit about how many of these techniques feel odd. Two points he makes:
I would also add that it's more important to stick to things you believe in than to try to literally apply something that you feel is bad or wrong. If you're convinced that, in a given situation, a label of "it sounds like you're very happy with the way this turned out" is a gaming of the other person, don't use it. But if in a situation you think it's actually an accurate label that helps summarize the situation and correctly shows the other person that you are tuned in to what they are feeling and expressing, do it! Just keep an open mind to the possibility of using labels.
Summary (added): Basically I think if you use low-stakes practice and only selectively apply to the real world the skills you are comfortable with, you don't need to experience an intermittent dip in effectiveness due to not feeling authentic.
Thanks for the summary. Just a quick note: I found that multiple headers & jumping between different "lenses" made it much harder to me to read the post. I guess I'd wish that those blocks would be larger, possibly all the way separated.
Thanks for the feedback! It seems like you're saying I should first have done "negotiation techniques" then "do these negotiation techniques have a place in rational discourse?" as separate sections. So if we make a table with rows as techniques and columns as lenses, then I should have traversed it column major instead of row major.
Did I misunderstand or miss an angle to what you're saying?
Having read the book "Never Split the Difference", I can say it is decent. One of the best books in the self-help genre I've read. Very actionable.
It seems worth mentioning that leverage is hugely important. Both 1) having it, and 2) understanding it. For example, suppose you are a programmer applying to companies. 1) It's helpful to be good at interviews and have a lot of companies interested in you. 2) It's helpful to be aware of this fact, and to be aware of what sort of leverage the companies have. Ie. BATNA.
Maybe you can call what I am referring to as hard skills and what you are referring to as soft skills? I feel like that isn't a great way to categorize is, but nothing better is coming to me. Whatever the categories are, I think it would be good to explicitly mention that this article is targeting one of them, and that there are other things that are important for the bigger picture of being able to negotiate well.
Good point; I added a section clarifying this focus: https://www.lesswrong.com/posts/CRAzG386t3suSqDgd/chris-voss-negotiation-masterclass-review#Low_level_execution_focus_rather_than_domain_specific_tactical_or_business_school_style_strategy_focus
I really like how you anticipated and addressed all of these questions, and had solid, practical answers to them.
This post is about the negotiation MasterClass taught by Chris Voss as well as a related MasterClass Session by Chris Voss and more broadly about the techniques and worldviews shared in those. I watched the original MasterClass in April 2020 and watched the Session in December 2021.
My post discusses some of the relationship -- similarities and differences -- with rationalist advice. These portions should be of particular relevance to LessWrong and are part of the reason for posting to LessWrong.
This is a fairly lengthy post, so please use the sections to navigate to the portions you are most interested in.
I cover a lot in this review:
ETA 2021-11-24: I added a few more subsections and made some edits to existing content. You can see the change set here.
ETA 2021-11-25: I added a subsection in response to a point raised in the comments. You can see the change set here.
ETA 2022-01-02: I expanded and restructured the review to cover the MasterClass Session that I consumed in December 2021. The original post, written in November 2021, did not cover the Session. You can see the change set here.
Christopher Voss worked at the FBI as a lead hostage negotiator. In 2008 he left to form his own company, The Black Swan Group, that offers coaching to people on negotiation techniques. Their customers include people in real estate and many other kinds of business.
MasterClass is a player in the growing edutainmment space, combining education and entertainment. There are 100+ "classes" in MasterClass. Each class is taught by a subject matter expert (the teacher), and includes several videos (ranging from 2 to 30 minutes, usually 5 to 20 minutes). The typical video format just includes the teacher speaking to the camera, but there are occasionally interactive sessions with other participants, or other co-instructors.
Course notes can be downloaded. There are also community features on MasterClass.
You may have seen video ads for MasterClass on YouTube!
In late 2021, MasterClass introduced Sessions, a more interactive form of MasterClass that is designed a bit more like a course. A few differences between traditional MasterClasses and sessions:
There are stronger nudges to consume a session chronologically; later activities in a session only unlock after you mark earlier activities as completed.
While most of the activities in both regular MasterClass and sessions are video consumption, the sessions tend to feature more videos where you, the viewer, is expected to do something active, such as observing or counting something. Although so far these videos don't have interactivity at a technological level (e.g., you are not supposed to click when/where something happens), a conscientious viewer can get the benefit of active learning.
The sessions also include explicit prompts for viewers to enter information on their progress or thoughts. You have to enter something to proceed (you can't just skip it) though if you really just want to proceed, you can enter gibberish and check the box to not share it with others.
As far as I can make out, unlike regular MasterClasses, you cannot buy sessions individually; they're only available as part of a subscription.
A MasterClass annual subscription all-access pass costs $180; this would allow you to consume both the regular MasterClass and the MasterClass Session within the year you got the subscription for.
A single MasterClass (such as Chris Voss's) can be purchased for $90. As far as I can make out, sessions cannot be purchased individually, and you need to have a subscription.
There are 18 videos and the total video time is 3 hours 4 minutes.
MasterClass doesn't include a total length, and it was tedious to add up the exact lengths, so I just added up the rounded lengths available on the session page. The total that I could count came to 5 hours 50 minutes.
The MasterClass includes three live exercises, that I do not discuss here (since they are more for demonstration purposes of a range of techniques). I also do not discuss the historical hostage negotiation examples that Chris Voss discusses (a bank robbery, the Jill Carroll kidnapping case, and the Dwight Watson (tobacco farmer) case). I also skip some other material, including discussion of bargaining (using the Ackerman model), that is outside of the main negotiation techniques.
The class guide is 14 pages long, including a cover page (this blog post overlaps quite a bit with the class guide).
By and large, I do not discuss the session separately; I mostly only mention ideas in the session that are not in the regular MasterClass. Also, I consumed the session and added information about it well after writing the original post, and have relatively limited time.
