03 March, 2021

All Measures Short of a Cross Straits Invasion

Much of what I have written about Taiwan defense issues assumes that the primary challenge facing Taiwanese forces and their allies is defeating (and thus deterring) a proper amphibious invasion. Two recent reports argue—convincingly, I think—that this assumption is wrong. 

In his testimony to Congress a few weeks ago, former DIA analyst Lonnie Henley asks Americans to ponder what happens after the forces of right defeat the PLA invasion:

If ordered to compel reunification by military force, the PLA would bring every tool to bear. Among its most effective lines of operations would be a long-term air, maritime, and information blockade of Taiwan. Such a blockade could be the main effort, eschewing an attempted landing altogether, or it could be part of a larger invasion campaign. Most importantly, even if the landing failed, the PLA could continue the blockade indefinitely and neither US nor Taiwan forces would have much ability to overcome it.  

The Communist Party (CCP) leadership could not afford to accept defeat. The passions aroused by the war itself and by the propaganda effort in support of the war would not allow the Party to stop short of a political outcome they could credibly sell as a victory. If such a formula were available in the immediate wake of a failed landing, they might be tempted to take it. If not, they would have no choice but to continue the conflict by whatever means remained.  

Even at this point in the conflict, after weeks or months of intense fighting, the loss of much of its landing force, expenditure of its ballistic missile inventory, and very severe attrition of its navy and air force, the PLA could still have the upper hand in enforcing the blockade. The distinctive geography of the Taiwan theater would finally start working in the PLA’s favor and its remaining short-range strike assets would still be useful.

...In my assessment, China could continue the blockade operation indefinitely even with the severely diminished force that remained after a failed landing and months of air and naval attrition. US forces could probably push through a trickle of relief supplies, but not much more. [1]

Henley is disturbed by how little we know about how this scenario will play out:

Unlike Taiwan’s will to resist, its ability to resist a long-term blockade is something we should know much more about than we do. I am aware of no study in the United States or Taiwan examining Taiwan’s wartime consumption rate of critical materials, its peacetime stockpiles, or which stockpiles would likely be lost to PLA fires. There is no assessment of what must get through a blockade to keep Taiwan alive, what types of materiel in what quantities, or what Taiwan’s domestic production of food, water, supplies, and equipment might be under wartime conditions. And to the best of my knowledge, no one has considered in detail how to get enough materiel through a PLA air and maritime blockade, day after day, week after week, while working to break down the blockade itself.[2]

I am aware of one such study from 2013 that attempts to assess Taiwanese stocks of oil and jet fuel. It concludes:

Taiwan’s  total  stockpile  from  government  and  commercial  sources  could provide  732  days,  or  about  two  years,  of  civilian  consumption.  Such  an  amount  is beyond  adequate for  a  plausible  emergency scenario,  assuming  there  is  no  active  military  component.  Factoring  in military  demand,  however,  causes  the  days  of  available  supply  to  drop  precipitously.  In  an  air  war such as the one modeled above, Taiwan could supply 100 percent of military needs and 100 percent of civilian needs for 152 days, or about five months—still a solid emergency cushion, though a much smaller  one.[3]
Now that study is seven years old. I do not know how accurate it is now. Moreover, it assumes that the Taiwanese do not lose any fuel to PLAF or PLARF attacks, a wildly implausible operating assumption. But it does present a rosier picture of Taiwanese preparedness than Henley seems to fear.

It would not surprise me if nestled away in Taiwanese military journals are a few more studies of this type, but I am not aware of them. In English there is nothing else to draw on. What we have is not enough. Henley argues that given how little this has been studied, security analysts' confidence that the Chinese will not resort to blockade, and that if Chinese did so resort their blockade would not succeed, is unfounded. 

It is hard to argue with this logic.

Henley's blockade worries give him an unorthodox take on the true "center of gravity" of a cross straits conflict. Current Taiwanese defense planning sees the center of gravity on the landing beach, and their strategy is built around destroying the PLA landing force before it can dig in there. Americans, perhaps influenced by the traditional naval leadership of INDOPACOM, instead tend to see victory in terms of the number of PLAN vessels sunk.[4]  Talk cross straits contingencies with enough American military officers and you will see this pattern. Their optimism or pessimism usually rests on their personal assessment of America's submarine forces and Chinese ASW capabilities. Those confident that American attack submarines can pierce the Chinese anti-access bubble are bearish on the PLAN successfully getting a proper invasion force across the strait. Those who think the Chinese can keep American submarines at bay are far more pessimistic.[5]

Henley focuses on a different domain altogether:

The center of gravity for this entire conflict, in my judgment, is the PLA air defense network. Over many years of participating in Taiwan Strait war games and tabletop exercises, I observe that Taiwan’s air defenses are almost always disabled within the first few days of the conflict, but China’s integrated air defense system (IADS) along the Taiwan Strait remains effective for as far into the conflict as the exercise examines. This in turn limits the United States to long-range stand-off weapons or precision-strike incursions by stealth platforms. I assume I am not the only person to have observed this and that US forces are working on the issue.

 Success in this area would have the greatest impact on the overall conflict, more even than finding a way to defend US air bases from Chinese missile strikes. Poorly defended bases will still generate some combat sorties, particularly as the conflict drags on and the Chinese expend their inventory of theater-range missiles. But a functioning air defense network greatly reduces the impact those sorties can have. Conversely, defeating the Chinese IADS would open the door to the kind of air campaign that has proven decisive against less capable opponents. More specifically, enabling US air operations over the Strait would be our best hope for getting cargo into Taiwan’s western ports. The PLA’s short-range anti-ship assets can be extremely effective under a tight air defense umbrella, but much less so in the glare of US air power. The PLA air blockade, meanwhile, simply ceases to exist without the IADS. [6]

If Henley is right and the Chinese IADS are the critical node in the Chinese war effort, this has important strategic and diplomatic consequences. Most important is this: the core of the Chinese IADS are mobile surface to air missile systems (like the Russian bought S400, S300, and the indigenous HQ-9) and the mobile radar systems that support them. These forces—which the Pentagon admits outclass their American counterpartsare ground based.[7]- The PLAN's destroyers and the PLARF's small fleet of airborne early warning and control ("AEWAC") planes are also a part of this system, but they play the smaller part. Any fight to destroy the Chinese IADS is a fight to destroy ground based missile launchers—missile launchers that will be located on mainland Chinese soil.

These ground based systems will not be destroyed quickly. As one helpful review of the Chinese IADS system concludes that, while "completely rolling back an IADS the size, depth and complexity of those of Russia or China" is possible, it "would most likely take weeks and possibly months of full-scale warfighting."[8] 

Part of the appeal of the "destroy the invasion fleet" strategy is precisely that it avoids months of full-time warfighting, especially months of attacks on mainland targets. Those who want to avoid escalation eagerly search for ways to keep the fight "in theater." But with surface to air missiles being fired from mainland locations hundreds of kilometers inland, the distinction between military platforms "in theater" and on the mainland grows blurry.

So there are real stakes then to this debate. It is interesting to see how much of this debate turns on essentially political questions. Just how willing is Beijing to accept the costs of a multi-month blockade after having lost much of its Navy or an invasion force? Henley thinks they will be willing to keep plugging away, confident that if nothing else they can outlast the Taiwanese:

Most of the operational approaches available to US forces would not serve to end the war. Defeating the landing operation is feasible given a large enough US effort, but to repeat my earlier point, that would merely move the war into the next very extended phase. It would not end the military conflict, nor would it necessarily go very far toward creating the conditions for a political settlement...

The problem is that Chinese leaders certainly would think far more than twice before going to war against the United States. The military cost is only one of myriad reasons not to do it, and not the most important reason by far. If they decide they must do so anyway, they will have made that decision in full acceptance that the war will be economically devastating to China for decades to come and that its failure would severely endanger the Communist Party’s hold on power. At that point, the “cost imposition” dial is at 11; it won’t go any higher.

