Razib Khan's Unsupervised Learning

Subscribe now

Give a gift subscription

Share

This week on Unsupervised Learning Razib catches up with Leighton Woodhouse, a documentarian and journalist (with a Substack!), to discuss the rise of political polarization and the disintegration of traditional parties and coalitions on both the left and the right. Leighton, whose activism began in 1999 at the WTO protests in Seattle, reflects on the financial, geopolitical and social shocks of the last twenty years, how they’ve transformed the movements he came up in, and the new elitist vs. populist dichotomy he sees developing around him.

Originally a labor organizer and Marxist, he argues that the decline in rigorous debate in favor of moralistic platitudes and ideological conformity is both a symptom and cause of America’s current listless cultural malaise. As someone with an academic background, Leighton also bemoans universities now sacrificing difficult intellectual discourse on the altar of easy performativity.  

The conversation confronts the reality that in 2021 heterodox thinkers who refuse to perform are often pilloried. They observe how Glenn Greenwald, once lionized by the left for his courageous exposés and uncompromising defenses of civil liberties, is now demonized as a right-wing shill. Liberals even accuse Greenwald of undermining the security state, as if that’s a bad thing. Equally detrimental from Leighton’s viewpoint is the reduction of issues that should be nonpolitical, like humane treatment of animals, into political football games where there’s only one winner. 

Despite starting from very different political positions Razib and Leighton find common ground in critiquing a cultural moment in which appearance and outrage have taken center stage, and those claiming the mantle of leadership make only token changes while making sure to keep a grip on their power and privilege.

Subscribe now

Give a gift subscription

Share

Subscribe now

Give a gift subscription

Share

This week on the Unsupervised Learning podcast Dr. Xiaotong Yao, a computational biologist specializing in cancer research at Cornell’s Weill Institute, joins Razib.  

 They first dig deep into genomics, considering the efficacy and costs of expanding whole-genome sequencing to assemble massive population-sized datasets. Not a thousand people, but a billion. Next, they probe the implications of wide-scale sequencing as it becomes integral to clinical diagnosis and treatment. As for treatment specifically, they flesh out CRISPR/Cas9’s exciting possibilities for curing Mendelian diseases. 

Razib also asks Xiaotong about his experience as a millennial growing up in China, and his impressions of the current state of affairs between the United States, where he lives, and his homeland. They discuss ways in which the development of the Chinese government’s proactive role in education and economic development over the last 30 years has resulted in increased GDP, and how this has created a dynamic between the two global superpowers that has been mutually beneficial. US intellectual property and Chinese manufacturing have operated synergistically to make technological wonders like the smartphone globally accessible. 

Transcript

They finally touch on the reality that over the last few years tensions between China and the US have greatly increased, and nascent rivalry between what are now the world’s two great economic superpowers may lead to a Cold War redux. Razib remains hopeful that science and the pursuit of understanding can continue to provide a common ground to facilitate cross-cultural dialogue in pursuit of a better future for humanity, no matter the geopolitical winds.

Subscribe now

Give a gift subscription

Share

Subscribe now

Give a gift subscription

Share

This week on Unsupervised Learning author and Washington Post columnist (and former blogger) Megan McArdle join Razib for a  wide-ranging conversation reflecting on our reemergence after the year and a half ordeal of COVID lockdowns, rising violent crime rates, defunding policing, and the preposterous genetic distribution on Trantor, capital of Isaac Asimov’s Galactic Empire.  

An urbanite who has spent her life in the US’s own imperial capital cities - New York City, where she grew up in the 70’s and 80’s, and Washington D.C., where she now lives - Megan sees a functional law enforcement system as being a prerequisite to urban living. She, like Razib, is old enough to remember when crime and cities were synonymous, and they reflect on the murder spike of the last few years and how it might drive changes to the current pattern of gentrification.

Transcript

Throughout the conversation, they come back to the theme of generational reference points bracketing our experience and expectations – whether in young children for whom COVID lockdowns represent the majority of their remembered experience or Zoomers who can’t recall a time where network television created a shared narrative reality.