The session includes a lot of demonstrations involving Voss's colleagues at The Black Swan Group (Brandon Voss, Derek Gaunt, and Sandy Hein). I don't discuss any of the demonstrations individually (though I found them very useful to watch!) and I mainly cover only the insights from the.
You can sign up and cancel within 30 days to get a full refund (however, you should check if this option is available when you sign up, as MasterClass may change its policies on this front). This gives you enough time to consume the Chris Voss MasterClass.
You may be able to sign up for a free week; however, I'm not sure what subset of the videos in the MasterClass are available. More details here.
Many of the techniques discussed in the MasterClass are discussed online, including in videos from Chris Voss and others. I have more links for each of the individual techniques, so you can use my post to basically get a "free" version of the MasterClass. You can also Google around for more videos and written materials on the techniques.
Voss has also co-authored a book Never Split The Difference that covers these techniques. It's available on Amazon; at $11.99 for the Kindle edition it is considerably cheaper than the MasterClass. There is a Reddit thread comparing the book and the MasterClass and consuming both.
Masterwiki.how can be a helpful free version of MasterClass, but it does not seem to have Chris Voss's MasterClass.
My blog post didn't exist at the time I bought the MasterClass; I think in the absence of such an alternative, the MasterClass was worth paying something for. In fact, I would say that the exposure to Chris Voss's ideas was worth the price of two full years of MasterClass subscription for me, though I could have cancelled after 30 days once I had consumed the Chris Voss MasterClass.
Now that this blog post exists, and given the great amount of online material, the value-add of the MasterClass is less clear. I think there is still a case for it, but it probably isn't worth spending a full year's subscription of $180 just for this class, nor is it worth spending $90 to buy it. With that said, if you find several other MasterClasses that you consider worth watching, the Chris Voss MasterClass could tip the scales.
Given the ability to cancel after 30 days, it seems worth trying it out if you think there's a chance it will be worthwhile.
Since rationalists have often spent $1,000+ on CFAR workshops and found it worthwhile, I think there's a good chance that many will find the class worth paying for, even after having access to the material in the blog post.
I found Daniel Pink's sales and persuasion MasterClass to be similar and relevant. There are a lot of similarities and some differences between Pink's and Voss's advice on the topics where they overlap. Here is a good review of Pink's MasterClass as well as some discussion of the similarities and differences with Voss's.
In general I have found MasterClass to be worth the subscription cost for at least two years, but much of it might have to do with the specific topics I am interested in. You can check here what MasterClasses I have watched.
The MasterClass is mostly just Chris Voss speaking, with nobody challenging him. The session does a better job of showcasing interaction among multiple people, but they're still all people at The Black Swan Group, so it doesn't necessarily show as much as possible how their ideas could be challenged or questioned by outsiders.
I like to see people questioned about their ideas, ideally with question that I have or would have if I thought more about it. Since watching the original MasterClass, I've watched several interviews of Chris Voss; I list below the ones I found most interesting:
Chris Voss's MasterClass was the main reason I signed up for MasterClass, and I finished it within a day after signing up. I liked it quite a bit then. Over time, and after having thought about its ideas and watched other related material, I continue to think the ideas are pretty good.
I do not make claims about how much of the material originated from Voss versus was learned by him from others (including the FBI manual). Regardless of Voss's role as innovator versus peddler, the stuff covered is good.
Of the ideas I learned from the MasterClass, I've already shared several with friends. However, given that the MasterClass is paywalled, I think it would be helpful to have a public write-up of the ideas that I can link to. A public write-up would also benefit people outside my inner circle of friends.
Rationality offers a powerful way of viewing the world, but there are also large bodies of knowledge around human interaction developed by others. Fruitful exploration of these bodies of knowledge can help enhance and deepen our rationality.
Negotiation tactics are mostly focused outward: they're focused on how to deal with the other side (known as the "counterpart"), not how to deal with yourself.
A large part of the focus of rationality is inward: how to reason better, how to tame and use your emotions, how to achieve goals. Similarly, many techniques such as meditation and relaxation techniques have an inward focus too.
With negotiation tactics, one's own goals obviously matter, but the goals of the other person are front-and-center.
There are a couple of ways that negotiation tactics relate to inward-focused activities. First, in principle it is possible to apply some negotiation tactics in self-negotiation. Interestingly, some such applications have crude similarities to self-talk techniques (such as CBT or EFT).
Second, in order to negotiate effectively with others, it generally helps to have taken care of yourself before that. The idea is that in a negotiation, the counterpart and the situation take center stage. If you're bringing your own baggage to the situation, though, it becomes harder to apply best principles of keeping the focus on the situation and making the counterpart feel heard and addressed.
The negotiation MasterClass does not discuss the first point; it does bring up some ideas related to the second point but indirectly.
The negotiation techniques in the MasterClass grew out of efforts to deal with hostage kidnappers, and evolved to address difficult business negotiations. These people aren't necessarily irrational, but they are not aspiring to epistemic virtues. Pointing out their cognitive biases will not make them thank you for helping them achieve their epistemic goals.
The negotiation techniques therefore can be quite different from norms recommended for rationalist cultures, where there is an assumption that all parties are aspiring toward epistemic virtue. For instance, tell culture can be great as a rationalist norm but a naive application of tell culture principles would contradict many aspects of negotiation techniques (though reconciliation is possible).
One concern we might have about negotiation techniques is that they specifically rely on the irrationality of one's counterparts, and so would fall flat or even backfire in more epistemically virtuous environments. After thinking about the various techniques, I think this actually isn't much of a concern, as long as the person applying the negotiation techniques adjusts the mix effectively based on what is needed. Broadly, I do not think these negotiation techniques are what is sometimes called the dark arts on LessWrong (though they probably could be interpreted as such!). Later in this post, as I discuss each negotiation technique, I discuss what aspects or variants of it I consider "dark arts"y.
