Friday, March 31, 2006

Creationism Freud Evolution

Here's a good answer to Creationists: "dogs!"

Why? They clearly evolved from a common ancestor, we did it ourselves through artificial selection, and it was fast.

Here's a rebuttal. When I was studying psychology - an Open University module back in the 1970s at the University of Sussex - we had a lecturer who got pretty emotional about Freud. "This is not science!" she growled: she was one of the dying breed of behaviourists.

Like the creationists, I didn't buy her argument at all. Why? Because her scientific model was rubbish. It didn't even account for the phenomena ... denied that emotions, consciousness and inate psychological capabilities were real.

So you don't have to believe in scientific theories which don't account for the phenomena. Doesn't mean you have to believe anything else ... sometimes 'we just don't know' is the best answer. In fact, if you do believe in scientific theories which are not deep enough to account for the phenomena, you risk terrible mistakes. Behaviourism entailed scary child development practices.

However, the converse does apply. When theories do explain the phenomena, and make successful and counter-intuitive predictions, then you should believe the science and disbelieve contrary prejudices.

Sorry creationists, that's where your juggernaut leaves the road!

Sunday, March 26, 2006

Race Differences in Intelligence - what follows?

Professor Richard Lynn argues in his recent book, “Racial Differences in Intelligence” that the races differ markedly in IQ. This was reviewed in the previous post. Assuming his summary of the extensive research literature is correct, what, if any, are the implications?

A number of contributors, grouped around the website Vdare.com, seem to draw almost racist conclusions - from a Caucasian viewpoint. Contributors have seemed to suggest that higher-IQ peoples, East Asians and some Jews, should be differentially taxed to create a level playing field. And migration into the US from countries with lower mean IQs seems to worry many of the contributors.

The science, however, is neutral and does not at all force these unpleasant conclusions.

The current human race is the only one we have, and the one which has evolutionarily worked so far. The only strong case for altering it is to remove obvious bugs (genetic diseases), and even that forecloses the possibility that bugs can evolutionarily become features later .

Where do we draw the circle of empathy? Do we restrict it to our genetic kin group, so that everyone else is a rival who needs to be dealt with? Or do we draw it across the whole human race, as the great religions suggest? Marginal analysis suggests we attempt the latter - if we meet a stranger, we can either attempt to make them our friend or count them as an enemy. The former increases our sociality, delivering scale, division of labour and greater economic opportunities; the latter delivers impoverishment. Trying to make friends doesn’t always work, but it should be the default option. Stopping at contingent national boundaries is just plain silly from an ethical point of view (obviously not from the point of view of national interest groups, of course).

People have argued that races with low IQ cannot make democracy function, or can’t support a modern technological civilisation. But mean IQ is just an average - IQ within a population is normally distributed. Suppose we have a population of one million with mean IQ = 80 (Sub-Saharan Africans, according to Lynn). Then 500,000 will in fact have an IQ above 80, around 23,000 will have an IQ above 110 and more than 1,000 will have an IQ above 125. The statistics of the bell curve.

All societies show a grading of IQ with social status. To make democracy work, you need enough people to lead the institutions: mass competence is not necessary, otherwise democracy would work nowhere. Even with a base IQ of 80, there are enough smart people to make it work.

Another fact the IQ obsessives forget. What is so great about intellectuals? Would you depend on them for fighting wars, or caring for the sick, or for day-to-day running any large enterprise? In running complex human societies, as any recruiter knows, the personality dimensions of, say, the five factor model are as important as IQ, and high IQ correlates strongly with only the conceptual personality type (Myers-Briggs NT/NF, or ‘openness and conscientiousness’). So let’s have less about eugenically changing everyone into an intellectual - nothing would ever get done!

People worry too much about immigration. In the long-run it is determined by economics. People come for jobs and flow in or out depending on economic, social and legal conditions. These are everywhere subject to governmental policy, and most of these policies are skill and wealth-based, both proxies for IQ in the large.

It needs to be stated that today, we do not have a problem. In advanced countries we produce enough people with the right skills for all niches in our society and with more investment in education we could produce more skilled individuals if we wished. If different races are differentially stratified across job types, so what? Everyone is an individual and should find the jobs and mode of living which suit them. If in some remote future, the lowest IQ members of society could not economically function (they will be of all races, but not in equal proportions) we should leave that to the policy makers of the time - and it will be a people problem, not a race problem.

Saturday, March 25, 2006

"Race Differences in Intelligence: An Evolutionary Analysis"

Review of Professor Richard Lynn's new book also posted here.

It would be difficult to fail to discriminate between Pygmies, Swedes, Japanese, American Indians and Australian Aborigines. Wearing evolutionary spectacles, the physical differences are clear racial adaptations to different environmental conditions, with climatic adaptations being particularly evident. Why would levels of intelligence not also have been differentially selected for between races?

Chapter 1 defines ‘intelligence’ and motivates why IQ is a measure of it. Chapter 2 justifies the concept of ‘race’ - depressing that this is thought to be necessary - and makes the argument of the first paragraph above.

