Not a bad article at Wired
Tuesday, March 30, 2004
Wednesday, March 24, 2004
In some ways, I'm not very happy with the writing in this post, and some of the problems I've had writing it I'll put down to the awful problems I'm having with my back, which make it so hard to think clearly when I'm sitting down at the computer....
I've talked before (here and here) about how the structure of our language and thought can easily mislead our thinking; here's another example, derived from Richard Dawkin's excellent book The Selfish Gene.
It's known that most birds regulate their clutch size -- the number of eggs they lay -- and it seems as if they do it for the good of the species, so they don't overtax the available resources. Dawkins argues, however, that this seemingly altruistic behaviour is in fact better explained as the result of the workings of selfish genes.
How could this behaviour arise from selfish genes? If the selfish gene view is correct, shouldn't we be expecting the birds to be wanting to have as many eggs as they can, so their genes are propagated as much as possible? Why then would these birds be limiting the number of eggs they are having - why would they be practising birth control? It's hard to imagine!
Actually, it's not that hard to imagine if we frame the problem in a more appropriate way. It's hard to imagine because of the way I've put things. I don't mean to say I've used some sort of special language-trick, however; what I mean is that we can, if we are not careful, be subtly mislead by the language we are using. To see what I mean you'll have to read on and see the details.
So what have I done to misleadingly frame the issue? What I've done is exemplified by the use of the term birth control. But what is wrong with that? If birds are laying less eggs than they could, are they not practicing a form of birth control? Would we not say that if people were limiting the number of children they have they are not practicing a form of birth control?
Perhaps we can clarify the problem by considering what we mean by 'birth control'. I think the essence the term's meaning can be captured by the following: birth control is limiting the number of children born; birth control controls (limits) the population; if you want to go about limiting the population you need to practicse birth control. It's hard to think how this meaning for birth control could be wrong, but it is -- in fact, as Dawkins explains, birth control can have exactly the opposite effect!
Dawkins says (pg 116):
That is, birds are practicing birth control in order to increase the number of children they have!
[Each] selfish individual [chooses] the clutch size that maximizes the number of children she rears. If three is the optimum clutch size for swifts, what this means [is] that any individual who tries to rear four will probably end up with fewer children than rivial, more cautious individuals who only try to rear three. The obvious reason for this would be that food is so thinly spread between the four babies that few of them survive to adulthood.
Lets now consider how and why the language has been getting us confused. How can birth control limit the number of children and also increase the number of children? The answer: it depends on the context. In the case of humans, at least in countries with a high-standard of living, birth control will limit the number of children. This is because baies have a high chance of surviving into adulthood, so any child born will increase the population and any child that isn't won't. In the case of the birds, and no doubt many other creatures, however, child mortality may be high, with the chances of a child surviving being dependent upon how much time and effort the parent(s) can devote to it. And thus having more children may, because you can't care for any of them adequately, mean you end up with less surviving children.
That's an explanation of how it's possible for birth control to either limit or expand the number of children, but how was it that we got confused in the first place? It was because the problem was framed in terms of number of children born, that this was simply the causal factor that effected population size, and that an increase in births meant an increase in population size (or, put the other way around, that an increase in population size comes from an increase in births). If we'd been careful about what we were thinking, we would have realised that we aren't really interested in the number of births but the number of children that survive. We'd assumed that the two meant the same thing.
We assumed the two meant the same thing because we didn't really think about the relationship between number of children born and population size. If we give the issue a quick think over in terms of the concepts such as 'giving birth to children' and 'population size', it's seems quite natural that there's a direct causal relationship between them, and if you take this answer it's hard to look at it and see why it is wrong. But there's a big difference about thinking about the concepts and really thinking about what the situation could be like (thinking 'what if the mother had twenty eggs rather than three?'). If we'd really thought it through, it ought to have been easy to see that having more children born could mean less children surviving.
I was originally planning to have a second part to this post, where I would discuss where our thinking had been going wrong, in terms of some vocabulary I've been developing for talking about this kind of stuff. But now I'm going to do that in a separate post which I'm going to start writing next.
