Conversations are better with four people | Hacker News Reader

Conversations are better with four people

thetimes.com

230 pointsnomilk149 comments

The article quotes Dunbar as saying "In the normal run of things, when a fifth person joins a group, it’ll become two conversations within about 20 seconds." I've noticed something similar. The article suggests this might be because of the difficulty we have imagining what multiple other people are thinking, but I suspect other causes also play a part.

If I'm in a group of five or more, and I think too long about what somebody said, or about how to respond, I find there are two conversations going on without me. In a group of four, the only way it can split into two conversations is if I participate. If I don't, the other three have no choice but to maintain (at most) a single conversation, making it easier for me to rejoin.

That's not so much because it's five people as it's a new person.

in my experience it is more dynamic than that. based on tech meetings that i have been to, it depends on the topic currently being discussed. larger groups will break apart if not everyone is interested. it also depends on the arrangement. if a few people stand in a circle and more join the circle, the conversation won't split, because for that to happen they would actually have to leave the circle and create a new one.

This is a really interesting observation, because I'm in an adult discussion class in Sunday school that is arranged like a circle, and I've been trying to figure out why it seems that structure works so well at limiting conversational divergence.

It seems that when everyone is forced to look at each other, it's harder to divest from the main conversation without drawing your attention away from the remainder of the group. It seems better for fostering discussion with a single speaker at a time since everyone can look at that person all at once.

It's not perfect but for larger groups the "circle strategy" definitely seems to work well.

Thanks for sharing!

This is the social upside of something like a “blunt rotation” forcing every person to periodically break from carrying the conversation.

I'm going to shamelessly borrow your phrasing, because what you've described is exactly what happens: It's a rotation through other participants so neither the study leader nor other individuals have to "[carry] the conversation." It's really interesting, because it fosters conversation that can lead to interesting questions, observations, or other information that might not otherwise come to light. There are some significant deficiencies, of course, but I think works fairly well depending on the group.

Where it breaks down is if one person starts to dominate the conversation for the duration of the class and carries it off-topic, or if someone becomes combative. So, the group composition and personalities can influence the relative success.

The circle shape makes you feel engaged and able to converse with everyone so it feels cohesive. It's also pretty obvious you were arranged in that shape to emphasis a group conversation activity is taking place.

Kind of the opposite as when you get seated at a restaurant in a long rectangle table. You immediately know it will be difficult to talk to the whole table and will have multiple conversations taking place. If you're at one end, you'll likely only talk to 3-4 people. If you're at the center, you might talk to 3-4 to the left and right but they'll be different conversations. Not that this is a strict law of table talk, just kind of what usually takes place. Sometimes there's something that comes up that gets everyone's attention and the whole table is shouting. An even more extreme example, is a bar top seating, where you are just a line facing the same direction. You might only talk to your neighbors and possibly their neighbor but it's not great at facilitating larger group discussion.

If you ever go to a banquet or wedding where they have round tables but a very tall floral decor piece thing in the middle, it completely breaks the circle advantage. I think eye contact and your ability to convey body language is a major component.

I really hate trying to carry conversation in restaurants for that reason, and because the background noise levels can often get to a point where you can only hear the person sitting right next to you.

You're absolutely right that 1) group composition, 2) room structure, and 3) motive(s) are all important factors. As someone else observed, having a "discussion leader" is also important in that sort of setting.

It's interesting that you see a large group of people having one conversation as a success. Personally I really dislike those situations and would much rather it splinter into smaller groups.

I think it depends. In the context of a discussion class, it makes sense, because you don't want to deviate too far from the purpose of the study. As someone else pointed out, it doesn't work as well for unstructured conversation.

Is there a clear leader/facilitator in that discussion class? Someone who keeps the discussion "on track" and prevents it from wandering from the chosen topic?

That would be different from the spontaneous, unstructured conversations the article is talking about.

Yup, and that's another reason it works so well.

Now, it does break down somewhat when the core discussion is over or if someone is a bit disruptive (which has happened recently).

As a sibling comment to yours wrote, it DOES help that the topic is understood among everyone present, there is a clear intent to the gathering, and everyone has approximately the same motives.

Your case is relatively unusual and unlike general conversations, since you have:

  - a topic that everybody knows will be the focus, and
  - an audience that actively values your conversation.

Outside your core discussion time is when you'll see general discussion tropes. If people are (generally) neither leaving nor arriving, in my experience we see transient circles of 6-8 that split (if more than 8 people, or if multiple topics persist) and merge (if fewer than 6 people). The article's limit of 4 doesn't apply if movement between different groups is considered a feature rather than a bug.

That's exactly it, and I really like your observation that the "limit of 4" doesn't apply in some cases. Before/after class, we have the exact experience you mentioned here where people will transiently go around talking to different groups (often 2 or 3 individuals, sometimes more where there are more listeners than speakers).

What's really interesting about your observation is how the rule-of-thumb breaks down when the conversation is limited to more confined topics or: If the individuals see each other with some regularity outside the group setting (no need to engage in extended conversation about who's doing what) or some of the conversations involve topics brought up in the course of the class.

That's a really interesting observation too. I've always preferred round dining tables for six people at home and hadn't really made the connection with the idea that it does allow for a single conversation rather than a rectangle with two rows of three people facing each other when it easily breaks into separate conversations.

