Items that are hopefully Likeable, Enlightening, or at least Nuts.
Eric Hoffer was an interesting man – a self-educated philosopher who worked days as a migrant laborer and longshoreman. With that unorthodox background came some unorthodox views – Hoffer disliked socialism, supported unions, and thought automation was A-OK, since it would give him more time to think and write books.
Hoffer’s earliest books may have been his best – I particularly like The True Believer, in which Hoffer applied his aphoristic style to understanding mass movements. His thesis – not fully original but still well-argued – was that mass or “populist” movements are all cut from the same bolt of cloth. A disenfranchised membership and extreme views are common features, with details of philosophy or dogma coming as an afterthought. Written in early 1950’s, The True Believer is chock-full of examples showing the similarities between movements ranging from offshoot religions, to fascism, to communism. If you’ve wondered about the extreme but random belief systems of current populist movements, you might enjoy this short and interesting book. Hoffer never met an aphorism he didn’t like, so while the topic isn’t light, the style usually is.
Ian Kershaw’s To Hell and Back: Europe 1914-1949 has a well-established theme – that the Second World War served to finish what the First World War started. However, by focusing on events between the two wars – he covers WWII in a scintillating four pages – Kershaw establishes that Europe was never really at peace during this period. 1914-1949 was less a period bracketed by war, than a period of one long war using shifting modes and weaponry. I thought To Hell was nicely executed, and a compelling indictment of isolationist policies.
I have only read a review of Cordelia Fine’s Testosterone Rex: Myths of Sex, Science, and Society, but several items have already caught my eye. Fine, using slightly different language, argues that the analytics used to establish gender properties are suspect.
In some cases, researchers have simply assigned properties to the gender in aggregate, when the range within a gender is greater than the difference between the genders – essentially, reinforcing stereotypes.
What really caught my attention was a discussion about males assuming more risk than females. Fine argues that the questions researchers asked about risk were really questions about whether men and women are attracted to male-dominated risk – e.g. playing sports, skydiving, and riding motorcycles. So the answers, duly processed and reported, end up being about the games men like to play, rather than about whether which gender is more risk-prone.
Ironically, I wrote yesterday (before seeing the review) about the seminal importance of properly framing questions and interpreting answers – for if the questions and interpretations are biased, the analytics in the middle won’t help us. Indeed, analytics can makes things worse: we’ve now invested in the wrong answer to a poor question, further biasing us against additional inquiry. Fine seems to have stepped back from the data mechanics, and it looks like the result has been some intriguing insights – bravo.
All this talk about analytics reminds me: Maybe Samuel Clemens was right and statistics is the art of lying with the aid of numbers. OK, but it takes a probability to assure that no one will know which lie we’re really telling. Those 2016 elect-o-meters graphics still hurt, particularly if you’re a Clinton supporter.
If statistical thought processes can seem unnecessarily complex – and therefore error-prone – I’m inclined to agree. Have some of our mathematical statistical formulations become historical artifacts? I’ve increasingly wondered – many of these formulations were crafted when computing power was more limited, so approximate statistical models were necessary. But it’s interesting that R.A. Fisher – rightly cited as the father of modern statistics – would use simple, intuitive, and straightforward argumentation when he could. The well-crafted (if not ideally named) “Statistics for Experimenters” https://www.amazon.com/Statistics-Experimenters-Design-Innovation-Discovery/dp/0471718130 by George Box, J. Stuart Hunter, and William G. Hunter illustrates this with several examples. (Disclaimer: the authors don’t make the explicit point I’m making, so far as I’m aware.)
I’ve had several recent discussions with friends about Ph.D programs, newly-minted Ph.Ds (particularly in data-related fields), and the integration of these skill sets into industrial settings. Ph.Ds normally come equipped with excellent skills, and a specialized expertise. But specialized expertise is perhaps a secondary point of Ph.D training – there are many roads to the goal of building up subject matter expertise. The original idea, and still the best application, of Ph.D training is to work on problems where there is no best practice, or maybe no practice at all. In short, to work on problems currently without solutions. Original research is a Ph.D. requirement, and learning to be unintimidated by the truly unsolved is what Ph.D’s bring to the party. It remains a challenge to assimilate those skills – most problems aren’t after all in the realm of the unknown, and even having the coolest and newest tool is different from knowing how to build an addition to the house.
Speaking of unknowns: if a frame completes the picture, what’s the picture that goes with this oddball gem?
A local art place has a spectacular array of framing samples, ranging from the sedate and normal, to a few even more far-fetched than this one. I had often looked at this frame and wondered what were they thinking? But then, when I brought in a Leonard Nimoy autograph for framing, I had my answer:
You know it’s true. There is a pink-glitter frame, however, that my art friends report lies unused and unfulfilled.
Have a good weekend.