Tag Archives: groups

Lone geniuses and wise crowds

After a decade that has worshipped the ‘crowd’, and the infinite capacities of the collective intelligence, the tide seems to be rolling back as people once more begin to suspect the fallibility of ‘Groupthink’ and once again idolize the tyrannical creative genius.

'People don't know what they want until you show it to them'

This underlying dialogue affects everything – our attitudes to politics, the way we structure our organisations, even the art we admire as we fall in and out of love with auteurism.

On a more banal level, it impacts the way that we go about effecting tasks, and how we construct the goldfish bowl of collective actions – from conferences, to workshops, to committees and boards. (For the practical minded, I will end this post with some suggestions on how to apply this thinking to structuring working groups, if you are really busy, you can go straight there.)

The basic truth I am exploring is that different unit sizes of people are better at different things.

Without doubt technology has changed those dynamics, taking the theoretical concept of the ‘general will’ and hardwiring it, making group action possible at much greater scale and speed than ever before.

One of many beautiful visualizations of the creative community of the internet.

But humans remain substantially the same in their cognitive capabilities and emotional instincts. Which means that nothing foolproof can be created by lone geniuses, and nothing truly beautiful can be created by wise crowds.

So how do you know how many people you should get to do what?

1) Individual Brilliance

I spend a lot of the last year or more tirelessly advocating an open-minded an open-hearted sense of collaboration, and an openness to letting ideas or projects be strengthened and improved by exposure to as much external stimulus as possible. 99% of which I stand by.

But then I rediscovered Rousseau’s Confessions…and read for the first time Steinbeck’s incredible East of Eden:

“And this I believe: that the free, exploring mind of the individual human is the most valuable thing in the world. And this I would fight for: the freedom of the mind to take any direction it wishes, undirected. And this I must fight against: any idea, religion, or government which limits or destroys the individual. This is what I am and what I am about.”

This is an incredible book. Disturbing but compelling. And best read alone.

These both reminded me that any really original, well-developed and imaginative idea, a synthesis that has integrity and power, is developed in a period of solitude and meditation. The same even applies to blog posts.

2) Pairs Solve Problems

Budding philosophy students are sometimes puzzled to flip open Plato and discover not well-reasoned essays, but these rather strange little two-character plays, often featuring Socrates and some randomly selected, often very confused 4th Century BC Greek called Phaedo or Phaedrus or Parmenides.

The pairs are rarely evenly matched in intellect or moral clarity, but the dialectic celebrates the fact that two heads are better than one. Generally one of the partnership has the driving thesis, but by dealing with the confusion, agitation and indignation of their interlocutor, the thesis will gradually become better and better and rise to the level of genuine synthesis.

Dialectic - so powerful that even Watson can do it.

The same works for being a detective, or indeed for comedy. And as anyone who has ever spent the late hours of the evening doing a wall review of an important presentation with someone who simply refused to understand it, it works there too. Ultimately, pairs are great at solving problems, and if you want strategic solutions, putting two heads together is generally the best way to do it.

On the other hand, this is what happens when one smart person tries to do dialectic on their own, or with a hostile partner:

3) Triumvirates Create Friction

Ancient Rome at the close of the Republic was a fanatically warlike culture. As a result, it is deeply unsurprising that they opted twice for the triumvirate form of government, first with Caesar, Crassus and Pompey, and then much more finally with Octavius, Lepidus and Mark Anthony. Because nothing is more certain than that a triumvirate will end up fighting.

Three heads are fightier than one.

The competitive urge create by triumvirates, both for dominance and for each other’s exclusive attention, can briefly create a highly creative and expressive dynamic. If you are looking to open up a question, find some disruptive ideas, or get beyond obvious solutions, a trio left alone to fight for an hour might get you there. If you are looking to create a stable form of government, look elsewhere.

4) Fours Execute Missions

Four is my desert island group size, being both the size of my current team, but also the size of all the greatest bands in the history of music. It is also apparently the size on Navy Seal units.

A group of four is however generally considered highly unstable – which is not my experience at Jumptank, and hopefully is not true of Navy Seals, but obviously and spectacularly is true of most of the greatest bands in the history of music, who very often after a decade of intense brotherhood and success shift into a couple of subsequent decades not speaking to each other.