Probably because people can focus on 3 - 4 things at a time.

90 years ago in Serve it Forth, the glorious M. F. K. Fisher stated definitively that the maximum number of people at an ideal dinner party is six, and probably three or four. Glad to see she's being upheld! Of course, the individuals probably matter more:

>It is, though, very dull to be at a table with dull people, no matter what their sex. Dining partners, regardless of gender, social standing, or the years they've lived, should be chosen for their ability to eat—and drink!—with the right mixture of abandon and restraint. They should enjoy food, and look upon its preparation and its degustation as one of the human arts.

> 90 years ago in Serve it Forth...

I wish there were a tool that could quickly connect new articles to older ideas, highlighting how much of what seems original today often builds on existing work, or just was mentioned before.

It wouldn't need to be as precise as mathematics or the hard sciences, just probabilistic enough to reveal meaningful connections.

That would be _tres cool_.

Thinking out loud here...That could be like an LLM that is tokenized on ideas as opposed to words/fragments, no?

I'm sure someone has worked on this, but how would a computer extract an idea from a bunch of words (sentences?) ?

Looks like another rabbit hole to add to the list...

I would say that tracing concepts back to their origins could be a good benchmark for evaluating AI models.

In the context of the article you mentioned, Robin Dunbar’s research and the Max4 Principle are referenced [1], along with a relevant Wikipedia article [2]. Expanding further, one can trace earlier foundational works, such as “The Primary Group as Cooley Defines It” [3], and even earlier sociological contributions by thinkers like Auguste Comte.

As for the “Serve it Forth” answer, I haven’t come across that yet except when, obviously?, I named it explicitly.

[1] https://newcreate.org/max-4-principle/

[2] https://en.wikipedia.org/wiki/Communication_in_small_groups

[3] https://www.jstor.org/stable/4105179

It could only go back to 1992 or so. For most people, history is whatever is linkable or the 5% of pre-1992 stuff that is popular enough to rate a dead wiki footnote.

The Google Ngram Viewer indicates that data was indexed well before 1992, even if the resource isn’t publicly accessible [1]. This likely reflects the reality that new content often garners more attention than old content, which doesn’t align with the "pay-per-click" model of monetization.

[1] https://books.google.com/ngrams/graph?content=darwin&year_st...

When I was a full-time IT industry analyst, one of the things that regularly annoyed me was overly large tables at dinner. A lot of factors were in play including the fact that many analysts felt very strongly about being at a table with the "important people" (tm) in the room. But if you were at even a round table seating 8+ people, it meant in practice you only talked with a few people who were reasonably adjacent.

I would say a 4-6 person table is about the largest where everyone can be talking with all the other people.

Six can be a bit of magic number for a table in a larger crowd, because people tend to be in groups of 2 or 4, so a table for six (apparently) means you’re inviting/ causing groups to mix.

Three seats is also magic? To invite/cause one person in each such group of 4 to go and find new friends. Plus some drama

At weddings, table sizes seem to be 8 or 10. Is that deliberate, to create 2 conversation groups?

I sometimes find myself trying to follow both conversations, or switching to the more interesting one. Is that bad etiquette?

Could be - making sure that everyone at the table always has a choice of conversation to listen to, instead of being trapped in one they have no interest in.

I doubt it's that explicit, though. Maybe it's just worked out better that way over the years, without anyone fully realizing the why?

I suspect that it's more about the geometry of best fitting people into the space. It allows the right amount of room for centerpieces and bread bowls and such. At least as far as the caterer and wedding planner are concerned, conversation is the last thing on their minds.

Those tables are intended for eight comfortably. They stretch to ten. You can do twelve, but everyone is gonna be cramped.

It's a size and shape that's so omnipresent one expects that it's some combination of geometrical efficiency, ability to pass/reach things, some flexibility in self-organizing conversational groups, and probably some other things such as not being too large to setup, teardown, and store.

Dinner parties at someone's house are such a lost art. I'm going to use this winter to start to bring them back!

My parents used to host dinner parties. Actual sit-down dinners at a table with place settings, not pizza with paper plates. They are a lot of work, especially if you do the cooking yourself. I was pretty young and only remember them vaguely, but they stopped because nobody ever reciprocated.

The problem with dinner parties is that they only work if all attendees like each other equally. Any group of people talking at once will boil down to the lowest common denominator, so one bad guest can easily ruin the whole event. Meanwhile during a pizza party it's okay to chat in a smaller group, which means that when you have a complex network of "A likes B but doesn't like C who is best friend of B", there's much higher chance of making it possible to somehow divide the guests into sensible subgroups.

Well that brings up another thing that we seem to have lost in society: how to be pleasant at a social event even if you don't like or find agreeable everyone else there.

Oh, I can perfectly do this! Actually, I'm very good at this! It's just that after 8 hours of doing so for 5 days in a row, I don't have any more energy for that.

Do you have any tips on how to perfectly coexist with disagreeable people?

It sort of depends on a mutual politeness or the sense that a dinner party is not the time or place to bring up (or persist in discussing) something controversial. If someone insists on being an asshole and spoiling the event there's not a lot you can do, but it seems that more people used to have the social graces to not to spoil an event they were invited to, and would be embarassed if they did.

Consider I, myself, might be disagreeable to others and they seem to perfectly coexist with me (so far).