Chapters 3 - 12 then itemise in great detail the results of numerous intelligence tests given to nine racially-distinguished populations: Europeans; Sub-Saharan Africans; Bushmen (South Africa) and Pygmies; South Asians (Middle-East, India, Pakistan) and North Africans; Australian Aborigines; Pacific Islanders; East Asians (China, Japan); Arctic People and Native Americans.

Base-lining Europeans at IQ = 100, Sub-Saharan Africans come out at around 67. Corrected for poor environmental conditions, Lynn estimates the genotypic IQ (the mean IQ Africans would have if raised in the same environment as Europeans) as around 80. Conversely, East Asians seem to have IQs centred around 105 (p. 130) while some populations of Ashkenazim Jews have mean IQs between 107-115 (p. 94).

Chapters 13-17 summarise racial differences, and propose an explanation based on the geographic radiation of homo sapiens out of Africa, the resulting geographical isolation of sub-populations, and the impact of two ice-ages (the first from 70,000 to 50,000 years ago, and then the more severe Wurm glaciation, 28,000 to 10,000 years ago). These culled the less-intelligent in those racial groups most exposed to arctic conditions as well as driving the more obvious physiological adaptations. The East Asians were particularly stressed by harsh conditions north of the Himalayas and east of the Gobi Desert.

I have some quibbles. Lynn’s timeline of geographic dispersal makes no mention of the Toba event, 70-75,000 years ago, which was said to have created a genetic bottleneck. If today’s races emerged from a radiation out of east Africa which post-dated that event, how would that affect the argument?

Secondly, the early evolution within east Africa (p. 225) is poorly argued. The contemporary IQ of 67 is not the one to use, as it factors in present-day malnutrition. In the ‘environment of evolutionary adaptedness’, this would not have been the case, so why not use 80? Lynn then appears to suggest there is a ‘continual directional selection for intelligence’ based on its utility, as if species always got smarter and smarter. However intelligence comes with large costs, in terms of the energy required for big brains, so one would expect instead an equilibrium where a species is no smarter than it has to be. So rather than a drift to smartness in Africa, isn’t it more likely that we saw waves of replacement populations radiating from groups who got smarter in more isolated niches where they were stressed more?

This is not a coffee-table read. It is somewhere between a scientific book for the academic community and a popularisation. I think Lynn hopes to move the goal posts so that we can move onto some of the interesting consequential issues clearly identified in his research programme. If at times he seems to stray into IQ-reductionism, then this is probably symptomatic of the existing research community being currently below critical mass.

There may well be public policy implications of the research results aggregated, summarised and theorised in this book. But they are rightly not addressed here.

Wednesday, March 22, 2006

Time to roll out the Wagner

According to American vets from Iraq, interviewed on BBC's Newsnight last night, US troops in the Sunni North were regularly driving around shooting anyone they saw. Apparently they carried shovels which they could leave with the corpses, if challenged, to 'prove' they had shot insurgents digging-in bombs. It's all very 'Apocalypse Now'.

A related fact from the same programme. UK and Australian troops in Iraq and Afghanistan are not handing suspects over to the Americans, for fear of their mistreatment.

The ex-SAS soldier on the programme supplied the key. "The Americans have not made the transition from war-fighting to counter-insurgency." he said.

There seem to be a few more elements to add to the mix. The honour culture of the American South, from where so many recruits are drawn, which can flip instantly from warmth to a lethal response if the subject feels 'dissed'. I guess seeing your buddies blown up would count. Also a kind of US parochialism mixed in with naive patriotism and a deep sense of American superiority over all other cultures.

And then there are the usual factors in the Sunni north:
  • social distance between the troops and the locals
  • lack of a shared culture and language
  • dissimilar appearance and dress
  • the lethal insurgents are indistinguishable from the 'civilians'.
All very predictable, and all just the same in that other war.

Thursday, March 16, 2006

Special Cats

A tall, elegant woman speaking into her mobile phone.

"I'm worried about Gucci."

"Yes, they're special cats, a rare breed. I can't let them go out. They spend all day on our rather comfortable couches and playing in our deep-pile carpets, but I'm worried."

"No, Gucci has taken to vomiting and excreting on our bed. Imagine! I thought it was attention-seeking, but the vet said she's probably stressed."

"No, I think she's being bullied by Chanel ..."

Overheard at 'The Cottage Inn', Maiden's Green, near Bracknell while waiting for a business colleague to arrive for lunch, 15/3/06.

Wednesday, March 15, 2006

The martian anthropologist

"The martian anthropologist" is a mythical entity appealed to in the social sciences to achieve distance from human prejudices. If you (the 'martian') were shown a number of alien entities communicating via a public protocol, there would be several routes to analysis, including:

  1. Take the protocol, examine its lexicon and syntax, and try to derive a mathematical semantics for it.
  2. Take an anthropological, ecological view and ask what these alien entities were trying to achieve through their communication. Then do step one.

The second approach looks to have more potential, somehow, but most natural language research has chosen the first course, on the basis that it is easier.