While Voss gives a lot of importance to emotions, I think it would be wrong to think of the negotiation techniques as primarily emotion-focused. Rather, I think the negotiation techniques try to pick at underlying motivations, some of which could be emotions! I think there is a moderately sound cognitive and epistemic grounding for the negotiation techniques, though the role of human emotions is very important to understanding how successful the techniques will be.
Many of the negotiation techniques are applicable to synchronous interactions, including in-person and phone interactions. In general, the techniques seem optimized for high-bandwidth communication with frequent back-and-forth. I found some techniques as well as some principles to have broader applicability, including to more asynchronous and one-to-may communication contexts.
In the comments, Adam Zerner writes:
It seems worth mentioning that leverage is hugely important. Both 1) having it, and 2) understanding it. For example, suppose you are a programmer applying to companies. 1) It's helpful to be good at interviews and have a lot of companies interested in you. 2) It's helpful to be aware of this fact, and to be aware of what sort of leverage the companies have. Ie. BATNA.
Maybe you can call what I am referring to as hard skills and what you are referring to as soft skills? I feel like that isn't a great way to categorize is, but nothing better is coming to me. Whatever the categories are, I think it would be good to explicitly mention that this article is targeting one of them, and that there are other things that are important for the bigger picture of being able to negotiate well.
I also got a related comment on my Facebook share of the post:
Also if I may seek more details about the Masterclass , does it also delve into concepts taught in management schools (albeit with a different name) like ZOPA(Zone of Possible Agreement) & BATNA ( Best Alternative to Negotiated Agreement )
These are great points. There is some discussion in Voss's MasterClass about understanding who has leverage. He also goes into a few nuts-and-bolts bargaining tools like the Ackerman method and provides general advice on price negotiation. He also has a mock job negotiation.
Nonetheless, for the most part, the aspect of negotiations that Voss covers has much more to do with the low-level execution of how to share and receive information, as opposed to the what of communication. And to the extent that Voss covers other angles, my review skips over them.
Obviously, the what of negotiation matters a lot, and there's great advice online around it -- this post does not compete with such advice. In fact, when a lot of people think of negotiation, they're focused on the what aspect -- what price should I settle for?
There are a few reasons I think Voss doesn't focus on these much in his MasterClass, and my review here doesn't talk about it at all:
The what is highly domain-specific, and even within a domain requires a fair amount of market research and even getting into the nitty-gritties after your counterpart gives you situation-specific information. A 3-hour MasterClass intended to appeal to a wide audience can't really get into the what too much. With that said, in the MasterClass and in several of his interviews that I've linked to, Voss talks about some of the nuts and bolts of the what question in the context of real estate negotiation, business partnerships, and job interviews. He's often relying on information he learns back from people who come to the Black Swan Group for training and apply the ideas in real life.
Highly prescriptive approaches to the what question are in tension with the thrust of the approach that Voss is trying to push for -- namely to be open and curious and let the other side reveal more information to enable collaborative problem-solving. In some of his interviews Voss talks about how he thinks some of the techniques like BATNA are not that useful, but he has a lot of respect for Roger Fisher, who championed these techniques. Voss thinks Fisher's success comes not so much from the techniques as from the emotional intelligence he has when applying them. Voss thinks his own techniques come closer to what needs to be done execution-wise to achieve those sorts of results.
In a video in the MasterClass session, Voss talks a bit more about his main issue with BATNA, which is that there are many situations where your alternatives are actually pretty bad. This probably comes partly from his own background as a hostage negotiator, where the alternative to negotiated agreement can literally be fatal. So, while it's good to have alternatives, a good foundation of negotiation techniques is one that can work even in the situations where you don't have alternatives.
This section covers my own interpretation of key "principles" behind the individual negotiation techniques. The way I frame it doesn't always match how Voss presents ideas in the MasterClass or elsewhere, and may not be endorsed by Voss.
A lot of the negotiation techniques boil down to a show, don't tell approach of demonstrating good faith. Many of these "show"s are constructive proof that would be hard to fake for somebody who is not listening and not interested in a good faith collaboration (examples include labeling and the accusations audit). Many of them also directly create value by making substantive progress in zeroing in on the issues involved.
A key point: the "show, don't tell" applies at the meta level of your sincerity and competence; obviously there will be cases where you have to tell the other person factual information or ideas. This is only loosely related to pedagogy's "show, don't tell", so actions that are "show"s in terms of demonstrating sincerity could be "show"s or "tell"s at the object level.
I think this is good advice in general. I do think there are cases where "telling" works, but telling is much more likely to backfire than showing. So I'd say one must always show, but whether to tell depends on how much trust has already been built.
An important aspect of the negotiation techniques is to center them on your counterpart, i.e., the person you are negotiating with. Things like, be curious about what they have to say, be tactically empathetic, always show that you care about the impact of your words and actions on them. Let them go first. By and large, avoid first-person pronouns, so don't say "What I'm hearing is ..." or "I want ..." or "I need ..." (this is not an absolute injunction, but good to start with). Don't make the other person feel you are putting them on the spot.
While I think this makes general sense, there are important considerations of asymmetry here that deserve a more detailed treatment. I cover the asymmetric nature of counterpart focus in a later section.
Often, in high-stress situations, people are thinking impulsively, defensively, and carelessly. The many negotiation techniques Voss teaches are designed partly to get your counterpart to slow down, relax, and think more reflectively (Kahneman's "system two" thinking). This frame of mind is more conducive to solving challenging problems collaboratively. An additional side-effect of these techniques is also to slow down your own thinking and make you a calmer, better thinker.
I'm generally in favor of slowing down and thinking deeply and reflectively. This doesn't translate to being slow in absolute terms -- speeds could vary a lot based on context and familiarity. But I do think it's important to avoid "rushing" things and to combat the tendency for stressed, anxious thought patterns.
A general theme in negotiation is how both parties perceive it, and whether, at the end of it, they feel like it was a worthwhile endeavor. Starting with the difficult portions and gradually making progress to end on a positive note is important. Voss says that "the last impression is a lasting impression" and emphasizes that much of the positive messaging we are inclined to use to open an interaction may be better suited to the close.