This brings us back to my central point: if we can’t defeat the blockade, we can’t prevail.[9]

I would not be so confident. One of the odd things about American strategic culture is the cataclysmic nature of our strategic thinking. We jump immediately to total ends: for two centuries we have equated "war termination" with total surrender. But this is not the historical norm. It certainly is not a marked feature of Chinese military history. I can imagine many scenarios where the Chinese conclude that, after having destroyed the ROC Navy and Air Force, killed Taiwan's leaders, reduced her communications bases, air defense system, and radar stations to rubble, and inflected incalculable suffering on the Taiwanese people, they have successfully "taught a lesson" to the Taiwanese and seek a termination of hostilities in exchange for a return to the status quo (knowing full well they will recover from their losses far faster than the Taiwanese can hope to).

I can also imagine a scenario of rising discontent at home. It is it so fanciful to picture a Chinese people dispirited by battlefield defeats or afraid of escalating destruction as the Americans and Japanese get involved in a war they assumed would be full of quick and easy victories? Behind this scenario is the recognition that all costs are marginal costs. In the face of domestic discontent marginal domestic costs of continuing operations—especially operations like a blockade, which don't offer any morale-boosting headlines—may exceed the marginal costs of admitting defeat. 

This sounds more plausible when we consider the types of costs such a war will impose on the Chinese people by default. A recent report by Robert Blackwill and Philip Zelikow reminds us of these costs:

In this context, it is useful to recall and review what happens when the United States goes to war with a country, even a small war. First, the United States would freeze all assets owned by that country, or its citizens, in the United States. In this case, the United States could have to recognize an independent Taiwanese government, even in extremis a government-in-exile of some kind, in order to distinguish and protect Taiwanese-owned assets from being seized and possibly forfeited along with all Chinese-owned assets.

Second, the United States would cut off, and strictly control, any business transactions or dollar transactions with China. No trading with the enemy would be conducted. This would necessarily end any payment of interest on American securities, government or private, held by Chinese citizens or the Chinese government. It would include at least the suspension of interest payments on Treasury bonds held by the Chinese government or Chinese citizens.

 These are profound measures. They would affect trillions of dollars of assets in the United States and around the world. We presume that China would retaliate in kind. These moves would immediately trigger a large and devastating financial and economic crisis (as also happened immediately after the outbreak of war in August 1914, because  of  the  asset  controls and commercial closures, not because of the impact of the military operations themselves.

 These effects would be so great that it is not credible to threaten them as sanctions. We are not proposing a strategy of coercive diplomacy. This is a strategy to spell out how  world  politics and the world economy are likely to fracture after such a terrible break. That is why robust U.S. and allied local military capability is so essential. Without the impetus of an outbreak of fighting, Washington’s deterrence threat of such gigantic measures could seem hollow.However,  if  a violent local conflict does occur, such extraordinary measures would not only be adopted but also likely implemented practically overnight. If Japanese citizens had also been killed, and Japan was embroiled in the local war, then Japan would probably adopt a similar set of draconian measures.[9]
And all of that is without cyber-attacks, blockades, and other measures less automatic than the ones that Blackwill and Zelikow describe. Of course, Henley is right: if Beijing opts for war that means it opts for war with full knowledge all of this may happen. But if the Party leadership (or just as importantly, the Chinese people) go to war with the belief that victory will be a quick and easy thing, their tolerance for bearing these costs over an extended period of time might be much smaller than Henley predicts.

Given their sensitivity to the costs of full campaigning, Blackwill and Zelikow suggest another measure short of armed invasion that the PLA may opt for. They call it a "quarantine": 

By quarantine, we do not mean blockade. In a quarantine scenario, the Chinese  government  would  effectively  take  control  of  the  air  and  sea  borders of Taiwan. It would declare control over Taiwan’s airspace so that,  in  effect,  Taipei’s  Taoyuan  International  Airport  was  no  longer  its  own  international  gateway,  and  Kaohsiung  was  no  longer  its  own  international  port.  The  Chinese  government  would  run  effectively  a  clearance  operation  offshore  or  in  the  air  to  screen  incoming  ships  and aircraft. The screeners could then wave along what they regarded as innocent traffic. Or they could request that suspect ships or aircraft divert  for  full  Chinese  customs  clearance  at  a  neighboring  airport  on  the mainland or in a neighboring port, such as Fuzhou or Guangzhou, or  Xiamen  or  Shantou.  China  has  excellent  “domain  awareness,”  having plenty of ships from its navy, coast guard, and maritime militia at its disposal to accomplish this task....

The Chinese government could run such a quarantine without actually trying to take effective control of the Taiwanese people themselves. In  this  scenario,  the  Chinese  government  would  allow  the  people  in  Taiwan  to  run  their  own  affairs  on  the  island,  at  least  for  some  time,  as China showed that it controlled who came (and perhaps who went).

This  scenario  includes  variations,  in  which  China  engages  Taiwan’s air and naval forces that contest this quarantine, or fires salvoes of missiles into Taiwan to intimidate its citizens into compliance. The point is that, in all of these scenarios, China neither invades Taiwan nor attempts to cut off supplies of food or energy, as it would in a full siege. The goal is to force Taiwan to accept a loss of control, cutting Taiwan off from, at least, transfers of military equipment and associated foreign experts. This scenario places a heavy burden on foreigners to decide whether they will deliberately choose to make a military challenge to this assertion of Chinese sovereignty. In this context, with the Chinese making these  arguments,  the  outsider,  such  as  the  United  States  or  Japan,  would first have to negotiate the divided and contentious domestic politics surrounding such a deliberate and dangerous military challenge. At the same time, all would be watching the behavior of Taiwan’s citizens and their political divisions and quarrels about how to proceed....

This  scenario  could  then  seem  to  offer  the  prospect,  to  a  Chinese  planner, of relatively manageable risks and high rewards. China would not  only  demonstrate  meaningful  sovereignty;  it  would  gain  the  military benefit of effectively blocking the further military modernization of Taiwan’s defenses, possibly permanently. If that were successful, the longer-term effects on Taiwan’s domestic politics could be imagined.[10]

I find this scenario particularly disturbing: I do not think the United States and Japan currently have any easy proportional counter-response to employ against a "quarantine." While Japanese defense officials have recently sent some very strong signals about the importance of Taiwan to Japan's national defense, they have not communicated this effectively to the Japanese public. Moreover, American and Japanese officials have not done joint contingency planning for nor gamed out responses to this sort of Taiwan challenge. It will be difficult to do so on the fly, with the American and Japanese publics unenthusiastic about Taiwanese liberties and significant business interests in both countries resolutely opposed to actions that might threaten their China-buoyed bottom lines. 

Something similar could be said for Chinese attacks on Taiwan's outer islands. In all cases, we will be in a far better position to respond to Chinese aggression if Taiwanese, Japanese, and American officials have been coordinating, contingency planning, and vigorously preparing public opinion for a robust response well before the crisis arrives. The time to think through these problems is now.

For more writing on Taiwanese affairs, you might also like the posts "Taiwan's Past Matters Less than Taiwan's Present," "Why I Fear For Taiwan,"  "Why Taiwanese Leaders Put Political Symbolism Above Military Power," "Taiwan Will Be Defended by the Bullet or Not At All," and "Losing Taiwan is Losing Japan." To get updates on new posts published at the Scholar's Stage, you can join the Scholar's Stage mailing list, follow my twitter feed, or support my writing through Patreon. Your support makes this blog possible.

[1] Lonnie Henley, "PLA Operational Concepts and Centers of Gravity in a Taiwan Conflict," Testimony before the U.S.-China Economic and Security Review Commission (18 February 2021), 4.

[2] ibid., 6.

[3] Rosemary Kelanic, "Oil Security and Conventional War: Lessons From a China-Taiwan Air Scenario," Council on Foreign Relations report (October 2013), 8.