Every invitation to fight doesn't have to be accepted.

Beat poem about that fifth guest: https://youtu.be/eYJ60eWprlQ?si=VnWBdn-1iqJT3bOL

Well, it depends. If those people are willing to accept our differences, then I reciprocate, and we can be friends. I accommodate them, they accommodate me. But if gap is too big to be bridged (they have a completely different personality), or they don't want to reciprocate (they insist on doing things their way because obviously that's normal and better and everything) then I just stop caring, and treat them as necessary evil. Coworker is passionate about ruining a project and manager won't do anything about it? Not my problem, just make sure that my suggestions were written down and then rejected. I'm not paid to do engineering, I'm paid to put up with stupid people and office politics while doing minimal amount of engineering. Aunt feels the need to spread the word of Jesus? Well, looks like next time I won't make it to visit her, and if I absolutely need to because of complex family relationships, I'll just nod and smile while being mentally absent.

At the same time, I refuse to spend my free time with people who challenge my patience. There's a clear distinction between professional/family me, and the leisure time me.

Yes. Adlerian philosophy. Start with the book ‘the courage to be disliked’ and work from there.

Be polite, even if they’re not. Choose to ignore their bait and slights. Choose to not engage with their negativity, and only say what is necessary and nothing more. If they complain, whine, or speak ill of others, affirm that you see they’re upset, and move on.

Anyone who has raised children ought to recognize these tactics.

The nice thing about dinner parties is that you can choose who to invite (and not invite).

Both are great! We love having people over for a nicely presented dinner or pizza. The import part is getting people together.

People are still doing it with game nights: have dinner, then play a board/card game together (usually with drinks.) In the 50s, Bridge or Dominoes was a big married couple date.

I'd like to see a revival of the old setup where everyone has dinner, then everybody in turn, individually and/or in groups, does a musical performance, poetry reading, or interesting lecture to entertain everyone else.

Pro tip 1: Potluck! Cooking and organizing can be a lot of work.

Pro tip 1a: Hot pot! You supply the pot and broth; guests bring the rest.

Pro tip 2: The first one will be waaay harder; then it’ll get easier, as you get a track record and you don’t need to recruit (or “train”) guests; they’re already in and already trained :)

Pro tip 3: They get easier, fast, as you both 1) get practice and 2) learn what doesn’t matter

Pro tip 3: The first people to show up will ask to help; plan something for them to do (I do a make-a-pizza party; I have them grate the mozzarella)

Thanks for this! I hope to put it in practice soon!

I found "The Two-Hour Cocktail Party" on Hacker News, and the advice really resonated with me although I haven't had the opportunity to host yet. The author suggests doing a few of these cocktail parties before a dinner party, as he considers those to be advanced and quite stressful for someone not used to hosting. I have to agree, I read through it right after hosting a Thanksgiving dinner party that was a lot of fun, but was... trying.

https://news.ycombinator.com/item?id=32879851

Some variations:

- Formal. Host does all the planning and provides all of the food. At most, some people help in the kitchen.

- Informal. Don't use the good china or fancy napkins, might eat in the kitchen or living room. Often used before a game night or similar.

- Collaboratively assigned. People are assigned a particular set of foods (bread, potatoes, etc.) to prepare (usually about 2 except for the host who might do more), common at holidays (usually formal in winter and informal in summer).

- Semi-assigned. People are assigned general categories of what food (meat, side, salad, dessert) to prepare. Useful if the meal is for a group smaller than the chaotic-potluck threshold, or if you don't have a potluck culture ingrained.

- Location. In a house? In a rented facility? In a public park, or at the beach? Sometimes "on your front lawn / in your back yard" is a reasonable answer.

That's because everyone works for someone else these days. Back then more people (particularly women) worked for themselves, building their own wealth (house, social life, family etc) and their own expertise. Now everyone works for someone else, building someone else's wealth and becoming an expert in someone else's domain etc.

I like that even as someone nearly 60 years old there are still new words to learn. TIL "degustation" - the act of tasting with relish; savoring.

"three couples" makes an excellent group!

I think it needs clarification. What I've found is that:

* 2 people is ideal for serious, deep conversation

* 3-4 is ideal for more humorous, relaxing conversation, but precludes the deepest intellectual topics

I wonder if that's why some people are introverts, without even realising the cause.

I'll do small talk, but it's more boring than mowing the lawn with a pair of scissors. I find large gatherings completely boring, with nothing but noise involved because, well.. it's all just meaningless chatter.

Maybe some people aren't introverts, just "talking to more than a few people means this is dumb" verts.

I concur, and this can be solved once you get the hang of actually creating a small bubble inside crowds and accept the FOMO of missing the rest of the event to have a quality conversation for a few hours.

Not everybody is made to ride a wave of people. But you can find a way to enjoy it in a different fashion. I find it true for a lot of situations, actually.

You don't have to follow all the rules.

There's probably a certain unconference aspect to larger groups. (i.e. if something isn't working for you feel free to move on.) "This is really interesting but I really need to say hello to my friend over there who I haven't seen in ages." Of course, there's some art to politic transitions.

I’m an introvert. I prefer long stretches of solitude to just about anything else. I need people and interaction, but in very small doses.

Me too. I think this what introversion means.

Being quiet at bigger gatherings is a different thing.

