The local citizens had problems pronouncing such an Anglo-Saxon name, so over time it dropped the “l” and added a “b” and became Ambattan Bridge. But that was not its name. Just the way the locals pronounced Hamilton.

Along came a new officious official. “What’s that bridge called?” “Ambattan Bridge, sir”. “Ambattan, what’s that?” “It means barber, sir” [It so happened that in the mangling of Hamilton, the locals produced a sound which was close to the Tamil for "barber"].

“well, it’s not going to be called Ambattan, no dashed native names here, get it renamed Barber’s Bridge immediately!”

His wish was their command. So Hamilton became Barber.

Just to point out that lawyers thought they had this covered in the fifteenth century, according to David Crystal :-)

At the time legal language in the UK was moving from Latin and French to English because hardly anybody spoke French at home any more, even the nobility of Norman origin and not even the clergy spoke Latin in conversation except to show off. However, to maintain continuity they simply added the new English term to the existing French or Latin word in the description of common concepts.

Thus we still say “will and testament” when both words mean the same thing – they just come from different source languages (English and Latin). Similarly “goods and chattels” (English and French), “fit and proper” (English and French) and so on.

As an attempt at future-proofing the meaning of legal documents it wasn’t entirely successful as it really just created a corpus of difficult-to-read technical gobbledygook. But full marks for being aware of the problem.

Even in the example given above, where birds learnt to peck bottle tops to get at milk, relies upon our subjective judgement that any two birds share ‘the same idea’. This is extremely hard to do, due to the fact that, to date, our most compelling evidence for identity of ideas comes from introspection. From introspection, we realise that attention, recognition and understanding all contribute to the ‘refactoring’ of ideas as in the example I gave above where someone made a complex joke involving several contemporary ideas.

Another example is given by asking if when two people use the same word, they are necessarily referring to the same idea. Or, as was once famously argued, to ask on what basis my using a given word today might mean that I used it in the same sense yesterday. I.e. that ‘in my understanding’ or ‘in my mind’ the word referred to the same idea today as it did yesterday.

To conclude: the bird example tempts to believe something that it is hard to argue for in a sustainable way due to problems such as those illustrated above, as well as Nic’s groups example.

Patents are attempts to put artificial hard boundaries around ideas, by using language as a tool. This is more successful when

(a) the language used is clear and has a relatively stable widely agreed meaning.
(b) this is also true for the idea, such that if it gets refactored, the heritage of ‘derived ideas’ is extremely clear (i.e the. child idea is in breach of the patent of the parent idea)

Opponents of the patent system argue that (a) and (b):

- are difficult to achieve,

- rely upon the questionable statement that two ideas
I and J are the same if and only if their descriptions
in language, D(I) and D(J), are the same across time
and space (eg description created London in 1920,
vs in New York in 2006)

- rule out complex chains of creation involving
many sources over time, as JP argued

- encourage artificially broad descriptions of ideas
(which can in turn cause an extreme reaction
eg descriptions that are so narrow that the patentor
gains nothing)

- are expensive and arbitrary to legislate, as many
have complained

- favour vested interests over innovators, as was
argued above, with alternatives being equally
interventionist and unsuccessful

- and are counterproductive because they discourage
innovation based on subtle refactorings ‘close to the
idea’. By describing the idea such that it cannot be
redescribed, you effectively kill it.

The last thing is really interesting. Imagine if you could patent mathematical proofs, instead of, as is current practice using attribution which is much better it seems to me. Then, ipso facto, a proof B that was ‘too close to’ a previous proof A, would break the patent. This would stop people from improving proofs which is one way that maths progresses. Ho hum.

So to summarise, the ‘memetic assumption’ that you can wholly describe ideas as if they were genes in some pseudo-Darwinian system, appears to me to underly much of what is wrong with patents.

It seems to me that applying the concept to groups (or even humans these days) is tough because of a combination of the following:
1) the goals of groups are multi-faceted and stretch beyond survival
2) groups morph more quickly and unpredictably than living organisms

The spread of the idea was certainly too rapid for any genetic or generational influence to have been the vector. Whether the idea spread by social learning from one Einstein of the bird population or was simultaneously invented in a number of discrete events is really not important.

I think there is a good reason for protecting a small organisation’s ability to operate fairly in the market. Otherwise large corporations would steal their ideas and undercut them until they were out of business. Copy protection and electronic DRM isn’t the right way to do this though.

A great example of this is the idea of a JOKE. This is closely related to ‘fair use’ and comment issues in copyright. To illustrate this point, please let me link to this highly apposite blog post by Tim Malbon:

http://staff.interesource.com/tim/aug2006/slithr.htm

This post is great because, to ‘get it’, you have to follow Tim’s refactoring of a number of ideas (call them memes if you insist), which simultaneously forces you to understand their ancestry (note genetic analogy), and yet appreciate that Tim is creating something new.

As Dave said, you shouldn’t think economic/technology/science evolution as analogous to biological evolution. Instead, you should understand the general evolutionary algorithm, will all its complexities and game-theoretic twists…

If you will decide to read it, please write your thoughts about it :)