Subscribe now

Give a gift subscription

Share

Subscribe now

Give a gift subscription

Share

This week on Unsupervised Learning Charles C. Mann, author of 1491, 1493, and The Wizard and the Prophet joins Razib, to delve into the history of the Americas, and a broader theme that runs through Mann’s work – how human societies and their environment are inseparably intertwined.   

Mann’s work goes a long way towards dispelling the myth that the Americas were an untamed wilderness before the arrival of Europeans, scarcely populated and unshaped by the hand of man prior to Christopher Columbus. He describes a New World then peopled by complex societies with huge populations, possessing a well-developed toolkit of biological technologies for engineering the natural world, managing ecological succession, and diversifying food production strategy, all arguably superior to that of their European conquerors.  

Ultimately, when the Old and New Worlds collided, it was the calamitous impact of disease, rather than a significant technological advantage in weaponry, that eased the European conquest of the Americas. Through highlighting the fall of the Triple Alliance of Tenochtitlan, Tetzcoco, and Tlacopan, which comprised the Aztec Empire, and the Wars of Succession of the Inca, Mann also provides insight into human choices that also contributed to the end of these societies. 

During this pivotal period, for the first time, a global exchange of subsistence crops, slaves and luxury goods circulated throughout the whole world. The effects of the discovery of the New World were felt on every continent, as new crops were adopted in regions where they alleviated local food security issues and reshaped the local ecology (often increasing pressure on the landscape and further degrading it over time).

As the world transitioned to the 20th century it was a precarious landscape of food insecurity that motivated William Vogt, whom Mann styles as “the Prophet,” to preach on the importance of environmental carrying capacity and overpopulation. In contrast, Mann’s “Wizard,” agronomist Norman Borlaug, a pioneer of the technological techniques underpinning the Green Revolution, came to prominence applying science to enable our adaptive ingenuity in the face of ecological constraints.  For his part, Mann does not take sides or offer us a clear winner – but believes the discussion between these two intellectual strands to be of utmost importance when considering how we interpret our past and consider our future. 

Subscribe now

Subscribe now

Give a gift subscription

Share

This week on Unsupervised Learning, Razib is joined by Tim Lee, a former columnist at the Washington Post, Ars Technica, and Vox.com, to discuss his new project, Full Stack Economics, a newsletter on economics, technology, and public policy.

The conversation jumps directly into a major issue facing many Americans today: the cost of housing. In many US cities, access to affordable housing is the most economically important issue facing individuals and families. While nearly everyone agrees that working-class Americans should be able to afford rent or a mortgage, sharp differences in public policy positions like rent controls and freezes on property tax rates can create major distortions in the supply of housing and the incentives to build and maintain units. Numerous interests are at play between those who already own property in a neighborhood, whose property value often benefits from a housing crunch, those who want to purchase or rent in a new neighborhood, and developers who wish to increase supply but have to deal with established stakeholders. Though homeowners often make arguments about the character and aesthetics of their neighborhoods, the reality that their home values increase when they limit building has a major impact on their incentives from a purely economic perspective.

Though these issues are widely discussed in an ad hoc fashion every day in this country, Tim and his co-author, Alan Cole, attempt to highlight and flesh out the issue in a more explicit manner at Full Stack Economics, bringing both economic analysis and the public policy context to the fore.

Then Razib and Tim switch to discussing how the erosion of public trust in media organizations, particularly those that are principally funded by ad revenue, has created a market for subscription-based email newsletters. Tim’s goal with Full Stack Economics is to bring highly researched and deeply reported pieces directly to subscribers without the pressure of having to fulfill a weekly quota of copy or to be pressured by click-bait advertising incentives. Though his audience is smaller, he hopes it is more willing to be patient as well as more deeply.

They conclude their conversation by discussing a variety of technologies in the pipeline that might be potential economic and social game changers, including artificial meat, genetically modified mosquitoes, preimplantation genetic diagnosis, self-driving cars, and green energy.  Due to our leadership in science and technology, Tim remains bullish on the American economy and its potential in the coming decade, while being concerned by the political polarization and American culture’s social ills.

Subscribe now

Give a gift subscription

Share

Subscribe now

Give a gift subscription

Share

This week on Unsupervised Learning, researcher, blogger, and essayist Tanner Greer joins Razib to consider the challenges facing conservatism in America today, the future of China and its relationship to the US. Much of Tanner’s extensive research and analysis are featured on his excellent weblog, The Scholar’s Stage, and the conversation also touches on the current state of blogging (and its past). 