This is why I always go outside to hang out with the smokers at parties, even though I don’t smoke. The conversations are always more interesting in those small circles of 2-3 people who went outside for a cigarette than in the main party.

It is worse when story tellers come in and hog conversations. It is even worse worse when there are more than one and it is one upmanship on the war stories. Find this unpleasant and tiring. I prefer listening from more people.

Yeah I dislike it when people always jump in with their experience rather than relating to the person who is talking.

I figured out that some people are "relators" and some people are "analyzers and askers" and it's usually the latter that is much more fun to talk to. But I know that two "relators" who seem to talk past each other often also seem to enjoy this method.

> people always jump in with their experience rather than relating to the person who is talking.

For at least some of them, that is them relating to the person.

My significant others' family we see several times a year has several such folks, and at least three of the top level people at the small company where I work are the same way. I've grown to be able to handle it much better than I could 20 years ago, but it's still deeply exhausting.

On a bad day it will drain my energy to the point where I truly lose most of my ability to function and need significant time to recover.

Agree, and I often derange the conversation to cut their story shot lol ( sorry ). Unless they're really determined to continue.

I notice other people do this to. And there are even different skill levels to how smooth they can cut into the story tellers.

> derange the conversation

I was expecting to read "derail" there, but that's not bad either - I mean, it's bad, but in a deranged way ...

> I'll do small talk, but it's more boring than mowing the lawn with a pair of scissors. I find large gatherings completely boring, with nothing but noise involved because, well.. it's all just meaningless chatter.

The trick I’ve found is to make the conversation more interesting, if you think it’s boring. 9 times out of 10 everyone is just jonesing for someone to take it somewhere fun, but afraid to make the first step. YOU can make that step.

> The trick I’ve found is to make the conversation more interesting, if you think it’s boring. 9 times out of 10 everyone is just jonesing for someone to take it somewhere fun, but afraid to make the first step.

Not in 2024. You're 100% going to get a person who's going to get offended and make a scene.

Which is why I refuse to hang out with people who fulfill these two conditions at once:

1. They're boring

2. Their opinion on me matters

I don't think I agree with your statement.

But even if it's true, at least it won't be boring, right?

What are some concrete tips?

Sometimes I ask people about their internal monologue, can they change its perceived voice? Do they even have one? That often sparks some interesting conversations.

To me this means 1) You aren't hanging around the right people 2) You aren't talking about stuff you're interested in 3) You need to drink more.

I mean most people are just pretty boring. I find that people talk about:

Work (generally not a great idea at parties, maybe a bit more acceptable if it's hosted at work, or if you do something really unusual that others find interesting, and are able to talk about it in layman's terms).

Sports (I like sports well enough but am not passionate about any team or sport. I cannot add much to a conversation about specific players, games, coaches, statistics, strategy).

Wine, whiskey, tequila, food: See Sports, above.

Their kids or their vacations or other bragging. Nobody cares.

Reminiscing about some experience that a group of them had together. Hard to join in, if you're not part of that group or that experience.

Politics, conspiracies. Just no.

Interesting people who talk to people they don't already know are rare.

One useful trick is to treat each conversation as a game in which you have to discover the most interesting thing about the other person. Often, a conversation partner that seems boring is reserved, shy, or waiting for a social signal that it's okay to deviate from the topics you've listed (which can provide an 'in' to the really interesting stuff).

This is so interesting! I had this experience just a few weeks ago at a restaurant with a group of people where I talked about the most introverted and extroverted in the group, with another person who seemed to be introverted and that got them interested in talking a lot more just with me. It was so much more fun than listening to the 2 other loud people in the group.

There are 8 billion people whose "things" I am not interested in discovering. Why would I be interested in a random person's things just because they happened to be physically proximate?

How do you know you're not interested in it when you don't know what it is?

I don't know but I can give a rough estimate, based on my experience. The estimate equals 0.42%.

Alex? :p

I'm sure not, but fun to imagine this written by someone I know.

Beware small sample sizes

Because "things" are interconnected. There's not just one definition of Thing, whatever your thing might be. There's your perspective on Thing, and there's other peoples' perspective on Thing. So, whatever it is, Thing is not something that you can understand in solitude.

Understanding it means understanding both your perspective on it, and other peoples' perspective on it, at the same time. That's most peoples' reason for wanting to interact with random people. They see it as an opportunity to refine their understanding of Thing, whatever it might be.

I can't help but wonder what topics you do find interesting?

Exactly. This is the type of person I avoid like the plague in real life.

Same for me, in most situations I feel like 2 people (me included) is the perfect number, as I find it to be the only way for a conversation not to diverge from its pivotal argument ~2-3 minutes into it at maximun. Though sometimes when I feel more talkative 3 people is somewhat like a sweetspot, in disagreement with the article (for me personally) when the irl server reaches four it usually feels to me like a point of no return I wished we didnt get ourselves into.

I don't know if that's the reason but I couldn't agree more. The more people are in a group the more boring it tends to become for me too.

Maybe it's because it's more difficult to find shared topics of conversation. Or maybe people tend to avoid deeper topics because intimacy is lost. Or maybe there's some other group dynamic at play (family stuff etc).

You get into more people and you're likely to just be hesitant to go there on a lot of topics--at least unless you know everyone pretty well.

I think the easiest way to look at it is that you start with every topic available, then you filter down to what people are comfortable with. So, you can only ever lose topics as you add people, looking for the lowest common denominators.

