The Logical Fallacies of Alasdair Thompson

 

 

If women are supposed to be less rational and more emotional at the beginning of our menstrual cycle when the female hormone is at its lowest level, then why isn't it logical to say that, in those few days, women behave the most like the way men behave all month long?

Gloria Steinem

Former President of Harvard University Larry Summers created controversy in 2005 when in a speech he suggested that the under-representation of women in science and engineering could be due to a "different availability of aptitude at the high end," and less to patterns of discrimination and socialization.  The subsequent uproar forced his eventual resignation.

New Zealand is having a superficially similar outcry over comments made by EMA CEO Alasdair Thompson.  Mr Thompson’s contention is that the pay equity gap between men and women is due to differing patterns of sick leave:  "Look at who takes the most sick leave," said Mr Thompson. "Because you know, once a month they have a sick problem. Not all women, but some do. They have children and they have to take leave off."

Before looking at the implications of Mr Thompson’s statement, let’s see if there is any evidential support for his assertion.

The evidence

We know there is a pay equity gap.  Pay equity is a different concept to pay equality, which is the requirement to pay equal pay for equal work regardless of gender, and is a legal requirement under the Equal Pay Act 1972.  Prior to April 1973 it was quite legal to pay a woman less for doing the same job as a man.  Despite this requirement, women’s average earnings in NZ in 2009 were still around 12% less than men’s average earnings.   Effectively women’s work is valued less than men’s i.e. there is gender based inequity in pay rates.

There is evidence that women on average take more sick leave than men.  Although there are no NZ wide statistics, there are some from the public service, which the PSA has obtained through an Official Information Act request.   On average women took 8.4 days per annum and men 6.8, but these figures don’t show how much overlap there is between the two populations.  

Brenda Pilott from the PSA stated of the difference:  “Given that the majority of responsibility for children and other family members falls on women, I’m surprised there isn’t a larger difference in the amount of days taken”.  So the PSA agree with the last part of Mr Thomson’s statement “They have children and they have to take leave off". 

He also stated that women take more sick leave because of “monthly sick problems”.  The National Business Review has cited a study published in American Economic Journal: Applied Economics in support:  “Using the personnel dataset of a large Italian bank, we show that the probability of an absence due to illness increases for females, relative to males, approximately 28 days after a previous illness.  This difference disappears for workers age 45 or older. We interpret this as evidence that the menstrual cycle raises female absenteeism.  Absences with a 28-day cycle explain a significant fraction of the male-female absenteeism gap.”

Non cause for cause

 

Cartoon from NZ Herald

Although he may have been factually correct about the difference in leave use he went further, drawing a causal relationship between these facts and pay equity;

Menstruation & family commitments cause higher sick leave use causes lower productivity causing lower average rates of pay.

Has he fallen into a logical fallacy non causa pro causa? 

The American Economic Journal study tried to model this and calculated that the earning cost associated with menstruation to women accounted for just under 12% of the earning gap between men and women.  This is a single study, and the result was arrived at by modelling – so it isn’t a compelling conclusion.  It provides both some support for Mr Thompson’s statement and undermines it, because it states menstruation only accounts for 12% of the difference, not 100% of it.  At the very least his statement is a causal oversimplification, ascribing a single cause rather than taking account of all contributing factors.  And that’s just one of its many problems.

By focussing on an inalterable physical characteristic as the cause, rather than addressable social factors, Mr Thompson – in the words of Mai Chen in the New Zealand Herald “tacitly endorsed paying women less because they, unlike men, have periods which makes them less productive.”

 

The Department of Labour suggests the three main mechanisms for the gender pay difference are;

• the jobs women do; New Zealand has a relatively high level of concentration of women workers in female-dominated occupations, with 47% of women working in occupations where 80% or more of employees are women.  Women are more likely to be found at the bottom or middle of an organisation and find it difficult to move up into higher-level positions.
• the value put on women’s jobs; the skills and knowledge that women bring to the work in female-dominated occupations may not be recognised and therefore not valued appropriately in comparison to other jobs i.e. discriminatory stereotyping.
• the work arrangements and caring responsibilities; more women than men combine primary care giving with part-time work. This limits women’s access to better paying jobs and positions, since part-time work is more readily available in lower-paid occupations and positions.