I definitely agree with this. One of the things that makes interactions stressful is when people keep dropping bombshells throughout the conversation. This keeps the other side wary throughout the conversation. Getting the tough parts in the open quickly makes the rest of the interaction more relaxed.
I also agree with the importance of ending on a positive, collaborative note.
Voss says that people often fight to be in control or in charge, e.g., to be the ones speaking in the room or having more overt control of the situation. By relinquishing one's own desire for control, and letting the other person take charge -- while collaborating with them in the process, you can acquire the upper hand by getting more information. Several of the techniques discussed later, such as mirroring, labeling, mislabeling, dynamic silence, and calibrated questions help the other side feel more in control while also giving you the upper hand by learning more.
While I do think being in control is overrated, and it's often more important to learn more than to be in charge, some aspects of this framing didn't resonate much with me. The "upper hand" framing is a little bit in tension with the whole idea of negotiation as being helpful to both sides.
As I discuss in a later section, it can be counterproductive if people start competing to not go first -- just as it can be counterproductive if people are competing to go first. So my main takeaway from this point is that if you have a tendency to want to go first and dominate a situation, rethink that. But don't be too singularly focused on not going first in all situations.
Mirroring is the technique of repeating about 1 to 3 words of the last sentence the other person said.
Mirroring in general is helpful as it is a low-effort way of showing the other side that you're listening and engaged with what they say.
Mirroring with upward inflection (i.e., a kind of questioning tone) is helpful as a prompt to get the other person to continue expanding and elaborating. This can be helpful if you don't quite understand what the other person said, or you want them to elaborate more.
Instead of the MasterClass, you can watch a free Chris Voss YouTube video on mirroring.
I have not tried to use mirroring much in my life. My impression has been that mirroring is most useful as a low-effort way to engage another person and learn more about what's on their mind. It may be a bit less useful in cases where other, more high-effort and high-reward techniques, can be used.
I used mirroring once in a low-stakes situation with success; a colleague and I were on a call with a third party who ended up not showing up, so we were just waiting for about ten minutes. My colleague was just chatting about stuff going on with his work. I had no particular agenda in terms of information I wanted to know, but I also didn't mind hearing him, so I decided to try using mirroring to help show I was engaged without putting in a lot of effort. This seemed to work well; at the end of it, my colleague said it was great chatting, despite me basically saying nothing.
Mirroring as information-gathering doesn't seem dark arts-y at all to me.
Mirroring as a way of showing you're listening can be dark arts-y if you're not actually listening. Somebody practiced enough at mirroring could probably do it automatically without paying close attention to what the counterpart is saying. The counterpart then thinks you were engaged and listening (because they hear your mirrors and don't even realize you were mirroring) but you actually weren't. To be clear, this is not the sort of mirroring that Voss encourages; he emphasizes genuine curiosity and interest.
I think mirroring has a place in rational discourse, but a relatively small one. I'm much more excited about the other techniques discussed (that are both high-effort ad high-reward), including labeling.
Labeling is the act of providing a short summary of the underlying emotions, thoughts, and ideas behind what your counterpart is saying or doing.
A tactical aspect of labeling: Voss recommends using "It seems/sounds/feels/looks like ..." or "You seem/sound/look like ..." before articulating the label (in a single sentence where possible), and being deferential in tone. It is also open to correction (see the next section, mislabeling). The use of first-person pronoun framings such as "I think ..." or "What I'm hearing is ..." is discouraged because you want it to be about your counterpart and the situation, not about yourself.
Labeling is more high-effort than mirroring: rather than just using short-term memory to remember and pick words from the last sentence, you need to listen to the entirety of what the other person is saying (as well as tone of voice and nonverbal cues where applicable) and summarize it. But it's also more high-reward, because it shows a deeper understanding, helps provide clarity to both sides, and can lead to real progress.
There's a video going over mirroring and labeling.
Voss claims that labeling negatives "always" diffuses them. The key point is to label a negative by accepting it, rather than by denying or contesting it.
For instance, instead of saying "I don't want to sound too demanding" you say "This is going to sound demanding". Or instead of saying "Don't get upset" say "It seems like you are upset".
Instead of the MasterClass, you can check out this video on labeling negatives.
Voss claims that labeling positives tends to reinforce them (the opposite of the impact of labeling negatives). For instance, "you sound really excited about this" or "it sounds like you're very happy with the way this turned out".
I've generally found labeling to be more useful than mirroring, particularly when a lot of raw information is being conveyed and it needs to be processed.
Labeling generally seems like a positive thing and not like a dark art. While labels could be misused, I don't see the potential for misuse as big enough to make labeling a dark art.
Absolutely! I think labeling is pretty valuable. A lot of rational discourse already follows similar practice; for instance, providing labels/summaries of what another person said is viewed positively in the community.
The rationalist community also places more importance on self-labeling, something that negotiation strategy frowns upon (it's not about you, it's about your counterpart).
Mislabeling refers to the (usually intentional, or at least probabilistically intentional) application of an incorrect or exaggerated label to give the counterpart the opportunity to correct you and reveal more information. So instead of directly asking for the reasons for something, a mislabel might attribute a reason that's probably wrong, and let the other person correct it.
For instance, if a person declines to do an activity, you may mislabel it as "it sounds like you really hate doing this" which gives them the opportunity to correct by saying "actually the time doesn't work, I would love to do it next time".
Brandon Voss (Chris Voss's son and business partner) has a video on mislabeling here.
There is a "dark arts" form of mislabeling, where your goal is not to get information from the other person, but rather to make them claim something that's beneficial to you. That happens when your mislabel is done with a desire to have the other person reassure you. Mislabeling, when done with curiosity, deference, and a genuine openness to being corrected, and with the goal of getting information rather than manipulate the other person into claiming something, seems good to me.