I don't think this is true at all. If you have a bunch of engineers at the table, they have plenty of topics available, yes, but say you add a medical doctor to the mix, you've suddenly unlocked the topic of how the two professions compare, interact, overlap, etc. With just the engineers, you don't have that, outside of surface level speculation.

Very interesting, but there's no link or reference to a paper here, just

> But his research has also explored how people act in smaller groups.

My assumption is that Rhys Blakely was at Cheltenham Science Festival enjoying a lecture, and decided to fluff up a minute of it into an article. He seems to have quoted from one of Dunbar's popular science books; maybe he should have checked the footnotes or the bibliography?

We're on the internet now, we don't have to "Authorities say..." anymore. This sounds interesting to me, but I don't want to look through every paper Dunbar has written to find it.

-----

edit: curiosity is annoying. I found this:

Robin IM Dunbar, Neill DC Duncan, Daniel Nettle "Size and structure of freely forming conversational groups" (1995)

Abstract: Data from various settings suggest that there is an upper limit of about four on the number of individuals who can interact in spontaneous conversation. This limit appears to be a consequence of the mechanisms of speech production and detection. There appear to be no differences between men and women in this respect, other than those introduced by women’s lighter voices.

https://scholar.google.com/citations?view_op=view_citation&h...

-----

edit 2: and this:

Guillaume Dezecache, R.I.M. Dunbar "Sharing the joke: the size of natural laughter groups" (2012)

> Our results confirm, with a considerably larger sample, the upper limit of N≈ 4 on conversation group size reported by Dunbar et al. (1995). In addition, they suggest that there is a similar limit on the number of individuals that can be involved in a laughter event.

https://doi.org/10.1016/j.evolhumbehav.2012.07.002

Someone I know goes to a moms group where they play this dice game called Bunco¹ They've been playing it for 20 years.

Tables of four people quickly form, play and move.

The game is based on luck, not skill, so people are social not competitive. They're also concentrating on each other not so much the game.

I got to attend it once and play, and found it to be an exceptional way to get to know absolutely everyone in the room.

[1] https://en.wikipedia.org/wiki/Bunco

"over 59 million women have played bunco and over 27 million play regularly"

One of the reasons I like D&D and board games is because it gives something to do while socializing.

The idea of inviting someone over just to hang out - or being invited to do so - without an underlying activity fills me with dread. But if I'm being invited over to play a game, then that's great - and I can chat and socialize while doing so.

I don't think it has to be Bunko, or any one particular thing. Hell, even helping friends move is fun in a way, as long as your spine is up to the lifting :P

I am the exact opposite!

If the hangout is activity-based, I find I feel forced to care about and talk about the activity, but I do not really give a flying fuck about the activity. For example going to do axe throwing? I don't care at all about axe throwing and have no interest in talking about it, I do have interest in talking about how my friends are doing, anecdotes, politics... I just want to sit down and _talk_ (a drink or meal works very well for me). Board games are hell to me, I don't care at all how many beans I can earn this round, I wanna talk about your home renovation or the trip you just went on.

For example I had a first date at an arcade. The games were actually pretty cool, but I wasn't really interested in them in the moment (I was interested in my date) and the chit-chat about the games felt very forced: I really didn't care on the moment. As soon as we took a break to grab soft drinks, suddenly conversation was flowing and infinitely more interesting.

Of course there are activities that I enjoy for their own sake and like doing with friends (eg rope climbing), but it's a very different type of hanging out which doesn't bring me the same pleasure at all.

http://archive.today/WTDBs

I suspect this is related to why a string quartet is the right number of musical voices. Two violins, viola, and cello give you a very fulfilling number of separate ideas to track without overwhelming you.

That's also a "classical" rock band - vocals, guitar, bass, drums (e.g. Beatles any many others)

Most rock bands have more than one guitar.

Which could be in the vocalists hands in the gp's example keeping the number at four.

Bass is a guitar in 99% rock bands.

I have not seen a good band with more than 4 people on stage simultaneously.

AC/DC. Foo Fighters. Guns 'n Roses. Queens of the Stone Age...

well I have not seen them in particular ^^

There are good bands like that but not most.

Let's not forget the synthesists

Rolling Stones, Lynyrd Skynyrd, Aerosmith, Journey, Radiohead, Guns and Roses, Fleetwood Mac, AC/DC, The Eagles

Have not seen them either.

You pick really popular names but in most regular bands when there is more people there is worse link between them...

"Music for 18 Musicians" by Steve Reich is probably one of the masterworks of the second half of the 20th century.

Any vaguely disco-adjacent band will have more than 4 people on stage because there will be at least keyboards and horns in addition to drums, bass, guitar and vocals. Even a band that simply adds an additional person player percussion to a typical 4 piece exceeds your limit yet can wonderfully enhance the music.

If you haven't seen any of those bands, then that's your loss, but provides no reason to try to generalize about the right size for a live band.

It's anecdotal and subjective but I go see many rock bands and good ones always have 4, 3 or 2 people :shrug:

Vocals are often a person who is also playing an instrument. So in a 4-person band you can have up to four voices, lead and rhythm guitar (or maybe keyboard), drums, and bass.