There are studies showing evidence for unconscious discriminatory stereotyping – for example white male doctors being rated as more approachable and competent than equally-well performing women or minority doctors.

Perception through language

 

Another issue is his choice of words in describing menstruation as a monthly sick problem.  The daughter of Gichin Funakoshi, the founder of Karate in Japan, described how as a traditional gentleman her father refused to say certain words (socks being the example used).   Let’s charitably assume that Mr Thompson is of a similar ilk.  The euphemism he has chosen though is inappropriate; while symptoms such as nausea or queasiness can technically be described as sickness, the word also carries a connotation of disease and an undesirable or disordered condition.  This concept can be traced back to a number of primitive cultural traditions and is prevalent in Christian writing and texts from other religions, and it’s unfortunate that it continues to be perpetuated: naturalia non sunt turpia - what is natural is not dirty.

In her article Ms Chen focuses on what is important in an employee; to look at their ability, work ethic, and other productivity factors as an individual, and ignore irrelevant group stereotyping that may lead to the devaluing of otherwise superior employees - the antithesis of the generalisation inherent in Mr Thompson’s statement.

In conclusion

There is evidence that women take more sick leave than men, and weaker evidence that some of that is due to menstruation.  If you accept that second piece of evidence, then Mr Thompson is guilty of a single cause fallacy, oversimplifying an issue that has complex causes.  If you don’t accept that evidence then he has committed a false cause fallacy – assuming correlation implies causation.

Implying that because women have more sick leave they have lower productivity is a false inference, as there is no evidence that their productivity is lower, and if it is in specific cases then to assume it is for all women is a faulty generalisation leading to stereotyping.  In this instance is his thinking lazy, confused or devious?

Focusing on menstruation, a natural, gender specific phenomenon rather than social causes is an argument by innuendo – implying there is nothing we can do about the gender wage gap.  There is plenty of evidence that the gap has at least some cause in discriminatory stereotyping and in social norms.

His use of words with a negative connotation shows his prejudice, not only his use of the term “sick problem”, but his characterisation of time off to look after dependents as “they have children and they have to take leave off".  Most people in New Zealand would consider putting family before work a good and socially acceptable thing, as is perpetuating the species through having children.  Mr Thompson however implies at best it’s a bit of an inconvenience and at worst quite undesirable.

What makes Mr Thompson’s statements damaging is that the weight of the statement is proportional to the influence of the speaker; and as CEO of the EMA he is perceived as having influence with employers – the people who currently pay a premium for male employees.  Whatever the outcome from his statements, one thing we can say for certain – logic and reason were the losers on the day.

Climate Change?

There are no passengers on spaceship earth. We are all crew.  Marshall McLuhan

My main bugbear with the subject of human caused environmental change is the misleading arguments, bad science, emotion and dogma rather than rational evidence based approach used by many people on all sides of the debate.  This of course is common where science and politics intersect: where either science may be used as a pretext for belief based policy, or where belief is held up as being superior to science.  Science is based on reason, logic, and empirical scientific methodology; which is often counter-intuitive to the “common sense”, beliefs, superstitions and compliance with social networks we are programmed to give precedence to.

I was therefore interested to read the Guardian article headed “Climate change should be excluded from curriculum, says adviser”.   As it turns out he was presenting an argument, not on the validity of climate change, but instead on a principle as opposed to rule based approach to how the curriculum should be written, as well as concentrating on the basics of science.  I remember concentrating on the basics of science when I was at school, and don’t think it will do anything to reverse the decline in science graduates we are experiencing.  But that’s anecdotal - the research seems to suggest that the way you teach science rather than content is the most important factor.  However, I digress.

Ok, I know nothing about climate science – so where do I start?

I am told that there is a “scientific consensus” is that human generated climate change is real, and will cause many perhaps irreversible adverse changes to the environment.  The term adverse is used in a human centric way here; Venus demonstrates that planets can exist quite comfortably with runaway greenhouse gas effects.

I’m slightly uncomfortable with the idea of “scientific consensus” – science is surely based on evidence, not agreement?  This may be based on both scientific argument and the scientific method, but still represents people’s judgement and opinion – both notoriously fallible – even if they are masters of the particular field, and the judgement represents the majority viewpoint.   