This is a little unclear. I think mislabeling is okay in rational discourse if the mislabel is still the leading individual candidate. For instance, if you have several hypotheses to explain something, and the leading hypothesis has 30% chance, higher than any other, formulating that hypothesis as a label seems reasonable (and it's probably a mislabel because there's a 70% chance it's wrong). On the other hand, deliberately choosing a low-probability hypothesis as a mislabel seems not very rational.
Dynamic silence (also called effective pause) is the idea of just staying silent, usually either right after you mirror, label, or mislabel, or when things fall silent in general. The idea is to give the other person space to keep going on and to correct you. According to Voss, dynamic silence works best after you have established, through mirroring and labeling, that you are listening, engaged, and understanding.
There's a video discussing dynamic silence here.
I have found dynamic silence to be generally useful when the other side is reasonably articulate. I haven't really needed to consciously practice it; it seems to come naturally.
Calibrated questions (also called open-ended questions) are what/how questions designed to both elicit information from the counterpart and start the collaborative process with them by introducing to them the considerations and challenges on your side (Voss calls this "forced empathy"). According to Voss, the main purpose of calibrated questions isn't necessarily to get clear answers, but to get the other side to slow down and think. Two of his favorite questions are:
There's a video of Chris Voss explaining these questions here.
Voss is generally against the use of "why" questions because he claims that, universally across languages and cultures, it gets people defensive. When people hear "why" it sounds to them like you think there's something wrong, and they want to defend themselves against accusations. That's because about half the time people use "why" they are being accusatory, and the other half, they genuinely want to know -- and a person on the receiving end can have their defensiveness triggered even when it shouldn't because they have uncertainty about the asker's state of mind.
Another related idea that Voss goes over in the MasterClass is to ask "legitimate" questions (and/or raise legitimate concerns) -- basically, the calibrated questions that, if the other side were to hear, they would have to agree that this is a valid point. For instance, rather than asking "How will you guarantee me that X?" ask the underlying question which might be "How do we deal with the fact that I am operating with uncertainty regarding X?" Voss gives the example of the proof-of-life question that hostage negotiators must ask the kidnapper; he suggests a "How do I know the victim is alive?" rather than demanding a proof of life.
I have generally found the calibrated questions idea to work well, though I've generally used it more for information-gathering than for forced empathy. Avoiding "why" in particular seems to make a lot of intuitive sense and I've generally found it to be effective at reducing both the defensiveness of the other side and the perceived antagonism in the interaction to third parties.
I don't see any clear mechanism by which calibrated questions are dark arts-y.
Yes! In fact, I think the idea of using what/how questions instead of why questions has a purely rational basis, in addition to the reasons based on the emotional triggers that "why" could produce. Namely, a "what"/"how" question tends to be more constructive or specific. A "why" question is much more vague, like an injunction to "explain yourself".
For instance, if you ask "Why did you do X?" it's a very difficult question to address. Instead, a question like "What was your motivation for X?" has an appropriate level of focus on motivation. On the other hand, a question like "What events led you to do X?" has a focus on history. By picking a specific aspect, it avoids the very vague "explain yourself" character of the "why" question.
The "accusations audit" is a comprehensive list of the negative assumptions, thoughts, and feelings that the other side has about you -- both things that they already might harbor and things that they are at risk of harboring once you start revealing the information you're planning to reveal. The accusations audit is then communicated to the other side to show that you're aware of the issues and also to get out in front of them so that both sides can be more relaxed and make progress.
The accusations audit is a proactive version of labeling negatives, where instead of waiting for negatives to pop up before labeling them, you get out in front of them. Some of the remarks made previously about Voss's claim that labeling negatives diffuses them also apply here.
As with labeling negatives in general, it's important in the accusations audit to not try to preemptively defend against or contest the negatives. The first step is to acknowledge the negative, and let the other side acknowledge your acknowledgement. In many cases, the act of acknowledgement itself diffuses the negative enough; in other cases, there's enough time after the acknowledgement to address the substantive issues raised by the negative.
Chris Voss explains the accusations audit here.
Brandon Voss has a practical video on the accusations audit here.
The accusations audit can be tricky without a good mental model of the other person. It can also be tricky to do if you have a lot of self-esteem issues of your own. I think a key ingredient to the success of the accusations audit is the ability to genuinely think from the other side's perspective, rather than project your own insecurities.
With that caveat, I've found the accusations audit useful, particularly for being able to start discussions that may otherwise not have happened (unlike reactive labeling, the accusations audit can be started asynchronously and used as input to try to get the other side engaged).
I don't think the accusations audit is "dark arts"y. With that said, a variant of it could be, where you trump up exaggerated accusations with the goal of manipulating the other person into consoling you. The key way to overcome that is that the "accusations" have to be the other person's accusations toward you, not your accusations toward them. There should be no judgment from your end conveyed as you describe the accusations.
I think so! I think bringing out negative assumptions, thoughts, and feelings early creates the scope to actually address them in the interaction.
The idea behind no-oriented questions is to frame yes/no questions in a way that the answer that will make your life easier is the "no" answer. I found this one of the most interesting and thought-provoking negotiation techniques.
For instance, instead of asking "Is it ok to publish this?" ask "Are there any concerns with publishing this?" Or, in a sales call context, instead of asking "Do you have a few minutes?" ask "Is now a bad time?"
The simplest reason, as given by Voss, is that "yes" can make us feel trapped, like we're being pressured or tricked into agreeing with something. On the other hand, "no" makes us feel freer and safer, like we have protected ourselves. I think there are a few other factors that make no-oriented questions valuable, that I explore in the next few paras.
Chris Voss has a video on no-oriented questions here.
Among all the negotiation techniques, I feel that the switch to no-oriented questions has been the one that's influenced my day-to-day actions the most. Part of this might be that this technique is highly applicable to low-bandwidth, asynchronous interaction, that forms a large proportion of my interaction.