>> I have not seen a good band with more than 4 people on stage simultaneously

Glenn Branca Orchestra.

what does it have to do with the point

> I have not met a religious programmer

> Donald Knuth

Yes but parent poster was almost certainly talking about lead guitar and rhythm guitar.

You've never seen a big band or an orchestra? Sir please.

I saw an orchestra once or twice. We're talking about rock music here.

I've seen bigger bands but they suck in comparison.

Guessing from your response, you might not know of https://en.wikipedia.org/wiki/Big_band, which is fine. But now you do :)

Edit: I thought you linked to yet another famous band. People keep doing that... 99% of bands a normal person would see in normal life don't even have a wikipedia page.

However your link looks about jazz.

A common rock band is rarely good with more than 4 members because people lose unity and it's just technically harder and people are rarely professionally trained.

Honestly I wouldn't know anything about rock bands because I don't listen to it at all so you could be very right. I just responded because this thread was talking about string/barbershop quartets, which are almost definitely 4 parts because that is the minimum number of people required to make a 7th chord, and then again because you didn't know what a Big Band was, which I suppose is very understandable, but as somebody that grew up around and playing on them it's super foreign to me.

Ah. I was replying to "Most rock bands have more than one guitar". I don't listen to barbershop quartets, not even sure what it is

Also the traditional barbershop quartet for acapella.

Interestingly, I like the 5-piece versions of all 3 of these: add a keyboardist to the rock band, a piano or harpsichord to the classical string or woodwind quartet, a female vocalist to the acapella group. Having two leads lets you do much more intricate countermelodies and harmonies.

A string quartet consists of 4 tonally adjacent instruments, and is thus much more like 4 humans talking.

A "classical" rock band consists of 4 utterly different instruments from a tonal perspective, and is thus nothing like 4 humans talking. Same thing for jazz - and its why you can have multiple instruments performing simultaneously and in ways that are not obviously connected to each other.

I recall that Charles Rosen wrote somewhere that one of the reasons the string quartet took off in the classical period was that it allowed the playing of all the notes in a dominant seventh chord without double stops. Although this was probably a better explanation for the relative paucity of string trios in the output of Mozart (1) and Beethoven (0). The establishment of four parts as the "standard" scoring for vocal ensembles can be traced back to the 15th century.

On the other hand the second and more famous dining (and conversation) club founded by Dr Johnson had originally 9 members, and gradually grew from that to dozens. Although many including Johnson may have not been entirely happy with the expansion.

I think you're taking the metaphor about a string quartet as a "conversation among equals" too literally.

In terms of perception, I'm not sure there's much of a relationship to a human conversation. To make things equal, the string players would need to take turns soloing while the others wait more or less silently to respond, each with their own solo response. You'd be bored out of your gourd if string quartets were written that way.

But more to the point, the vast majority of time in a string quartet is devoted to two or more of the players producing phrases of music in parallel, and that is musically coherent and pleasing to the players and audience. Most humans cannot track two humans speaking in parallel at all. That alone tells us that music cognition is a very different phenomenon than speech cognition.

In short, I'm not sure why a string quartet would be considered the optimal genre for humans to produce music together. And even if it is, the reasons why are even less likely to do with the protocols around human speech cognition, and certainly not with some bizarre equivalent of the "theory of mind" associated with the musical phrase produced by one of the instruments[1].

1: Small digression-- In Elliott Carter's 2nd String Quartet he actually started with a concept that each instrument was a kind of "character" in a play among the quartet. In this case, the problem with OP's metaphor becomes obvious even in the introduction-- the homogenous timbre of a string quartet makes it difficult to hear the differences among the characters. (IIRC I think even Carter admitted this.)

Counterpoint may leave too much implied with only two voices; with four or more voices one must increasingly break or relax various rules that promote voice independence, e.g. the use of parallel motion where additional voices simply double some other line (they can't all be independent, there's too bleeding many of them!), or to drop voices for a thinner texture, for example where there are five instruments but only three or four of them are sounding together most of the time. That's a long way to say that around three to four voices is ideal if you want independent lines (except they're not really independent, like two people shouting past one another; there's a weird mix of both working together while each yet manages to stand out in good counterpoint) though even better than this claim would be to compare, say, Bach's two-part inventions to works that have more voices.

For those who do not know counterpoint, you have only three motions a voice (a horizontal line of music, traditionally sung) can make relative to another voice (move closer, apart, or to hold steady) combined with limited voice ranges (say, a doubling of frequency, or so) and limited interval choices (seven, or so) within an octave or frequency doubling, and the voices are very close to one another but only rarely cross one another, on top of all that various rules systems that forbid or frown on such things as the tritone, parallel fifths, and so on into the weeds such that with more than a few voices you quickly run out of valid options for all the voices to move independently.

I think the big question is the components of the groups’:

1) willingness or desire to contribute (hopefully everyone does

2) the discipline of the participants (I find a growing trend of people wanting to talk over each other, start talking before someone has finished their thought, people wanting to dominate the conversation)

I’ve had conversations that are intolerable with 2/4 people because they’re either hard to talk to or talking over you, conversely I’ve had lovely conversations with 8 people that have gone extremely well.

> I find a growing trend of people wanting to talk over each other, start talking before someone has finished their thought, people wanting to dominate the conversation

Been noticing this at work meetings. It’s annoying because if you want to be heard, you have to become part of the problem… I hate it. I wonder if social media has anything to do with this, since it created a way to “win” conversations, and heavily encourages you to do so. Zoom meetings do amplify this problem, so maybe it’s always been like this and I forgot?