The tricky thing about fours is that they create a lot of opportunities for pairing off into cabals. But if this can be avoided, a four can be highly effective. There are two key rules to getting the most out of a four:

a) Make sure they have a clear, emotional shared mission. For the Seals, I guess this is easy – for bands it is harder because the mission is so often success…which means that when you achieve it, the band breaks up.

The Stone Roses - firm friends up until the moment they made a single penny of money

b) Make sure the group has a diverse group of skills and personalities. This is the whole strength of fours – it is probably the largest group where a diversity of skills can be effectively hammered together. When everyone has a unique contribution to make, the power of the whole can be incredible. When everyone is stepping on each other’s toes, it can spiral quickly. The Beatles worked great when there was a funny drummer, a thoughtful guitarist, a melodic bass player and an angry rhythm player. When they all wanted to be everything, their time was up. Whereas the Stones, where Charlie Watts is happy to be ‘the bed I lie on’ (Keith Richards) are still just about bumbling along fifty years in.

5) Democracy is Odd, not Even

Anything over 4 is basically hopeless for actually creating anything, but if you are making decisions, there is definitely some wisdom in numbers. Groups are good at criticizing and judging (as long as something is pretty well thought through already.)

One secret quirk of decision-making groups is that it can actually be pretty important to think about whether the size of a group is odd or even, if there is any chance that they might formally or informally need to vote on something. The group will want to act effectively as a group – which might lead to some consensus voting that could lead you in the wrong direction:

Never ask a band to make a 'majority rules' decision

6) 12 Apostles, not 12 Angry Men

Once you get to double figures, the group begins to lose the ability to even make decisions effectively. I mentioned this to someone recently, and they raised the test case of jury sizes – if a group of say 12 (also an even number) is so bad at making decisions, why has it been a key historical unit for judging?

My answer is this:

The point of having 12 people on a jury is not to make quick, effective decisions – but to maximize doubt. If that is your goal, by all means have a 12-person decision-making panel. If not, I would think 12 apostles – great for spreading the word, for going and doing likewise, but not a nimble and accurate decision-making group.

7) Dunbar’s Number: 100-230

The final unit size I’d like to consider is the ideal maximum size for a team, or a company. The key element to remember here is that you aren’t just herding cattle – you are trying to create highly effective and sustainable networks. The ideal number here is ‘Dunbar’s number’ – that a group shouldn’t exceed roughly 150, a number limited by the size of a human’s neocortex.

If it feels to you like a healthy community ought to be able to be bigger than this, that’s because you’re thinking about the wrong number. The question isn’t how many people are in the group, but how many relationships exist within the group, and therefore how well integrated it is. This increases exponentially, as below – so 150 people actually means 11,175 effective working relationships to maintain. Ow.

The number is actually supposed to vary a bit with intelligence levels, meaning you can stretch the effective group to about 230 in academic circles for example. Though I personally am yet to see a cohesive unit of 230 academics.

8) The Wisdom of Crowds

Many people have already written brilliantly on this subject, both on ‘community crowds’ (eg very large groups of programmers working on improvements to open source software) and the intuitive capabilities of ‘real crowds’ (ie randomly selected, much better at Who Wants to be a Millionaire than your best friend, but occasionally spectacularly wrong.)

My only contribution would be that real crowds are good for research; and community crowds, whilst connected to a larger effort by software platforms, tend to work in practice in ones, or twos, or sometimes, very briefly, in triumvirates.



1. Give well-stimulated individuals the scope to create and craft ideas.

2. Encourage dialectic pairs to improve and substantiate those ideas.

3. Allow threes to get a couple of disruptive new ideas on the table.

4. Give a well-balanced four the mission of bringing the ideas to fruition.

5. Get a group of 7 or a 9 to sift and choose the best ideas.

6. Use the broader group (12, 14, 18…?) to go forth and spread the word.

7. Don’t get a team of any larger than 150 to execute anything, ever.

8. Test your solution against the crowd – don’t expect them to design it.

1 Comment

Filed under Uncategorized