It is often pointed out as a criticism of the concept that there are historical examples of it being wrong.  Let’s ignore Galileo, who is sometimes used, as he was a pioneer of the scientific method arguing against religious dogmatism.  Both the theories of continental drift and the bacterial cause of stomach ulcers are better examples.  However they both actually demonstrate the strength of scientific consensus – in the face of evidence the consensus will change.  Where there is a significant change from established science needed (as in both these cases) a high standard of evidence and documentation is required – or as Carl Sagan might say “extraordinary claims require extraordinary proof”.

Of course, while this might strengthen consensus as being a valid tool from a scientific standpoint, it does nothing to make it useful for creating public policy.

This is taken care of by the precautionary principle; that if an action or policy has a suspected risk of causing harm to the public or to the environment but there isn’t scientific certainty about the likelihood, size or nature of the harm, the burden of proof that it is not harmful falls on those taking the action.  In an absolute sense this is nonsense, as a complete lack of harm is impossible to prove, and if it had been used historically we may not have cars, cellphones, aspirin, electricity networks or the Nintendo Wii.  There are correlations between the level of education and nearsightedness in individuals, and industrial development and nearsightedness in populations.  Yes, under an absolute interpretation of the precautionary principle we should all go back to living in pre-literate rural communities.

But the principle is much more nuanced than that.  It is based partly on the old English common law of “duty of care” which requires you, when doing something which could harm others, to take precautions not to do so.  Most versions require some evidence of the likelihood and severity of harm.  Many allow for a consideration of the cost of precautionary measures.  A good analogy of the principle is found in obligations under health and safety legislation for both employees and employers, and in workplace hazard management practices.

While irreversible or highly damaging human caused climate change is a prediction – and therefore open to debate - human caused environmental degradation is incontrovertible.  We destroy natural habitats, we deplete resources, and we poison the soil, the air and water.   On a localised scale we can see the economic consequences of issues such as overfishing and agricultural land loss through soil salinity from irrigation; desertification; health, environmental and economic costs of industrial accidents such as Bhopal and Chernobyl, and large scale environmental destruction like the Aral Sea. 

  Formerly part of Aral Sea

The problem of extrapolating long term global consequences of this degradation is because it is part of a chaotic system.  The characteristics of this sort of system are that what happens is very dependant on the initial conditions, so unless you know those initial conditions in their entirety, you cannot predict the systems future behaviour with accuracy.  As scientists modify their predictive models based on evidential feedback they reduce the uncertainty around those initial conditions – which may or may not make them more accurate - and the models indicate long term global warming and consequent environmental change.

Natural variations in temperature and weather have occurred in the past and are believed to have had significant effects for human populations – from the ice age to the theory that drought caused the collapse of Mayan civilisation, so we can expect any human caused change to have consequences.  So we have a situation in which there is uncertainty, effects that for all practical purposes on a human timescale are irreversible, and consequences to us and future generations of those effects.  

 

Yet we continue down this path.  This is analogous to a situation called the “Tragedy of the Commons”, based on an example of medieval European common land on which each herder is entitled to let their cows graze.  It’s in each herder’s interest to put as many cows as possible onto the commons, even though it will lead to overgrazing, because the herder receives all the benefits from each additional cow, while the damage is shared right across the group.  If all the herders do this though eventually the commons will be destroyed, to the detriment of all.  This is summarised by Wikipedia as “a dilemma arising from the situation in which multiple individuals, acting independently and rationally consulting their own self-interest, will ultimately deplete a shared limited resource, even when it is clear that it is not in anyone's long-term interest for this to happen.”

We can see this analogy in these comments about bringing agriculture into New Zealand’s emissions trading scheme:  Imposing the emissions trading scheme on New Zealand agriculture would hurt the country's economy unless other countries were doing the same thing, says Fonterra Cooperative Group chairman Henry van der Heyden.  "Why do anything to the detriment of our competitive advantage?"

This is a classic commons argument.

There are two solutions to the commons dilemma.   The first is to privatise the resources.  This is proven to be effective, but it is unlikely that we will ever have (or want) Global Inc.  The second is to regulate, but as we’ve seen from Kyoto and other attempts, this seems equally unlikely.