In large social groups I shutdown if people talk over me. I don’t want to fight to be part of a conversation. It’s fine, I’ll have moments here or there, though it does make me look quiet and weird… but there are other friends that don’t talk over me and I doubt they would say the same thing (the quiet part at least lol)

Zoom converstations are impossible with more than a few people, because there is an unnatural latency (even if slight) and the subtle body language cues that let you know it's a good time to speak are totally hidden. So three people start talking at once, then everyone stops and is quiet for a few seconds, then two people say something, etc. It's even difficult with just two people sometimes.

I do think social media has something to do with it, but I think to me it’s more the shortening of attention span e.g long form journalism -> tweet, Books -> TikTok’s, I think people just actually struggle to contain themselves if someone is speaking for more than ~20 seconds

I find large groups exist in three states of matter:

1. crystalline / couple position: Couples sit around the table, there is about 15cm distance between each molecule. One person at the table holds the conch and speaks.

2. bipolar / sex-segregated: One end of the table holds all men, the other all women. The women talk about men and the men talk about poop and race cars.

3. gaseous: All people move freely around the room and bounce off each other, there is vigourous chatter everywhere. Often there is alcohol involved.

Your job as a host is to increase the temperature to where all molecules break apart and you reach gaseous forms of communication.

Absolutely an overlooked responsibility in hosting is making sure everyone is included and enjoying the experience. Or at least making the effort.

This is genius. Did you come up with it on your own? Also, any tips?

Yes thanks :p I used to be extremely uncomfortable in social situations so I have to reflect a lot.

I don't like having a lot of attention on me but that actually works great because people love talking about themselves. In group situations I try to identify super extroverted people and people with a high energy level that day and bounce random stuff off them to get the ball rolling, they will entertain the group. Then I'll single out a low-energy person and get them talking about their life.

But that's just what works for me. It also sounds weird and formulaic when I say it like that but it's not a premeditated thing, I just natually act that way.

Probably the most important thing when hosting is to project that you enjoy having these exact people around at this exact moment. If you can get yourself into that mood good things will follow. It's easier with people who fit into your dynamic.

My experience is more that 4-5 is optimal, and groups splinter into 3+3 at 6. It could be because I'm an introvert and tend to hang out with other introverts, though. A group of 5 will usually have 2-3 people actively talking and 2-3 listening and occasionally contributing only if they have something worthwhile to say. 5 extroverts, in my experience, are unmanageably loud and will usually split into 2+3.

Note the sample bias in this. If you're studying social gatherings, introverts tend to disproportionately not attend these, because they are energy-draining. If you are an introvert, however, then any gathering you are part of will by definition have at least one introvert in it.

Good point!

Three-way group chats have been really great for me (as a 45 year old male) for keeping connection with my guy friends. Have a dozen of them or so. Small enough to still feel intimate. Big enough that someone replies / keeps the conversation going.

Four people seems to be an interesting threshold when it comes to groups. The British special forces, for example, started doing four man teams and several units of other countries followed their example (SEALS, KSK, etc).

https://www.sasregiment.org.uk/patrol.html

When a group is small enough, people will often 'wait their turn' (i.e. wait until the current speaker is done saying something) before talking. Once a group gets a little bigger, two or more people will try to speak at the same time. This causes the discussion to split into separate groups, each carrying on a different conversation.

The bit I could get to in the HTML:

> Why conversations are better with four people

> Three might be a crowd but four appears to be the magic number when it comes to conversation. And, according to an academic who has spent decades studying how we socialise, William Shakespeare instinctively understood that.

> Professor Robin Dunbar, an anthropologist and evolutionary psychologist at the University of Oxford, is known for “Dunbar’s number,” which alludes to his theory that most of us are able to sustain about 150 social connections.

> But his research has also explored how people act in smaller groups. At Cheltenham Science Festival he explained that when it comes to having an enjoyable chat, the upper limit is a gang of four. When social groups have five or more members, the chances of them laughing together plummet.

If anyone's got the whole thing, please post.

Three might be a crowd but four appears to be the magic number when it comes to conversation. And, according to an academic who has spent decades studying how we socialise, William Shakespeare instinctively understood that.

Professor Robin Dunbar, an anthropologist and evolutionary psychologist at the University of Oxford, is known for “Dunbar’s number”, which alludes to his theory that most of us are able to sustain about 150 social connections.

But his research has also explored how people act in smaller groups. At Cheltenham Science Festival he explained that when it comes to having an enjoyable chat, the upper limit is a gang of four. When social groups have five or more members, the chances of them laughing together plummets.

He said: “You very rarely get more than four people in a conversation. In the normal run of things, when a fifth person joins a group, it’ll become two conversations within about 20 seconds.” Alternatively, a “lecture” situation develops in which one person holds court and the others act as an audience.

In larger groups, “you have to decide whether the person who is speaking is really so important you’d rather be standing there saying nothing”, he said. If the speaker is not very interesting, the audience tends to splinter into groups of four or fewer. Dunbar believes that the underlying reason is that we can only track what a certain number of people are likely to be thinking at one time.