In particular, the majority of my questions, particularly around getting consent/approval for joint actions, are now no-oriented. While I think in most cases it doesn't end up mattering, I feel like it does make things smoother when it does matter (i.e., when the other side's answer is the opposite of what's easier for me).
Other than the psychological benefits of "no" making us feel more protected, are there other benefits to no-oriented questions? I think so, and I think it would be good to generally push more toward no-oriented questions.
First, I think the act of formulating a no-oriented question helps us, as question-creators, think of the nature of the objection/challenge more clearly. The act of formulating the question itself therefore improves the likelihood of catching issues.
Second, formulating a no-oriented question makes the other side more comfortable responding in either way: with a yes (i.e., the answer that makes life harder) and with a no (i.e., that makes life easier). By using a no-oriented formulation, you're signaling to the other side that you are prepared for bad news (in the form of a "yes"). This also means that if they deliver bad news, it feels more collaborative -- both sides are actively trying to unearth the bad news and address it rather than hide from it -- and less of a fight.
Third, responding to a no-oriented question also triggers more thinking from the other side, partly because you've already shown that you are willing to hear bad news.
The "summary" is a technique not covered explicitly in the regular MasterClass, but covered more in the session. A summary is usually a paragraph or two that summarizes what the counterpart has been saying and/or what all has been discussed so far.
The summary is something that should be used once you've extracted relevant information using mirrors and labels. A summary can be somewhat similar to labels, in that individual sentences in the summary can be similar to labels, but it differs in that it puts everything together.
Summaries could also be used at the beginning to start with what's known already, but it's usually better to begin with an accusations audit if that applies, and to wait till there's more information before attempting a summary.
Voss says that getting the other side to say "you're right" isn't great, but getting the other side to say "that's right" is great. "That's right" is an acknowledgement by the other side that you've understood the situation.
While "that's right" is great to hear at any time (e.g., in response to a good label), the summary is the time in the conversation when you're most likely to get it, because a good summary that exactly speaks to the other side gets the most satisfied response from them.
I had not come across summaries as a negotiation technique until I watched the session in December 2021. I have done summaries in the past, and generally think they are useful.
I have a little caveat / nuance to add to the "you're right" versus "that's right" distinction. While "that's right" is almost always a good sign, "you're right" isn't always a bad sign; the tone of voice and the context matters. If somebody says "you're right" right after you go on a tirade, it's probably a bad sign and suggests they may want you to shut up. However, if the counterpart has a spontaneous epiphany and says "you're right about X!" that's good. So the context and tone of voice matter a lot in interpreting "you're right."
Paraphrasing, like mirroring, involves repeating the things your counterpart said back to them. However, paraphrasing isn't just the last few words, but goes over a larger portion of what they said, picking out key ideas. Paraphrasing doesn't need to use the exact words, but it should still retain the meanings.
Paraphrasing is similar to summaries but involves less synthesis and more just a literal reading and compression of what was said.
When your counterpart is unloading a lot of information, interrupting them to paraphrase it, and then handing the control back to them, might bee a good idea. While people don't like being interrupted, so they may feel a little bit or irritation in the moment, their fear that you interrupted them just to say your own thing will dissipate once you hand control back to them.
Roughly speaking:
Mirrors : Paraphrasing :: Labels : Summaries
I have not used paraphrasing much, but generally agree with some of the benefits.
This is a little related to the accusations audit, but a slightly different tack. Sometimes, in a negotiation, you need to introduce a tricky constraint from your side. If you expect that the first reaction of the other side to you introducing the constraint would be for them to think "that's your problem, not mine" then -- preface with "that's my problem" before introducing the constraint. That way, it becomes clear to the counterpart that while you are seeking their help to work within the constraint, you aren't allocating blame or responsibility for the constraint to your counterpart.
I like this idea. I hadn't specifically encountered it before the MasterClass session, and haven't had many opportunities to apply it, but I expect I will.
This is not so much a single technique as an underlying idea that influences other techniques; nonetheless it's close enough to a technique. The idea behind "tactical empathy" is to demonstrate -- through actions -- that you understand and respect the situation your counterpart is facing and the impact it's having on them (of course, if you don't yet understand, things like mirroring, labeling, and calibrated questions help get you there). Tactical empathy is different from but related to sympathy ("I feel how you feel"). The goal of tactical empathy is to get the other side to trust that you have a good enough understanding of their situation that you can collaborate with them to solve problems.
There is a lengthy video by Voss on tactical empathy here.
In the MasterClass and elsewhere, Voss uses the term "trust-based influence" to describe the kind of influence that you can build through the use of tactical empathy, supported by other methods discussed earlier (mirroring, labeling, calibrated questions). With trust-based influence, the other side understands that you understand their situation and respect them, and therefore trusts that you'll be able to collaborate with them to solve problems.
I have not actively applied tactical empathy as a technique, but I think my practice has moved a bit in that direction after being exposed to the concept.
I think tactical empathy can be "dark arts"y if you actually don't understand the counterpart's situation.
It's still likely less "dark arts"y than false sympathy or false agreement.
Tactical empathy seems closely related to proto-rationalist ideas such as Rogerian argument and the similar "ideological Turing test".
Voss identifies three kinds of tones of voice:
He also identifies two kinds of inflections:
Inquisitive inflections are good when you want to get the other side to talk more.
Since a lot of my communication is text-based rather than voice-based, these concepts have had limited utility to me. One general idea that I've taken from this is to speak more slowly (one of the aspects of the late-night FM DJ voice). I might also have reduced my use of assertive voice and increased my use of playful/accommodating voice as a result of being influenced by these ideas.
Some of the principles also carry over from tone of voice to tone of text communication. Even prior to this MasterClass, I adopted a upbeat style of communication, sprinkling exclamation points and smileys in future drafts. I've continued with this practice.
I don't see anything "dark arts"y about using using a playful/accommodating tone of voice, or the late night FM DJ voice.