[4] The best English language overview of the current Taiwan defense concept is Lee Hsi-min and Eric Lee, "Taiwan's Overall Defense Concept, Explained," The Diplomat (3 November 2020). For examples of American views see  Michelle Flournoy's comments in Foreign Affairs last year:

if the US military had the capability to credibly threaten to sink all of China's military vessels, submarines, and merchant ships in the South China Sea within 72 hours, Chinese leaders might think twice before, say, launching a blockade or invasion of Taiwan; they would have to wonder whether it was worth putting their entire fleet at risk.
Robert Work said something similar a year before:

 Pentagon leaders should challenge the armed services to solve very hard, very specific problems, Work said: Sink 350 Chinese navy and coast guard vessels in the first 72 hours of a war.
Blackwill and Zenilow, cited below, cite both of these examples on p. 37 as senior defense officials seeing a "need to develop a capability to destroy every Chinese vessel in the South China Sea within seventy-two hours." I credit them for drawing my attention to the parallels between both statements, but am somewhat mystified at their reference to the South China Sea, when both Work and Flournoy are discussing Taiwan contingencies. For the original statements see Michèle A. Flournoy, “How to Prevent a War in Asia,” Foreign Affairs (18 June 2020) and Sydney Freedberg Jr., “US ‘Gets Its Ass Handed to It’ In Wargames: Here’s a $24 Billion Fix,Breaking Defense (7 March 2019).

[5] This is also the argument you will hear from the defenders of Taiwan's indigenous submarine program (see for example Mark Stokes comments here), and I agree with these folks in the abstract. But the subs cost a lot and will be slow in coming. Too slow, perhaps, to make a difference.

[6] Henley, "PLA Operatioal Concepts," 6.

[7] For that judgement see Military and Security Developments Involving the People's Republic of China: 2020 (Washington DC: Office of the Secretary of Defense, 2020), vii.

[8] Justin Bronk, "Modern Russian and Chinese Integrated Air Defence Systems: The Nature of the Threat, Growth Trajectory and Western Options," RUSI Occasional paper (January 2020), 32.

[9] Robert Blackwill and Philip Zelikow, The United States, China, and Taiwan: A Strategy to Prevent War (Washington DC: Council on Foreign Relations, 2021), 45-46.

[10] ibid., 35-37.

27 February, 2021

Longfellow and the Decline of American Poetry

 Last summer the New Yorker published an essay by James Marcus that asks the following question: why was Henry Wadsworth Longfellow so loved in his own lifetime when today he is so little read or respected? There is one very compelling answer to this that the article that does not discuss—indeed, that the article itself is an unconscious illustration of. This answer touches on not only Longfellow’s reputation, but the broader place of poetry in American intellectual and social life. Longfellow’s fall from the top of the American pantheon foreshadows the fall of American poetry as a popular art form. 

Now this “decline of poetry” is the subject of many long debates and essays. It is an over-determined phenomena. Everybody has their own favorite explanation for why poetry is no longer read and why poets have lost so much of their past esteem. The explanation most popular among dogmatists (as well as most non-readers of poetry) has to do with the disappearance of rhyme and meter. There is some truth to that—and Longfellow was nothing if not a man of easily discerned rhymes and meters—but the story I want to tell today is more subtle and fundamental than a move towards free verse. 

Near the end of Marcus’ essay we find the following tidbit about Longfellow’s writing:

It was not in Longfellow’s nature to write about himself. He once described “I” as “that objectionable pronoun.” But he did produce a handful of lyrics throughout his career that seemed to spring directly from his own suffering.[1]
What Marcus does not note is that this aversion to “I” was once the dominant mode of English poetry. Historians of the English lyric note a 19th century shift in the form and focus. We are the heirs of this shift and our conceptions of the poetic are framed entirely inside of it. We think of poetry as the art of poignantly capturing interior experience on the page. The natural focus of the poet, we believe, is emotions, perceptions, and other subjective experiences. But this was not always the case. Here how one scholar of English literature describes the shift:
Many readers nowadays tend to think of poetry as a form of self-expression—a way to convey thoughts and feelings that are pent up inside of us. But what might happen to our reading of older poetry if we were to find out that this common assumption is, historically speaking, relatively recent? 
In fact, thinking of poetry as the expression of a poet’s unique inner life is a legacy of cultural shifts that began to take place in the West during the period of Romanticism starting around the last decades of the eighteenth century. By contrast, during earlier centuries poets and their audiences did not automatically assume that the main purpose of poetry was to express the feelings of a unique individual. Instead, they lived in a culture that valued the skillful use of language because of its potential effects on an audience. Educated readers and writers shared the assumption that the function of poetry is to delight, teach, and move us. If we are stuck with the post–Romantic expectation that all poetry is about the self-expression of a unique individual, we will miss out on much of what remains valuable and interesting in earlier lyric masterpieces. Our understanding and enjoyment of older poetry will be greatly increased if we try to orient ourselves to the rhetorical culture that helped shape it.[2]

The rest of that book (The English Lyric Tradition: Reading Poetic Masterpieces of the Middle Ages and Renaissance) goes on to explore how most lyric poems in the renaissance era were “rhetorical” lyrics, not "interiorized" ones. But the bulk of Renaissance and Restoration poetry were not even lyric poems to start with. They were dramatic or narrative poems. Poets like Chaucer, Spencer, Shakespeare, Milton, Dryden, and Pope, who reigned supreme as the exemplary poets of their respective centuries, made their fame not through short lyric poems, but through plays, masques, epics, dramatic dialogues, poetic narratives, and poetic essays. 

The poet most responsible for elevating the interior over the rhetorical was William Wordsworth. Wordsworth famously saw poetry as “the spontaneous overflow of powerful feeling: it takes its origin from emotion recollected in tranquility.” [3]  Wordsworth’s poetry was thus mostly about himself— or the wider world as he personally perceived it. There is nothing inherently wrong with this. Wordsworth’s poems are beautiful and touching, and because of that they were hugely influential. So influential, in fact, that they completely changed what the intellectual castes of the English-speaking world considered the purpose of poetic art. Thus by 1860 we already see John Stuart Mill arguing things like this:

At what age is the passion for a story, for almost any kind of story, merely as a story, the most intense? In childhood. But that also is the age at which poetry, even of the simplest description, is least relished and least understood; because the feelings with which it is especially conversant are yet undeveloped, and, not having been even in the slightest degree experienced, cannot be sympathized with.
In what stage of the progress of society, again, is story-telling most valued, and the story-teller in greatest request and honor? In a rude state like that of the Tartars and Arabs at this day, and of almost all nations in the earliest ages. But, in this state of society, there is little poetry except ballads, which are mostly narrative,---that is, essentially stories,---and derive their principal interest from the incidents. Considered as poetry, they are of the lowest and most elementary kind: the feelings depicted, or rather indicated, are the simplest our nature has; such joys and griefs as the immediate pressure of some outward event excites in rude minds, which live wholly immersed in outward things, and have never, either from choice or a force they could not resist, turned themselves to the contemplation of the world within.
Passing now from childhood, and from the childhood of society, to the grown-up men and women of this most grown-up and unchild-like age, the minds and hearts of greatest depth and elevation are commonly those which take greatest delight in poetry: the shallowest and emptiest, on the contrary, are, at all events, not those least addicted to novel-reading. This accords, too, with all analogous experience of human nature. The sort of persons whom not merely in books, but in their lives, we find perpetually engaged in hunting for excitement from without, are invariably those who do not possess, either in the vigor of their intellectual powers or in the depth of their sensibilities, that which would enable them to find ample excitement, nearer home. The most idle and frivolous persons take a natural delight in fictitious narrative: the excitement it affords, is of the kind which comes from without. Such persons are rarely lovers of poetry, though they may fancy themselves so because they relish novels in verse. But poetry, which is the delineation of the deeper and more secret workings of human emotion, is interesting only to those to whom it recalls what they have felt, or whose imagination it stirs up to conceive what they could feel, or what they might have been able to feel, had their outward circumstances been different.
Poetry, when it is really such, is truth; and fiction, if it is good for anything, is truth: but they are different truths. The truth of poetry is to paint the human soul truly: the truth of fiction is to give a true picture of life. The two kinds of knowledge are different, and come by different ways, come mostly to different persons. Great poets are often proverbially ignorant of life. What they know has come by observation of themselves: they have found within them one, highly, delicate and sensitive specimen of human nature, on which the laws of emotion are written in large characters, such as can be read off without much study. Other knowledge of mankind, such as comes to men of the world by outward experience, is not indispensable to them as poets: but, to the novelist, such, knowledge is all in all; he has to describe outward things, not the inward man; actions and events, not feelings; and it will not do for him to be numbered among those, who, as Madame Roland said of Brissot, know man, but not men.[4]

