the flawed genius of William Playfair [book review]

David Bellhouse has written a new book on the history of statistics, focussing on William Playfair this time (following his fantastic book on Abraham de Moivre). The Flawed Genius of William Playfair (The Story of the Father of Statistical Graphics) got published a few months ago by the University of Toronto Press.

“[Playfair] was an ideas man whose ideas often did not come to fruition; or, when they did, they withered or exploded.” [p.121]

The impressions I retained from reading this detailed account of a perfect unknown (for me) are of a rather unpleasant, unappealing, unsuccessful, fame-seeking, inefficient, short-sighted, self-aggrandising,  bigoted, dishonest, man, running from debtors for most of his life, with jail episodes for bankruptcy, while trying to make a living from all sorts of doomed enterprises, short-lived blackmailing attempts, and mediocre books that did not sell to many. Similar to David Bellhouse’s colleague earlier wondering at the appeal of exposing such a rogue character, I am left with this lingering interrogation after finishing the book…

“[Richard] Price liked what Playfair had written. He found [in 1786] Playfair to be “agreeable” and “useful”.” [p.64]

Not that I did not enjoy reading it!, as it gives a most interesting of the era between the 18th and the 19th Centuries, in particular in its detailed narration of the first months of the French Revolution of 1789, and of the impact of the Industrial Revolution on economics and politics as the birth of capitalism. The book abounds in crossing lots of historical characters, like Richard Price (Bayes’s friend who published his most famous paper), Adam Smith (whose book Playfair reprinted with poor additions), Edward Gibbons (whose book along with Smith’s inspired the title of his Inquiry Into the Permanent Causes of the Decline and Fall of Powerful and Wealthy Nations), Thomas Malthus (competing for an annotated edition of Smith’s book), not to mention the political class of Britain at the time. David Bellhouse’s book demonstrates academic and historical excellence, constantly being very detailed, with a wealth of references, documents, and definite support for or against the rumours that accompany the life and deeds of Playfair. (Frankly, rarely a name has been that inappropriate!) This includes for instance the pictures pointing out to his first (?) forged signature [p.140] and the evacuation of the myth of Playfair as a spy for the British Crown—which the Wikipedia page happily reproduces, pointing out the need for an in-depth revision of said page. Similarly, the book delivered a convincing discussion of arguments for and mostly against Playfair “being the key player in the British operation to forge [French] assignats” towards destroying its economy. A lot of the book is touching upon the then novel issue of paper money, which Playfair only and negatively considered through his own (and catastrophic) experiences. At times, the book is almost too scholarly as it makes reading less fluid than was the case his Abraham de Moivre for instance. (And obviously less than in the contemporary Jonathan Strange & Mr. Norrel!)

It may be that my very relative lack of enthusiasm stems from the realisation that the story of Playfair is overall rather little connected with statistical inference, if not with descriptive statistics (albeit with a complete disregard for the quality and sources of his data), as when  publishing a Statistical Breviary on descriptive statistics for a series of countries (and surprisingly sold on Amazon!).  Or Statistical Account of the United States of America. And of course for his innovative graphical representations like the one represented on the cover of the book or the pie chart. I feel that the book is much more engaged in Playfair’s contributions to the then nascent science of economics, as for instant about the shallow and mostly misguided views of his’ on banking and running the economy, while conducting his personal finance and investments so disastrously that it negatively advertised against confidence in such views.

On a very personal level, I noticed that some graphs were provided by my friend and statistics historian Stephen Stigler [who also wrote a review of the book] while an analysis of the poor French involved in a coding scam of Playfair about Napoléon’s escape from Elba was by Christian Genest (whom I first met at a statistics conference dinner on the Lac de Neufchâtel in 1986).

[Disclaimer about potential self-plagiarism: this post or an edited version will eventually appear in my Books Review section in CHANCE. As appropriate for a book about Chance!]

