I think that treating needs and preferences as two different kinds of human desire facilites more understanding and compassionate decision making. In this post, I’ll explain why.

These ideas are quite sketchy and fresh, so I’d really appreciate feedback on what parts you found wrong or not useful :)

Two situations

Jimmy and Rachael are housemates. One evening, Jimmy asks Rachael if she wants to go see a film. Rachael is busy at work, and would strongly prefer to stay until 8pm and skip the movie. But she thinks that Jimmy is probably feeling lonely, and meeting his need for companionship is more important than her own preferences, so she leaves work for the movie. But in fact, Jimmy didn’t have a strong preference to see the movie, and would have also been happy to just spend an hour with Rachael when she arrived home after 8pm.

Jimmy and Rachael are co-organising a party. Jimmy’s in charge of music, and his preference is for mostly EDM. Rachael is feeling apprehensive about this, because in the past, she’s often felt outsided and unsafe at parties with electronic music. Still, she decides to tell Jimmy that whilst she might prefer another genre, she doesn’t mind it being mostly EDM.

What connects these two situations?

I think that both are cases of needs and preferences being confused in decision making:

• In the first case, Jimmy had a need for connection and companionship. Rachael saw this, but confused it with a need for watching the movie (which was only a weak preference). So in their preference aggregation, her strong preference to do more work was unfortunately overridden.

• In the second case, Rachael had a need for community and safety at their party, which meant she didn’t want EDM (or at least needed to talk about her feelings around that). Jimmy couldn’t see this, instead thinking she only had a weak preference against EDM. He had a strong preference for EDM, so in their preference aggregation, her need for community was not accounted for.

How could they do better?

I’d suggest that a better framework for Jimmy and Rachael’s decision making is:

1. Firstly, both parties express their needs (and associated feelings), and strive to find ways to meet both their needs (without compromise). I claim that it’s almost always possible to meet the needs of all parties.

2. Then, if some aspects of the decision are still undetermined, both parties:

   a. Figure out their own preference, assuming the other party has the same preference.

   b. Express these “first-order” preferences (without updating them as they hear what the other party wants).

   c. Once both parties understand and have considered each others’ first-order preferences, update their own preferences and act accordingly. (Thanks to Charlie for making this precise in the Decision Making section of this post).

What does this look like in practice?

In first case above:

1. Jimmy expresses his need for connection and companionship. Rachael may also have needs in that moment. They figure out that their needs can be met by spending at least an hour together that evening, in any capacity.

2. This leaves many aspects of the decision undetermined (e.g. what time, where, for what activity will they meet?). Jimmy says his first-order preference is to meet somewhere warm, do something relaxing, but other than that he doesn’t mind. Rachael says her first-order preference is to meet at home after 8pm. They update on each others’ preferences and agree to meet at home for tea and a film.

In the second case:

1. Rachael expresses her need for community and safety. She says she suspects electronic music at their party will make this hard to meet this need. Jimmy reassures her that he really cares about her feeling safe at the party, and so if it’s necessary, there won’t be EDM.

2. Jimmy says his first-order preference is for mostly EDM, because he wants to dance with his friends and that’s his preferred music for dancing. Rachael says her first-order preference is for dance, funk and folk, but that she’ll probably leave the party before 11. Conditioned on Rachael’s need and first-order preference, Jimmy says his preference is now to have funk and dance music before 11 (he also likes these genres), and only play EDM once Rachael has gone home. Rachael says that this would meet her need, and she’s happy with this plan.

I think this decision making procedure, which treats needs and preferences as different kinds of desire, facilitates more connection and compassion than treating them as “on the same level” (as in the first two paragraphs).

But isn’t the line between needs and preferences blurry?

Or: aren’t needs just very strong preferences?

Sure, it’s possible to model them that way. But I have found that viewing them as different categories of desire maps onto the human condition better, and helps me to relate more compassionately to myself and others.

The examples with Jimmy and Rachael capture some of my intuition about why this is.

So, how can needs and preferences be distinguished?

I’m a fan of Marshall Rosenberg’s work on nonviolent communication, and I think he’s mostly right when he says that all humans have seven basic needs:

• connection (acceptance, intimacy, mutual respect, safety, trust, to see and be seen)

• physical well-being

• honesty (authenticity, integrity)

• play (joy, humour)

• peace (headspace, harmony)

• autonomy (freedom, independence, “space”)

• meaning (contribution, growth, stimulation, self-expression, to matter).

All other desires are preferences. Some preferences are ways of meeting needs; others are not. For example, my preference for having a heater in my room is a way of meeting my need for physical well-being (but there are other ways of meeting that need). In contrast, my preferences to watch a thriller movie is a preference that isn’t meeting any need.

I’ve summarised some other feature which help to distinguish preferences from needs in this table:

Needs	Preferences
Universal (we all have basically the same needs)	Particular (we all have different preferences)
Can probably be met by many states of the world (since needs are for general things)	Can be met by a smaller number of states of the world (preferences are for particular things)
We shouldn’t update our needs when we learn about the needs of others (no one can need you to not have your needs met)	We should consider updating our preferences when we learn about the preferences of others (someone can legitimately have preferences about your preferences)
If someone’s needs are not met, this tends to make them feel distressed, attacked or hurt	If someone’s preferences are not fulfilled, this probably won’t hurt them too much (unless those preferences were actually meeting some need)

Two further applications of the distinction

Understanding compromise

I used to find compromising on some issues very difficult, because I felt like it threatened my (need for) autonomy, and yet also feel guilty about that (as in, I “should” be more okay with compromise). Now I know that compromising is fine and healthy on preferences, but not on needs. Before, I couldn’t see the difference, and so my bad experiences of compromising on needs led me to be averse to compromise in general.

Better self-care

One other application of this distinction is in how I’ve changed my attitude to self-care. I used to model self-care as fulfilling some preference. Combined with my belief that preferences are malleable, this was a precarious situation to be in. For instance, when I felt tired at the end of a day’s work:

• part of me wanted to rest

• another part wanted to work more

• aggregating these preferences, the part that wanted to work more used to speak louder, so I let its preferences dominate

• but now, I see that the part which wanted to rest was asking for me to meet a basic human need, rather than a preference, so I give it what it wants, without guilt, because I’m fully onboard with the fact that meeting needs takes priority.

As I said, these ideas are all sketchy and fresh, so I’d really appreciate feedback on what parts you found wrong or not useful :)