By the time the modernists roll around a half century later poetry had been completely “interiorized”—or in Mill’s terms, had become the art of the human soul, not that of human life. We still speak about poetry this way. See, for example, Edward Hirsch's 2006 column celebrating poetry as the medium that “delivers on our spiritual lives precisely because it simultaneously gives us the gift of intimacy and interiority, privacy and participation.”[5] Not surprisingly, our canon of great poets has shifted towards the poets of “the intimate.” 

This impulse runs through Marcus’s essay. He compares Longfellow unfavorably to Emily Dickinson—but Dickinson 20th century reputation as America’s best 19th century poet rests entirely on her mastery of compact statements of interior self-expression. It is not surprising that Marcus’ favorite Longfellow poems are those few which deal with the most interior of topics: grief. 

This century long shift away from rhetorical or narrative poems towards intimate, interiorized lyrics is one of the central engines of poetry’s declining popularity as an artform. This is not because lyric poetry is less meaningful than narrative or dramatic poetry. Rather, it lends itself to a different sort of social experience. Marcus notes with some wonder that Longfellow’s epic Evangeline “went through six printings in a matter of months.” What were those printings used for? What exactly did Evangeline mean to the antebellum Americans who bought it up in droves? 

Here is what they were not doing: retreating to some secluded corner to ponder Longfellow’s verse and puzzle out the subtleties of its real intent! Those who bought Longfellow’s epic bought it to read aloud or recite to friends and family. In the 19th century this was one of most common types of household entertainment. Families would gather together around the hearth to recite dramatic scenes from plays or read long sections of narrative poems. Poetry was a social experience.[6]

This helps account far Longfellow’s metrical choices, which often sound silly or forced to modern ears. We feel this way because we do not actually experience Longfellow’s verse with our ears. We experience it with our eyes. The ceaseless ta-tum, ta-tum of Longfellow’s Hiawatha is ideal for recitation and memorization. Marcus pokes fun at Longfellow’s extended similes, but here again we find a poetic device designed for someone listening to poetry. The listener cannot ‘pause’ the poem to ponder out what a given metaphor might mean, but must instead catch this meaning before the reciter moves forward with the narrative. Spending extra lines on a clever metaphor ensures that a listener is able comprehend it in time.

 Modernism, free verse forms, and the puzzle-like nature of so much modern poetry are all fruits of this transition--the transition from the social and spoken to the solitary and written. Could this shift have occurred if poets had not abandoned older forms in favor of dense, subjective lyrics? [7] To be sure, there are older poems of the puzzling sort, poems almost certainly written to be read, not spoken. John Donne’s poems are a perfect example of this—they must be read closely before subtle meanings can be drawn out of them. But then again, John Donne’s verse was largely ignored until the 20th century, when T.S. Eliot agitated for his central place in the canon. [8] As with Emily Dickinson, Donne’s rehabilitation is a story of meditative, cryptic lyricists lionizing poets from the past who most approached their own ideals. We understand poetry through the lens of early 20th century critics (not coincidentally, the last years poetry or its critics mattered to the wider world). 

Line for line and word for word the interior lyricists write with more beauty and insight than most dramatic or narrative poets ever managed (much less the exterior, rhetorical lyric poets of centuries past). But the cost of this transition are clear: poetry has become less approachable for the general educated public, less enjoyable to those without special training, and less central to intellectual life writ large. This may have been inevitable—I do not know if bedside poetry jams would ever have been able to compete with the television.[9] But resistance to narrative and dramatic forms does make things harder on professional poets. Today's poets give little honor to poems which do not reach an absurd standard of insight and irony. No one starts out loving that sort of stuff. Poetry lovers find their love young, but today’s young adults never get much exposure to more accessible forms—a love of which might one day propel them on to more subtle and sensitive lyric works. 

I submit that a 21st century poetry revival, if it is to occur, must occur as a revival of verse as a vehicle for telling stories. Two books come to mind as admirable examples of what this might look like in practice—thoroughly modern, 21st century narrative poems. The first is Christopher Logue's War Music, a work from last century. War Music is an attempt to retell the story of the Iliad in a more modern poetic form (complete with epic similes comparing ancient Greek battle to “the joy of the Uzi shuddering warm against your hip”).[10] Unfortunately Logue's work assumes a close familiarity with the Iliad and will not make sense to people who have not read the original. But for those who have read their Homer it is a wild ride—the first book I give people when they want to see the potential that poetry might have as a narrative form today. 

Less profound, but more fun, is Aaron Poochigian’s Mr. Either/Or. The protagonist of this noir-mystery is a secret agent who must defeat a cult who wants to use a secret artifact to destroy New York City. Really! On the face of it is complete pulp, a mishmash of Spielberg films and Bond tropes—but a mishmash of Spielberg films and Bond tropes written entirely in iambic pentameter. 

Look: I have read novels in verse before. Most of them suck. They are not readable. Poochigan understands that narrative poetry only works if the poetry carries the reader forward. And his does! It takes a few minutes to acclimate yourself to the form, but once you have gotten used to the style you will not want to put Poochigan's mystery-action-science fiction thriller down. It isn’t Shakespeare or Paradise Lost or even Idylls of the King. It is more fun than deep. But in terms of craftmanship it is a near perfect book. This is what 21st century narrative poetry should read like. When some man or woman does write our century’s Idylls of the King that poet will be writing in the style that Poochigan pioneered. A poet who can marry the depth of War Music to the poetic style of Mr. Either/Or will have produced a masterpiece. 

Hopefully that masterpiece will be written. Of course, not all dramatic or narrative poetry needs to be book length—most of Longfellow’s certainly wasn’t. But unless we see more poets as committed to narrative art as Longfellow was, I doubt poetry will ever be able to reclaim an honored place at the center of American high culture.

If this post on the social role of literature in American life has caught your interest, consider reading my earlier posts "Shakespeare in American Politics," "Where Have all the Great Works Gone?" "On Adding Phrases to the Language," and "Living in the Shadow of the Boomers," To get updates on new posts published at the Scholar's Stage, you can join the Scholar's Stage mailing list, follow my twitter feed, or support my writing through Patreon. Your support makes this blog possible.

[1] James Marcus, "What is There to Love About Longfellow?," The New Yorker (1 June 2020)

[2] James Goldstein, The English Lyric Tradition: Reading Poetic Masterpieces of the Middle Ages and Renaissance (Jefferson, NC: McFarland Publishing, 2017), 11.

[3] William Wordsworth, "Preface to Lyrical Ballads" (or. published 1802, available at Wikisource).

[4] John Stuart Mill, "Thoughts on Poetry and Its Varieties," The Crayon, vol 8 (April 1860), 94.

[5] Edward Hirsch, "In the Beginning Is the Relation," Poetry Foundation Website (or. pub. Washington Post, 2006; accessed 27 Feb 2020).

[6] See Joan Rubin, Songs of Ourselves: The Uses of Poetry in America (Cambridge: Cambridge University Press, 2009) and Michael C Cohen, The Social Life of Poems in 19th Century America (Philadelphia: University of Pennsylvania Press, 2015) for many vignettes of exactly this sort of thing.

