Archive for October, 2014

Progress(?) in philosophy: the Gettier case

Wednesday, October 15th, 2014

I’m of two minds about this article by Massimo Pigliucci. While I continue to be dismissive of the claim that there is anything of substance in the Gettier examples I can agree that there may be progress in the game of clearly expressing why that is the case. But the role of philosophers in advancing that progress is often more obstructive than constructive.

My own initial response to the Gettier problems was basically what Pigliucci refers to as the false premise objection – namely that the claimed “justification” for the belief in the allegedly problematic cases may be make the belief “excusable” or “blameless” but is not justification in the intended (logical) sense because it is based on a false premise. And my respect for the discipline is not enhanced by the proposed example of “more sophisticated Gettier cases that do not seem to depend on false premises”.

What Pigliucci proposes is as follows:

I am walking through Central Park and I see a dog in the distance. I instantly form the belief that there is a dog in the park. This belief is justified by direct observation. It is also true, because as it happens there really is a dog in the park. Problem is, it’s not the one I saw! The latter was, in fact, a robotic dog unleashed by members of the engineering team from Bronx High School. So my belief is justified (it was formed by normally reliable visual inspection), true (there is indeed a dog in the park), and arrived at without relying on any false premise [2].

But here the “justification” clearly involves the false premise that what looks like a dog is a dog. Without that premise, any claim of justification “by direct observation” is just clearly nonsense.

What is “Natural Variability”?

Wednesday, October 15th, 2014

Briggs is right to complain that “natural variability” is an ambiguous and easily abused term, but what I would be most inclined to use it for is different from either of the usages that he identifies.

He notes that some use “natural variability” of a phenomenon (such as some average of temperature measurements) to refer to the actual values taken by the data at different points in time, and others use it for the values that would be expected in the absence of some “unnatural” factor (such as CO2 emissions from human use of technology). But to me it seems much more natural to use it to refer to the unexplained deviations of the data from what would be predicted by a (partially) explanatory model.

I misunderstood Briggs’ claim that theoretical and/or statistical modellers claim to “skillfully” predict the natural variability in his first sense as meaning that they claim to predict it completely or accurately, whereas he was referring to the technical definition used in meteorology where one prediction is said to be relatively skillful compared to another if its mean squared deviation from the observed data is less. But this depends both on the reference model used for comparison and on the interval over which the comparison is made. A model that is skillful over a long interval may well have substantial shorter intervals over which it is not skillful, and even though a prediction of an upward trend in global temperature may appear not to be skillful over the interval from 2008 to 2014, that made by Arrhenius in 1898 does seem to be so (and would be even more so if he had predicted a faster rate of increase by reducing his estimated doubling time for CO2 to account for the subsequent increase in both population and per capita energy use).

Misreading Statistics

Saturday, October 11th, 2014

Briggs, points out a very important real issue here – though I wouldn’t call the confusion between group and individual differences a matter of “exaggeration” exactly.

No matter how carefully one tries to express a claim about population differences, the risk of feeding prejudice about individuals is always substantial – so much so that I think there may be many true statements that would be best left unsaid.

OneTab shared tabs for 2014-10-06

Monday, October 6th, 2014

OneTab shared tabs.

The Story of a Disaster

Sunday, October 5th, 2014

On reading the comments on this old Tyee article I was struck by one particular exchange which follows a pattern that is sadly all too common.

When commenter Chris Abel made the claim that “NOBODY was harmed by radiations at Fukushima” the response by G West included a number of links which were presumably intended to convey the impression that in fact MANY were harmed “by radiation”. But if one reads them it becomes clear that the claims of actual and prospective medical harms due to radiation, (while obviously non-zero if one includes delayed rather than just immediate effects), are really very modest (and essentially negligible in comparison to the number of deaths and injuries caused by the tsunami itself)

For example the first link is to a report from Physicians for Social Responsibility which, while striking me as somewhat alarmist, does stick reasonably close to actual facts and so does not make any claims as to the actual number of expected morbid or fatal outcomes.

The second link is to the World Health Organization which reports that:

The WHO report ‘Health Risk Assessment from the nuclear accident after the 2011 Great East Japan Earthquake and Tsunami based on preliminary dose estimation’ noted, however, that the estimated risk for specific cancers in certain subsets of the population in Fukushima Prefecture has increased and, as such, it calls for long term continued monitoring and health screening for those people.