Back in the beginning, Montague semantics translated words, phrases and sentences of English into a massively complicated logic (higher-order with numerous modal operators capturing temporal and intentional attitudes). The semantics were standard but not recognisable as the world you and I inhabit.

Situation semantics created a richer semantic model in which the denotation of propositions was not collapsed to mere truth values, but was taken to be 'complex collections of relationships between entities and attributes at a location in time'. But still, intentionality and agency were either missing or untheorised.

I have always thought that building models of agency in paradigms drawn from economics and games theory was likely to be more promising. After all, isn't evolution just applied microeconomics? The result will be mathematical models capable of providing semantic valuations for formalised languages, but they will also constitute theories of human-style social psychology.

This must be the next big thing after todays' evolutionary psychology, surely?

A related point. Religions are also a public protocol, dual to something more hidden in the ecology of groups and social psychology. As they say, it was no accident that the great monotheistic religions developed 2-3,000 years ago at the onset of the agrarian revolution (+ the zerotheistic Buddhism and philosophical Taoism).

The God of the Old Testament (shared by Jews, Christians and Muslims) is unquestionably a muscular, patriarchic entity bestriding a dominance hierarchy - the embodiment of the agrarian pre-state embodying rules rather than the arbitrary dictates of the current warlord.

Note: Islam was later refined through a nomadic honour culture, while the New Testament represented a transactional innovation, a subversive, but constrained ideology in the interstices of the Roman Empire (Crossan). It is unlikely that Jesus would have signed up to the later evolution of 'Christianity' as an arm of that very same empire.

The principles underlying Buddhism and 'philosophical Taoism' are also transactional. Is this anything to do with the proposed greater cooperation found in east-asiatic sociaties, based on selection pressures for these traits during the time of the last ice-age (c.f. Richard Lynn, here)?

Saturday, March 11, 2006

Net neutrality and rents in the value chain

Today's Economist (11th March 2006) makes the point that we should avoid knee-jerk reactions when it comes to regulating the Internet to ensure equal access ('net neutrality').

The issue is the desire of US carriers investing in core and access networks to charge content providers such as Google and Yahoo a premium rate for carrying their traffic - at a higher QoS. The carriers make the point that they are investing heavily in new fibre networks, especially in access, and it is difficult to cover these investments based solely on household monthly broadband subscriptions. There are moves in Congress to pre-emptively regulate against this.

At the moment, everyone who connects to the Internet gets a single 'best effort' service, but due to historic massive investment in Internet infrastructure, this 'best effort' is actually pretty good. "Too good", the carriers say, but nevertheless, how could they get any buyers for putatively 'superior' classes of service?

The carriers say they will not actively damage anyone's traffic as they introduce superior classes of service at various price-points. However, a superior service class has to buy something extra, so the most likely story is that the carriers will slowly permit utilisation levels on the Internet to rise until the resulting congestion separates out an increasingly tardy 'best-effort' experience from superior 'gold', 'silver' and 'bronze' services.

The implications of this go beyond the attempt to directly extract economic rents from content providers upstream in the value chain. After all, Google et al could choose to buy their Internet access from many other ISPs. AT&T and Verizon will have to modify the current Internet peering arrangements, charging their peers to carry ingress premium traffic: look out for the arrival of a settlements model.

Now, there is nothing wrong in principle with offering a portfolio of products at different price points - train operators also sell first and second class tickets. The problem is monopoly pricing - exploiting market power. The forthcoming North American duopoly rings a few warning bells.

The best antidote to monopolistic practices is competition rather than regulation. I hope those rumours about Google quietly buying up a lot of transcontinental dark fibre, and investing heavily in WiFi/WiMAX access networks turn out to be true.

Thursday, March 09, 2006

Joining the ranks of idiot users

I took my car into the local garage today - for the last week the front passenger window has not been working. Was the motor burned out or was the electric switch faulty? I didn't know.

Half an hour after I had delivered the car, the garage called. The window worked fine. Had I perhaps depressed the window-lock tab, they wanted to know.

On the driver's door ledge, the window-lock button is physically next to the electric door-lock, just in front of the two levers which raise/depress the front-passenger and driver windows.

I have to say I have not given one thought to the window lock mechanism since I bought the car in 2003. I vaguely believed it was there to stop the rear windows being used by kids, as on previous cars I have owned (in fact in my current car, the rear windows are manual).

"No", I confidently told the garage, "this has nothing to do with the windows lock - the switch fault must be intermittent".

Ando so with those words I have damned myself into the same category as all those computer users who believe that the CD drive is actually a plastic coffee cup holder. Trust me, it doesn't feel good.

Wednesday, March 01, 2006

A dying bird

Clare saw a bird die this morning. The Chaffinch flew down from the top of the garage, after a freezing cold night. About a foot off the grass, it just fell out of the air. She picked it up and carried it into the house. Although still warm, it was quite dead.

Apparently it is too beautiful to bury, so it is currently resting in the wood pyre in the garden.

Birds spend the night in the open, and rely on their stored energy to keep them sufficiently warm. A cold snap like the present one kills many.

I have been instructed to get up at the crack of dawn tomorrow and lavishly spread birdseed on the back lawn.