[7] Probably not. One of the interesting piece of evidence I might provide here is that this movement from spoken to read, and from external to internal, also occurred in Chinese poetry—and there occurred twice. See

[8] As in T. S. Eliot, review of Metaphysical Lyrics and Poems of the Seventeenth Century: Donne to Butler. Selected and edited, with an Essay, by Herbert J. C. Grierson (Oxford: Clarendon Press. London; Milford) in the Times Literary Supplement (October 1921). 

[9] For more on how television came to replace the role that poetry once played, see Tanner Greer, "On Adding Phrases to the Language," Scholar's Stage ( 13 October 2019).

[10] Christopher Logue, War Music: An Account of Homer's Iliad (New York: Farrar, Straus and Giroux, 2016), Kindle Locations 3162-3177.

25 February, 2021

The Framers and the Framed: Notes On the Slate Star Codex Controversy

Let's talk about the grand Slate Star Codex brouhaha.

A lot of people have already written about this. Here is the original New York Times piece that started the controversy. [1] Against the Grey Lady we have Cathy Young, Robby Soave, Micah Meadowcroft, Matthew Yglesias, Freddie DeBoer, Scott Aaronson, Noah Smith, and Dan Drezner, as well as Scott Alexander himself. [2] The most compelling brief in the Gray Lady’s favor was written by Elizabeth Spiers, but Will Wilkinson and Elizabeth Sandifer have weighed in as well.[3] Gideon Lewis Kraus’ New Yorker essay from last June is probably the best “neutral” piece that has been written yet; If you do not know anything about Slate Star Codex or why so many people are writing about it now, start there. Sebastian Benthall's commentary is also a very good middle ground analysis.[4]

I am sure there has been a great deal of debate on twitter as well. I have not read it and thus cannot link to it: I unfollowed everyone on Twitter except a handful of newspapers and thus dwell in blissful ignorance. Indeed, from the perspective of one slowly letting go of Twitter following this debate has been great fun. Linking to all those blogs and substacks feels like reliving a memory from an older, better internet.

The least charming things about this entire debate is how every participant feels compelled to declare their loyalties before they state whatever they actually have to say. I'm not going to make my general readership slog through that kind of thing here; if you are the type who must know every intersection between my personal biography and the worlds of journalism or rationalist blogging, you will find that information in the fifth footnote.[5]

What sticks out to me when reading all of these pieces, aside from the biographical digressions, is that the participants are not actually debating the same thing. There are a half dozen separate questions being fought over. Some folks have a pressing interest in conflating them with each other. I do not think this is helpful.

At a minimum, these questions include:

1) Was it ok to “out” Scott Alexander’s true identity as Scott Siskind?

2) Did this specific New York Times article (“Silicon Valley’s Safe Space”) misrepresent the content of Slate Star Codex, the contours of the broader rationalist community, or the nature of their connection with Silicon Valley?

3) Assuming things were misrepresented, why did that happen? Was it a premeditated “hit job” or revenge piece? Or is there a better explanation for what happened than that?

4) Do journalists have the right to uproot the lives of their subjects lives with negative coverage? Do communities so targeted have the right to impose costs on journalists (say, by harassing them on twitter or flooding their inboxes) that are “just doing their job?” (A simpler way of phrasing this question: who is “punching up” here?)

5) Is this a controversy specific to the New York Times, or does the incident point to broader problems in the way American journalism works as a whole?

6) Do powerful figures in the consumer tech sector really expect journalists to play the role of a glorified PR agent? (Or to flip the question around: are journalists unfairly biased against tech?)

7) Does the entire Slate Star Codex affair prove the Silicon Valley decentralist argument right? Has the time come to overthrow old “East Coast” hierarchies and replace them with new “West Coast” institutions?

Now look, if you are Balaji Srinivasan you are going to want a negative answer on question #1 to translate to positive answer on question #7. Given his chosen project I cannot fault him for trying to equate one with the other. But in truth question #7 is not the same as question #1. You can be an East Coast climber and still view this particular piece as libelous. Or you can think Alexander had unrealistic expectations for personal privacy while also believing journalists are biased against tech. We have reduced these separate threads into one big debate: who is the real enemy here? This sort of Schmittian friends-and-enemies game is stupid. We are smarter than this.

Let us go through these questions one-by-one. Some of these questions are more interesting than the others. I will not be giving them equal space. I cannot promise I will finish them all today, but may instead defer some of the questions to a second post to keep this at a readable length.

Question One: Was it ok to “out” Scott Alexander’s true identity as Scott Siskind?

Of course not. Look: both psychiatrists and online community leaders have a certain sort of relationship with their clients and followers. By design these relationships are bounded in a specific domain and are wildly inappropriate outside of those bounds. With a big online following comes a host of parasocial relationships. These relationships are often, for lack of a better word, creepy. It would not be good for Scott’s office if his parasocial hanger-ons (be they his most ardent haters or his most rabid fans) show up looking for treatment.

Likewise, the relationship between psychiatrists and their clients are inherently imbalanced. A client may view their psychiatrist as their friend, but psychiatrists are not friends. By design it is a one-way relationship. In the world of counseling, clients who have an outside connection to their therapist, counselor, psychiatrist, or so forth are described as being part of a “dual” or “multiple relationship”—an ethical no-no for the profession.

While none of the various psychological codes of ethics specifically mention parasocial relationships in their lists of improper dual relationships, it should be obvious to anyone who thinks about the issue for 60 seconds that a psychiatrist cannot have a healthy working relationship with his or her client when said client has a parasocial attachment to them (especially when said psychiatrist regularly blogs about his experiences treating patients).

This isn’t hard folks. Scott created two separate identities because—irrespective of the actual content of his writing—the role of “psychiatrist” and “online community leader” cannot be played by the same person at the same time.

I do not think this solution was ever sustainable on the long term. At some point Scott would have to choose which role he wanted to play full time. But that should have been his decision. What right did the New York Times have to make that decision for him? Where do they get the moral authority to decide when that decision must be made? A great deal of the anger directed at the Times comes back to that question: just who are they to make that call—and to make it for the sake of C-section column most Times readers will have forgotten about a week after they read it?

Question two: Did this specific New York Times article (“Silicon Valley’s Safe Space”) misrepresent the content of Slate Star Codex, the contours of the broader rationalist community, or the nature of their connection with Silicon Valley?


Lots of folks have discoursed on this one already. I have little original to add and will not rechew others' cud. I will only note that the quest to connect the rationalists—who are not representative of Silicon Valley writ large and have far, far less influence there than the Times would have its readers imagine—to the broader sins of big tech is in part an attempt to pre-empt the question I posed above. "If we can explain why Google is sexist, surely that will give the New York Times the right to break apart this one man’s life for the sake of a column, right?" On the other hand, if Slate Star Codex’s influence is limited to claiming a handful of tech billionaires and a few dozen Heterodox Academy types among its readership, the justification for forcing Scott to choose between his career as a psychiatrist and his success as an internet writer begins to fade away. [6]

When I read through the debates on this question I often want to ask: did y'all read Gideon Lewis Kraus’ New Yorker piece on Slate Star Codex from last year? His piece is everything Cade Metz’s Time article is not. Kraus has a far stronger grasp of what the rationalist movement is all about, the types of personalities attracted to it, its actual relationship with the tech world, and where it fits into the broader story of American intellectual life. Because Kraus did his homework, his inevitable critiques of the community and their favorite blog land. The rationalist response to that piece was muted: It did not prompt a flurry of angry blogposts and twitter threads. Most Codex readers did not agree with all of it but accepted it as a fair piece of journalism—and the difference in that reaction tells us something about whether the Slate Star Codex audience’s expectations are really that crazy. For my part, I see Kraus’ essay as the best evidence that Metz and his editors overstepped their bounds, engaging in journalistic malpractice—or something very close to it.

3) Assuming things were misrepresented, why did that happen? Was it a premeditated “hit job” or revenge piece? Or is there a better explanation for what happened?