Experts estimated risks in the general population in Fukushima Prefecture, the rest of Japan and the rest of the world, plus the power plant and emergency workers that may have been exposed during the emergency phase response.

“The primary concern identified in this report is related to specific cancer risks linked to particular locations and demographic factors,” says Dr Maria Neira, WHO Director for Public Health and Environment. “A breakdown of data, based on age, gender and proximity to the nuclear plant, does show a higher cancer risk for those located in the most contaminated parts. Outside these parts – even in locations inside Fukushima Prefecture – no observable increases in cancer incidence are expected.

In terms of specific cancers, for people in the most contaminated location, the estimated increased risks over what would normally be expected are:

all solid cancers – around 4% in females exposed as infants;
breast cancer – around 6% in females exposed as infants;
leukaemia – around 7% in males exposed as infants;
thyroid cancer – up to 70% in females exposed as infants (the normally expected risk of thyroid cancer in females over lifetime is 0.75% and the additional lifetime risk assessed for females exposed as infants in the most affected location is 0.50%).
For people in the second most contaminated location of Fukushima Prefecture, the estimated risks are approximately one-half of those in the location with the highest doses.

The report also references a section to the special case of the emergency workers inside the Fukushima NPP. Around two-thirds of emergency workers are estimated to have cancer risks in line with the general population, while one-third is estimated to have an increased risk.

The almost-200-page document further notes that the radiation doses from the damaged nuclear power plant are not expected to cause an increase in the incidence of miscarriages, stillbirths and other physical and mental conditions that can affect babies born after the accident.

For the benefit of any readers who can’t read, the “up to 70%” increase (in usually non-fatal thyroid cancers) does NOT mean that 70% of those exposed (in the worst area) will get the cancer but that the number of cancers might increase by 70% of what it was previously – ie from about 1.5 in 200 to 2.5 in 200 – and the increases in other cancers are all by less than 10% of the background rate. So the actual number of expected extra cancers is indeed quite small.

But poor Mr Christian Abel apparently didn’t bother to follow the links, and just like many other readers assumed the WHO supported G West in contradicting him – and so resorted to foolishly dismissing them without realizing that they essentially supported his assertion (even to the extent that they were criticized in another of G West’s links)

G West’s third link is to the Health Physics Society which in turn links to many useful sources – most of which are consistent with the assertion by Robert Gale in their panel discussion of the event that (with regard to probable increases in the lifetime cancer rate over Japan’s pre-Fukushima rate which was about 50%) “You can see that these are incredibly small increases that would never be detectable, especially in light of a very steeply increasing
incidence in cancer deaths in Japan over the last 60 years.”

Next is an article which asserts that:

In theory there is a possibility of cancer among people exposed in the accident at the Fukushima Daiichi NPP. Assuming the LNT model represents the reality of radiation-induced cancer at low doses, however, significant excess risk due to exposure is unlikely to be detected for the emergency workers and the public living around the site unless their doses have been seriously underestimated

and concludes that:

In the Fukushima accident, no acute radiation injuries have been observed even among people associated with the operation of the plant or responding to the accident in contrast to the Chernobyl accident where a number of people suffered acute radiation injuries. The anxiety among most of the civilian population is the future increase in the possibility of tumorigenesis.

West’s fifth link is to the rebuttal of WHO that I mentioned above (about which he helpfully says “And I suppose you thing these guys are biased too?” as if to identify them as even more mainstream than the rest when in fact, whether right or wrong, they are by far the most extreme in their assessment of the likely harm.

Finally, the last link is to the Science Daily report of a Stanford University study which unfortunately does set up the straw man claim that “There are groups of people who have said there would be no effects”, but the effects it does claim are really quite modest (even at the high end of a very wide range).

Radiation from Japan’s Fukushima Daiichi nuclear disaster may eventually cause anywhere from 15 to 1,300 deaths and from 24 to 2,500 cases of cancer, mostly in Japan, Stanford researchers have calculated.

The numbers are in addition to the roughly 600 deaths caused by the evacuation of the area surrounding the nuclear plant directly after the March 2011 earthquake, tsunami and meltdown.

So the expected number of cancer deaths is anywhere from many times less to maybe a couple of times more than the number of people killed by the decision to evacuate. (Not to mention the almost 20,000 immediate fatalities resulting from the tsunami itself and the totally ignored number of cancers etc that may result from other kinds of pollution caused by the destruction of various toxic chemical repositories!)