Razib and Tanner first tackle the evolution of a new strand of modern conservative thought that has labeled itself the ‘New Right’ which, despite the recycled name, was not intended to suggest any relationship to the previous ‘New Right’ of the 1960’s through the 1980’s – that culminated in Reaganite conservatism. Apparently, the 25-year-old staffers and writers who are promoting the New Right have no clear memory, let alone awareness of the movement of the same name that dominated the last third of 20th-century politics in America. Either that or they lack the creativity to come up with something original. 

Rather than being a reaction to a movement that young conservatives don’t remember, the 2020’s New Right grows out of a rejection of the Neoconservatism of the turn of the millennium and the libertarianism that failed to meet the challenges of the 2008 financial crisis. Tanner and Razib draw on the works of historians David Walker Howe and David Hackett Fischer to examine how the constitution of the New Right may have its roots in American history. But the New Right is less a grassroots movement than a collection of intellectuals gathering around journals and think tanks. It takes a suspicious view of the valorization of individualism that typified 20th-century conservatism, but its relevance to populist mass politics remains to be seen.

Tanner believes that the New Right is a positive development. He argues that different generations may require different means to tackle disparate problems and that the lack of political and economic power enjoyed by the young often leads to a tyranny of orthodoxy that eventually fades and disappears. Reaganism will only exhaust itself when the generations that remember Reagan retire from the scene.

Eventually, the conversation forays across the Atlantic for a brief look at China’s current prospects.. That nation faces its own set of challenges due in no small part to its own generational divide and disparities between rural and urban demographics.  Razib asks Tanner why everyone in DC suddenly seems to be a China hawk and whether we are headed into a new Cold War. Though today Tanner is writing about the American political scene, he was until recently more well known as a China watcher and observer.

Finally, they wrap up the conversation by discussing blogging, and the indispensability of a private and informal intellectual space for exploring ideas and interacting in some depth.  

Subscribe now

Give a gift subscription

Subscribe now

Give a gift subscription

Share

This week on the Unsupervised Learning podcast Razib turns his gaze to space with Eric Berger, Senior Space Editor at Ars Technica and author of Liftoff: Elon Musk and the Desperate Early Days that Launched SpaceX.

They ask who is Elon Musk anyway, and how did SpaceX come to win the early race to dominate private spaceflight? What does the privatization of the space fleet mean for the prospects and goals of NASA? How has NASA’s mission evolved, and how does the privatization of spaceflight complement NASA’s competencies? Eric shares his in-depth research and direct experience interviewing pivotal early members of SpaceX’s team to answer these questions. He also outlines Musk’s vision for the future of space travel and the prospects of a human colony on Mars. 

As the conversation rocks on, Razib and Eric go beyond the Kuiper belt and straight into a “science fiction” world that may be on an accelerated path to becoming “science fact” thanks to SpaceX’s early efforts pushing the envelope. They discuss why Jupiter’s moon Europa is so interesting to space-nuts, and when asteroid mining might become economically feasible.  Eric talks about how long we might have to wait for orbital power-generating stations and Dyson spheres. 

Finally, the discussion comes back to earth as they take a look at some of the problems and limitations that the space industry is currently trying to overcome in its efforts to make the “final frontier” something more than a television catchphrase. Eric talks about how the complementary strategies of competing billionaires, Musk, Jeff Bezos, and Richard Branson, are setting the stage for what may be the next exciting chapter in the human epoch.

Subscribe now

Give a gift subscription

Share

Subscribe now

Give a gift subscription

Share

This week on the Unsupervised Learning podcast, Harvard professor Carole Hooven joins Razib to discuss her new book T: The Story of Testosterone, the Hormone that Dominates and Divides Us. Though they do talk about the science of testosterone, Razib and Carole end up exploring the public reaction to her writing a book on sex and biology in 2021, as well the culture of censorship and shunning that has become the norm in much of academia. Hooven’s recent experience is in the context of backlash to her brief appearance on Fox News.