So why did this happen? Here is where I break ranks with the rationalists: all the talk about “hit jobs” is silly and conspiratorial. Prestige media will play this game with very important people—CEOs, politicians, generals, and the like.[7] Scott Alexander is not at that level. Not even close. Scott Aaronson might think that the demise of Slate Star Codex would be “an intellectual loss on the scale of, let’s say, John Stuart Mill or Mark Twain burning their collected works,” but just about no one outside of the rationalist community would agree with him. [8]

On this account Elizabeth Spiers is absolutely correct:
“Only in a bubble as insular and tiny as the SSC community would this theory be even remotely plausible. To put this in context: SSC is influential in a small but powerful corner of the tech industry. It is not, however, a site that most people, even at The New York Times, are aware exists—and certainly, the Times and its journalists are not threatened by its existence. They are not out to destroy the site, or “get” Scott, or punish him. At the risk of puncturing egos: they are not thinking about Scott or the site at all. Even the reporter working on the story has no especial investment in its subject. That reporter is also probably working on six other stories at the same time, thinking about their friends, family, what their kid needs to do in Zoom school tomorrow, the book they want to read, whether Donald Trump will get arrested, whether rats dream of boredom. They do not sit around thinking about how they’re going to “get” people they write about, and when subjects think they do, it’s more a reflection of the subject’s self-perception (or self-importance) and, sometimes, a sprinkling of unadulterated narcissism. [9]

Now this does not excuse Metz’s shoddy reporting—in a way it makes it even less ethical, a point we will return to later—but it does change the nature of the problem. If only this saga were a matter of malice or an empty chase for clicks! [10] That would be easy to solve. Unfortunately the central problem here is larger, and more difficult to fix.

Not to get too meta, but it might be useful to think of the issue like this: every person on the Earth perceives not the Earth itself but a representation of the Earth that their brain has built. Walter Lippman explained it a century ago:
The real environment is altogether too big, too complex, and too fleeting for direct acquaintance. We are not equipped to deal with so much subtlety, so much variety, so many permutations and combinations. And although we have to act in that environment, we have to reconstruct it on a simpler model before we can manage it.[11]

Azar Gat made a similar observation more recently:
In order to cope with their environment, humans strive to identify, understand, and explain the forces operating within and behind it, so that they can at least predict, and if possible, also manipulate these forces and their effects to their advantage. They are predisposed to assume that such forces are there. With respect both to their natural and human environment, humans achieved impressive successes in using these methods. The quest for an understanding thus evolved into a fundamental human trait. Humans must have answers as to the reasons and direction of the world around them. Stretching this faculty the furthest, humans have a deep emotional need for a comprehensive interpretive framework, or set of interpretive 'stories' that would explain, connect the various elements of, and give meaning to their world and their own existence within it. They need a cognitive map of, and a manipulative manual for, the universe, which by lessening the realm of the unknown would them a sense of security and control, allay their fears, and alleviate their pain and distress. [12]

This is true even for even the most concrete of human experiences: the hunter, logger, and geologist will walk through the same patch of wilderness and see an entirely different forest, for each eye is trained to notice something different. The more abstract the things observed the greater individual variance there will be. For intangible social processes like market exchange, mass movements, and elections, our understanding is all model, no matter. [13]

This need to reduce reality to a simple mental model is an inherent feature of human cognition. For the most part it is done automatically without much thought. We cannot avoid simplification—we speak of London doing this or China doing that not because such simplifications are true (there is no unitary agent named “London” or “China” doing anything) but because it is impossible to act in a complex world without such short cuts.[14]

The problems of journalism are the problems of cognition on steroids. For the journalist, historian, or social scientist, the drive to reduce is acute and explicit. On top of the normal simplification we all do unconsciously, nonfiction writers must reduce twice more: The first round of reduction comes with investigation. Any subject is too large to be understood in toto. The investigator must decide where to focus her efforts, how to spend limited time, what sources to consult, what questions to ask, and what sort of evidence to be on the lookout for. Many of these things are not explicitly decided, but are forced upon the investigator by the nature of her tools and sources or by her preconceived sense of what is notable and what is not.

The second round of simplification, just as inherent to the journalistic enterprise as the first, is built into act of writing. The investigator has collected in her brain more that can ever be put on a page. Journalists in particular must condense what they have learned onto a very small space. This double reduction process is often described as “framing” a story. Reducing an entire movement—the histories, controversies, disagreements, defeats, glories, and quirks of thousands of unique individuals—to one comprehensible frame will always cut important things out. It is inevitable that some members of the covered group will be dissatisfied with the frame they have been forced into.

This process, far more than any explicit ideological agenda, is the source of most bias in journalism. This source of bias cannot be escaped. Stories without a frame are just an incoherent collection of facts too long and too varied to fit on a page. The bias imposed by framing is necessary—and sometimes even a good thing.

I am here reminded of my Chinese friends who complain about Western journalists' disproportionate focus on dissident and human rights stories in China. My friends are right to complain, in their own way: the story of China's persecuted minorities is only a bit part in the vast universe of experiences and events that is China. I don't mind this particular bias, however. Nor do I find the dominance of diplomatic, security, and macroeconomic stories about China particularly distressing. Journalists and their editors carry with them a set of a priori beliefs on what is actually important (“newsworthy”). In this case I find it difficult to argue that these beliefs are wrong.

The trouble comes when attachment to a given frame leads journalists into misperceiving their subjects, forcing them into a framework that does not really fit them. If you are primed to think of internet subcultures through the gamergate frame, gamergate is all you will ever find. In the terminology of the rationalists, it is a problem of “priors.” All that was required for a mess like this was a writer with wildly different priors and tight time demands to come into contact with a community they only had a superficial understanding of. No active malice is necessary.

Unfortunately, if this is a problem inherent to journalism, the particular practices of the New York Times editorial team aggravate the issue. Listen to one ex-Times editor compare his time at the Gray Lady to his earlier career at the Los Angeles Times:
For starters, it’s important to accept that the New York Times has always — or at least for many decades — been a far more editor-driven, and self-conscious, publication than many of those with which it competes. Historically, the Los Angeles Times, where I worked twice, for instance, was a reporter-driven, bottom-up newspaper. Most editors wanted to know, every day, before the first morning meeting: “What are you hearing? What have you got?”

It was a shock on arriving at the New York Times in 2004, as the paper’s movie editor, to realize that its editorial dynamic was essentially the reverse. By and large, talented reporters scrambled to match stories with what internally was often called “the narrative.” We were occasionally asked to map a narrative for our various beats a year in advance, square the plan with editors, then generate stories that fit the pre-designated line. 
Reality usually had a way of intervening. But I knew one senior reporter who would play solitaire on his computer in the mornings, waiting for his editors to come through with marching orders. Once, in the Los Angeles bureau, I listened to a visiting National staff reporter tell a contact, more or less: “My editor needs someone to say such-and-such, could you say that?” [15]

All humans naturally tack new developments to pre-existing mental models. Far from fighting this mental tick, the New York Times mandates that its reporters explicitly commit their reporting to certain narrative arcs before they have even begun to really investigate! This is not usual journalistic practice. It has predictable consequences. About two years ago the New York Times signaled that an increasing percentage of their reporting would be devoted to a new guiding narrative. That narrative goes something like this: “shed light on the ideas, institutions, and personalities that exacerbate racial and gender inequity in American life, creating the sort of world where Donald Trump can be president.” Times reporters went searching for stories that might fit the bill. Little wonder that they have found so many!

I suppose if your sole definition of “newsworthy” is “something that exacerbates racial and gender inequity in American life” then you will struggle to see what was wrong with this decision. If that is your definition of “newsworthy,” fine, I can’t criticize you—it isn’t really any different from my finding security stories the most newsworthy things to come out of China. Just be open about the trade-offs of this approach. A journalist who conceives of her beat in terms of a predetermined frame will end up finding them—but inevitably there will be stories shoehorned into a framing that they poorly fit.