Thursday, March 18, 2004
I've been reading Richard Dawkin's book about evolution, The Selfish Gene. It's apparent that a lot of ideas about evolution are tested out through computer simulations. Of course, it's not just evolutionary theory where computer simulations are heavily used - there's other areas such as Economics, Physics and Meteorology. But despite the fairly widespread use of simulation, despite its utility, may people seem to consider it somehow lesser than other techniques for understanding the world.
Many people would consider verbal or mathematical means of understanding the world superior to simulations. I believe the argument is that simulations seem so paltry against the complexity of the real world, with the assumption that the other techniques are much better at dealing the nature of the world, perhaps because they don't try explicitly representing it. Against this idea, it occurred to me before -- and this probably has occurred to others -- that all thinking is fundamentally simulation.
Any reasoning about the world -- verbal, mathematical, etc -- involves some conception of the way the world is. In short, when you're reasoning you're manipulating this model in some way in order to see what it can tell you. What does this have to say about the relative efficacy of a computer simulation? That's something I'll consider another time.
This post is really just a quick sketch - I want to come back to this and consider it more fully another time, but now it's beddy-bies time and I need my beauty sleep :-).
Sunday, March 07, 2004
A while back, I gave a glowing review of Harold McGee's The Curious Cook. This post is about a single sentence from that book, which I'm using to illustrate a facet of the way our minds seem to work. This post doesn't provide a complete argument - at the moment I'm just trying to collect examples in order to build a fuller case, one that I'm hoping to present in the future. And you might find its conclusion rather lightweight - but after all, it's only a single, small example...
In the book's article on making mayonnaise, he explains how a microwave can be used to kill the salmonella that can sometimes be found in eggs. A microwave is suitable for the task because it can kill the bacteria without destroying the "rawness" of the egg yolk, which is important for making mayonnaise. It can do so because it can cook yolks at precise "temperatures" for a precise duration of time.
"As for the cooking itself, after years of owning a microwave oven, I finally found a task that it alone can accomplish well. On the stovetop, it's very difficult to ..." (pg 99)
It's easy to nod along with what he's saying, and I think it gives us a glimpse of something interesting about the way our minds work to see why he's not really right. When you think about it, a task that microwaves alone can accomplish is the task of heating something up quickly. That's not to say the results are always fantastic, but if it's speed you're after, a microwave is your best bet.
That microwaves can heat things up quickly is a rather obvious fact, and yet you may have noticed that its doesn't come obviously to mind when you read the quoted sentence. Why is that? I think the reason is that our minds have a bias towards thinking in terms of physical entities over more abstract properties such as the duration of something. "Cooking something fast" is definitely valid as a task a microwave can alone do well, but it's far less obvious a task than something physical such as "frying something", "baking a cake" or "making rice".
There's an objection to my argument that you may have found yourself thinking. That is, that when he said "I finally found a task that it alone can accomplish well" he was only talking about physical tasks such as "frying something", "baking a cake" or "making rice". I can't say this wasn't the case, but even if it was, I don't think it really says anything against the point I'm making.
If he intended the sentence to refer only to physical tasks, then perhaps it's illustrative that he didn't feel the need to qualify himself, despite the raher obvious thing microwaves do well (cooking things quickly), even if you don't consider that a task. And regardless of how he intended it, it's still illustrative if people still read that sentence without realising the possibility of a task being "cooking quickly".
People will often reject a notion on the grounds that it's "ridiculous". I want to eventually write a longer post about this, but for the moment I'm just collecting examples. Tim Cahill provides one in his book Remote Journies Oddly Rendered. The particular story in the book recounts the time he camped in Yellowstone National Park near a set of remote geysers, and it notes:
Early white explorers sent back accounts of the boiling water and steam erupting from the earth. One eastern publication, in 1870, rejected what now seems like a solid descriptive manuscript with the words, `Sorry, we do not print fiction.’ And it was in 1870 that men who wondered at the supposedly fictional geysers first expressed the thought that the entire areas would be set aside as a National park. (1st ed, pg 200. Note that in the US the book seems to be called "Pass the Butterworms: Remote Journeys Oddly Rendered").