Scientists call this the “theory of mind”, which involves being able to see the world from another person’s perspective. Also known as “mentalising”, it is crucial for conversation because people often use imprecise language, which makes context important. “The language we use is full of metaphors and unfinished sentences. The listener has to be able to figure out what it is the speaker is trying to say,” said Dunbar.

Additionally, the speaker must track whether those they are talking to are following their meaning. Dunbar believes that the limits of our ability to predict the thoughts of others in this way explains why groups that work in challenging situations — such as SAS patrols and surgical teams — tend to do best when there are four members.

In The Social Brain, co-authored by Dunbar, he argues that Shakespeare must have intuitively known about this phenomenon as it is rare for his plays to have more than four significant characters speaking in one scene.

He wrote: “[Shakespeare] instinctively understood the mentalising capacities of his audience. He was anxious to ensure his audience wasn’t cognitively overloaded by the number of minds in the action on stage. [It is] a masterclass in the study of human psychology.”

Not exactly Shakespeare but this reminds me what I enjoyed so much about the opening act of Cirque du Soleil's O show was how it used so many acrobats and dancers to choreograph a concurrent scene, you were just overloaded with what you could pay attention to at any one moment. Their specially designed stage was large enough to show multiple things going on in a 3D space.

Bach's fugues using 4 voices and other such classical music is also an interesting example, the complexity is often written to be beyond the reach of even an experienced listener to grasp everything concurrently, and the performance of this kind of music is necessary highly challenging for the pianist as well.

Always assumed the reason why conversations or business meetings, should not have more than four persons was due to Metcalfe's law...

https://en.wikipedia.org/wiki/Metcalfe%27s_law

It also depends on the people a lot. There are at least two types of people who are tiring to be around:

1. The one who is very slow to get to the point. At around 10 second mark you can see where the story is going, but still you have to sit there for several minutes listening the person get there.

2. The one who has endless amount of stories about themselves. Initially you think it's interesting and entertaining, but soon it dawns on you that the stories are actually boring and waste your time. Typically people like this have developed a sense that people don't like to listen to them, and have excellent defenses against all attempts to shut them down politely.

Podcasts can illustrate some dynamics. The dominant format is two people. Three is rare. Four seems to work, but in a more shallow way. Five? Never heard one.

Is it worth counting/considering the listener? Multiple people listening also leads to interesting things like pausing the podcast for a side conversation.

Incidentally, half my regularly listened to podcasts have three hosts or two plus guest.

personally I enjoy conversations in groups! it gives me time to consider what I'm going to say and gather my thoughts, I have ADHD so in a one-on-one convo I tend to overwhelm people by being all over the place and often not making sense.

I get the point about conversations naturally splitting when there are 5+ people... definitely happens to me all the time. But I wonder if it’s really just about mental limits, or if there’s something else going on? Like, isn’t it also about the vibe of the group or the type of people involved? Some groups are just better at keeping everyone engaged no matter the size- like certain friend groups or teams that have great dynamics.

Also, not sure if the "four is magic" thing holds up everywhere... In my experience, some of the best conversations happen with just two people. Like really deep, meaningful stuff you can’t get with more people. And for bigger groups, there’s often this chaos energy that can be fun in its own way. Yeah, it’s not the same as an intimate chat, but it’s not worse, just different.

That said, I do like the idea that our brains are wired for certain sizes - makes sense when you think about the mental juggling it takes to track other people’s thoughts and reactions. And I love how Dunbar tied this to Shakespeare—kinda cool that he instinctively kept scenes small to avoid "cognitive overload." Makes me wonder if modern writers and creators even think about this stuff or just stumble into it.

So yeah, the mental limits idea is interesting, but I feel like the type of people, the setting, and even the purpose of the group matter a lot too. Sometimes it’s not the number of people, but how good they are at making everyone feel included... which is maybe a rarer skill than we think.

Sex is also better with 4 people, requires a bit of organization though.

I'm a D&D player, and run campaigns on a regular basis - and I've learned that for me, the upper limit for a party is 4 players.

It's nice to see a headline of an article referencing a study that backs up my personal preferences.

Maybe I should read the article - or dare I say, even the study? - and make sure it's not complete bullshit :P

This makes me wonder if our preference for group size might extend beyond conversation to other areas, like collaboration or even entertainment

A lot of TV shows have three presenters. The fourth is the viewer, I guess.

Pair programming is pretty popular. Has anyone tried "trio" or "quad" programming?

I've tried a few three person programming sessions with a live pycharm collaborative session (whatever they call it) - what I thought was interesting is we naturally divided responsibilities.

2 people would do more direct pair programming of one doing more thinking / design / instructing and the other is the workhorse and supplemental designer. The third would be the polisher and tester, and participate in design as well when able to.

Idk if there is a role-based analogue in larger social groups but interesting to think about.

mob programming is a thing. one person at the keyboard and a huge screen or a projector and everyone else discussing what to do. it works because the display is the focus. people might break into smaller groups to have discussions, but they are still focused on the task and potentially they discuss different aspects or ideas and present results to the group.

in the end it is like every classroom or presentation or group discussion. people are either engaged or they aren't. but breaking into smaller groups is not necessarily a problem in itself if everyone is still focused.

Conversations between four people very often devolve into two conversions between two people. The optimal number is either two or three.

4 people just means 3 people talking and me eating food.

Reply on news.ycombinator.com