Tone of voice (or tone of text) provides another dimension of communication that influences discourse. Using it in a good way seems consistent with the idea of rational discourse. Since a lot of rational discourse is centered around open exchange of ideas, the playful/accommodating tone of voice seems suited to it.
Voss talks about loss aversion -- he calls it "fear of loss" -- and says that this distorts people's thinking a lot. So what do you do if the idea on the table isn't about a loss? His suggestion seems to boil dow to framing it in terms of a loss by using the opportunity cost framing: basically point to the other side what thy lose by not doing the deal.
Voss talks about loss aversion in this video where he cites the academic field of prospect theory. He says that reframing a forgone gain as a loss can be so powerful that the term bending reality can be used for it.
To the extent that this is about combating and neutralizing an existing loss aversion bias, I think it's not a dark art. But to the extent it's about invoking loss aversion and creating distortionary fear, I think it is a dark art. A lot depends on the implementation.
I do not remember any situations where I consciously applied this technique; also, I was already broadly aware of the ideas of loss aversion and opportunity cost so the marginal impact of the MasterClass was low for me.
The term "black swan" refers to a hidden piece of information that, once revealed, changes the shape of the negotiation. Part of the goal of negotiation is the (collaborative) discovery of these black swans. This can be thought of as one reason it's so important to get the other side to talk and reveal more private information (combining this private information with your private information can help unearth the black swans).
I have not had any major success unearthing black swans! But I still think it's a valuable idea.
I don't see the idea of trying to discover black swans as "dark arts"y.
Absolutely! I think a lot of rational discourse is about discovering new ideas, and some of the more novel ones could qualify as black swans.
The related ideas of fairness and reciprocity come up a lot in negotiation. Awareness of these can help.
Voss talks of a few ideas related to these:
Avoid triggering reciprocity by e.g., making asks/demands, when there are other alternative ways: For instance, the use of legitimate calibrated questions can engage your counterpart to collaboratively solve the problem with you, without triggering the sense that they are doing you a favor.
Rather than say things like "I only want what's fair" (that can be read as an accusation of unfairness) reassure the other side that if at any point they feel that they aren't being treated fairly, they should speak up.
Offer things to the other side (this could include goodies or information) that aren't costly for you, but that are either directly valuable to the other side, or at least signal that you are there to help them and/or that they are squeezing out good value from you. An example is mentioned in the final stage of Ackerman bargaining: once you are at the limit of the budget you are willing to pay, offer some non-monetary good that is cheap for you -- and may even be something not valuable to the other side (if you can't think of anything valuable to them) -- but that shows the other side that you are stretched to the limit with the money side and they've gotten a good deal.
In the MasterClass session, Voss and colleagues talk about three negotiating styles/personalities; you can see more details here:
Assertive style/personality: The assertive style is to assert, to confront, and to make sure one's point of view is heard.
Analyst style/personality: Analysts care more about getting data and information, and tend to be reticent to say much, especially early on.
Accommodator style/personality: Accommodators want and try to be liked; they value harmony and feel the need to diffuse any tension or awkwardness.
The personalities correspond to the tones of voice previously discussed; assertive style corresponds to assertive tone of voice, analyst style corresponds to the "late night FM DJ" voice, and accommodator style corresponds to the playful/accommodating tone of voice. However, the underlying style and tone of voice don't need to match. For instance, the assertive tone of voice should probably never be used, but the assertive style/personality can still be expressed through other tones of voice and can be important.
The general advice for different personalities is to focus on incorporating more of the techniques that don't naturally jive with that personality. Here are some of the ideas if you're one of these personalities:
Assertive style/personality: Assertives need to make sure they are not using an assertive tone of voice. Also, they may need to be more conscious of shutting up and listening so as to hear what the other side has to say.
Analyst style/personality: Analysts need to make sure that they are sufficiently warm and friendly, as the analyst personality can appear too reticent at times. Active application of negotiation techniques such as mirroring and labeling can help with this.
Accommodator style/personality: Accommodators tend to generally be good listeners, but tend to have more trouble with using tools like dynamic silence to get the other side to talk. They may also end up talking or revealing too much, or overpromising.
There is also general advice for dealing with the different sorts of personalities in your counterpart:
Assertive style/personality: When dealing with assertives, it's particularly important to let them have their say first. This is important even in general, but less so for analysts and accommodators, who don't generally have that much of an urge to have their say first.
Analyst style/personality: When dealing with analysts, it's helpful to draw them out more. Instead of asking questions, using labels and mislabels can help draw out the analyst due to their desire for correctness.
Accommodator style/personality: When dealing with accommodators, it's important to get them to feel more comfortable to voice their own concerns and get out of the mode of trying to accommodate. Also, accommodators may tend to overpromise when they feel pressure to please, so it's important to be careful about that.
One general idea presented in the session, and demonstrated in various demonstrations, is that the path to progress isn't always monotonic -- there can be some ups and downs along the way. One of the mechanisms is that as people get more comfortable in the negotiation, they reveal more of their challenges and negative feelings, which can sometimes cause them to heat up. Also, sometimes the application of labels and mirrors may be off; for instance, a mislabel can cause the counterpart to get more angry sometimes.
However, even if the emotional journey isn't always one of progress, the informational journey usually is -- even with the rocky emotional moments, new information is emerging that can ultimately be used to make more progress.
This section goes into detail on general concerns that I've had or that have been raised to me about negotiation techniques. This includes concerns raised from rationalist perspectives.
While there is a lot of homage paid to the idea of negotiation not being a zero-sum game, the fact is that most of the negotiation techniques are applied asymmetrically between you and your counterpart. For instance, "let them go first" -- what happens if both you and your counterpart are trained in negotiation techniques and trying them on one another?
Another way of thinking of it game-theoretically: are negotiation techniques like defecting in a prisoner's dilemma, achieving gains at the other person's expense, but if both people did it, both sides would lose out?