 

futuristic statistical science [editorial]

This special issue of Statistical Science is devoted to the future of Bayesian computational statistics, from several perspectives. It involves a large group of researchers who contributed to collective articles, bringing their own perspectives and research interests into these surveys. Somewhat paradoxically, it starts with the past—and a conference on a Gold Coast beach. Martin, Frazier, and Robert first submitted a survey on the history of Bayesian computation, written after Gael Martin delivered a plenary lecture at Bayes on the Beach, a conference held in November 2017 in Surfers Paradise, Gold Coast, Queensland, and organised by Bayesian Research and Applications Group (BRAG), the Bayesian research group headed by Kerrie Mengersen at the Queensland University of Technology (QUT). Following a first round of reviews, this paper got split into two separate articles, Computing Bayes: From Then ‘Til Now , retracing some of the history of Bayesian computation, and Approximating Bayes in the 21st Century, which is both a survey and a prospective on the directions and trends of approximate Bayesian approaches (and not solely ABC). At this point, Sonia Petrone, editor of Statistical Science, suggested we had a special issue on the whole issue of trends of interest and promise for Bayesian computational statistics. Joining forces, after some delays and failures to convince others to engage, or to produce multilevel papers with distinct vignettes, we eventually put together an additional four papers, where lead authors gathered further authors to produce this diverse picture of some incoming advances in the field. We have deliberated avoided topics which have excellent recent reviews— such as Stein’s method, sequential Monte Carlo, piecewise deterministic Markov processes— and topics which are still in their infancy, such as the relationship of Bayesian approaches to large language models (LLMs) and foundation models.

Within this issue, Past, Present, and Future of Software for Bayesian Inference from Erik Štrumbelj & al covers the state of the art in the most popular Bayesian software, reminding us of the massive impact BUGS has had on the adoption of Bayesian tools since its early introduction in the early 1990s (which I remember discovering at the Fourth Valencia meeting on Bayesian statistics in April 1991). With an interesting distinction between first and second generations, and a light foray of the potential third generation, maybe missing the role of LLMs in coding that are already impacting the approach to computing and the less immediate revolution brought by quantum computing. Winter & al.’s The Future of Bayesian Computation [TITLE TO CHANCE] is making a link with machine learning techniques, without looking at the scariest issue of how Bayesian inference can survive in a machine learning world! While it produces an additional foray into the blurry division between proper sampling (à la MCMC) and approximations, additional to the historical Martin et al. (2024), it articulates these aspects within a (deep) machine learning perspective, emphasizing the role of summaries produced by generative models exploiting the power of neural network computation/optimization. And the pivotal reliance on variational Bayes, which is the most active common denominator with machine learning. With further entries on major issues like distributed computing, opening on the important aspect of data protection and guaranteed  privacy. We particularly like the clinical presentation of this paper with attention to automation and limitations. Normalizing flows actually link this paper with Heng, Bortoli and Doucet’s coverage of the Schrödinger bridge, which is a more focussed coverage of recent advances on possibly the next generation of posterior samplers. The final paper, Bayesian experimental design by Rainforth & al., provides a most convincing application of the methods exposed in the earlier papers in that the field of Bayesian design has hugely benefited from the occurrence of such tools to become a prevalent way of designing statistical experiments in real settings.

We feel the future of Bayesian computing is bright! The Monte Carlo revolution of the 1990s continues to be a huge influence on today’s work, and now is complemented by an exciting range of new directions informed by modern machine learning.

Dennis Prangle and Christian P Robert

Shane MacGowan (1957-2023)

The punk singer Shane MacGowan has died a few days ago. He formed the fantastic Pogues Celtic punk band in the early 1980’s, mixing punk raw energy with Celtic tunes and instruments, and Irish nationalism. (In true punk fashion, the name of the band come from the Gaelic póg mo thóin that I let readers check for translation! In the same spirit, his earlier band was called the Nips with no connection with the Neural Information Processing Systems conference! However, he could have claimed a connection with Bayes since he was raised in Tunbridge Wells.) His early death is sadly unsurprising given his lifelong issue with alcohol, which started as a young boy when he started drinking Guinness… (As a minor trivia, the Dirty Old Town song was originally written about Salford, near Manchester, England, although adopted by several Irish bnds.)

Bayes impact

irreverent Mike [in memoriam]

While I could not find an on-line picture of Mike Titterington, another testimony to his modesty and selflessness, I remembered this series of sketches on priors he made for the Bulletin in Applied Statistics in 1982, under the title Irreverent Bayes!