Hooven recounts in detail the repercussions she’s weathered for defending the terms ‘male’ and ‘female’ as being real, concrete and meaningful descriptors of observable phenomena associated with reproductive organs and the particular set of gametes one possesses. She stresses that these terms should in no way diminish any individual’s struggles with identity, sexuality and belonging - but that they are necessary to have accurate descriptors for the dimorphic nature of humans and other species, which begins in utero and is largely determined by prenatal hormonal balance affecting the physical development of both body and brain.

Transcript

They both talk about their disappointment with institutions and the erosion of academic vigor in favor of creating an environment that is supposedly ‘inclusive’ – an ironic assertion since inclusion seems to involve creating a climate that has a chilling, exclusionary effect on opinions that are not de rigueur.  

Eventually, the conversation circles around to the hormonal differences in the development of gendered traits, noting that some traits – such as play fighting – seem to be highly segregated between the sexes, in ways that are observably echoed across many species. Despite the reality of human culture and the complexity of individuals, they agree that some patterns and differences in distributions are real, robust and have to be addressed and acknowledged.

Subscribe now

Give a gift subscription

Share

Subscribe no

Give a gift subscription

Share

This week on the Unsupervised Learning Podcast, Razib gets into the genetics weeds again with Alex Young of the Social Sciences Genetic Association Consortium (SSGAC). They discuss the heritability of complex traits and how the SSGAC develops predictive models using genetics to tackle questions that have traditionally been the purview of social sciences (and why that’s controversial, but shouldn’t be). 

Alex explains how large datasets where many individuals’ whole genomes are sampled at high coverage (meaning the genome is sequenced over and over to get an accurate read) are necessary to develop the most effective complex models. This is due to the reality that most complex traits are very polygenic – controlled by numerous genes, thousands in the case of intelligence.  

Then they discuss the weaknesses in traditional twin studies, and Alex outlines why looking at family pedigrees more generally, particularly parents and siblings work out to be a natural controlled experiment. He foresees this as the future of complex trait analysis. Alex also discusses the statistical methods he has developed for analyzing genomes for heritability and complex traits, especially as they relate to “genetic nurture”. 

Transcript

Razib then asks a series of clarifying questions about the nature of the research on the heritability of educational attainment based on Alex’s extensive work on the subject in Iceland. They talk about how Iceland is perhaps the ideal place to perform an experiment attempting to distinguish sociological explanations from biological ones due to its isolation and homogeneity.  

They conclude their conversation about genetic nurture by clearly stating that there are complex interactions between societal structures and heritable traits, but emphasize that it is becoming apparent that there is significant evidence that characteristics such as educational attainment are mediated to a large extent by genes inherited from one’s parents. While ethical, social and cultural implications are unavoidable, fearmongering about racism, eugenics and the specter of a dystopian GATTACA-like future are unhelpful and uninformed. 

Subscribe now

Give a gift subscription

Subscribe now

Give a gift subscription

Share

This week on the Unsupervised Learning podcast, Razib is joined by genetic genealogist Josh Lipson for a deep dive into the history and genetics of the Ashkenazi Jewish population in Europe.  

They review the historical demographics of the Jews of both Northern Europe and the Mediterranean, as well as the possible founding source populations from the Levant (Palestine) and Mesopotamia (Babylon). They discuss the cultural and genetic differences between the Sephardim and Ashkenazim and consider whether the preponderance of evidence suggests a continuity of European Jews with Classical Antiquity or a more recent migration in the early Middle Ages. Josh believes that the recent genetic research indicating a more recent migration is probably wrong and that Y-chromosomal evidence implies Jews were present in the Western Mediterranean 2,000 years ago.

Josh brings a comprehensive understanding of historical, textual, linguistic and onomastic evidence to the table in explaining how dialect differences and family names can be used to tease out distinctions within the Ashkenazim, with clear regional identities in the Rhineland, Galicia (Southwest Poland), and the Baltic already being evident during the Middle Ages. He also argues that there is good evidence Rhineland Jews are subtly different from Central/Eastern European groups who presumably descend from them.

Finally, Razib discusses some of the finer points of genetic testing, statistical inferences, and how genetic models continue to improve over time, as well as where flaws in the testing algorithms might gloss over minute details. These details often provide subtle clues that might offer deeper insight if carefully analyzed by a human expert. 

Subscribe now

Give a gift subscription

Share