Question four: Do journalists have the right to uproot the lives of their subjects lives with negative coverage? Do communities so targeted have the right to impose costs on journalists (say, by harassing them on twitter or flooding their inboxes) that are “just doing their job?”

Historians face all the same problems that journalists do. But when historians debate framing and bias the stakes are low. At the end of the day, their debates are strictly academic. Journalists deal instead with living, breathing people. On this front, many of the folks manning the battlements at high prestige publications lack self-awareness. It can be difficult to get them to understand the true nature of their work. Whatever else it might aspire to be, journalism is an exercise in power. No one who wields power should be surprised when those subjected to it resist.

Part of the problem is that full-time writers thrive in the limelight. For a public intellectual like Will Wilkinson all press is good press; he lives in a world of ceaseless self-promotion, and is not properly situated to understand what being targeted by an international media outlet feels like for folks outside of that world. In her book Liquidated, anthropologist-cum-Wall Street trader Karen Ho makes a parallel observation about investment bankers: thriving in a career marked by transience and risk, socialized to believe it is normal to be let go at any time, they have little compunction reshaping corporate America in their own image and even less sympathy for Americans not as adept at dealing with job insecurity as they are. [16] Fish will never understand fear of deep water.

To make things clear for the fish: humans have a strong, perhaps even innate, need to tell their own story. This is what makes social media so addicting—it allows you to endlessly curate your own self-image, forever perfecting your personal story for others’ consumption. Losing the ability to tell our own story feels like loss of agency—a violation. The teenage girl subject to a high school whisper campaign feels violated even though no one has touched her. I will not call it an act of “violence,” though some have.[17] “Violation” describes the experience well enough.

Recognizing this is a natural, human experience does not mean all humans have a natural right to narrative control. But it does mean that journalists must be more sensitive to the nature of their enterprise—especially if they write for an outlet with the reach of the New York Times. A story like this may very well be, as Spiers argues, just one of “six other stories” a journalist is “working on… at the same time,” of less interest to them than whether “rats dream of boredom.” It may just be a few hundred words out of hundreds of thousands the journalist will write, the end product of a few humdrum hours on the daily grind. Fine. But for the subjects of a piece—say for the 7000 rationalists committed enough to their community to take the annual seventy-plus question SSC questionnaire—this sort of thing is a once-in-a-lifetime event.

A journalist descends from the New York Times like an immortal from Olympus. If you read your myths you know how those moments go. The Olympians have their fun. They then return heavenward as glowing and unsoiled as they descended. It is for the mortals left behind to bear the scars of their exchange. 

Journalists resist this message. They have so long internalized that their role is "speaking truth to power” that they fail to see when they are the power. So let me be explicit: if you have a staff position at the New York Times you are the power. When you are writing for that paper you have the power to determine what millions of people will think about an individual, movement, or event. You have the power to decide the first thing people will find when they Google search your subject—forever. That is power. If every piece you file does not have you in awe at your own responsibility, you are doing this wrong. Unless your subject sits atop a corporate hierarchy, is an elected politician, an appointed high official, a commanding general, has a net worth of more than $5 million, or a loyal following with similar numbers, then your critical coverage is “punching down.” That is simply the nature of the institution you work for.

This fact can be obscured by tech titans eager to co-opt incidents like these as part of their own struggle. Just as the rest of us do, folks like Srinivasan interpret events like these through an a priori frame. As is true with most all of us, these frames can be self-serving. Srinivasan's cause has something to gain by reinterpreting “the New York Times causes controversy with lazy reporting” as “legacy East Coast institutions continue their unrighteous crusade against Silicon Valley upstarts.” The Times’ own reporting choices lend themselves to such a reading, of course, but that does not mean this is actually what happened. The powers that be would like it if this what was happening: Times journalists want to pretend they are adroitly interrogating big tech; tech titans want to pretend they slay the vampire institutions of America’s rentier class. Behind the pomp is a more pedestrian reality: to fit his reporting into a pre-existing frame, a sloppy journalist mischaracterized an internet subculture.

Few members of that subculture ever wished to be proxies in this battle. Mortals know that when Olympians feud, it is never Olympians who die. Besides, these people have their own concerns. It may be worth while listing some of these out. It will be a good reminder of what the rationalist community is actually about: 

  •  Slightly less than a third of Slate Star Codex readers have anxiety; slightly more than a third have depression. 
  • Just under a fourth of Slate Star Codex readers are on the Autism spectrum (or believe themselves to be). 
  • 10% of Slate Star Codex readers have attempted to commit suicide. Another 26% have seriously considered it. Two thirds of those who attempted suicide regret their attempt did not succeed. [18]

Behind the pretensions to rationalist perfection is a community of people acutely aware of their own atypicality. Sebsastian Benthall calls it a “therapeutic community.”[18] He is right to do so. At the end of the day, rationalism is a giant support group—a philanthropically minded attempt to provide that transcendent sense of community and meaning “normies" find at church.

Like all moral communities, its members react defensively when violated by outsiders. There is nothing surprising or wrong about that.

It sucks if you are the journalist involved, of course. And I concede that sometimes it might be necessary, and even good, for journalists at a high-prestige publication to “punch down.” But given the harm they are about to inflict, these journalists then bear a special responsibility to make sure they know what they are talking about. But how to ensure that they take this responsibility seriously? How to keep their violations in bounds?

The easiest answer, the only practicable one that doesn’t involve vast internal reforms to institutions like The New York Times or a complete transformation of the American media ecosystem, is something like the fabled “delicate balance of terror.” Tit meets tat; lazy harms are matched by rather more intentional ones. From this perspective it is hard to be sympathetic to Times editors upset with the torrent of angry e-mails they have received in response to all this. They are in a position of terrible power. If communities like these are not ready to defend themselves, who else will keep those wielding this power honest?

Well that is it for tonight folks. I have things yet to say about the destructive tendencies of America’s media ecosystem, which is too centered around one big newspaper; thoughts on why technologists and journalists are talking past each other when they debate “bias” in negative tech coverage; and some skeptical swipes at the assumptions behind the proposed tech rebellion against everything East Coast. But this post is already several thousands words long. My thoughts on these other questions will be published next week.

EDIT 26 February 2021: Balaji Srinivasan requested I remove the phrase "tech secessionism" from this post as it does not accurately describe the aims of his program, instead being imposed on it by outside observers (New York Times journalists!) who misconstrued its aims. I am happy to change these references to his preferred moniker ("decentralists"). But more on that next week!


Readers who enjoyed this post might find my other takes on internet discourse of interest: see the posts "The World Twitter Made," "Why Writers Feud So Viciously," "On the Angst of the American Journalist," and "Public Intellectuals Have a Short Shelf Life -- But Why?"  To get updates on new posts published at the Scholar's Stage, you can join the Scholar's Stage mailing list, follow my twitter feed, or support my writing through Patreon. Your support makes this blog possible.


[1] Cade Metz, “Silicon Valley’s Safe Space,” New York Times (13 February 2021). 

2] Robby Soave, “What The New York Times' Hit Piece on Slate Star Codex Says About Media Gatekeeping,” Reason (15 February 2021); Micah Meadowcroft, “On Writing Around Censors,” Conservative American (20 February 2021); Matthew Yglesias, “In Defense of Interesting Writing on Controversial Topics,” Slow Boring (13 February 2021); Freddie Deboer, “Scott Alexander is not in the Gizmodo Media Slack,” personal weblog (15 February 2021); Scott Aaronson, “A grand anticlimax,” Shetl-Optomized (14 February 2021); Noah Smith, “Silicon Valley Isn’t Full of Fascists,” Noahpinion (13 February 2021); Dan Drezner, "Everything Old is New Again in Mainstream Media," Washington Post (17 February 2021); Scott Siskind, “Statement on the New York Times Article,” Astral Codex Ten (13 February 2021).

