Thursday, November 11, 2010

The Secreted Life of Bees

Responding to Thomas Seeley:

By anthropomorphizing bees you are adding your own interpretations to what they do, and then making these interpretations the basis for their decision-making activities. This not only misrepresents bees, it creates a false picture of how bees (and networks in general) make decisions.

> 1. Remind the group's members of their shared interests and foster mutual respect...  There are no clashing curmudgeons in a bee swarm.


I doubt that bees are capable of a higher-order function such as mutual respect. Certainly bees disagree with each other as each bee that finds a potential new home will advocate for that home. Any sense of 'shared interest' is purely implicit - a bee is not capable of comprehending shared interests.

No individual bee has a sense of what is in the hive or swarm's best interests. Each bee manages its own little bit of it, and the hive's interest is represented through the interactions among the bees. It is therefore not necessary to have a sense of shared vision or responsibility for community interests.

> 2. Explore diverse solutions to the problem, to maximize the group's likelihood of uncovering an excellent option. The scout bees search far and wide to discover a broad assortment of possible living quarters.


Each bee explores one solution to the problem. It is only when viewed as a a collective that we see an exploration of diverse solutions to the problem. This is important, because it means that in a hive diversity is represented by the bees themselves being different from each other, and not being the same by embracing the same range of diversity.

> 3. Aggregate the group's knowledge through a frank debate. Use the power of a fair and open competition to distinguish good options from bad ones. The scout bees rely on a turbulent debate among groups supporting different options to identify a winner. Whichever group first attracts sufficient supporters wins the debate.

Bees don't debate. They simply present what they know. A bee will present more or less vigorously depending on the suitability of the home. There is no sense on the part of the bee of being 'in competition' with the other bees. There's no sense of 'winning the debate' - what happens is that the best new home is selected, not that some bees convinced the other bees.

> 4. Minimize the leader's influence on the group's thinking. By functioning as an impartial moderator rather than a proselytizing boss, a leader enables his group to use its combined knowledge and brainpower. The scout bees have no dominating leader and so can take a broad and deep look at their options.

Bees do not have leaders and there is no such thing as 'moderation' in bee debate. There is no sense of 'enabling his group' to use its knowledge and brainpower. Queen bees (not even remotely a 'his') function only as the reproductive element of a hive. Again, the hive as a whole - and not individual scouts - take a look at the different options.

> 5. Balance interdependence (information sharing) and independence (absence of peer pressure) among the group's members. Only if ideas are shared publicly but evaluated privately will the group be good at exploring its options and making good choices. Scout bees share freely the news of their finds, but each one makes her own, independent decision of whether or not to support a site.


There is no such balance in a beehive. There is no peer pressure. Bees are completely independent from the perspective of how to behave, but completely interdependent from the perspective of producing the resources needed to survive.

---

The reason why these distinctions are important is that the author, by anthropomorphizing the hive behaviour, introduces the idea that the correct decision is the result somehow of appropriate group management on the part of the bees (or the putative 'head bee' who would ensure that all of these pieces of advice are followed) rather than built-in as structural components of the networks.

What is remarkable about bees - and similar sorts of network behaviours - is that they reach the correct conclusion with no cognitive activities at all. They don't need to be mindful of respecting the others group dynamics, or choosing from multiple options. The hive functions best when each bee attends to its own business.

Indeed, it is the element of persuasion, leadership, and group dynamics that introduces the likelihood of error into the mix. If a bee becomes attached to its own choice, and begins to lobby for it, possibly hoping to be selected head bee and to be rewarded with special privileges, the the hive is more likely to be persuaded by the most personable bee, instead of the best choice.

In human society, if there is a lesson to be drawn from this, it is that the leader should play no role in leading the group at all. The only behaviour expected from the leader is to introduce new workers or drones into the mix. The workplace ought to be structured such that there is no extra reward for convincing people, no promotion to be obtained by subverting the group. It's not about competition at all, it's about cooperating - each person playing his or her own role and each benefiting from the success of the group as a whole.

3 comments:

  1. As a long-time beekeeper, I really appreciate your insight on this topic - now I need to see how I can apply your observations to my own business!

    ReplyDelete
  2. "...the leader should play no role in leading the group at all." That's fine if everything were figured out and immutably true but we know that in human societies that just isn't so.

    ReplyDelete
  3. Why do things have to be figured out and immutably true?

    Bees have to figure things out. There;s a great deal of uncertainty facing a hive. Where is there honey? How best to get to it? Is the bee approaching friend or foe? How many workers should there be? Where should the next hive be located?

    The point is, the hive as a whole figures these things out. The hive doesn't defer them to some head bee, who is somehow in possession of privileged knowledge. Through a distributed decision-making process, the hive can do without a leader. But it *still* makes decisions.

    It's the same with human society. We tend to think that we need a leader in order to make decisions and figure things out, But we don't. We can do this on our own, without a leader.

    ReplyDelete

I welcome your comments - I'm really sorry about the moderation, but Google's filters are basically ineffective.