In one of his interviews, Voss addresses this in some depth. He says that he actually does't mind, and even prefers, if people use his negotiation techniques on him. He says, for instance, that most of his co-workers ask no-oriented questions all the time. He's used to it and doesn't consider it manipulative.
What if both sides were using negotiation techniques on each other? In particular, what if both sides are trying to let the other side go first? Could this result in an impasse? Some more advanced treatments of negotiation techniques (including some parts of the MasterClass) discuss this. Generally, if the other side is really passionate about wanting you to go first, it's a good opportunity to learn what concerns they have about going first, for which you can use mirroring, labeling, and calibrated questions. Ultimately, if there's something that's a deal breaker for them, it's good to know that in a way that doesn't involve confrontation and anger, and that's what the negotiation techniques are for.
So, whereas the fundamental criticism that negotiation techniques are asymmetric is true, they can be adapted easily to a symmetric world, and that symmetric world is likely even better for both parties than if only one side is applying negotiation techniques.
The bad, entitled way to use your knowledge of negotiation techniques is to start expecting that people around you will start using the nice parts of them on you. For instance, maybe you realize how great it is to hear calibrated questions instead of "why" questions, or how much better no-oriented questions sound (on the receiving end) than yes-oriented questions. An entitled application of this enhanced knowledge would be to start suggesting/demanding that the people around you start using these techniques with you so that your lived experience is nicer. However, you rarely have the level of control over other people to do this in a big enough way, and it's entitled to expect that they do. I do think it's worth sharing these ideas with friends so that they can be more effective, just not primarily for the purpose of them providing a better experience to you!
A better way to apply your knowledge of these is to come to situations with more awareness of your own subconscious triggers. When somebody asks a why question, and you feel defensive or irritated, notice that, and think about how much of this is the "why" framing of the question. In some cases, the other person's use of "why" might reflect genuine irritation and hostility on their part. In other cases, though, it may be an innocuous word choice. One thing I have found useful is to notice my slight defensiveness at being asked why questions, then pause, and then answer them instead as if I had been asked a corresponding what/how question. In almost all cases, this works really well. In rare cases where I detect extreme hostility in the why question, or extreme lack of clarity in it, some other methods such as mirroring, labeling, or asking calibrated questions back can help.
The same goes with the use of yes-oriented questions: I now tend to notice my sense of feeling pressured when somebody asks me a yes-oriented question that I do not fully want to say yes to. In such cases, I sidestep the use of a binary response and answer in the same sort of way as if I'd been asked a no-oriented question.
In addition to helping me respond better to cases where others ask questions or make remarks to which my initial response is negative, this awareness also gives me a better lens when viewing interactions (written or oral) as a third party. When an interaction that starts off cordially becomes openly antagonistic, or when an interaction seems to have undertones of hostility, I can often locate things like why questions and yes-oriented questions in there.
The outward focus of negotiation techniques counters a lot of advice on the importance of expressing yourself. If everybody were busy negotiating, would people avoid expressing themselves?
I think this is an important criticism and it's important to remember that in a context where the other side is interested in what you have to say, expressing yourself is good (because it's actually meeting a need of both sides, i.e., it is a coincidence of wants). Negotiation techniques are more the scaffolding you put around self-expression, they aren't about the self-expression itself but they help support it by helping create a safe environment for that expression.
For instance, the idea of letting the other side go first and extracting information from them is important because, generally, only once people have had their say do they feel relaxed enough to hear you.
Similarly, the accusations audit helps diffuse the other person's negative valence around you, putting them in a position to actually listen to you.
Scott Alexander introduced the concept of asymmetric weapons in two blog posts (here and here). The first blog post highlighted reason as an asymmetric weapon that generally helps push toward epistemic progress. The second blog post pointed out several ways that the asymmetric weapon of reason could actually make things worse. A question about negotiation principles and techniques is whether they are symmetric, or asymmetric in a predictable way.
I think negotiation principles and techniques are slightly asymmetric in a positive direction. Overall, if more people adopted these successfully, interactions would be calmer and more collaborative, and this would lead to better outcomes for all.
In many cases, your counterpart is an agent of an organization or another entity (the principal). For instance, you might be negotiating with an employee of a company you are doing business with. The employee is the agent. The business is the principal. In any situation where the agent and principal differ, there is potential for a principal-agent problem.
The bulk of negotiation techniques are focused on the agent, and as such, they may exploit the principal-agent problem to get good deals for yourself. Many of the examples provided by Chris Voss, including free hotel room upgrades, free flight ticket upgrades, and special coupons at stores, seem to be open to this criticism.
I do think this is a valid but ultimately minor criticism. In most of these cases, the principals have already made a macro decision as to how much discretion to grant agents, and your actions are operating within that discretionary framework. For instance, if the hotel front desk staff has the ability to offer you a late checkout, that's because the hotel management decided to grant them this flexibility.
That said, there could be exceptions, and I think as an individual you can decide/choose not to apply negotiation techniques to situations where you feel it's exploiting a principal-agent problem in a particularly bad way.
The ultimate criticism is that this all sounds cool, but how consequential is it to the challenges facing civilization? Surely we can't negotiate our way out of AI risk!
I don't have a solid answer, but here is a tentative reason to think this is important. First, improving the quality of cooperation in general -- both in terms of the objective results produced and the positive vibes around it that make it more sustainable -- seems extremely important for tackling difficult challenges. People like Brian Tomasik have written about the broad theme.
Negotiation techniques done right seem like a good way to improve collaboration, coordination, and cooperation, at least at the micro-level. The handwavy part is getting from that to macro-level improvements in cooperation, in ways that meaningfully improve the world. I don't have a lot of confidence in how strong that connection is. But I think it might be enough to at least give negotiation techniques a try. The magnitude is uncertain but I think the effect is likely positive.
Overall I am glad to have been exposed to the negotiation techniques and ideas popularized by Chris Voss. I think many of them could be valuable to readers on LessWrong. Thank you for reading all the way till here, and please don't hesitate to share your thoughts in the comments here!