[3] Elizabeth Spiers, “Slate Star Clusterfuck,” My New Brand Is (14 February 2021); Will Wilkinson, “Gray Lady Steelman,” Model Citizen (19 February 2021); Elizabeth Sandifer, “The Beigeness, or How to Kill People with Bad Writing: The Scott Alexander Method,” Erudotirum Press (20 February 2021). 

[4] Gideon Lewis Krass, “Slate Star Codex and Silicon Valley’s War Against the Media,” The New Yorker (9 July 2020); Sebastian Benthall, “Social justice, rationalism, AI ethics, and libertarianism,” Digifesto (21 Feb 2021)

[5] OK, here we go. 

Although I have done a bit of journalism myself, I am best described as an essayist. My work has been published in numerous magazines and journals of medium prestige. I have never written for the New York Times. But I make my living as a writer, and the writer's milieu is now my own. In addition, I am personal friends with many members of the China reporting corps and editors for major newspapers and magazines in the Asia trade. I go to journalist wedding showers and play board games with them over pizza. I have had more debates on issues like these with them in person than I can count. 

As for the rationalist side of the equation—I am not a rationalist nor would anyone who has read me for a long time consider me in their number. I am not even part of the broader “grey tribe.” I read poetry, not code. I am a red tribe member with the basic social attitudes and opinions you would expect to find in the type of person who voted for George W. Bush in 2004 or John McCain in 2008. I like Scott's blog, but never read it religiously, and never bothered commenting on it after he put in the fancy log-in system to filter out bad comments.

 Despite not being a proper Scott Stan, I was one of the five people who helped organize the 2020 petition imploring the NYT editors to refrain from releasing Scott's true identity. I did this because I don't think it is right for a major newspaper to make an anonymous writer choose between writing online and their existing profession, and because I viewed Slate Star Codex and its long comment threads as an exemplary model of reasoned debate and humane liberalism, something America has in short supply these days. 

[6] As an aside, were I the one trying to be "critical" of the rationalists, I would write much less about how their comment threads are hostile to feminism--which isn't true in any case, there were plenty of feminists in those threads, more of those commenting than there ever were neoreactionaries--and more about how they continually agitate for people to give part of their income to stopping Skynet. I kid you not. If this is not a grift, what is? How did that not make it into the article while all this nonsense about the Silicon Valley psyche did?

Well, we know the answer to that. The absurdity of the AI risk project is so far outside of the NYT's existing narrative frame that this detail did not even register.

[7] Cue Hamilton Noah, public editor for the Washington Post:

Journalism, particularly at the highest level, is about raw power. It is about bringing important people to heel, on behalf of the public. Politicians and officials and business leaders don’t want to talk to the press, subjecting themselves to the possibility of being made to look bad; they do it because they have always felt they had no choice. They felt that way because papers like the Post could offer the carrot of great exposure to those who needed it, but also, always, the stick of negative coverage to those who spurned them. There is nothing devious or ignoble about this; a powerful press, for all its flaws, is good for democracy, and tends to promote equality by holding the big shots in check. Anyone who has ever negotiated to land a contentious interview with a famous person knows that you only get those interviews when your subject fears what will happen if they don’t do the interview. Today, that fear is disappearing. We all need to figure out what to do about that. 

Hamilton Noah, "Washington Post public editor: The powerful have realized they don’t need the Post," Columbia Review of Journalism (20 October 2020).

[8] Scott Aaronson, “Pseudonymity as a trivial concession to genius,” Shetl-Optomized (23 June 2020). 

[9] Spiers, “Slate Star Clusterfuck.” 

[10]This idea that the New York Times published things for “the clicks” is common but inaccurate. Vox publishes things for the clicks. The New York Times, like other top tier publications such as The New Yorker or the Washington Post, make their money from subscriptions and side services--like that high school trip to Peru that got a certain Times reporter fired. The New York Times is rolling in dough, and that dough has nothing to do with the virality of any given article. In fact, there is a good chance that uber-viral articles cost them more readers  than they gain from them. No one subscribed because of 1619 or the Cotton op-ed, but a lot of people did unsubscribed because of them!

Likewise most writers, regardless of publication, care very little about their hit count. At most publications individual writers are not even told site traffic stats for individual pieces. Only in rare cases is payment tied to popularity. What motivates writers and journalists is not clicks but prestige. They measure their self worth through the esteem of their fellow writers, and write to that end. For more on this see my post "Why Writers (And Think Tankers) Feud So Viciously."

[11] Walter Lippman, Public Opinion (New York: Harcourt, Brace, and Company: 1922), pp. 8-9. 

[12] Azar Gat, War in Human Civilization, (Oxford: Oxford University Press, 2006), p. 101. 

[13] You know, gorilla experiment and all that jazz.

[14] Paul Boyer, Minds Make Societies: How Cognition Explains the Worlds Humans Create (New Haven: Yale University Press, 2018), 

[15] Michael Cieply, “Stunned By Trump, The New York Times Finds Time For Some Soul-Searching,” Deadline (10 November 2016).

  [16] Karen Ho, Liquidated: An Ethnography of Wall Street (Chapel Hill: Duke University Press, 2009). 

 [17] See for example this statement in Benthall, "Social justice, rationalism, libertarianism, and AI ethics:"

 When the NYT notices something, for a great many people, it becomes real. NYT is at the center of a large public sphere with a specific geographic locus, in a way that some blogs and web forums are not. So whether it was justified or not, Metz’s doxing of Seskind was a significant shift in what information was public, and to whom. Part of its significance is that it was an assertion of cultural power by an institution tied to old money in New York City over a beloved institution of new money in Silicon Valley. In Bourdieusian terms, the article shifted around social and cultural capital in a big way. Seskind was forced to make a trade by an institution more powerful than him. There is a violence to that.

I agree with the sentiment and find it well expressed, but draw a hard line on restricting the word "violence" to physical acts of coercion and injury.

[18] Slate Star Codex Reader Survey 2020, results available here: https://docs.google.com/forms/d/e/1FAIpQLSd4I-x9oArWW1Tz5mEK4uHmxcJzVKGA28RfKPsDvW8hzZNViw/viewanalytics

 [19] The entire passage is worth quoting:

 Certainly in elite intellectual circles, the idea that participation in a web forum should give you advanced powers of reason that are as close as we will ever get to magic is pretty laughable. What I think I can say from personal experience, though, is that elite intellectuals seriously misunderstand popular rationalism if they think that it’s about them. 
….Many of these people (much like Elizabeth Spiers, according to her piece) come from conservative and cloistered Christian backgrounds. Yudkowsky’s Harry Potter and the Methods of Rationality is their first exposure to Bayesian reasoning. They are often the best math students in their homogeneous home town, and finding their way into an engineering job in California is a big deal, as is finding a community that fills an analogous role to organized religion but does not seem so intellectually backwards. I don’t think it’s accidental the Julia Galef, who founded CFAR, started out in intellectual atheist circles before becoming a leader in rationalism. Providing an alternative culture to Christianity is largely what popular rationalism is about. 
From this perspective, it makes more sense why Seskind has been able to cultivate a following by discussing cultural issues from a centrist and “altruistic” perspective. There’s a population in the U.S. that grew up in conservative Christian settings, now makes a living in a booming technology sector whose intellectual principles are at odds with those of their upbringing, is trying to “do the right thing” and, being detached from political institutions or power, turns to the question of philanthropy, codified into Effective Altruism. This population is largely comprised of white guys who may truly be upwardly mobile because they are, relative to where they came from, good at math. The world they live in, which revolves around AI, is nothing like the one they grew up in. These same people are regularly confronted by a different ideology, a form of left wing progressivism, which denies their merit, resents their success, and considers them a problem, responsible for the very AI harms that they themselves are committed to solving. If I were one of them, I, too, would want to be part of a therapeutic community where I could speak freely about what was going on.
Bentham, "Social justice, rationalism, libertarianism, and AI ethics,"